摘要目的:评估在脑MRI检查普及的条件下，全脑预防性放疗（PCI）对根治性放化疗后不同疗效的局限期小细胞肺癌（SCLC）患者预后的影响。方法:病例来源于本课题组2002年6月—2017年1月在中山大学肿瘤防治中心及浙江省肿瘤医院开展的关于局限期SCLC胸部放疗的开放标签、前瞻性临床试验。本研究回顾分析其中接受根治性放化疗后疗效评价达完全缓解（CR）或部分缓解（PR）的患者。按不同疗效分层，按是否接受PCI分组，比较患者的生存情况。采用Kaplan-Meier法计算生存数据，采用Cox比例风险模型进行预后影响因素的多因素分析。结果:接受根治性胸部放疗及化疗的309例局限期SCLC患者中，133例患者疗效评价达到CR，140例患者疗效评价达到PR，将这273例患者纳入本研究。133例CR患者中，29例未行PCI，其余接受PCI的104例患者中89例（85.6%）行脑MRI排除脑转移。中位随访时间为22.1个月，在接受和未接受PCI患者中，累积脑转移发生率分别为18.3%和37.9%（ P=0.020）；中位OS期分别为30.2、30.5个月；1、3、5年OS率分别为93.3%、41.9%、27.7%和82.8%、44.8%、40.8%（ P=0.910）。基线卡诺夫斯凯计分（KPS）为90分是CR患者OS的独立预后有利因素（ HR=0.93，95% CI为0.89~0.98， P=0.006）。而140例PR患者中，52例未行PCI，其余88例患者中80例（90.9%）在PCI前行脑MRI排除脑转移。中位随访时间为18.9个月，在接受和未接受PCI患者中，累积脑转移发生率分别为10.2%和44.2%（ P<0.001）；中位OS期分别为26.0、18.0个月，1、3、5年OS率分别为86.4%、37.9%、32.2%和75.0%、17.3%、10.8%（ P=0.001）。基线KPS为90分（ HR=0.93，95% CI为0.89~0.97， P=0.001）以及PCI（ HR=0.54，95% CI为0.36~0.80， P=0.002）是PR患者OS的预后有利因素。 结论:对于胸部放化疗后疗效评价CR及PR的局限期SCLC患者，PCI可显著降低其脑转移发生率及改善PR患者OS，但未能显著改善CR患者OS。

abstractsObjective:To evaluate the impact of prophylactic cranial irradiation (PCI) on the prognosis of patients with limited-stage small cell lung cancer (SCLC) in the era of widespread application of MRI.Methods:Clinical data were collected from an open-lable prospective clinical trial on thoracic radiotherapy target volumes for limited-stage SCLC conducted in Sun Yat-sen University Cancer Center and Zhejiang Cancer Hospital between June 2002 and January 2017. In this study, patients who achieved complete remission (CR) or partial remission (PR) after definitive chemoradiotherapy (CRT) were retrospectively analyzed. Stratified analysis was performed according to different clinical efficacies. Patients were divided into different groups according to whether PCI was conducted or not. Survival analysis of patients was carried out. Survival data were calculated by Kaplan-Meier method, and Cox proportional hazards model was applied for multivariate prognostic analysis.Results:Among 309 patients with limited-stage SCLC who received CRT, 133 patients achieved CR and 140 cases obtained PR. These 273 patients were enrolled in this study. Among 133 patients with CR, 29 of them did not receive PCI, and 89 (85.6%) of the remaining 104 patients receiving PCI underwent brain MRI to exclude brain metastasis (BM) before PCI. With a median follow-up time of 22.1 months, the cumulative BM rates were 18.3% and 37.9% in patients who received or did not receive PCI ( P=0.020). The median overall survival (OS) was 30.2 and 30.5 months, and the 1-, 3- and 5-year OS rates were 93.3%, 41.9%, 27.7% and 82.8%, 44.8%, 40.8%, respectively ( P=0.910). Multivariate analysis indicated that baseline Karnofsky performance status (KPS) = 90 was a favorable independent prognostic factor for OS in CR patients ( HR=0.93, 95% CI: 0.89-0.98, P=0.006). Among 140 patients achieving PR, 52 cases did not receive PCI and 80 (90.9%) of the remaining 88 patients received brain MRI before PCI. With a median follow-up time of 18.9 months, the cumulative BM rates were 10.2% and 44.2% ( P<0.001). The median OS was 26.0 and 18.0 months, and the 1-, 3-, and 5-year OS rates were 86.4%, 37.9%, 32.2% and 75.0%, 17.3%, 10.8%, respectively ( P=0.001). Baseline KPS = 90 ( HR=0.93, 95% CI: 0.89-0.97, P=0.001) and PCI ( HR=0.54, 95% CI: 0.36-0.80, P=0.002) were favorable prognostic factors for OS in PR patients. Conclusions:PCI significantly reduces the incidence of BM and prolongs the OS in patients with limited-stage SCLC who achieve PR after CRT, but it fails to significantly prolong the OS of CR patients.