Head and neck squamous cell carcinomas (HNSCCs) represent the most common histologic subtype within the broader and highly heterogeneous category of head and neck tumors [1]. HNSCC is the seventh most common cancer worldwide and accounts for ~4% of all cancers in the United States, with an estimated 66,000 new cases diagnosed each year. Globally, more than 500,000 new cases and 250,000 deaths attributed to HNSCC occur each year ^[2][3][2,3]. Diagnosis of HNSCC has long been associated with smoking, alcohol, exposure to environmental carcinogens, age, male sex, and unique viral etiologies [4]. Although early-stage tumors are often associated with a good prognosis, most patients present with locally advanced disease, of whom about 50% will relapse within 1 year. Median survival rates remain modest, with 5 -year survivals ranging from 50 to 60% in the era before immunotherapy ^[5][6][5,6]; current immunotherapy approaches demonstrate significant albeit modest gains of 2–4 months in the relapsed and refractory setting [7]. Patients who do achieve a cure from the disease often suffer consequences of complicated and often morbid surgeries or long-term sequelae from chemotherapy and radiation to the head and neck, resulting in temporary or permanently altered eating, speech, and quality of life [8]. Among cancer survivors, those with HNSCC have the second highest rate of suicide after those with pancreatic cancer, which is attributed to psychological stress and compromised quality of life [9]. New diagnostic tools and therapeutic strategies are needed to inform patient care, provide meaningful improvement to progression-free and overall survival, and minimize long-term toxicity of therapy.

Outcomes in HNSCC remain inadequate, particularly in the metastatic and relapsed setting. Thus, there is a tremendous need to identify molecular biomarkers that provide prognostic and therapeutic opportunities, which may identify subgroups of patients who could benefit from novel approaches based on clinical or molecular characteristics.

Within non-nasopharyngeal cancers, two distinct forms of HNSCC have emerged with different risk factors: tobacco and alcohol exposure, and human papillomavirus (HPV) infection ^[10][11][10,11]. Patients with HPV-negative HNSCC present more frequently with HNSCC affecting the oral cavity and larynx and have prolonged exposure of this anatomy to carcinogens via smoking and alcohol consumption. Generally, patients with HPV-positive HNSCC present with oropharyngeal origins and have more favorable outcomes in response to chemotherapy and radiotherapy [12]. A recent Surveillance, Epidemiology, and End Results Program analysis noted increased rates of 5-year survival for patients with HNSCC, from 55% to 66%, when comparing outcomes in 1992–1996 vs. 2002–2006. This improvement was mainly attributed to better treatment options and increased survival in patients with HPV-positive HNSCC who tend to be younger and more fit [13]. Significantly higher incidence of HPV-negative HNSCC has been reported for men compared to women (incidence rate 0.19 vs. 0.06 per 1000 person-years, respectively). The imbalance favoring men remains 2.9-fold higher even when comparing subsets of never-smokers/drinkers, which suggests an intrinsic risk factor associated with male sex [14]. Similar results have been observed in other databases [15] and in patients with HPV-positive oropharyngeal disease [16]. Although increased exposure to tobacco and alcohol among men vs. women may partially explain the higher incidence of HPV-negative HNSCC in men, recent understanding of mutational patterns has suggested additional mechanisms that underlie a gender bias in men. Recurrent loss-of-function mutations in X-linked tumor suppressors are observed in HNSCC, including DDX3X, KDM5C, and KDM6A [17]. Mutations in these genes appear more common in men than women with HNSCC. Because men have only one copy of X-linked genes, a single mutation is sufficient to inactivate X-linked tumor suppressor function, thus accelerating transformation. Analysis of the ~800 X-linked coding genes identified 90 genes that modify TP53-dependent pathways [18]. Because TP53 is one of the most commonly mutated genes in HNSCC, X-linked tumor suppressors that modulate TP53-dependent pathways present a particular vulnerability for HNSCC transformation in men. Since females have two X chromosomes and males have only one, which is maternally derived, Y-linked effects are more associated with a male bias. Mosaic loss of chromosome Y (LOY) in blood cells is found in approximately a third of men with HNSCC, is disproportionally present among tobacco users, and is associated with shorter survival [19]. Therefore, genomic effects involving recurrent mutations in X-linked tumor suppressors likely contribute to a higher incidence of HPV-negative HNSCC in men, and this bias is not likely only due to greater exposure to tobacco.

Immune evasion occurs as a downstream result of accumulating genetic mutations driven by dysfunction of the tumor suppressor activity and amplification of oncogenes. Therefore, therapies that reestablish recognition and elimination of tumor cells may provide patient-specific therapeutic approaches if tumor-specific targets can be identified. Activity of these types of therapies provides further evidence for the pathologic relevance of neo-epitopes, inducing immune surveillance in HNSCC.

Checkpoint inhibitors with combination approaches have emerged as a therapeutic strategy that enables immune re-activation to overcome tumor immunosuppressive effects. Results of these trials suggest that the combination may not improve durability of response, particularly in patients with a history of either therapy ^[20].[20]

Non-structural mechanisms such as epigenetic silencing of B2M, HLA class I, or other antigen-processing genes are common mechanisms of immune evasion and can be seen in nearly three-quarters of patients with cancer, including those with HNSCC ^[21][22].[21,22]

HLA Loss of Heterozygosity (LOH) is an irreversible genomic event that is propagated through subsequent sub-clonal evolution. HLA LOH may provide a means to distinguish tumor from normal tissue in a definitive manner due to this irreversible, clonal loss within tumor cells. A recent strategy has emerged to target these cell-surface epitopes that are absent on malignant cells through use of a logic-gated chimeric antigen receptor (CAR)-T receptor referred to as Tmod [23]. Tmod incorporates an activator with a blocker. The blocker is generated using LIR1 extracellular and intracellular domains and fuses this to an scFv domain targeting HLA-A*02, although additional haplotypes may also be targeted. This provides an inhibitory signal when the CAR-T cell engages a cell with HLA-A*02 expression (i.e., a non-malignant cell) [23]. Integrating blocking receptors that recognize loss of HLA expression may enable an improved therapeutic index in patients with tumors that have acquired HLA LOH. This strategy may reduce on-target/off-tumor toxicities and improve the therapeutic index sufficiently to enable effective CAR-T targeting in HNSCC.

Methods that can reverse immune evasion (e.g., epigenetic targeted therapies) or exploit irreversible genomic events (e.g., logic-gated CAR-Ts) can potentially transform survival outcomes for these patients and offer a future direction for therapeutic development.