Multidimensional biomarker predicts disease control in response to immunotherapy in recurrent or metastatic head and neck squamous-cell carcinoma

Abstract

Purpose: Anti-PD-1 therapy provides clinical benefit in 40-50% of patients with relapsed and/or metastatic head and neck squamous cell carcinoma (RM-HNSCC). Selection of anti- PD-1 therapy is typically based on patient PD-L1 immunohistochemistry (IHC) which has low specificity for predicting disease control. Therefore, there is a critical need for a clinical biomarker that will predict clinical benefit to anti-PD-1 treatment with high specificity.

Methods: Clinical treatment and outcomes data for 103 RM-HNSCC patients were paired with RNA-sequencing data from formalin-fixed patient samples. Using logistic regression methods, we developed a novel biomarker classifier based on expression patterns in the tumor immune microenvironment to predict disease control with monotherapy PD-1 inhibitors (pembrolizumab and nivolumab). The performance of the biomarker was internally validated using out-of-bag methods.

Results: The biomarker significantly predicted disease control (65% in predicted non-progressors vs. 17% in predicted progressors, p < 0.001) and was significantly correlated with overall survival (OS; p = 0.004). In addition, the biomarker outperformed PD-L1 IHC across numerous metrics including sensitivity (0.79 vs 0.64, respectively; p = 0.005) and specificity (0.70 vs 0.61, respectively; p = 0.009).

Conclusion: This novel assay uses tumor immune microenvironment expression data to predict disease control and OS with high sensitivity and specificity in patients with RM-HNSCC treated with anti-PD-1 monotherapy.

Keywords: Biomarker; HNSCC; Immune checkpoint inhibitors; PD-1; PD-L1; Pembrolizumab.

Fig. 1

RNA-seq data was used to build the biomarker. A A multidimensional biomarker was trained using tumor specimen RNA-seq gene expression data and clinical response data from 103 HNSCC patients. Logistic regression and forward feature selection were used to integrate the data (“Predictive Immune Modeling”) and produce a multidimensional biomarker. B Patients were segregated by outcome and evaluated for progression free survival (PFS). CR, complete response; PR, partial response; SD, stable disease, PD, progressive disease. CR, PR, and SD together constitute “non-progressors”. PD patients are “progressors”. C Candidate feature relative abundance is shown for each patient sample. Features are gene expression data or gene signatures associated with specific immune cell types. Patient samples are sorted by outcome label (true outcome), then by OncoPrism Score and PD-L1 IHC prediction

Fig. 2

Higher OncoPrism Scores are associated with disease control with anti-PD-1 therapy A ROC curve showing the performance of OncoPrism-HNSCC compared to PD-L1 IHC across all thresholds. An out-of-bag (OOB) method was used to assess performance. OncoPrism-HNSCC (orange) has an area under the curve (AUC) of 0.76, while PD-L1 IHC (grey) has an AUC of 0.65. The dashed line (red) is performance equivalent to chance. B Patients were ranked along the x-axis according to their OncoPrism Score (y-axis) and colored according to their actual clinical outcome (progressors = grey, non-progressors = orange) C Samples were ranked according to their OncoPrism Score and divided into quartiles for assessment. The disease control rate (DCR) for all patients is represented by the solid line. The blue bars represent the actual DCR for each quartile not a mean. Significant differences among quartiles were determined using Fisher’s Exact test (*p < 0.05, **p < 0.01, ***p < 0.001)

Fig. 3

OncoPrism-HNSCC predicts disease control and overall survival (OS) better than PD-L1 IHC. A DCR is significantly higher for OncoPrism-HNSCC predicted non- progressors than for OncoPrism-HNSCC predicted progressors (p < 0.001). DCR for PD- L1 CPS ≥ 20 (“non-progressor”) is significantly higher than CPS < 20 (“progressor”; p = 0.02). Bars represent the actual DCRs not a mean. Significant differences among quartiles were determined using Fisher’s Exact test. B OncoPrism-HNSCC predicted non-progressors have significantly longer OS than predicted progressors (p = 0.004; n = 103). OS is measured from the time of first anti-PD-1 treatment. C PD-L1 IHC CPS ≥ 20 patients (corresponding to a PD-L1 predicted non-progressor) do not have longer OS than CPS < 20 (p = 0.7; n = 100). OS is measured from the time of first anti-PD-1 treatment

Fig. 4

Cox proportional hazards model of overall survival (OS). Patients with complete OncoPrism-HNSCC and PD-L1 data were included (n = 100). OncoPrism-HNSCC predicted non-progressors have a Hazard Ratio of 0.51 relative to predicted progressors (p = 0.008). There is no significant difference in OS between PD-L1 < 20 and PD-L1 ≥ 20