180O IMpassion132 double-blind randomised phase III trial of chemotherapy (CT) ± atezolizumab (atezo) for early-relapsing unresectable locally advanced or metastatic triple-negative breast cancer (aTNBC)

Immune checkpoint inhibitors (ICIs) improve efficacy of first-line CT for some patients (pts) with aTNBC but data in early relapse are limited. IMpassion132 (NCT03371017) enrolled pts with aTNBC relapsing <12 mo after the last of any anthracycline- and taxane-containing (neo)adjuvant CT or primary TNBC surgery. PD-L1 status was centrally assessed by SP142 before randomisation. Initially pts were enrolled irrespective of PD-L1 status. In Aug 2019 the protocol was amended to enrich for PD-L1+ (tumour immune cell ≥1%) aTNBC. Stratification factors were investigator-selected CT (capecitabine 1000 mg/m2 bid d1–14 q21d X or carboplatin AUC2 + gemcitabine 1000 mg/m2 d1 subgroup results were consistent. PFS was similar across treatment arms and populations (median ∼4 mo). ORRs were 28% with placebo + CT vs 40% with atezo + CT. Adverse events (predominantly haematological) were broadly similar between arms and as expected with atezo + CG or X in early-relapsing aTNBC.Table: 180OFinal OS analysis (prespecified after ∼247 deaths in the PD-L1+ population); median follow-up 9 monthsOS (95% CI)PD-L1+mITT*Placebo + CT (n=177)Atezo + CT (n=177)Placebo + CT (n=192)Atezo + CT (n=188)Events, n (%)128 (72)124 (70)160 (83)158 (84)Stratified hazard ratio0.93 (0.73–1.20); p=0.590.94 (0.76–1.18)Median, mo11.2 (9.0–13.3)12.1 (10.1–15.1)9.8 (8.4–12.0)10.4 (8.9–12.9)12-mo rate, %48 (40–55)50 (43–58)42 (35–50)46 (39–54)18-mo rate, %32 (25–40)34 (26–41)26 (19–32)27 (20–34)*All-comer pts randomised before Aug 2019. Open table in a new tab *All-comer pts randomised before Aug 2019. Pts with early-relapsing aTNBC have a dismal prognosis that is not improved with atezo.A biology-based definition of intrinsic resistance to ICIs in aTNBC is urgently needed to optimally treat these pts and design next-generation (combination) clinical trials.