AtezoTRIBE: ctDNA (ctF) non-responders show larger PFS gain with PD-L1 atezolizumab add-on in 1L mCRC

In the Phase 2 AtezoTRIBE trial (NCT03721653), adding the PD-L1 inhibitor atezolizumab to first-line FOLFOXIRI/bevacizumab in untreated metastatic colorectal cancer was evaluated here with a translational liquid-biopsy analysis using the Tempus xF 105-gene cfDNA assay to measure plasma TMB (pTMB; high ≥26 mut/Mb) and circulating tumor fraction (ctF) dynamics from baseline (T0) to end of induction (T1). Baseline pTMB was evaluable in 123/155 (79%), with 16% pTMB-high (20/123); pTMB-high was prognostic for longer PFS (HR 0.69; 90%CI 0.43–1.10; p=0.095) but not predictive for atezolizumab benefit (treatment interaction p=0.871)—consistent with the broader reality that pMMR/MSS mCRC rarely benefits meaningfully from ICIs outside niche biomarker-defined pockets. Among patients with paired ctF (n=89), “molecular responders” (top 30% relative ctF reduction/clearance) were common (71%; 63/89), yet the larger relative benefit from adding atezolizumab appeared in ctF non-responders (HR 0.33; 90%CI 0.16–0.69) vs molecular responders (HR 0.76; 90%CI 0.03–3.95), with an interaction signal using the authors’ pre-specified lenient threshold (p_int=0.132; sensitivity analysis p_int=0.108). For context, first-line FOLFOXIRI/bevacizumab already delivers median PFS ~12.3 months in TRIBE vs ~9.7 months with FOLFIRI/bevacizumab (HR 0.77), so any incremental ICI effect needs to be clearly distinguishable from chemotherapy-driven ctDNA clearance.