Efforts to reduce the late toxicity associated with chemoradiotherapy for locally advanced head and neck squamous cell cancer have focused on radiation therapy dose de-escalation in select populations, according to James Melotek, MD, a radiation oncologist at the University of Chicago.
“Patients with favorable response to induction chemotherapy for head and neck cancer have significantly decreased rates of locoregional failure, and the majority of locoregional failures after definitive chemoradiation are ‘in-field’ failures that occur in areas of previous gross tumor,” said Dr. Melotek at the 2016 Multidisciplinary Head and Neck Cancer Symposium in Scottsdale, Arizona.
Dr. Melotek et al sought to increase induction chemotherapy response by adding the mTOR inhibitor everolimus (Afinitor) to induction chemotherapy,1 based on preclinical and clinical data suggesting its activity across multiple tumor types. The investigators incorporated a novel response-adapted volume de-escalation (RAVD) approach that used induction chemotherapy response to guide the extent of radiation therapy volume reduction in a nonselected population of patients with locally advanced head and neck squamous cell cancer.
Study Details
Patients with measurable locally advanced head and neck squamous cell cancer were enrolled in a phase I/II randomized trial and received up to two cycles of induction chemotherapy. It consisted of cisplatin at 75 mg/m2 on day 1; paclitaxel at 175 mg/m2 on day 1; and cetuximab (Erbitux) at 400 mg/m2 then 250 mg/m2 weekly, with or without 5 mg of daily everolimus. Responses were measured radiographically after completion of induction chemotherapy.
Patients with ≥ 50% reduction in sum of gross tumor diameters were classified as “good responders” and received TFHX2 (paclitaxel, fluorouracil, hydroxyurea, and 1.5 Gy twice daily radiation therapy every other week) to a total of 75 Gy with the planning target volume (PTV1) encompassing exclusively gross disease. Patients with < 50% reduction were classified as “nonresponders” and treated with volumes encompassing PTV1 and the next nodal station at risk (PTV2) to 45 Gy, followed by a sequential boost to PTV1 to 75 Gy.
“The good responders and nonresponders were treated with reduced radiation volumes compared to conventional treatment volumes in this trial,” said Dr. Melotek. “And nonresponders were treated to only the first echelon of uninvolved nodal disease.”
Baseline Characteristics
Between 2010 and 2014, 94 patients were enrolled on the protocol. The study population was 84% male, with a median age of 57 years (range = 27–76 years). Cancer of the oropharynx was present in the majority of subjects, and good responders were significantly more likely to have tumors positive for the human papillomavirus (HPV; 81% vs 51% of nonresponders).
There was no significant difference between good responders and nonresponders in the proportion of T3 tumors (> 4 cm) and N2b cancers (involvement of at least one ipsilateral lymph node, none > 6 cm). Fifty-six percent of patients had T3 or greater disease, and 88% had N2b or higher disease.
Overall Results
On interim analysis, change in mean tumor size was –42% with everolimus and –49% with placebo. “Therefore, everolimus was discontinued for futility, and the protocol was amended accordingly,” said Dr. Melotek.
Chemoradiotherapy Toxicity in Head and Neck Cancer
- Efforts to reduce late toxicity associated with chemoradiotherapy for locally advanced head and neck squamous cell cancer typically rely on radiation dose de-escalation.
- A novel response-adapted volume de-escalation approach uses induction chemotherapy response to guide the extent of radiation therapy volume reduction.
- Response-adapted volume de-escalation resulted in improvements in late toxicity and will be studied further in patients with human papillomavirus–positive oropharyngeal cancer.
Induction therapy was completed by 89 patients, 37 of whom (41.6%) were good responders and 52 of whom (58.4%) were nonresponders. Of the 37 patients with good response, 30 had HPV-positive oropharyngeal squamous cell cancer. One patient died during chemoradiation, with a total of 88 patients completing chemoradiation.
There were no significant differences in progression-free survival or overall survival between good responders and nonresponders. Three-year progression-free survival and overall survival were 78.7% and 78.2% for good responders and 72.1% and 85.2% for nonresponders, respectively.
In analyzing locoregional and distant control for the entire cohort, the investigators observed a strong trend toward improved locoregional control among good responders vs nonresponders (94.4% vs 79.7%). However, there was no significant difference in distant control between good responders and nonresponders.
Overall, late toxicity was improved with the use of response-adapted volume de-escalation. Nonresponders were significantly more likely to undergo gastrostomy tube placement during treatment (73.5% vs 50.0%, P = .04) and to be gastrostomy tube dependent at 6-month follow-up (32.6% vs 5.7%, P = .005).
HPV-Positive Subset
“No statistically significant difference was observed between overall and progression-free survival for good [responders] and nonresponders in the HPV-positive subset,” said Dr. Melotek. Three-year progression-free survival and overall survival were 84.2% and 84% for HPV-positive oropharyngeal squamous cell cancer good responders and 79.6% and 96% for HPV-positive oropharyngeal squamous cell cancer nonresponders, respectively. There was a strong trend toward improvement of locoregional control in good responders, with 3-year locoregional control at 100%, but no significant difference was observed in distant control.
The researchers performed a regression analysis on progression-free survival. “On univariate analysis, T stage and HPV status were significant, but on multivariate analysis, only HPV positivity was a predictor of progression-free survival,” said Dr. Melotek. Neither response-adapted volume de-escalation nor everolimus was associated with progression-free survival.
On univariate analysis, only T stage was a predictor of overall survival. “Again, neither receiving response-adapted volume de-escalation nor everolimus was significantly associated with overall survival,” he added.
Based on these results, the investigators concluded that “the elimination of elective nodal coverage in good responders did not appear to compromise outcomes in the entire cohort, nor specifically in the HPV-positive oropharyngeal squamous cell cancer subgroup, and resulted in improved toxicity,” Dr. Melotek told The ASCO Post. ■
Disclosure: Dr. Melotek reported no potential conflicts of interest.
Reference
1. Melotek JM, et al: 2016 Multidisciplinary Head and Neck Cancer Symposium. Abstract 3. Presented February 18, 2016.
