Postoperative radiation for lung cancer metastatic to the brain.

Abstract

PURPOSE Although resection of single brain metastases and postoperative whole-brain radiation therapy (WBRT) improves survival, compared with treatment using WBRT alone, the value of postoperative WBRT after resection of brain metastases is controversial. We analyzed the largest reported series of lung cancer patients with resected brain metastases to evaluate the impact of postoperative WBRT. MATERIALS AND METHODS Between 1974 and 1989, 185 patients with non-small-cell lung cancer (NSCLC) underwent resection of brain metastases. Patients who had received preoperative WBRT (23%, 42 of 185) were excluded. The remaining patients were divided into group A (no WBRT; n = 32), group B (patients received WBRT and were prognostically matched to group A; n = 32), and group C (all other WBRT patients; n = 79). Most patients received postoperative doses of 30 Gy in 10 fractions. Higher doses were used in 16% of group B and 18% of group C patients. RESULTS Overall 5-year survival rates were as follows: group A, 12%; B, 8%; C, 16%. Overall brain failures occurred in 38% of patients in group A, 47% in group B, and 42% in group C. The use of WBRT (group A v groups B plus C) had no apparent impact on survival or on overall brain failure rates. In particular, no improvement in either of these parameters could be demonstrated when group B was compared with group A. Focal failure (defined as failure within the brain adjacent to the site of the resected brain metastases) occurred as follows: group A, 34% (11 of 32); groups B plus C, 23% (25 of 111) (P = .07). WBRT significantly reduced focal failure for patients with adenocarcinoma (group A, 33% [eight of 24]; groups B plus C, 14% [11 of 79]; P = .05). Nonfocal failure (anatomically distinct from the resected metastasis) occurred in 9% of patients in group A (three of 32), 21% in groups B plus C (23 of 111) (P = .07). CONCLUSION Long-term survival is possible when NSCLC brain metastases are resected. Postoperative WBRT as used in this series only had an impact on the focal control of brain metastases and this effect was of borderline significance. The lack of conclusive benefit supports the need for ongoing randomized trials to test the value of adjuvant postoperative WBRT. Brain failures were relatively common in all three groups of patients, which suggests that doses greater than 30 Gy need to be studied.