Neoadjuvant chemotherapy plus serplulimab and immuno-response molecular subtype for pMMR locally advanced rectal cancer (FIRM).

110

Background: Immune checkpoint blockade (ICB) shows limited efficacy in mismatch repair proficient (pMMR) locally advanced rectal cancer (LARC). Recent clinical studies have shown that neoadjuvant chemoradiotherapy (nCRT) plus ICB improve response. However, radiotherapy is associated with toxicities, including radiation proctitis, anastomotic leakage, and anal dysfunction. Moreover, radiotherapy is not feasible for high position LARC. Radiotherapy-free neoadjuvant strategy warrants further exploration. The FIRM trial is the first phase 2, proof-of-concept study evaluating neoadjuvant immuno-chemotherapy (nICT) with mFOLFOX6 plus PD1 blockade for pMMR LARC. Methods: Patients (pts) with T3/4 or N+, pMMR LARC located ≤15 cm from anal verge were eligible. Pts received 6 cycles of mFOLFOX6 and serplulimab followed by surgery. Primary endpoints were pathological complete response (pCR) rate and major pathological response (MPR) rate. Enrollment of 30 pts was planned, with hypothesis of an increased pCR of 24% compared with reported data of 6.6% after mFOLFOX6 (FOWARC trial, JCO 2016). Multi-omics profiling (single cell RNAseq, bulk RNAseq, whole exome sequencing) was performed to elucidate response mechanisms and develop an immuno-response molecular subtype (IRMS) for immunotherapy stratification. Results: 30 pts were enrolled, 28 completed ≥4 cycles of nICT. 2 discontinued due to adverse events (AE). The pCR and MPR rate were 42.9% and 67.9%, respectively. Tumor regression grade (TRG) 0, 1, 2, and 3 were observed in 44.4%, 11.1%, 40.7%, and 3.7% of pts. 6 pts experienced grade 3 AEs, with no grade 4-5 AE or anastomotic leakage observed. Baseline profiling identified two novel subsets predicting nICT response: inflammatory/migratory tumor cells (IMT) and ITGAX⁺ activated B cells (IAB). Post-nICT samples showed increased enrichment of CD8⁺ T, cDC1, pDC, memory B and Tfh cells, together with reduced exhausted CD8⁺ T, Tregs, extracellular matrix deposition and tumor angiogenesis. IRMS stratified pts into immune-sensitive and immune-resistant types and showed strong predictive performance in internal cohort (AUC =0.87, sensitivity =0.74, specificity =1.0, accuracy =0.81), outperforming conventional biomarkers including CPS (AUC = 0.48) and TMB (AUC =0.47). Notably, IRMS was validated across multiple external immunotherapy cohorts (melanoma, colorectal, lung, breast cancers), showing robust predictive value (AUC =0.70-0.86). Conclusions: These findings provide the first evidence supporting immuno-chemotherapy as a promising radiotherapy-free neoadjuvant strategy for pMMR LARC. nICT converts the immune-cold tumors into immune-activated state by orchestrating a dual-pathway immune remodeling involving IMT-cDC1-CD8 and IAB-CD4 axes. The IRMS provides a high-precision, pan-cancer tool to guide patient selection and personalized immunotherapy. Clinical trial information: NCT06688786 .