Does MRI Restaging of Rectal Cancer After Chemoradiotherapy Actually Permit a Change in Surgical Management?
We read with interest the recent publication by Battersby et al based on data derivative of the Magnetic Resonance Imaging and Rectal Cancer European Equivalence study.1 In the abstract, this study highlights the importance of Magnetic resonance imaging (MRI) in the postneoadjuvant setting. The way that we interpret these data, however, is that they show that in its current form, even in a world center of excellence, MRI is still a limited tool for restaging patients after neoadjuvant chemoradiation, and thus is of limited clinical utility.
As an example, of the patients who were predicted to have a positive CRM on posttreatment MRI, true CRM positivity was found in only 19%, that is, MRI was 81% inaccurate in showing down-staging in this group. Interestingly, from within this entire group, only 17% of patients underwent sphincter-preserving surgery. As it stands, therefore, the vast majority of these patients were probably overtreated from an oncological perspective.
It is also interesting to note that of the 33 patients who were initially predicted to have an “unsafe” CRM that became MRI “safe” after neoadjuvant chemoradiotherapy (CRT), none had an involved pathological circumferential resection margin; yet only 30% actually underwent sphincter-saving surgery, thus potentially negating the potential change in surgical strategy offered by down-staging. If the radiological technique described in this study truly does approach 100% accuracy at predicting a negative CRM as the data suggest, and if reproducible in other centers, then this is of extreme clinical importance. To this end, we feel that it would be very helpful if the authors could more fully describe the exact manner in which they determine “good response.” It seems, from the manuscript, that this is a composite endpoint based on both T stage and magnetic resonance tumour regression grade (mrTRG), which the authors have not really fully described in their methodology, nor have they fully described before to the best of our knowledge.2,3 For example, if the tumor appears to be yT2, but there is poor regression, do they still call that margin involved? Can the authors please provide the sensitivity and specificity of both yT stage and mrTRG alone and in combination? Should this technique become standard of care in light of the poor utility of MRyT stage alone at predicting ypT stage in previous studies, including the one conducted in our own unit?4–6
Can the authors please also state how they suggest that surgeons use post-CRT MRI to counsel male patients with persistent unsafe positive margins? In the graphic of Table 6, it suggests that 60% of extramural vascular invasion positive anterior tumors <4 cm from the anal verge would be predicted to have a positive margin by post-CRT MRI; however, in the results, it seems that the pathological correlate of this is that only 20% of predicted “unsafe” CRMs are truly positive. Who should be offered an anterior exenteration?
It is also notable that the authors do not seem to routinely perform restaging MRI if patients do not have a threatened margin at baseline scan, but after CRT, there was nearly an 8% involved margin rate. Do the authors feel that they should in fact rescan after CRT in light of these findings? Also, at the other end of the scale, from an anatomical and surgical perspective, predicted MRI CRM positivity of mesorectal fascia or internal sphincter on MRI for tumors that involve the levator plate and internal sphincter is irrelevant for patients who are planned to undergo an extralevator abdominoperineal excision, because it removes these structures in their entirety with wide margins anyway.