Can We Predict Complete or Major Response after Chemoradiotherapy for Rectal Cancer by Noninvasive Methods? Results of a Prospective Study on 61 Patients

Abstract

Rectal preservation has been proposed as an alternative to radical resection in patients with presumed complete or major response to chemoradiotherapy (CRT). The aim of this prospective study was to evaluate the accuracy of digital rectal examination (DRE) and magnetic resonance imaging (MRI) to predict major or complete rectal cancer response to CRT. Over 2 years, 61 patients underwent radical resection after CRT for rectal cancer. DRE and MRI were carried out before and 6 to 8 weeks after the end of CRT. Data from DRE and MRI post-CRT were compared with pathological examinations. At pathological examination, major/complete responses were recorded for tumors classified ypT1N0 and ypT0N0, respectively. DRE post-CRT showed major/complete response in 26 cases, of which 14 (54%) were confirmed by pathology. The positive (PPV) and negative (NPV) predictive values of DRE to predict major/complete response were 54 and 88 per cent, respectively. MRI post-CRT showed major/complete response in 12 cases, of which nine (75%) were confirmed by pathology. The PPV and NPV of MRI to predict major/complete response were 75 and 82 per cent, respectively. Data from DRE and RMI post-CRT were concordant in 45 patients. The PPV and NPV of concordant DRE and MRI to predict major/complete response were 82 and 91 per cent, respectively. DRE and MRI do not appear to be sufficiently accurate for safe selection of patients appropriate for a rectum-sparing strategy because the risk of leaving an invasive tumor untreated is 18 per cent.