SAN FRANCISCO – National trends in gastrointestinal stromal tumors suggest that diagnosis of this malignancy has become more accurate and survival has improved in recent years, thanks to some key advances in molecular medicine.
In a population-based study, the number of new cases of gastrointestinal stromal tumors (GIST) annually rose almost fivefold between 1998 and 2002, according to results reported at a meeting on gastrointestinal cancers sponsored by the American Society of Clinical Oncology. It leveled off thereafter.
Lead investigator Dr. Elan R. Witkowski attributed the sharp rise mainly to the advent of testing for c-kit mutations, which drive the disease in the vast majority of cases. "Since [we’ve had] the ability to test for c-kit mutations, the diagnosis became much more precise," he explained in an interview, so tumors that would have otherwise been misclassified are now correctly recognized to be GIST.
Researchers should take note, according to Dr. Witkowski, a surgical resident at the University of Massachusetts in Worcester. "Prior to 2002, we probably shouldn’t be using this code to do research, [but] from 2002 on, we think it’s safe to do research with that one specific code," he said.
The study also found that survival was significantly better for patients with GIST that was diagnosed in 2005-2007 than for their counterparts with GIST that was diagnosed in 2002-2004.
"Our hypothesis is that the availability of Gleevec [imatinib] and the broadening of its indications are what are primarily responsible for that survival benefit," he commented, noting that there were no major advances in surgery or radiation therapy for the disease during that same time period.
Imatinib, manufactured by Novartis, is an inhibitor of the c-KIT tyrosine kinase (as well as the BCR-ABL and platelet-derived growth factor receptor tyrosine kinases). It was initially approved for GIST by the Food and Drug Administration in 2002, for the treatment of KIT-positive, unresectable and/or metastatic malignant tumors. Its GIST indication was later expanded to include adjuvant treatment in adults after resection of KIT-positive tumors.
For the study, the investigators analyzed data from the SEER (Surveillance, Epidemiology, and End Results) database for the years 1998-2007. They identified cases of GIST using the GIST-specific diagnostic code 8936 from the International Classification of Diseases for Oncology, third revision (ICD-O-3).
Analyses were based on 3,568 adult patients with a new GIST diagnosis. They were 63 years old on average, and the majority was white (71%) and male (53%), according to results reported in a poster session. Their tumors predominantly were in the stomach (52%) or small bowel (30%), and had a SEER stage of localized (51%).
Results showed that the annual number of new GIST cases in the database rose sharply between 1998 and 2002 (from approximately 100 to nearly 500), but remained fairly steady thereafter. The same pattern was also evident across individual centers participating in SEER, according to Dr. Witkowski.
"One of the things you see in the literature – particularly in prior years – is that it approximated the incidence of GIST by using a number of different variables because the diagnosis was not certain. So you had 10, 12, 15 variables that were often being combined together to approximate GIST," he commented. "Coinciding with [the increase in GIST diagnoses] was a decrease of all those other diagnosis codes that used to be used to approximate it."
The proportion of the patients who had a recommendation for surgery in their records fell between 2002 and 2007, from 85% to 80% (P = .0008). And the proportion of patients who actually underwent surgery fell as well, although not significantly so.
"Before the availability of Gleevec, surgery was really the only chance for significant response," said Dr. Witkowski. "So I think people were generally more aggressive."
Median overall survival improved in a comparison of patients receiving a new diagnosis in 2002-2004 with their peers receiving a new diagnosis in 2005-2007, with the actuarial 3-year rate increasing from approximately 70% to 75% (P = .03). Improved survival was noted both for patients who underwent resection and for those who did not.
In adjusted analyses for the years 2002-2007, patients had a decreased risk of death with each increase in the year of diagnosis (hazard ratio, 0.9) and if they underwent surgery (HR, 0.5).
On the other hand, they had an increased risk of death if they had a distant or regional stage of disease vs. a localized stage (HR, 2.4 and 1.6) or were male (HR, 1.2). Also, the risk rose with age (HR, 1.1).
Dr. Witkowski reported having no conflicts of interest related to the study.