Evaluation of MRI for detection and staging of relapse in stage I seminoma: Data from independent central scan review in the randomised TRial of Imaging and Surveillance in Seminoma Testis (TRISST)


Year:

Session type:

Aslam Sohaib1, Sarah Swift2, Laura Murphy3, Fay Cafferty3, Robert Huddart1, Gordon Rustin4, Dipa Noor3, Simona Wade3, Francesca Schiavone3, Elizabeth James3, Richard Kaplan3, Johnathan Joffe5
1Institute of Cancer Research (ICR), 2St James’s Hospital, 3Other, 4Mount Vernon Cancer Centre, 5Calderdale and Huddersfield NHS Foundation Trust

Abstract

Background

CT surveillance in stage I seminoma is an international standard of care, avoiding adjuvant therapy. In this young population, minimising irradiation is vital. The TRISST trial recently demonstrated non-inferiority of MRI vs CT in this setting. With MRI, relapse detection was slightly earlier with fewer advanced relapses. Here, we report results from the trial’s independent, central scan review, assessing impact of radiologist experience.

Method

TRISST was a randomised, phase III, non-inferiority trial. Eligible men from 35 UK centres had undergone orchiectomy for stage I seminoma with no adjuvant therapy. Factorial (2x2) randomisation was to MRI vs CT and more vs fewer scans; n=669, all contributing to both comparisons. Imaging was of the retroperitoneum. CT: spiral or multi-detector scanner; maximum reconstructed slice thickness 5mm; oral and intravenous contrast media injection. MRI: at least a 1 Tesla system with phased array coils and contiguous axial 5mm section T1 and T2 weighted images.

Imaging (from baseline to relapse) for relapsing patients was independently assessed by one of two reviewers. Descriptive analyses compare investigator-reported timing/size of relapse with central review.

Results

82 patients relapsed; 37/41 MRI and 34/41 CT underwent central review. 2 relapses were missed on scheduled 6-month MRI (one identified on markers at 8 months; one on 12-month MRI). 2 were missed on scheduled 6- and 24-month CT (identified on scans at 18 and 36 months respectively). Additionally, for 1 MRI and 2 CT patients with early relapse, review identified relapse on baseline CT. All 7 patients had complete response to treatment. There was one false positive on CT (review indicated diagnostic threshold not met); none on MRI. Measurement discrepancies were small and similar for MRI and CT (MRI: median 0.25mm, range -0.4 to +1.7mm; CT: 0.20mm, -0.6 to +3.5mm).

Conclusion

MRI is accurate for diagnosing relapse in seminoma; discrepancies between radiologists are uncommon and similar to those seen for CT, with no impact on outcomes. MRI is recommended to avoid irradiation.

Impact statement

This study adds further evidence to support use of MRI for surveillance in stage I seminoma, with potential to avoid harmful irradiation in these young patients.