Prospective study of algorithm-based chemoradioselection to optimize treatment intensity for advanced head and neck cancer.

BACKGROUND

For the treatment of locally advanced head and neck squamous cell carcinoma (LA-HNSCC), a new strategy to reduce the excessive doses in the current standard high-dose cisplatin (CDDP) (100 mg/m x 3) chemoradiotherapy (CRT) is urgently needed.

METHODS

After 40 Gy of CRT (CDDP 80 mg/m), 131 patients with clinical stages Ⅲ or Ⅳ HNSCC were segregated into responders (≥50 % primary tumor reduction) who proceeded to an additional 30 Gy of CRT (CDDP 80 mg/m) (Res-CRT treatment group) or non-responders who either underwent radical surgery (non-Res-ope) or refused surgery in favor of an additional 30 Gy of CRT (CDDP 80 mg/m) (non-Res-CRT).

RESULTS

The algorithm stratified patients into 106 (81 %) responders (Res-CRT) and 25 non-responders (11 in the non-Res-CRT and 14 in the non-Res-ope group). Favorable 5-yr overall survival OS (73 %) and 5-yr laryngoesophageal dysfunction-free survival (LEDFS) (49 %) were obtained for LA-HNSCC. The responders (Res-CRT) showed significantly better progression-free survival (PFS) than non-responders (HR: 2.144; 95 % CI: 1.236-3.719; P = 0.0054). Among the three groups, the non-Res-CRT showed the worst PFS. In addition, the non-Res-CRT demonstrated significantly worse PFS (HR: 2.864; 95 % CI: 1.394-5.883; P = 0.0027) and LEDFS (HR: 2.758; 95 % CI: 1.219-6.241; P = 0.0376) compared to the Res-CRT. Good health-related QOL was maintained in patients with preserved larynx.

CONCLUSIONS

Algorithm-based chemoradioselection accurately selected patients who were good candidates for optimal intensity CRT (CDDP, maximum 160 mg/m) or radical surgery, achieving favorable survival, functional laryngeal preservation, and substantially good QOL. This platform may provide an alternative to high-dose CDDP CRT in LA-HNSCC.