The paradigm shift from conventional chemotherapy and targeted therapy to immunotherapy represents an exciting frontier in the medical management of human cancer. Head and neck cancers are ideal candidates for these immune-based approaches due to their high degree of immune cell infiltration. Two PD-1 inhibitors (pembrolizumab and nivolumab) were recently FDA approved for the treatment of head and neck cancer. Ongoing studies are investigating the role of checkpoint inhibition in the neoadjuvant setting and in combination with other novel immune modulating agents to maximize patient responses.
Immunotherapy represents an exciting frontier in the development of nonsurgical treatments for human cancers. Oral cavity cancer is one of the most promising avenues for the field. General approaches to triggering immune eradication of cancers have included tumor vaccines, adoptive cell transfer (ACT), and immune checkpoint blockade (ICB). The unprecedented successes of checkpoint inhibitors in a variety of cancer types have led to their approval by the FDA for the treatment of melanoma, non-small cell lung, renal cell, head and neck, and bladder cancers. These are the first agents since conventional chemotherapy to be applied successfully across a variety of tumor types, and are the subject of a rapidly developing body of research to expand responses and refine treatment regimens.
Immunotherapeutic strategies are distinct from conventional chemotherapy or targeted therapy, and represent a paradigm shift toward harnessing the body’s own immune system to attack a tumor. The nonspecific destruction of rapidly dividing cells achieved with conventional chemotherapy is associated with significant morbidity and toxicity. Targeted therapies have shown promise in subsets of tumors with specific mutations; however, the whirlwind of growth and mutational evolution that define most cancers can also lead to rapid development of resistance. Immunotherapy, on the other hand, seeks to harness the power of the body’s immune system against invading tumor cells to generate a widespread and durable response.
Responses to chemoradiation in the oral cavity cancer population have historically been modest, and surgery is increasingly preferred as the primary treatment modality when feasible. 1 Oral cavity cancer is an area with unmet need for improved nonsurgical alternatives, and these patients are poised to benefit greatly from advances in immunotherapy research. This chapter will explore our baseline understanding of immune infiltration in head and neck cancer, particularly in oral cavity cancer, and highlight the successes and failures of recent immunotherapy clinical trials for head and neck cancer. Additionally, this chapter will address clinical biomarkers for response, resistance mechanisms, and ongoing clinical trials and research efforts to improve immunotherapy for head and neck cancer.
50.2 Immune Infiltration in Head and Neck Cancer
Cancer types that are susceptible to checkpoint inhibition tend to be immunologically active at baseline with a high level of immune cell infiltration in the tumor microenvironment. These tumors generally have a high mutational load, creating neoantigens that are displayed on the surface of antigen-presenting cells (APCs), and lead to activation of the T lymphocytes and natural killer (NK) cells that destroy tumor cells. The development of a clinically relevant cancer requires evasion of this immune response. Immunotherapeutic approaches such as ICB aim to reverse this immune escape, restoring the body’s ability to reject invading tumor cells.
Head and neck squamous cell carcinoma (HNSCC) lies at the high end of the spectrum of immune-infiltrated cancers, making it a particularly appealing candidate for the use of immunotherapy. Human papillomavirus (HPV) + HNSCC is the third most immune-infiltrated human cancer, and HPV- HNSCC ranks seventh. 2 The degree of tumor-infiltrating lymphocytes (TILs) in HNSCCs varies both by HPV status and by anatomic subsite, with HPV + oropharyngeal squamous cell carcinoma (OPSCC) at the top. 3 The fact that oral cavity cancer has lower immune infiltration than some other subtypes provides a rationale for therapeutic approaches aimed at enhancing immune infiltration (i.e., Toll-like receptor [TLR], Stimulator of Interferon Genes [STING]) in addition to reversing immune suppression (i.e., ICB).
A variety of studies have demonstrated a correlation between tumor-infiltrating immune cells and positive outcomes. Infiltration of T cells into HNSCCs has been associated with increased locoregional control and survival, regardless of treatment modality. 4 – 6 Not only is the number of lymphocytes in the tumor microenvironment an independent prognostic factor, 7 but lymphocyte functionality is also associated with prognosis. 8 The prognostic value of immune cell infiltration extends far beyond HNSCC in the landscape of human cancers, and has also been demonstrated reproducibly in melanoma, 9 ovarian, 10 and colorectal cancers. 11
The inflamed microenvironment of HNSCC sets the stage for reversal of immune exhaustion. A variety of co-stimulatory and co-inhibitory immune checkpoint receptors modulate T-lymphocyte activation by APCs at the immune synapse (▶ Fig. 49.1). The use of monoclonal antibodies to block co-inhibitory immune checkpoints (i.e., programmed death-1 [PD-1], CTLA-4), or activate co-stimulatory checkpoints (i.e., CD40, OX40, 4-1BB, CD28) can alter the balance in favor of antitumor immunity.
PD-1 is of particular interest due to FDA approval of antibodies targeting this checkpoint. PD-1 is expressed on the surface of activated T cells and NK cells, and limits an overactive immune response, preventing autoimmune destruction. However, long-term antigen exposure in the setting of tumor development may lead to overexpression of PD-1 and T lymphocyte exhaustion. When engaged by its ligands, PD-L1 or PD-L2 (on tumor cells), PD-1 inhibits immune activation and allows tumor growth to proceed. Reversal of this immunologic suppression can lead to profound changes in immune response and patient prognosis.
50.3 Immunotherapy for Head and Neck Cancer
50.3.1 Previous Attempts
Prior to the advent of ICB, cetuximab was the first new agent to be approved by the FDA for HNSCC in many years. Cetuximab is a monoclonal antibody (mAb) that targets the epidermal growth factor receptor (EGFR) on the surface of HNSCC cells, and is approved by the FDA for the treatment of advanced HNSCC in combination with radiation, as monotherapy, or in combination with chemotherapy for recurrent/metastatic HNSCC. Cetuximab functions as an immunotherapeutic agent through activation of antibody-dependent cellular cytotoxicity (ADCC). The cetuximab antibody serves as a bridge that at one end (Fab region) binds EGFR on the surface of tumor cells and at the other end (Fc region) binds CD16/FcRIII on the surface of NK cells. This binding activates NK cells to release perforin and granzyme B, which lyse the tumor cell. Although 80% of HNSCCs overexpress EGFR, only 10-20% of patients respond to cetuximab, spurring efforts to combine it with additional agents to circumvent this resistence. 12
Additional approaches that have been evaluated with limited success in early phase HNSCC trials include vaccination and ACT. Antitumor vaccines aim to augment immune recognition of tumor cells through increased presentation of antigen by dendritic cells, and induce immunologic memory to prevent recurrence. Vaccination approaches involve injection of tumor-associated antigens 13 or injection of autologous dendritic cells loaded with p53 antigen ex vivo. 14 However, compared to other cancer types with well-defined tumor antigens (i.e., prostate cancer, for which the Sipuleucel-T vaccine is FDA approved), identification of a highly immunogenic peptide for broad use in HNSCC remains a challenge.
ACT is another approach that has shown promising results in melanoma, 15 , 16 and was recently approved by the FDA for treatment of acute lymphoblastic leukemia. Traditionally, ACT involves purification of autologous TILs from a surgical specimen, ex vivo stimulation, expansion, and re-injection of TILs that are active against the patient’s tumor. 15 , 16 These autologous T lymphocytes may also be engineered to express T-cell receptors (TCRs) or chimeric antigen receptors (CARs) with high specificity for particular tumor-associated antigens, thereby increasing the potency and specificity of the antitumor response. In HNSCC, a clinical trial evaluating CARs specific for HPV-16 E6 has recently been completed (NCT02280811).
50.3.2 Current State
Currently, the use of checkpoint inhibitors targeting PD-1 is the only immunotherapeutic strategy that has demonstrated clear clinical benefit in the treatment of HNSCC. Data from recent trials evaluating the PD-1 inhibitors pembrolizumab and nivolumab resulted in FDA approval in 2016 of both agents for the treatment of recurrent or metastatic HNSCC that progressed either during or after treatment with platinum-based chemotherapy.
The phase Ib KEYNOTE-012 trial of pembrolizumab in recurrent or metastatic HNSCC demonstrated an 18% overall response rate with median progression-free survival of 2 months and overall survival of 13 months, and a 17% rate of grades 3-4 adverse events. 17 The phase III CheckMate 141 trial compared nivolumab to standard therapy in patients with recurrent or metastatic, platinum-refractory HNSCC. 18 The identification of a significant survival benefit led to early conclusion of the trial: nivolumab reduced the risk of death by 30% compared to standard treatment with median overall survival extending from 5.1 to 7.5 months. With an overall response rate of 13.3% compared to 5.8% with standard therapy, nivolumab also more than doubled 1-year survival from 16.6 to 36%. Additionally, nivolumab was better tolerated, with only 13.1% experiencing grades 3-4 adverse events compared to 35.1% in the standard therapy group.
The KEYNOTE-040 phase III trial compared pembrolizumab to standard therapy in patients with recurrent or metastatic platinum-refractory HNSCC. 19 This study failed to meet its primary endpoint: a 19% improvement in overall survival achieved with pembrolizumab did reach the pre-specified criteria for statistical significance. Median survival with pembrolizumab was only marginally improved to 8.4 months compared to 7.1 months with standard therapy. However, the use of salvage immunotherapy in the standard arm may confound these results, and could contribute to 2 months longer median survival in the standard arm of KEYNOTE-040 compared to CheckMate 141. Similar to the results of CheckMate 141, only 13.4% experienced grades 3-5 adverse events with pembrolizumab compared to 36.3% with standard therapy.
The KEYNOTE-055 phase II trial also evaluated pembrolizumab in HNSCC that was refractory to both platinum and cetuximab. 20 In this population, an overall response rate of 16% was identified, with responses lasting for a median of 8 months. Although the response rates to ICB in these studies remain modest, the durability of benefit and translation into improved clinical outcomes in these heavily pre-treated populations provide the foundation for current and future work aimed at expanding responses and identifying patients likely to benefit.
50.4 Ongoing and New Trials
50.4.1 Ongoing Monotherapy Trials
The KEYNOTE-048 (NCT02358031) phase III trial is currently evaluating pembrolizumab as first-line treatment compared to standard of care in patients with recurrent/metastatic HNSCC. In addition, phase II trials are evaluating adjuvant pembrolizumab vs. placebo, in patients at high risk for recurrence (PATHWay, NCT02841748) or after salvage surgery in patients with relapsed, locally recurrent HNSCC (NCT02769520). A randomized phase II trial comparing pembrolizumab to methotrexate in frail or cisplatin-ineligible HNSCC patients has not yet begun recruitment (ELDORANDO, NCT03193931). Pembrolizumab is also under evaluation for patients with residual disease following definitive radiation (XRT) (NCT02892201).
50.4.2 Neoadjuvant Trials
When feasible, surgical resection or radiation remains the primary treatment for oral cavity cancer, with chemoradiation typically reserved for the adjuvant setting in patients with high risk features, or as upfront treatment for patients with unresectable disease. However, with the addition of checkpoint inhibitors to the armamentarium, interest has been renewed in the use of neoadjuvant therapy for HNSCC at large. 21 Theoretical advantages of immunotherapy in the neoadjuvant setting include intact lymphatics allowing for improved infiltration of immune cells to the tumor, reversal of lymphocyte exhaustion at an early stage, and presence of significant intact antigen to prime immune memory and keep postsurgical recurrence at bay. The successful use of neoadjuvant therapy may lead to early identification of responsive tumors, and the possibility of decreasing tumor burden to minimize the degree of resection or downstage unresectable to resectable disease. 21
A phase II trial evaluating neoadjuvant pembrolizumab in surgically resectable stage III/IV HNSCC is currently underway (NCT02296684). 22 Preliminary analysis identified pathologic response to a single dose of neoadjuvant pembrolizumab in 43% of patients at the time of surgery, and 48% of patients experienced clinical-to-pathologic downstaging. 22 The CheckMate-358 phase I/II study is an ongoing international multicenter study evaluating nivolumab in virus-associated cancers. A cohort of 29 patients with HPV + and HPV- resectable HNSCC were treated with two doses of nivolumab prior to surgery, and 48% experienced reduction in tumor size by the time of surgery. 23
50.4.3 Rationale Surrounding Combination Studies
Although clinical response rates to ICB have been low in the heavily pretreated recurrent/metastatic HNSCC population (13.3-18%), the clinical improvements seen in responders are impressive. 17 , 18 Tumors may downregulate co-stimulatory receptors or increase signaling through alternative immune checkpoints at the immune synapse (▶ Fig. 50.1) to maintain immunosuppression in the face of ICB. Efforts have turned to the development of combination therapy regimens to broaden responses through the simultaneous targeting of multiple immune mechanisms. For example, 4-1BB (CD137) is a co-stimulatory receptor that is the target of two mAbs (utomilumab and urelumab) in clinical development. Of additional interest are studies combining PD-1 inhibition with the CTLA-4 inhibitor ipilimumab, which was the first checkpoint inhibitor to achieve FDA approval in 2011. CTLA-4 is expressed on the surface of regulatory T lymphocytes (Tregs), and binds CD80/86 on the surface of APCs to competitively inhibit T-cell activation (▶ Fig. 50.1). 24 A phase III clinical trial comparing a combination of nivolumab (PD-1) and ipilimumab (CTLA-4) to the EXTREME regimen (cetuximab, cisplatin/carboplatin, and fluorouracil) is currently recruiting (NCT02741570).
Combinations of ICB with conventional chemotherapy and radiation are also of interest due to the tremendous release of antigen achieved with these cytotoxic therapies. Several studies are evaluating the timing of ICB administration, either concurrent with conventional therapy or as an adjuvant treatment. See ▶ Table 50.1 and ▶ Table 50.2 for a summary of active clinical trials evaluating nivolumab or pembrolizumab in combination therapy regimens.