The year 2012 has been an eventful year for head and neck squamous cell cancer (HNSCC), with several trials from this year's ASCO Annual Meeting showing disappointingly negative results. While some novel immunotherapeutic approaches to squamous cell carcinoma of the lung may have broad implications for other similar tumors such as HNSCC, ASCO 2012 will be remembered by the HNSCC field for the disappointingly negative findings of the randomized Phase III induction chemotherapy trials.
Head and neck cancer represents a heterogeneous group of cancers of varying primary sites, 95 percent of which are squamous cell cancers, including tumors of the oral cavity, oropharynx, larynx, hypopharynx, nasal pharynx, paranasal sinuses, lips, and salivary glands. Most patients with head and neck cancer present with locally advanced disease, also classified as stage III and IV.
For newly diagnosed patients with resectable or unresectable disease, platinum-based concomitant radiotherapy and sequential therapy were the major standard approaches. Surgery or radiation alone have long been the standard for early or limited stage disease (stage I or II). For recurrent or metastatic disease, there are different paradigms, but in general, surgery followed by radiation for resectable disease or concomitant chemoradiation for unresectable are the standards for recurrent disease.
Combination three-drug chemotherapy with cisplatin, fluorouracil (5-FU), and cetuximab has emerged as the standard of care, based on the Vermorken publication in the New England Journal of Medicine, 1 for patients with metastatic HNSCC and good performance status.
Two studies in locally advanced resectable (TAX 323) and unresectable (TAX 324) HNSCC were published back to back in the New England Journal of Medicine in 2007.2,3 These studies showed a clear survival advantage from combined docetaxel, cisplatin, and 5-FU (TPF) versus cisplatin and 5-FU induction chemotherapy, therefore establishing TPF as the optimal induction chemotherapy. Toxicity was tolerable in the TPF arm, and this became the standard of care that was used as the basis for the paradigm in the DeCIDE (docetaxel-based chemoradiotherapy plus or minus induction chemotherapy to decrease events in head and neck cancer) trial, a Phase III multicenter international study presented by Ezra Cohen at ASCO 2012.
In this study, patients with locally advanced HNSCC, the preponderance of patients with N2C or N3 disease, were randomized to receive either two cycles of TPF induction chemotherapy (75 mg/m2 each for docetaxel and cisplatin, and 750 mg/m2 5-FU continuous infusion for 5 days) followed by docetaxel, 5-FU, hydroxyurea, and concurrent radiotherapy), or to the control arm of concurrent docetaxel, 5-FU, hydroxyurea, and chemoradiotherapy.
The University of Chicago-pioneered hyperfractionated radiotherapy regimen was administered. Accrual lagged well behind the projected accrual of 400 patients, and the study was closed after 280 patients had been accrued by May 2009. Of the 280 patients, 142 were randomized to induction chemotherapy and 138 to concurrent chemoradiotherapy. The two arms were well balanced by age, performance status, and primary tumor site.
Evaluating toxicity by arm, the number of grade 3–4 adverse events during induction chemotherapy was insubstantial (9.6% neutropenia, 8.8% mucositis, and 7.4% fatigue, anorexia or hyperglycemia). Substantial adverse events occurred during chemoradiotherapy on both arms, which did not appear to be dramatically different except for the higher rates of neutropenia (25.6% vs. 11.3%) and leukopenia (13.6% vs. 3.8%) on the induction chemotherapy versus chemoradiotherapy arm, respectively.
Induction chemotherapy was highly effective in inducing responses: almost nine percent of patients had a complete response, 55 percent had a partial response, and only 4.4 percent had disease progression. The 64 percent overall response rate to induction chemotherapy was consistent with that shown by Vermorken and Posner.
Numerical improvements in recurrence-free survival (p=0.16) and distant failure were seen in the induction- chemotherapy arm, which did not achieve statistical significance, but overall survival was not meaningfully different between the two arms. In this study, induction chemotherapy did not improve survival, but it improved the cumulative incidence of distant failure, with excellent overall survival and disease control.
The PARADIGM study,4 a Phase III study comparing sequential chemotherapy and radiation to concurrent chemoradiotherapy in locally advanced head and neck cancer, was presented by Robert Haddad of Dana-Farber Cancer Center. Emory participated in this study, which sought to compare the primary three-year survival achieved by TPF-based sequential therapy followed by concurrent radiation with use of carboplatin versus cisplatin-based concurrent chemoradiotherapy.
The goal was a 15 percent improvement in three-year survival, from 55 percent for chemoradiotherapy to 70 percent for TPF.
The study planned to randomize close to 400 patients, but accrued only 145—70 on the induction chemotherapy arm and 75 on the concurrent chemoradiotherapy. The results were reported after a median follow-up of 49 months, when the data locked in January 2012.
Fewer Events than Expected
While the arms were well-balanced, what was notable was that the number of events was substantially lower than expected, particularly on the control arm. There was no difference in overall survival or three-year progression-free survival (PFS; 67% on the induction chemotherapy arm and 69% on the chemoradiotherapy arm).
Numerically, induction chemotherapy appeared to provide an 11% advantage in non-oropharyngeal cancer and a 16% disadvantage in oropharyngeal cancer when compared with the cisplatin-based concurrent chemoradiotherapy arm. An interesting difference was seen between the two chemoradiation arms of the experimental arm. Patients who achieved a very good complete response on induction chemotherapy received radiotherapy concurrent with carboplatin, whereas patients who achieved a less impressive response and had bulky residual disease after induction chemotherapy received concurrent radiotherapy with docetaxel.
The docetaxel arm appeared to have significantly better three-year progression-free survival and overall survival rates, but this was not the primary endpoint of the study. Acute toxicities were as expected. Interestingly, both of these studies were underpowered and failed to achieve their primary endpoint.
At this point, concurrent chemoradiotherapy remains the standard of care, and while the induction chemotherapy arm of DeCIDE5 showed some suggestions of efficacy, this was not shown in the PARADIGM study, potentially due to the use of single-agent carboplatin as a radiosensitizer, resulting in inferior results.
Other studies of interest at ASCO included the phase II randomized CONCERT I trial6 of chemoradiotherapy with or without panitumumab in patients with unresected locally advanced HNSCC. This study, presented by Jordi Giralt of Hospital Vall d'Hebron in Barcelona, Spain, had a primary endpoint of locoregional control at two years in patients receiving panitumumab plus chemoradiotherapy versus chemoradiotherapy alone.
A total of 153 patients were randomized at a 3:2 ratio, favoring the experimental arm. Despite excellent median dose delivery of cisplatin and panitumumab throughout the study, the panitumumab arm generally performed worse, although not statistically significantly so, compared with the cisplatin and radiation with regard to progression-free and overall survival, where there was a trend favoring concurrent cisplatin and radiation, and the addition of panitumumab increased the hazard ratio of death to 1.63 (p=0.1223).
In summary, adding panitumumab to concurrent chemoradiation offered no advantage for patients with locally advanced HNSCC.
Most Promising Study
The most promising study was a Phase III trial of cabozantinib (XL184),7 a pan tyrosine kinase inhibitor that targets hepatocyte growth factor receptor (MET), vascular endothelial growth factor 2 (VEGFR2) and RET in medullary thyroid cancer.
This study randomized patients with medullary thyroid cancer with documented RECIST progression in a 2:1 manner to cabozantinib (n=219) versus placebo (n=111) until progression. No crossover unblinding was allowed, and overall survival was the primary endpoint.
There was a highly significant difference in one-year PFS in favor of cabozantinib (47.3% vs. 7.2%, p=0.0001). Median PFS was 11.2 months versus four months when assessed by an independent review committee.
These data, as well as the impressive waterfall plot, clearly show that cabozantinib results in a significant response rate in patients with metastatic medullary thyroid cancer with changes in calcitonin and CEA lowering with clinical response. These findings result in hopes that this highly efficacious agent with an overall response rate of 28 percent and a median duration of response of 14.6 months will soon be approved for the treatment of metastatic medullary thyroid cancer.
In conclusion, while the DeCIDE and PARADIGM trials, two underpowered Phase III trials, failed to show a meaningful advantage for induction chemotherapy versus concurrent chemoradiotherapy, only the Radiation Therapy Oncology Group has the capacity to run adequately powered Phase III trials in this setting to determine whether a role exists for induction chemotherapy in the treatment of locally advanced high-risk HNSCC.
Furthermore, while panitumumab failed to replicate the benefit of cetuximab in locally advanced HNSCC, approaches targeting RET with cabozantinib for metastatic medullary thyroid cancer continue to look promising, and are likely to enhance treatment for advanced diseases.
From OT Clinical Advisory Editor for Oncology Ramaswamy Govindan, MD
The term “head and neck cancer” encompasses a wide variety of malignancies in terms of both histology and site of origin. Dr. Fadlo Khuri, an expert in this area, summarizes here the key papers presented at this year's ASCO Annual Meeting. The role of induction therapy in locally advanced squamous cell cancer arising in the head and neck region and molecularly targeted therapy in medullary thyroid cancer feature prominently in his discussion.