Abstract 684: Multi-site external reproducibility of PD-L1 IHC 22C3 pharmDx performance on Dako Omnis and comparison to Autostainer Link 48 for head and neck squamous cell carcinoma using the Combined Positive Score

Abstract Background: PD-L1 IHC 22C3 pharmDx is a diagnostic assay indicated as an aid in identifying biomarker PD-L1 expression levels for use on the Autostainer Link 48 (ASL48) platform for multiple indications. However, to date PD-L1 IHC 22C3 pharmDx on Dako Omnis platform has only been validated for non-small cell lung cancer. This study sought to demonstrate external reproducibility of PD-L1 IHC 22C3 pharmDx for head and neck squamous cell carcinoma (HNSCC) using Dako Omnis (Investigator Use Only/Performance Evaluation Only). This study also investigated the concordance of PD-L1 IHC 22C3 pharmDx on HNSCC on both ASL48 and Dako Omnis at external sites. Methods: This study was a three-site, blinded, and randomized study of PD-L1 IHC 22C3 pharmDx on HNSCC specimens. To investigate inter-site and intra-site reproducibility, five replicate sets of pre-qualified unstained specimens were stained using Dako Omnis and evaluated at each of the three testing sites. For the inter-platform performance evaluation, pathologists at each of the three testing sites evaluated one set of HNSCC slides that was pre-stained using PD-L1 IHC 22C3 pharmDx on ASL48. Pathologists determined the PD-L1 expression level using a Combined Positive Score (CPS) at diagnostic cutoffs CPS ≥ 1 and CPS ≥ 20. The analytical agreements of the diagnostic outcome (positive/negative) were estimated by evaluating negative percent agreement (NPA), positive percent agreement (PPA), and overall percent agreement (OA) based on comparisons to the consensus diagnostic outcome with corresponding percentile bootstrap confidence intervals. The pre-determined acceptance criteria (AC) for all endpoints are that the lower-bound (LB) of a two-sided 95% confidence interval (CI) for NPA, PPA, and OA must meet or exceed 85.0%. Results: The study met all AC for the CPS ≥ 1 and CPS ≥ 20 cutoffs. For CPS ≥ 1, the 95% CI LB of the inter-site reproducibility endpoints were 90.5% for NPA, 98.7% for PPA, and 95.0% for OA and intra-site reproducibility endpoints were 98.9% for NPA, 98.0% for PPA, and 98.5% for OA. For CPS ≥ 20, the 95% CI LB of the inter- site reproducibility endpoints were 87.7% for NPA, 91.1% for PPA, and 91.2% for OA and intra-site reproducibility endpoints were 92.7% for NPA, 95.8% for PPA, and 95.0% for OA. For CPS ≥ 1, the 95% CI LB of the inter-platform performance comparison endpoints were 96.0% for NPA, 98.7% for PPA, and 98.0% for OA. For CPS ≥ 20, the 95% CI LB of the inter-platform performance comparison endpoints were 90.6% for NPA, 94.2% for PPA, and 93.3% for OA. Conclusions: This study demonstrates reproducibility of PD-L1 IHC 22C3 pharmDx (with Dako Omnis) in HNSCC specimens at the CPS ≥ 1 and CPS ≥ 20 cutoffs. The results of the inter-platform performance comparison study demonstrate concordance between ASL48 and Dako Omnis in HNSCC specimens at the CPS ≥ 1 and CPS ≥ 20 cutoffs. Citation Format: Monika D. Polewski, Lining Hutchinson, Jianan Ren, Nichole LaPointe, Siena Tabuena-Frolli, Kurt Mathews, Gregory A. Multi-site external reproducibility of PD-L1 IHC 22C3 pharmDx performance on Dako Omnis and comparison to Autostainer Link 48 for head and neck squamous cell carcinoma using the Combined Positive Score [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 684.