LAUSR

http://dx.doi.org/10.1016/j.jtho.2016.12.002 In the literature, there are limited data on thoracic surgery in patients with SCLC and even more scarce data on prophylactic cranial irradiation (PCI) in these same patients. The main reason is that because of the rapid doubling time of SCLC, most patients present with bulky mediastinal disease (stage III) not accessible by surgery; furthermore, two old randomized trials led to the conclusion that surgery had no place in SCLC. There has, however, been a reappraisal of surgery in recent years, especially in early-stage SCLC (stages I and II), with several surgical series coming from Asia and large data base studies. Surgery can thus be considered a valid option in the case of complete resection. This trend may be partly due to the advances in diagnostic imaging, such as chest computed tomography (CT)–positron emission tomography (PET) and brain magnetic resonance imaging that have contributed to better selection of patients with more limited SCLC. It is indeed patients with pathologic stage (p-stage) I disease who benefit the most from a surgical strategy. In a recent retrospective series of 277 Japanese patients treated between 1974 and 2011, the 5-year survival rates among patients with and without surgical resection according to clinical stage were as follows: 62% and 25% in stage I (p < 0.01), 33% and 24% in stage II (p 1⁄4 0.95), 18% and 18% in stage III (p 1⁄4 0.35), respectively. In a Surveillance, Epidemiology, and End Results database study of 3566 patients with stage I or II disease treated between 1988 and 2007, 25% of patients underwent an operation. Of patients with stage I disease, those who had lung resection had a median survival of 38 months (95% confidence interval [CI]: 30.1–45.9) compared with 16 months (95% CI: 15.1–16.9, p < 0.001) among nonsurgical patients with stage I SCLC; among patients with stage II disease, median survival was 25 months (95% CI: 19.7–30.3) in those patients who had a lung operation compared with 14 months (95% CI: 12.9–15.1, p < 0.001) among nonsurgical patients. One of the main limitations of the Surveillance, Epidemiology, and End Results study is that there is no information regarding chemotherapy, which is the pivotal treatment of SCLC. In another more recent North American data base study, the outcome seemed improved with adjuvant chemotherapy and cranial irradiation among the 954 patients with completely resected pT1-2N0M0 SCLC who were analyzed. The multivariable Cox modeling suggested that adjuvant chemotherapy delivered to 57% of patients or chemotherapy with brain radiation delivered to 10% of patients was associated with improved survival when compared with no adjuvant therapy. This article does not give any information on the patterns of failure or, most particularly, on brain failure. Is brain failure a major event in resected SCLC? Several authors have shown that it was one of the main causes of failure. Xu et al. from the Shangai Chest Hospital have to be congratulated for their retrospective study collecting data from 349 SCLC resected patients, which adds evidence to the surgical strategy in SCLC, and to the unanswered question of PCI in this population. This article gives us several valuable pieces of information and also raises several questions. The selection of patients and their treatment are well described. As they were treated between 2006 and 2014, more than 70% of the patients had a PET scan and most had adjuvant chemotherapy as recommended in guidelines, but there is no information regarding baseline brain imaging. The authors address the question of PCI in this surgical population according to stage. As in NSCLC, this study seems to confirm that the risk for development of brain metastasis (BM) is higher in more advanced stage: the