Evaluating surgery and first-line imatinib (ima) treatment patterns for patients with gastrointestinal stromal tumors (GIST).

Session Type and Session Title: 
General Poster Session, Sarcoma
Abstract Number: 
J Clin Oncol 31, 2013 (suppl; abstr 10590)
Michael Eaddy, Brian S. Seal, Mark R. Green, Dana Stafkey-Mailey; Xcenda, Palm Harbor, FL; Bayer HealthCare Pharmaceuticals, Wayne, NJ

Abstracts that were granted an exception in accordance with ASCO's Conflict of Interest Policy are designated with a caret symbol (^).

Abstract Disclosures


Background: In 2002, ima was FDA approved for advanced metastatic GIST, followed in 2008 with an indication for adjuvant therapy. Real-world data on changing patterns of ima therapy, including adjuvant management, is sparse. This study evaluated changes in treatment patterns, resection rates, and corresponding costs of care among patients receiving first-line (1LT) ima for GIST over an 8-yr period, with a breakpoint around time of initial reporting of phase III adjuvant therapy data showing survival benefit. Methods: The study was conducted Jan 1, 2003–Dec 31, 2010 utilizing pharmacy/medical claims data from a large health plan. Patients with a GIST-related ICD9 code who also received ima were eligible, and excluded if they received any chemotherapy prior to date of first diagnosis or ima. All were designated into 1 of 3 cohorts: metastatic disease at first recognition (MP), non-metastatic non-surgical (NS) patients, and non-metastatic patients undergoing a GIST-related surgical procedure before ima (SP). Patients were followed from diagnosis to end of study period or loss of continuous eligibility. Results: Patient assignment across the disease extent and surgery/no surgery cohorts by the phase III reporting event is shown in the Table. A change in distribution of primary disease presentation and/or management across two time intervals is suggested by marked decrease in percentage of MP (54.9%→36.1%) and greater than doubling of SP patients presumably receiving adjuvant therapy following initial resection attempt. In the study population, 18.8% of patients who initiated 1LT with ima went on to 2LT (73.7% sunitinib); 5.9% receive 3LT (44.4% nilotinib and 38.9% sorafenib); <1% receive 4LT (66.7% sorafenib). In general, 1LT was more costly than subsequent therapies. Across all lines of care, oral TKI drug acquisition cost was the dominant determinant of total cost of care. Conclusions: A change in distribution of primary disease presentation and/or management across the two time intervals is suggested. In each cohort, overall cost of care was largely a function of TKI acquisition cost.

2003–2006 (n=173) 2007–2009 (n=130) Total (n=303)
MP 54.9% 36.1% 46.9%
NS 34.7% 39.2% 36.6%
SP 10.4% 24.6% 16.5%