Annals of Surgical Oncology

, Volume 14, Issue 6, pp 1784–1786

Should Surgical Resection Be Combined with Imatinib Therapy for Locally Advanced or Metastatic Gastrointestinal Stromal Tumors?


  • Sam S. Yoon
    • Division of Surgical Oncology, Department of SurgeryMassachusetts General Hospital, Harvard Medical School
    • Division of Surgical Oncology, Department of SurgeryMassachusetts General Hospital, Harvard Medical School

DOI: 10.1245/s10434-006-9344-x

Cite this article as:
Yoon, S.S. & Tanabe, K.K. Ann Surg Oncol (2007) 14: 1784. doi:10.1245/s10434-006-9344-x

Before the 1980s, the majority of stromal neoplasms of the gastrointestinal tract were thought to be of smooth muscle origin and were often termed leiomyomas and leiomyosarcomas. Electron microscopy and immunohistochemical studies in these tumors found both smooth muscle and neural features in many of these tumors,1 and the term gastrointestinal stromal tumor (GIST) was coined.

Kindblom et al.2 in 1998 noted that GISTs were similar in phenotype to the interstitial cells of Cajal, which were an obscure population of cells in the gastrointestinal tract thought to act as intestinal pacemaker cells. Interestingly, mice with a mutation in the c-kit gene failed to develop interstitial cells of Cajal.3 The c-kit gene encodes for the KIT receptor tyrosine kinase that binds stem cell factor, and Hirota et al.4 subsequently found that mutations in the c-kit gene led to constitutive activation of the receptor in five GISTs.

Larger studies have subsequently demonstrated that the majority of stromal neoplasms of the gastrointestinal tract are in fact GISTs, and we now know that approximately 95% of GISTs stain positive for the KIT receptor.5 GISTs have a unique tumor biology compared with most other solid tumors in that they are often highly reliant on a single pathway, the KIT receptor tyrosine kinase pathway, for neoplastic growth. In patients with metastatic GISTs, treatment with imatinib (Gleevac, STI-571) results in high rates of response (approximately 45%) or stable disease (approximately 30%) and a median progression-free survival of nearly 2 years.6

Imatinib is taken as a daily pill (400–800 mg) and is generally well tolerated, with the major side effects being mild fatigue, edema, diarrhea, and muscle cramping.7 Unfortunately, metastatic GISTs ultimately behave as other malignant solid tumors and develop alternative pathways to promote tumor growth, resulting in imatinib resistance. Given the efficacy of imatinib against GISTs, should surgical resection be combined with imatinib therapy for locally advanced or metastatic GISTs?

In a recent issue of Annals of Surgical Oncology, Bonvalot et al. from the Institut Gustave Roussy in France reported their series of 22 patients with initially unresectable or metastatic GISTs who were treated with imatinib and then underwent surgery.8 Five of these patients had emergent surgery due to bleeding, perforation, or fistulization, and three of these five patients died in the early postoperative period. An additional five patients with locally advanced primary tumors underwent complete macroscopic resection. Of the remaining 12 patients with metastatic GISTs, 10 had complete macroscopic resection. The median length of time on imatinib before surgery was 12 months. After a median follow-up of 32 months, median progression-free survival in the 17 patients who underwent elective operation was 23.4 months.

This group’s approach to the surgical treatment of locally advanced or metastatic GISTs considers new paradigms in treating GIST that are usually not considered for other gastrointestinal malignancies, where chemotherapeutic and biologic therapies are often not as effective as imatinib. Perhaps the treatment of locally advanced or metastatic GIST should be considered more analogous to ovarian epithelial cancer, which is highly responsive to platinum-based chemotherapy. Ovarian cancer is usually discovered at an advanced stage and metastasizes most commonly to the peritoneal cavity. Standard treatment for stage III ovarian cancer is initial surgical cytoreduction followed by intraperitoneal or intravenous platinum-based chemotherapy. Median survival in optimally cytoreduced patients receiving chemotherapy is >4 years.9

The optimal timing of surgery for patients with locally advanced or metastatic GISTs in relation to imatinib therapy remains to be determined. In the Bonvalet et al. study, patients were treated with imatinib for a median of 12 months and a range of 1 to 30 months. The development of imatinib resistance is likely directly related to the number of tumor cells and the duration of imatinib therapy. By using this rationale, one could argue that surgical resection or debulking should be instituted relatively early in the course of imatinib treatment. Positron emission tomography scans can determine response to imatinib after a week or less of therapy,10 and so one need not wait several months to see tumor shrinkage on the basis of serial computed tomographic scans. Approximately 5% of patients receiving imatinib experience intra-abdominal or gastrointestinal hemorrhage (usually from necrosis of large, bulky tumors),7 and early surgical resection or debulking may reduce this risk as well.

The arguments for surgery later in the course of imatinib treatment include possibly a higher rate of complete resection, increased organ preservation, and less surgical morbidity. In addition, only a small fraction of tumors cause marked hemorrhage. Although the number of patients in this study is small, this study, along with other small series,11,12 supports the use of surgical resection or debulking in select patients with locally advanced or metastatic GISTs. As with all small retrospective series, it is difficult to discern the precise beneficial effect of surgery against possible confounding issues such as selection bias. The 17 patients selected for elective surgery in this series are a small subset of 180 patients who had been initially treated with imatinib.

Clearly some patients with locally advanced or metastatic GISTs will not benefit from surgical intervention, particularly those whose disease progresses while receiving imatinib therapy. In a recent study from Poland, eight patients were operated on for salvage therapy after their disease progressed while they were receiving imatinib.13 Complete macroscopic resection was possible in only two patients. One patient died perioperatively, and only one patient was alive beyond 1 year. In the series by Raut et al.,14 32 patients underwent surgery after localized disease progression and 14 patients after generalized disease progression while receiving imatinib. Median progression-free survival was only 8 months and 3 months, respectively. For patients with de novo resistance or who develop resistance to imatinib, effective alternatives exist. SU11248, a small molecule inhibitor of KIT, platelet-derived growth factor receptor, and other receptor tyrosine kinases have been shown to improve survival in patients with imatinib-resistant GISTs.15 For patients with initially resectable GISTs, standard therapy remains complete surgical resection.

Several studies evaluating the efficacy of imatinib after complete resection have been completed or are being conducted. For example, in the American College of Surgeons Oncology Group Z9001 trial, subjects with completely resected GISTs of >3 cm in size are randomized to imatinib or placebo for 1 year, and similar adjuvant imatinib trials are underway in Europe. The Radiation Therapy Oncology Group is examining the potential role of imatinib before surgery in patients with resectable GISTs. In the study RTOG-S0132, patients with resectable primary or recurrent GISTs receive an initial 4-week course of imatinib followed by repeat cross-sectional imaging and positron emission tomography scan. Stable or responding patients receive an additional 4–6 weeks of imatinib followed by surgery and postoperative imatinib for 2 years. Results of this trial may define whether reduction in tumor mass early in the course of imatinib treatment will reduce the likelihood of resistance and lead to prolongation in survival compared with historical control patients. Molecular analyses in resected tumors will allow for correlation of clinical parameters to specific mutations, genomic profile, and protein activation.

The importance of close and frequent communication between treating medical and surgical oncologists experienced in management of GIST patients cannot be overemphasized. Numerous factors should be considered, including the potential for emergence of resistance with further imatinib therapy, the potential reduction in surgical morbidity with further response to imatinib, and the likelihood of a complication with continued therapy. Data emerging from several centers strongly support a role for resection of residual disease in patients with initially unresectable GISTs that respond to imatinib. However, we should heed the words of the Greek poet Hesiod, who noted in the 8th century bc in Works and Days, “Observe due measure, for right timing is in all things the most important factor.”

Copyright information

© Society of Surgical Oncology 2007