Development of effective therapies for metastatic head and neck cancer has been challenging, and progress has been slow.
Charu Aggarwal, MD, MPH
Head and neck squamous cell cancer (HNSCC) is relatively uncommon in the United States compared with other malignancies such as non—small cell lung cancer (NSCLC) and breast cancer. There are about 500,000 cases worldwide every year. The majority of patients present with locally advanced disease, and most of them are treated in the multidisciplinary setting with multimodality approaches combining surgery, radiation, and chemotherapy. Human papillomavirus (HPV)-associated head and neck cancer has emerged as an almost separate disease entity, with a distinct tumor biology, patient characteristics, and a lack of associated conventional risk factors—such as smoking and drinking—typically associated with HNSCC.
Development of effective therapies for metastatic head and neck cancer has been challenging, and progress has been slow. In fact, the last approval for a drug for metastatic head and neck cancer, prior to 2016, was for cetuximab (Erbitux) in 2006. Bevacizumab (Avastin), a vascular endothelial growth factor receptor antagonist, has been associated with overall survival (OS) advantage in metastatic disease including lung cancer and colorectal cancer, where it has an established role, with a secure FDA approval. The role of bevacizumab in combination with a platinum-based backbone of first-line therapy for metastatic HNSCC was recently evaluated in a large, randomized phase III trial of over 400 patients. [The addition of bevacizumab did not provide a statistically significant improvement in OS but did improve progression-free survival (PFS) and response rates. Table.1]
A greater understanding of the immune microenvironment in HNSCC and the availability of active agents in the form of anti—PD-1/PD-L1 antibodies have made immunotherapy an attractive option for advanced HNSCC. Carcinogenesis-associated head and neck cancer, often seen in the HPV-negative setting, has many similarities to other tobacco-induced cancers such as lung cancer and bladder cancer. One of the characteristics is high mutational burden, which we know, based on several studies in melanoma and NSCLC, is associated with a higher response to immunotherapy, including improvements in PFS and OS. HPV-positive disease has a distinct biology. Despite its usual location within the lymphoid organ, it is able to escape immune detection and does so by several mechanisms, including upregulation of PD-L1, infiltration of tumor-infiltrating lymphocytes, and high expression of the PD-L1 receptor on T cells, all of which can promote adaptive resistance.
We have recently seen results of trials conducted in the metastatic, recurrent HNSCC setting that evaluated use of nivolumab (Opdivo), and pembrolizumab (Keytruda). KEYNOTE-012 was a phase Ib study with several treatment cohorts of patients with refractory solid tumors that initially enrolled patients with PD-L1 expression, followed by another cohort of patients who were not selected based on PD-L1 expression. KEYNOTE-055 was a phase II trial that included patients who were platinum and cetuximab refractory, and all of these patients were treated with a fixed dose of pembrolizumab.
Results from both of these trials were presented at the 2016 American Society of Clinical Oncology (ASCO) Annual Meeting. Response rates across the 2 trials were similar, at 17%. Responses were durable, with about half of the complete responses occurring after 6 months of therapy. Results from the CheckMate 141 study, evaluating nivolumab were presented at the same meeting. Again, response rates were very similar, but for the first time we observed an OS benefit, compared with investigator choice chemotherapy, in a platinum-refractory population. Results of a randomized phase III trial of pembrolizumab compared with investigator-choice chemotherapy for the treatment of metastatic/recurrent HNSCC were recently presented at the 2017 ESMO Congress. [In the KEYNOTE-040 trial, pembrolizumab reduced the risk of death compared with investigator-choice chemotherapy but the difference fell short of statistical significance. Table.1,2]
Although these agents are currently approved by the FDA, and have been associated with remarkable success in this otherwise difficult-to-treat refractory/metastatic population, there is a large unmet need for evaluation of combination immunotherapy and rationally sound scientific approaches that might move the needle to a higher response rate, improve outcomes, and prolong survival. At the 2017 ASCO Annual Meeting, we saw interesting data on combination immunotherapy—pembrolizumab plus indoleamine 2,3-dioxygenase inhibitors [epacadostat]—as well as incorporation of immunotherapy in the neoadjuvant and definitive treatment settings [Table].
The strategy for HPV-associated head and neck cancer, perhaps, should be different. Biologically, it is a distinct subset, with a strikingly different mutational profile. Tumor-directed “targeted immunotherapy” approaches are being explored, including DNA vaccine approaches, with preliminary results presented at the 2017 ESMO Congress.
While PD-L1 is an established predictive biomarker for NSCLC, its role as a predictive and prognostic biomarker for HNSCC is less clear. It is not currently used to guide clinical decision making. Other biomarkers are being evaluated, which may predict a response to immunotherapy. Mutational load and gene expression profile are 2 such biomarkers. Mutational load reflects tumor antigenicity while gene expression profile reflects activated T cells in the tumor microenvironment. Similar to data in other solid tumors, early data in metastatic HNSCCs reveal that these biomarkers may be predictive of response to immunotherapy. These are provocative data, as they may help us select the patients who may benefit the most from single-agent immunotherapy.
There have also been recent advances in the locally advanced disease setting. Chemoradiation therapy with cisplatin improves survival in patients with locally advanced HNSCC. High-dose cisplatin, administered at a dose of 100 mg/m² intravenously every 3 weeks, has been established as the standard of care in both the adjuvant and definitive chemoradiation therapy settings. There is, however, significant toxicity from the use of highdose cisplatin, and less than 50% of patients are able to receive all 3 doses. Weekly cisplatin is used as an alternative regimen; however, there are no large prospective clinical trials that directly compare these different administration schedules. At ASCO, we heard results of a randomized phase III noninferiority trial of weekly cisplatin, compared with high-dose cisplatin administered in the locally advanced setting for HNSCC. This trial was conducted in India in a primarily non-HPV population. Additionally, we eagerly await results of 3 large trials being conducted through RTOG Foundation, and ECOG-ACRIN Cancer Research Group, which will help us determine the best systemic therapeutic agent, dose, and schedule to use in combination with radiation therapy in the definitive and adjuvant disease settings. RTOG 1016 is clinical trial that randomized p16-positive patients to receive either cisplatin or cetuximab in combination with radiation therapy, for patients undergoing definitive chemoradiation therapy. ECOG 3311 is an adjuvant trial for p16 patients where adjuvant therapy is being determined based on pathology and risk features following transoral robotic surgical resection. RTOG 1221 is a similar trial, where adjuvant therapy is decided based on pathology and risk features in the p16-negative subset of patients with locally advanced disease.
In summary, we have witnessed several practice-changing advances in HNSCC. Further research is needed to optimize patient selection to help us identify patients who may benefit from single-agent immunotherapy, targeted therapy, combination therapies, or targeted immunotherapy.