Immunotherapy in head and neck cancers

Summary

Immunotherapy is a promising therapeutic modality that has rapidly gained its place in the treatment of many malignancies. The assumption of its effect in head and neck tumors stems from the high mutational load of these tumors. Phase I clinical trials were consistent with preclinical assumptions. Subsequent phase II and III trials confirmed the efficacy of the check-point inhibitors nivolumab (OPDIVO) in second-line palliation, and in the case of pembrolizumab, in first-line treatment of recurrent and metastatic head and neck cancer. In the Czech Republic, the use of pemobrolizumab was approved in 2014. In clearly defined indications, pembrolizumab therapy (Keytruda) is now also covered by public health insurance. However, in the Czech Republic, surgical treatment followed by adjuvant oncological treatment still remains the gold standard in the first-line treatment of head and neck cancer. Nevertheless, the approach is always individual and the best treatment modality is chosen based on the extent of the primary disease and the overall condition and preferences of the patient.

Keywords:

nivolumab – HPV – immunotherapy – checkpoint inhibitors – head and neck cancer – pembrolizumab

Introduction

Head and neck tumors represent a heterogeneous group of diseases in which multiple factors are involved. There are known modifiable risk factors such as smoking, alcohol consumption, and exposure to carcinogens such as nickel, chromium, lead, asbestos, fluorine, arsenic, hydrocarbons in soot, tar and industrial oils, and exposure to radioactive radiation [1]. Other risk factors for head and neck cancers are infection with human papillomaviruses (HPV) [2] that are associated with oropharyngeal tumors, or infection with Epstein-Barr virus (EBV), which gives rise to nasopharyngeal tumors [3].

Epidemiology

Per the latest GLOBOCAN estimates (2020), squamous cell carcinomas of the head and neck represent the seventh largest group of malignant tumors globally [4]. The incidence of head and neck tumors worldwide is reported to be 890,000/year and the mortality rate is 450,000/year [4, 5]. In the Czech Republic, around 1,800 head and neck cancers are newly diagnosed annually and 800 patients die every year because of this disease [6].

The main prognostic factors that influence the survival of patients with head and neck cancer are the stage of the disease, HPV status (in practice, immunohistochemical staining of the p16 molecule is used [7]), comorbidities, the general condition of the patient, and the response to previous therapy.

The prognosis of patients is influenced by the etiologic agent of malignant transformation. In general, HPV-induced oropharyngeal tumors have a better prognosis than HPV-negative head and neck tumors [8]. Both HPV-positive and HPV-negative tumors have a relatively high tumor mutational burden (TMB). Whereas in HPV-negative tumors, mutagenesis is mainly associated with smoking, and mutations in HPV-positive tumors accumulate due to inhibition of the tumor-suppressor protein p53 and retinoblastoma protein (pRb) via viral oncoproteins E6 and E7 [9]. Tumors with high TMB are considered immunogenic and according to in silico analyses, HNSCC tumors indeed belong among the 10 most immune infiltrated malignancies [10]. However, in depth analyses of the tumor microenvironment showed markedly higher levels of tumor-infiltrating lymphocytes in HPV-associated cancers compared to HPV-negative tumors [10]. As high densities of tumor-infiltrating immune cells, especially CD8+ T cells and B cells, were shown to be significantly correlated with patient’s prognosis [10], the immune response could at least partly explain a better therapeutic response of patients with HPV-positive tumors [9, 10]. Lifestyle also contributes to the longer survival of patients with HPV-positive tumors, with the majority of patients being non-smokers and only irregular alcohol consumers [11].

Treatment

Once the diagnosis is verified histologically and the extent of the tumor is determined, the choice of optimal treatment depends on several factors, including its location, the physical condition of the patient, consideration of the possible side effects of the planned treatment, and also on the patient‘s wishes.

The principal modalities of curative therapy for locally or locoregionally confined HNSCC are resection, radiation, and systemic therapy. Treatment planning should aim for the most highly curative approach, while optimizing preservation of function. In patients with a small primary cancer with no clinical nodal involvement, or involvement of only a single node, cure rates of over 80% can be achieved with single modality intervention (resection or radiation) [12]. Surgery is commonly used for oral cavity cancers, whereas radiation might be more commonly employed for pharyngeal and laryngeal cancers [13]. For laryngeal cancers, a moderately hypofractionated radiation schedule results in better locoregional control and survival than standard radiation therapy [14]. Advances in minimally invasive resection, including transoral robotic or laser resection and larynx-preserving partial laryngectomy, as well as improved reconstructive techniques, have extended the indications for primary surgical management in the hands of highly specialized head and neck surgical oncologists. Occult metastases in the draining cervical lymph nodes might be present even in patients with small, invasive primary tumors, and use of elective neck dissection improves survival. In the event of treatment failure after single modality radiation or surgery, salvaging with the alternative modality offers a high chance of being cured. For tumors with more advanced tumor or nodal stage, postoperative radiation or chemoradiation, guided by pathological risk factors, reduces the risk of recurrence and improves survival [12].

Nowadays in the Czech Republic, the basic treatment modality for head and neck tumors is surgical resection followed in most cases by adjuvant radiotherapy, which may be supplemented or combined with chemotherapy. Another possible treatment is primary concomitant chemoradiotherapy or radiotherapy alone. In case of inoperable local recurrence of the primary disease in the already irradiated area and in case of metastatic process, systemic therapy is the only option. For a long time, it was based on a platinum derivative or taxane. However, with the development of biological therapy, the anti-epidermal growth factor receptor (anti-EGFR) antibody cetuximab was introduced [1].

Cetuximab with chemotherapy represented the gold standard in the first--line treatment of relapsed and/or metastatic squamous cell carcinomas of the head and neck before the introduction of checkpoint inhibitors [1].

Immunotherapy

Programmed death-1 receptor (PD-1) expressed on activated and exhausted T cells has been shown to promote immunosuppression upon interaction with its ligands, PD-L1 and PD-L2, located on cancer cells and tumor-infiltrating immune cells. PD-L1 expression on immune cells in pre-treatment tumor biopsies is indicative of a formerly elicited anti-tumor adaptive immune response and has been associated with improved treatment outcomes. Consequently, blocking PD-1/PD-L1 interaction with anti-PD-1 or anti-PD-L1 monoclonal antibodies (mAbs) promotes the reactivation of immunity, resulting in durable anti-tumor effects in a fraction of patients with various solid tumors. Current evidence in HNSCC suggests that a combined positive score (CPS), the number of PD-L1 positive cells including tumor cells, lymphocytes, and macrophages, in relation to the total number of tumor cells, offers a more effective evaluation than the tumor proportion score (TPS) (Fig. 1), which measures PD-L1 expression on tumor cells alone [15].

When checkpoint inhibitors are considered for administration, it is now common practice to determine PD-L1 expression in tumor tissue or stroma to predict the response to treatment [16]. The combined positive score (CPS) used in most clinical trials is calculated as the number of PD-L1-positive cells (tumor cells, macrophages, and lymphocytes) /total number of tumor cells × 100 [17].

The cutoff value of PD-L1 positivity chosen by different clinical institutions varies, mainly consisting of 1, 5, 10, and 50%. A study in renal cell carcinoma showed that the objective response rate of atezolizumab in patients with PD-L1-positivity was 56% when the cutoff value of CPS was 5%, whereas the objective response rate of patients with negative PD-L1 expression was 25%. However, when the cutoff value was set to 1%, the objective response rates of positive and negative patients were 50% and 55%, respectively, with no statistical significance. Keynote-024 set the TPS cutoff value to 50% or greater and showed that pembrolizumab significantly improved the prognosis of patients with NSCLC and PD-L1 TPS ≥50%, with OS extended to 30 months, compared with 14.2 months for patients receiving chemotherapy. Generally, a higher cut- -off value indicates a positive correlation with a more accurate prediction; however, it also indicates reduced sensitivity of the test and increased false negatives, which might neglect some patients who could benefit from anti-PD-1/PD-L1 therapy. Therefore, a method to determine practicable cutoff values of PD-L1 positivity is still required [15].

Anti-PD-1 monoclonal antibodies – pembrolizumab (KEYTRUDA) and nivolumab (OPDIVO) – were used in most clinical trials and are the only immune checkpoint inhibitors approved for HNSCC treatment so far. Pembrolizumab is a humanized monoclonal antibody against the PD-1 receptor (IgG4/k isotype with a stabilizing sequence change in the Fc region) generated by recombinant DNA technology in Chinese hamster ovary cells [18]. Nivolumab is a fully human, monoclonal IgG4 antibody which binds to the PD-1 receptor with high specificity and affinity. PD-1 is expressed on T cells and dampens their immune response upon binding ligands PD-L1 and PD-L2 on antigen-presenting cells (APCs) [19, 20].

The effect of treatment with check-point inhibitors was demonstrated in the MK-3475-012/KEYNOTE-012 study in phase 1b, in the multicenter MK-3475-055/KEYNOTE-055 study in phase II, and subsequently in the MK-3475-040/KEYNOTE-040 study in phase III. In each case, the treatment effect was evaluated in patients with recurrent or metastatic head and neck cancer who progressed within 6 months or 3 to 6 months of completion of platinum chemotherapy [1].

In the KEYNOTE-048 trial (NCT023 58031), which tested pembrolizumab in first-line palliative treatment, patients were randomized into three arms – the first group of patients was treated with pembrolizumab, the second group with platinum/5-FU chemotherapy and pembrolizumab, and the third group with platinum/5-FU chemotherapy and cetuximab. In the overall patient population, pembrolizumab alone did not significantly improve survival (11.5 vs. 10.7 months). However, this study demonstrated the benefit of combining pembrolizumab with chemotherapy compared to chemotherapy with cetuximab (median survival 14.1 vs. 11.0 months in patients with PD-L1 CPS ≥20, and median survival 13.6 vs. 10.4 months in patients with PD-L1 CPS ≥1). In the overall population, a prolonged survival of 13.0 vs. 10.7 months was observed. Monotherapy with pembrolizumab compared with combination chemotherapy with cetuximab improved overall response rate in the CPS ≥20 group; median survival was 14.9 vs. 10.7 months and 12.3 vs. 10.3 months in the CPS ≥1 group [18]. The toxicity of combination therapy was comparable in the group combining chemotherapy with pembrolizumab and chemotherapy with cetuximab-the EXTREME trial (G3–4 85 vs. 83.3%). The group with pembrolizumab immunotherapy alone had the lowest toxicity, with G3–4 toxicities reported in 54.7% [21].

Immunotherapy is only applicable in a few indications. KEYTRUDA (pembrolizumab) is indicated in the Czech Republic as monotherapy or in combination with platinum and fluorouracil (5-FU) chemotherapy for the first-line treatment of metastatic or surgically unresectable recurrent squamous cell carcinoma of the head and neck in adults whose tumors express PD-L1 with a CPS ≥ 1. It is also indicated in monotherapy for the treatment of recurrent or metastatic squamous cell carcinoma of the head and neck in adults whose tumors express PD-L1 with TPS ≥50% and who are undergoing or have undergone platinum-containing chemotherapy [22]. Similarly, OPDIVO (nivolumab) is registered for monotherapy in the treatment of recurrent or metastatic squamous cell carcinoma of the head and neck in patients who have relapsed from disease during treatment involving a platinum derivative [23].

Immune checkpoint inhibitor toxicity in HNSCC

Immunotherapy, like any other treatment modality, has side effects. The most common reported adverse events in HNSCC are dermatologic, endocrine, and gastrointestinal.

Dermatologic toxicities are one of the most reported immune-related adverse events (irAEs) associated with immune checkpoint inhibitors and occur in 8–19% of HNSCC patients treated with immune checkpoint inhibitor agents. The majority of cases are usually low grade, ranging from pruritus and rash to dermatitis. However, a serious skin reaction is less common (grades 3–4 are less than 2%). Cases of Stevens-Johnson syndrome are practically notable, and one case resulted in a treatment-related death with pembrolizumab treatment [24].

Endocrine toxicity is a common side effect of immune checkpoint inhibitor therapy in HNSCC patients. The most frequent endocrinopathies are hypothyroidism (6.8–16%), hyperthyroidism (3%), and hypophysitis (less than 1%). Rare cases of primary adrenal insufficiency, hypercalcemia, and immune-related type 1 diabetes mellitus, leading to hyperglycemia and diabetic ketoacidosis, have been reported [24].

Diarrhea is one of the more frequently observed irAEs with immune checkpoint inhibitors (5.4–10.4%), and the majority of cases are mild [24].

Less common side effects are hepatitis (1–8% of HNSCC patients treated with immune checkpoint inhibitors) or asymptomatic elevation of aspartate aminotransferase and alanine transferase, and pneumonitis (a noninfectious inflammation of the lungs) that occurs in 1–4% of HNSCC patients receiving immune checkpoint inhibitor therapy. Arthralgia and myalgia are the most common musculoskeletal side effects occurring in 2–6% of HNSCC patients treated with immune checkpoint inhibitor agents [24].

General management of irAEs includes treatment interruption, immunosuppression or immunomodulatory agents, and hormone replacement. Glucocorticoid therapy was commonly used for moderate to severe irAEs, but this routine therapy may not be suitable for all adverse events [24].

Future outlook – neoadjuvant immunotherapy

The established efficacy of immune checkpoint inhibitors in the recurrent and metastatic setting has produced widespread interest in their neoadjuvant use. Neoadjuvant immunotherapy in HNSCC is appealing for a host of reasons. Incorporation of immunotherapy in the curative setting might reduce the risk of local recurrence and distant metastases thereby improving overall survival. Opportunities to potentially de-escalate adjuvant therapy or the potential to reduce morbidity and the necessary extent of surgical resection are intriguing as well. Investigational use of neoadjuvant immunotherapy also offers an ideal platform to perform correlative studies to assess whether improved predictors of response to immunotherapy might be found. However, neoadjuvant immunotherapy is expected to be more easily tolerated than TPF (docetaxel, cisplatin, fluorouracil), and as a result would be unlikely to limit dose intensity of adjuvant cisplatin-based chemoradiation when indicated [25].

Predictors of the response to neoadjuvant therapy in HNSCC are lacking. Many recent studies have built-in robust pre-specified correlative analyses to potentially identify signatures that predict the response or lack thereof to immunotherapy. While much data remains preliminary, there are early signals toward association of a response to therapy with PD-L1 expression and high pre-treatment TMB. Additionally, at the time of surgery in HNSCC, increased infiltration of CD8, CD26, and Tim3 positive TIL (tumor-infiltrating lymphocytes) were observed in responders, while an increase in effector memory T cells was noted in non-responders [25]. Increased NK cell infiltrate has also been observed in responders at the time of surgery [26].

The promise of defining predictors of a response to immunotherapy is increasingly real and may help select patients who would benefit from neoadjuvant immunotherapy and those who should instead proceed to current standard of care treatments. Immunotherapy has been changing the treatment landscape of recurrent and metastatic disease and now promises to change the face of curative intent therapy as well [27].

Discussion

Head and neck tumors are a very heterogeneous group of diseases and therefore require personalized treatment. The gold standard is surgical removal of the tumor and local metastases followed by radiotherapy if indicated. For inoperable tumors, primary oncological treatment with concomitant chemoradiotherapy or radiotherapy alone is the method of choice [1]. In the case of recurrent or metastatic head and neck cancer, it is necessary to individually assess which treatment will be most beneficial for the patient. Previously, the only treatment of choice was cetuximab, an antibody against the epidermal growth factor receptor, in combination with chemotherapy. In the Czech Republic, this therapy is reserved for patients who have already been treated with platinum-derivative chemotherapy, and who, despite this, have relapsed or developed metastatic disease. A newer treatment option for these patients is the anti-PD-1 antibody pembrolizumab or nivolumab, which according to studies in combination with chemotherapy, prolong overall survival while reducing treatment toxicity [1]. In addition, studies show that pembrolizumab prolongs overall survival even when administered as monotherapy in the first line of treatment and is also the least toxic treatment [21]. In the Czech Republic, Keytruda is used in the above-mentioned indications [22]. Immunotherapy would certainly represent a promising therapeutic modality for first-line treatment of primary disease in indicated cases, but this option is currently not considered in the Czech Republic. Many studies are now ongoing to determine whether neoadjuvant immunotherapy would be appropriate for some patients. The object of further research is to select patients who might benefit from this new promising therapeutic approach [27].

Conclusion

Immunotherapy for head and neck cancer is currently approved in the Czech Republic for the first-line treatment of metastatic or unresectable recurrent squamous cell carcinoma of the head and neck in adults whose tumors express PD-L1 with CPS ≥1. In addition, KEYTRUDA is indicated in monotherapy for the treatment of recurrent or metastatic squamous cell carcinoma of the head and neck in adults whose tumors express PD-L1 with TPS ≥50% and who are receiving or have received platinum-based chemotherapy. Pembrolizumab prolongs overall survival, and in the case of combination with chemotherapy, reduces the dose of chemotherapy required, thereby reducing the overall toxicity of this treatment. In monotherapy, this checkpoint inhibitor shows comparable results and the lowest toxicity ever. Immunotherapy already has an important place in the treatment of head and neck cancer that may be further expanded in the near future.

In conclusion, surgical treatment followed by adjuvant oncological treatment remains the gold standard in the first-line treatment of head and neck cancer in the Czech Republic. Nevertheless, the best treatment modality is chosen individually based on the extent of the primary disease and the overall condition and preferences of the patient.

Funding

This work was supported by the Charles University Cooperation Program, research area “Surgical Disciplines”.

Conflict of interest statement

The author of this paper declares that she has no conflict of interest in relation to the topic, development, and publication of this paper and that the development and publication of this paper was not supported by any pharmaceutical company. This declaration also applies to all co-authors.