Long-term results of Intergroup RTOG 91–11: A phase III trial to preserve the larynx—Induction cisplatin/5-FU and radiation therapy versus concurrent cisplatin and radiation therapy versus radiation therapy

Abstract

5517 Background: The 2-year results of Intergroup RTOG 91–11 were published in 2003 (NEJM 349:2091–8,2003). We now present the 5-year results (after median follow-up for surviving patients of 6.9 years) of 515 eligible pts with resectable stage III or IV (excluding T1 and high volume T4), cancer of the glottic or supraglottic larynx. Methods: Patients were randomized to induction cisplatin/5-FU (CF) with responders then receiving RT (I+RT) (n = 173); or concurrent cisplatin (100 mg/m2 q 21 days × 3) and RT (CRT) (n = 171); or RT alone (R) (n = 171). Laryngectomy was performed for < partial response to induction CF, for persistent/recurrent disease or for laryngeal dysfunction. Results: At 5 years, laryngectomy-free survival (LFS) was significantly better with either I+RT (44.6%, p = 0.011) or CRT (46.6%, p = 0.011) compared to R (33.9%). There was no difference in LFS between I+RT and CRT (p = 0.98). Laryngeal preservation (LP) was significantly better with CRT (83.6%) compared to I+RT (70.5%, p = 0.0029) or R (65.7%, p = 0.00017). Local-regional control (LRC) was significantly better with CRT (68.8%) compared to I+RT (54.9%, p = 0.0018) or R (51%, p = 0.0005). I+RT compared to R for LP and LRC showed no significant difference (p = 0.37 and 0.62, respectively). The distant metastatic rate was low (I+RT 14.3%, CRT 13.2%, R 22.3%) with a trend (p ∼0.06) for benefit from chemotherapy. Disease-free survival (DFS) was significantly better with either I+RT (38.6%, p = 0.016) or CRT (39%, p = 0.0058) compared to R (27.3%). Overall survival rates were similar for the first 5 years (I+RT 59.2%, CRT 54.6%, R 53.5%); thereafter I+RT had a non-significant lower death rate. Compared to CRT, significantly more pts in the R group died of their cancer (34% vs 58.3%, p = 0.0007); the rate for I+RT was 43.8%. Conclusion: These 5-year results differ from the 2-year analysis by a significant improvement in LFS now seen for both I+RT and CRT treatments compared to R. For the endpoints of LP and LRC, CRT is still the superior treatment with no advantage seen to the addition of induction CF to R. There is no significant difference in overall survival. [Table: see text]