Department of Surgery, Brigham and Women's Hospital and Dana-Farber Cancer Institute, Boston, MA.
Center for Sarcoma and Bone Oncology, Dana-Farber/Brigham and Women's Cancer Center, Boston, MA.
Ann Surg. 2018 Aug;268(2):296-302. doi: 10.1097/SLA.0000000000002281.
To refine treatment recommendations for patients with metastatic gastrointestinal stromal tumors (GISTs) treated with tyrosine kinase inhibitors (TKIs) and surgery.
Early reports suggested that patients with metastatic GIST responding to TKIs treated with surgery may have favorable outcomes. However, identification of prognostic factors was limited by small cohorts.
Progression-free survival (PFS) and overall survival (OS) from time of surgery and from start of initial TKI was determined. Multivariate analysis was conducted on all patients undergoing GIST metastasectomy between 2001 and 2014 at 2 institutions.
We performed 400 operations on 323 patients with metastatic GIST on TKIs. Radiographic response at time of surgery was classified as responsive disease (RD, n = 64, 16%), stable disease (SD, n = 100, 25%), unifocal progressive disease (UPD, n = 132, 33%), and multifocal progressive disease (MPD, n = 104, 26%). For patients on imatinib before surgery, radiographic response was predictive of PFS from time of surgery (RD 36 months, SD 30 months, UPD 11 months, MPD 6 months; P < 0.001) and from imatinib initiation (RD 71 months, SD 51 months, UPD 47 months, MPD 33 months; P < 0.001). Radiographic response was predictive of OS from time of surgery (RD not reached, SD 110 months, UPD 59 months, MPD 24 months; P < 0.001), and from imatinib initiation (RD not reached, SD 144 months, UPD 105 months, MPD 66 months; P = 0.005). Radiographic response was not predictive of PFS/OS for patients on sunitinib. Metastatic mitotic index ≥5/50 HPF, MPD, and R2 resection were prognostic of worse PFS/OS; primary mutation was not.
Surgery in metastatic GIST patients in the absence of MPD on imatinib is associated with outcomes at least comparable with second-line sunitinib and may be considered in select patients.
为接受酪氨酸激酶抑制剂（TKI）和手术治疗的转移性胃肠间质瘤（GIST）患者制定治疗建议。
早期报告表明，对 TKI 有反应并接受手术治疗的转移性 GIST 患者可能有较好的预后。然而，由于队列规模较小，对预后因素的识别受到限制。
分别从手术时间和初始 TKI 开始时间确定无进展生存期（PFS）和总生存期（OS）。对 2001 年至 2014 年在 2 家机构接受 GIST 转移切除术的所有患者进行多变量分析。
我们对 323 名接受 TKI 治疗的转移性 GIST 患者进行了 400 次手术。手术时的影像学反应分为反应性疾病（RD，n=64，16%）、稳定疾病（SD，n=100，25%）、局限性进展性疾病（UPD，n=132，33%）和多发性进展性疾病（MPD，n=104，26%）。对于术前接受伊马替尼治疗的患者，影像学反应可预测手术时的 PFS（RD 36 个月，SD 30 个月，UPD 11 个月，MPD 6 个月；P<0.001）和伊马替尼起始时的 PFS（RD 71 个月，SD 51 个月，UPD 47 个月，MPD 33 个月；P<0.001）。影像学反应可预测手术时的 OS（RD 未达到，SD 110 个月，UPD 59 个月，MPD 24 个月；P<0.001）和伊马替尼起始时的 OS（RD 未达到，SD 144 个月，UPD 105 个月，MPD 66 个月；P=0.005）。对于接受舒尼替尼治疗的患者，影像学反应不能预测 PFS/OS。转移有丝分裂指数≥5/50 HPF、MPD 和 R2 切除与较差的 PFS/OS 相关，而原发突变则无相关性。
在无 MPD 的情况下，伊马替尼治疗转移性 GIST 患者的手术与二线舒尼替尼的疗效相当，在某些患者中可以考虑使用。
