1428P - Risk factors and features of recurrence after curative resection of gastrointestinal stromal tumor: results of the Kinki GIST Registry study

Date 29 September 2014
Event ESMO 2014
Session Poster Display session
Topics GIST
Surgery and/or Radiotherapy of Cancer
Presenter Junya Fujita
Citation Annals of Oncology (2014) 25 (suppl_4): iv494-iv510. 10.1093/annonc/mdu354
Authors J. Fujita1, T. Takahashi2, Y. Yamashita3, S. Sato4, K. Yamamto5, Y. Nakajima6, Y. Asao7, K. Tominaga8, T. Omori9, T. Tsujinaka10
  • 1Department Ofsurgery, NTT West Osaka Hospital, 543-8922 - Osaka/JP
  • 2Department Of Gastroenterological Surgery, Osaka University Graduate School of Medicine, Osaka/JP
  • 3Department Of Gastroenterological Surgery, Osaka City General Hospital, Osaka/JP
  • 4Department Of Surgery, Shizuoka General Hospital, Shizuoka/JP
  • 5Department Of Surgery, Osaka National Hospital, Osaka/JP
  • 6Department Of Surgery, Osaka Red-Cross Hospital, Osaka/JP
  • 7Department Of Abdominal Surgery, Tenri Hospital, Nara/JP
  • 8Department Of Gastroenterology, Osaka City University Garduate School of Medicine, Osaka/JP
  • 9Department Of Surgery, Osaka Police Hospital, Osaka/JP
  • 10Department Of Surgery, Kaizuka City Hospital, Kaizuka/JP



To develop the optimal method of postoperative adjuvant therapy for gastrointestinal stromal tumor (GIST), it is necessary to estimate the risk of recurrence and to predict the prognosis after resection.


Patients histologically diagnosed as GIST between Jan 2002 and Dec 2007 were enrolled into the GIST registry study by the Kinki GIST study group. A total of 570 patients in 40 institutiions who underwent curative resection of the tumor and were not given adjuvant therapy were investigated in this study. The prognostic factors and features of recurrence were assessed. And the patients were stratified by the modified-Flecher's (m-F) risk classification and validated for predicting the recurrence.


73 patients developed tumor recurrences. 5-year overall survival (OS) and disease-specific recurrence-free survival (RFS) were 89.8%, 86.0%, and 10-year OS and RFS were 73.1%, 72.5%, respectively. The independent risk factors for recurrence were non-gastric location (HR: 2.63, 95%CI: 1.49-4.65), tumor size >5cm (HR: 3.70, 95%CI: 1.96-7.04), mitotic index (MI) >5/50HPF (HR: 5.49, 95%CI: 3.40-10.0), presence of tumor rupture (HR:3.90, 95%CI: 1.72-8.85). The median period of recurrence after resection was 541 (range 34-3128) days and 67% of the recurrence occurred within 2 years, 86% within 3 years, 95% within 5 years. Patients with high mitotic count or with tumor rupture developed signifcantly earlier recurrece (MI >5 vs. <5; 370 vs. 855 days, rupture yes vs. no; 248 vs.578 days). The type of recurrence were hepatic; 48, peritoneal; 22, lymph node; 3, local site of the surgery; 2. According to the m-F classification, 5-yr RFS of low-risk, intermediate-risk, high-risk group were 97.9%, 97.3%, 54.6%, respectively. The sensitivity and specificity of recurrence of the high-risk patients by m-F classification were 89.2%, 81.9%, respectively.


Modified-F classification discriminates the risk of recurrence of GIST properly. Patients classified as high risk by m-F criteria should be given adjuvant therpy. The high MI or tumor rupture are at particularly high risk of recurrence, thus, intensive precaution should be considered for the patients meeting these criteria.


All authors have declared no conflicts of interest.