Is Pembrolizumab Appropriate Before EGFR TKI Therapy in Lung Cancer?

Immune checkpoint inhibition with pembrolizumab prior to EGFR tyrosine kinase inhibitor (TKI) therapy might be unsafe for TKI-naive patients with EGFR-mutant non–small-cell lung cancer (NSCLC), reported researchers of an open-label, uncontrolled, single-cohort phase II study (abstract 9014). The findings were presented at the 2018 American Society of Clinical Oncology (ASCO) Annual Meeting, held June 1–5 in Chicago.

An objective response rate of 0% in the 10 programmed death ligand 1 (PD-L1)–positive patients with EGFR-mutant NSCLC “suggests pembrolizumab prior to an EGFR TKI is not an appropriate treatment strategy,” said lead study author Aaron Elliott Lisberg, MD, of David Geffen School of Medicine at UCLA, Los Angeles, and coauthors, in a poster presentation.

The small number of patients (n = 11) in the study “precludes definitive conclusions” regarding the safety of administering pembrolizumab before EGFR TKI therapy, the authors cautioned.

But 3 of the 11 patients experienced adverse events on pembrolizumab or subsequent TKI therapy that were “particularly concerning,” the team reported. One patient experienced grade 3 pembrolizumab-related transaminitis, leading to treatment discontinuation. Another experienced grade 2 pembrolizumab-related adrenal insufficiency that persisted after disease progression. A third patient died of erlotinib-related pneumonitis. The researchers expressed concern that prior pembrolizumab may have contributed to this patient’s death.

Only 1 of the 10 EGFR-mutant patients experienced tumor regression greater than 10% with the pembrolizumab plus TKI treatment regimen.

“Our single-center experience on the KEYNOTE-001 trial suggested that pembrolizumab-treated EGFR-mutant patients, who were TKI naive, had superior clinical outcomes to those previously treated with a TKI,” the study authors reported. “As TKI-naive EGFR mutants have generally been excluded from pembrolizumab studies, data to guide treatment decisions in this patient population are lacking, particularly in patients with PD-L1 expression ≥ 50%.”

Among the 30 patients with EGFR-mutant NSCLC treated at UCLA in the KEYNOTE-001 trial, four TKI-naive patients had significantly longer median progression-free survival (PFS) and overall survival (OS) than the 26 who had received TKIs prior to pembrolizumab (median PFS, 157.5 vs 56 days; median OS, 559 vs 120 days).

To further explore this possible sequencing benefit, in 2017, the researchers enrolled 11 TKI-naive patients with advanced EGFR-mutant NSCLC and PD-L1 expression levels of at least 1%. The patients received 200 mg of pembrolizumab every 3 weeks until tumor progression. After disease progression, they were treated with EGFR-targeted TKIs and evaluated for TKI efficacy and safety following pembrolizumab.

“Seventy-one percent (5/7) of patients treated with an EGFR TKI after pembrolizumab remained on TKI therapy at the time of data cutoff, suggesting TKI efficacy is not affected by preceding PD-1 axis inhibition, but durability of this efficacy is unknown, given the short median duration of TKI therapy in these patients (109 days),” the authors concluded.