Journal
Vol. 27 No. 3, 2024
Table of Contents
ORIGINAL ARTICLES
The Significance of Computed Tomography–Detected Breast Lesions
ORIGINAL ARTICLE CME
Hong Kong J Radiol 2024 Sep;27(3):e140-6 | Epub 4 September 2024
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
ORIGINAL ARTICLE CME
Hong Kong J Radiol 2024 Sep;27(3):e147-55 | Epub 16 September 2024
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
ORIGINAL ARTICLE
Hong Kong J Radiol 2024 Sep;27(3):e156-63 | Epub 28 August 2024
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
ORIGINAL ARTICLE
Hong Kong J Radiol 2024 Sep;27(3):e164-70 | Epub 10 September 2024
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
CASE REPORT
Hong Kong J Radiol 2024 Sep;27(3):171-5 | Epub 19 September 2024
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
CASE REPORT
Hong Kong J Radiol 2024 Sep;27(3):e176-82 | Epub 2 September 2024
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
PICTORIAL ESSAY
Hong Kong J Radiol 2024 Sep;27(3):e183-91 | Epub 12 September 2024
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]
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]
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
PICTORIAL ESSAY
Hong Kong J Radiol 2024 Sep;27(3):e192-201 | Epub 9 September 2024
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