5350 - The efficacy and safety of pembrolizumab as a first-line therapy in PD-L1 50% positive advanced NSCLC (HOPE-001)

Background

Pembrolizumab (Pem) for NSCLC and PD-L1 TPS ≥50% as a first-line therapy showed the longer PFS and OS compared with chemotherapy in some clinical trial. However, only limited patients in good general condition without organ failure can participate in them and their outcomes may not be entirely representative of real-world setting.

Methods

We conducted a multicenter retrospective study across 11 medical centers (Hanshin Oncology clinical Problem Evaluation group (HOPE)). We analyzed clinical data from NSCLC patients receiving Pem as a first-line therapy between February 1st 2017 and April 30th 2018. We aimed to evaluate the efficacy and safety and to identify which patients will become more suitable candidates for Pem monotherapy.

Results

213 patients were enrolled in this study. The median age was 71 years. Out of 213 patients, 176 (82.6%) were male, 20 (9.4%) were never smokers (Median brinkman index: 900), 172 (80.8%) had ECOG PS of 0-1, 55 (25.8%) had SQ, and PD-L1 TPS were 50-74%: 97 (45.5%), 75-89%: 55 (22.1%), and 90-100%: 69 (32.4%). 39 (18.3%) of all had AEs of grades ≥3. The most frequently severe AEs was pneumonitis (10 (4.7%) including in 1 grade 4), and no patient died of severe AEs. The overall RR/DCR were 51.2%/73.2%, the median PFS/OS was 8.3/18.4 months (M). In the univariate analysis, the ECOG PS (0-1 vs. ≥2: 9.0 vs. 4.0 M, HR: 2.11, p = 0.00061), CRP/ALB (<0.3 vs. ≥0.3: NA vs. 5.9 M, HR: 1.88, p = 0.00148), and steroid usage (not usage vs. usage: 8.7 vs. 2.0 M, HR: 3.17, p = 0.00034) were significantly correlated with PFS of Pem. In the multivariate analysis, ECOG PS (0-1 vs. ≥2: HR: 1.69, p = 0.03138), CRP/ALB (<0.3 vs. ≥0.3: HR: 1.92, p = 0.00153), steroid usage (not usage vs. usage: HR: 2.94, p = 0.00267), and PD-L1 TPS (50-89% vs. 90-100%: HR: 0.65, p = 0.04984) were significantly and independently correlated with PFS of Pem.

Conclusions

Our results was consistent with the efficacy and safety of previous key clinical trials, although our study had various backgrounds. Furthermore, poor PS, high inflammatory state (CRP/ALB≥0.3), and steroid usage at the time of Pem treatment commencement were independently correlated with a shorter PFS of Pem. On the other hands, higher PD-L1 TPS (90-100%) was independently correlated with a longer PFS of Pem.

Clinical trial identification

UMIN (University Hospital Medical Information Network in Japan; number 000032470).

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

M. Tamiya: Speaker Bureau / Expert testimony: MSD; Speaker Bureau / Expert testimony: Taiho Pharmaceutical. A. Tamiya: Speaker Bureau / Expert testimony: MSD; Speaker Bureau / Expert testimony: Taiho Pharmaceutical. Y. Taniguchi: Speaker Bureau / Expert testimony: MSD. T. Yokoyama: Speaker Bureau / Expert testimony: Taiho Phermaceutical. K. Hirano: Speaker Bureau / Expert testimony: MSD; Speaker Bureau / Expert testimony: Taiho Phermaceutical. T. Hirashima: Speaker Bureau / Expert testimony: Taiho Phermaceutical. M. Kanazu: Speaker Bureau / Expert testimony: MSD. T. Kumagai: Speaker Bureau / Expert testimony: MSD; Speaker Bureau / Expert testimony: Taiho Phermaceutical. D. Fujimoto: Speaker Bureau / Expert testimony: MSD; Speaker Bureau / Expert testimony: Taiho Phermaceutical. All other authors have declared no conflicts of interest.