BACKGROUND: The results of the GEMCAD-1402 phase II randomized trial suggested that adding aflibercept to modified FOLFOX6 (mFOLFOX6) induction, followed by chemoradiation and surgery, could increase the pathological complete response (pCR) rate in patients with high-risk, locally advanced rectal cancer. Here we update results up to 3 years of follow-up and evaluate the predictive value of consensus molecular subtypes identified with immunohistochemistry (CMS-IHC).
METHODS: Patients with magnetic resonance imaging-defined T3c-d/T4/N2 rectal adenocarcinoma in the middle or distal third were randomized to mFOLFOX6 induction, with (mF+A, N = 115) or without aflibercept (mF, N = 65), followed by capecitabine plus radiotherapy and surgery. The risk local relapse (LR), distant metastases (DM), disease-free survival (DFS), and overall survival (OS) were estimated at 3 years. Selected samples were classified via immunohistochemistry into immune-infiltrate, epithelial, or mesenchymal subtypes.
RESULTS: mF+A and mF had 3-year DFS of 75.2% (95% confidence interval [CI]: 66.1% to 82.2%) and 81.5% (95% CI: 69.8% to 89.1%), respectively; 3-year OS of 89.3% (95% CI: 82.0% to 93.8%) and 90.7% (95% CI: 80.6% to 95.7%), respectively; 3-year cumulative LR incidences of 5.2% (95% CI: 1.9% to 11.0%) and 6.1% (95% CI: 1.7% to 15.0%), respectively; and 3-year cumulative DM rates of 17.3% (95% CI : 10.9% to 25.5%) and 16.9% (95% CI: 8.7% to 28.2%), respectively. pCR were achieved in 27.5% (N = 22/80) and 0% (N = 0/10) of patients with epithelial and mesenchymal subtypes, respectively.
CONCLUSION: Adding aflibercept to mFOLFOX6 induction was not associated with improved DFS or OS. Our findings suggested that CMS-IHC subtypes could be predictive of pCR with this treatment.
