Journal

Vol. 27 No. 3, 2024

Table of Contents

ORIGINAL ARTICLES

The Significance of Computed Tomography–Detected Breast Lesions

   CME

LY Lam, KM Chu, HHC Tsang, WC Wai, JLF Chiu

ORIGINAL ARTICLE    CME
 
The Significance of Computed Tomography–Detected Breast Lesions
 
LY Lam, KM Chu, HHC Tsang, WC Wai, JLF Chiu
Department of Radiology and Imaging, Queen Elizabeth Hospital, Hong Kong SAR, China
 
Correspondence: Dr LY Lam, Department of Radiology and Imaging, Queen Elizabeth Hospital, Hong Kong SAR, China. Email: lly858@ha.org.hk
 
Submitted: 14 March 2023; Accepted: 5 October 2023.
 
Contributors: LYL, HHCT and WCW designed the study and acquired the data. All authors analysed the data. LYL drafted the manuscript. KMC, HHCT, WCW and JLFC critically revised the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.
 
Conflicts of Interest: All authors have disclosed no conflicts of interest.
 
Funding/Support: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
 
Data Availability: All data generated or analysed during the present study are available from the corresponding author on reasonable request.
 
Ethics Approval: This research was approved by the Kowloon Central Cluster and Kowloon East Cluster Research Ethics Committee of Hospital Authority, Hong Kong (Ref No.: KC/KE-22-0059/ER-4). The requirement for patient consent was waived by the Committee due to the retrospective nature of the research.
 
 
 
 
 
Abstract
 
Introduction
 
With ever-increasing computed tomography (CT) utilisation, more breast lesions are incidentally detected. We sought to investigate the yield of undiagnosed cancers from incidental CT-detected breast lesions. Imaging features were compared with the pathological diagnoses.
 
Methods
 
A retrospective analysis of CT examinations in a regional hospital in Hong Kong between January 2018 and December 2020 was performed. Patients without a history of breast diseases whose CT reports contained the keyword ‘breast’ and who were referred for a formal breast examination were included. Two radiologists reviewed the CT reports and lesion characteristics were recorded. The diagnostic accuracy of different CT features was evaluated.
 
Results
 
A total of 219 breast lesions were included. Forty-eight lesions (21.9%) were malignant. Patients with malignant diagnoses were older in age than those with benign diagnoses (mean age: 67.58 vs. 56.60 years; p = 0.05). Spiculation, irregularity, rim and heterogeneous enhancement, as well as the presence of abnormal lymph nodes were useful in predicting malignancy. Lesion size and presence of calcifications were not useful for predicting the pathological diagnosis.
 
Conclusion
 
More than 20% of breast lesions detected incidentally in CT examinations were malignant. The presence of spiculation and irregularity were positive predictive factors. A careful review of the breasts on CT studies including the chest should always be part of routine practice.
 
 
Key Words: Breast; Breast neoplasms; Diagnostic imaging; Incidental findings; Lymph nodes
 
 
中文摘要
 
電腦斷層掃描檢測到乳腺病變的重要性
 
林樂宜、朱嘉敏、曾凱晴、衛頴莊、趙朗峰
 
引言
隨着電腦斷層掃描使用愈趨普及,越來越多的乳腺病變被偶然發現。我們調查電腦斷層掃描偶然檢測到的乳腺病變中未確診癌症的發生率,並比較影像學特徵與病理診斷。
 
方法
我們就2018年1月至2020年12月期間香港某地區醫院的電腦斷層掃描檢查進行回顧性分析。本研究納入沒有乳腺疾病史而電腦斷層掃描報告中包含「乳腺」關鍵字並轉診進行正式乳腺檢查的患者。兩名放射科醫生審查了電腦斷層掃描報告。本研究記錄了患者的病變特徵,並評估不同電腦斷層掃描特徵的診斷準確性。
 
結果
本研究共納入219個乳腺病灶,48個病變(21.9%)為惡性。診斷為惡性病變的患者年齡比診斷為良性病變的患者年齡大(平均年齡:67.58歲與56.60歲;p = 0.05)。毛邊、不規則、邊緣和不均勻強化以及異常淋巴結的存在有助預測惡性腫瘤。病灶大小和鈣化的存在無助預測病理診斷。
 
結論
電腦斷層掃描檢查偶然發現的乳腺病變中20%以上為惡性。存在毛邊和不規則是陽性預測因素。透過電腦斷層掃描檢查仔細檢查胸部乳腺應成為常規。
 
 
 
INTRODUCTION
 
Breast cancer is one of the commonest cancers in Hong Kong, comprising 28.4% of female cancers in 2020.[1] In Hong Kong, mammography and sonography are the preferred initial modalities in the evaluation of breast lesions.[2] In some other countries, mammography is the mainstay for breast cancer screening.[3] In an era with increasing utilisation of computed tomography (CT), more breast lesions are detected incidentally when CT scanning is performed for other indications such as pulmonary or cardiac conditions.[4] [5] [6] Although dedicated mammography and sonography are still required for a better evaluation of the lesions, it is still important for radiologists to detect imaged breast lesions and to characterise them when such lesions are encountered on a CT scan. With an increased detection of undiagnosed breast cancer, a prompt referral of suspicious lesions for further investigation can help improve patient outcome. We aimed to investigate the yield of breast cancers from incidentally detected breast lesions on CT in Hong Kong. Their imaging features were correlated and compared with the final pathological diagnoses.
 
METHODS
 
A retrospective analysis of the CT examinations scanned in Queen Elizabeth Hospital from 1 January 2018 through 31 December 2020 was performed. Patients whose CT reports contained the keyword ‘breast’ and were referred for a formal breast examination were included. The CT reports were derived from the Radiological Information System and Picture Archiving and Communication System, which is a system managed by Hospital Authority. Patients who had a history of breast diseases or breast surgeries were excluded from the study.
 
The CT scanners used in this study included SOMATOM Force Ultra-Fast Dual Source CT Scanner (Siemens Healthcare, Erlangen, Germany), Aquilion CXL 128 Slice CT Scanner (Toshiba, Tochigi, Japan), and Aquilion Prime CT Scanner (Canon Medical Systems, Tochigi, Japan), with a section thickness of 5 mm. The types of contrast used for enhanced procedures were iohexol (Omnipaque 350; GE HealthCare, Milwaukee [WI], US) and iodixanol (Visipaque 320; GE HealthCare, Milwaukee [WI], US), with a standard adult dose of 90 mL, administered via a pump injector. In examinations of the abdomen requiring injected contrast material imaged in different phases, the contrast was administered for the arterial phase at 3.5 mL/s or in the portal venous phase at 2.5 mL/s with a standard 70-second delay.
 
The CT images were reviewed by and commented on by two experienced radiologists (with 6 and 8 years’ experience in breast imaging, respectively) who were blinded to the diagnostic outcome. As there is no formal lexicon for breast lesions detected in a CT scan, the descriptors used in this study were adapted from the Breast Imaging and Reporting Data System terminology for magnetic resonance imaging lexicons (5th edition).[7] [8] The axillary lymph nodes were considered abnormal if: (1) their longest-to-shortest axis ratio was <2; (2) they lacked a fatty hilum; (3) there was cortical thickening of >3 mm; or (4) their cortices were eccentric.[9]
 
Continuous variables were presented as mean ± standard deviation, and categorical variables were presented as frequencies. The Mann-Whitney U test was used to evaluate the distribution of continuous data. Fisher’s exact test was performed to assess the correlation of the CT features with final pathological diagnosis. The specificity and sensitivity for malignancy were calculated for the significant CT features. A p value of < 0.05 was considered to be statistically significant.
 
RESULTS
 
Demographics and Study Cohort
 
A total of 22,255 CT studies of the thorax with or without abdominal regions were performed during the study period. Among these CT examinations, 2,575 of the reports contained the keyword ‘breast’. A total of 347 patients without history of prior breast disease or surgery were noted to have one or more incidental breast lesions in the CT studies; 345 were women and two were male. A total of 224 patients were referred for further formal breast assessment and investigation, among which 188 had subsequent formal breast investigations, nine had defaulted appointments, and 27 were still pending appointments at the time of the study. Among these 188 patients (186 female and 2 male), 164 patients had a solitary lesion, while 17 patients had two lesions and seven patients had three lesions, for a total of 219 incidentally detected breast lesions (Figure 1).
 
Figure 1. Patient selection
 
Referral Rates
 
The overall referral rate for formal breast investigations during the study period was 64.6%. It was lowest in 2018 (61.0%) and highest in 2019, which was 70.8%.
 
Formal Breast Assessment Findings
 
Among the 219 breast lesions undergoing formal breast assessment, 88 were classified as ‘normal’ or ‘with benign appearance’ by clinical examination, mammography, and ultrasonography. The remaining 100 patients have undergone ultrasound-guided biopsy in our institute. In these patients, 83 lesions were found to be benign and 48 lesions were malignant (Figure 2), most of which were invasive ductal carcinoma (Table 1).
 
Figure 2. Results of formal breast assessments
 
The malignancy rate was 21.9% (48 out of 219 lesions). Patients with malignant lesions were likely to be older in age compared with those with benign findings (mean age: 67.58 vs. 56.60 years, p = 0.05).
 
Table 1. Core biopsy pathology of incidental breast lesions (n = 131).
 
Lesion Characteristics
 
CT measurement showed that the malignant lesions (mean size: 1.56 cm) were larger than the benign lesions (mean size: 1.23 cm) but not statistically significant (p = 0.08). Among the morphological characteristics of the breast lesions (Table 2), more lesions with an irregular shape or non-circumscribed margin were diagnosed as malignant (p ≤ 0.001). The malignancy rate (i.e., positive predictive value [PPV]) of all irregular lesions was 80%. The sensitivity and specificity of an irregular shape were 25% and 98.2%, respectively, while the malignancy rate of all non-circumscribed lesions was 53.1%. The calculated sensitivity and specificity of a non-circumscribed margin were 54% and 86.5%, respectively. Among the two descriptors for non-circumscribed margins, a spiculated margin had a malignancy rate of 100% with sensitivity and specificity of 16.6% and 100%, respectively, and is more indicative of malignancy (p < 0.05) [Table 3 and Figure 3].
 
Table 2. Morphology and enhancement patterns of the breast lesions (n = 219)
 
Table 3. Diagnostic performance of suspicious computed tomography features in the differentiation of malignant from benign incidental breast lesions
 
Figure 3. A 67-year-old woman underwent a plain computed tomography (CT) scan of the thorax for prolonged cough. (a) Axial CT scan reveals an incidental irregular shaped mass with spiculated margins in the upper outer quadrant of the right breast (arrow), with suspicious involvement of the right pectoralis muscle. (b) Ultrasound scan of the right breast confirms the hypoechoic mass with irregular shape and spiculated margin at the right 9 o'clock position (arrow). (c, d) Mammograms showing the right breast mass as a high-density irregular mass with spiculated margins that closely abuts the pectoralis muscle (arrows). Biopsy was performed and pathological examination confirmed invasive ductal carcinoma
 
Four biopsy-proven malignant lesions contained calcification. However, the association between calcification and malignancy of the lesions was not statistically significant (p > 0.05) [Table 4 and Figure 4].
 
Table 4. Presence of abnormal axillary lymph nodes or calcifications in the breast lesions (n = 219)
 
Figure 4. A 59-year-old woman underwent a contrast-enhanced computed tomography scan of the thorax for shortness of breath. An incidental breast mass showing peripheral enhancement is seen in the upper part of the left breast (arrow). Subsequently she underwent biopsy in the private sector that revealed invasive ductal carcinoma
 
Contrast Enhancement
 
A total of 172 lesions were evaluated in contrast-enhanced CT scans, with 113 of them showing contrast enhancement (Table 2). All of the malignant lesions showed enhancement. The sensitivity and specificity of contrast enhancement were 100% and 44.4%, respectively, for a negative predictive value of 100% for lack of enhancement (Table 3). Furthermore, all lesions that showed rim enhancement were malignant, giving a malignancy rate of 100% (p < 0.05) with specificity of 100% (Table 2 and Figure 5).
 
Figure 5. A 60-year-old woman with a contrast-enhanced computed tomography scan of the abdomen for abdominal pain. (a) An incidental breast mass with lobulated shaped and circumscribed margins is seen at the upper outer quadrant of the left breast with subtle internal calcifications (arrow). (b) Mammography showing a high-density, irregularly shaped mass with spiculated margins and internal pleomorphic calcifications (arrow). (c) Sonography showing the large left breast mass with internal echogenic foci suggestive of calcifications. This mass was later biopsied and confirmed to be invasive ductal carcinoma.
 
Presence of Axillary Lymph Nodes
 
The axillary regions were included in the CT scan range in 126 patients. A total of 19 patients were found to have abnormal axillary lymph nodes, of which 12 of them have biopsy-proven malignant breast lesions. The association between presence of abnormal-looking axillary lymph nodes and breast malignancy was found to be statistically significant (p < 0.001) [Table 4].
 
DISCUSSION
 
Our study shows an increase in referral rate for dedicated breast imaging from 2018 to 2020, with increased reporting of CT-detected breast lesions. Despite a similar number of total CT scans done in our institute annually, this may be due to an increased awareness of CT-detected breast lesions leading to referral to the breast imaging units for characterisation.
 
In our study, 21.9% of the incidental CT-detected breast lesions were proven to be malignant after biopsy in the breast unit, i.e., out of the 22,255 CT studies performed in 2018 to 2020, 43 patients (five of them with more than one incidental breast tumour) were ultimately diagnosed with unsuspected breast cancer. Hence, the extrapolated breast cancer detection rate by CT scans is 1.9 cases per 1000 population based on our findings. A retrospective review by a local public hospital performed in the 5-year period from 1998 to 2002 showed that the breast cancer detection rate by mammogram is 5 cases per 1000 population,[2] in agreement with the concept that CT scan alone is not a better screening test than mammography.
 
Of the incidentally detected breast cancers in our study, most cases were invasive ductal carcinoma (70.8%) [Table 1], similar to the incidence of invasive ductal carcinoma in the general population of 75%.[10] [11] Ductal carcinoma in situ accounted for 20.8% of the incidentally detected breast cancers in our study (Table 1). Although CT lacks the resolution for microcalcifications, these cases were detected as breast masses.[12]
 
The most suspicious features for malignancy were found to be an irregular shape (malignancy rate of 80%) and spiculated margin (malignancy rate of 100%) [Table 2]. These results were in keeping with other studies across different modalities including mammography and sonography. Liberman et al[13] reported a PPV for malignancy of 73% for irregular shape and 81% for spiculated margins for mammographic studies. Inoue et al[14] reported a PPV for malignancy of 99% for irregular shape and 100% for spiculated margins for CT using dynamic dedicated breast CT. Stavros et al[15] reported a PPV for malignancy of 91.8% for spiculated lesions on sonography.
 
On the other hand, we found that oval shape (malignancy rate of 9.8%) and circumscribed margins (malignancy rate of 12.9%) are more indicative of benignity (Table 2). These results are also similar in the study by Moyle et al.[16]
 
The presence of CT-detected calcifications in breast lesions does not show a statistically significant association with the final pathology diagnosis (Table 4). In this study, only four biopsy-proven malignant lesions contained calcification, while the other visible calcifications were associated with benign entities. This is likely due to the fact that CT has limited spatial resolution. Microcalcifications <0.5 mm that are more likely associated with malignancy cannot be detected on non-dedicated CT.[17] Lindfors et al[18] found that CT was worse than mammography for visualisation of microcalcifications.
 
In our study, all of the biopsy-proven breast malignancies showed contrast enhancement with different enhancement patterns. Among these patterns, rim enhancement and heterogeneous enhancement were more indicative of malignancy (malignancy rate of 100% and 81.8%, respectively) [Table 2]. These results are similar to the findings by Moyle et al[16] and Agrawal et al[19] but are opposite from the study by Inoue et al,[14] who made use of dedicated breast CT for their study. The discrepancy can be due to the difference in timing of image acquisition in the CT studies in our study. Also, malignant breast tumours show rapid contrast uptake and washout, which is well known as a type 3 curve.[20] However, one limitation of our study is that the timing of the contrast administration was not fixed for all the CT studies, therefore such contrast enhancement pattern cannot be demonstrated.
 
In our study, the association between the presence of abnormal-looking axillary lymph nodes and breast malignancy was found to be statistically significant (Table 4). Therefore, evaluation of axillary lymph nodes is essential as part of the triple assessment and before sentinel lymph node biopsy. Although axillary ultrasound is more convenient, it is found that the combination of axillary ultrasound, breast CT, and magnetic resonance imaging of the breast yields a better accuracy rate than the use of a single imaging modality.[21]
 
CONCLUSION
 
The breasts are an area for review by CT radiologists, as more breast lesions are being detected incidentally in CT examinations. This study has shown that nearly one in four incidental breast lesions leads to a diagnosis of breast cancer, particularly in older adults, lesions demonstrating spiculation, irregularity or rim enhancement, and in the presence of abnormal axillary lymph nodes. The detection of these incidental lesions can facilitate a timely referral for a formal breast examination, prompt patient management, and better disease outcome.
 
REFERENCES
 
1. Centre for Health Protection, Department of Health, Hong Kong SAR Government. Breast cancer. January 2024. Available from: https://www.chp.gov.hk/en/healthtopics/content/25/53.html#:~:text=It%20accounted%20for%2028.4%25%20of,of%20breast%20cancer%20in%20males.&text=In%202020%2C%204956%20new%20cases,per%20100%20000%20female%20population. Accessed 2 Mar 2023.
 
2. Lui CY, Lam HS, Chan LK, Tam KF, Chan CM, Leung TY, et al. Opportunistic breast cancer screening in Hong Kong; a revisit of the Kwong Wah Hospital experience. Hong Kong Med J. 2007;13:106-13.
 
3. Zeng B, Yu K, Gao L, Zeng X, Zhou Q. Breast cancer screening using synthesized two-dimensional mammography: a systematic review and meta-analysis. Breast. 2021;59:270-8. Crossref
 
4. Molteni T, Patrique PO, Niasme E, Haefliger L, Vo QD. Incidental breast lesion on chest CT scan: a review. Obstet Gynecol Int J. 2022;13:15-7. Crossref
 
5. Son JH, Jung HK, Song JW, Baek HJ, Doo KW, Kim W, et al. Incidentally detected breast lesions on chest CT with US correlation: a pictorial essay. Diagn Interv Radiol. 2016;22:514-8. Crossref
 
6. Monzawa S, Washio T, Yasuoka R, Mitsuo M, Kadotani Y, Hanioka K. Incidental detection of clinically unexpected breast lesions by computed tomography. Acta Radiol. 2013;54:374-9. Crossref
 
7. Spak DA, Plaxco JS, Santiago L, Dryden MJ, Dogan BE. BI-RADS ® fifth edition: a summary of changes. Diagn Interv Imaging. 2017;98:179-90. Crossref
 
8. Magny SJ, Shikhman R, Keppke AL. Breast Imaging Reporting and Data System. 2022 Aug 29. In: StatPearls [Internet]. Available from: https://www.ncbi.nlm.nih.gov/books/NBK459169/. Accessed 30 Dec 2022.
 
9. Lee HJ, Nguyen AT, Song MW, Lee JE, Park SB, Jeong WG, et al. Prediction of residual axillary nodal metastasis following neoadjuvant chemotherapy for breast cancer: radiomics analysis based on chest computed tomography. Korean J Radiol. 2023;24:498-511. Crossref
 
10. Mannu GS, Wang Z, Broggio J, Charman J, Cheung S, Kearins O, et al. Invasive breast cancer and breast cancer mortality after ductal carcinoma in situ in women attending for breast screening in England, 1988-2014: population based observational cohort study. BMJ. 2020;369:m1570. Crossref
 
11. American Cancer Society. Breast Cancer Facts & Figures 2019-2020. 2019. Available from: https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/breast-cancer-facts-and-figures/breast-cancer-facts-and-figures-2019-2020.pdf. Accessed 3 Jan 2022.
 
12. Vaidya Y, Vaidya P, Vaidya T. Ductal carcinoma in situ of the breast. Indian J Surg. 2015;77:141-6. Crossref
 
13. Liberman L, Abramson AF, Squires FB, Glassman JR, Morris EA, Dershaw DD. The breast imaging reporting and data system: positive predictive value of mammographic features and final assessment categories. AJR Am J Roentgenol. 1998;171:35-40. Crossref
 
14. Inoue M, Sano T, Watai R, Ashikaga R, Ueda K, Watatani M, et al. Dynamic multidetector CT of breast tumors: diagnostic features and comparison with conventional techniques. AJR Am J Roentgenol. 2003;181:679-86. Crossref
 
15. Stavros AT, Thickman D, Rapp CL, Dennis MA, Parker SH, Sisney GA. Solid breast nodules: use of sonography to distinguish between benign and malignant lesions. Radiology. 1995;196:123-34. Crossref
 
16. Moyle P, Sonoda L, Britton P, Sinnatamby R. Incidental breast lesions detected on CT: what is their significance? Br J Radiol. 2010;83:233-40. Crossref
 
17. Gazi PM, Yang K, Burkett GW Jr, Aminololama-Shakeri S, Seibert JA, Boone JM. Evolution of spatial resolution in breast CT at UC Davis. Med Phys. 2015;42:1973-81. Crossref
 
18. Lindfors KK, Boone JM, Nelson TR, Yang K, Kwan AL, Miller DF. Dedicated breast CT: initial clinical experience. Radiology. 2008;246:725-33. Crossref
 
19. Agrawal G, Su MY, Nalcioglu O, Feig SA, Chen JH. Significance of breast lesion descriptors in the ACR BI-RADS MRI lexicon. Cancer. 2009;115:1363-80. Crossref
 
20. Yang SN, Li FJ, Chen JM, Zhang G, Liao YH, Huang TC. Kinetic curve type assessment for classification of breast lesions using dynamic contrast-enhanced MR imaging. PLoS One. 2016;11:e0152827. Crossref
 
21. Aktaş A, Gürleyik MG, Aydın Aksu S, Aker F, Güngör S. Diagnostic value of axillary ultrasound, MRI, and 18F-FDG-PET/CT in determining axillary lymph node status in breast cancer patients. Eur J Breast Health. 2021;18:37-47. Crossref
 
 
 

Early Local Community Data on Safety and Efficacy of Fruquintinib in Metastatic Colorectal Cancer

   CME

HK So, TTS Lau, NSM Wong, M Tong, JJ Huang, CY Shum

ORIGINAL ARTICLE    CME
 
Early Local Community Data on Safety and Efficacy of Fruquintinib in Metastatic Colorectal Cancer
 
HK So1, TTS Lau1, NSM Wong2, M Tong3, JJ Huang3, CY Shum1
1 Department of Oncology, Princess Margaret Hospital, Hong Kong SAR, China
2 Department of Clinical Oncology, Tuen Mun Hospital, Hong Kong SAR, China
3 Cluster Quality and Safety Division, Tuen Mun Hospital, Hong Kong SAR, China
 
Correspondence: Dr HK So, Department of Oncology, Princess Margaret Hospital, Hong Kong SAR, China. Email: hk.so@ha.org.hk
 
Submitted: 28 April 2024; Accepted: 2 July 2024.
 
Contributors: HKS, NSMW, TTSL and CYS designed the study. HKS, NSMW, MT and JJH acquired the data. HKS analysed the data and drafted the manuscript. NSMW, TTSL and CYS critically revised the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.
 
Conflicts of Interest: All authors have disclosed no conflicts of interest.
 
Funding/Support: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
 
Data Availability: All data generated or analysed during the present study are available from the corresponding author on reasonable request.
 
Ethics Approval: This research was approved by the Central Institutional Review Board of Hospital Authority, Hong Kong (Ref No.: CIRB-2023-006-2). Informed patient consent was waived by the Board due to the retrospective nature of the research and the use of anonymised data.
 
 
 
 
 
Abstract
 
Introduction
 
Fruquintinib, a selective inhibitor of vascular endothelial growth factor receptor-1, -2, and -3 tyrosine kinases, is indicated for late-line treatment of metastatic colorectal cancer (mCRC). This retrospective study aimed to review the safety and efficacy of fruquintinib in the Hong Kong population.
 
Methods
 
Patients with mCRC who had failed at least two standard chemotherapy regimens were treated with fruquintinib at two tertiary centres in Hong Kong between December 2021 and July 2023. We reported overall survival, event-free survival (EFS), disease control rate, and toxicity. EFS was defined as the time from starting treatment to an event, which could be disease progression, discontinuation of treatment for any reason, or death.
 
Results
 
A total of 26 mCRC patients were treated with fruquintinib. The median overall survival and median EFS were 8.9 months and 4.2 months, respectively. Among the 22 patients who experienced an event, 15 (57.7%) had disease progression, six (23.1%) discontinued treatment for any reason, and one (3.8%) died. The disease control rate was 38.5%, including two (7.7%) patients with partial response and eight (30.8%) patients with stable disease. Grade ≥3 adverse reactions occurred in 69.2% of patients, the most common of which were hypertension (53.8%), hand-foot syndrome (19.2%), and diarrhoea (11.5%). There were no treatment-related deaths.
 
Conclusion
 
Fruquintinib demonstrated reasonable clinical efficacy and a manageable safety profile, consistent with the findings of international clinical studies. It is a valid option for later-line mCRC patients.
 
 
Key Words: Carcinoembryonic antigen; ErbB receptors; Hand-foot syndrome; Vascular endothelial growth factor A
 
 
中文摘要
 
使用呋喹替尼治療轉移性大腸癌的安全性及有效性的早期本地社區數據
 
蘇衍錕、劉芷珊、黃善敏、唐雯、黃嘉杰、岑翠瑜
 
引言
呋喹替尼是一種血管內皮生長因子受體(VEGFR)-1、-2及-3酪胺酸激酶選擇性抑制劑,適用於轉移性大腸癌的後期治療。本回顧性研究旨在調查於香港人口使用呋喹替尼的安全性及有效性。
 
方法
經歷最少兩次標準化療方案失敗的轉移性大腸癌患者於2021年12月至2023年7月期間在香港兩所三級醫療機構接受呋喹替尼治療。我們報告整體存活期、無事件存活期、疾病控制率及毒性數據。無事件存活期的定義為開始治療起計至有事件發生的時間,可能包括病情惡化、因任何原因導致停止治療或死亡。
 
結果
共有26名轉移性大腸癌患者接受呋喹替尼治療。整體存活期中位數及無事件存活期中位數分別為8.9個月及4.2個月。在22名有事件發生的患者當中,15名(57.7%)病情惡化,6名(23.1%)因任何原因導致停止治療,1名(3.8%)死亡。疾病控制率為38.5%,包括兩名(7.7%)部分反應患者及8名(30.8%)反應穩定患者。共有69.2%患者出現三級或以上不良反應,最常見為高血壓(53.8%)、手足症候群(19.2%)及腹瀉(11.5%)。本研究沒有與治療有關的死亡個案。
 
結論
呋喹替尼具有合理的臨床有效性及易於管理的安全狀況,與國際臨床研究結果一致,是後期轉移性大腸癌患者的有效選項。
 
 
 
INTRODUCTION
 
Globally, colorectal cancer is the third most common type of cancer and the second leading cause of cancer-related deaths.[1] In Hong Kong, colorectal cancer was not only the second most common cancer but also the second most common cause of cancer-related deaths in 2020.[2]
 
The primary treatment for metastatic colorectal cancer (mCRC) is chemotherapy, often supplemented by targeted therapy and, in certain cases, immunotherapy for patients with mismatch repair deficient tumours. For chemotherapy, standard chemotherapy regimens include 5-fluorouracil (or its oral prodrug capecitabine) plus either oxaliplatin or irinotecan, or both.[3] [4] Targeted therapy agents include bevacizumab and aflibercept, which target the vascular endothelial growth factor (VEGF) pathway; and cetuximab and panitumumab, which target the epidermal growth factor receptor (EGFR) pathway.[3] [4] Later-line treatment regimens include trifluridine-tipiracil[5] and regorafenib.[6] These two agents offer only modest improvements in overall survival (OS) and progression-free survival. However, even after failure on multiple different treatment strategies, patients may still maintain good performance status. This underscores the necessity for more safe and effective treatment options.
 
Fruquintinib is a selective inhibitor of vascular endothelial growth factor receptor (VEGFR)-1, -2, and -3 tyrosine kinases.[7] In the phase III FRESCO randomised clinical trial (Fruquintinib Efficacy and Safety in 3+ Line Colorectal Cancer Patients), fruquintinib significantly improved the median overall survival (mOS) compared with that of the placebo group (9.3 months [95% confidence interval (CI) = 8.2-10.5] vs. 6.6 months [95% CI = 5.9-8.1]) in Chinese patients with mCRC who progressed after at least two prior chemotherapy regimens (i.e., third- or later-line use).[8] It was approved in Mainland China in 2018 and was granted a fast-track designation by the US Food and Drug Administration in June 2020 for the above indication.[9] Another recent phase III FRESCO-2 randomised clinical trial also showed significant improvement in mOS with fruquintinib compared with placebo (7.4 months [95% CI = 6.7-8.2] vs. 4.8 months [95% CI = 4.0-5.8]).[10] Fruquintinib received its approval from the US Food and Drug Administration on 8 November 2023, for adult patients with mCRC who had previously received 5-fluorouracil, oxaliplatin and irinotecan-based chemotherapy, anti-VEGF therapy, and anti-EGFR therapy (if the tumour was RAS-wild type).[11]
 
The FRESCO trial recruited 416 Chinese patients from Mainland China, where fruquintinib was developed.[8] On the other hand, the FRESCO-2 trial included patients from North America, Europe, Australia and Japan, but Japanese patients comprised <10% of the trial population.[10] The FRESCO trial excluded patients who had been previously exposed to regorafenib,[8] while patients who progressed on or were intolerant to trifluridine-tipiracil or regorafenib could enter the FRESCO-2 trial.[10] Hong Kong was not a study site in either trial, and local experience in the use of fruquintinib was scarce.
 
This study aimed to analyse the safety and efficacy of fruquintinib in mCRC patients. To the best of our knowledge, this is the first retrospective study of fruquintinib in public healthcare setting in the local population.
 
METHODS
 
Data Collection and Participants
 
Clinical data from 26 patients who received fruquintinib between 31 December 2021 and 22 July 2023 were retrospectively reviewed and collected from the institutional databases of two tertiary centres in Hong Kong, namely, Princess Margaret Hospital and Tuen Mun Hospital. The inclusion criteria for fruquintinib use were modified from the FRESCO[8] and FRESCO-2[10] trials, which were as follows: (1) age ≥18 years; (2) an Eastern Cooperative Oncology Group (ECOG) performance status score of 0 to 1; (3) histologically confirmed mCRC; (4) failure (progressive disease or intolerance) on at least two standard chemotherapy regimens using fluoropyrimidine, irinotecan, oxaliplatin, anti-VEGF antibodies (bevacizumab and aflibercept), or anti-EGFR antibodies (cetuximab or panitumumab); (5) measurable disease by Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1; (6) adequate bone marrow reserve (absolute neutrophil count ≥1.5 × 109/L, platelet count ≥100 × 109/L, and haemoglobin level ≥9.0 g/dL); (7) renal function (serum creatinine level ≤1.5 × upper limit of normal [ULN] or creatinine clearance ≥60 mL/min; urine dipstick protein of ≤1+ or 24-hour urine protein level <1.0 g/24 h); and (8) liver function (serum total bilirubin level ≤1.5 × ULN; alanine aminotransferase and aspartate aminotransferase level ≤2.5 × ULN in subjects without hepatic metastases; and alanine aminotransferase and aspartate aminotransferase level ≤5 × ULN in subjects with hepatic metastases). There were no exclusion criteria involving prior use of trifluridine-tipiracil or regorafenib.
 
Study Design
 
This was a local, single-arm, retrospective analysis of patients with mCRC conducted at two tertiary centres in Hong Kong. These patients had either progressed or shown intolerance after receiving at least two lines of chemotherapy. The included patients underwent repeated 28-day treatment cycles of fruquintinib, with a schedule of 3 weeks on the medication (5 mg oral daily) followed by a 1-week break. This treatment cycle was continued until disease progression, death, occurrence of unacceptable toxicity, or discontinuation by the physician. Dose reduction was allowed to manage treatment-related adverse effects and followed the protocol of the FRESCO trial.[8]
 
Clinical Assessment Outcomes and Endpoints
 
The primary endpoint was OS, defined as the time from the start of treatment using fruquintinib to death from any cause. Tumour response assessment was performed at intervals subject to the availability of imaging and physician discretion, and response was defined by RECIST version 1.1. The secondary endpoints were event-free survival (EFS) [defined as the time from starting treatment to an event, which could be disease progression defined as the first documentation of disease progression assessed by the investigator according to RECIST version 1.1, discontinuation of treatment for any reason, or death], duration of treatment (defined as the time from starting treatment to last study treatment dose), objective response rate (defined as confirmed complete or partial response), disease control rate (defined as the sum of the complete response, partial response and stable disease rates), and carcinoembryonic antigen (CEA) response. EFS was selected instead of progression-free survival because the imaging intervals in real-world settings vary. In heavily pretreated patients, quality of life (QoL) is important, and discontinuation of treatment for any reason can also indicate the tolerability of a drug. For CEA response, a definition modified from the RECIST criteria was used to evaluate treatment response, and responses were classified into three groups, namely, CEA-RD (responsive disease), CEA-SD (stable disease), and CEA-PD (progressive disease).[12] [13] CEA-RD was defined as a decrease of >30% from the original level; CEA-PD was defined as an increase of >20% from the original level.[12] [13] A change in the CEA level that did not meet the criteria for CEA-RD and CEA-PD was defined as CEA-SD.[12] [13]
 
Adverse events (AEs) were recorded throughout the study from the start of treatment to the end of the study period or the start of the next line of treatment. They were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events version 5.0.[14]
 
Statistical Analysis
 
For OS and EFS, the Kaplan-Meier method was used to estimate the median survival time and 95% CI. Relationships between individual patient characteristics and OS or EFS were analysed using the Cox proportional hazards model to estimate hazard ratios (HRs) and 95% CIs. All analyses were performed using commercial software SPSS (Windows version 28.0; IBM Corp, Armonk [NY], US). A p value of < 0.05 was considered statistically significant.
 
RESULTS
 
The baseline demographics and disease characteristics of the patients are shown in Table 1.
 
Table 1. Baseline characteristics of the study population (n = 26).
 
Efficacy
 
Survival Outcomes and Duration of Treatment
 
The median duration of treatment was 4.3 months (range, 0.6-15.4) and the median number of treatment cycles was 3 (range, 1-17). The median follow-up time was 7.3 months. The mOS was 8.9 months (95% CI = 4.5-13.3). The Kaplan-Meier plot for OS is shown in Figure 1. The proportion of patients still alive at 6 months was 78.7% and that at 12 months was 41.2%.
 
Figure 1. Kaplan-Meier estimates for overall survival in patients with metastatic colorectal cancer receiving fruquintinib.
 
The median EFS was 4.2 months (95% CI = 2.5-5.9). Among the 22 of 26 patients who experienced an event, 15 (57.7%) had disease progression, six (23.1%) discontinued treatment for any reason, and one (3.8%) died. The Kaplan-Meier plot for EFS is shown in Figure 2.
 
Figure 2. Kaplan-Meier estimates for event-free survival in patients with metastatic colorectal cancer receiving fruquintinib.
 
Subgroup analyses of OS and EFS were carried out with a Cox proportional hazards model (simple and multivariable), but only a few clinical, tumour, or treatment factors exhibited a statistically significant correlation (Tables 2, 3 and 4). Simple analysis also revealed that OS was worse for patients with liver metastasis and with multiple sites of metastasis (Table 2). Patients with an ECOG performance status score of 0 had better EFS (HR = 0.34; 95% CI = 0.14-0.85) [Table 4 and Figure 3]. Previous use of trifluridine-tipiracil, regorafenib, or both did not significantly affect OS or EFS (Tables 2, 3 and 4).
 
Table 2. Simple analysis of risk factors associated with overall survival.
 
Table 3. Multivariable analysis of risk factors associated with overall survival.
 
Table 4. Simple analysis of risk factors associated with event-free survival.
 
Figure 3. Kaplan-Meier estimates for event-free survival in patients with metastatic colorectal cancer receiving fruquintinib (with Eastern Cooperative Oncology Group performance status score of 0 and 1).
 
The starting dose of fruquintinib was 5 mg daily (3 weeks on, 1 week off). 11 patients had their dose reduced, with 5 patients (19.2%) being reduced to 4 mg dailiy and 6 patients (23.1%) being reduced to 3 mg dailiy. Dose reduction of fruquintinib was associated with better OS (HR = 0.19, 95% CI = 0.05-0.72; p = 0.014) [Table 2 and Figure 4], but not in EFS (HR = 0.52, 95% CI = 0.22-1.21; p = 0.129) [Table 4].
 
Figure 4. Kaplan-Meier estimates for overall survival in patients with metastatic colorectal cancer receiving fruquintinib (with or without dose reduction).
 
Radiological Response
 
In patients treated with fruquintinib, the disease control rate was 38.5% (10 of 26 patients), which included two (7.7%) patients with partial response and eight patients (30.8%) with stable disease. There was no patient with complete response.
 
Carcinoembryonic Antigen Response
 
Regarding serum CEA responses, the differences in OS and EFS were not statistically significant between patients with CEA-RD, CEA-SD and CEA-PD. Numerically, patients with CEA-RD had better OS and EFS than patients with CEA-PD (Tables 2, 3 and 4).
 
Adverse Events
 
Twenty-five of 26 patients (96.2%) had at least one AE of any grade (Table 5). The most frequently reported AEs of any grade were hypertension (84.6%), proteinuria (57.7%), hand-foot syndrome (HFS) [50%], and hypothyroidism (50%). Severe AEs (grade ≥3) occurred in 18 patients (69.2%), with the most common being hypertension (53.8%), HFS (19.2%), and diarrhoea (11.5%) [Table 5]. There were no treatment-related deaths in the study population.
 
Table 5. Adverse events of the study population (n = 26).
 
Six of 26 patients (23.1%) discontinued fruquintinib due to treatment-related AEs. The most frequent AE that led to treatment discontinuation was HFS in two patients (7.7%). Treatment interruption due to AEs occurred in five patients (19.2%), and the most common AE associated with treatment interruption was hypertension in two patients (7.7%). Dose reduction due to AEs occurred in 11 patients (42.3%). The most frequent AEs leading to dose reductions were HFS (11.5%), proteinuria (11.5%), and diarrhoea (7.7%).
 
DISCUSSION
 
This retrospective study investigated the local population treated with fruquintinib at two tertiary institutions in Hong Kong, which included patients who had experienced disease progression following at least two lines of chemotherapy. This study did not have a placebo arm; the analysis of the results focuses on early experience of safety and efficacy in our locality.
 
Efficacy
 
In this study, the mOS was 8.9 months and the median EFS was 4.2 months, with a disease control rate of 38.5%. In Hong Kong, third-line or beyond monotherapy options for mCRC include trifluridine-tipiracil and regorafenib. In the RECOURSE trial (trifluridine-tipiracil vs. placebo in patients with previously treated mCRC), the treatment group had an mOS of 7.1 months and a disease control rate of 44%.[5] In the CONCUR trial (regorafenib vs. placebo in Asian patients with previously treated mCRC), the treatment group had an mOS of 8.8 months and a disease control rate of 51%.[6] Our data suggest that fruquintinib is a feasible monotherapy option in the third line and beyond setting for mCRC patients in Hong Kong.
 
OS and EFS analyses did not reveal any statistically significant differences in most of the subgroups (Tables 2, 3 and 4), although OS was shown to worsen when patients had liver metastasis or more than one site of metastasis (Table 2). EFS was shown to be related to ECOG performance status score (Table 4). There was no statistically significant correlation between CEA response and OS or EFS (Tables 2, 3 and 4), although OS tended to improve in patients with CEA-RD and tended to worsen in patients with CEA-PD.
 
In terms of the radiological and CEA (tumour marker) response, there were more patients with CEA-RD than there were with an objective response. This could be due to less intensive imaging schedules in public hospital settings, such that metabolic or relatively short-lived treatment responses could not be captured radiologically. Further study is needed to confirm the correlation between CEA response and survival, and studying the CEA response could help determine whether it can supplement the suboptimal scanning schedule in assessing the treatment response.
 
Adverse Events
 
The incidence of AEs and serious AEs was considerably high in the study population. The most frequently reported grade ≥3 AEs were hypertension, HFS, and diarrhoea (Table 5). These AEs were manageable by supportive measures and dose modification. The discontinuation of fruquintinib in this study was 26.9%, whereas the rate in FRESCO and FRESCO-2 were 15.1%[8] and 20%,[10] respectively. Further QoL analysis would be valuable to correlate the relatively high incidence of AEs and their impact on patient’s QoL.
 
Baseline hypertension was a strong risk factor for high-grade hypertensive toxicity. Among the 14 patients who experienced grade ≥3 hypertension, all had preexisting hypertension at baseline. The baseline hypertension rate (88.4%, n = 23) was relatively high when compared to that of the general population (57.4% in the 65-84 age-group).[15] The odds ratio associated with grade ≥3 hypertension was 7.00 for patients with baseline hypertension of any grade (95% CI = 0.34-144.06; p = 0.20). Of those patients with baseline hypertension, only eight (34.8%) received intervention. Home blood pressure monitoring was started or ensured in seven patients, while antihypertensive therapy was started or titrated only in one patient. Of the eight patients who underwent intervention, seven (87.5%) still developed grade ≥3 hypertension. These findings suggested that more aggressive intervention by managing oncologists is needed for patients with baseline hypertension.
 
According to the evidence from regorafenib, which is also a VEGFR inhibitor, when encountering grade 2 hypertension, treating physicians can consider starting a single antihypertensive agent (such as an angiotensin-converting enzyme inhibitor).[16] For grade 3 hypertension, an additional agent (such as a beta blocker) should be considered, and if it remains refractory, a third agent (such as a calcium channel blocker) may be added. Diuretics should be avoided because diarrhoea is also a common side-effect of fruquintinib, which may cause dehydration.
 
Another important adverse reaction was HFS. In HFS management, preventive measures include reducing skin friction, reducing exposure to heat, using skin barriers and early identification of skin abrasions.[17] The use of urea-based cream in combination with sorafenib (also a VEGFR inhibitor) has been shown to reduce the incidence of HFS.[18] Other commonly used measures include analgesics, topical anaesthetics, topical high-potency corticosteroids, keratolytic, and emollients.[17] If the above supportive measures are not able to improve tolerance, physicians can consider reducing the dose according to the drug’s prescription information.
 
A total of 99% of the patients experienced any grade of AE in the fruquintinib (FRESCO-2) trial,[10] 97% in the regorafenib (CONCUR) trial[6] and 98% in the trifluridinetipiracil (RECOURSE) trial,[5] with 63%, 54%, and 69% of patients experiencing grade ≥3 AEs, respectively. For specific grade ≥3 AEs, we compared the same class of drugs (i.e., fruquintinib vs. regorafenib), and the two drugs had similar severe AE profiles. Compared with drugs of a different class (i.e., fruquintinib vs. trifluridine-tipiracil), the toxicity profiles of these agents clearly differed, with the chemotherapy class (trifluridine-tipiracil) having more haematological toxicity, as one would expect. The most common severe AEs associated with trifluridine-tipiracil were neutropenia (38%), leukopenia (21%), and anaemia (18%).[5]
 
Patients in whom the dose was reduced had better OS and tended to improve EFS. As the dose reduction was mainly in response to toxicity, toxicity may be a predictor of the VEGFR inhibitor treatment response. A similar phenomenon was observed with anti-EGFR therapy, and a worse skin reaction was proven to be associated with a better response. In the OPUS study (untreated EGFR-expressing advanced colorectal cancer, FOLFOX4 vs. FOLFOX plus cetuximab),19 patients with grade 3 to 4 skin toxicity had a 66.7% response rate, while grade 1 patients and grade 0 patients had response rates of 42.2% and 13%, respectively. In the EPIC study (The European Prospective Investigation into Cancer and Nutrition, second-line treatment after oxaliplatin-based therapy, cetuximab plus irinotecan versus irinotecan),20 the median survival was 5.8 months for grade 0 toxicities, 11.7 months for grade 1 to 2 toxicities, and 15.6 months for grade 3 to 4 toxicities. However, further studies are needed to confirm this hypothesis in fruquintinib therapy.
 
Limitations
 
This study has several limitations. First, this was a retrospective study based only on data from two public tertiary centres in Hong Kong. Second, the sample size was small, and some subgroup analyses did not show statistical significance. Third, QoL data were not collected in this study.
 
CONCLUSION
 
Fruquintinib demonstrated reasonable clinical efficacy and a manageable safety profile and is a valid option for later-line mCRC patients. Hypertension is the most common high-grade toxicity, and preexisting hypertension is a strong risk factor. Proactive management of hypertension is strongly advocated. Prompt AE management can optimise its clinical utility, and dose reduction did not compromise efficacy. Further study of treatment sequence and patient QoL among the approved third-line or beyond options is needed.
 
REFERENCES
 
1. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71:209-49. Crossref
 
2. Hong Kong Cancer Registry, Hospital Authority. Top ten cancers. 2021. Available from: https://www3.ha.org.hk/cancereg/topten.html. Accessed 14 May 2024.
 
3. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology: Colon Cancer, version 3. 2023. Available from: https://www.nccn.org/professionals/physician_gls/pdf/colon.pdf. Accessed 23 Sep 2023.
 
4. Cervantes A, Adam R, Roselló S, Arnold D, Normanno N, Taïeb J, et al. Metastatic colorectal cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol. 2023;34:10-32. Crossref
 
5. Mayer RJ, Van Cutsem E, Falcone A, Yoshino T, Garcia-Carbonero R, Mizunuma N, et al. Randomized trial of TAS-102 for refractory metastatic colorectal cancer. N Engl J Med. 2015;372:1909-19. Crossref
 
6. Li J, Qin S, Xu R, Yau TC, Ma B, Pan H, et al. Regorafenib plus best supportive care versus placebo plus best supportive care in Asian patients with previously treated metastatic colorectal cancer (CONCUR): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol. 2015;16:619-29. Crossref
 
7. Cao J, Zhang J, Peng W, Chen Z, Fan S, Su W, et al. A phase I study of safety and pharmacokinetics of fruquintinib, a novel selective inhibitor of vascular endothelial growth factor receptor-1, -2, and-3 tyrosine kinases in Chinese patients with advanced solid tumors. Cancer Chemother Pharmacol. 2016;78:259-69. Crossref
 
8. Li J, Qin S, Xu RH, Shen L, Xu J, Bai Y, et al. Effect of fruquintinib vs placebo on overall survival in patients with previously treated metastatic colorectal cancer: the FRESCO randomized clinical trial. JAMA. 2018;319:2486-96. Crossref
 
9. HUTCHMED. Chi-Med announces fruquintinib granted U.S. FDA fast track designation for metastatic colorectal cancer [Press Release]. 2020 June 18. Available from: https://www.hutch-med.com/fruquintinib-granted-us-fda-fast-track-designation-for-mcrc/. Accessed 2 Sep 2024.
 
10. Dasari A, Lonardi S, Garcia-Carbonero R, Elez E, Yoshino T, Sobrero A, et al. Fruquintinib versus placebo in patients with refractory metastatic colorectal cancer (FRESCO-2): an international, multicentre, randomised, double-blind, phase 3 study. Lancet. 2023;402:41-53. Crossref
 
11. United States Food and Drug Administration. FDA approves fruquintinib in refractory metastatic colorectal cancer. 2023 November 8. Available from: https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-fruquintinib-refractory-metastatic-colorectal-cancer. Accessed 2 Sep 2024.
 
12. Su YT, Chen JW, Chang SC, Jiang JK, Huang SC. The clinical experience of the prognosis in opposite CEA and image change after therapy in stage IV colorectal cancer. Sci Rep. 2022;12:20075. Crossref
 
13. Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer. 2009;45:228-47. Crossref
 
14. United States Department of Health and Human Services. Common Terminology Criteria for Adverse Events (CTCAE) Version 5.0. 2017 November 27. Available from: https://ctep.cancer.gov/protocoldevelopment/electronic_applications/docs/CTCAE_v5_Quick_Reference_5x7.pdf. Accessed 2 Sep 2024.
 
15. Centre for Health Protection, Department of Health, Hong Kong SAR Government. Hypertension. May 2023. Available from: https://www.chp.gov.hk/en/healthtopics/content/25/35390.html. Accessed 20 Jan 2024.
 
16. Krishnamoorthy SK, Relias V, Sebastian S, Jayaraman V, Saif MW. Management of regorafenib-related toxicities: a review. Therap Adv Gastroenterol. 2015;8:285-97. Crossref
 
17. Kwakman JJ, Elshot YS, Punt CJ, Koopman M. Management of cytotoxic chemotherapy-induced hand-foot syndrome. Oncol Rev. 2020;14:442. Crossref
 
18. Ren Z, Zhu K, Kang H, Lu M, Qu Z, Lu L, et al. Randomized controlled trial of the prophylactic effect of urea-based cream on sorafenib-associated hand-foot skin reactions in patients with advanced hepatocellular carcinoma. J Clin Oncol. 2015;33:894-900. Crossref
 
19. Bokemeyer C, Bondarenko I, Makhson A, et al. Cetuximab plus 5-FU/FA/oxaliplatin (FOLFOX-4) versus FOLFOX-4 in the first-line treatment of metastatic colorectal cancer (mCRC): OPUS, a randomized phase II study. J Clin Oncol. 2007;25(Suppl 18):4035. Crossref
 
20. Jonker DJ, Karapetis CS, Moore M, Zalcberg JR, Tu D, Berry S, et al. editors. Randomized phase III trial of cetuximab monotherapy plus best supportive care (BSC) versus BSC alone in patients with pretreated metastatic epidermal growth factor receptor (EGFR)–positive colorectal carcinoma: a trial of the National Cancer Institute of Canada Clinical Trials Group (NCIC CTG) and the Australasian Gastro-Intestinal Trials Group (AGITG). In: Proc Am Assoc Cancer Res; 2007.
 
 
 

Efficacy and Safety of Preoperative Embolisation of Bone Tumours: A Tertiary Centre Experience

FFY Wan, TWY Chin, KC Lai, MK Chan

ORIGINAL ARTICLE
 
Efficacy and Safety of Preoperative Embolisation of Bone Tumours: A Tertiary Centre Experience
 
FFY Wan, TWY Chin, KC Lai, MK Chan
Department of Diagnostic and Interventional Radiology, Queen Elizabeth Hospital, Hong Kong SAR, China
 
Correspondence: Dr FFY Wan, Department of Diagnostic and Interventional Radiology, Queen Elizabeth Hospital, Hong Kong SAR, China. Email: wfy471@ha.org.hk
 
Submitted: 22 May 2023; Accepted: 19 October 2023.
 
Contributors: All authors designed the study, acquired and analysed the data. FFYW drafted the manuscript. All authors critically revised the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.
 
Conflicts of Interest: All authors have disclosed no conflicts of interest.
 
Funding/Support: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
 
Data Availability: All data generated or analysed during the present study are available from the corresponding author on reasonable request.
 
Ethics Approval: This research was approved by the Kowloon Central / Kowloon East Cluster Research Ethics Committee of Hospital Authority, Hong Kong (Ref No.: KC/KE-23-0038/ER-4). The requirement for informed patient consent was waived by the Committee due to the retrospective nature of the research.
 
 
 
 
 
Abstract
 
Introduction
 
Preoperative embolisation of bone tumours minimises risk of major intraoperative haemorrhage. Technical success is defined as obliteration of tumour vascularity by ≥70% on post-embolisation angiography. We retrospectively reviewed the technical success, efficacy, and safety of preoperative embolisation of bone tumours in our centre.
 
Methods
 
Nineteen patients underwent preoperative embolisation of bone tumours from December 2010 to July 2022. Subsequent surgery was performed 1 day post-embolisation. Patient demographics, tumour histology and location, presence of pathological fracture or spinal cord compression, primary embolic agent used, technical success, intraprocedural blood loss, need for blood transfusion, and major complications related to embolisation or subsequent surgery were assessed.
 
Results
 
Most of the bone tumours were metastases (n = 14) with the majority being hypervascular metastases from renal cell carcinoma or thyroid cancer. The primary bone tumours (n = 5) included vertebral haemangioma (n = 2), plasmacytoma (n = 2), and chordoma (n = 1). Pathological fractures were present in 11 patients. Among the 11 tumours in the spine, eight of them were complicated by spinal cord compression before embolisation. Particles were used as the main embolisation agent in all cases, with 89% technical success. There were no major embolisation-related complications. In patients after successful embolisation, the estimated intraprocedural blood loss ranged from 20 to 3,000 mL.
 
Conclusion
 
Preoperative embolisation of bone tumours is safe and feasible with high technical success.
 
 
Key Words: Bone neoplasms; Hemangioma; Radiology, interventional; Spinal cord compression
 
 
中文摘要
 
骨腫瘤術前栓塞的有效性和安全性:一個三級醫療中心的經驗
 
尹芳盈、錢永恩、黎國忠、陳文光
 
引言
骨腫瘤術前栓塞盡量減少了術中大出血的風險。技術成功的定義是栓塞後血管攝影中腫瘤血管消失≥70%。我們對本中心骨腫瘤術前栓塞的技術成功率、有效性和安全性進行回顧性分析。
 
方法
2010年12月至2022年7月期間,19例患者接受了術前骨腫瘤栓塞治療,他們於栓塞後1天進行手術。我們分析了患者基本數據、腫瘤組織學和位置、是否存在病理性骨折或脊髓壓迫、使用的主要栓塞劑、技術成功率、術中失血、輸血需求以及與栓塞或後續手術相關的主要併發症。
 
結果
大多數骨腫瘤是轉移瘤(n = 14),其中大多數是腎細胞癌或甲狀腺癌的富血管轉移瘤。原發性骨腫瘤(n = 5)包括椎體血管瘤(n = 2)、漿細胞瘤(n = 2)和脊索瘤(n = 1)。11名患者存在病理性骨折。11個脊椎腫瘤中,有8個在栓塞前併發脊髓受壓。所有病例均使用以顆粒為主的栓塞劑,技術成功率為89%。沒有嚴重的栓塞相關併發症個案。栓塞成功的患者的預計術中失血量為20至3,000 mL。
 
結論
骨腫瘤術前栓塞是安全且可行的,技術成功率亦高。
 
 
 
INTRODUCTION
 
The management of bone tumours is complex and requires a multidisciplinary approach. In general, the best line of treatment is surgical resection. Bone tumours with impending or completed pathological fractures require early surgical intervention to prevent or stabilise the fractures. Nevertheless, surgery may be technically difficult due to large or hypervascular tumours, difficult anatomical locations, or close proximity to adjacent vital structures such as the spine. In these scenarios, arterial embolisation plays a pivotal role as a preoperative methodology to achieve devascularisation of the tumour, thus minimising intraoperative bleeding and complications. In this study, we aimed to evaluate the technical success, efficacy, and safety of preoperative embolisation of bone tumours in our tertiary musculoskeletal tumour centre.
 
METHODS
 
This retrospective study evaluated 19 patients who underwent preoperative embolisation of bone tumours followed by surgery at our centre from December 2010 to July 2022. Patient demographics, tumour histology and location, presence of pathological fractures or spinal cord compression, and choice of primary embolic agent were recorded. The technical success of embolisation, defined as reduction of tumour arterial blush by ≥70% on postoperative angiography,[1] [2] as shown in our case (Figure 1), was assessed. In cases of no definite tumoural staining identified on preoperative angiography, which was seen in one of our patients, technical success could not be reliably evaluated. Clinical notes as well as operative and anaesthetic records were reviewed for major surgical complications, intraoperative blood loss, and requirements for transfusion. Categorical data are presented as percentages, while numerical data are presented as medians with ranges.
 
Figure 1. (a) Bone metastasis from renal cell carcinoma in the left proximal tibia manifests as a large expansile lytic bone lesion (arrow) on radiograph. (b) It shows significant hypervascularity (arrow) on angiogram with supply from multiple genicular arteries and the anterior recurrent tibial artery. (c) Post-embolisation angiogram confirms technical success of the procedure with minimal residual tumoral vascularity (arrow).
 
Techniques
 
Case selection for preoperative embolisation requires a multidisciplinary team discussion. Factors to consider include tumour histology, location, size and vascularity, and the risk of significant intraprocedural haemorrhage.[2] Tumour histology is confirmed by image-guided core needle biopsy. The location, size, and vascularity of the tumour are assessed on imaging. Preprocedural review of imaging, in particular computed tomography angiography, is important for identifying blood supply and drainage, tumour extension into adjacent structures, and proximity to vital structures potentially sharing the arterial supply. Before the embolisation procedure, the results of laboratory tests including clotting profile, platelet count, haemoglobin level, and creatinine values are reviewed. Abnormal coagulation should be corrected since many of the embolic agents require a functioning intrinsic clotting mechanism.
 
All patients in our centre had surgery performed 1 day following the embolisation. In view of potential revascularisation with a prolonged interval between embolisation and surgery, the timing of embolisation should be as close as possible to that of the operation, ideally within 3 days after embolisation.[2] The procedure was performed by radiologists with 8 to 26 years of experience in vascular interventional radiology. The embolisation procedure was done under local anaesthesia in the angiography suite. Vascular access was obtained via femoral arterial puncture. A 5-Fr or 6-Fr vascular sheath and a 4-Fr or 5-Fr pre-shaped catheter were used. A pre-embolisation angiogram was obtained to assess the degree of tumour vascularity, identify the major supplying arteries, and confirm the safety of embolisation. For instance, careful attention must be paid to ensure there is no opacification of a spinal pial artery such as the artery of Adamkiewicz. If embolisation was not contraindicated for any of these reasons, a microcatheter was introduced coaxially through the catheter to achieve superselective catheterisation of tumour feeding arteries and reduce the chance of non-target embolisation. Micron-sized solid embolic particles were primarily used in all cases. They lodged in the tumour vessels proximal to or at capillary level, thus occluding vessels within the tumour to achieve distal tumour microvasculature penetration. The particles were suspended in non-ionic contrast medium to enable visualisation during the angiographic procedure. The choice of particle diameter was determined by vessel size and desired distal embolisation. Injection of embolic agents must be performed under fluoroscopic guidance to guard against reflux into non-target vessels. All embolisation procedures were performed under continuous fluoroscopic guidance. Multiple angiograms were acquired to evaluate the degree of vessel occlusion. The endpoint of the procedure was reached when all the major tumour-supplying vessels were occluded with near-complete obliteration of tumour blush. Finally, a post-embolisation angiogram was performed to assess the technical success of embolisation, which was defined as catheterisation of the major tumour feeding arteries with reduction of the tumour blood supply by ≥70%.[1] [2]
 
RESULTS
 
There were 10 female and nine male patients who underwent preoperative embolisation during the study period (Table). Patient age ranged between 22 and 77 years and the median age was 61 years. The majority of the tumours were bone metastases (n = 14, 74%) and most of them were either metastases from renal cell carcinoma (n = 6, 32%) or thyroid carcinoma (n = 5, 26%). The rest of the bone tumours (n = 5, 26%) included vertebral haemangioma (n = 2, 11%), plasmacytoma (n = 2, 11%), and chordoma (n = 1, 5%). More than half (n = 11, 58%) of the tumours were located within multiple vertebrae. The rest were located in the extremities (n = 6, 32%) or the pelvis (n = 2, 11%). Pathological fractures were present in 58% of the patients (n = 11). Among the 11 vertebral tumours, cord compression was seen in eight (73%) of them.
 
Table. Demographics, clinical, and pathological characteristics of patients (n = 19).
 
Technical success was achieved in 16 out of 18 (89%) patients and selected case examples are shown in Figures 2, 3, and 4. Only partial embolisation could be performed in two patients due to the proximity of the tumour feeding arteries to the spinal artery in one patient and the occurrence of chest pain during the procedure in another patient. Technical success could not be reliably evaluated in one patient since no definite tumour staining was evident on pre-embolisation angiogram for comparison. The primary embolic agents used included trisacryl gelatin microspheres (Embosphere; Merit Medical, Warrington [PA], US) [n = 8], polyvinyl alcohol (PVA) particles (Contour; Boston Scientific, Marlborough [MA], US) [n = 7], and PVA hydrogel microspheres (Bead Block; Terumo Medical, Tokyo, Japan) [n = 4].
 
Figure 2. (a) Preoperative embolisation performed for bone metastasis from renal cell carcinoma at the right femoral intertrochanteric region (arrow) with pathological fracture. (b) Angiogram of the right femoral artery shows the hypervascular tumour (arrow) supplied by branches of lateral and medial circumflex femoral arteries. (c) Post-embolisation angiogram demonstrates >90% reduction in tumour vascularity (arrow), suggestive of technical success. (d) The patient underwent bipolar hip arthroplasty on the subsequent day with intraoperative blood loss of around 100 mL.
 
Figure 3. Histologically proven spinal metastasis from solitary fibrous tumour undergoing preoperative embolisation. Computed tomography shows the L5 spinal tumour (arrows) with intraspinal (a) and paraspinal (b) extension. Pre-embolisation angiograms confirm the hypervascular tumour (arrows) to be supplied by branches of the left fourth lumbar artery (c) and iliac branch of the left iliolumbar artery (d). Superselective pre-embolisation angiograms by microcatheters advanced into the branches of the left fourth lumbar artery (e) and the branches of the iliac branch of the left iliolumbar artery (f) reveal significant tumour blush (arrows). Post-embolisation angiograms of branches of the left fourth lumbar artery (g) and the iliac branch of the left iliolumbar artery (h) demonstrate absent tumour blush, suggestive of technical success.
 
Figure 4. (a) Pathological fracture through a renal cell metastasis in the distal left humerus (arrow) is seen on the radiograph. (b) Brachial angiogram shows the hypervascular tumour (arrow) with arterial feeders from the brachial artery and radial recurrent artery. (c) Completion angiogram demonstrates successful devascularisation. (d) The patient underwent partial resection of the humerus and total elbow replacement with minimal blood loss.
 
There was no mortality related to embolisation. Minor complications in the form of post-embolisation syndrome and pain from ischaemic necrosis of tumours occurred in six patients (32%) and these were treated with analgesics and fluid. In patients with embolisation of vertebral tumours, there were no procedure-related neurological deficits. The median of intraprocedural blood loss was 700 mL (range, 20-14,000). Two patients (11%) suffered major haemorrhages requiring massive intraprocedural blood transfusions. One of them had a spinal metastasis from renal cell carcinoma with supply from the bilateral T6 segmental arteries. However, successful embolisation was only achieved at the right T6 segmental artery because the spinal artery was seen in repeated angiograms of the left T6 segmental artery (Figure 5). To minimise the risk of spinal cord infarction, the procedure was abandoned after only light embolisation of the left T6 segmental artery and the target of technical success could not be achieved. The patient had significant intraprocedural blood loss requiring massive transfusion and the transfused blood volume was around 3 L. Postoperatively, there was diplegia of the lower limbs, suggestive of spinal cord injury. Another patient had sacral chordoma with no definite tumour staining on preoperative angiography. As a result, technical success of the embolisation procedure could not be reliably evaluated. Embolisation was performed pre-emptively in view of the possibility of massive intraoperative bleeding. Unfortunately, major intraprocedural haemorrhage was still encountered, necessitating massive transfusion with transfused blood volume of around 4 L.
 
Figure 5. (a) Sagittal view of computed tomography of the thoracic spine reveals a T6 bone metastasis from renal cell carcinoma with extension into the spinal canal (arrow). (b) Superselective catheterisation of the right T6 segmental artery shows significant tumour staining (arrow) and non-visualisation of the spinal artery. (c) Post-embolisation angiogram of the right T6 segmental artery shows absent tumour blush. (d) The spinal artery (arrow) was visualised on repeated angiogram of the left T6 segmental artery after initial light embolisation, therefore no further embolisation was performed.
 
DISCUSSION
 
Successful embolisation of bone tumours may potentially decrease intraoperative blood loss and improve visualisation of the surgical field, thus minimising risks of major complications and enabling safer and more complete resection. It is particularly beneficial when there is a high risk of intraoperative bleeding, spinal involvement with cord or neural encroachment or in technically difficult locations with expected prolonged surgery,[3] such as hypervascular spinal and pelvic bone metastases. In our case series, the median estimated intraprocedural blood loss was 700 mL, which was lower than that reported in other studies.[4] [5]
 
Apart from its role as an adjuvant therapy to surgery, embolisation may also be performed as a palliative treatment for symptomatic relief of bone metastases. It may be done as a standalone treatment or combined with ablation or cementoplasty.[6] It has been used successfully to achieve neurological improvement in patients with hypervascular vertebral metastases causing acute spinal cord compression[7] and symptomatic relief in patients with painful bone metastases from renal cell carcinoma.[8] There have been studies supporting embolisation as a primary treatment for benign bone tumours such as aneurysmal bone cysts and giant cell tumours.[9] [10] [11] It is particularly beneficial in tumours located in the spine or pelvis, where surgery and radiation are associated with high rates of morbidity and recurrence. Serial embolisation of these tumours is usually performed until there is symptomatic relief or near complete resolution of tumour vascularity.[10] Radiological response can also be assessed and it manifests as reduction in tumour vascularity and increase in ossification. In patients with vertebral haemangiomas complicated with spinal cord compression or spinal pain, surgery or radiotherapy has been the traditional treatment of choice. However, surgery alone is associated with risk of significant bleeding from these highly vascular tumours. Preoperative embolisation has been shown to be a useful adjunctive therapy to minimise bleeding risk.[12] [13]
 
Particulate materials, namely PVA particles and microspheres, are primarily used for embolisation. PVA is water-soluble synthetic polymer made from polyvinyl acetate through partial or full hydrolysis to remove the acetate groups, with size ranging from 50 to 1000 μm. It has the ability to penetrate and occlude the tumour blood supply. It is compressible after drying and will expand to up to 15 times its compressed size after rehydration.[14] Most interventional radiologists have extensive experience in using it and it is relatively easy to deliver. It is safe without any long-term side-effects. The conventional preparation (Contour PVA) has irregular outlines and therefore occludes vessels larger than its diameter due to aggregation of particles. Some newer preparations, e.g., Bead Block PVA hydrogel microspheres, are engineered PVA particles with relatively uniform size. Their microporous nature also enables them to be compressible and facilitates delivery through small catheters. Embosphere microspheres are trisacryl gelatin microspheres with size ranging from 40 to 1200 μm. Their compressibility allows smooth passage through microcatheter with a diameter smaller than its size. They are more uniform in size than PVA and their sizes do not change in liquids. They also have less tendency to clump after injection. The choice of the primary particulate embolic agent is mainly determined by the operator’s experience and preference. There is currently little published literature comparing the efficacy of different embolic materials in preoperative embolisation of bone tumours. A study performed to assess the intraprocedural blood loss post-embolisation showed no clinically significant difference between trisacryl gelatin microspheres and PVA particles.[15] Liquid embolic agents may induce more tumour necrosis than particles and be beneficial when definitive treatment is aimed. Nonetheless, they are technically more difficult to handle and their use requires an experienced operator. They are also associated with a higher risk of non-target embolisation and non-target necrosis compared with particles.[6] As a general rule, if embolisation is performed as preoperative or palliative treatment, liquid agents have little advantage over particulate agents.
 
Complications of embolisation of bone tumours include arterial dissection, pain due to ischaemic necrosis of tumour, non-target embolisation, infection, haemorrhage, and post-embolisation syndrome.[8] [16] Post-embolisation syndrome is a common but usually self-limiting side-effect. Patients present with symptoms such as pain, fever, and malaise, which could be treated with analgesics and fluid. Non-target embolisation is another potential complication. Aside from the use of microcatheters to reduce its risk, coils may be employed to embolise and protect the non-target vessels more proximally, which could not be navigated beyond to get close to the tumour feeding vessels.[7]
 
Limitations
 
There were limitations in our study. First, it was retrospective in nature and a non-embolisation group was not available for comparison. With reference to other studies from the literature, it still offered a reasonable view of preoperative embolisation as a potentially helpful procedure in the management of bone tumours. Another limitation was the heterogeneous study population with different tumour pathologies and surgeries performed, but this reflected the diversity of primary and metastatic bone tumours that could be considered for preoperative embolisation. Ideally, a prospective randomised controlled trial with a larger study population would be optimal for determining the exact value of preoperative embolisation compared with non-embolisation. Other factors that could affect intraprocedural blood loss, including patient factors, and the surgery performed, should also be taken into account.
 
CONCLUSION
 
Preoperative embolisation is safe, technically feasible, and potentially useful in the treatment of bone tumours, although a high risk of intraoperative bleeding should be taken into consideration.
 
REFERENCES
 
1. Kwon JH, Shin JH, Kim JH, Gwon DI, Yoon HK, Ko GY, et al. Preoperative transcatheter arterial embolization of hypervascular metastatic tumors of long bones. Acta Radiol. 2010;51:396-401. Crossref
 
2. Geraets SE, Bos PK, van der Stok J. Preoperative embolization in surgical treatment of long bone metastasis: a systematic literature review. EFORT Open Rev. 2020;5:17-25. Crossref
 
3. Barton PP, Waneck RE, Karnel FJ, Ritschl P, Kramer J, Lechner GL. Embolization of bone metastases. J Vasc Interv Radiol. 1996;7:81-8. Crossref
 
4. Owen RJ. Embolization of musculoskeletal bone tumors. Semin Intervent Radiol. 2010;27:111-23. Crossref
 
5. Manke C, Bretschneider T, Lenhart M, Strotzer M, Neumann C, Gmeinwieser J, et al. Spinal metastases from renal cell carcinoma: effect of preoperative particle embolization on intraoperative blood loss. AJNR Am J Neuroradiol. 2001;22:997-1003.
 
6. Kickuth R, Waldherr C, Hoppe H, Bonel HM, Ludwig K, Beck M, et al. Interventional management of hypervascular osseous metastasis: role of embolotherapy before orthopedic tumor resection and bone stabilization. AJR Am J Roentgenol. 2008;191:W240-7. Crossref
 
7. Gottfried ON, Schmidt MH, Stevens EA. Embolization of sacral tumors. Neurosurg Focus. 2003;15:E4. Crossref
 
8. Munk PL, Legiehn GM. Musculoskeletal interventional radiology: applications to oncology. Semin Roentgenol. 2007;42:164-74. Crossref
 
9. Smit JW, Vielvoye GJ, Goslings BM. Embolization for vertebral metastases of follicular thyroid carcinoma. J Clin Endocrinol Metab. 2000;85:989-94. Crossref
 
10. Forauer AR, Kent E, Cwikiel W, Esper P, Redman B. Selective palliative transcatheter embolization of bony metastases from renal cell carcinoma. Acta Oncol. 2007;46:1012-8. Crossref
 
11. Lin PP, Guzel VB, Moura MF, Wallace S, Benjamin RS, Weber KL, et al. Long-term follow-up of patients with giant cell tumor of the sacrum treated with selective arterial embolization. Cancer. 2002;95:1317-25. Crossref
 
12. Luther N, Bilsky MH, Härtl R. Giant cell tumor of the spine. Neurosurg Clin N Am. 2008;19:49-55. Crossref
 
13. Rossi G, Rimondi E, Bartalena T, Gerardi A, Alberghini M, Staals EL, et al. Selective arterial embolization of 36 aneurysmal bone cysts of the skeleton with N-2-butyl cyanoacrylate. Skeletal Radiol. 2010;39:161-7. Crossref
 
14. Bandiera S, Gasbarrini A, De Iure F, Cappuccio M, Picci P, Boriani S. Symptomatic vertebral hemangioma: the treatment of 23 cases and a review of the literature [in English, Italian]. Chir Organi Mov. 2002;87:1-15.
 
15. Acosta FL Jr, Dowd CF, Chin C, Tihan T, Ames CP, Weinstein PR. Current treatment strategies and outcomes in the management of symptomatic vertebral hemangiomas. Neurosurgery. 2006;58:287-95. Crossref
 
16. Sheth RA, Sabir S, Krishnamurthy S, Avery RK, Zhang YS, Khademhosseini A, et al. Endovascular embolization by transcatheter delivery of particles: past, present, and future. J Funct Biomater. 2017;8:12. Crossref
 
 
 

Stereotactic-Guided Magnetic Seed Localisation Versus Radioguided Occult Lesion Localisation: A Comparison of Total Resection Volumes

RYS Mak, AHC Wong, CKM Mo, KH Chin, WWC Wong, PL Chau, YH Ling, LWY Ma, JSY Lee, JYW Chan, CY Choi, AYT Lai

ORIGINAL ARTICLE
 
Stereotactic-Guided Magnetic Seed Localisation Versus Radioguided Occult Lesion Localisation: A Comparison of Total Resection Volumes
 
RYS Mak1, AHC Wong1, CKM Mo1, KH Chin1, WWC Wong1, PL Chau2, YH Ling2, LWY Ma3, JSY Lee3, JYW Chan3, CY Choi3, AYT Lai1
1 Department of Radiology, Pamela Youde Nethersole Eastern Hospital, Hong Kong SAR, China
2 Department of Surgery, Ruttonjee Hospital, Hong Kong SAR, China
3 Department of Surgery, Pamela Youde Nethersole Eastern Hospital, Hong Kong SAR, China
 
Correspondence: Dr RYS Mak, Department of Radiology, Pamela Youde Nethersole Eastern Hospital, Hong Kong SAR, China. Email: mys877@ha.org.hk
 
Submitted: 4 April 2023; Accepted: 5 October 2023.
 
Contributors: RYSM and AYTL designed the study. RYSM, AHCW, CKMM, PLC, YHL, LWYM, JSYL, JYWC and AYTL acquired the data. RYSM, KHC, WWCW, YHL, LWYM and AYTL analysed the data. RYSM, AHCW and PLC drafted the manuscript. CKMM, KHC, WWCW, PLC, YHL, LWYM, JSYL, JYWC, CYC and AYTL critically revised the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.
 
Conflicts of Interest: All authors have disclosed no conflicts of interest.
 
Funding/Support: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
 
Data Availability: All data generated or analysed during the present study are available from the corresponding author on reasonable request.
 
Ethics Approval: This research was approved by the Hong Kong East Cluster Research Ethics Committee of Hospital Authority, Hong Kong (Ref No.: HKECREC-2022-059). The requirement for patient consent was waived by the Committee due to the retrospective nature of the research.
 
Declaration: This paper was presented orally in the 30th Annual Scientific Meeting of Hong Kong College of Radiologists (12-13 November 2022; virtual meeting).
 
 
 
 
Abstract
 
Introduction
 
Cosmetic outcome after breast conservation surgery has a major impact on patients’ quality of life. Previous research demonstrated the use of non-radioactive magnetic markers (Magseed) to be safe and effective. There have been few studies comparing magnetic seeds and radioguided occult lesion localisation (ROLL). This study compares the total resection volume in lumpectomies of mammographically detected non-palpable lesions using magnetic seeds with the volume resulting from ROLL guidance.
 
Methods
 
This was a retrospective cohort study comparing lumpectomy cases guided by one or the other technique. Total resection volume was calculated based on pathology reports. Margin clearance and reoperation rates were analysed.
 
Results
 
Each cohort included 11 patients with similar baseline characteristics and comparable histopathology from the vacuum-assisted biopsy specimens. The technical success rates of magnetic seed deployment and ROLL injection were both 100%. The total resection volume in the magnetic seed cohort was significantly lower than that in the ROLL cohort. If the cases with involved or close margins were excluded from analysis of total resection volume, the magnetic seeds group still achieved a significantly lower total resection volume. No significant difference was found in the final histopathological diagnosis, margin clearance, or reoperation rates between the two groups.
 
Conclusion
 
Magnetic seed localisation is a safe and effective technique that can reduce total resection volume compared with ROLL, without compromising margin clearance and reoperation rates.
 
 
Key Words: Breast; Carcinoma; Psychosocial functioning
 
 
中文摘要
 
立體定位磁粒子定位與無線電導引隱匿性病灶定位:總切除體積的比較
 
麥恩善、黃可澄、巫冠文、錢凱、黃慧中、周珮鈴、凌若熙、馬慧欣、李雪盈、陳盈穎、蔡自怡、黎爾德
 
引言
乳房保留手術後的美容效果對患者的生活品質有重大影響。先前的研究證明使用非放射性磁性標記(Magseed)是安全有效的。比較磁性粒子和無線電引導隱匿性病灶定位(ROLL)的研究很少。本研究比較了使用磁粒子對乳房X光檢查檢測到的不可觸及病變進行腫塊切除術的總切除體積與ROLL引導結果的體積。
 
方法
這是一項回顧性隊列研究,比較由這兩種技術指導的腫瘤切除術病例。我們根據病理報告計算總切除體積,並分析切緣清除率和再手術率。
 
結果
每個隊列包括11名具有相似基線特徵和真空輔助活檢標本組織病理學相似的患者。磁粒子部署和ROLL注射的技術成功率均為100%。磁粒子組的總切除體積顯著低於ROLL組。如果將涉及或接近切緣的病例排除在總切除體積分析之外,磁粒子組的總切除體積仍顯著較低。兩組之間的最終組織病理學診斷、切緣清除或再手術率沒有顯著差異。
 
結論
磁粒子定位是一種安全有效的技術,與ROLL相比,可以減少總切除體積,且不影響切緣清除率和再手術率。
 
 
 
INTRODUCTION
 
The use of non-radioactive magnetic seed markers (Magseed; Endomagnetics, Cambridge, United Kingdom) is a relatively new technique for localisation of non-palpable breast lesions requiring surgical resection, including early breast cancers and high-risk lesions, which are being increasingly detected due to advancements in breast imaging techniques and more widespread breast cancer screening. Several studies have already demonstrated magnetic seeds to be a safe, effective method that is non-inferior to wire-guided localisation.[1] [2] [3] A recent retrospective cohort study showed that localisation with magnetic seeds resulted in reduced resection volumes without an increased margin positivity rate compared with wire-guided localisation. Minimising resection volumes is important for optimal cosmetic outcome.[4]
 
In contrast, there have been fewer studies comparing magnetic seeds and radioguided occult lesion localisation (ROLL). Initial experience in a regional hospital in Hong Kong showed comparable operation times, surgical specimen sizes, margin clearances and reoperation rates compared with ROLL, with magnetic seeds having the added advantage of being non-radioactive and allowing decoupling of radiological and surgical schedules.[5] Due to similar experience, magnetic seed localisation has become the preferred technique since its adoption in our unit.
 
It is known that the cosmetic outcome after breast conservation surgery (BCS) can affect the psychosocial functioning of patients.[6] Women with pronounced breast asymmetry are more likely to feel stigmatised, experience depressive symptoms, and have a worse quality of life.[6] The major determining factor for cosmetic outcome is resection volume.[7] Ideally, the resection volume should be as small as possible without jeopardising the margin status. This is in turn related to surgical accuracy that relies heavily on the localisation technique for non-palpable breast lesions.[7] We aim to compare the total resection volume along with other outcome measures, including margin status and reoperation rate, in lumpectomies of mammographically detected non-palpable lesions under magnetic seed and ROLL guidance performed in two regional hospitals in Hong Kong.
 
METHODS
 
This retrospective cohort study first identified all lumpectomy cases aiming at vacuum-assisted biopsy (VAB) markers guided by stereotactic-guided magnetic seed localisation. To control for the targeted amount of tissue to be excised, only the lumpectomy cases aiming at radiopaque VAB markers that were placed after VAB were included. These were malignant or high-risk lesions identified from the VAB specimens, which required further surgical excision. These excisions were guided by magnetic seed placement. The cases were then matched with a control group of older consecutive lumpectomy cases aiming at VAB markers stereotactically guided by ROLL, beginning immediately before the adoption of magnetic seeds for such cases, until the same number was reached. Theoretically, as the lesions were non-palpable and sonographically occult, and the VAB markers were the common mammographically localised targets, the expected total resection volume would be comparable between the two cohorts. It was not feasible to draw both cohorts from the same time period as ROLL was rarely utilised after the introduction of magnetic seeds.
 
Electronic medical records were reviewed. Baseline characteristics including age, laterality of lesion, initial mammographic abnormality, pathology of the VAB specimen, and the time interval between localisation and operation were recorded. The preoperative mammographically detected post-VAB residual lesion sizes including the VAB marker (3 mm) were measured. The total span of the preoperative mammographically detected post-VAB residual lesion (with inclusion of the VAB marker) and the localisation agent (magnetic seed or iodinated contrast injected during ROLL) was likewise measured.
 
Surgical specimen volume was calculated using the ellipse volume formula Volume = 4/3 × π × A × B × C, where A, B, and C are the lengths of all three semi-axes as documented in the pathology reports. If additional margins were excised intraoperatively, their volumes were likewise calculated. In cases where intraoperative additional margins measurements were not fully documented, they were assumed to be of negligible volume. Specimens of other breast lesions, contralateral breast surgery, and sentinel lymph node biopsy were considered irrelevant in the calculation of specimen volume in this study. The total resection volume was yielded by the sum of all relevant specimens.
 
The technical success rate of localisation for magnetic seeds and ROLL were recorded. The technical success of magnetic seed localisation was defined as deployment of the seeds without significant migration (>1 cm), and subsequent excision of the seeds and the VAB marker. In the ROLL cohort, additional iodinated contrast (0.1-0.25 mL) was injected at the original site of isotope injection, followed by post-procedural mammographic spot images and planar scintigraphy to ensure accurate localisation and absence of ductograms. Technically successful ROLL was defined as iodinated contrast seen at the site of the VAB marker and its subsequent complete excision of the VAB marker.
 
Other relevant outcome measures, including operation duration, pathology of the lumpectomy specimen, margin status, successful surgical removal of all VAB markers and magnetic seeds, and reoperation within 6 months, were also analysed.
 
Statistical Analysis
 
Statistical analysis was performed using SPSS (Windows version 28.0; IBM Corp, Armonk [NY], United States). Graphical representations were made using commercial software GraphPad Prism (Windows version 9.3.1; GraphPad Software Inc, San Diego [CA], United States).
 
Frequencies and percentages were calculated for categorical data and compared using Fischer’s exact test. Continuous data were reported as medians with interquartile range and compared using the Mann-Whitney U test.
 
RESULTS
 
A total of 11 consecutive lumpectomy cases with stereotactic guidance targeting VAB markers and localised with magnetic seeds were identified between 1 April 2021 and 28 February 2022, after excluding a case in which multiple lesions localised with magnetic seeds were excised in one specimen. A control group of 11 consecutive lumpectomy cases with stereotactic guidance targeting VAB markers managed with ROLL between 2 May 2019 and 31 March 2021 was identified. All lumpectomies were performed by at least one specialist surgeon. There were seven specialist surgeons in the magnetic seed cohort, two of whom performed the lumpectomies in the ROLL cohort.
 
Baseline characteristics of the patients and lesions in both groups are shown in Table 1. Both groups of patients had similar age ranges. All the lesions initially manifested as microcalcifications, with or without architectural distortion. The histopathology from the VAB specimens in both cohorts were comparable, with most lesions being either atypical ductal hyperplasia (ADH) or ductal carcinoma in situ (DCIS), and a minority of lobular carcinoma in situ or invasive ductal carcinoma. None of the patients included in the study received neoadjuvant treatment prior to lumpectomy.
 
Table 1. Baseline characteristics of patients and lesions localised with magnetic seeds or radioguided occult lesion localisation.
 
The cases localised by magnetic seeds underwent lumpectomy with or without sentinel lymph node biopsy, with the magnetic seeds placed at a median of 9 days (interquartile range, 4-10) before surgery. All cases guided by ROLL underwent lumpectomy within the same day (Table 1).
 
The outcomes of the lumpectomy cases localised with magnetic seeds and ROLL are shown in Table 2. The technical success rates of marker deployment and ROLL injection were 100%. The preoperative mammographically detected post-VAB residual lesion size with inclusion of the VAB marker in both groups did not differ significantly (3 mm vs. 3 mm; p = 0.652). After localisation, the total span of the preoperative mammographically detected post-VAB residual lesion (with inclusion of the VAB marker) and the localisation agent (magnetic seed or iodinated contrast injected during ROLL) was significantly lower in the magnetic seed cohort compared with the ROLL cohort (6.5 mm vs. 15 mm; p < 0.001) [Table 2].
 
Table 2. Comparison between outcome measures of lumpectomy cases localised by magnetic seeds and radioguided occult lesion localisation.
 
The median operative time was lower in the magnetic seed cohort compared with the ROLL cohort (32 min vs. 52 min; p = 0.028), after exclusion of the cases that also underwent sentinel lymph node biopsy, excision of additional breast lesion(s) or contralateral breast surgery in the same setting. However, there were six such excluded cases in the magnetic seed cohort and one in the ROLL cohort, rendering the sample size small (Table 2).
 
The total resection volume of the relevant specimens in the cases localised by magnetic seeds was significantly lower than in the group using ROLL (11.5 cm3 vs. 21.2 cm3; p = 0.028) [Table 2 and Figure]. If the cases with involved or close margins (three in the magnetic seed cohort and two in the ROLL cohort) were excluded from analysis, the magnetic seeds group still achieved a lower total resection volume (11.3 cm3 vs. 25.2 cm3; p = 0.015). There was a higher proportion of cases with additional margins resected intraoperatively in the magnetic seeds group (54.5% vs. 36.4%), but it did not reach statistical significance (p = 0.670) [Table 2]. In two cases of the ROLL cohort and one case in the magnetic seed cohort, the measurements of the additional margins taken were not fully documented and they were assumed to be of negligible volume.
 
Figure. Column scatter graph of the total resection volume in the lumpectomy cases guided by magnetic seeds and radioguided occult lesion localisation. The median is represented with the interquartile range (horizontal lines). The difference between the two groups was significant (p = 0.028).
 
There was no incidence of magnetic seed migration in the entire cohort. All magnetic seeds and VAB markers were successfully removed from all patients (Table 2).
 
The final histopathological diagnoses of the lesions, which were taken as the higher of the grades between the VAB and lumpectomy specimens, were comparable in both groups with no significant difference (p = 0.565) [Table 2]. Two cases in the magnetic seed group and one in the ROLL group were upgraded from ADH to DCIS after surgical excision.
 
The margin clearance rates (72.7% vs. 81.8%) and reoperation rates (18.2% vs. 18.2%) were similar in both groups with no statistically significant difference (both p = 1.000) [Table 2]. In all cases localised by magnetic seeds, there was no margin involvement. Two cases with DCIS did have close margins (<1 mm and 0.5 mm) and underwent re-excision of margin and mastectomy, respectively. No evidence of malignancy was detected in the subsequent specimens obtained in either case, except a small focus of ADH found in the patient who had mastectomy. One patient who had DCIS with a close margin (1.8 mm) opted against reoperation, proceeded to adjuvant radiotherapy and has remained in remission up to the time of this writing (9 months after her operation). No significant postoperative complications were recorded in the magnetic seed cohort. In the group localised with ROLL, one case of DCIS had focally involved margins and underwent re-excision. Residual DCIS was found in the re-excision specimen. There was one case of DCIS with a close margin (<1 mm) upon which a radiotherapy boost instead of re-excision was decided after multidisciplinary team discussion. One patient underwent wound exploration and clot evacuation due to postoperative hematoma.
 
DISCUSSION
 
This study showed that a statistically significant smaller total resection volume could be achieved with magnetic seed localisation compared with ROLL, while maintaining a similar margin clearance and reoperation rate (Table 2). Although additional margins were excised intraoperatively in a non-significantly higher proportion of cases in the magnetic seed group, it did not lead to an overall increased total resection volume, which is the main determinant for cosmetic outcome.[7] Previous studies on cosmetic outcomes after BCS found that exceeding a resection volume of 50 to 85 cm3 was associated with a higher rate of cosmetic failure.[7] [8] [9] [10] [11] In our study, there was only one case in the magnetic seed cohort in which the total resection volume reached this range (50.1 cm3). Two cases from the ROLL cohort (77.8 cm3 and 71.3 cm3) fell within this range. None in the entire cohort exceeded 85 cm3.
 
The magnetic seeds were placed at a median of 9 days before surgery. In contrast, all cases guided by ROLL underwent same-day operation due to the constraint of the nature of radioisotopes. Successful removal of the magnetic seeds from the patient was achieved in all cases (Table 1). The median operative time was lower in the magnetic seed cohort (Table 2). No complications were observed in the group localised with magnetic seeds, while there was one case of postoperative hematoma requiring surgical wound exploration in the group guided by ROLL.
 
A possible reason for the improvement in resection volume using magnetic seeds could be its ability to more precisely localise lesions, thus enhancing surgical accuracy. With ROLL, the radioisotope, and hence the area with highest radioactivity detected by the gamma probe, infiltrates and disperses to adjacent tissues upon injection, as supported by our data, resulting in the surgeons resecting additional margins if residual activity is detected in the surgical bed.[12] Intraductal injections can also occur in ROLL, causing the radioisotopes to be even more widely dispersed, in which case a salvage localisation procedure would be necessary.[7] [13] [14] Moreover, in the post-excision specimen radiograph, it is easier to visualise the centre of the target if it is guided by magnetic seeds than by ROLL.[15] One case from the ROLL cohort in the study took up to four specimen excisions before the VAB marker was seen included within one of the specimens. It is possible that the diffuse distribution of the radioisotopes could have contributed to the need for repeated excisions in this case. In comparison, a magnetic seed can pinpoint the exact location of a lesion. The main factor that could undermine its accuracy would be migration. Previous studies showed that magnetic seed migration appears to occur more frequently when performed under stereotactic guidance owing to the ‘accordion’ effect, which the release of breast compression causes the magnetic seed to migrate along the direction of compression.[16] [17] This effect could be mitigated by using less compression before deploying the marker and by slowly releasing the breast from compression after placement.[17] The reported migration rates of magnetic seeds in previous studies were low[17] [18] [19]; none of the magnetic seeds migrated in this study.
 
Overall, the observations in this study echo those of previous works.[1] [2] [3] [5] It shows that magnetic seeds are a safe and effective localisation technique, and further suggests that they can reduce total resection volume compared with ROLL, without negatively impacting the margin clearance and reoperation rates. They could potentially benefit patients with early-stage breast cancer, for which BCS with adjuvant radiotherapy is the standard treatment. This is a large group of patients with good survival rates, who face substantial psychological stress.[6] [20] Although BCS preserves the breast, it results in different degrees of breast asymmetry, which runs the risk of affecting the psychosocial functioning of these surviving cancer patients.[6] Improving the cosmetic outcome without compromising the oncological margin status may contribute to their psychosocial well-being and quality of life.[7]
 
One major drawback of magnetic seeds is their high cost in comparison with ROLL and other localisation techniques, which may be a barrier to its adoption in some centres. The logistical advantage of magnetic seeds due to the decoupling of localisation and surgery may be able to reduce delays in surgery and increase overall efficiency.[7] Future cost-effectiveness analysis is required and should take into consideration the overall efficiency, which include surgical outcomes including cosmetic results and patient satisfaction.
 
Limitations
 
This study has several limitations. First, the sample size was small, including patients only from two regional hospitals, during the initial stage of magnetic seed adoption. Second, surgeons might have resected a larger amount of tissue in selected cases to ensure a clear margin, particularly if previous VAB pathology results already confirmed malignancy. This was assumed to be balanced out by the relatively comparable distribution of high risk versus malignant pathology of the VAB specimens in both cohorts. Contrary to the concern for selection bias, in the magnetic seed cohort of the current study, there were two more invasive cancer cases than in the ROLL cohort, which would presumably require a wider margin. Third, the specimen weights were not available in some cases, thus specimen volumes were retrospectively calculated using the ellipse volume formula, assuming that the surgical specimens were ellipsoids. In practice, however, they are often irregularly shaped. In a few cases, the measurements of some of the additional margins excised were not fully documented, which may affect the accuracy of the results.
 
CONCLUSION
 
This study demonstrated that localisation of non-palpable breast lesions with magnetic seeds can achieve a smaller total resection volume compared with ROLL, without affecting the margin clearance or reoperation rate. Multicentre studies with larger sample size are required to substantiate this finding and compare other surgical outcomes of magnetic seeds and ROLL.
 
REFERENCES
 
1. Gera R, Tayeh S, Al-Reefy S, Mokbel K. Evolving role of Magseed in wireless localization of breast lesions: systematic review and pooled analysis of 1,559 procedures. Anticancer Res. 2020;40:1809-15. Crossref
 
2. Miller ME, Patil N, Li P, Freyvogel M, Greenwalt I, Rock L, et al. Hospital system adoption of magnetic seeds for wireless breast and lymph node localization. Ann Surg Oncol. 2021;28:3223-9. Crossref
 
3. Thekkinkattil D, Kaushik M, Hoosein MM, Al-Attar M, Pilgrim S, Gvaramadze A, et al. A prospective, single-arm, multicentre clinical evaluation of a new localisation technique using non-radioactive Magseeds for surgery of clinically occult breast lesions. Clin Radiol. 2019;74:974.e7-11. Crossref
 
4. Redfern RE, Shermis RB. Initial experience using Magseed for breast lesion localization compared with wire-guided localization: analysis of volume and margin clearance rates. Ann Surg Oncol. 2022;29:3776-83. Crossref
 
5. Tsui HL, Fung EP, Kwok KM, Wong LK, Lo LW, Mak WS. Magnetic marker wireless localisation versus radioguided localisation of nonpalpable breast lesions. Hong Kong J Radiol. 2021;24:247-56. Crossref
 
6. Waljee JF, Hu ES, Ubel PA, Smith DM, Newman LA, Alderman AK. Effect of esthetic outcome after breast-conserving surgery on psychosocial functioning and quality of life. J Clin Oncol. 2008;26:3331-7. Crossref
 
7. Krekel N, Zonderhuis B, Muller S, Bril H, van Slooten HJ, de Lange de Klerk E, et al. Excessive resections in breastconserving surgery: a retrospective multicentre study. Breast J. 2011;17:602-9. Crossref
 
8. Cochrane RA, Valasiadou P, Wilson AR, Al-Ghazal SK, Macmillan RD. Cosmesis and satisfaction after breast-conserving surgery correlates with the percentage of breast volume excised. Br J Surg. 2003;90:1505-9. Crossref
 
9. Olivotto IA, Rose MA, Osteen RT, Love S, Cady B, Silver B, et al. Late cosmetic outcome after conservative surgery and radiotherapy: analysis of causes of cosmetic failure. Int J Radiat Oncol Biol Phys. 1989;17:747-53. Crossref
 
10. Taylor ME, Perez CA, Halverson KJ, Kuske RR, Philpott GW, Garcia DM, et al. Factors influencing cosmetic results after conservation therapy for breast cancer. Int J Radiat Oncol Biol Phys. 1995;31:753-64. Crossref
 
11. Vrieling C, Collette L, Fouquet A, Hoogenraad WJ, Horiot JH, Jager JJ, et al. The influence of patient, tumor and treatment factors on the cosmetic results after breast-conserving therapy in the EORTC ‘boost vs. no boost’ trial. EORTC Radiotherapy and Breast Cancer Cooperative Groups. Radiother Oncol. 2000;55:219-32. Crossref
 
12. Postma EL, Verkooijen HM, van Esser S, Hobbelink MG, van der Schelling GP, Koelemij R, et al. Efficacy of ‘radioguided occult lesion localisation’ (ROLL) versus ‘wire-guided localisation’ (WGL) in breast conserving surgery for non-palpable breast cancer: a randomised controlled multicentre trial. Breast Cancer Res Treat. 2012;136:469-78. Crossref
 
13. Luini A, Zurrida S, Paganelli G, Galimberti V, Sacchini V, Monti S, et al. Comparison of radioguided excision with wire localization of occult breast lesions. Br J Surg. 1999;86:522-5. Crossref
 
14. Rampaul RS, MacMillan RD, Evans AJ. Intraductal injection of the breast: a potential pitfall of radioisotope occult lesion localization. Br J Radiol. 2003;76:425-6. Crossref
 
15. Zacharioudakis K, Down S, Bholah Z, Lee S, Khan T, Maxwell AJ, et al. Is the future magnetic? Magseed localisation for non palpable breast cancer. A multi-centre non randomised control study. Eur J Surg Oncol. 2019;45:2016-21. Crossref
 
16. Esserman LE, Cura MA, DaCosta D. Recognizing pitfalls in early and late migration of clip markers after imaging-guided directional vacuum-assisted biopsy. Radiographics. 2004;24:147-56. Crossref
 
17. Lamb LR, Bahl M, Specht MC, D’Alessandro HA, Lehman CD. Evaluation of a nonradioactive magnetic marker wireless localization program. AJR Am J Roentgenol. 2018;211:940-5. Crossref
 
18. Singh P, Scoggins ME, Sahin AA, Hwang RF, Kuerer HM, Caudle AS, et al. Effectiveness and safety of Magseed-localization for excision of breast lesions: a prospective, phase IV trial. Ann Surg Open. 2020;1:e008. Crossref
 
19. Harvey JR, Lim Y, Murphy J, Howe M, Morris J, Goyal A, et al. Safety and feasibility of breast lesion localization using magnetic seeds (Magseed): a multi-centre, open-label cohort study. Breast Cancer Res Treat. 2018;169:531-6. Crossref
 
20. Institute of Medicine (US) and National Research Council (US) National Cancer Policy Board. Hewitt M, Herdman R, Holland J, editors. Meeting Psychosocial Needs of Women with Breast Cancer. Washington (DC): National Academies Press (US); 2004.
 
 
 
CASE REPORTS

Diabetic Ketoacidosis after Pembrolizumab Treatment in a Patient with Thymic Carcinoma and No Known Diabetes Mellitus: A Case Report

HCY Wong, HF Hung, CH Kwok

CASE REPORT
 
Diabetic Ketoacidosis after Pembrolizumab Treatment in a Patient with Thymic Carcinoma and No Known Diabetes Mellitus: A Case Report
 
HCY Wong1, HF Hung2, CH Kwok1
1 Department of Oncology, Princess Margaret Hospital, Hong Kong SAR, China
2 Department of Medicine and Geriatrics, Princess Margaret Hospital, Hong Kong SAR, China
 
Correspondence: Dr HCY Wong, Department of Oncology, Princess Margaret Hospital, Hong Kong SAR, China. Email: henrywong3011@gmail.com
 
Submitted: 2 May 2023; Accepted: 5 October 2023.
 
Contributors: All authors designed the study, acquired the data, analysed the data, drafted the manuscript, and critically revised the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.
 
Conflicts of Interest: All authors have disclosed no conflicts of interest.
 
Funding/Support: This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
 
Data Availability: All data generated or analysed during the present study are available from the corresponding author on reasonable request.
 
Ethics Approval: This study was approved by the Kowloon West Cluster Research Ethics Committee of Hospital Authority, Hong Kong [Ref No.: KW/EX-22-065(175-04)]. The requirement for patient consent was waived by the Committee as the patient had passed away at the time of writing.
 
Supplementary Material: The supplementary material was provided by the authors and some information may not have been peer reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by the Hong Kong College of Radiologists. The Hong Kong College of Radiologists disclaims all liability and responsibility arising from any reliance placed on the content.
 
 
 
 
CASE PRESENTATION
 
A 64-year-old Chinese man presented in December 2020 with a 3-month history of neck pain. Contrast-enhanced magnetic resonance imaging of the cervical spine revealed a large mass at the C6 vertebra (Figure 1). Computed tomography (CT)–guided biopsy of the mass revealed poorly differentiated carcinoma, with immunohistochemistry tests positive for p40, cytokeratin, CD5, PAX8 and c-kit, and negative for thyroid transcription factor 1, CDX2, leukocyte common antigen, S100 protein, desmin, synaptophysin and CD56. These results were suggestive of thymic squamous cell carcinoma. Positron emission tomography–CT showed a hypermetabolic thymic mass and multiple bone metastases, confirming the diagnosis of metastatic thymic carcinoma (Figure 2). No metastases were observed in the pancreas or adrenal glands. He had a past medical history of hypertension well controlled on amlodipine 5 mg daily. His cell counts, organ function, fasting glucose level and lipid profile were normal 1 month before the diagnosis of malignancy. He had no family history of diabetes mellitus.
 
Figure 1. C6 bone metastases (asterisk) on T1-weighted magnetic resonance imaging.
 
Figure 2. Positron emission tomography–computed tomography images of the thymic tumour (a) [arrow] and extensive bone metastases (b) [square bracket].
 
The patient received palliative radiotherapy to the painful cervical and thoracic spine bone metastases at a dose of 22.5 Gy in five daily fractions over 1 week with anterior-posterior opposing fields. The thymic tumour was covered in the radiation portal (Figure 3). Subsequently, he was started on palliative chemotherapy with etoposide and cisplatin (etoposide 100 mg/m2 and cisplatin 30 mg/m2 daily from day 1 to day 3 every 3 weeks) in January 2021. Regular zoledronic acid every 4 weeks was also given. He developed grade 4 neutropenia requiring granulocyte colony-stimulating factor support, treatment deferrals, and dose reduction. After six cycles of etoposide and cisplatin, CT showed mixed response with stable bone metastases but enlarging thymic tumour. As the patient was asymptomatic, he opted for a drug holiday.
 
Figure 3. Radiation portal covering the thymic tumour and bone metastases at the cervical (a) and thoracic spine (b).
 
Six months later, the patient complained of increasing lower back pain, and CT confirmed disease progression. He was started on pembrolizumab 200 mg every 3 weeks in November 2021. His fasting glucose level before treatment was 5.8 mmol/L. Baseline morning cortisol level was low at 36 nmol/L (normal: 133-537) while thyroid function was normal (thyroid-stimulating hormone level: 3.67 mIU/L [normal: 0.27-4.20], thyroxine level: 16.9 pmol/L [normal: 12.0-22.0]). He was given hydrocortisone replacement of 10 mg twice daily before starting immunotherapy. No significant side-effects were observed during the first three cycles.
 
The patient was admitted to the hospital for coma in January 2022, 3 days after the fourth cycle. Blood results showed severe hyperglycaemia (blood glucose level: 55.7 mmol/L) and metabolic acidosis (pH value: 7.22, bicarbonate level: 9.6 mmol/L). Multistix urine test revealed large amounts of ketones. Coupled with an elevated beta-hydroxybutyrate level, the clinical diagnosis of diabetic ketoacidosis (DKA) was suggested. He was treated with insulin infusion and fluid resuscitation. Subsequent investigations after stabilisation showed glycated haemoglobin level of 10.8% and low C-peptide (0.06 nmol/L; normal: 0.30-2.40) and insulin (1.6 mIU/L; normal 2.6-24.9) levels. Anti-GAD65 and anti-IA2 antibodies were negative. Insulin infusion was weaned off and switched to subcutaneous insulin glargine.
 
The oncology team decided to stop pembrolizumab as the severe hyperglycaemia and DKA could be related to the treatment. The plan was to consider second-line chemotherapy if there was progressive disease. A CT performed 2 months after the presentation of hyperglycaemia showed stable disease.
 
The patient’s diabetic control was brittle and he required three admissions within 2 months for insulin titration. The first admission was due to hyperglycaemia, whereas the latter two were for hypoglycaemia. In the third admission, he had persistent hypotension requiring escalation of hydrocortisone replacement for stabilisation. He ran a progressive downhill course with deconditioning and was readmitted for Klebsiella pneumoniae chest infection. He succumbed in May 2022, 4 months after the presentation of DKA and 17 months after the diagnosis of thymic carcinoma. Details about the patient’s timeline of events are illustrated in online supplementary Figure 1.
 
DISCUSSION
 
This patient developed life-threatening DKA following pembrolizumab treatment. Since the patient’s baseline fasting glucose level was normal and type 1 diabetes mellitus (T1DM) was considered unlikely for the patient’s age, his condition was most probably related to pembrolizumab.
 
In the past decade, immune checkpoint inhibitors (ICIs) have revolutionised the field of oncology. Pembrolizumab, a programmed cell death protein 1 (PD-1) inhibitor, has been studied in thymic carcinoma and shown promising efficacy in this entity with a poor prognosis.[1] [2] Despite important clinical benefits, ICIs are known for their immune-related adverse events (irAEs). These can target virtually any organ system and their severity can range from mild to life threatening. ICI-associated autoimmune diabetes mellitus (CIADM) is a rare complication of therapy, with an incidence of 0.2% to 1.4%.[3] With increasing clinician awareness of CIADM, its incidence is likely to increase.
 
The pathophysiology of CIADM involves the development of autoreactive T cells to pancreatic beta cells in response to a previous environmental trigger in genetically predisposed individuals. These T cells are generally controlled by immune checkpoints but pathology may result when activated by anti–PD-1/programmed death-ligand 1 (PD-L1) therapy.[2] [3]
 
The presentation of CIADM is variable, ranging from asymptomatic hyperglycaemia to severe diabetic complications. This patient’s presentation with DKA is the most common presentation of CIADM. In a pooled analysis of 200 case reports, 67.5% of CIADM patients presented with DKA.[4] The onset of CIADM varies with a median of 6 to 9 weeks but can occur as early as 1 week and as late as after the end of ICI treatment.[5]
 
The diagnosis of CIADM is characterised by two hallmark features of hyperglycaemia and low C-peptide level. When C-peptide level is normal, alternative causes of hyperglycaemia during ICI therapy should be considered, including exacerbation of type 2 diabetes mellitus, steroid-induced hyperglycaemia, autoimmune pancreatitis, and lipodystrophy.[3] Compared with T1DM where autoantibodies are present in >90% of cases, autoantibody positivity is lower in CIADM, ranging from 0% to 71%.[3] Therefore, negative values for this patient did not exclude CIADM.
 
Due to the rarity of CIADM, evaluating its risk factors based on clinical characteristics and biomarkers is challenging. A recent systematic review identified that close to 60% of CIADM patients had susceptibility haplotypes for T1DM, and patients with positive T1DM antibodies had an earlier onset of CIADM.[4] Although this provides important information about the disease nature and clinical course of CIADM, it does not help clinicians assess which patients need enhanced surveillance. Suazo-Zepeda et al[6] demonstrated that high PD-L1 expression is associated with the development of immune-related adverse reactions in patients with non–small cell lung cancer. Whether this correlation is also observed for CIADM and patients with thymic carcinoma is uncertain. Unfortunately, our patient had passed away at the time of writing this case report, and it was not possible to retrieve his archival specimen for PD-L1 testing.
 
The mainstay of treatment for CIADM is insulin. In contrast to other irAEs, treatment with glucocorticoids or immunosuppressants is not effective in these patients due to the almost complete destruction of beta cells.[3] [5] Steroids will likely negatively influence diabetes control in these patients and are not advised. In view of the irreversible damage to beta cells, similar to that in T1DM, a multi-dose basal-bolus regimen or continuous insulin pump is recommended to achieve glycaemic targets.[3] Our patient was prescribed long-acting insulin glargine only and discharged before C-peptide result was available, possibly one of the reasons for his labile glycaemic control.
 
Close surveillance for irAEs is essential while using ICIs. The 2021 American Society of Clinical Oncology guideline suggests testing of baseline fasting glucose level and monitoring of random glucose level before each dose of ICI.[7] Although CIADM is rare, regular monitoring to facilitate early endocrine team referral and insulin treatment to prevent life-threatening diabetic complications should be advocated. This patient had an elevated glycated haemoglobin level at presentation with DKA, suggesting he may have been hyperglycaemic during the preceding months. If regular surveillance of glucose level was performed, CIADM could have been diagnosed at an earlier stage. The suggested workflow for monitoring and treatment of the condition is depicted in online supplementary Figure 2.
 
In general, treatment of severe irAEs requires permanent discontinuation of the checkpoint inhibitor. Nonetheless similar to other immune-related endocrinopathies where the damage is irreversible, restarting treatment may be considered with close monitoring of diabetic control once glucose levels stabilise.[7]
 
In the two prospective phase II studies of the role of pembrolizumab in patients with thymic carcinoma, around 15% of patients developed grade >3 irAEs,[1] [2] much higher than the pooled incidence of <2% in a systematic review and meta-analysis of clinical trials evaluating anti-PD1 and anti–PD-L1 checkpoint inhibitors.[8] Notably, the types of high-grade irAEs in these patients were rarely seen in other tumour histologies. Of the 66 thymic carcinoma patients in the two studies, three developed myasthenia gravis (4.5%), two developed myocarditis (3.0%), one developed myositis (1.5%), and one developed myoclonus (1.5%).[1] [2] CIADM was observed in one patient (1.5%).[1] [2] The higher incidence and unusual clinical presentations of irAEs in patients with thymic carcinoma warrant further study and validation in larger patient cohorts.
 
Another reason this patient developed CIADM is that he may have had an underlying autoimmune condition. This patient had a low baseline cortisol level before treatment with pembrolizumab. It is possible that he had undiagnosed autoimmune adrenalitis since thymic carcinomas are associated with autoimmune paraneoplastic syndromes, albeit at lower rates compared with thymomas.[9] In retrospect, further workup with blood tests for adrenocorticotropic hormone level and antiadrenal antibodies should have been performed. Patients with preexisting autoimmune conditions are known to have higher risks for irAEs and have flare-ups during immunotherapy.[10] This may also explain the need to escalate our patient’s hydrocortisone dose after commencing pembrolizumab.
 
This case highlights the need for a heightened degree of suspicion amongst physicians for CIADM when treating patients with immunotherapy, especially those with thymic carcinoma, malignancies prone to paraneoplastic syndromes, or a past history of autoimmune diseases. Blood testing for C-peptide in patients who present with hyperglycaemia following immunotherapy aids the diagnosis of CIADM.
 
REFERENCES
 
1. Giaccone G, Kim C, Thompson J, McGuire C, Kallakury B, Chahine JJ, et al. Pembrolizumab in patients with thymic carcinoma: a single-arm, single-centre, phase 2 study. Lancet Oncol. 2018;19:347-55. Crossref
 
2. Cho J, Kim HS, Ku BM, Choi YL, Cristescu R, Han J, et al. Pembrolizumab for patients with refractory or relapsed thymic epithelial tumor: an open-label phase II trial. J Clin Oncol. 2019;37:2162-70. Crossref
 
3. Wu L, Tsang VH, Sasson SC, Menzies AM, Carlino MS, Brown DA, et al. Unravelling checkpoint inhibitor associated autoimmune diabetes: from bench to bedside. Front Endocrinol (Lausanne). 2021;12:764138. Crossref
 
4. Lo Preiato V, Salvagni S, Ricci C, Ardizzoni A, Pagotto U, Pelusi C. Diabetes mellitus induced by immune checkpoint inhibitors: type 1 diabetes variant or new clinical entity? Review of the literature. Rev Endocr Metab Disord. 2021;22:337-49. Crossref
 
5. Clotman K, Janssens K, Specenier P, Weets I, De Block CE. Programmed cell death-1 inhibitor–induced type 1 diabetes mellitus. J Clin Endocrinol Metab. 2018;103:3144-54. Crossref
 
6. Suazo-Zepeda E, Bokern M, Vinke PC, Hiltermann TJ, de Bock GH, Sidorenkov G. Risk factors for adverse events induced by immune checkpoint inhibitors in patients with non–small-cell lung cancer: a systematic review and meta-analysis. Cancer Immunol Immunother. 2021;70:3069-80. Crossref
 
7. Schneider BJ, Naidoo J, Santomasso BD, Lacchetti C, Adkins S, Anadkat M, et al. Management of immune-related adverse events in patients treated with immune checkpoint inhibitor therapy: ASCO Guideline update. J Clin Oncol. 2021;39:4073-126. Crossref
 
8. De Velasco G, Je Y, Bossé D, Awad MM, Ott PA, Moreira RB, et al. Comprehensive meta-analysis of key immune-related adverse events from CTLA-4 and PD-1/PD-L1 inhibitors in cancer patients. Cancer Immunol Res. 2017;5:312-8. Crossref
 
9. Padda SK, Yao X, Antonicelli A, Riess JW, Shang Y, Shrager JB, et al. Paraneoplastic syndromes and thymic malignancies: an examination of the international thymic malignancy interest group retrospective database. J Thorac Oncol. 2018;13:436-46. Crossref
 
10. Tison A, Quéré G, Misery L, Funck-Brentano E, Danlos FX, Routier E, et al. Safety and efficacy of immune checkpoint inhibitors in patients with cancer and preexisting autoimmune disease: a nationwide, multicenter cohort study. Arthritis Rheumatol. 2019;71:2100-11. Crossref
 
 
 

Computed Tomography and Magnetic Resonance Imaging Features of Pedal Ectrodactyly with Lateral Hindfoot Syndrome: A Case Report

JK Fung, JHM Cheng, JKC Chan, BWT Cheng, CY Chu, KH Chin

CASE REPORT
 
Computed Tomography and Magnetic Resonance Imaging Features of Pedal Ectrodactyly with Lateral Hindfoot Syndrome: A Case Report
 
JK Fung1, JHM Cheng1, JKC Chan2, BWT Cheng1, CY Chu1, KH Chin1
1 Department of Radiology, Pamela Youde Nethersole Eastern Hospital, Hong Kong SAR, China
2 Department of Orthopaedics and Traumatology, Pamela Youde Nethersole Eastern Hospital, Hong Kong SAR, China
 
Correspondence: Dr JK Fung, Department of Radiology, Pamela Youde Nethersole Eastern Hospital, Hong Kong SAR, China. Email:
 
Submitted: 11 June 2023; Accepted: 5 October 2023.
 
Contributors: JKF and JHMC designed the study. JKF, JHMC, JKCC, BWTC and CYC acquired and analysed the data. JKF, JHMC, JKCC and BWTC drafted the manuscript. JKF, JHMC, JKCC, CYC and KHC critically revised the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.
 
Conflicts of Interest: All authors have disclosed no conflicts of interest.
 
Funding/Support: This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
 
Data Availability: All data generated or analysed during the present study are available from the corresponding author on reasonable request.
 
Ethics Approval: This study was approved by the Hong Kong East Cluster Research Ethics Committee of Hospital Authority, Hong Kong (Ref No.: CIRB-2023-069-1). Informed patient consent for the study and publication was obtained.
 
 
 
 
INTRODUCTION
 
Ectrodactyly, also known as split hand and foot deformity, is a rare congenital skeletal deformity characterised by deficiency or absence of the central digital rays. The central cleft simulates the appearance of a lobster claw.[1] It can present as an isolated deformity or part of a syndrome—usually with autosomal dominance inheritance—such as the ectrodactyly-ectodermal dysplasia-clefting syndrome or limb-mammary syndrome.[2] [3]
 
We present a case with non-syndromic pedal ectrodactyly. The patient presented with chronic pain over the lateral ankle. To the best of our knowledge, this is the first report of imaging findings, highlighting consequent soft tissue findings in an adult with split foot deformity.
 
CASE PRESENTATION
 
A 31-year-old female with known left foot deformity since birth presented with a 2-year history of increasing pain over the lateral ankle. The pain was worse on movement and weight-bearing. There was no preceding injury. She was born full term by normal spontaneous delivery. She enjoyed good past health and had no other known congenital anomalies. Physical examination revealed clefting deformity and abnormal configuration of the left foot. The lateral ankle joint was tender with mild local swelling.
 
Radiographs and reconstructed three-dimensional computed tomography of the left foot and ankle revealed the presence of two complete rays and a singular tarsal bone articulating with the first ray. An incomplete ray was seen in between, comprised of two phalanges and a metatarsal head. The base of the proximal phalanx and the metatarsal head formed anomalous articulation with the first metatarsal head (Figure 1).
 
Figure 1. (a) Dorsopalmar oblique view of X-ray of the left foot. An incomplete ray and a single metatarsal articulating with the first ray are demonstrated. Synostosis of the talus and calcaneum is evident. (b) Frontal view of X-ray of the bilateral ankle joints shows abnormal configuration of the left calcaneum (notched arrow) and shortened and widened left distal fibula (arrow). Osteoarthritic changes of the left calcaneofibular neo-facet are evident. (c) Reformatted computed tomography in bone window of the left foot. The proximal phalanx (PP) and metatarsal head (MT) of the incomplete ray articulate with the first metatarsal head.
 
The left hindfoot was formed by a single bony structure with anterior bifid appearance, simulating non-segmentation of the talus and calcaneus (hereby termed the talocalcaneal complex) [Figures 1 and 2]. A calcaneofibular neo-facet was demonstrated. The distal fibula was shortened while the lateral malleolus appeared widened (Figures 1 and 2).
 
Figure 2. Reconstructed three-dimensional computed tomography of the left foot and the ankle, demonstrating the bony anatomy in lateral oblique (a) and frontal (b) projections, as well as the soft tissue components in the lateral (c) and frontal (d) projections.
 
Pes planar deformity was evident with loss of both medial and lateral longitudinal arches. The talus-first metatarsal angle (Meary’s angle) measured 15° while the calcaneal inclination angle approached 0°, signifying loss of the medial and lateral arches, respectively (Figure 3a). There was widening of the talocalcaneal angle, suggestive of hindfoot valgus deformity (Figure 3b). Osteophytes and subchondral sclerosis were seen around the first tarsalmetatarsal joint, suggestive of midfoot arthrosis (Figure 3b).
 
Figure 3. (a) Lateral view of X-ray of the left foot showing pes planar deformity and a talus-first metatarsal angle of 15˚ convex downwards (Meary’s angle in normal individuals is normally 0˚) [black dotted lines]. The calcaneal inclination angle, measured between the calcaneal inclination axis and the supporting horizontal surface, approaches 0˚ (normal, 20-30˚) [white dotted line]. (b) Dorsopalmar view of X-ray of the bilateral feet. Widened talocalcaneal angle (38˚) [black solid lines] is compared with normal right side (20˚) [white solid lines] (Kite’s angle in normal individuals is normally 25-40˚). It also shows arthrosis of the first tarsal-metatarsal joint (white arrow).
 
Computed tomography of the left ankle joint revealed bony remodelling with flat neo-facets at the calcaneofibular articulation. Secondary osteoarthritic changes and osteophytes were observed (Figure 4a). Magnetic resonance imaging showed associated marrow oedema, moderate synovial thickening, and interposed soft tissue oedema. Mild thickening of the adjacent peroneus brevis and longus tendon sheaths was observed (Figures 4b to 4e). Overall features were suggestive of lateral hindfoot impingement syndrome.
 
Figure 4. (a) Computed tomography and (b-e) magnetic resonance imaging of the left ankle joint. (a) Reformatted coronal projection showing osteophytosis (black arrows) and osteoarthritic changes of the calcaneofibular neo-facet. (b) Sagittal proton density (PD)–weighted short tau inversion recovery image. Increased marrow signals (hollow notched arrows) over distal fibula (DF) and talocalcaneal complex are suggestive of oedema. (c) Coronal PD-weighted fat-saturation image showing moderate soft tissue oedema and thickened sheath of the peroneus longus tendon (solid notched arrow). (d, e) Sequential PD-weighted axial slides through the talofibular joint showing moderate synovial thickening and interposed soft tissue oedema (black arrows).
 
Normal anterior talofibular ligament and calcaneofibular ligament were not well delineated, possibly due to either congenital absence or secondary to chronic impingement-related tearing (Figure 5a and 5b). The posterior talofibular ligament and anterior inferior tibiofibular ligament were small in calibre. Moderate thickening and high heterogeneous short tau inversion recovery signals of the peroneus longus tendon was seen, suggestive of tendinosis and interstitial tear (Figure 5c and 5d). It was possibly a sequala of the altered hindfoot biomechanics.
 
Figure 5. (a) Proton density (PD)–weighted axial view through the right distal calf for comparison. (b) PD-weighted axial view through the left distal calf. Normal anterior talofibular ligament (white arrow) is not well demonstrated. All muscles in the anterior, lateral, and posterior compartments are atrophic. (c) PD-weighted reformatted sagittal image and (d) PD-weighted fat-saturated sequence axial image showing thickening of the plantar portion of the peroneus longus tendon (notched arrows), with high heterogeneous signals suggestive of tendinosis and interstitial tear.
 
The left peroneus longus and brevis muscles were atrophic compared with the right side. Normal infra-malleolar portion of the peroneus brevis tendon was not demonstrated (Figure 6). Other calf muscles had smaller bulk than the right side, most evident at the lateral and posterior compartments (Figure 5b). The plantar portions of the tendons in the anterior and posterior compartment were attenuated.
 
Figure 6. Proton density–weighted axial magnetic resonance images through the right (a) and left (b) distal calves, with the anterior, posterior, and lateral compartments highlighted in green, blue, and yellow, respectively. The left peroneal brevis (PB) and peroneal longus (PL) muscles are atrophic. Muscles in the posterior compartment (tibialis posterior [PT], flexor digitorum longus [FDL], and flexor hallucis longus [FHL]) are also atrophic. Muscle bulks in the anterior compartment (left extensor hallucis longus [EHL] and extensor digitorum longus [EDL]) are preserved. The plantar portion of the anterior and posterior compartment tendons are mostly attenuated (not shown).
 
DISCUSSION
 
Ectrodactyly derives from the Greek words ‘ektromo’ and ‘daktylos’, meaning abortion and fingers, respectively. It is nonetheless not limited to the digits or upper limbs. Reportedly it represents a spectrum of limbic deformity, from aphalangia, adactylia and acheiria, to hemimelia or amelia. Most cases of sporadic ectrodactyly are unilateral.[4] The pathogenesis involves failure of initiation of apical ectodermal ridge, or subsequent signalling pathways, that contributes to truncation of all skeletal elements over the distal developing limb bud.[3]
 
Although ectrodactyly commonly presents as an isolated finding, it may form part of a syndrome. Ectrodactyly-ectodermal dysplasia-clefting syndrome is the most reported. It is an autosomal dominant condition, affecting structures derived from the ectoderm such as hair, skin, nails, and teeth, with such presentations as skin hypopigmentation, sparse hair, and dental defects. Genitourinary and lacrimal duct anomalies are also common. Limb-mammary syndrome is also well studied. It is characterised by mammary gland and nipple hypoplasia.[5] Other less common syndromes associated with ectrodactyly include Karsch–Neugebauer-syndrome (congenital nystagmus), Patterson-Stevenson- Fontaine syndrome (mandibulofacial dysostosis), and Adams-Oliver syndrome (scalp defect).[5] [6] Despite the characteristic features, marked phenotypic variability is reported, possibly related to variable expression and incomplete penetrance.
 
Blauth and Borisch[7] proposed a radiological classification for cleft foot that describes a spectrum of metatarsal defects ranging from types I to VI. The characteristics and possible associated features are shown in the Table.[7] With an incomplete metatarsus and two absent rays, our patient fell between type IV and V. In line with the findings, our patient demonstrates possible synostosis of the talus and calcaneum. From the case series, the authors suggested that cleft formation begins at the second or third ray, then proceeds in a distal to proximal fashion.[7] The fifth ray is usually the last affected (Table). Synostosis is commonly seen at the margin of the cleft. Associated features such as syndactyly, polydactyly and cross-bone deformities may be observed, yet appear widely variable.
 
Table. Radiological classification for cleft foot[7]
 
The presence of an incomplete ray caught our attention. According to the known pattern of deformation, the distal portion of the ray should be first affected. For the incomplete ray, provided the absence of a proximal metatarsal, the development of the distal metatarsal and phalanges are not to be expected. In split-hand deformities, syndactyly of the remaining digits is reportedly common. There are also rare cases of triphalangeal thumbs.[5] In the case series by Blauth and Borisch,[7] a few cases displaying central polydactylous element were also reported. Two of the cases showed duplicated phalanges of the second digit.[7] We remain open to the possibility that the incomplete metatarsal head in our patient might arise from the first ray as a polydactylous deformity.
 
The Bluman-Myerson classification stages adult acquired flatfoot deformity (AAFD) based on severity and rigidity of the flatfoot deformity.[8] The bony deformities in our patient caused severe and irreversible hindfoot valgus and rigid flatfoot deformity, classified as stage III AAFD. The altered bony configuration leads to lateral hindfoot (or subfibular) impingement with associated soft tissue fibrosis, midfoot arthrosis, peroneal tendinopathy, and calcaneofibular ligament entrapment.[9] Stage III AAFD is also not uncommonly present in patients with congenital tarsal coalition due to arch flattening and associated rigid hindfoot valgus.[10] The principles of treating stage III AAFD include correcting hindfoot valgus, realigning the midfoot from abduction deformity, and relieving the lateral compartment. In normal individuals, arthrodesis of the talonavicular and subtalar joints is usually performed, where most deformities are found. The calcaneocuboid joint may also be fused, after balancing against the risks of ankle valgus, failure of the deltoid ligament and worsened foot rigidity.[11] In our case, the aim of surgical treatment will be to relieve lateral compartment pressure and the associated pain. A bony procedure is needed as the primary pathology is bone anomaly. Treatment options are limited by a deformed talocalcaneal complex, the absence of few mid- and hind-foot bony structures and hence such joints as the subtalar and calcaneocuboid joints. Medialising calcaneal (talocalcaneal complex, in our case) osteotomy is the most feasible option. In view of the osteoarthritis of the calcaneofibular neo-facet, this procedure may not completely relieve the pain. Ankle fusion would be the last resort to treat the arthritic pain since severe functional disability would result.
 
CONCLUSION
 
Common features of ectrodactyly, including absent metatarsals and tarsal synostosis, are present in this case. The presence of an incomplete distal ray around the cleft is nonetheless discordant with current knowledge; we propose that it may arise from the first ray as a polydactylous deformity. Pes planus deformity and associated lateral hindfoot impingement syndrome were well demonstrated across different imaging modalities and consistent with the clinical presentation. Future case series directed at associated soft tissue injuries may be helpful in planning rehabilitation programmes and surgical interventions.
 
REFERENCES
 
1. Jindal G, Parmar VR, Gupta VK. Ectrodactyly/split hand feet malformation. Indian J Hum Genet. 2009;15:140-2. Crossref
 
2. Pinette M, Garcia L, Wax JR, Cartin A, Blackstone J. Familial ectrodactyly. J Ultrasound Med. 2006;25:1465-7. Crossref
 
3. Duijf PH, van Bokhoven H, Brunner HG. Pathogenesis of split-hand/split-foot malformation. Hum Mol Genet. 2003;12 Spec No 1: R51-60. Crossref
 
4. Krakow D. The dysostoses. In: Rimoin D, Pyeritz R, Korf B, editors. Emery and Rimoin’s Principles and Practice of Medical Genetics. 6th ed. Amsterdam: Elsevier; 2013. p 1-22. Crossref
 
5. van Bokhoven H, Hamel BC, Bamshad M, Sangiorgi E, Gurrieri F, Duijf PH, et al. p63 gene mutations in EEC syndrome, limb-mammary syndrome, and isolated split hand-split foot malformation suggest a genotype-phenotype correlation. Am J Med Genet. 2001;69:481-92 Crossref
 
6. Hernandez-Andrade E, Yeo L, Goncaives LF, Luewan S, Garcia M, Romero R. Fetal musculoskeletal system. In: Norton ME, Scoutt LM, Feldstein VA, editors. Callen’s Ultrasonography in Obstetrics and Gynaecology, 6th ed. Amsterdam: Elsevier; 2017. p 272-345.
 
7. Blauth W, Borisch NC. Cleft feet. Proposals for a new classification based on roentgenographic morphology. Clin Orthop Relat Res. 1990;(258):41-8. Crossref
 
8. Bluman EM, Title CI, Myerson MS. Posterior tibial tendon rupture: a refined classification system. Foot Ankle Clin. 2007;12:233-49. Crossref
 
9. Donovan A, Rosenberg ZS. Extraarticular lateral hindfoot impingement with posterior tibial tendon tear: MRI correlation. AJR Am J Roentgenol. 2009;193:672-8. Crossref
 
10. Flores DV, Mejía Gómez C, Fernández Hernando M, Davis MA, Pathria MN. Adult acquired flatfoot deformity: anatomy, biomechanics, staging, and imaging findings. Radiographics. 2019;39:1437-60. Crossref
 
11. Vulcano E, Deland JT, Ellis SJ. Approach and treatment of the adult acquired flatfoot deformity. Curr Rev Musculoskelet Med. 2013;6:294-303. Crossref
 
 
 
PICTORIAL ESSAYS

The Different Faces of Osler-Weber-Rendu Syndrome on Radiological Imaging

Rajoo Ramachandran, Nishita Goyal, Sheelaa Chinnappan, Harini Gnanavel, Jagadeesan Dhanasekaran, Rajeswaran Rangasami

PICTORIAL ESSAY
 
The Different Faces of Osler-Weber-Rendu Syndrome on Radiological Imaging
 
Rajoo Ramachandran1, Nishita Goyal1, Sheelaa Chinnappan1, Harini Gnanavel1, Jagadeesan Dhanasekaran2, Rajeswaran Rangasami1
1 Department of Radiodiagnosis, Sri Ramachandra Institute of Higher Education and Research, Chennai, India
2 Department of Intervention Radiology, Sri Ramachandra Institute of Higher Education and Research, Chennai, India
 
Correspondence: Dr N Goyal, Department of Radiodiagnosis, Sri Ramachandra Institute of Higher Education and Research, Chennai, India. Email: drgoyalnishita@gmail.com
 
Submitted: 4 September 2022; Accepted: 9 May 2023.
 
Contributors: All authors designed the study and acquired the data. R Ramachandran, NG, SC, HG and JD analysed the data. All authors drafted the manuscript and critically revised the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.
 
Conflicts of Interest: All authors have disclosed no conflicts of interest.
 
Funding/Support: This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
 
Data Availability: All data generated or analysed during the present study are available from the corresponding author on reasonable request.
 
Ethics Approval: This study was approved by the Publication Oversight Committee of Sri Ramachandra Institute of Higher Education and Research, India [Ref No.: 11026.docx(D110578897)]. The patients were treated in accordance with the Declaration of Helsinki and informed verbal consent was obtained from the patients.
 
 
 
 
INTRODUCTION
 
Osler-Weber-Rendu syndrome (OWRS), also known as hereditary haemorrhagic telangiectasia (HHT), is characterised by multiple mucocutaneous telangiectasias and multiorgan arteriovenous malformations (AVMs) due to abnormal vascular remodelling traceable to a genetic defect in a binding protein for transforming growth factor. The telangiectasias and AVMs have a propensity to bleed due to their fragile nature. The multisystemic involvement of OWRS makes it imperative for clinicians and radiologists to identify the different clinical manifestations and radiological features of this syndrome. In this pictorial essay, we give an overview of the radiological features found in patients with OWRS and the role of interventional radiology in management.
 
RADIOLOGICAL FEATURES OF THE CASES
 
Three patients with a radiological diagnosis of OWRS were included (Table 1). [1] [2] [3] They underwent contrastenhanced computed tomography (CECT) and magnetic resonance imaging (MRI) examinations followed by vascular interventions at our hospital from February 2019 to September 2020.
 
Table 1. Curaçao clinical criteria for diagnosis of hereditary haemorrhagic telangiectasia.[1] [2] [3]
 
Computed Tomography and Magnetic Resonance Imaging Acquisition
 
CECT examinations of the abdomen and the pelvis were acquired using a Revolution EVO Gen 2 128-slice unit (GE HealthCare, Beijing, China). CT images were obtained with parameters of 120 kV and current in auto mode. Axial thin sections were acquired from the dome of the diaphragm to the symphysis pubis and the data were processed into multiplanar reconstruction (MPR) images with 5-mm slice thickness and three-dimensional images. The reconstruction interval was 0.625 mm. Approximately 80 mL of non-ionic contrast material (iohexol 350 mg iodine/mL; GE HealthCare, Milwaukee [WI], US) was administered with a power injector at a rate of 3.5 mL/s. Image data were acquired 18 to 20 seconds (arterial phase) and 60 seconds (portal venous phase) post contrast injection. For CT pulmonary angiography, 60 mL of non-ionic contrast was administrated intravenously at the rate of 5 mL/s followed by 30 mL of a normal saline chaser. Axial thin sections acquired from the sternal notch to the xiphisternum were used to reconstruct three-dimensional and MPR images.
 
For MRI of the brain, the sequences were axial T1-weighted, T2-weighted, diffusion-weighted imaging, sagittal T1-weighted, coronal fast fluid-attenuated inversion recovery, two-dimensional time-of-flight, and three-dimensional time-of-flight acquired on a Signa HDxt 1.5T scanner (GE HealthCare, Milwaukee [WI], US). The CT and MRI images were sent to a picture archiving and communication system and interpreted at workstations. Interventional procedures were performed using a biplane system (Allura Biplane FD20/20; Philips, Best, the Netherlands).
 
Case 1
 
A 43-year-old female, who had three pregnancies and three live births previously, presented with abnormal uterine bleeding for 1 month. On pelvic ultrasound, an anterior wall uterine fibroid and a right ovarian cyst was noted. A CT of the abdomen and pelvis showed a right paraovarian cyst and a heterogeneous lesion within the right lobe of the liver. CECT of the abdomen showed extensive telangiectasias[4] (Figure 1a, 1b, and Tables 2 and 3) and confluent masses[5] [6] within both lobes of the liver, with the right one larger than the left one. Early opacification of the left portal vein was noted in the arterial phase, suggestive of AVMs[5] (arteriovenous shunts) [Figure 1c]. The extra- and intrahepatic arteries were dilated, with the common hepatic artery measuring around 10 mm in diameter.[5] [6] On clinical examination, telangiectasias were noted in the fingertips and oral cavity of the patient.
 
Figure 1. Case 1. Axial views of the abdomen in arterial phase at the level of the liver. (a) Multiple tiny areas of vascular puddling (arrowheads) in the hepatic parenchyma are evident, suggestive of telangiectasias. There is also early opacification of the left portal vein (yellow arrow), suggesting an arteriovenous malformation (Tables 2 and 3). (b) Medium-to-large sized vascular abnormalities representing confluent vascular masses (blue arrows) involving the right lobe of the liver are shown. (c) Multiple tortuous corkscrew-like vessels (yellow arrows) scattered in the hepatic parenchyma and the hepatic artery is dilated (white arrow), suggestive of telangiectatic vessels in Osler-Weber-Rendu syndrome.
 
Table 2. Salient features of arteriovenous malformations within different organs with presenting signs/symptoms and treatment options.[4] [8] [11]
 
Table 3. Different types of hepatic shunts on radiological imaging.[5] [6]
 
The diagnosis of OWRS was made (Table 1). The patient needed intervention in the liver AVMs to prevent development of portal hypertension as a complication. However, due to logistical reasons and the patient’s financial difficulties, follow-up imaging was suggested.
 
Case 2
 
A 2-year-old boy presented with a sudden onset episode of generalised seizures. His father had a history of brain AVMs. On brain MRI, a bilobed extra-axial lesion was seen in the right transverse temporal sulcus with blooming on gradient echo sequences and flow void on other sequences. A prominent feeder from the M4 segment of the right middle cerebral artery and a prominent vein from the superior anastomotic vein of Trolard were seen draining the lesion into the superior sagittal sinus, suggestive of AVM (Figure 2a and 2b).[1] [7]
 
Figure 2. Case 2. (a) Axial magnetic resonance angiography showing a well-defined extra-axial lesion (star) in the right transverse temporal sulcus communicating with a prominent feeder from the M4 segment of the right middle cerebral artery (white arrow). There is associated left cerebral hemisphere atrophy. (b) Reformatted magnetic resonance venography showing a bilobed extra-axial lesion with a prominent vein representing the superior anastomotic vein of Trolard (curved yellow arrow) seen draining the lesion into the superior sagittal sinus. (c) Axial T2-weighted image (T2WI) showing a pachygyria-polymicrogyria complex (white arrows) involving the right parietal lobe adjacent to the abovementioned lesion. The lesion appears hypointense on T2WI, likely due to flow voids, suggestive of a high-flow arteriovenous fistula. (d) Axial T2WI shows an associated porencephalic cyst (star) involving the left cerebral hemisphere communicating with the frontal horn of the left lateral ventricle. (e) Preprocedural digital subtraction angiography image showing pial arteriovenous fistula involving the right fronto-parietal region fed by a right middle cerebral artery–M4 branch (yellow arrow) and draining via the vein of Trolard (blue arrow) into the superior sagittal sinus with a large venous sac (star). (f) The arteriovenous malformation (AVM) was embolised and post-procedure angiography image shows complete exclusion of the AVM from the circulation (blue arrow). (g) A non-contrast axial computed tomography of the brain shows streak artifacts secondary to post-embolisation status of the AVM (yellow star). Associated left cerebral hemisphere atrophy is noted (curved green arrow).
 
A pachygyria-polymicrogyria complex involved the right parietal lobe adjacent to the lesion (Figure 2c and 2d).[7] A large left-sided porencephalic cyst[7] communicating with the ipsilateral lateral ventricle exerted mass effect in the form of a midline shift of 7 mm to the right. There was associated left cerebral hemiatrophy (Figure 2c and 2d). The diagnosis of OWRS was made (Table 1). In view of the high-flow AVM, embolisation was advised. Endovascular glue embolisation of the right parietal pial arteriovenous fistula was performed by administering iodixanol contrast (GE HealthCare, Milwaukee [WI], US) using a microcatheter (Excelsior XT-17; Stryker, Fremont [CA], US), microwire combination (Transcend; Meditech, Watertown [MA], US) and a HyperGlide balloon (eV3 Endovascular Inc, Irvine [CA], US) to inject 66% glue [n-butyl-2-cyanoacrylate liquid embolic system (Trufill; Cordis Neurovascular, Miami Lakes [FL], US)] as an embolisation agent mixed with ethiodised oil in various dilution ratios depending on the application to control polymerisation rate. Post-procedure angiography and brain CT were performed, which showed complete exclusion of the AVM from the circulation (Figure 2e to 2g). The patient was started on antiepileptic drugs and did not have any other seizure episode during the course in the hospital.
 
Case 3
 
A 43-year-old male presented with one episode of haemoptysis (around 10 mL of blood) and two episodes of generalised tonic-clonic seizures. Multiple scattered non-haemorrhagic lacunar infarcts were noted in the right occipital lobe and both fronto-parietal regions. Chest radiography showed multiple homogenous mass lesions in the right lower zone and left upper and lower zones (Figure 3a). Subsequent CT pulmonary angiography revealed multiple AVMs involving the anterior basal segment of the right lower lobe, the apicoposterior segment of the left upper lobe, and the lateral basal segment of the left lower lobe. Multiple tiny AVMs were identified in the right lower lobe.[8] [9] The feeding artery diameter of a few of the AVMs was >3 mm[10] (Figure 3b). The diagnosis of OWRS was made (Table 1). In view of feeding artery diameter size being >3 mm, the patient was advised to undergo curative embolisation to prevent pulmonary as well as cerebral complications. The preprocedural angiogram (Figure 3c and 3d) showed pulmonary AVMs supplied from the basal segmental arteries. Endovascular glue embolisation of the pulmonary AVMs was performed by administering iohexol contrast using a microcatheter and/or microwire and coils with 50% glue as an embolisation agent mixed with ethiodised oil in various dilution ratios. Post-embolisation angiography (Figure 3e) showed complete exclusion of the AVMs from the circulation.
 
Figure 3. Case 3. (a) Chest radiograph showing multiple homogenous mass lesions (stars) seen in the right lower zone and left upper and lower zones. (b) Reformatted coronal maximum intensity projection image showing multiple arteriovenous malformations (AVMs) involving the antero-basal segment of the right lower lobe, apicoposterior segment of the left upper lobe, and the lateral basal segment of the left lobe (red arrows). Multiple tiny AVMs are seen in the right lower lobes (yellow arrowheads). (c) Pre-procedure digital subtraction angiography image of the right lung showing pulmonary AVM (star) with supply from the basal segmental arteries (blue arrow). (d) The AVM nidus (star) was progressively embolised using micro coils with 60% glue and subsequent opacification was seen. Blue arrow indicates vascular supply to the nidus from the basal segmental arteries. (e) Post-procedure angiogram shows complete exclusion of AVM from circulation (blue arrow).
 
DISCUSSION
 
Aetiologically, OWRS has been classified into different types based on the genetic mutations found in these patients. HHT type 1, found in nearly 61% of cases,[8] shows a mutation in the endoglin gene located on chromosome 9 and is found on the inner cell membrane of the endothelial cells lining the blood vessels. Patients with these mutations are generally predisposed to cerebral and pulmonary AVMs.[8] HHT type 2, found in nearly 37% of cases, shows a mutation in the activin A receptor-like type 1 (ACVRL-1) gene or the activin receptor-like kinase-1 (ALK-1) gene located on chromosome 12.[8] [11] These patients have been shown to have liver AVMs.[4] Mutations in the mothers against decapentaplegic homolog 4 (SMAD4) gene,[9] which encodes protein for signal transmission from the transforming growth factor beta receptor, has been implicated in 2% of cases of HHT with juvenile gastrointestinal polyposis.[9] Patients having bone morphogenetic protein-9 (BMPR-9) and RSA-1 gene mutations have also shown phenotypic overlap with telangiectasia.[9]
 
Clinical Diagnosis, Signs, and Symptoms
 
The clinical diagnosis of OWRS is made using the four Curaçao clinical criteria,[1] [2] [3] namely: (1) recurrent epistaxis; (2) telangiectasias involving sites including the lips, oral cavity, nose, and fingers; (3) visceral lesions with the gastrointestinal tract, liver, pulmonary, cerebral or spinal involvement; and (4) family history of HHT in a first-degree relative (Table 1).
 
Recurrent epistaxis is the most common symptom which can begin in childhood or adolescensce.[8] Low-pressure packing techniques can be used to manage such episodes. Telangiectasias affected individuals can present post puberty or in adulthood. It happens when capillaries fail to develop between arterioles and venules, commonly involving the face, lips, tongue, palm, and fingers (periungual and nail bed).[8] Telangiectasias can also develop in the gastrointestinal tract, presenting most commonly in the fourth decade of life with stomach and duodenum being the most common sites.[6] [8] AVMs, which are direct communications between blood vessels having a calibre greater than telangiectatic vessels, are also seen in the patients.[9] [11]
 
Brain Involvement
 
Distal emboli containing blood clots or bacteria from pulmonary AVMs may result in abscess formation and ischaemic stroke (Table 2).[4] [8] [11]
 
The brain abscesses are generally multiple and recurrent, and involve the superficial layers of the cerebral lobes, most commonly occurring in the parietal lobe. A higher incidence is seen between the third and the fifth decades of life, corresponding to increased pulmonary AVMs.[4] [11]
 
Imaging Features of Cerebral Arteriovenous Malformations
 
Cerebral AVMs are seen as serpiginous areas of flow void with invasion into the brain parenchyma on MRI. The feeding artery and the draining veins can be identified on different sections (Figure 2a and 2b). Some patients may have high signal intensity on T1-weighted imaging within the basal nuclei that can be a result of the metabolic disorder caused by hepatic artery-portal venous shunting. In equivocal lesions, cerebral angiography is performed, which may show high flow pial AVFs occurring due to the lack of an intervening capillary bed (Figure 2a and 2b).[1]
 
Cortical Development Malformation
 
Cortical developmental malformation is another feature which can be seen in the paediatric population. It involves two main entities: polymicrogyria and heterotopia.[7]
 
Patients with polymicrogyria can present with developmental delay, cognitive abnormalities, and epilepsy (about 78% of cases).[4] Epilepsy shows earlier onset in patients having higher degrees of polymicrogyria.[7] A favourable prognosis is present in patients with unilateral and localised polymicrogyria. The imaging features of polymicrogyria on MRI are smaller gyri with thin, shallow sulci separating them (Figure 2c and 2d). The cortex appears thickened, with an irregular surface and abnormal vasculature in close proximity. It is most commonly seen in the perisylvian region, followed by parietal, parietotemporal, and frontal regions.[7]
 
Heterotopia is an abnormal location of normal neuronal cells due to abnormal migration. The most commonly seen variant in OWRS is the periventricular nodular type. Bilateral occurrence is more common in the frontal lobes.[7]
 
Brain and pulmonary AVMs seem to have a higher incidence in patients with cortical developmental malformations.[7] [8]
 
Lung Involvement
 
Pulmonary AVMs are the most striking features of lung involvement, seen in nearly 50% of HHT cases (Table 2).[8] The anatomical structure of AVMs can be simple, with one feeding artery and one draining vein, or complex with ≥ 2 arterial branches and draining veins.[12]
 
Chest radiography shows well-defined nodules within the lung (Figure 3a). Cardiomegaly and prominent pulmonary arteries can also be seen (Table 2).
 
Chest CT with MPR and maximum intensity projections show one or multiple serpiginous masses/nodules with ≥1 enlarged feeding artery (diameter of ≥3 mm) and draining vein (Table 2 and Figure 3b).[10] [12] Contrast-enhanced magnetic resonance angiography can show all pulmonary AVMs with feeding arteries having a diameter >3 mm.[10]
 
The treatment of choice is transarterial pulmonary AVM vaso-occlusion with coils (Table 2).[9]
 
Liver Involvement
 
Ultrasound imaging can show an increase in common hepatic artery calibre (>7 mm) and intrahepatic hypervascularity.[6] Doppler imaging shows pulsatile portal flow in cases of arterioportal shunting and pulsatile hepatic venous flow in cases of arteriovenous shunting.[6]
 
Focal nodular hyperplasia and hepatic AVMs are more commonly seen in patients with ALK-1 gene mutation. Increased sinusoidal blood flow leads to portal hypertension, pseudocirrhosis of the liver, and hepatic encephalopathy in later stages.[4] [6]
 
Liver Telangiectasias
 
CECT of the abdomen with MPR and maximum intensity projections can show telangiectasias in proximity to large vessels[5] and also in the subcapsular regions. These are seen as focal hyperattenuating rounded nodular lesions in the arterial and late arterial phases, which become isodense with the hepatic parenchyma in the hepatic phase (Figure 1a, 1b, and Tables 2 and 3).[4] [5] [6] MRI shows high signal intensity on T2-weighted imaging and appears hypointense on T1-weighted imaging.
 
Large Confluent Masses
 
On CECT of the abdomen, large confluent masses within the liver (>10 mm) may be seen enhancing in enhancing arterial phase with persistent enhancement in hepatic phase.
 
Hereditary Haemorrhagic Telangiectasia and Cirrhosis
 
The differentiating features between HHT and cirrhosis on imaging for hepatic perfusion abnormalities are listed in Table 4.[5]
 
Table 4. Important differentiating points between hereditary haemorrhagic telangiectasia and cirrhosis on radiological imaging.[5]
 
Pancreatic Arteriovenous Malformations
 
Pancreatic AVMs are the most common extrahepatic AVMs. CT imaging shows focal lesions (diameter: 5-10 mm) with increased vascularity. Arteriovenous shunting to the splenic vein or superior mesenteric vein may be noted.[5] [6]
 
Gastrointestinal Involvement
 
The most common manifestation is telangiectasias, both in the small bowel (around 60%) and the stomach (around 30%). Patients with HHT have an increased incidence of small bowel polyps as compared to the general population.[6] [8] Gastrointestinal haemorrhage is a common presentation, usually seen in the fourth to the fifth decades[6] (Table 2). On colonoscopy, 31% to 32% of OWRS patients show colonic AVMs associated with the HHT1 genotype having the endoglin mutation.[4] SMAD4 mutations in OWRS patients result in juvenile polyposis, which is difficult to differentiate from the juvenile polyposis caused by BMPR1A mutations in the general population. Rarely, OWRS cases show intramural haematomas on endoscopic evaluation.[6] [8] [9]
 
Ocular Manifestations
 
The ocular signs and symptoms include conjunctival telangiectasias or AVMs, bloody tears, conjunctival post-haemorrhagic granulomatous lesions, and the recently described association of choriocapillaris atrophy with HHT. However, retinal involvement prevalence is only around 1%.[3] [9]
 
Management
 
Embolisation of pulmonary and cerebral AVMs is important to avoid serious complications. Pulmonary AVMs having a feeding artery diameter >3 mm are ideal candidates for embolisation to decrease the risk of pulmonary haemorrhages and paradoxical emboli to the brain.[10] Patients should be advised about the side-effects related to embolotherapy for pulmonary AVMs, such as transient post-procedural chest pain and self-limiting pleurisy. Brain AVMs need treatment to prevent the occurrence of stroke and abscess. Microsurgical resection, stereotactic radiation surgery, and endovascular embolisation can be performed. Embolisation can be pre-microsurgical, pre-radiation surgery, curative, or palliative depending on the patient. Preprocedural CT angiography and antibiotic prophylaxis for the risk of bacteraemia is advised. Digital subtraction angiography is used for the procedure.[13] In our patients, since the AVMs were relatively small with few feeding pedicles and without perinidal angiogenesis, curative embolisation was performed (Table 2).
 
Follow-up is an important step in management of these patients. A baseline post-embolisation scan is performed at 6 months and then repeated at 12 months to ensure sac involution, followed by follow-up intervals of 2 years to detect growth of untreated pulmonary AVMs and reperfusion of treated AVMs.[14] Chest CT or contrast-enhanced MRI are used for follow-up post-coiling to look for reperfusion of occluded pulmonary AVMs.[15] The risk of reperfusion increases with larger feeding artery diameter, using less number of coils, oversized coils, or more proximal placement of the coil within the feeding artery. In such cases, it is increasingly difficult to treat these reperfused AVMs, resulting in higher recurrence rates.[15]
 
CONCLUSION
 
AVMs associated with OWRS are ticking time bombs which can result in catastrophic events such as pulmonary haemorrhages, brain abscess, stroke, chronic gastrointestinal bleeds, high-output cardiac failure, paraparesis, and, rarely, paraplegia. Since affected individuals are generally asymptomatic, the lesions are often discovered incidentally. Radiologists must always be on the lookout for such classical findings to not only aid the diagnosis but also lower the risk of complications.
 
REFERENCES
 
1. Geibprasert S, Pongpech S, Jiarakongmun P, Shroff MM, Armstrong DC, Krings T. Radiologic assessment of brain arteriovenous malformations: what clinicians need to know. Radiographics. 2010;30:483-501. Crossref
 
2. Ha M, Kim YJ, Kwon KA, Hahm KB, Kim MJ, Kim DK, et al. Gastric angiodysplasia in a hereditary hemorrhagic telangiectasia type 2 patient. World J Gastroenterol. 2012;18:1840-4. Crossref
 
3. Rinaldi M, Buscarini E, Danesino C, Chiosi F, De Benedictis A, Porcellini A, et al. Ocular manifestations in hereditary hemorrhagic telangiectasia (Rendu-Osler-Weber disease): a case-series. Ophthalmic Genet. 2011;32:12-7. Crossref
 
4. Singh A, Suri V, Jain S, Varma S. Rare manifestations in a case of Osler-Weber-Rendu disease. BMJ Case Rep. 2015;2015:bcr2014207852. Crossref
 
5. Siddiki H, Doherty MG, Fletcher JG, Stanson AW, Vrtiska TJ, Hough DM, et al. Abdominal findings in hereditary hemorrhagic telangiectasia: pictorial essay on 2D and 3D findings with isotropic multiphase CT. Radiographics. 2008;28:171-84. Crossref
 
6. Jackson SB, Villano NP, Benhammou JN, Lewis M, Pisegna JR, Padua D. Gastrointestinal manifestations of hereditary hemorrhagic telangiectasia (HHT): a systematic review of the literature. Dig Dis Sci. 2017;62:2623-30. Crossref
 
7. Palagallo GJ, McWilliams SR, Sekarski LA, Sharma A, Goyal MS, White AJ. The prevalence of malformations of cortical development in a pediatric hereditary hemorrhagic telangiectasia population. AJNR Am J Neuroradiol. 2017;38:383-6. Crossref
 
8. Macri A, Wilson AM, Shafaat O, Sharma S. Osler-Weber-Rendu disease. Available from: http://www.ncbi.nlm.nih.gov/books/NBK482361/. Accessed 8 Nov 2022.
 
9. National Organization for Rare Disorders. Hereditary hemorrhagic telangiectasia. 2021. Available from: https://rarediseases.org/rare-diseases/hereditary-hemorrhagic-telangiectasia/. Accessed 8 Nov 2022.
 
10. Majumdar S, McWilliams JP. Approach to pulmonary arteriovenous malformations: a comprehensive update. J Clin Med. 2020;9:1927. Crossref
 
11. Maher CO, Piepgras DG, Brown RD Jr, Friedman JA, Pollock BE. Cerebrovascular manifestations in 321 cases of hereditary hemorrhagic telangiectasia. Stroke. 2001;32:877-82. Crossref
 
12. Lee HN, Hyun D. Pulmonary arteriovenous malformation and its vascular mimickers. Korean J Radiol. 2022;23:202-17. Crossref
 
13. Vollherbst DF, Chapot R, Bendszus M, Möhlenbruch MA. Glue, Onyx, Squid or PHIL? Liquid embolic agents for the embolization of cerebral arteriovenous malformations and dural arteriovenous fistulas. Clin Neuroradiol. 2022;32:25-38. Crossref
 
14. Hong J, Lee SY, Cha JG, Lim JK, Park J, Lee J, et al. Pulmonary arteriovenous malformation (PAVM) embolization: prediction of angiographically-confirmed recanalization according to PAVM Diameter changes on CT. CVIR Endovasc. 2021;4:16. Crossref
 
15. Maruno M, Kiyosue H, Hongo N, Matsumoto S, Mori H. Where is the origin of the last normal branch from feeding artery of pulmonary arteriovenous malformations? Cardiovasc Intervent Radiol. 2018;41:1849-56. Crossref
 
 
 

Magnetic Resonance Imaging Findings of Cardiac Metastases: A Pictorial Essay

Eda Cingoz, Rana Gunoz Comert, Mehmet Cingoz, Memduh Dursun

PICTORIAL ESSAY
 
Magnetic Resonance Imaging Findings of Cardiac Metastases: A Pictorial Essay
 
Eda Cingoz1, Rana Gunoz Comert2, Mehmet Cingoz3, Memduh Dursun2
1 Department of Radiology, Istanbul Bagcilar Training and Research Hospital, Istanbul, Turkey
2 Department of Radiology, Istanbul University, Istanbul, Turkey
3 Department of Radiology, Basaksehir Cam and Sakura City Hospital, Istanbul, Turkey
 
Correspondence: Dr E Cingoz, Department of Radiology, Istanbul Bagcilar Training and Research Hospital, Istanbul, Turkey. Email: edacanipek@gmail.com
 
Submitted: 31 July 2023; Accepted: 19 October 2023.
 
Contributors: EC and MD designed the study. RGC acquired the data. MC analysed the data. EC, RGC and MC drafted the manuscript. MD critically revised the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.
 
Conflicts of Interest: All authors have disclosed no conflicts of interest.
 
Funding/Support: This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
 
Data Availability: All data generated or analysed during the present study are available from the corresponding author on reasonable request.
 
Ethics Approval: This study was approved by the Istanbul Medical Faculty Clinical Research Ethics Committee, Turkey (Ref No.: 2022/709). The requirement for informed patient consent was waived by the Committee due to the retrospective design of the study. The data used in the study were de-identified.
 
 
 
 
INTRODUCTION
 
Tumours metastatic to the heart may involve the pericardium, epicardium, myocardium and/or endocardium.[1] [2] These metastases are approximately 30 times more common than primary cardiac tumours.[3] Cardiac metastases occur with an incidence of 1.5% to 20% according to postmortem statistics.[4] Echocardiographic data have suggested an increase in the incidence of cardiac metastases over the past 30 years due to increased life expectancy in patients with a known malignancy who have benefitted from progress in cancer treatment.[5]
 
Over 90% of cardiac metastases remain clinically silent, explaining the lack of antemortem diagnosis.[2] Tumours that most commonly involve the heart include lung cancer, breast cancer, melanoma, and lymphoma, reflecting the relatively high prevalence of these malignancies in the population.[6] In all, 36% to 39% of cardiac metastases originate from primary lung cancer, followed by 10% to 12% from breast cancer and 10% to 21% from haematological malignancies.[1] [4] Tumours such as melanoma have a much higher propensity (nearly 50%) to involve the heart.[1] [7] Following melanoma, the tumours that tend to metastasise to the heart include ovarian, gastric, renal, and pancreatic carcinomas.[1] [4]
 
Cardiac metastases may manifest a variety of appearances. A mass stemming from the lung or mediastinum can directly invade the heart. Additionally, tumour cells that reach the heart via the pulmonary veins (haematogenous spread) can manifest as a central mass.[8] Metastases may present as pericardial effusion and nodularity, as well as myocardial nodules.
 
Echocardiography is the most frequently used initial modality for the diagnosis of any cardiac mass, though there are some limitations regarding its diagnostic capabilities. First, it is difficult to differentiate a cardiac thrombus from an endocardial mass with echocardiography unless it is performed with contrast imaging.[9] Moreover, metastases to the heart with extracardiac extension cannot be evaluated solely by using echocardiography. Cardiac magnetic resonance imaging (MRI) provides a more comprehensive anatomical evaluation by demonstrating the entire thoracic cavity and serves as an excellent diagnostic tool in patients with suspected cardiac metastases. The use of contrast medium adds extra value to cardiac MRI images since it may allow the distinction between a mass that shows contrast enhancement versus a non-enhancing thrombus. Cardiac MRI offers excellent soft tissue contrast resolution, allowing the clinician to distinguish between metastatic lesions and myocardial tissue. The differentiation between benign and malignant tumour, thrombus and blood may be provided by MRI with relative ease.
 
This pictorial essay presents our experience and highlight the diverse appearances of cardiac involvement by metastases.
 
CARDIOVASCULAR MAGNETIC RESONANCE PROTOCOL
 
A total of 1119 consecutive cardiac MRI studies that were carried out at our institution from January 2015 to March 2022 were reviewed and 22 cases of metastases involving the heart were detected. These 22 patients were aged 14 to 98 years, and their demographics as well as features and locations of the lesions were recorded.
 
Cardiac MRI studies were performed using a 1.5T scanner (Aera; Siemens, Erlangen, Germany) with phased array coil systems. The protocol was the standardised protocol as previously described by the Society for Cardiovascular Magnetic Resonance,[10] which includes steady-state free precession cine imaging, bright-blood and dark-blood single-shot imaging, T1-weighted and T2-weighted fast spin-echo imaging, and early and late perfusion imaging during and after the administration of contrast medium.[10] [11] [12]
 
IMAGING FINDINGS
 
All cardiac MRI images were evaluated by a radiologist specialising in cardiac imaging with experience of >20 years. Table 1 shows the origins of the primary tumours, while Tables 2 and 3 show the sites of cardiac involvement. Half of the patients were male, suggesting an absence of gender predilection. The mean age of the patients was 58.5 years, with a standard deviation of 21.1 years.
 
Table 1. Distribution of cardiac metastases according to the origin of the primary tumour (n = 22).
 
Table 2. Distribution of cardiac metastases according to the involvement of the cardiac chambers (n = 22).
 
Table 3. Distribution of cardiac metastases according to the involvement of the cardiac tissue layers (n = 22).
 
Lymphatic Spread
 
Figure 1 depicts metastatic involvement of pericardial fat surrounding the right coronary artery from a mediastinal lymphoma. Lymphatic drainage of the pericardial space is by lymphatic channels located in the pericardium that converge at the root of the aorta, where these channels are most often obstructed, giving rise to pericardial effusion.[8] Figure 2 shows an example of pericardial metastasis from renal cell carcinoma. Metastatic involvement of the pericardium gives rise to pericarditis initially, followed by haemorrhagic effusion.[13] The development of symptoms during the progression of pericardial effusion depends on the rate of accumulation of fluid. Although the accumulation of large amounts of fluid over time may not cause symptoms, rapid accumulation of small amounts of fluid may cause serious symptoms.[13] In addition to pericardial effusion, deposits of malignant cells on the pericardium may also result in constrictive pericarditis, leading to the deterioration of heart function.[2]
 
Figure 1. Cardiac magnetic resonance imaging of a 74-year-old female patient with a diagnosis of mediastinal lymphoma (white arrow in [a]). (b-d) Images showing lymphoma infiltration around right coronary artery and its branches (red arrows). Pericardial and bilateral pleural effusions were also noted (green arrows in [b]).
 
Figure 2. Three-chamber cardiac magnetic resonance image of a 77-year-old male patient with known renal cell carcinoma showed the presence of a pericardial mass (white arrow) at the level of mid-lateral segment of the left ventricle. The tumour showed slight myocardial invasion at the outermost portion of the left ventricular muscle (red arrow).
 
Haematogenous Spread
 
Figures 3 and 4 demonstrate the myocardial metastasis of a uterine leiomyosarcoma and an iliopsoas muscle sarcoma, respectively. Figures 5 and 6 depict the metastasis of gastric carcinoma to different chambers of the heart. Figure 7 shows the myocardial involvement of leukaemia that diffusely involved the left ventricular myocardium. Figure 8 depicts a nasopharyngeal carcinoma metastasis that caused left myocardial involvement.
 
Figure 3. Two-chamber cardiac magnetic resonance image of a 52-year-old female patient diagnosed with uterine leiomyosarcoma (a). The mass is contiguous and indistinguishable from the right ventricular wall (black arrow). (b) A four-chamber image demonstrates the metastatic lesion (black arrow) in the right ventricle. The patient presented with several metastases, and two of them are clearly seen in the posterior right lung (white arrows).
 
Figure 4. Cardiac magnetic resonance imaging of a 28-year-old male patient with a history of non-Hodgkin lymphoma. The patient was treated with both chemotherapy and radiotherapy in childhood and was later diagnosed with high-grade iliopsoas leiomyosarcoma. The images showed the same lesion at different sequences and projections, which is a mass arising from the interventricular septum with a broad base extending into the right ventricular chamber (arrows), consistent with a metastasis from the leiomyosarcoma. (a) Short-axis cine steady-state free precession gradient echo sequence. (b) Short-axis T1-weighted double inversion black-blood turbo spin echo sequence. (c) Axial steady-state free precession gradient echo sequence. (d) Axial T1-weighted double inversion black-blood turbo spin echo sequence.
 
Figure 5. Cardiac magnetic resonance imaging of a 49-year-old female patient with a diagnosis of gastric carcinoma with multiple bone metastases. The images showed the same lesion at different sequences, which is a mass attached to the left atrial wall and extending into the atrial cavity that was considered to be gastric cell carcinoma metastasis (white arrows). Bilateral pleural effusions and left costal metastases (green and red arrows) are shown in (c). (a) Four-chamber cine steady-state free precession gradient echo sequence. (b) Two-chamber cine steady-state free precession gradient echo sequence. (c) Axial T2-weighted triple inversion turbo spin echo sequence.
 
Figure 6. Cardiac magnetic resonance imaging of a 66-year-old male patient with known gastric cancer who was referred to the hospital with symptoms of right heart failure revealed the presence of a right ventricular mass from a gastric cancer metastasis. Two-chamber (a), dark blood four-chamber (b), pre-contrast four-chamber (c), and post-contrast four-chamber (d) images showing the mass located at the right ventricular apex extending into the right ventricular space (arrows), not only filling most of the right ventricle but also infiltrating the interventricular septum at the apical and mid anteroinferoseptal levels.
 
Figure 7. Four-chamber magnetic resonance images at two sequences revealing the left ventricular myocardial infiltration (arrows) in a patient with leukaemia. (a) Four-chamber T2-weighted triple inversion turbo spin echo sequence. (b) Four-chamber phase-sensitive inversion recovery sequence with late gadolinium enhancement.
 
Figure 8. Cardiac magnetic resonance imaging at different sequences of a 24-year-old male patient with a diagnosis of nasopharyngeal carcinoma. Two-chamber planes showed an irregular expansion of the inferior wall of the left ventricle (arrows in a-c). (d) The short axis image demonstrates asymmetrical involvement of the left ventricle by the metastatic mass (arrow). The patient also had bilateral adrenal and lung metastases (not shown). (a) Two-chamber cine steady-state free precession gradient echo sequence. (b) Two-chamber T1-weighted double inversion black-blood turbo spin echo sequence. (c) Two-chamber phase-sensitive inversion recovery sequence with late gadolinium enhancement. (d) Short-axis phase-sensitive inversion recovery sequence with late gadolinium enhancement.
 
Local Extension
 
Locally aggressive tumours can directly extend into the pericardium and cause frank invasion.[2] This typically occurs in patients with massive lung carcinomas; however, oesophageal carcinomas and mediastinal lymphomas may also directly invade the heart due to anatomical proximity.[9] Figures 9 and 10 show a central primary lung carcinoma invading the pericardium and myocardium. A large neuroblastoma in the thoracic cavity invading the heart is shown in Figure 11. Similarly, a mediastinal teratoma involving the pericardium is shown in Figure 12. Large masses occurring in organs close to the heart may involve the heart via anatomical proximity as shown in Figure 13.
 
Figure 9. Cardiac magnetic resonance images showing the same lesion at two sequences, which is a central lung mass (arrows) with a broad base and direct invasion to the heart causing loss of the normal myocardial signal at the outermost part of the left ventricle. The patient had no cardiac symptoms. (a) Short-axis cine steady-state free precession gradient echo sequence. (b) Short-axis phase-sensitive inversion recovery sequence with late gadolinium enhancement.
 
Figure 10. Cardiac magnetic resonance image of a 67-year-old male patient who presented with a cough that had lasted for >6 months. Computed tomography of the thorax showed a central lung mass (not shown) invading the heart. The image depicted the lesion (white arrow) and its extension through the pericardium and myocardium. A pericardial effusion, most probably due to the tumoural involvement, was also noted (red arrow).
 
Figure 11. Cardiac magnetic resonance images of a 15-year-old female patient diagnosed with intrathoracic (a) and intraabdominal (b) neuroblastoma (arrows). The tumour has invaded the right atrium (c) [arrow].
 
Figure 12. (a, b) Cardiac magnetic resonance images at two sequences of a 14-year-old male patient with a large mediastinal teratoma occupying most of the left hemithorax (arrows). The mass invaded the pericardium of the left ventricle at the level of the midlateral and midanterior myocardial segments with possible invasion of the myocardium. (a) Short-axis cine steady-state free precession gradient echo sequence. (b) Two-chamber cine steady-state free precession gradient echo sequence.
 
Figure 13. Cardiac magnetic resonance images in sagittal (a), coronal (b), and short axis (c) views of a 57-year-old female patient who was diagnosed with primary hepatic sarcoma arising from the left lobe of the liver. (a), (b), and (c) demonstrate the presence of a large mass invading the heart with atrial, vascular, and ventricular involvement, respectively (arrows).
 
Some tumours, including renal cell carcinoma and hepatocellular carcinoma, may extend into the inferior vena cava (IVC), allowing for growth into the right atrium via transvenous extension.[2] Figure 14 shows a hepatocellular carcinoma causing cardiac metastasis via the IVC. The superior vena cava may also serve as a transportation route for cancer cells to the heart, as seen with thoracic and mediastinal tumours.[13] Figures 15 and 16 demonstrate a case of invasive thymoma and a thyroid carcinoma, respectively, in which the malignant tissue arising from the thymus and thyroid gland reached the right atrium through the superior vena cava. Figure 17 shows left atrial metastatic involvement of a melanoma case through the left pulmonary vein enabling the tumour cells to reach the left atrium from the left lung mass.
 
Figure 14. Cardiac magnetic resonance image from a patient with hepatocellular carcinoma arising from the dome of the liver (arrow). The tumour invaded the hepatic vein leading to the haematogenous dissemination of the cells via the inferior vena cava, eventually resulting in a mass that filled the right atrium.
 
Figure 15. Cardiac magnetic resonance images of a 74-year-old male patient diagnosed with invasive thymoma. Coronal (a) and sagittal (b) images showing the tumour (arrows) reaching the right atrium via the superior vena cava.
 
Figure 16. Cardiac magnetic resonance imaging of a 98-year-old female patient with superior vena cava syndrome revealed a retrosternal superior mediastinal mass that was continuous with the thyroid gland. (a) The mass entered the left brachiocephalic vein, expanding its lumen (arrow). (a, b) The tumour cells travelled through the vasculature all the way to the left brachiocephalic vein and superior vena cava, reaching the right atrium (arrow in [b]). The presence of an irregular area at the distal end of the mass that showed no contrast enhancement was consistent with an extension of the tumoural mass accompanying a distally located thrombus. The retrosternal mass was biopsied under ultrasound guidance and the patient was diagnosed with thyroid carcinoma.
 
Figure 17. Cardiac magnetic resonance images of a 63-year-old female patient with a 2-year history of metastatic melanoma, which are sequential slices that clearly and continuously demonstrate the invasion of the mass. Metastatic masses in the right (white arrows) and left lung (red arrows) were seen. The mass in the left lung caused invasion of the left pulmonary vein leading to involvement of the left atrium.
 
DISCUSSION
 
Metastatic dissemination to the heart from noncardiac tumours may occur via the lymphatics, or via haematogenous routes that include both arterial and transvenous dissemination.[9] While lymphatic spread or direct invasion targets the pericardium first, myocardial or endocardial involvement is more common in haematogenous metastases for anatomical reasons.[1] [2]
 
Metastatic cardiac tumours have a poor prognosis, camouflaging themselves until a serious complication develops. The symptoms are broad and range from mild chest pain to cardiac rupture leading to sudden death. Pericardial and myocardial metastases may especially mimic acute coronary syndrome, and the onset of a new cardiac symptom in any cancer patient should be approached with the suspicion of cardiac metastases. Imaging findings of cardiac metastases are diverse.[14] There was one melanoma patient with left atrial metastasis in our cohort. It was demonstrated in a recent study of 23 patients with melanoma metastatic to the heart that although all chambers may be involved, right ventricular involvement was most common.[15] If hepatocellular carcinoma metastasises to the heart, the route is usually extension into the IVC, allowing for growth into the right atrium via transvenous access as with our cases.[16] [17] The two cases of uterine leiomyosarcoma in our patient group showed metastasis to the ventricles. However, the atria can also be involved.[18] Cardiac metastases from renal cell carcinoma are not frequently encountered and they may have varying imaging appearances.[19] [20] [21] A new cardiac symptom in a patient with a known renal cell carcinoma should alert the clinician to a possible cardiac metastasis. Although there were no examples in our patient cohort, malignant neuroendocrine tumours and benign uterine leiomyomas may also metastasise to the heart.[22] [23] [24]
 
CONCLUSION
 
Cardiac metastases are far more common than previously thought and should be taken into consideration in oncology patients presenting with a new cardiac symptom. The clinical scenario of cardiac metastases includes a variety of signs and symptoms depending on the anatomical site of the involvement. Although echocardiography is the preferred initial diagnostic modality owing to its relatively easy accessibility and availability, cardiac MRI may also provide a comprehensive visualisation of both cardiac and extracardiac involvement.
 
REFERENCES
 
1. Bussani R, De-Giorgio F, Abbate A, Silvestri F. Cardiac metastases. J Clin Pathol. 2007;60:27-34. Crossref
 
2. Goldberg AD, Blankstein R, Padera RF. Tumors metastatic to the heart. Circulation. 2013;128:1790-4. Crossref
 
3. Leja MJ, Shah DJ, Reardon MJ. Primary cardiac tumors. Tex Heart Inst J. 2011;38:261-2.
 
4. Al-Mamgani A, Baartman L, Baaijens M, de Pree I, Incrocci L, Levendag PC. Cardiac metastases. Int J Clin Oncol. 2008;13:369- 72. Crossref
 
5. Yusuf SW, Bathina JD, Qureshi S, Kaynak HE, Banchs J, Trent JC, et al. Cardiac tumors in a tertiary care cancer hospital: clinical features, echocardiographic findings, treatment and outcomes. Heart Int. 2012;7:e4. Crossref
 
6. Abraham KP, Reddy V, Gattuso P. Neoplasms metastatic to the heart: review of 3314 consecutive autopsies. Am J Cardiovasc Pathol. 1990;3:195-8.
 
7. Allen BC, Mohammed TL, Tan CD, Miller DV, Williamson EE, Kirsch JS. Metastatic melanoma to the heart. Curr Probl Diagn Radiol. 2012;41:159-64. Crossref
 
8. Chiles C, Woodard PK, Gutierrez FR, Link KM. Metastatic involvement of the heart and pericardium: CT and MR imaging. Radiographics. 2001;21:439-49. Crossref
 
9. Mankad R, Herrmann J. Cardiac tumors: echo assessment. Echo Res Pract. 2016;3:R65-77. Crossref
 
10. Kramer CM, Barkhausen J, Bucciarelli-Ducci C, Flamm SD, Kim RJ, Nagel E. Standardized cardiovascular magnetic resonance imaging (CMR) protocols: 2020 update. J Cardiovasc Magn Reason. 2020;22:17. Crossref
 
11. Velangi PS, Choo C, Chen KA, Kazmirczak F, Nijjar PS, Farzaneh-Far A, et al. Long-term embolic outcomes after detection of left ventricular thrombus by late gadolinium enhancement cardiovascular magnetic resonance imaging: a matched cohort study. Circ Cardiovasc Imaging. 2019;12:e009723. Crossref
 
12. Hooks M, Okasha O, Velangi PS, Nijjar PS, Farzaneh-Far A, Shenoy C. Left ventricular thrombus on cardiovascular magnetic resonance imaging in non-ischaemic cardiomyopathy. Eur Heart J Cardiovasc Imaging. 2021;22:1425-33. Crossref
 
13. Lewis MA, Hendrickson AW, Moynihan TJ. Oncologic emergencies: pathophysiology, presentation, diagnosis, and treatment. CA Cancer J Clin. 2011;61:287-314. Crossref
 
14. Castro-Martín JJ, Di Silvestre-Alonso MA, Rivero-García M, Muñoz-Rodríguez R, Izquierdo-Gómez MM, Baeza-Garzón F, et al. Magnetic resonance imaging in the study of cardiac masses: a case series. Medicina (Kaunas). 2023;59:705. Crossref
 
15. Balinski AM, Vasbinder AL, Kerndt CC, Catalan TC, Parry NP, Rehman RA, et al. Metastatic melanoma of the heart: retrospective cohort study and systematic review of prevalence, clinical characteristics, and outcomes. Cancer Med. 2023;12:2356-67. Crossref
 
16. Amin S, Shahab A, Qazi AR, Yunus H, Saeed L, Khan AA, et al. A rare incidence of hepatocellular carcinoma with tumor thrombus extending to the right heart. Cureus. 2023;15:e43965. Crossref
 
17. Sung AD, Cheng S, Moslehi J, Scully EP, Prior JM, Loscalzo J. Hepatocellular carcinoma with intracavitary cardiac involvement: a case report and review of the literature. Am J Cardiol. 2008;102:643-5. Crossref
 
18. Patel T, Aswal P, Jakhetiya A, Meena V, Pandey A. A novel case of left atrial and right lung mass turned out to be unconventional metastasis of uterine leiomyosarcoma with a review of literature. Indian J Pathol Microbiol. 2023;66:601-4. Crossref
 
19. Ali A, Shah S, Furrukh M. Renal cell carcinoma metastasis to the left atrium. Tex Heart Inst J. 2022;49:e207452. Crossref
 
20. Li Fraine S, Coman D, Durand M, Laskine M. Renal cell carcinoma with cardiac metastases. World J Oncol. 2021;12:124-6. Crossref
 
21. Romejko K, Rytel A, Rozmyslowicz T, Niemczyk S. Heart metastases of clear cell renal cell carcinoma. Diagnostics (Basel). 2023;13:1600. Crossref
 
22. Arnfield EG, Tam L, Pattison DA, Younger J, Chikatamarla VA, Wyld D, et al. Cardiac metastases from neuroendocrine neoplasms: complementary role of SSTR PET/CT and cardiac MRI. J Nucl Cardiol. 2023;30:2676-91. Crossref
 
23. El Ghannudi S, Ouvrard E, Mikail N, Leroy Freschini B, Schindler TH, Imperiale A. Cutting-edge imaging of cardiac metastases from neuroendocrine tumors: lesson from a case series. Diagnostics (Basel). 2022;12:1182 Crossref
 
24. Li J, Zhu H, Hu SY, Ren SQ, Li XL. Case report: cardiac metastatic leiomyoma in an Asian female. Front Surg. 2022;9:991558. Crossref