Value of PD-L1 testing in NSCLC debated

By Shreeya Nanda

Two viewpoint articles published in JAMA Oncology provide opposing opinions on the benefit of testing for programmed death-ligand 1 (PD-L1) expression in patients with non-small-cell lung cancer (NSCLC) treated with checkpoint inhibitors.

Nathan Pennell, from Cleveland Clinic Taussig Cancer Institute in Ohio, USA, believes that PD-L1 test results should not direct the use of checkpoint inhibitors in this patient population.

He cites the conflicting results of the CheckMate 017 and CheckMate 057 trials comparing the programmed death 1 (PD-1) inhibitor nivolumab with docetaxel in patients with squamous and nonsquamous NSCLC, respectively, to support his view.

The CheckMate 017 trial found no association between PD-L1 expression and survival, with both high- and low-expressing patients deriving a significant overall survival (OS) benefit with nivolumab relative to docetaxel.

In the CheckMate 057 trial, the magnitude of benefit with nivolumab was greater for participants with higher (≥1%, 5% or 10%) than lower levels of PD-L1 expression.

However, nivolumab-treated PD-L1-negative (<1%) patients participating in CheckMate 057 had a comparable OS and overall response rate to those given docetaxel, and "there was no evidence that these patients did worse with nivolumab", says Pennell.

Highlighting that the safety profile also favoured nivolumab in CheckMate 057, he believes that "there is no reason to exclude patients without PD-L1 expression from treatment with nivolumab and thus no real rationale for routinely testing for the biomarker."

"Patients with previously treated NSCLC can simply be treated with second-line nivolumab, reserving docetaxel for third-line treatment."

However, Aaron Lisberg and Edward Garon, from the University of California-Los Angeles in the USA, think that the value of PD-L1 testing lies in identifying "the appropriate timing and use of" anti-PD-1 therapy.

Given the lack of a uniform assay and the dynamic nature of PD-L1 expression, they agree that testing for the biomarker is "not precise enough to exclude patients from receiving therapy". But they believe that "it is sufficiently informative to help select patients for second-line therapy with PD-1 inhibitors vs deferring PD-1 inhibitor therapy to later-line therapy."

And Lisberg and Garon conclude: "Until we are able to predict with confidence which patients will not respond to therapy, PD-L1 testing will remain a valuable tool that clinicians should use to identify which patients are most likely to respond to therapy, allowing for the proper sequencing of treatments."

Licensed from medwireNews with permission from Springer Healthcare Ltd. ©Springer Healthcare Ltd. All rights reserved. Neither of these parties endorse or recommend any commercial products, services, or equipment.