Abstract: |
AimPatients with locally advanced rectal cancer may avoid surgery if they develop a clinical complete response (cCR) to total neoadjuvant therapy (TNT). During surveillance, detection of local regrowth can be challenging. The ability of MRI to predict local regrowth is uncertain, especially when it differs from endoscopic findings. This study evaluates the ability of MRI and endoscopy to predict local regrowth in patients under surveillance for cCR after TNT.MethodAll rectal cancer patients in our institution from 2006 to 2020 who achieved cCR following TNT, and entered watch and wait (WW), with findings suspicious for local regrowth during surveillance and subsequently underwent surgery were assessed. The main outcomes were the ability of MRI to correctly identify the local regrowths and its correlation with endoscopy and pathology.Results1426 patients were diagnosed with locally advanced rectal cancer. 388 patients achieved cCR after TNT and underwent surveillance. 112 patients developed clinical and/or radiological findings suspicious of local regrowth. Of 99 patients who proceeded to surgery, 12 (12%) had a pathological complete response (pCR). For patients with an abnormal MRI only and normal endoscopy, pCR rate was 4/9 (44%). For patients with an abnormal endoscopy, pCR rates in those with a normal MRI and those with an abnormal MRI were 6/34 (18%) and 2/56 (4%), respectively (chi 2-test, p = 0.001). Normal endoscopy was significantly associated with higher odds of pCR (OR 8.2, p = 0.012) whereas normal MRI showed a non-significant association (OR 2.11, p = 0.33).ConclusionIn rectal cancer patients achieving cCR after TNT, local regrowth without detectable endoscopic findings is rare. When isolated abnormal MRI findings occur, repeat radiologic surveillance prior to deciding on surgery may minimize unnecessary surgical interventions. |