胃肠间质瘤肝转移的临床特征及生存分析

目的 探讨胃肠间质瘤（GIST）肝转移患者的临床病理特征和预后因素。方法 回顾性分析2002年1月至2014年1月入组的复发或转移性GIST共138例患者，其中肝转移者74例。根据治疗前病灶部位将全组病例分为单纯肝转移组34例、腹盆腔转移组64例和肝腹盆腔转移组40例。均给予伊马替尼起始剂量400 mg／d口服。用Logistic回归分析近期疗效相关因素，生存分析用Kaplan-Meier法，预后多因素分析用Cox回归模型。结果 135例可评价近期疗效。74例肝转移患者中，CR 11例，PR 41例，SD 18例，PD 4例，有效率（RR）为70.3％，疾病控制率（DCR）为 94.6％。3组RR、DCR的差异均无统计学意义（P＞0.05）。Logistic回归分析显示，是否伴同时性肝转移是影响RR的独立因素。全组患者的中位无进展生存期（PFS）为52个月，中位总生存期（OS）为66个月，1、2、3和5年生存率分别为97.0％、89.6％、82.3％和60.0％。肝转移患者的中位PFS为45个月，中位OS为68个月，1、2、3和5年生存率分别为97.2％、92.5％、87.4％和59.2％。单纯肝转移组、腹盆腔转移组和肝腹盆腔转移组的中位PFS分别为61、56和30个月，中位OS分别为75、65和63个月。是否合并其他部位转移和近期疗效是影响PFS的独立因素，是否合并其他部位转移和年龄是影响其OS的独立因素。3组患者的主要不良反应为水肿、白细胞减少和腹泻，多为1～2级，3组不良反应发生率的差异均无统计学意义（P＞0.05）。结论 肝转移并未影响伊马替尼治疗晚期GIST的近期疗效，肝外病灶、年龄和近期疗效是影响GIST肝转移患者远期生存的重要因素。

Analysis of clinical features and prognostic factors of gastrointestinal stromal tumors with hepatic metastasis

Objective To discuss the clinical features and prognostic factors of gastrointestinal stromal tumor( GIST) with he-patic metastasis. Methods From January 2002 to January 2014, 138 patients with recurrent or metstatic GIST were enrolled, among whom 74 cases occurred with hepatic metastasis. According to the recurrent sites, patients were divided into 3 groups:hepatic metasta-sis only group ( n=34) , abdominal metastasis group ( n=64) , and abdominal and hepatic metastasis group ( n=40) . All patients were given imatinib mesylate orally with initial dose of 400 mg per day. Logistic regression model was used to analyze factors affecting short-term efficacy;Kaplan-Meier method was employed to analyze long-term efficacy; prognostic factors were analyed by Cox proportional hazards regression model. Results A total of 135 cases could be evaluated for curative effect. In 74 cases with hepatic metastasis, 11 received CR, 41 received PR, 18 received SD, and 4 received PD. The response rate ( RR) and disease control rate ( DCR) were 70. 3% and 94. 6%, respectively. There were no statistical differences in RR and DCR among the 3 groups ( P>0. 05) . According to the Logistic regression model analysis, simultaneous liver metastasis was the independent factor affecting RR. The median progression-free survival ( PFS) and median overall survival ( OS) of the whole group were 52 months and 66 months, respectively. The 1-, 2-, 3-and 5-year survival rates were 97. 0%, 89. 6%, 82. 3% and 60. 0%, respectively. The median PFS and OS of patients with hepatic me-tastasis were 45 months and 68 months, respectively. The 1-, 2-, 3-and 5-year survival rates were 97. 2%, 92. 5%, 87. 4% and 59. 2%, respectively. The median PFS of hepatic metastasis only group, abdominal metastasis group, and abdominal and hepatic me-tastasis group were 61, 56 and 30 months, respectively. The median OS of the 3 groups were 75, 65 and 63 months, respectively. Ac-cording to the Cox proportional hazards regression model analysis, companying with other organ metastasis and short term efficacy were the independent factors affecting PFS, while companying with other organ metastases and age were the independent factors affecting OS. The main adverse reactions of the three groups were edema, leukopenia and diarrhea, mainly in grade 1-2. There were no statistical differences in adverse reactions among the 3 groups. Conclusion Short term efficacy is not affected by hepatic metastasis in GIST pa-tients taking imatinib mesylate. Extrahepatic lesions, age and response rate are important factors affected the survival of the GIST pa-tients who have hepatic metastasis.