Three-Weekly Cisplatin Superior to Weekly Dose in Head and Neck Cancer

A phase III trial found that 3-weekly cisplatin is superior to weekly cisplatin with regard to preventing locoregional relapses when combined with radiotherapy for locally advanced head and neck squamous cell carcinoma (HNSCC). The 3-weekly schedule, however, does carry increased risk of severe acute toxicities (abstract 6007).

Multiple clinical trials established a 3-weekly dose schedule of cisplatin in locally advanced HNSCC. However, “this is a very toxic regimen to deliver, and as a result more than half of the patients are not able to receive all three planned doses of cisplatin,” said Vanita Noronha, MD, of Tata Memorial Hospital in Mumbai, noting that this causes dose intensity and outcomes to be compromised.

Though a weekly schedule has been used to aid with toxicity, it has never been robustly compared with the 3-weekly schedule. Noronha presented these results at the 2017 American Society of Clinical Oncology (ASCO) Annual Meeting.

The study included 300 patients randomized to weekly or 3-weekly cisplatin (150 in each group), followed for a median of 22 months. Patients had a median age of 44 years, most were male (89%), and the primary tumor site in most patients was the oral cavity (87.3%). Most patients were either smokeless tobacco users (71.3%) or smokers (19.7%).

In the weekly group, 88.7% of patients received at least 6 cycles of chemotherapy; in the 3-weekly group, 95.3% received at least 2 cycles (P = .1). Fourteen weekly patients (9.3%) and 12 3-weekly patients (8%) required a dose reduction. Radiotherapy treatment delivery was similar in the two groups.

The 2-year locoregional relapse rate was 38.67% with weekly cisplatin, and 24.67% with 3-weekly cisplatin, yielding a hazard ratio (HR) of 1.76 (95% CI, 1.11–2.79; P = .014).

Disease-free survival (DFS) differences did not reach significance, but favored the 3-weekly group. The median DFS was 20.8 months with weekly cisplatin, and 37.7 months in the 3-weekly group, for an HR of 1.38 (P = .069). Progression-free survival was similar, at 17.7 months with weekly cisplatin and 28.6 months with 3-weekly cisplatin, for an HR of 1.24 (P = .21). The median overall survival was 39.5 months in the weekly patients, and it was not yet reached in the 3-weekly patients (P = .48).

Acute toxicity was worse with the 3-weekly schedule, with 84.6% of those patients experiencing a grade 3 or higher adverse event, compared with 71.6% of weekly patients (P = .006). More 3-weekly patients required hospitalization (45.6% vs 32.4%; P < .001) and the use of growth factors (26.8% vs 3.4%; P < .001), though the rate of feeding tube use was no different. There was one toxic death in the 3-weekly group and none in the weekly group. The rate of severe chronic toxicities was generally low.

Noronha said that the superior results with 3-weekly cisplatin suggests it should remain the chemoradiotherapy regimen of choice for locally advanced HNSCC. Robert I. Haddad, MD, of the Dana-Farber Cancer Institute in Boston, was the discussant for the session, and he noted that a bolus administration of cisplatin is the standard of care in the postoperative setting, but the specific dosing is important.

“For all other patients with head and neck cancer, weekly cisplatin is not a de-intensification strategy” based on these data, he said. “In my opinion, bolus cisplatin is the standard of care for patients with head and neck cancer.”