Predicting Progression Free Survival after Systemic Therapy in Advanced Head and Neck Cancer: Bayesian regression and Model development
Abstract
Background
Advanced Head and Neck Squamous Cell Cancer (HNSCC) is associated with a poor prognosis, and biomarkers that predict response to treatment are highly desirable. The primary aim was to predict Progression Free Survival (PFS) with a multivariate risk prediction model.
Methods
Blood samples from 56 HNSCC patients were prospectively obtained within a Phase 2 clinical trial (<ext-link xmlns:xlink="http://www.w3.org/1999/xlink" ext-link-type="clintrialgov" xlink:href="NCT02633800">NCT02633800</ext-link>), before and after the first treatment cycle of platinum-based chemotherapy, to identify biological covariates predictive of outcome. A total of 42 baseline covariates were derived pre-treatment, which were combined with 29 covariates after one cycle of treatment. These covariates were ranked and selected by Bayesian multivariate regression to form risk scores to predict PFS, producing “baseline” and “combined” risk prediction models respectively.
Results
The baseline model comprised of CD33+CD14+ monocytes, Double Negative B cells and age, in a weighted risk signature which predicted PFS with a concordance index (C-index) of 0.661. The combined model composed of baseline CD33+CD14+ monocytes, baseline Tregs, after-treatment changes in CD8 effector memory T cells, CD8 Central memory T cells and CD3 T Cells, along with the hypopharyngeal primary tumor subsite. This weighted risk signature exhibited an improved C-index of 0.757. There was concordance between levels of CD33+CD14+ myeloid cells in tumor tissue, as demonstrated by imaging mass cytometry, and peripheral blood in the same patients. This monocyte subpopulation also had univariate predictive value (log-rank p value = 0.03) but the C-index was inferior to the combined signature.
Conclusions
This immune-based combined multimodality signature, obtained through longitudinal peripheral blood monitoring, presents a novel means of predicting response early on during the treatment course.
Funding
Daiichi Sankyo Inc, Cancer Research UK, EU IMI2 IMMUCAN, UK Medical Research Council, European Research Council (335326), National Institute for Health Research and The Institute of Cancer Research.
Related articles
Related articles are currently not available for this article.