Double-blind, randomized trial of granulocyte-colony stimulating factor (GCSF) versus (v.) placebo during postoperative radiation (RT) for advanced resectable squamous cell head and neck cancer (SCCHN): Impact on mucositis

Abstract

5587 Background: Mucositis is common in patients (pts) with SCCHN receiving RT, contributing to the need for tube feeds (PEG) and treatment delays. GCSF has decreased mucositis in pts receiving chemotherapy (CT). We evaluated its ability to decrease mucositis and related dysfunction during postoperative RT using a randomized, placebo-controlled design. Methods: Eligible pts had completely resected, stage II-IV SCCHN, no prior RT or CT, and signed informed consent. Pts were stratified by primary site, then randomized to receive GCSF 3 mcg/kg or placebo daily by subcutaneous injection during RT (63 Gy, 1.8 Gy/day). Pts and treating physicians were blinded to treatment assignment. Endpoints: need for PEG feeds (primary), severity of mucositis/level of intervention for mucositis (secondary; both scored biweekly, grades 0–3). Results: 41 pts (132 planned) were enrolled at MSKCC from 1/92 to 12/96. Study was closed due to slow accrual. Pt characteristics (GCSF v. placebo): median age 59 v. 54; pathologic stages: T3/4 32 v. 50%, N2/3 68 v. 59%, Stage IV 79 v. 68%. Primary site distribution was comparable between groups. All pts Karnofsky ≥ 80. Mean absolute neutrophil count during RT 24.1 v. 4.1 x109/L (p<0.0001). 9 pts treated with GCSF required dose reduction (white blood count ≥ 40 x109/L, radiotherapist blinded), 2 pts had grade 2/3 bone pain, 3 pts total refused injection (2 GCSF, 1 placebo). 39 pts were evaluable. All received 63 Gy. Outcomes (GCSF v. placebo): PEG rate 0 v. 10% (p=0.16), RT interruption 0 v. 10% (p=0.16), mean mucositis grade 0.84 v. 1.11 (p=0.11), mean level intervention 0.59 v. 0.95 (p=0.076), mean # of treatment days 35.9 v. 38.1 (p=0.007). 5-year actuarial survival (intent to treat, 41 pts, median followup 7.6 years) was 79 v. 47% (p=0.062, log rank test). Conclusions: GCSF during RT was feasible, led to trends for less PEG feeds and mucositis (underpowered), significantly less # of treatment days, and no decrement in survival. Survival data are not clearly explained by known prognostic factors, and warrant further study. No significant financial relationships to disclose.