O11-5 Durvalumab(D) ± tremelimumab(T) + platinum-etoposide(EP) in 1L ES-SCLC: Characterization of long-term benefit in CASPIAN

In the phase 3 CASPIAN trial, 1L D+EP significantly improved OS vs EP (HR 0.73 [95% CI 0.59-0.91; p=0.0047]) in pts with ES-SCLC, with sustained benefit after >2 yr median follow-up (HR 0.75 [95% CI 0.62-0.91; nominal p=0.0032]). Landmark analyses indicated 22% of pts were alive at 24m with the addition of D±T to EP. Here we assess the clinical characteristics and outcomes of pts deriving long-term benefit, as well as the relationship between TMB and efficacy outcomes in the ITT population. 805 pts with ES-SCLC were randomised 1:1:1 to D+EP, D+T+EP, or EP. Exploratory subgroup analyses defined long-term clinical benefit as PFS ≥12m. Tumour tissue was mandated at screening, if available. TMB was assessed in tissue (tTMB) using the FoundationOne CDx platform. 45 (17%), 42 (16%), and 12 (5%) pts treated with D+EP, D+T+EP, and EP had PFS ≥12m, respectively (data cutoff 27 Jan 2020). In all arms, the PFS ≥12m subgroup had a higher incidence of favorable prognostic factors (more women and pts with PS 0, fewer pts with brain/liver metastases). In the D+EP arm, pts with PFS ≥12m received more D (median 25 vs 7 cycles) and had improved ORR (96% vs 63%), median DoR (NR vs 4m) and OS at 24m (77% vs 11%) compared with the PFS <12m subgroup. Similar results were observed with EP and when both IO arms were combined. Safety and additional efficacy outcomes in the subgroups will be presented. Across all 3 arms, 283 pts (35% of ITT) were evaluable for tTMB. tTMB was not predictive of a differential treatment effect for D±T+EP vs EP (OS, PFS, or ORR). Across all arms, pts with PFS ≥12m had exceptional 2 yr OS rates >75%, despite some having poor prognostic factors such as baseline brain or liver metastases. There were >3 times more pts deriving long-term benefit when treated with durvalumab + EP vs EP alone. Further investigation into predictive factors for long-term benefit with durvalumab is ongoing.