KeywordsSystemic therapy, active agents, radiation therapy, chemotherapy
The role of systemic therapy in the treatment of head and neck cancer has increased in recent decades. Randomized clinical studies of integrated chemotherapy/radiotherapy programs have demonstrated improvements in locoregional control, organ preservation, and overall survival. However, high-grade acute toxicity and long-term sequelae of treatment remain a significant problem.
In contemporary multimodality therapy, chemotherapy may be given before surgery or concurrently with radiation therapy. For patients who undergo primary surgery for a disease that carries a poor risk of recovery, concurrent administration of high-dose cisplatin with postoperative radiation therapy often is recommended. The prognosis for patients with recurrent or metastatic disease remains poor in most cases, but treatment options are expanding with the development of novel targeted therapies that are based on a gradual improvement in the understanding of the molecular pathology of the disease and harnessing the power of the immune system to mediate tumor rejection. It is also now well appreciated that human papilloma virus (HPV) status of tumors favorably influences the prognosis for patients with newly diagnosed disease amenable to localized, combined modality therapy ( Ang et al., 2010 ). The significant impact of HPV on clinical outcomes is now reflected in the revised staging system for head and neck cancers (eighth edition of the American Joint Committee on Cancer staging manual [ Amin et al., 2017 ]). While decisions regarding standard of care treatment for head and neck cancers are currently not informed by HPV status, the evaluation of new treatment paradigms for HPV-positive head and neck cancer is currently a strong research focus that potentially could influence practice in the years to come.
Active Cytotoxic Agents in Head and Neck Squamous Cell Cancer
Platinum agents, taxanes, and antimetabolites all have clinical activity against some head and neck cancers ( Table 20.1 ). Cetuximab, a chimeric monoclonal antibody directed against the extracellular domain of the epidermal growth factor receptor (EGFR), was more recently added to the list of systemic therapies for head and neck cancer. Other drugs such as ifosfamide, vinca alkaloids, hydroxyurea, and bleomycin also have some activity in this disease but are less commonly used in clinical practice.
|Class of Agent||Examples||Main Mechanism of Action|
|Platinum||Cisplatin, carboplatin||Form DNA cross-links|
|Taxane||Paclitaxel, docetaxel||Stabilize microtubules to block M-phase|
|Antifolate (antimetabolite)||Methotrexate||Inhibit dihydrofolate reductase (DHFR) during S-phase|
|Fluoropyrimidine (antimetabolite)||5-Fluorouracil||Inhibit thymidylate synthase (TS) during S-phase|
Cisplatin ( cis -diamminedichloroplatinum), the agent with the most data to support its use in the combined modality setting, is a platinum coordination complex that cross-links with deoxyribonucleic acid (DNA) and also can cross-link proteins with DNA. Cisplatin was discovered serendipitously in 1965 when it was observed that a substance produced by platinum electrodes was toxic to bacteria in laboratory experiments and was subsequently shown to have significant activity against a variety of human tumor cell lines. Cisplatin is now a standard therapeutic agent for patients with a wide range of malignancies, including head and neck squamous cell cancer. Renal toxicity is the dose-limiting toxicity. It is now widely understood that safe administration of cisplatin requires extensive intravenous hydration before and after the cisplatin infusion.
As with other cytotoxic chemotherapies, cisplatin causes transient myelosuppression, during which time the patient may be at increased risk of infection. Cisplatin is potently emetogenic, but cisplatin-induced emesis usually can be well controlled or prevented with contemporary antiemetic medications. Permanent peripheral neuropathy and ototoxicity (most typically, high-frequency hearing loss) may occur with use of cisplatin. As with most cytotoxic chemotherapies, fatigue tends to worsen with cumulative treatment cycles. Carboplatin appears to have the same general mechanism of action as cisplatin, but the adverse effect profile differs in that carboplatin is less nephrotoxic but more myelosuppressive than cisplatin. Carboplatin has not been studied as extensively as cisplatin in persons with head and neck cancer, but the available evidence suggests that carboplatin has a lower antitumor efficacy than cisplatin in persons with this disease.
Paclitaxel and docetaxel have significant activity against head and neck cancer. Paclitaxel was isolated from the bark of the Pacific yew tree in 1971, and docetaxel is a semisynthetic taxane. Taxanes work by binding to microtubules as cells attempt to divide, thereby causing cell cycle arrest in the M phase. Common adverse effects of taxanes include fatigue, alopecia, myelosuppression, and peripheral neuropathy. Taxanes are significantly less emetogenic than cisplatin, and the peripheral neuropathy is more likely to be reversible. Taxanes are metabolized in the liver, and these drugs generally are not administered to patients with significant liver dysfunction. Patients must receive steroid premedications to reduce the risk of an anaphylactic reaction to components of the lipid solvent in which paclitaxel is administered. As with platinum compounds, taxanes can serve as radiosensitizers in the treatment of patients with head and neck cancer at the cost of increased mucositis in the radiation field.
Methotrexate and 5-fluorouracil (5-FU) are antimetabolites that can yield clinically useful activity against head and neck cancer. Methotrexate was developed in the 1950s and inhibits dihydrofolate reductase (DHFR), the enzyme that catalyzes the production of tetrahydrofolate. The resulting intracellular folate deficiency limits DNA synthesis and cell division. Methotrexate typically is administered to patients with advanced head and neck cancer as a palliative agent at low doses, and it is generally well tolerated in this setting. Low-dose palliative methotrexate may be associated with myelosuppression, fatigue, nausea, and mucositis. The drug 5-FU inhibits thymidylate synthase, thereby interfering with DNA synthesis and cell division. In head and neck cancer treatment, 5-FU most commonly is administered as a continuous infusion over 4 to 5 days in combination with other chemotherapy and/or radiotherapy. Common adverse effects of 5-FU are bone marrow suppression, diarrhea, and mucositis. Patients with low levels of dihydropyrimidine dehydrogenase, the enzyme that catabolizes 5-FU, can increase the risk of severe toxicity when treated with 5-FU.
Cetuximab is a chimeric monoclonal antibody therapy that targets the extracellular domain of the EGFR ( Fig. 20.1 ) and is approved for the treatment of head and neck cancer. Almost all head and neck squamous cell cancers express EGFR, which is the prototype of the erb-B/HER family of type I receptor tyrosine kinases. Dysregulation of EGFR signaling activity appears to contribute to the malignant phenotype in a significant percentage of head and neck cancers, although it is unclear if the clinical efficacy of this therapeutic antibody is linked to inhibiting EGFR signaling or mediating antibody-dependent cell-mediated cytotoxicity. Cetuximab possesses modest activity as a single agent and can be an effective radiosensitizer in head and neck cancer. The most common adverse effect of cetuximab is an acne-like rash. A small risk exists of a severe allergic infusion reaction during a patient’s first exposure to cetuximab.
Of these agents, cisplatin is the drug with the most data to support its use in combination with curative-intent radiotherapy. For patients with medical comorbidities that render them suboptimal candidates for treatment with cisplatin concurrent with radiation therapy, other options such as concurrent cetuximab may be preferable. Combination chemotherapy regimens have significant activity in the induction chemotherapy setting for previously untreated patients, although the activity of these drugs is lower in patients with recurrent disease. In the palliative treatment setting, the choice of combination chemotherapy versus monotherapy must be individualized, balancing quality of life issues against the goal of tumor control.
The remainder of this chapter will review the clinical literature that guides the selection of systemic therapies for patients with head and neck cancer.
The initial proof-of-concept for induction chemotherapy for this disease dates back to The Head and Neck Contracts Program Trial, the first large-scale study to evaluate the role of sequential chemotherapy in the management of advanced, resectable squamous cell cancer of the head and neck ( Fig. 20.2 ). No significant differences could be demonstrated in locoregional control or 5-year disease-free and overall survival times among the three arms of the study. However, patients treated with neoadjuvant/adjuvant chemotherapy had a lower rate of distant recurrence. Of note, only 9% of patients received all six cycles of single-agent cisplatin adjuvant therapy, potentially contributing to the lack of observed survival benefit. Upon subsequent multivariate analysis, only patients with N2 disease had a significant survival benefit with the neoadjuvant/adjuvant chemotherapy. The neoadjuvant arm demonstrated the feasibility of administering primary chemotherapy in patients with head and neck cancer but failed to demonstrate a significant impact of induction chemotherapy on clinical outcome.
In the 1980s, the combination of cisplatin plus 5-FU (PF) emerged as the standard induction chemotherapy regimen. In a randomized phase III study, Paccagnella compared the PF induction chemotherapy regimen (cisplatin, 100 mg/m 2 on day 1 plus 5-FU, 1000 mg/m 2 /day with a 24-hour infusion on days 1–5 administered every 21 days for four cycles) before locoregional therapy versus locoregional therapy alone for patients with stage III or IV disease without distant metastases. Patients with either operable or inoperable disease were eligible. Although no significant difference in overall survival between the two treatment groups was found when all patients were considered, a modest survival advantage was associated with induction chemotherapy in a subgroup analysis restricted to patients with inoperable disease.
The Veterans Affairs Laryngeal Cancer Study Group established the larynx preservation paradigm using the PF regimen. Dramatic responses were seen in a significant number of patients with preservation of a functional larynx ( Fig. 20.3 ). As Fig. 20.4 shows, patients were randomly assigned to standard treatment (i.e., primary laryngectomy followed by radiotherapy) or organ preservation therapy (i.e., induction chemotherapy with cisplatin and 5-FU, followed by radiotherapy). The overall and median survival rates were not significantly different between the two arms, but larynx preservation was reported in 66% of the survivors in the induction chemotherapy arm. Induction PF also was used in a study of the larynx-preservation approach for squamous carcinoma of the hypopharynx, which was conducted by the European Organization for Research and Treatment of Cancer (EORTC) ( Fig. 20.5 ). Disease control and survival rates were comparable between the two arms, and the larynx could be preserved in 35% of patients in the chemoradiation arm.
The addition of taxanes to PF induction chemotherapy for patients with stage III or IV disease with no distant metastases yields superior outcomes compared with PF alone. The TAX 323 and TAX 324 studies randomly assigned patients to three cycles of PF versus three cycles of TPF (PF plus docetaxel, 75 mg/m 2 on day 1). In both studies, the total dose per cycle of 5-FU was reduced in the TPF regimens compared with the PF regimens. TAX 323 was restricted to patients with unresectable disease. Subtle differences in the dose and schedule of cisplatin and 5-FU existed in the TPF regimens in these two studies ( Figs. 20.6 and 20.7 ). After induction chemotherapy, patients received definitive radiation therapy alone in TAX 323 or with concurrent weekly carboplatin (area under the curve of 1.5) in TAX 324.
Both studies demonstrated superior outcomes with TPF compared with PF. In TAX 323, overall response rates after induction chemotherapy (68% versus 54%) and 3-year overall survival (37% versus 26%) were significantly higher for patients treated with TPF versus PF. Similarly, in TAX 324, the overall response rates after induction chemotherapy (72% versus 64%, p = .07) and 3-year overall survival were superior for the TPF group (62% versus 48%). These results are consistent with those of a randomized trial reported by Hitt and colleagues that evaluated the addition of paclitaxel to cisplatin and 5-FU in patients with stage III or IV disease without distant metastasis. The addition of paclitaxel yielded a significant improvement in response rate to induction chemotherapy and a trend toward improvement in overall survival.
These trials, however, were not designed to compare the strategy of induction chemotherapy followed by chemoradiation versus primary chemoradiation. Several phase III randomized clinical trials that followed failed to demonstrate a significant efficacy advantage with this sequential approach ( Cohen et al., 2014 ; Haddad et al., 2013 ; Hitt et al., 2014 ). The negative results in the PARADIGM and DeCIDE trials have been attributed to poor accrual and unexpected favorable outcomes in the control arms ( Cohen et al., 2014 ; Haddad et al., 2013 ), and the Spanish Head and Neck Cancer Cooperative Group (TTCC) study has only been reported with relatively short follow-up ( Hitt et al., 2014 ). The only positive phase III clinical trial demonstrating superior outcomes with TPF induction followed by chemoradiation over chemoradiation alone was reported in abstract form by Italian investigators from Gruppo di Studio Tumori della Testa e del Collo (GSTTC). Taken together, these results suggest that more investigation is required to better elucidate the benefit of induction chemotherapy, and perhaps more importantly better defining the patient population who benefits from the sequential approach.