Neoadjuvant and Adjuvant Pembrolizumab in Resectable Locally Advanced, Human Papillomavirus-Unrelated Head and Neck Cancer: A Multicenter, Phase II Trial

Abstract

Purpose: Pembrolizumab improved survival in patients with recurrent or metastatic head and neck squamous-cell carcinoma (HNSCC). The aims of this study were to determine if pembrolizumab would be safe, result in pathologic tumor response (pTR), and lower the relapse rate in patients with resectable human papillomavirus (HPV)-unrelated HNSCC.

Patients and methods: Neoadjuvant pembrolizumab (200 mg) was administered and followed 2 to 3 weeks later by surgical tumor ablation. Postoperative (chemo)radiation was planned. Patients with high-risk pathology (positive margins and/or extranodal extension) received adjuvant pembrolizumab. pTR was quantified as the proportion of the resection bed with tumor necrosis, keratinous debris, and giant cells/histiocytes: pTR-0 (<10%), pTR-1 (10%-49%), and pTR-2 (≥50%). Coprimary endpoints were pTR-2 among all patients and 1-year relapse rate in patients with high-risk pathology (historical: 35%). Correlations of baseline PD-L1 and T-cell infiltration with pTR were assessed. Tumor clonal dynamics were evaluated (ClinicalTrials.gov NCT02296684).

Results: Thirty-six patients enrolled. After neoadjuvant pembrolizumab, serious (grades 3-4) adverse events and unexpected surgical delays/complications did not occur. pTR-2 occurred in eight patients (22%), and pTR-1 in eight other patients (22%). One-year relapse rate among 18 patients with high-risk pathology was 16.7% (95% confidence interval, 3.6%-41.4%). pTR ≥10% correlated with baseline tumor PD-L1, immune infiltrate, and IFNγ activity. Matched samples showed upregulation of inhibitory checkpoints in patients with pTR-0 and confirmed clonal loss in some patients.

Conclusions: Among patients with locally advanced, HPV-unrelated HNSCC, pembrolizumab was safe, and any pathologic response was observed in 44% of patients with 0% pathologic complete responses. The 1-year relapse rate in patients with high-risk pathology was lower than historical.

Figure 1.. Trial profile and tumor responses (pathologic and radiologic) to neoadjuvant pembrolizumab

(A) Trial profile: Patients (n=36) with locally advanced, Stage III/IV, HPV-negative HNSCCs underwent baseline tumor and blood sampling and received neoadjuvant pembrolizumab 2–3 weeks before surgery. Of the 18 patients with high-risk pathology, 12 received adjuvant pembrolizumab. Patients with low/intermediate-risk pathology did not receive adjuvant pembrolizumab. (B) pathologic tumor response-2 (pTR-2) was observed in similar proportions of patients with low/intermediate and high-risk pathology. (C) Baseline and (D) post-neoadjuvant treatment (at surgery) images of oral cavity primary cancer in Patient 20 showing dramatic decrease in the size. (E) Representative CT images at baseline and (F) after neoadjuvant pembrolizumab (day prior to surgery), confirmed tumor response seen on physical exam. Notably, the pre-treatment CT and FDG-PET/CT scan (not shown) showed multiple large and necrotic FDG-avid neck lymph nodes, which are radiologic signs of SCC. Of note, the internal jugular vein (white arrow) was compressed on the baseline scan (E) and appeared fuller on the post-treatment scan (F). (G) Representative H&E slide of pTR highlighting changes noted in surgical specimens. (H) Nineteen patients had CT evaluations at baseline and prior to surgery following neoadjuvant pembrolizumab: 16 with stable disease (SD) and 3 with progressive disease (PD) by RECIST criteria. *Indicates two patients with pTR-1.

Figure 2.. Pathologic tumor response correlates with tumor PD-L1 and immune infiltrates

(A) Representative PD-L1 staining of tumor biopsies at baseline. (B) PD-L1 H-score correlated with pathologic tumor response (PTR). Baseline PD-L1 primary tumor expression levels by IHC and percent PTR were significantly positively correlated for 32 evaluated patients (rho=0.43; 95%CI 0.079–0.668) (C) representative multiplex immunofluorescence (MIF) images showing patient 5 with minimal CD8+ T cell infiltrates and Patient 10 with higher CD8+ T cell infiltrates (white arrows) in baseline biopsies. (D) Extent of PTR was correlated with number of CD8+ T cells in the baseline biopsy tumor microenvironment (TME). Baseline number of TME CD8+ T cells assessed by MIF and percent PTR were significantly positively correlated for 23 evaluated patients (rho=0.72; 95%CI: 0.443–0.875).

Figure 3.. Immune infiltrate and activity correspond to patient response

A. Heatmap shows genes (n=41) associated with hallmark gene sets (right panel) that were differentially expressed (p<0.01, adjusted p<0.25) between patients with PTR and those without at baseline. Genes are sorted first by hallmark geneset, then by Ward’s hierarchical clustering. Patients are sorted by decreasing maximum PTR (at either the tumor or lymph node site), then by Ward’s hierarchical clustering. Expression is displayed as the gene-normalized expression across samples. B–C. Baseline- and post-treatment RNA was assessed for patterns of infiltrating immune cells, and values were summarized by the absolute levels of immune cells. These are indicated as the sum of all immune cell populations (Absolute score) or by subpopulation at baseline (B). Matched samples were available for a subset of patients (n=15) and the changes over the course of treatment are depicted by connected lines between baseline and resection timepoints in (C). Wilcoxon tests were used to evaluate statistical significance across responder groups and timepoints; * indicates p<0.05. D. Expression (Log2 FPKM, y-axis) of six genes in baseline and post-treatment bulk tumor RNAseq data. Points and lines are colored by either PTR (pTR-1/2) or without (pTR-0). Paired samples per individual are connected by lines, and p-values (labeled above) indicate the comparison of paired pre- and post-treatment samples.

Figure 4.. Tumor mutational burden and neoantigen burden does not positively correlate with PTR

(A) The number of nonsilent mutations or (B) putative neoantigens predicted, as detected by WES and available RNAseq data from baseline samples (y-axes) was compared to the extent of PTR (x-axis). ‘Extent of PTR’ indicates the maximum PTR observed in the primary and/or resected lymph nodes. Points are colored by category of PTR, indicating whether there was no PTR (pTR-0), pTR-1 (<50% PTR at either site), or pTR-2 (>50% at either site).

Figure 5.. Varied clonal dynamics due to spatial heterogeneity.

A) Clonality plots comparing VAF of SNVs and Indels at baseline (x-axis) and resection (y-axis). These values represent aggregate metrics of all post-treatment samples available, and colors represent variants that clustered together by comparing these aggregate metrics per individual. Open circles represent variants that were either unassigned or were assigned to clusters with less than 5 variants. B) The scaled allelic frequency (AF) is depicted on the y-axis of each variant (point) that was assigned a cluster in (A). The solid line represents the average scaled AF of the cluster. Shaded ribbons represent the standard error from the mean scaled AF of the cluster. Eight non-overlapping regions were isolated from Patient 2 post-treatment (PT) resected tumor, 7 regions were isolated from patient 10, and 8 regions were isolated from Patient 2 and analyzed (Supplementary Figure 10).