PD-L1 testing in non-small cell lung cancer patients: a study on interlaboratory variation

Abstract

Lung cancer is the second most commonly diagnosed type of cancer worldwide. Approximately 85% of patients with lung cancer have non-small cell lung cancer (NSCLC). Since 2015, patients with advanced NSCLC can be treated with immunotherapy, which supports the patient’s own immune system in killing cancer cells by blocking the programmed death-1 (PD-1) receptor or its ligand programmed death-ligand 1 (PD-L1). If a patient qualifies for treatment with these PD-1/PD-L1 blockade agents, and if a patient should also receive chemotherapy in addition to immunotherapy, depends on the percentage of tumor cells that express PD-L1. This percentage is determined by pathologists through immunohistochemistry, which visualizes microscopic PD-L1 expression on the membrane of tumor. Given its importance in guiding treatment decisions, it is crucial that this expression is determined accurately and reproducibly.

In this thesis, we assessed the impact of various pre-analytic, analytic, and post-analytic variables on PD-L1 immunostaining results, and investigated the interlaboratory variation in PD-L1 testing in NSCLC patients in the Netherlands. We found that considerable variation in PD-L1 positivity exists between pathology laboratories in clinical practice. Most likely, interobserver variability in PD-L1 scoring plays an important part in causing this. When it comes to PD-L1 testing performed on cytology samples, which contain free cells or tissue fragments rather than whole tissues, differences in fixation methods between laboratories partially explain the interlaboratory variation in PD-L1 positivity, too. Moreover, it is likely that differences between laboratories in the use of antibodies for PD-L1 testing influence interlaboratory variation in PD-L1 positivity as well.

The degree of interlaboratory variation in PD-L1 positivity that we found is problematic for NSCLC patients, since it could lead to different PD-L1 results depending on the laboratory where their PD-L1 test is performed. With this thesis, we hope to raise awareness of this considerable interlaboratory variation in PD-L1 positivity among all clinicians involved in NSCLC patient care, and stimulate joint efforts to diminish this degree of variation.