• Sara Pilotto, Luisa Carbognin, Antonio Rossi, Giampaolo Tortora, and Emilio Bria.
• University of Verona, Azienda Ospedaliera Universitaria Integrata, P.zza L. A. Scuro, 10, Verona, Italy. Electronic address: sara.pilotto@univr.it.
• Semin. Oncol. 2018 Jun 1; 45 (3): 176-180.

AbstractPatients with nonsmall-cell lung cancers expressing high levels of PD-L1 present a therapeutic dilemma for clinicians who have to choose between pembrolizumab as a single agent or in combination with chemotherapy. In order to help them as they ponder over this decision we performed a meta-analysis using the data available from randomized clinical trials that enrolled patients with untreated advanced nonsmall-cell lung cancers with PD-L1 expression level ≥50%. We evaluated interactions according to type of treatment-add-on strategy: pembrolizumab plus chemotherapy versus chemotherapy or head-to-head strategy: pembrolizumab alone versus chemotherapy. Hazard and Odds Ratios (HR/OR) for primary (overall survival, OS) and secondary endpoints (progression-free survival, PFS and objective response rate, ORR) were extracted and cumulated by adopting a random-effect model with 95% confidence interval. Four clinical trials that enrolled 2,754 patients including 1,252 with PD-L1 expression in ≥50% of cells were examined. We did not find a significant interaction (P = 0.16) between an add-on strategy and head-to-head comparisons with pembrolizumab for OS (HRs in favor of immunotherapy of 0.50 and 0.67, respectively). A significant quantitative interaction favoring the add-on strategy was found for PFS and ORR (P < 0.001), with a HR for PFS of 0.36 with the add-on strategy and 0.65 in head-to head comparisons, and an OR for ORR of 5.35 and 1.58, respectively. In absence of planned prospective noninferiority trials addressing this issue, addition of chemotherapy to pembrolizumab appears to decrease tumor size and delay disease progression significantly more than pembrolizumab alone, but has no impact on OS. We conclude that the data support deciding between both treatment options on an individual basis by considering a patients' clinical status and disease characteristics.Copyright © 2018 Elsevier Inc. All rights reserved.

27 keywords...

hide…