The never-ending story: finding a role for neoadjuvant chemotherapy in the management of head and neck cancer.

Abstract

The management of head and neck cancer has evolved dramatically over the decades. Surgical and radiation (diagnostic and therapeutic) techniques have improved, resulting in better outcomes for patients with head and neck cancer. The addition of systemic therapy has also resulted in improved outcomes, principally through the use of chemotherapy concurrently with radiation. Highlighting this phenomenon is the work of the Meta-Analysis of Chemotherapy on Head and Neck Cancer (MACH-NC) Collaborative group, which has pooled data on over 17,000 patients with head and neck cancer treated on randomized trials testing the impact of adding chemotherapy in the overall management schema. Fifty trials of concomitant chemoradiation demonstrated that the addition of chemotherapy resulted in an absolute survival benefit of 6.5% at 5 years. Neoadjuvant chemotherapy is an approach used as a component in the treatment of many solid tumors, and has also been investigated in the management of head and neck cancers. The enthusiasm for induction therapy took off in the late 1970s when cisplatin and fluorouracil (FU) were established as the two principal drugs that were able to achieve reasonable response rates in the treatment of metastatic head and neck cancer. However, randomized trials demonstrating a survival benefit with induction chemotherapy were uncommon, and in their initial report in 2000, MACH-NC stated that induction therapy tested in 31 randomized trials, did not improve survival. A flicker of hope remained for induction therapy in this report, as the authors did state that limiting the analysis to 15 trials using cisplatin and FU, there was a small and significant survival advantage to induction therapy. This small benefit and the discovery that taxanes were also efficacious for treating head and neck cancers led to trials investigating 3 drug regimens containing a platin, taxane, and FU (TPF). Several trials, culminating in the TAX-324 trial demonstrated that regimens including a taxane were superior to those with platin-FU alone. The article that accompanies this editorial presents the formal reporting of the DeCIDE trial, a randomized phase III study that attempts to answer the question of whether the addition of TPF-based induction therapy improves outcomes in patients planned for concurrent chemoradiotherapy as treatment for locally advanced head and neck cancer. The study is somewhat reminiscent of the approaches of the Radiation Therapy Oncology Group in the 2000s; combine a second strategy to chemoradiotherapy and anticipate improved outcomes. Unfortunately, neither modifying radiation by acceleration with cisplatin nor adding cetuximab to platin-based chemoradiotherapy improved survival. Similarly, the results of DeCIDE demonstrate that adding neoadjuvant chemotherapy to chemoradiotherapy does not improve survival. While DeCIDE did not show the anticipated survival advantage hoped for by adding induction therapy to chemoradiotherapy, Cohen et al state in the conclusions of their abstract that the trial was underpowered due to not meeting planned accrual, and had higher than predicted survival rates in both arms. What is the relevance of this conclusion? Should the reader infer from this statement that we should dismiss the negative results based on these two factors? The study had planned a sample size of 400, and was only able to accrue 285 patients. The authors address this issue by demonstrating that with the actual sample size, the power of the study decreased from 88% to 56%. This, however, misses the point, as it ignores the data they did collect. The real question is, having achieved a hazard ratio (HR) of 0.91 with a 95% CI of 0.59 to 1.41, what is the likelihood of achieving the goal of an HR of 0.625? Even without elegant statistics, it is obvious that the probability of success is extremely low. The second issue is the impact of higher-than-anticipated survival rates. While we should not discount the impact the therapies had on outcomes, the relatively high survival rates are more likely due to the disease and not the therapy. When tested for human papillomavirus (HPV) in 31% of the patients with oropharyngeal cancer, the positivity rates were extremely high, and HPV-associated disease has been demonstrated to have high survival rates. Granted, higher survival rates than anticipated would impact the sample size calculations. However, the epidemiologic shifts should not have undermined the premise that lowering distant recurrence would improve survival, particularly since while demonstrating that patients with HPVassociated disease have improved survival relative to patients with HPV-nonassociated disease, studies on oropharyngeal cancers have suggested that distant recurrence rates between the two diseases are not noticeably different. Decreasing the emergence of distant metastases is one component of the challenge to improve survival rates in patients with head and neck cancer. Improvements in outcomes, particularly locoregional control have led some to contend that distant disease has JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 32 NUMBER 25 SPTEMBER 1 2014