Background: IMvigor130 evaluated atezo + plt/gem (Arm A) vs placebo + plt/gem (Arm C) as 1L mUC treatment, allowing pts with at least SD to continue atezo or placebo “maintenance” after 4-6 cycles of plt/gem. The primary analysis showed statistically significant PFS benefit for Arm A vs C (Galsky Lancet 2020). The 1st interim OS data for Arm A vs C were encouraging but immature. This exploratory analysis used 2nd interim OS data (ITT median follow-up, 13.3 mo) to examine OS outcomes by response during atezo or placebo + chemo “induction” in Arm A vs C.

Methods: Pts were randomized 1:1:1 to Arm A, B (atezo monotherapy) or C. In Arms A and C, chemo was gem + investigator's choice of cis or carbo. Co-primary EPs of PFS and OS in Arm A vs C (ITT) and OS in Arm B vs C (ITT and IC2/3) were tested hierarchically. This post hoc analysis evaluated post-induction (post-wk 18) OS in pts who completed 4-6 (but not >6) cycles of chemo without PD during induction (through wk 18), followed by ≥1 dose of atezo or placebo maintenance. Post-PD OS was evaluated in pts with PD during induction. HRs were adjusted to account for potential imbalances in known prognostic factors between arms.

Results: Efficacy data are in the Table. Post-induction OS HR (95% CI) for pts with no PD was 0.86 (0.64, 1.16) (cis: 0.60 [0.35, 1.03], carbo: 0.97 [0.67, 1.41]). Post-progression OS HR for pts with PD was 0.74 (0.53, 1.03) (cis: 0.52 [0.28, 0.96], carbo: 0.78 [0.51, 1.18]).

Conclusions: In this exploratory analysis from IMvigor130, OS benefit with atezo + plt/gem vs placebo + plt/gem in both pts with and without PD during induction was consistent with ITT results from the 1st interim analysis. Preliminary data suggest that cis-treated pts may derive greater OS benefit from adding atezo to chemo than carbo-treated pts, which warrants further investigation. OS follow-up is ongoing.

IMvigor130 efficacy (second interim OS analysis)

No PD during inductiona (post-induction OS)bPD during induction (post-progression OS)
Arm A (n=166)Arm C (n=152)Arm A (n=91)Arm C (n=93)
OS events, n (%) 
ITT 97 (58.4) 92 (60.5) 82 (90.1) 84 (90.3) 
Cisplatinc 27/57 (47.4) 35/59 (59.3) 24/29 (82.8) 30/33 (90.9) 
Carboplatinc 70/109 (64.2) 57/93 (61.3) 58/62 (93.5) 54/60 (90.0) 
Median OS (95% CI), mo 
ITT 20.5 (17.6, 25.9) 18.8 (14.4, 21.9) 4.3 (3.5, 5.8) 3.3 (2.5, 4.4) 
Cisplatin 28.2 (19.8, NE) 19.9 (11.0, 26.5) 6.6 (4.8, 10.1) 2.5 (1.6, 5.2) 
Carboplatin 18.5 (12.9, 21.6) 18.8 (12.9, 22.8) 3.8 (3.2, 5.6) 3.4 (2.6, 4.5) 
OS HR (95% CI)d 
ITT 0.86 (0.64, 1.16) 0.74 (0.53, 1.03) 
Cisplatin 0.60 (0.35, 1.03) 0.52 (0.28, 0.96) 
Carboplatin 0.97 (0.67, 1.41) 0.78 (0.51, 1.18) 
No PD during inductiona (post-induction OS)bPD during induction (post-progression OS)
Arm A (n=166)Arm C (n=152)Arm A (n=91)Arm C (n=93)
OS events, n (%) 
ITT 97 (58.4) 92 (60.5) 82 (90.1) 84 (90.3) 
Cisplatinc 27/57 (47.4) 35/59 (59.3) 24/29 (82.8) 30/33 (90.9) 
Carboplatinc 70/109 (64.2) 57/93 (61.3) 58/62 (93.5) 54/60 (90.0) 
Median OS (95% CI), mo 
ITT 20.5 (17.6, 25.9) 18.8 (14.4, 21.9) 4.3 (3.5, 5.8) 3.3 (2.5, 4.4) 
Cisplatin 28.2 (19.8, NE) 19.9 (11.0, 26.5) 6.6 (4.8, 10.1) 2.5 (1.6, 5.2) 
Carboplatin 18.5 (12.9, 21.6) 18.8 (12.9, 22.8) 3.8 (3.2, 5.6) 3.4 (2.6, 4.5) 
OS HR (95% CI)d 
ITT 0.86 (0.64, 1.16) 0.74 (0.53, 1.03) 
Cisplatin 0.60 (0.35, 1.03) 0.52 (0.28, 0.96) 
Carboplatin 0.97 (0.67, 1.41) 0.78 (0.51, 1.18) 

Data cutoff: June 14, 2020. 1L, first-line; carbo, carboplatin; chemo, chemotherapy; cis, cisplatin; CR, complete response; EP, endpoint; HR; hazard ratio; ITT, intention to treat; mets, metastases; PD, progressive disease; PFS, progression-free survival; PR, partial response; SD, stable disease. a Pts who achieved at least SD (CR, PR, or SD without PD) during induction. b Post-week 18 OS. c Denominators refer to the number of pts in each chemotherapy subgroup/treatment arm. d Multivariable Cox proportional hazards model stratified by enrollment stage (1/2); adjusted for age, ECOG PS (0, 1, 2), cisplatin ineligibility (Y/N), liver mets (Y/N), lymph node mets only (Y/N), ≥3 metastatic sites (Y/N), renal impairment (Y/N), alkaline phosphatase ≥upper limit of normal (Y/N), glomerular filtration rate (<60 vs ≥60 mL/min), PD-L1 status (IC0/1, IC2/3), Bajorin risk score (0, 1, 2 and/or liver mets). For pts without PD, model was additionally adjusted for best response (CR, PR, SD) during induction.

Citation Format: Enrique Grande, Aristotelis Bamias, Matthew D. Galsky, Eiji Kikuchi, Ian D. Davis, José Ángel Arranz, Arash Rezazadeh Kalebasty, Xavier Garcia del Muro, Se Hoon Park, Ugo De Giorgi, Boris Alekseev, Marina Mencinger, Kouji Izumi, Javier Puente, Jian-Ri Li, Xiaodong Shen, Beiying Ding, Qian Zhu, Jeremy Linsenmeier, Maria De Santis. Overall survival (OS) by response during “induction” from the global, randomized Phase III IMvigor130 study of atezolizumab (atezo) + platinum/gemcitabine (plt/gem) vs placebo + plt/gem in patients (pts) with previously untreated metastatic urothelial carcinoma (mUC) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr CT187.