Magnetic Resonance Imaging–Guided Cryotherapy for Precision Tumour Ablation

Hong Kong J Radiol 2023 Sep;26(3):168-73 | Epub 2 Aug 2023

 

 

Department of Radiology and Imaging, Queen Elizabeth Hospital, Hong Kong SAR, China

 

  Full Article in PDF

Percutaneous tumour ablation is important in the treatment of both benign and malignant diseases, allowing minimally invasive treatment for tumour destruction and pain control. Percutaneous tumour ablation is frequently performed under ultrasound (US) or computed tomography (CT) guidance, both of which have significant limitations. For example, US provides limited visualisation of deep structures and is readily deflected by overlying gas and bone. CT, on the other hand, emits radiation, rendering real-time needle insertion unfavourable. Additionally, CT has limited soft tissue spatial resolution, making it difficult if not impossible to visualise the tumour and important surrounding structures. Most importantly, both techniques of imaging guidance do not allow accurate visualisation of ablation zones.[1] Magnetic resonance imaging (MRI) guidance has swiftly achieved global prominence as it provides superior soft tissue contrast, radiation-free real-time imaging during needle insertion, and, most importantly, clear visualisation of the ablation zone. Superior soft tissue spatial resolution is particularly important as it allows accurate visualisation of tumour and important adjacent structures,[1] [2] which is otherwise difficult under other imaging techniques. With these qualities, MRI guidance allows safer, more precise, and radiation-free imaging guidance of tumour ablation. Recently, the technique has been available in Hong Kong. The aim of this retrospective single-institution study is to describe the technique and evaluate its safety while highlighting its use and benefits.

 

From 1 May 2019 to 31 January 2020, 14 patients (9 males, 5 females) underwent MRI-guided cryoablation in our institution. Thirteen procedures were performed on tumours with curative intent, and one as a staged procedure to treat chest wall fibromatosis. Patients were referred for local ablative treatment either because the patient was not a suitable candidate for surgery or because the lesions could not be treated by surgery (e.g., desmoid tumours, bone metastases or lymph node metastases). Patient and lesion characteristics are summarised in the Table.

 

Table. Patient details and tumour characteristics

 

All interventions were performed in an MRI suite dedicated to interventional procedures (Aera 1.5T; Siemens Medical Systems, Erlangen, Germany) under strict aseptic technique. Procedures were performed under local lidocaine anaesthesia (n = 13) or general anaesthesia (n = 1). Cryotherapy was performed with a cryoablation system (BTG Visual-ICE; Boston Scientific, Marlborough [MA], United States) with one or more cryoprobes (IceRod MRI or IceSeed MRI; Galil Medical, Arden Hills [MN], United States) depending on lesion characteristics, with manipulation of cryoprobes to create different shapes and sizes of ablation zones, often using the synergistic effects of multiple probes. Cryoprobe insertion was performed under real-time magnetic resonance fluoroscopic guidance using a prototype interactive balanced steady-state free precession sequence implemented for interactive realtime tip tracking with interactive real-time tip tracking module (BEAT IRTTT; Siemens Medical Systems, Erlangen, Germany), which allows active adjustment of scan plane orientation depending on the needs of the operator. Occasionally, and particularly for ice ball monitoring, multiplanar T2-weighted periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER [periodically rotated overlapping parallel lines with enhanced reconstruction]) or half-Fourier acquisition single-shot turbo spin-echo (HASTE) sequences were acquired.

 

Tumour treatment was performed by ensuring that the ablation zone included the entire tumour with an additional safety margin of 5 to 10 mm. Treatment of the chest wall fibromatosis was performed in this case as a staged procedure with partial coverage of the lesion for size reduction.

 

Patient information on the Clinical Management System of Hospital Authority from 1 May 2019 to 31 January 2020 was collected, which included patient age and sex; target tumour type, location, size, and proximity to important critical structures; procedure details including mode of anaesthesia and need for hydrodissection; complications; and duration of follow-up period. The size of the target lesion is defined as the maximum diameter on preprocedural MRI on the day of the procedure. Complications were graded using the National Cancer Instituteʼs Common Terminology Criteria for Adverse Events (CTCAE) version 5.0. Patient outcome was evaluated by postoperative imaging, on clinic visits, and electronic medical records.

 

In total, 14 procedures were performed (10 renal cell carcinomas, 1 hepatocellular carcinoma, 1 adrenal metastasis, 1 external iliac lymph node metastasis, and 1 chest wall fibromatosis) from 1 May 2019 to 31 January 2020 (Table). Mean age ± standard deviation of the patients was 66 ± 14.3 years. Most procedures (n = 13) were performed with curative intent; for these cases, mean tumour diameter was 1.6 ± 5.8 cm. One non-curative procedure performed was a staged cryoablation of chest wall fibromatosis measuring 24 cm, performed for lesion size control.

 

All procedures performed with curative intent were technically successful with adequate coverage of the tumour observed during intraoperative MRI. For the patient with cryoablation of chest wall fibromatosis performed as a staged procedure, technical success was achieved with coverage of the intended region of the tumour. MRI-guided hydrodissection was performed in two patients (14.3%) due to close proximity to the colon (n = 1) and external iliac vessels (n = 1). A total of 85.7% (12 out of 14) patients had two freeze/thaw cycles, whereas 14.3% (2 out of 14) required three freeze/thaw cycles.

 

Three minor complications occurred. Two (14.3%) developed CTCAE Grade 3 complications, both developing a small area of frostbite at the needle insertion site, one of which developed before activation of the freeze cycle. Both patients recovered after a minor debridement. One patient (7%) developed a minor perinephric haematoma (CTCAE Grade 1), which recovered without intervention. No major complications were observed.

 

The mean follow-up period was 7.2 ± 3.03 months. One patient was lost to follow-up and excluded from analysis. Of the patients with malignant disease (n = 12, mean follow-up period = 7.6 months), tumour coverage with no imaging evidence of residual or recurrence was seen in 91.7% (n = 11). One patient with ablation of an external iliac lymph node showed suspicious fluorodeoxyglucose avidity in the iliac fossa on follow-up positron emission tomography–CT, undetermined but possibly residual disease. In the case of chest wall fibromatosis, lesion size reduction was seen in the first month after cryoablation. In this case, the disease that was not cryoablated also showed interval shrinkage and is softer on palpation (Figure 1), attributed to abscopal effect. No other clinical complications, such as pain, were observed in any of the patients.

 

Figure 1. (a) Left chest wall fibromatosis (arrow) in a 24-year-old woman with a history of multiple prior resections with local recurrence. (b) Magnetic resonance imaging (MRI) showing the 10 cryoprobes that were inserted into the large mass (arrow). (c and d) The ice ball is clearly identified as a signal void (asterisk) allowing accurate visualisation of the ablation zone. As the lesion was too large for complete ablation, the procedure was performed with the goal of size reduction. (e) Post-ablation MRI 1 month after surgery showing a necrotic region (arrow) with the non-ablated regions reduced in size, attributed to immune reaction from cryoablation. Clinically, the mass is softer on palpation.

 

Cryotherapy is a thermal ablative technique that causes tumour destruction by inducing cell damage by freezing and thawing. Current systems use argon and helium gases delivered via cryoprobes that induce freezing and thawing based on the Joule-Thomson effect (temperature change as an effect of rapid expansion of certain gases).[3] Cryoablation has advantages over other ablation modalities such as radiofrequency ablation due to its intrinsic analgesic effects and potential treatment of large tumours.[4] Additionally, cryotherapy allows visualisation of the ice ball, particularly under MRI, which improves predictability of tumour coverage and prevents non-target ablation.

 

MRI is particularly useful in conjunction with cryotherapy as it allows accurate visualisation of ice ball as a signal void[1] (Figure 2) as well as continuous radiation-free multiplanar ice ball monitoring,[4] allowing accurate assessment of ablation zone, which ensures tumour coverage while avoiding non-target ablation. Due to MRI’s superior soft-tissue differentiation, MRI allows better tumour visualisation and hence improved accuracy with needle positioning. Many lesions may only be seen on MRI, such as non-exophytic renal masses or hepatic dome lesions,[5] where MRI greatly improves the accuracy of needle positioning.

 

Figure 2. Clear cell renal cell carcinoma detected on (a) contrast-enhanced computed tomography (CT) [arrow] which cannot be delineated on (b) non-contrast CT (arrow), which is commonly used as guidance modality. (c) The lesion is apparent on magnetic resonance imaging due to the superior soft tissue differentiation (white arrow) allowing accurate real-time cryoprobe (dashed arrow) positioning within the tumour and avoiding critical structures. (d) The ice ball is seen as a signal void (asterisk), allowing accurate visualisation of the ablation zone, ensuring adequate tumour coverage and reduced risk of non-target ablation. (e) Post-ablation contrast-enhanced CT showing hypodensity with no evidence of enhancement (asterisk).

 

MRI-guided intervention may be performed in an open- or closed-bore system.[1] Open-bore systems allow easier access to the patient and enhanced operator flexibility but are limited by lower magnetic field strength (0.2-1 T).[1] Closed-bore systems are more desirable as they provide higher magnetic field strength, which improves imaging rate and quality,[1] [2] [3] although at the expense on operator comfort and flexibility. Such systems have specific requirements to allow for interventional use, such as a large bore width (70 cm) and short bore length (125-150 cm).[1] This is necessary to allow the interventional radiologist to stretch into the bore to work at the isocentre, allowing real-time freehand manipulation. Additional body and surface coils may be used to increase contrast resolution of images[6] but of all the procedures described, the spine or body coil within the magnet may suffice.

 

MRI-guided cryoablation has been described in the literature as a treatment option for a variety of tumours. Ahrar et al[2] described percutaneous cryotherapy of small renal tumours in 18 patients in a 1.5-T MRI. The mean follow-up period was 16.7 months, with rates of overall survival, disease-specific survival, and metastasis-free survival of 88.9%, 100%, and 100%, respectively. Similarly, MRI-guided liver cryotherapy was described to be safe and feasible. Mala et al[7] described percutaneous cryoablation of six patients with liver metastases from colorectal carcinoma using an open 0.5-T MRI, which they concluded allowed good visualisation of tumour for cryoprobe positioning in order to puncture the tumour. Shimizu et al[8] treated 16 tumours with MRI-guided cryoablation on a 0.3-T open-bore magnet with 1- and 3-year overall survival rates of 93.8% and 79.3%, respectively. The complete ablation rate was reported as 80.8% at 3 years.

 

As the indications for ablation have expanded in both malignant and benign tumours and with palliative and curative intent, MRI guidance may be the preferred imaging technique in certain lesions. MRI guidance has proven beneficial for established ablation targets including renal,[1] [2] [9] prostate,[1] [10] and soft tissue[1] tumours such as desmoids. Although less established, we find MRI guidance highly useful in liver ablation, especially in recurrent tumours and lesions located in the hepatic dome, as described by Wang et al.[5]

 

This paper describes our early experience of MRI-guided cryoablation with multiple shortcomings including a small sample size, short follow-up periods, and heterogeneity of treated lesions. Existing studies have mainly been retrospective analyses of cohorts of MRI or CT guidance in heterogeneous patient groups; controlled trials with larger sample sizes and comparison with other methods of imaging guidance will be useful.

 

MRI guidance is a safe imaging technique that allows accurate tumour and non-target organ visualisation, real-time needle puncture, and precise delineation of the ablation zone.