◥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 (Clin-icalTrials.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 IFNg 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, HPVunrelated 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.

Head and neck squamous cell carcinomas (HNSCC) are an ideal immunotherapy target due to their high mutation burden and frequent infiltration with lymphocytes. Preclinical models to investigate targeted and combination therapies as well as defining biomarkers to guide treatment represent an important need in the field. Immunogenomics approaches have illuminated the role of mutation-derived tumor neoantigens as potential biomarkers of response to checkpoint blockade as well as representing therapeutic vaccines. Here, we aimed to define a platform for checkpoint and other immunotherapy studies using syngeneic HNSCC cell line models (MOC2 and MOC22), and evaluated the association between mutation burden, predicted neoantigen landscape, infiltrating T cell populations and responsiveness of tumors to anti-PD1 therapy. We defined dramatic hematopoietic cell transcriptomic alterations in the MOC22 anti-PD1 responsive model in both tumor and draining lymph nodes. Using a cancer immunogenomics pipeline and validation with ELISPOT and tetramer analysis, we identified the H-2Kb-restricted ICAM1 P315L (mICAM1) as a neoantigen in MOC22. Finally, we demonstrated that mICAM1 vaccination was able to protect against MOC22 tumor development defining mICAM1 as a bona fide neoantigen. Together these data define a pre-clinical HNSCC model system that provides a foundation for future investigations into combination and novel therapeutics. www.impactjournals.com/oncotarget/

6012 Background: Pembrolizumab has efficacy in metastatic HNSCC. We hypothesized that treatment intensification in surgically resectable HPV-negative, Stage III/IV HNSCC with neoadjuvant plus post-operative adjuvant (POA) pembrolizumab would be safe and reduce 1-year locoregional recurrence/distant metastases (LRR/DM) from 35% (historical: Cooper and Bernier NEJM 2004) to 15%. Methods: Phase II trial where all eligible patients received 1 dose of pembrolizumab (200 mg) prior to surgery and only those with high-risk pathologic features (HRPF: extracapsular extension/positive margin) were given POA cisplatin and radiation followed by pembrolizumab. PD-L1 staining was assessed by immunohistochemistry (9A11 antibody). Results: The study continues to enroll. Characteristics of 21 enrolled patients (pts) were median age 59 (32-87) yrs, tobacco use 81% (17 pts), clinical T2 (n = 2), T3 (n = 1), T4 (n = 18), and cN0/1 (n = 8), cN2 (n = 13). Preliminary analyses revealed five important findings: 1) No serious study drug-related AEs or unexpected surgical delays/complications, 2) No LRR/DM events in the first 10 patients with > 1-year follow-up after surgery 3) HRPF rate of 38% (95% CI: 18%-62%) (expected: 80%), 4) 43% of pts (95% CI: 22%-66%) with pathologic treatment response to neoadjuvant pembrolizumab (definition: tumor necrosis and/or giant cell/histiocytic reaction to keratinous debris in > 10% of tumor area), and 5) 48% of pts (95% CI:26%-70%) with clinical-to-pathologic downstaging. Pathologic treatment effect (TE) in ≥ 70% of the resected tumor or lymph node tissue area occurred in 6/21 pts (29%). Baseline tumor biopsies were PD-L1 positive ( > 1% of tumor cells) in 11/19 (58%) evaluable samples and in 7/8 (88%) evaluable pathologic responders. A significant correlation existed between baseline PD-L1 expression on tumor cells and pathologic treatment effect in the tumor (correlation coefficient: 0.72 and p = 0.0005). Conclusions: Neoadjuvant and adjuvant pembrolizumab was safe and well tolerated. We observed several lines of evidence supporting an anti-tumor effect in these pts with a single dose of pre-operative pembrolizumab. Further evaluation of this strategy is warranted. Clinical trial information: NCT02296684.

Purpose Ras/MEK/ERK pathway activation is common in oral cavity squamous cell carcinoma (OCSCC). We performed a neoadjuvant (pre-operative) trial to determine biomarker and tumor response of OCSCC to MEK inhibition with trametinib. Patients and Methods Patients with Stage II–IV OCSCC received trametinib (2 mg/day, minimum 7 days) prior to surgery. Primary tumor specimens were obtained before and after trametinib to evaluate immunohistochemistry staining for p-ERK1/2 and CD44, the primary endpoint. Secondary endpoints included changes in clinical tumor measurements and metabolic activity (maximum Standardized Uptake Values [SUVmax] by F-18 fluorodeoxyglucose positron emission tomography/computed tomography), and in tumor downstaging. Drug-related adverse events (AEs) and surgical/wound complications were evaluated. Results Of 20 enrolled patients, 17 (85%) completed the study. Three patients withdrew because of either trametinib-related (n=2:nausea, duodenal perforation) or unrelated (n=1:constipation) AEs. The most common AE was rash (9/20 patients, 45%). Seventeen patients underwent surgery. No unexpected surgical/wound complications occurred. Evaluable matched pre- and post-trametinib specimens were available in 15 (88%) of these patients. Reduction in p-ERK1/2 and CD44 expression occurred in 5 (33%) and 2 (13%) patients, respectively. Clinical tumor response by modified World Health Organization criteria was observed in 11 of 17 (65%) evaluable patients (median 46% decrease, range 14 to 74%). Partial metabolic response (≥25% reduction in SUVmax) was observed in 6 of 13 (46%) evaluable patients (median 25% decrease, range 6 to 52%). Clinical-to-pathologic tumor downstaging occurred in 9 of 17 (53%) evaluable patients. Conclusions Trametinib resulted in significant reduction in Ras/MEK/ERK pathway activation and in clinical and metabolic tumor responses in OCSCC patients.

Herein, we report an oral cavity squamous cell carcinoma (OCSCC) patient-derived xenograft (PDX) platform, with genomic annotation useful for co-clinical trial and mechanistic studies. Genomic analysis included whole-exome sequencing (WES) and transcriptome sequencing (RNA-seq) on 16 tumors and matched PDXs and additional whole-genome sequencing (WGS) on 9 of these pairs as a representative subset of a larger OCSCC PDX repository (n = 63). In 12 models with high purity, more than 90% of variants detected in the tumor were retained in the matched PDX. The genomic landscape across these PDXs reflected OCSCC molecular heterogeneity, including previously described basal, mesenchymal, and classical molecular subtypes. To demonstrate the integration of PDXs into a clinical trial framework, we show that pharmacological intervention in PDXs parallels clinical response and extends patient data. Together, these data describe a repository of OCSCC-specific PDXs and illustrate conservation of primary tumor genotypes, intratumoral heterogeneity, and co-clinical trial application.