Journal for ImmunoTherapy of Cancer (Nov 2024)

Multicenter validation of an RNA-based assay to predict anti-PD-1 disease control in patients with recurrent or metastatic head and neck squamous cell carcinoma: the PREDAPT study

  • Howard L McLeod,
  • A Dimitrios Colevas,
  • Georges Azzi,
  • Kevin C Flanagan,
  • Douglas Adkins,
  • Jessica Ley,
  • Ezra E W Cohen,
  • James Wade,
  • Jon Earls,
  • Jeffrey Hiken,
  • Rachel L Wellinghoff,
  • Michelle M Ponder,
  • William H Westra,
  • Vera Vavinskaya,
  • Leisa Sutton,
  • Ida Deichaite,
  • Orlan K Macdonald,
  • Karim Welaya,
  • Andrew W Pippas,
  • Jennifer Slim,
  • Bruce Bank,
  • Xingwei Sui,
  • Steven E Kossman,
  • Todd D Shenkenberg,
  • Warren L Alexander,
  • Katharine A Price,
  • David N Messina,
  • Jarret I Glasscock,
  • Eric J Duncavage

DOI
https://doi.org/10.1136/jitc-2024-009573
Journal volume & issue
Vol. 12, no. 11

Abstract

Read online

Background Despite advances in cancer care and detection, >65% of patients with squamous cell cancer of the head and neck (HNSCC) will develop recurrent and/or metastatic disease. The prognosis for these patients is poor with a 5-year overall survival of 39%. Recent treatment advances in immunotherapy, including immune checkpoint inhibitors like pembrolizumab and nivolumab, have resulted in clinical benefit in a subset of patients. There is a critical clinical need to identify patients who benefit from these antiprogrammed cell death protein 1 (anti-PD-1) immune checkpoint inhibitors.Methods Here, we report findings from a multicenter observational study, PREDicting immunotherapy efficacy from Analysis of Pre-treatment Tumor biopsies (PREDAPT), conducted across 17 US healthcare systems. PREDAPT aimed to validate OncoPrism-HNSCC, a clinical biomarker assay predictive of disease control in patients with recurrent or metastatic HNSCC treated with anti-PD-1 immune checkpoint inhibitors as a single agent (monotherapy) and in combination with chemotherapy (chemo-immunotherapy). The test used RNA-sequencing data and machine learning models to score each patient and place them into groups of low, medium, or high.Results The OncoPrism-HNSCC prediction significantly correlated with disease control in both the monotherapy cohort (n=62, p=0.004) and the chemo-immunotherapy cohort (n=50, p=0.01). OncoPrism-HNSCC also significantly predicted progression-free survival in both cohorts (p=0.015 and p=0.037, respectively). OncoPrism-HNSCC had more than threefold higher specificity than programmed death-ligand 1 combined positive score and nearly fourfold higher sensitivity than tumor mutational burden for predicting disease control.Conclusions Here, we demonstrate the clinical validity of the OncoPrism-HNSCC assay in identifying patients with disease control in response to anti-PD-1 immune checkpoint inhibitors.Trial registration number NCT04510129.