First-line pembrolizumab (pembro) with or without lenvatinib (lenva) in patients with advanced urothelial carcinoma (LEAP-011): A phase 3, randomized, double-blind study

432

Background: Pembro monotherapy is a standard of care for advanced urothelial carcinoma (UC) and showed antitumor activity and acceptable safety when combined with lenva in the phase 1b/2 KEYNOTE-146 study. We present results of LEAP-011 (NCT03898180), a randomized, double-blind, multicenter, global, phase 3 study of first-line pembro + lenva vs pembro + placebo in pts with locally advanced or metastatic UC who are cisplatin-ineligible with PD-L1–positive tumors or are ineligible to receive platinum-based chemotherapy. Methods: Adults with histologically confirmed, locally advanced/unresectable or metastatic UC who were cisplatin-ineligible with tumors expressing PD-L1 (combined positive score ≥10) or were ineligible to receive platinum-based chemotherapy regardless of PD-L1 status were randomly assigned 1:1 to receive pembro 200 mg IV Q3W for up to 35 cycles (̃2 y) + either lenva 20 mg orally once daily or placebo. Primary end points were PFS per RECIST v1.1 and OS. The key secondary end point was ORR per RECIST v1.1. An independent data monitoring committee (DMC) regularly reviewed safety data every 3 months; for the 6th review, a nonbinding futility analysis to evaluate ORR (−1%) and PFS (HR ≥1.1) was performed. There was no futility bound for OS. Results: Of 441 randomly assigned pts, 218 were assigned to receive pembro + lenva (median age, 74 y [range, 43-93]; ECOG PS 2, 83.5%) and 223 (median age, 73 y [range, 47-92]; ECOG PS 2, 83.0%) were assigned to receive pembro + placebo. Median duration of treatment was 3.8 mo (range, 0.0-20.7) for pembro + lenva and 3.4 mo (range, 0.0-22.0) for pembro + placebo. Median PFS was 4.2 mo (95% CI, 3.8-5.9) in the pembro + lenva group and 4.0 mo (95% CI, 2.7-5.4) in the pembro + placebo group (HR, 0.91 [95% CI, 0.71-1.16]). Median OS was 11.2 mo (95% CI, 7.4-14.9) with pembro + lenva vs 13.8 mo (95% CI, 9.8-18.8) with pembro + placebo (HR, 1.25 [95% CI, 0.94-1.67]; 6-mo OS rate, 63.6% vs 70.7%). ORR was 31.2% with pembro + lenva vs 26.5% with pembro + placebo. In 436 treated pts, treatment-related AEs (TRAEs) occurred in 186 of 214 pts (86.9%) in the pembro + lenva group and in 149 of 222 pts (67.1%) in the pembro + placebo group. Grade 3-5 TRAEs occurred in 107 pts (50.0%) in the pembro + lenva group and in 62 pts (27.9%) in the pembro + placebo group. Death from a TRAE occurred in 6 pts (2.8%) in the pembro + lenva group and in 1 pt (0.5%) in the pembro + placebo group. Conclusions: The safety profile of pembro + lenva was consistent with that of previous studies; no new safety signals were observed. The benefit/risk ratio for pembro + lenva was not considered positive vs pembro + placebo in platinum-ineligible pts with advanced UC. Antitumor activity of pembro + placebo was similar to what has been reported in previous studies, and pembro monotherapy remains standard of care as first-line therapy in platinum-ineligible pts with advanced UC. Clinical trial information: NCT03898180.

Real-world treatment outcomes from nationwide ONCO-colon Turkey registry in RAS wi̇ld-type patients treated with biologics first-line metastatic colorectal cancer

e15561

Background: Efficacy of anti-angiogenic and anti-EGFR agents has been demonstrated metastatic colorectal cancer (mCRC). Real-world evidence is especially important to detect the findings of patients outside of clinical trials. It complements together with clinical trials. However, there are a few studies that evaluated these treatments with biologics in the real-world setting. Recognizing the change that has occurred over the years will also shed light on future approaches. Therefore, we aimed to investigate the real-world data of patients with RAS-wild type mCRC. Methods: Medical records from 28 centers were collected for patients diagnosed with RAS wild-type mCRC between January 2016 and April 2019 and were included into the study. Histopathological, molecular and clinical characteristics of the patients were recorded. The treatment duration, response rate, progression-free survival and safety results were determined. Also, changes over the years were compared. Patients were compared according to the first-line biological treatments as anti-EGFR group (Group A and B) (panitumumab and cetuximab) and anti-VEGF group (group C). Results: Patients with KRAS mutant type were 43,6% and 6.1% patients were NRAS mutant type. A total of 1064 patients with documented RAS wild-type status were evaluated. 33%, 37% and 30% of all first line patients were treated with regimen including panitumumab, cetuximab and anti-VEGF, respectively. The median follow-up time was 24 (1-59) months. Median age was 61 (17-88) years. Thirty-five percent of the patients were female. Twenty percent of the patients had a right-sided colon tumor. Patients received median 6 cycles of treatment. Also, responded patients received median 6 cycles of treatment as maintenance treatment with biologics plus fluoropyrimidine. Overall response rate was 46,4%, 41,9% and 41,5% in A, B and C group respectively (p = 0,170). The median OS was 26, 27, and 23 months in A, B and C group respectively (p = 0.044). The median PFS of the patients in first-line setting that received panitumumab, cetuximab and bevacizumab were 11.6 (SE:0,6; 95% CI: 10.4-12.7), 11.0 (SE:0,5; 95% CI: 9.9-12.0), and 9.6 (SE:0,4; 95% CI: 8.8-10.4) months respectively (p = 0.012). In univariate analysis, female gender (p = 0.030), left sided tumors(p = 0.001), ECOG performance status (PS) 0-1 (p = 0.001), normal CEA level at initial diagnosis(p = 0.001) and treatment with anti-EGFR agents(p = 0.016) were found as favorable factors. PS 0-1 and normal CEA level at initial diagnosis were found as independent prognostic factors in multivariate analysis (p = 0.049, p = 0.031 respectively). Conclusions: This analysis of real-world data confirms the comparable efficacy of anti-EGFR agents in RAS-wild type mCRC. However, anti-EGFR treatment provides PFS and OS advantage when compared with anti-VEGF treatment in these patients.

Health-related quality of life for pembrolizumab (pembro) plus ipilimumab (ipi) versus pembro plus placebo in patients with metastatic NSCLC with PD-L1 tumor proportion score ≥ 50%: KEYNOTE-598

9038

Background: In the phase 3 KEYNOTE-598 study (NCT03302234), OS (HR, 1.08; 95% CI, 0.85–1.37; P = 0.74) and PFS (1.06; 95% CI, 0.86–1.30; P = 0.72) were not improved for pembro + ipi vs pembro + placebo in patients (pts) with previously untreated metastatic NSCLC with PD-L1 tumor proportion score (TPS) ≥50% and without EGFR/ALK genomic alterations. Incidence of treatment-related grade 3–5 AEs, fatal AEs, and AEs leading to discontinuation was higher with pembro + ipi vs pembro + placebo. We present prespecified patient-reported outcome (PRO) analyses from KEYNOTE-598. Methods: Pts (n = 568) with previously untreated stage IV NSCLC with PD-L1 TPS ≥50% were randomized 1:1 to pembro 200 mg Q3W for up to 35 cycles + ipi 1 mg/kg or placebo Q6W for up to 18 cycles. The EORTC QLQ-C30, QLQ-LC13, EQ-5D-5L, and NSCLC-SAQ were administered at cycles 1‒7, then every 3 cycles through cycle 19, and every 4 cycles until PD or a maximum of 35 cycles. Change from baseline in global health status (GHS)/quality of life (QoL) score from the QLQ-C30 and the time to true deterioration (TTD) in the composite endpoint of cough (LC13), chest pain (LC13), or dyspnea (C30) were secondary objectives in KEYNOTE-598. Change from baseline in GHS/QoL was analyzed using a constrained longitudinal data analysis model with missing at random assumption. Difference in TTD was evaluated using a Cox proportional hazards model and stratified log-rank test. PROs were analyzed in all pts who completed ≥1 PRO assessment and received ≥1 dose of study treatment. P values are two-sided and nominal. Results: As of data cutoff (Sept 1, 2020), PRO analyses included 280 pts in the pembro + ipi group and 280 pts in the pembro + placebo group. QLQ-C30 completion rates were 95.7% in the pembro + ipi group vs 96.1% in the pembro + placebo group at baseline and 63.6% vs 70.0% at week 18. QLQ-LC13 completion rates were 95.4% vs 96.4% at baseline and 63.6% vs 69.6% at week 18. Mean QLQ-C30 GHS/QoL scores at baseline were 62.8 in the pembro + ipi group and 64.2 in the pembro + placebo group and were similar between the groups across the follow-up period. Least squares (LS) mean (95% CI) change from baseline to week 18 in GHS/QoL scores was improved in both groups (pembro + ipi: 3.7 [0.9‒6.5]; pembro + placebo: 4.1 [1.4‒6.9]), with no significant difference between groups (LS mean difference −0.4 [−4.0 to 3.1], P = 0.82). Median TTD in composite of cough, chest pain, or dyspnea was not reached (NR; 95% CI, 13.0 mo–NR) in the pembro + ipi group vs 20.0 (95% CI, 12.7–NR) mo in the pembro + placebo group (hazard ratio, 0.98 [95% CI, 0.74‒1.30]; P = 0.91). Conclusions: There was no difference in health-related QoL or TTD in lung cancer symptoms between pembro + ipi and pembro + placebo in pts with previously untreated metastatic NSCLC with PD-L1 TPS ≥50%. Clinical trial information: NCT03302234.