Systemic Targeted Therapy for Recurrent and Metastatic Head and Neck Squamous Cell Cancer

Matthew G. Fury

David G. Pfister

This chapter reviews chemotherapy options for patients with recurrent and metastatic (R/M) head and neck squamous cell cancer (HNSCC) and discusses several avenues of clinical research for this patient population.

For locoregional recurrence of disease, surgical resection or ablative procedures may provide long-term disease control in a minority of patients.¹ In select patients, reirradiation may be applied, although most patients do not experience durable disease control.² Further details regarding resection and reirradiation are noted elsewhere in other chapters. Selected patients with limited metastatic lesions may also be considered for surgical resection if their performance status allows, or ablative procedures if surgery is not planned.³^,⁴

For patients with R/M disease that is not deemed suitable for local therapy, systemic therapy given with palliative, not curative, intent is appropriately considered. In patients with lungonly metastasis(es), it may be difficult to clearly distinguish metastatic disease from a new lung primary. Clinical factors may help differentiate these two possibilities. For patients in whom the primary was an oropharynx tumor associated with the human papillomavirus (HPV), detection of HPV in the lung lesion can be used to help establish the diagnosis of metastatic disease.⁵

As reviewed later in this chapter, several drugs have activity in HNSCC. However, no single chemotherapeutic agent has consistently or conclusively demonstrated superiority in terms of overall survival when compared with another drug in randomized clinical trials. It should be noted that the interpretability of some randomized trials, particularly the older ones, is compromised by modest sample sizes that increase the risk of false negative results. The selection of systemic therapy for patients with R/M HNSCC takes into account many variables, including patient’s performance status, medical comorbidities, and prior treatment history.⁶ For example, single-agent palliative therapy may be appropriate for minimally symptomatic patients with low-volume disease or frail patients with a decreased performance status, whereas doublet or triplet chemotherapy regimens may be appropriate for fit patients with more symptomatic or higher-volume disease. Table 30.1 lists clinical factors that appear to predict outcomes among patients with R/M HNSCC.⁷^,⁸^,⁹^,¹⁰ Tumor status with regard to the presence or absence of the HPV has robust associations with response to therapy, and prognosis in patients with locally/regionally advanced disease,¹¹ but the potential impact of HPV status on outcomes in the R/M disease setting has not been clearly established.

CONVENTIONAL CYTOTOXIC AGENTS

Platinum Agents and Antimetabolic Agents

Cisplatin was among the first agents to demonstrate significant activity against HNSCC.¹¹ A randomized phase III trial with a 2 × 2 factorial design (n = 116 subjects) compared (a) cisplatin, (b) bleomycin, (c) cisplatin + bleomycin, and (d) no chemotherapy.¹³ Among all patients treated with chemotherapy, the response rate was 25 %. Treatment with cisplatin versus “no chemotherapy” was associated with a 10-week improvement in median overall survival.¹² Subsequent studies generally have not included a “no chemotherapy” group, due to both logistical and ethical concerns about such designs after this initial proof-of-principle.

Cisplatin and methotrexate have antitumor efficacy as single agents for patients with advanced HNSCC. In a randomized comparison (n = 44 patients) of cisplatin versus methotrexate in patients with recurrent HNSCC, response rates (P, 29%; M, 24%) and median survivals (˜ 6 months) were similar in both arms.¹⁴ Cisplatin and methotrexate yielded comparable median survival times (18 and 20 weeks, respectively) in a randomized study for patients (n = 100) with advanced head and neck cancer.¹⁵

The PF (cisplatin, 5-fluorouracil) regimen achieved a response rate of 70% in a single-institution study,¹⁶ although response rates were lower in subsequent studies. A three-arm randomized comparison of (PF vs. P vs. F, n = 249 subjects) demonstrated superior response rate in the PF group (PF, 32%; P, 17%; F, 13%), but hematologic toxicity was also higher for PF, and median survival was approximately 5.7 months for all study arms.¹⁷

Cytotoxic chemotherapy combination regimens historically have yielded higher objective response rates compared with monotherapy, but generally do not improve overall survival for patients with R/M HNSCC. The combination regimens, generally, are associated with increased treatment-related toxicities compared with monotherapy, which is a major concern in the palliative setting. For example, the EORTC randomized 382 subjects patients with R/M HNSCC to either PF versus CABO (cisplatin, methotrexate, bleomycin, and vincristine) versus cisplatin monotherapy.¹⁸ PF and CABO yielded the higher response rates (PF, 31%; CABO, 34%; P, 15%) at the expense of greater hematologic and nonhematologic toxicity, but there was no significant difference in overall survival among the three arms.

Carboplatin is generally well tolerated and may be preferable to cisplatin due to favorable adverse event profile as palliative
therapy for some patients with advanced head and neck cancer. In a phase II study for patients (N = 31) with R/M HNSCC, carboplatin monotherapy was associated with a response rate of 26%.¹⁹ In a subsequent study, carboplatin achieved a response rate of 24% among 29 subjects with recurrent HNSCC.²⁰

TABLE 30.1 Factors Associated with Inferior Overall Survival in R/M HNSCC Patients

Low performance status

Lack of response to prior chemotherapy

Prior radiation therapy

Hypopharynx or oral cavity primary site

Weight loss >5%

Malignant hypercalcemia (end-stage disease)

Source: From Deleyiannis FW-B, Thomas DB, Vaughan TL, et al. Alcoholism: independent predictor of survival in patients with head and neck cancer. J Natl Cancer Inst. 1996;88:542-549; Lee J-J, Jeng J-H, Wang H-M, et al. Univariate and multivariate analysis of prognostic significance of betel quid chewing in squamous cell carcinoma of the buccal mucosa in Taiwan. J Surg Oncol. 2005;91:41-47; Argiris A, Li Y, Forastiere A. Prognostic factors and long-term survivorship in patients with recurrent or metastatic carcinoma of the head and neck: an analysis of two Eastern Cooperative Oncology Group randomized trials. Cancer. 2004;101: 2222-2229; Penel N, Berthon C, Everard F, et al. Prognosis of hypercalcemia in aerodigestive tract cancers: a study of 136 recent cases. Oral Oncol. 2005;41: 884-889, with permission.

TABLE 30.2 Results of Selected Chemotherapy Trials in Recurrent and Metastatic Head and Neck Squamous Cell Cancer

First Author	Year	Number of Subjects	Treatment Groups	Response Rate (%)	Median Overall Survival (Months)
Hong¹⁴	1983	44	Cisplatin	29	6.3
			Methotrexate	24	6.1
Grose¹⁵	1985	100	Cisplatin	8	4.1
			Methotrexate	16	4.6
Jacobs¹⁷	1992	249	Cisplatin + 5FU	32*	5.5
			Cisplatin	17	5.0
			5FU	13	6.1
Forastiere²¹	1992	227	Cisplatin+ 5FU	32*	6.6
			Carboplatin + 5FU	21	5.0
			Cisplatin	10	5.6
Clavel¹⁸	1994	382	CABO^a	34*	6.7^b
			Cisplatin+ 5FU	31*
			Cisplatin	15
Gibson²⁴	2005	218	Cisplatin + 5FU	27	8.7
			Cisplatin + Paclitaxel	26	8.1
Vermorken⁵⁴	2008	442	Platinum (Cis or Carbo) + 5FU + Cetuximab^c	36^d	10.1^d
			Platinum (Cis or Carbo) + 5FU	20	7.4
^aCABO = cisplatin, methotrexate, bleomycin, and vincristine. ^bOverall survival was 29 weeks (6.7 months) for the entire study, with no significant differences between groups. ^cIn the experimental arm, maintenance cetuximab monotherapy was allowed after 6 cycles of triplet therapy. ^dStatistically significant difference compared with control arm.

A three-arm randomized trial (n = 261 subjects) compared PF versus carboplatin plus 5-fluorouracil (CbF) versus methotrexate (M).²¹ The starting Cb dose was 300 mg/m², with guidelines for carboplatin dose escalation to achieve grade 2 myelosuppression. Response rates by group were PF 32%, CbF 21 %, and M 10 %. Adverse events were also more common in the PF group, and there was no significant difference in median overall surival between the three groups. Limitations of this study include (a) it was not statistically powered to provide a definitive comparison of PF versus CbF, (b) it used a dosing strategy Cb that is not used widely in clinical practice, and (c) the PF schedule was every 3 weeks and the CbF schedule was every 4 weeks.²¹

Table 30.2 summarizes results from selected clinical trials involving platinum-based regimens for patients with R/M HNSCC.

Paclitaxel and Docetaxel

Early nonrandomized studies of taxane monotherapy demonstrated impressive objective response rates (paclitaxel, 40%; docetaxel 42%) among patients who had not received prior chemotherapy for R/M HNSCC.²²^,²³ The combination of cisplatin
plus paclitaxel (PT) was compared with PF in a randomized study (n = 218 subjects). No significant difference in response rate (PF, 27%; PT, 26%) or overall survival (PF, 8.7 months; PT group, 8.1 months) was observed between the two groups. Gastrointestinal and hematologic toxicities were less common with PT.²⁴ PT and PF both became accepted as palliative regimens for patients with R/M HNSCC. In the context of concerns about the tolerability of cisplatin-based chemotherapy among some patients with R/M HNSCC, carboplatin plus paclitaxel has been evaluated in phase II studies, both in every 3 week²⁵^,²⁶ and weekly²⁷ dosing schedules. However, there has not been a rigorous prospective comparison versus cisplatin-based doublets in this patient population.

Another attempt to improve chemotherapy results has been the utilization of three-drug combinations utilizing a third drug in addition to a platinum compound and a taxane. TIP (paclitaxel, ifosfamide, cisplatin),²⁸ TIC (paclitaxel, ifosfamide, carboplatin),²⁹ and modified TPF (docetaxel, cisplatin, 5-fluorouracil)³⁰ were associated with encouraging response rates in (44%-59%) in phase II studies, but median survival times (8.8-11 months) were not clearly superior to those achieved with doublet chemotherapy with R/M disease.²⁴

Other Cytotoxic Agents that Have Been Evaluated for R/M HNSCC

Weekly vinorelbine achieved response rates of 7.5% and 16% in two phase II studies and was generally well tolerated.³¹^,³² Ifosfamide monotherapy also demonstrated palliative activity (response rates 4%-26%) in R/M HNSCC patients who had received prior chemotherapy.³³

Pemetrexed is a multitargeted antifolate, which inhibits multiple enzymes involved in both pyrimidine and purine synthesis, including thymidylate synthase (TS), dihydrofolate reductase (DHFR), and glycinamide ribonucleotide formyl transferase (GARFT). Gemcitabine is a pyrimidine antimetabolite, which is anabolized into a diphosphate form, which inhibits ribonucleotide reductase and a triphosphate form which is incorporated into the DNA, thereby blocking DNA replication.

A phase II study evaluated pemetrexed (500 mg/m² IV q21 days) in 35 patients with locally advanced or metastatic HNSCC not amenable to curative surgery or radiotherapy. Patients did not receive supplementation with vitamin B12 or folate. The objective response rate was 27% and median overall survival was 7.3 months.³⁴ The activity of pemetrexed against HNSCC might be viewed as somewhat surprising, in view of the apparent low efficacy of this agent against squamous cell cancer (SCC) of the lung.³⁵ This discordance underscores the risks of assuming that the activities of chemotherapeutic agents against R/M HNSCC and against SCC of lung will always be similar.

In R/M HNSCC, gemcitabine monotherapy was evaluated in two phase II trials. In a study of 47 patients with R/M HNSCC, the overall response rate was 13%.³⁶ In a more recent trial, no objective responses were seen among 26 patients with R/M HNSCC.³⁷ The results of these two small studies indicate that the activity of single-agent gemcitabine is low in R/M HNSCC, and gemcitabine monotherapy is rarely used in the management of these patients. It should be noted, however, that gemcitabine is an active drug in patients with R/M nasopharynx cancer and is regularly used in this disease setting.³⁸^,³⁹

Based on preclinical evidence of synergy and encouraging phase I results⁴⁰ the combination of gemcitabine and pemetrexed was evaluated in phase II studies in this patient population. In a study for 25 patients with R/M HNSCC, pemetrexed plus gemcitabine was administered every 2 weeks with vitamin support. Treatment was generally well tolerated, but the objective response rate was only 16%, indicating no evidence of additive or synergistic effects with these two drugs.⁴¹ Similarly, in another study of pemetrexed plus gemcitabine q2W, only two responses (one confirmed) were seen among 17 patients with R/M HNSCC.⁴²

Pemetrexed has shown encouraging efficacy when combined with bevacizumab⁴³ and remains of significant interest for further study. Pemetrexed plus cisplatin has been compared with cisplatin alone in a large, double-blind randomized trial in 795 patients with R/M HNSCC. There was no significant difference between the arms with regard to overall and progression-free survival, although on subset analyses, the combination arm had superior survival outcomes in performance status 0 and 1 patients and those with oropharynx primary tumors.⁴⁴ However, in the absence of randomized data to demonstrate superiority to other agents commonly used in the management of R/M HNSCC or a clear advantage when used in combination therapy, use of pemetrexed outside of a clinical trial in this patient population is not currently recommended.

Irinotecan resulted in objective responses in approximately 20 % of subjects as first-line therapy for R/M HNSCC, but was not active as second-line therapy.⁴⁵ Neither vinorelbine, ifosfamide, nor irinotecan has been evaluated in a randomized phase III study for R/M HNSCC. Older cytotoxic agents such as bleomycin, vinblastine, cyclophosphamide, and adriamycin also have activity,⁴⁶ but these agents have largely been supplanted in the contemporary era by the platinum agents, taxanes, 5-fluorouracil, methotrexate, and cetuximab (Table 30.3).

Antiepidermal Growth Factor Receptor Agents

The epidermal growth factor receptor (EGFR) is a transmembrane receptor tyrosine kinase that is expressed by most HNSCC tumors, and high levels of EGFR expression are associated with inferior clinical outcomes.⁴⁷^,⁴⁸ The receptor can be activated by a variety of ligands, including EGF, transforming growth factor alpha (TGF-α), and amphiregulin. The activated receptor forms homodimers or heterodimers with other ErbB family members, resulting in activation of the tyrosine kinase domain, stimulation of cell DNA synthesis, and cell proliferation through stimulation of multiple pathways. Two strategies of EGFR inhibition have been studied in clinical trials: monoclonal antibodies directed against the outer domain of the EGFR and small molecule inhibitors of the intracellular tyrosine kinase domain. Cetuximab is the only EGFR-targeting agent with approval from the United States Food and Drug Administration for the treatment of HNSCC.

Cetuximab, a human-murine chimeric immunoglobulin G (IgG1) monoclonal antibody, binds to the extracellular domain of EGFR.⁴⁹ As weekly monotherapy (400 mg/m² loading
dose, followed by 250 mg/m² weekly) for patients with advanced platinum-refractory HNSCC, cetuximab achieves objective radiographic response rates of 10% to 13% in phase II studies.⁵⁰^,⁵¹ Similar efficacy and tolerability has been observed with singleagent cetuximab 500 mg/m² every 2 weeks for patients with R/M HNSCC.⁵²

TABLE 30.3 Chemotherapeutic Agents Commonly Used in the Management of R/M HNSCC

Class	Agent
Platinum	Cisplatin, carboplatin
Taxane	Paclitaxel, docetaxel
Antimetabolite	5-Fluorouracil, methotrexate
EGFR inhibitor	Cetuximab

A randomized phase III study compared cetuximab plus cisplatin versus placebo plus cisplatin among 123 patients who had not received any prior chemotherapy for R/M HNSCC. Objective response rate favored the cetuximab arm (26% vs. 10%, p = 0.03), but progression-free survival and overall survival did not differ significantly between the groups.⁵³

However, in a larger randomized trial, cetuximab improved the efficacy of standard platinum-based chemotherapy for patients with R/M HNSCC. The EXTREME trial randomized R/M HNSCC patients (N = 442) with either cisplatin or carboplatin plus 5-fluorouacil compared with the same chemotherapy options with the addition of weekly cetuximab, including 6 months of maintenance cetuximab among responding patients tolerating therapy in the latter arm. Choice of cisplatin or carboplatin was per investigator discretion. Patients had not received any prior therapy for R/M. The cetuximab plus platinum doublet, compared with the platinum doublet alone, was associated with a significantly improved response rate (36% vs. 20%), progression-free survival (5.6 vs. 3.3 months), and overall survival (10.1 vs. 7.4 months).⁵⁴

Panitumumab is a fully humanized IgG2 monoclonal antibody directed against the extracellular domain of EGFR. A large (657 patients) randomized phase III study of the addition of panitumumab to the PF doublet was deemed negative because there was no significant improvement in overall survival associated with the addition of this antibody to the doublet.⁵⁵ Of interest, and discussed later in this chapter, on subsequent subset analysis, survival benefit was demonstrated in the HPV-negative subgroup as assessed by p16 status.

Zalutumumab, a fully humanized IgG1 anti-EGFR monoclonal antibody, was evaluated in a randomized phase III study for patients with noncurable HNSCC. Patients (n = 286) were randomly assigned to either weekly zalutumumab + best supportive care or best supportive care with optional methotrexate. Zalutumumab, versus the control group, did not result in a significant increase in overall survival (6.7 vs. 5.2 months, p = 0.06), which was the primary endpoint of the study.⁵⁶ As such, cetuximab remains the only monoclonal against EGFR that is commercially available for HNSCC patients in the United States.

Several inhibitors of the EGFR tyrosine kinase domain have been evaluated for patients with advanced HNSCC. In a phase II trial of gefitinib 500 mg/day (n = 47 subjects), the objective response rate was 10.6% and median survival was 8.1 months.⁵⁷ In a randomized phase III comparison with methotrexate, gefitinib (at either 500 or 250 mg/day) did not improve overall survival compared with methotrexate,⁵⁸ but the risk of tumor-hemorrhage events was greater on the gefitinib arms (11.4% and 8.5%, respectively) than with methotrexate (1.9%).

Erlotinib is an orally available, reversible, and selective inhibitor of the EGFR tyrosine kinase that is currently approved for the treatment of refractory non-small cell lung cancer (NSCLC). In a phase II study (n = 115 subjects), erlotinib yielded a major response rate of 4% among patients with R/M HNSCC.⁵⁹

Preclinical studies have suggested that resistance to EGFR inhibitors may be related to activity of other ErbB family members.⁶⁰ Strategies to broadly inhibit ErbB signaling are being explored clinically. Lapatinib, a dual inhibitor of EGFR and ErbB2 tyrosine kinases, is commercially available for the treatment of some women with advanced breast cancer. In a phase II study for patients with R/M HNSCC (27 subjects with prior exposure to EGFR inhibitor and 15 subjects with no prior exposure to EGFR inhibitor), no objective responses were seen.⁶¹ Afatinib, a potent inhibitor of the tyrosine kinase domains of EGFR, ErbB2, and ErbB4, is an investigational agent that is being studied for several tumor types. Afatinib was compared with cetuximab in a randomized phase II study for patients with R/M HNSCC (n = 124 subjects). The confirmed partial response rate was higher in the afatinib group versus the cetuximab group, but progression-free survival did not differ between the two arms.⁶²

Only gold members can continue reading. Log In or Register to continue