Should adjuvant imatinib be used as primary treatment for gastrointestinal stromal tumors?

Peter Hohenberger  About the author

Gastrointestinal stromal tumors (GISTs) represent a rare tumor entity that occurs in the muscular layer of the digestive tract. These tumors commonly metastasize to the liver and peritoneum, and systemic chemotherapy or radiotherapy is unable to control tumor growth. The KIT gene is mutated in over 80% of GISTs. Imatinib, a small-molecule inhibitor of a group of tyrosine kinases, including KIT and BCR-ABL, effectively inhibits tumor proliferation. Within a few days of treatment, tumor control can be achieved and median survival in patients with metastastic disease was 57 months in the B2222 study compared with 19 months prior to the imatinib era.¹ Consequently, the use of the drug in an adjuvant setting of patients considered at risk for development of metastasis was investigated.

No TNM system of tumor staging is established for GIST. On the basis of a combination of tumor size and mitotic count per 50 high power fields, four categories of risk of malignancy have been defined: very low, low, intermediate and high.²

Data from the first adjuvant study were presented at the ASCO 2007 annual meeting. The American College of Surgeons Oncology Group trial Z9001 randomized 708 patients to receive either placebo or treatment with imatinib for 1 year. Before randomization all patients had complete resection of any GISTs that were at least 3 cm in size. Recurrence-free survival was the primary end point and this criterion was met at interim analysis. The study was discontinued upon the advice of the Independent Data Monitoring Committee because a significant difference in relapse rate in favor of imatinib was observed (3% versus 17%).³ At a median follow-up of 13 months, 21 of the 325 evaluable patients in the treatment group had developed metastases compared with 62 patients in the placebo arm (n = 319). The conclusion from this study was that patients who have undergone complete resection of a GIST of at least 3 cm should receive adjuvant imatinib for 1 year. Many oncologists, however, remain ambivalent about these data, which have not yet been published.

The Z9001 study and its conclusions have drawn criticism for several reasons.⁴ To assess the risk of malignant behavior in GISTs, tumor size and mitotic rate are combined; thus, tumors smaller than 5 cm are considered at intermediate or high-risk for malignant behavior only if they show a high mitotic count. The Z9001 study did not use mitotic count of the tumors as an entry criterion because of concerns on whether this would generate reproducible data. Considerable proportions of patients belong to the low-risk group and will be over-treated. Subgroup analysis showed that in patients with tumors between 3 and 6 cm in size, adjuvant therapy did not confer a statistically significant benefit.

Duration of treatment is another important factor. Patients with metastatic GIST commonly experience relapse if their imatinib therapy is discontinued, as has been shown in the French BFR14 phase III study.⁵ Some of the patients did not respond to the drug after reintroduction in the French BFR14 trial and a similar result was found following interruption of imatinib after 3 years in patients who had achieved tumor control.⁶ Discontinuation of the drug in patients who had complete resection of residual tumor resulted in tumor recurrence.⁷ Unlike conventional chemotherapy, tyrosine kinase inhibitors prevent proliferation, causing subsequent apoptosis of tumor cells. Thus, lifelong treatment with imatinib seems a viable option. It could even be argued that 1 year of treatment selects cell clones that are less responsive to subsequent imatinib therapy once metastases occur. Development of secondary mutations during treatment has been demonstrated⁸ and the outcomes of patients who experience recurrence after 1 year of adjuvant imatinib will be of great interest.

The issue of duration of adjuvant treatment is being investigated in the European SSGXVIII-AIO study. The trial initially assessed patients with resected high-risk tumors but also those with single metastases who received complete resection of the primary tumor and the metastases. Since then, the study population has been changed and adapted on the basis of knowledge now available. Patients with high-risk tumors or with tumor rupture are now randomized to 1 year versus 3 years of adjuvant therapy with imatinib. Recurrence-free survival is the primary end point in this trial, which reached its target accrual of 345 patients. It would be no surprise if it was demonstrated that 3 years of adjuvant therapy improves recurrence-free survival compared with only 1 year.

The most important aim of adjuvant therapy in oncology is to prolong survival. From this perspective, the Z9001 study is a self-fulfilling prophecy. Patients with a high risk of metastatic spread were subjected to treatment with a highly effective drug. Clearly, one would expect such treatment to improve the recurrence rate in comparison with placebo. Whether this advantage actually translates into improved long-term survival has yet to be demonstrated. The EORTC 62024 study will address this issue. In this study, patients with intermediate-risk or high-risk tumors are randomized to 2 years of imatinib therapy versus no treatment with survival being the primary end point. The study accrued its target population of 750 patients. Results will not be available until 2010 at the earliest, illustrating the problem of proving the value of adjuvant therapy in prolonging survival in the presence of any effective palliative treatment. The patients will be closely followed-up, therefore, treatment for recurrent disease will start at a minimal tumor load in comparison to patients treated in the B2222 study, who received treatment only at a later stage. Ultimately, as effective second-line or third-line therapies become available, survival may not be influenced by initial adjuvant therapy. Imatinib-free survival could be an interesting end point in terms of toxicity and cost of therapy.

The study concept of Z9001 was finalized in 2002, at a time when much less data on the biology of GIST were available. Today we know that the location of the tumor may be more clinically important than its size; larger GISTs of the stomach have a better biological behavior than smaller GISTs of the small intestine.⁹ Molecular pathology data show that certain mutations in the KIT proto-oncogene, such as exon 9 mutations, deletion of codon 557/558 or wild-type status of KIT, put the patient at an increased risk for metastases and such tumors show a poor or no response to imatinib compared with other subentities.^{10, 11} Taking the rarity of GIST into account, a centralized pathology assessment is feasible before embarking on extensive and expensive treatment. Finally, GISTs are one of the best understood solid malignancies in terms of clarification of their mechanism of development, molecular characterization, the development of adequate drugs and understanding treatment failure by mutational data. It is, therefore, hard to accept that a crude measurement such as tumor size alone should determine whether a patient receives molecularly targeted adjuvant therapy.

Competing interests

The author is a Consultant for Novartis and Pfizer and receives grant/research support from Novartis.

ARTICLE NAVIGATION -ARTICLE TOOLS

Full text
• Download PDF
• Send to a friend
• Export citation
• Rights and permissions
• Order commercial reprints

ISSUE NAVIGATION - THIS ISSUE

• Table of Contents
• Previous article
• Next article