TOPLINE:

According to a bridging analysis, two PD-L1 immunohistochemical assays — Dako 22C3 and VENTANA SP263 — demonstrated 88% concordance in identifying patients with non-small cell lung cancer (NSCLC) for first-line cemiplimab treatment. Clinical outcomes and survival benefits were comparable between patient populations selected by either assay, supporting their interchangeable use in clinical practice.

METHODOLOGY:

• PD-L1 expression is used to select patients with advanced NSCLC who are most likely to respond to PD-1/PD-L1 monotherapy. Multiple PD-L1 immunohistochemical assays with differing antibodies and platforms exist, complicating the process of selecting patients for PD-1/PD-L1 therapy. This bridging analysis determined whether Dako 22C3 and VENTANA SP263 reliably identified patients with PD-L1 expression ≥ 50% for cemiplimab therapy, using samples from the EMPOWER-Lung 1 study.
• The phase 3 EMPOWER-Lung 1 trial randomly assigned 710 treatment-naive patients with advanced NSCLC — with tumor PD-L1 expression ≥ 50% determined by the Dako 22C3 assay and without aberrations in the epidermal growth factor receptor; ALK; or ROS1, receptor tyrosine kinase genes — to receive first-line cemiplimab monotherapy or platinum-doublet chemotherapy.
• Samples from 871 screened patients, including 481 enrolled patients and 390 of those who did not pass screening, were retested using VENTANA SP263. Overall, 768 samples yielded evaluable results for both assays.
• Concordance metrics (positive percent agreement, negative percent agreement, and overall percent agreement) were calculated using Dako 22C3 as reference. The primary endpoints — overall survival (OS) and progression-free survival (PFS) — were estimated in the SP263-positive population (ie, PD-L1 expression ≥ 50%).
• Researchers compared outcomes in the 22C3-positive/SP263-positive subpopulation with those in the original 22C3-positive population.

TAKEAWAY:

• A concordance analysis of 768 samples revealed an overall percent agreement of 88% between assays, with a positive percent agreement of 82.7% and a negative percent agreement of 93.6%.
• Among discordant cases, 69% (47 of 68) of the 22C3-positive/SP263-negative subpopulation had borderline PD-L1-positive status, with Dako 22C3-determined tumor proportion scores ranging from ≥ 50% to ≤ 60%. This “suggests that the assays were highly concordant, and the lower-than-expected positive agreement was likely due to normal variation in PD-L1 scoring by pathologists close to the cut point of ≥ 50%.”
• Overall survival analysis demonstrated similar efficacy between the 22C3-positive/SP263-positive subpopulation (hazard ratio [HR], 0.52) and the 22C3-positive population (HR, 0.57) for cemiplimab vs chemotherapy. Similarly, PFS was comparable between them (HRs, 0.43 and 0.54, respectively).
• Additionally, sensitivity analyses using imputation and worst-case scenario analyses yielded consistent results. Because the VENTANA SP263 assay was assessed retrospectively on archived samples, adverse events were neither expected nor reported.

IN PRACTICE:

In this analysis, “comparison with the VENTANA SP263 assay showed a high level of analytical concordance, and an equivalent association of clinical efficacy (OS and PFS estimates) was found in the PD-L1 ≥ 50% population,” the authors of the study wrote, concluding that “the VENTANA SP263 and Dako 22C3 assays can be used interchangeably to identify patients with PD-L1 ≥ 50% who may benefit from cemiplimab therapy.”

SOURCE:

The study, led by Javier Perez, PhD, Regeneron Pharmaceuticals, Inc., in Tarrytown, New York, was published online in JCO Precision Oncology.

LIMITATIONS:

The discordant 22C3+/SP263- subgroup was small, limiting power for subgroup efficacy analysis. Several outlier results were observed in the concordance analysis, such as instances where strongly positive Dako 22C3 results showed near-zero readings with the VENTANA SP263 assay, which might be attributed to tumor heterogeneity or sectioning differences.

DISCLOSURES:

The study was funded by Regeneron Pharmaceuticals, Inc., and Sanofi. Thirteen authors reported being employees of or holding stock and other ownership interests with Regeneron. Full disclosures are noted in the original article.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.