The addition of chemotherapy to the treatment of advanced head and neck cancer is a controversial subject, with vocal opponents and proponents. Clinical trials have been devised to elucidate the role of treating squamous cell carcinoma in the head and neck with chemotherapeutic drugs. There are many difficulties, however, in interpreting the results from these studies: (1) the heterogeneous sites of the head and neck seem to respond differently to similar treatments; (2) the effectiveness of a drug depends not only on its intrinsic properties, but also on the dose used, the schedule in which it is delivered, and the subsites treated; duplication of all these events is difficult, hence the limited number of corroboratory studies; (3) drugs are often used in combination to take advantage of the different modes of action, with widely diverse combinations; (4) previously treated cancer responds much differently than naive tumors, and nonresectable tumors may respond differently than resectable tumors; (5) patients themselves vary widely and should be stratified according to well-known criteria, such as performance, alcohol, and nutritional status; and (6) many trials are not randomized, making their evaluation difficult, as even prospective studies must then use historical controls, which may or may not be relevant.

For a complete understanding of the use of individual and combination regimens, a brief description of the more commonly used drugs is presented. Cisplatin (P) and carboplatin are platinum-based compounds. These agents differ in that carboplatin has fewer side effects (especially renal failure and hematogenesis) and is slightly less efficacious. Their mechanism of action is by binding to DNA strands and causing multiple chain breaks in dividing cells, eventually leading to cell death. Major toxicities include acute renal failure, neuropathy, ototoxicity, severe emesis, and occasional myelosuppression.1, 2

The antimetabolites include methotrexate (a folate antagonist), 5-fluorouracil (5-FU), and gemcitabine (pyrimidine analogues), and 6-mercaptopurine (a purine analogue). They exert their action by disrupting DNA chain elongation, either by directly incorporating into the chain or by depleting the cells of necessary enzymes. Toxicities include myelosuppression, thrombocytopenia, nausea and vomiting, and mucositis.^{1, 2}

The most commonly used antibiotics (anthracyclines) are doxorubicin (Adriamycin), bleomycin (B), and mitomycin C (Mito). These agents act by producing intracellular free radicals (requiring a well-oxygenated tumor), damaging the DNA chain in association with topoisomerase. Cardiac and pulmonary side effects predominate, particularly pulmonary fibrosis and arrhythmias and, with larger doses, congestive heart failure.^{1, 2}

Vincristine, vinblastine, and the newer agent, vinorelbine, are plant alkaloids that bind to tubulin, interrupting the mitotic spindle and preventing the cell from completing meiosis. Partially irreversible peripheral neuropathy, gastrointestinal cramping, and mild myelosuppression are side effects noted with these agents.^{1, 2}

The newest group is the taxoid family, consisting of paclitaxel (Taxol) and docetaxol (Taxotere). In contrast with the vinca alkaloids, these drugs stabilize the microtubules, eventually causing the cells to accumulate arrays of disorganized microtubules, leading to cell death. As with the vinca alkaloids, they are active during the S phase of the cell cycle. Myelosuppression is the most common side effect, and irreversible toxicity may be seen with doses above 250 mg m².³

Palliative Chemotherapy

CHEMOTHERAPY TRIALS: SINGLE AGENTS AND COMBINATION THERAPY

Chemotherapy as the sole modality of treatment for head and neck cancer has been limited to almost exclusively recurrent or persistent disease after failure of primary locoregional therapy. In this setting, the goal of treatment is limited to palliation, as no improvement has been demonstrated for either disease-specific or overall survival.^{1, 4–10} Treating recurrent or metastatic head and neck cancer is much different than treating a naive tumor. Previous therapy usually diminishes the effectiveness of chemotherapy by 30 to 50%.¹¹ Theories to explain this include more hypoxic cells, decreased blood flow to the tumor, and chemical resistance.^{1, 5, 11, 12} The difference in effectiveness can easily be seen when comparing the complete and partial responses achieved with induction (neoadjuvant) chemotherapy (30 to 92%)^12–17 to the response rate of palliative treatment for recurrences (10 to 40%).^{4–10, 13} This section discusses chemotherapy only as it applies to recurrent or metastatic head and neck cancer.

Deciding on which therapeutic endpoint to use in clinical trials becomes important, as the “gold standard” of increased survival is no longer applicable. The most easily and objectively measured outcome is tumor response (Table 8-1). Most trials measuring the benefit from chemotherapy use this as their main index of effectiveness, making the assumption that a decrease in tumor size will directly benefit the patient. For painful or obstructing lesions, as is often the case, this argument probably holds merit. However, a major concern relates to the toxicity of the treatment itself, as this group of patients receive no survival benefit. If the side effect profile of the treatment causes considerable discomfort, a few months of reduced tumor burden will not be worthwhile.

Modified from Catimel,¹ Stupp,² and Schrivers.³

A counterargument is that an increase in side effects is acceptable if the probability of a complete response (CR) is increased, as a longer survival time has been associated with CR.^18–20 However, as Adelstein¹¹ and Fu¹² point out, this association may simply reflect selection of those patients who have a better prognosis to begin with. Response to chemotherapy and improved survival are linked, both occurring in those patients with a better performance status, less prior treatment, and certain head and neck subsites (larynx, mobile tongue). This same selection bias has been used as the basis for the organ preservation strategy (as explained later). Other measurements of therapeutic improvement include an increase in quality-of-life scores and time to progression and, for those achieving a CR, time until recurrence.

SINGLE AGENTS

Does chemotherapy add any benefit beyond providing pain control? In one of the few randomized studies with a “no-treatment” control arm, Morton et al.²¹ compared the use of cisplatin, bleomycin, and the combination of cisplatin plus bleomycin. They found a statistically significant increased survival in the platinum arms (4.0 to 4.2 months) as compared with the control arm (2.1 months). The population in this study was relatively small, with about 24 patients in each treatment arm, and survival was relatively short, but this does provide some evidence that chemotherapy can have an impact on survival. Most trials, however, show that there is no survival advantage to chemotherapy over historical controls.^{1, 4–10}

Methotrexate (M) has been the standard therapy for palliation of advanced tumors for three decades. The dose is 40 to 60 mg m² /week given intravenously (IV). The average response rate is 30% with a range of 8 to 57% (CR range from 6 to 26%).^{22, 23} Toxicities are tolerable, with mild to moderate leukopenia, anemia, mucositis, and diarrhea the most commonly cited.⁵ The duration of the response is approximately 4 months.

Cisplatin has been used extensively with response rates of 8 to 41%, also averaging around 30% (0 to 8% CR).^{21, 24, 25} Additional side effects of nephrotoxicity, neuropathy, and ototoxicity must be accounted for when choosing this drug over methotrexate. Carboplatin has fewer side effects than cisplatin, but the response rate is also decreased to around 21% (14 to 26%).

Three randomized studies have conducted direct comparison of cisplatin with methotrexate. In the first two studies, consisting of 144 patients, no difference was seen in response rate, response duration, or survival.^{24, 26} The larger of these studies showed a low response rate to cisplatin (8%) and methotrexate (16%), with an overall survival of only 4.5 to 5 months. A third trial from the group in Liverpool showed a significantly higher response rate of the tumors to cisplatin (26% vs 12%) and, again, no improvement in survival.²⁷

Bleomycin also produces a 21% response rate (6 to 45%) in refractory tumors, with the effects lasting 1 to 3 months.⁵ The standard dose is 15 U IV on a weekly schedule. Pulmonary toxicity is the main drawback, especially once a total dose of 200 to 400 U is reached.

5-FU is not often used alone to treat head and neck cancer, but response rates average 15% (0 to 33%).²⁸ The optimal administration is a 5-day continuous infusion, rather than bolus treatment.^{20, 29} Side effects include bone marrow suppression, cardiac toxicity, and mucositis.

The taxoids are generating some interest, with one phase II report showing a 40% response rate and another a 32% response.^{30, 31} Hematologic toxicity was severe, however, even with the use of granulocyte-colony-stimulating factor (G-CSF).

Other less frequently used agents are hydroxyurea (25%); alkylating agents, including cyclophosphamide and ifosfamide (36%); vinblastine (29%); gemcitabine (13%); and the anthracyclines (5 to 8%)¹ (Table 8-1).

SINGLE AGENTS VERSUS COMBINATIONS

Combination chemotherapy takes advantage of two characteristics of chemotherapeutic drugs. The first is their mechanism of action. The rationale is that the addition of drugs that target different areas of the cell might overcome the selection pressure causing resistance to the agents. Also, separate subpopulations might be better targeted (mitomycin C and hypoxic cells). The second purpose of combining drugs is to take advantage of the different toxicity profiles and to reduce the severity of side effects experienced by the patient.

Do combination regimens improve tumor responsiveness as compared with single agents? In two studies comparing single-agent methotrexate with cisplatin plus 5-FU in recurrent or metastatic head and neck cancer, one showed a significantly higher response rate in the combination therapy arm, and the other showed a trend toward significance in the combination therapy arm.27, 32 Two other studies compared the single-agent approach of cisplatin and 5-FU with the combination and again found a significantly improved response rate to the combination but no change in survival rate.^{33, 34}

A meta-analysis was undertaken by Browman and Cronin⁷ in 1994 to evaluate single-agent versus combination chemotherapy. These investigators analyzed 15 trials that included 1916 patients and 17 different treatment regimens. Single-agent methotrexate was less effective at reducing tumor size than was found in the pooled combination studies. The odds ratio (OR) was used to define statistical significance, with an OR of >1 favoring comparison treatment); the OR was 1.71 with a 95% confidence interval (CI) of 1.28-2.31. Pooled combination treatments were also statistically better at achieving tumor regression than were the pooled single agents (M, P, B, 5-FU), with OR-1.59 and 95% CI-1.28-2.01. The combination of cisplatin and 5-FU was found to be the most efficacious. The side effects of the different regimens were also evaluated showing that the combinations have a greatly increased risk of nausea and vomiting. It appears that combination chemotherapy should be used for those patients with end-stage metastatic or recurrent squamous cell cancer if they have a life expectancy of more than 4 to 5 months and a good performance score and can tolerate the treatment.

Therapeutic Chemotherapy

COMBINED-MODALITY THERAPY

Radiotherapy has proved effective in head and neck cancer. Its mechanism of cell killing is postulated to be mediated by the direct interaction of superoxide radicals with cellular enzymes and DNA, indicating that adequate oxygen content of the tumor is necessary for efficacy of this treatment.³⁵ In addition, radiotherapy has been found to be most effective during the G2 and M stages of the cell cycle. Adding chemotherapy to radiation has several theoretical benefits. Chemotherapy can prevent the repair of sublethal damage inflicted by ionizing radiation. Certain drugs act as sensitizers (e.g., 5-FU) and improve the effect of radiation on tumors. Some drugs are particularly toxic to hypoxic cells (mitomycin C) and thus can target a population that is radioresistant. Both cause apoptosis, probably by different mechanisms, so each potentiates the other.¹² Chemotherapy often causes growth arrest of cells at a certain part of the cell cycle, allowing them to be more sensitive to radiation.³⁵

One of the difficulties of combining the two treatment modalities has been in determining the appropriate timing of each. The schedule for adding chemotherapy to radiation has traditionally been divided into three groups: induction or neoadjuvant, concomitant (including alternating), and adjuvant (posttreatment) administration.

NEOADJUVANT CHEMOTHERAPY

Giving chemotherapy before radiation has theoretical benefits and risks. By reducing the size of the tumor without interrupting the blood supply, the oxygen content of the mass should be higher, leading to greater sensitivity to radiation. In addition, responders to chemotherapy may be treated effectively with a single modality treatment, usually radiation, and thus spared surgery (at least as a primary modality). On the downside, Toohill et al.³⁶ reported a worse survival rate with the neoadjuvant arm than with the standard (radiotherapy) arm. This raises the possibility that delaying definitive therapy may allow the tumor to become further disseminated and reduce survival.

A large number of trials have been designed to evaluate the benefit of neoadjuvant chemotherapy.^{1, 13–17, 36–41} Many show high response rates, up to 98% in operable laryngeal cancer.¹⁷ But survival data show no statistically significant change in overall or (where available) disease-specific survival (Table 8-2). Many of these studies have been criticized with respect to low patient numbers, inadequate drug dosages, and variable “standard” treatment arms, but several well-controlled studies have been completed.

Is there a role for induction chemotherapy?

Stay updated, free articles. Join our Telegram channel

Join