Abstract: Background　Small cell lung cancer （SCLC） is highly invasive，characterized by rapid tumor doubling time，early metastasis，easy recurrence，and poor prognosis.At the time of diagnosis，only about 30% of patients are in limited disease.Although SCLC is highly sensitive to radiotherapy and chemotherapy，some patients will develop drug resistance and tumor proliferation after a period of treatment.Therefore，to find out the factors affecting the response and outcome following treatment in SCLC patients will play a certain role in stratified management and prognosis improvement of these patients.Objective　To analyze the serum levels of NSE and Pro-GRP in the evaluation of response and outcome following initial chemotherapy in SCLC patients.Methods　From July 2015 to July 2016，103 patients with SCLC confirmed by pathological examination were recruited from Department of Respiratory Medicine，Changhai Hospital.All of them received initial chemotherapy for 4 to 6 cycles.Serum NSE and Pro-GRP levels were measured before and after chemotherapy.The patients were followed up as of 2017-07-31，and the progression-free survival（PFS） were recorded.Results　Among the 103 patients，7 （6.8%） had complete response （CR），18 （17.5%） had partial response （PR），53 （51.4%） had stable disease （SD），and 25 （24.3%） had disease progression （PD）．Compared with baseline，patients with CR and PR had decreased average serum NSE and Pro-GRP levels after the chemotherapy，so did those with SD（P<0.05）．But there were no significant differences in average pre- and post-chemotherapy NSE and Pro-GRP levels in PD patients （P>0.05）．The average levels of NSE and Pro-GRP in CR+PR，SD，and PD groups were significantly different after chemotherapy （P<0.05）．Spearman rank correlation test revealed that pre-chemotherapy NSE and Pro-GRP levels had no correlations with treatment response （rs=0.019，P=0.845；rs=0.171，P=0.086），but postchemotherapy NSE and Pro-GRP levels did （rs=0.342，P<0.001；rs=0.350，P<0.001）．The distribution of responses to initial chemotherapy differed significantly between those with NSE< upper limit of normal（ULN） and those with NSE≥ULN （P<0.05）．But the distribution of responses to initial chemotherapy did not vary significantly between those with Pro-GRP < ULN and those with Pro-GRP ≥ULN （P<0.05）．The PFS changed obviously according to tumor stage，metastasis status，levels of postchemotherapy NSE and Pro-GRP（P<0.05）．Multivariate survival analysis using Cox's regression model showed that tumor stage 〔HR=2.19，95%CI （1.29，3.72）〕 and postchemotherapy NSE level 〔HR=2.67，95%CI （1.62，4.44）〕 were prognostic factors of SCLC patients （P<0.05）．Conclusion　Postchemotherapy NSE and Pro-GRP levels can well monitor the response to chemotherapy in SCLC patients.In particular，postchemotherapy NSE level may also be used for predicting the outcome.

Key words: Small cell lung carcinoma, Pro-gastrin releasing peptide, Neuron specific enolase, Treatment outcome, Prognosis, Root cause analysis

摘要： 背景　小细胞肺癌（SCLC）具有高度的侵袭性，其临床特点为肿瘤倍增速度快、早期发生转移、治疗后易复发、预后差。在诊断时，仅有约30%的患者处于局限期（LD）。尽管SCLC对于放疗和化疗的敏感性高，但经过一段时间的治疗后部分患者会出现耐药及肿瘤的扩散。故找出影响SCLC患者疗效及预后的因素，对于SCLC分层管理、改善患者预后将起到一定作用。目的　分析评价初治SCLC患者化疗前后神经元特异性烯醇化酶（NSE）、胃泌素释放肽前体（Pro-GRP）水平对化疗后效果及预后的评估价值。方法　选取2015年7月—2016年7月上海长海医院呼吸内科收治并病理证实为SCLC患者103例，对其行4~6周期的初始化疗治疗。于化疗前及化疗后检测血清NSE、Pro-GRP水平。对患者进行随访，末次随访时间为2017-07-31，记录患者无疾病进展期（PFS）。结果　103例患者化疗周期结束后完全缓解（CR）7例（6.8%），部分缓解（PR）18例（17.5%），疾病稳定（SD）53例（51.4%），疾病进展（PD）25例（24.3%）。CR+PR、SD患者化疗后NSE、Pro-GRP水平均低于化疗前（P<0.05）；PD患者化疗前与化疗后NSE、Pro-GRP水平比较，差异均无统计学意义（P>0.05）。CR+PR、SD、PD患者化疗后NSE、Pro-GRP水平比较，差异均有统计学意义（P<0.05）。经秩相关检验，化疗前NSE、Pro-GRP水平与疗效无相关关系（rs=0.019，P=0.845；rs=0.171，P=0.086）；化疗后NSE、Pro-GRP水平与疗效存在相关关系（rs=0.342，P<0.001；rs=0.350，P<0.001）。化疗后NSE<参考范围上限（ULN）与NSE≥ULN者疗效分布比较，差异有统计学意义（P<0.05）；Pro-GRP<ULN与Pro-GRP≥ULN者疗效分布比较，差异无统计学意义（P>0.05）。不同肿瘤分期、有无肿瘤转移、化疗后NSE水平、化疗后Pro-GRP水平者PFS比较，差异均有统计学意义（P<0.05）。多因素Cox风险比例回归分析结果显示，肿瘤分期〔HR=2.19，95%CI（1.29，3.72）〕、化疗后NSE水平〔HR=2.67，95%CI（1.62，4.44）〕是SCLC患者预后的影响因素（P<0.05）。结论　化疗后NSE、Pro-GRP水平对SCLC患者化疗效果有很好的监测作用，化疗后NSE水平可以预测SCLC患者预后。

关键词: 小细胞肺癌, 胃泌素释放肽前体, 神经元特异性烯醇化酶, 治疗结果, 预后, 影响因素分析