Revisiting Induction Chemotherapy for Head and Neck Cancer

Argiris and colleagues present a comprehensive review of 25 years of phase II/III trials using multimodality therapy for locally advanced head and neck squamous cell cancer (HNSCC). They focus on two approaches: induction chemotherapy followed by definitive local therapy (surgery and/or radiotherapy) and concurrent chemoradiotherapy. In sorting through these trials, the authors review the controversies in the management of locally advanced HNSCC, while also presenting a rationale for a unified approach- combining induction and concomitant chemoradiotherapy in a multimodality treatment paradigm. Evidence from several recent studies suggests that this strategy will benefit a subset of patients with locally advanced disease. The stage is set for the reevaluation of the benefit of induction chemotherapy prior to definitive chemoradiation. To that end, three different prospective phase III trials are under way in the United States. Evolving Standard
Many questions remain unanswered as we embark on the current generation of trials in multimodality therapy. Trials conducted in the 1980s focused on the addition of induction chemotherapy to definitive local therapy of radiation or surgery-as defined by tumor resectability. The focus in the 1990s on the use of concomitant chemoradiotherapy with the goal of organ preservation led to the establishment of new standards of care. The current focus is on reevaluating the combination of induction chemotherapy added to concomitant chemoradiotherapy for disease stages in which the latter is now a standard option. The idea of combining these approaches is based on two parallel and complementary observations from previous trials. While concomitant chemoradiotherapy mainly affects locoregional control such that distant metastases and second primary tumors are the greater risk in these patients, induction chemotherapy results in a lower rate of metastasis but does not improve locoregional control. From these two observations, one derives the simplistic inference that combining these two approaches may substantially improve survival. Although this is a reasonable hypothesis that is supported by promising phase II reports, future trials devised by the head and neck cancer research community need to (1) have adequate power to demonstrate a potentially small but significant increase in survival without unacceptable acute and late toxicities, and (2) define the patients who will benefit from this improvement in treatment. Creating the optimal combination of surgery, radiation, and systemic therapy required to manage the primary and the neck within each site and stage grouping is an enormous and continuing challenge. Available Data
How strong are the data that drive this interest in revisiting induction chemotherapy? Since the late 1970s, more than three dozen prospective, randomized trials comparing induction chemotherapy followed by surgery and/or radiotherapy have been published. To our knowledge, only one trial of induction chemotherapy demonstrated an improvement in survival, although neither locoregional control nor distant metastatic rates were altered.[1] A second often-quoted trial reported a survival benefit for a subset of patients with unresectable disease, but there was no benefit in the analysis of all patients.[2] These and other results are presented in a meta-analysis of updated individual patient data from 63 randomized trials that incorporated chemotherapy in the primary management of newly diagnosed patients with HNSCC.[3] This meta-analysis showed that neither induction (hazard ratio [HR] = 0.95; 95% confidence interval [CI]: 0.88-1.01) nor adjuvant (HR = 0.98; 95% CI: 0.85-1.19) chemotherapy resulted in improved survival. In contrast, an 8% absolute survival advantage at 5 years (HR = 0.81; 95% CI: 0.76-0.88) was observed with concomitant chemoradiotherapy. This benefit was subsequently confirmed in an updated meta-analysis that added 24 randomized trials completed between 1994 and 2000.[4] Given the likely time-dependent differences between these trials performed over several decades, exploratory subset analyses of the number and types of chemotherapy were performed. These revealed a 5% survival gain (HR = 0.88; 95% CI: 0.79-0.97) for induction trials employing cisplatin(Drug information on cisplatin) or carboplatin(Drug information on carboplatin) plus fluorouracil(Drug information on fluorouracil). The result of this subset analysis (though not corroborated by large, prospective individual trials) combined with the observations that (1) response to induction chemotherapy is predictive of response to subsequent radiotherapy; and (2) chemotherapy can suppress distant metastases provides a rationale for proceeding to test this hypothesis in prospective, randomized trials. Choice of Therapy
The next step in planning these phase III trials is to determine what chemoradiotherapy combination to use and in which population this strategy is most likely to provide survival benefit. The success of combinations of platinum, taxane, and fluoropyrimidine chemotherapeutics in inducing tumor regression and enhancing radiation cytotoxicity coupled with advances in radiotherapy techniques (eg, intensity-modulated radiation therapy, altered fractionation) has enabled investigators to intensify locoregional therapy. Further supporting the feasibility of intensified regimens are improvements in supportive care that include nutritional support, pain management, growth factors, and prophylactic antibiotics. The good news is that improved survival of patients with the worst prognosis, such as those with massive unresectable primaries and advanced neck disease, is now possible. The bad news is that the temptation to treat all stage III/IV patients with these aggressive therapies increases the risk of serious late effects and function impairment from normal tissue injury. Patient Population
Choosing the appropriate subset of patients with locally advanced disease for treatment with induction chemotherapy followed by concomitant chemoradiotherapy is crucial to maximizing survival and minimizing toxicity. Concurrent chemoradiotherapy is the standard of care for patients with unresectable locally advanced disease. Because the standard treatment of cisplatin, 100 mg/m² every 21 days during radiotherapy, achieves a 3-year survival rate of only 37%, this patient group seems ideal for a phase III trial of primary chemoradiation with or without induction chemotherapy.[ 5] Patients with locally advanced, resectable disease of the oropharynx and larynx are also treated nonsurgically with chemoradiotherapy when organ preservation is desired. Stage T3/T4 or N2/N3 oropharyngeal cancer carries a similar high risk of locoregional failure and distant metastases, making these patients another ideal subgroup for the induction followed by concomitant therapy approach. Pilot studies in this patient group are encouraging.[6,7] Several subgroups of patients with locally advanced disease may not benefit from intensified treatment. Intermediate- stage cancers (eg, T2, N0 and T2, N1) without other identified poor prognostic factors may derive little survival benefit but suffer added tox- icity from this intensification of treatment. Larynx cancer may also not be appropriate for this strategy. Data from the recent Intergroup trial R91-11 show an 88% rate of larynx preservation, 78% rate of locoregional control, 8% rate of distant metastases, and 76% overall survival at 2 years with concurrent chemoradiation.[8] Improving on these results by adding induction chemotherapy would be difficult to demonstrate. Instead, alternative induction approaches-for example, with biologically targeted agents-should be considered. In summary, randomized trials of induction chemotherapy added to definitive chemoradiation need to include high-risk locally advanced subsets of patients treated with platinum-based combinations in order to optimize this intensified combined-modality therapy. Enrollment of patients in carefully designed trials of induction chemotherapy followed by chemoradiotherapy that evaluate quality of life, function, and late toxicities along with survival end points is a logical and crucial next step in improving survival for patients with HNSCC.