Circulating tumor (ct)DNA-guided adjuvant atezolizumab (atezo) in muscle-invasive bladder cancer (MIBC): Exploratory analysis of ctDNA dynamics in the IMvigor011 trial.

633 Background: In the randomized, double-blind, Phase 3 IMvigor011 trial (NCT04660344), ctDNA-guided adjuvant atezo led to statistically significant improvements in investigator-assessed disease-free survival (INV-DFS) and overall survival (OS) vs placebo (pbo) after radical cystectomy (RC) in patients (pts) with MIBC. A higher proportion of pts in the atezo arm cleared ctDNA at either Cycle (C)3 or C5; all pts in the pbo arm who cleared ctDNA had low baseline concentrations (Powles, New Engl J Med 2025). Here we report exploratory analyses of ctDNA timing and concentration in ctDNA-positive (+) pts in surveillance, and on-treatment ctDNA dynamics in randomized pts, including correlation with clinical outcomes. Methods: Pts with MIBC and no radiographic disease were enrolled in surveillance within 6–24 weeks (wk) of RC. Serial ctDNA monitoring was done every 6 wk with imaging every 12 wk for up to 1 year post-RC. Pts who tested ctDNA+ at any time were screened for the treatment phase. Eligible pts were randomized 2:1 to atezo or pbo every 4 wk for 12 cycles or up to 1 year. Time to first ctDNA+ result and baseline ctDNA concentration were assessed in all pts who tested ctDNA+ during surveillance (n = 379); correlation with INV-DFS was evaluated in pts who did not receive active adjuvant treatment (n = 212). On-treatment ctDNA dynamics, including association with INV-DFS and OS, were evaluated in pts who received ≥1 dose of atezo (n = 165) or pbo (n = 83). Results: In pts who tested ctDNA+ during surveillance, the concentration of ctDNA at time of first + test ranged from 0.01–613.8 mean tumor molecules (MTM)/mL (median, 0.27 MTM/mL). The majority of pts (n = 281, 74.1%) tested ctDNA+ within 24 wk of RC. Early ctDNA-positivity and higher baseline ctDNA concentration were associated with higher tumor stage and/or node+ status at RC, as well as shorter INV-DFS. Late ctDNA-positivity was associated with low baseline ctDNA concentration (R = −0.28; P < 0.001). In randomized pts, atezo decreased ctDNA levels between C1 and C3 (log2 fold-change [FC] –0.4), while levels increased on pbo (log2FC 0.5; Wilcoxon P = 0.009). ctDNA clearance at C3 or C5 was associated with improved DFS and OS; landmark DFS and OS rates were numerically higher in atezo-treated pts (Table). Conclusions: For most pts identified as ctDNA+ via serial ctDNA monitoring after RC, positivity was evident ≤24 wk. Early positivity and high baseline concentrations were negative prognostic factors. Atezo treatment prevented the rapid increase in ctDNA levels observed in the pbo arm and led to enhanced ctDNA clearance, which was associated with superior clinical outcomes. Clinical trial information: NCT04660344 . Cleared ctDNA at C3 or C5 Did not clear ctDNA at C3 or C5 Atezo (n=42) Pbo (n=12) Atezo (n=123) Pbo (n=71) DFS, % 6 months (mo) 97.6 91.7 54.7 42.0 12 mo 84.4 62.9 31.2 23.7 OS, % 12 mo 97.1 100 80.8 65.4 24 mo 86.2 85.7 54.9 41.5 Median follow-up: 16.1 mo