Impact of adding concurrent chemotherapy to hyperfractionated radiotherapy for locally advanced non-small cell lung cancer (NSCLC): comparison of RTOG 83-11 and RTOG 91-06.

A hyperfractionated radiation therapy (HFX RT) trial (1.2 Gy twice daily, b.i.d.) (HFX) for non-small cell lung cancer (NSCLC) showed that 69.6 Gy resulted in better survival than did lower total doses (Radiation Therapy Oncology Group, RTOG 83-11) and that cisplatin concurrent with irradiation improved local control and survival over RT alone (Radiation Therapy Oncology Group, RTOG 91-06). Concurrent combination chemotherapy and HFX could improve both local and systemic control. In a phase II trial (RTOG 91-06) for inoperable NSCLC, two cycles of PE were used [cisplatin 50 mg/m2 intravenously (i.v.) days 1 and 8, etoposide 50 mg orally (p.o.) b.i.d., 75 mg/day if body surface area (BSA) < 1.7 m2, days 1-14] starting on day 1 of HFX (69.6 Gy) and repeated on day 29. HFX/PE was compared with HFX (69.6 Gy) from an earlier phase II trial (RTOG 83-11). Seventy-six patients treated with HFX/PE and 203 patients who received HFX alone were compared for toxicity, response, survival, and patterns of failure. The rates of grade 4 nonhematologic toxicity were similar (3.0% for HFX/PE, 3.0% for HFX), but grade 4 hematologic toxicity occurred only with HFX/PE 56.6%. Three (3.9%) HFX/PE patients had fatal toxicity (2 pulmonary, 1 renal); 1 HFX patient had fatal esophageal toxicity. Response and metastasis rates were similar for the two treatments, but infield (p = 0.054) and overall (p = 0.04) progression-free survival rates were better with HFX/PE. Median survivals were 18.9 months with HFX/PE and 10.6 months with HFX. Two-year survival rates were 36% for HFX/PE and 22% for HFX (p = 0.014). The differences in survival between HFX/PE and HFX remained borderline statistically significant (p = 0.0593) in the multivariate model, which included weight loss, Karnofsky performance status (KPS), sex, and stage. HFX/PE is an effective regimen in patients with inoperable NSCLC, although it is considerably more toxic, and is undergoing a comparison in a three-arm randomized phase III study against induction cisplatin/vinblastine plus standard once-daily RT and against cisplatin/vinblastine concurrent with standard RT.