Purpose

To evaluate the performance of an abbreviated MR protocol implemented at our centre for liver surveillance in high risk melanoma patients.

Material and methods

Patients diagnosed with melanoma considered at high risk for liver metastases were screened using a rapid non-contrast MRI protocol from August 2014, until January 2019.

Imaging was performed using a 1.5T GE SIGNA Explorer and a 3T Philips Ingenia.

Axial fat-suppressed 3D T1, axial fat-suppressed T2, and coronal T2 sequences were performed, together with diffusion weighted imaging (b=0, 50, 400, 800).

Scans were reported by two board-certified radiologists and categorized into “No evidence of hepatic metastases” or “Likely hepatic metastases”.

Results

Our cohort included 77 MRIs performed at 6-12 month intervals in a total of 35 melanoma patients. Melanoma was ocular in 26, cutaneous in 8 and anorectal in 1 patient.

Metastases were found in 8 ocular melanoma patients and 2 cutaneous melanoma patients.

Consensus was always reached between both observers, without the need for a full protocol contrast-enhanced MRI and confirmed by image-guided biopsy in all cases with suspected metastases.

2 patients underwent radioembolisation, 2 thermal ablation, 1 trans-arterial embolisation, and the remainder received immunotherapy/chemotherapy.

Conclusion

The use of abbreviated MRI protocol in screening for melanoma liver metastases is feasible, cost-effective and should play a role in the management of such patients. Larger scale trials regarding such protocols should be performed to test their efficacy in different clinical scenarios and compare them to other imaging modalities.