BACKGROUND:The objective of the study was to describe patterns of care of patients with gastrointestinal stromal tumors (GISTs) in the United States in the tyrosine kinase inhibitor (TKI) era.PATIENTS AND METHODS:From November 2004 through March 2009, data were collected regarding demographics, diagnostic history, treatment, relapse, and survival of 882 patients with GIST from 122 community and academic medical practices.RESULTS:The most common first-line treatment for the 719 patients presenting with localized GIST was surgery (87%). Use of adjuvant imatinib increased after June 2007; 47% of patients enrolled in the registry considered by the investigator to be at high risk for recurrence received adjuvant imatinib after June 2007 versus 18% before. Overall, 56% of patients received imatinib and 11% received sunitinib. The utilization of targeted therapy increased over time (45% and 0.4% of patients received imatinib and sunitinib, respectively, in 2006 versus 56% and 11%, respectively, in 2009).CONCLUSIONS:These are the first GIST registry data from the TKI era. The use of targeted therapy for GIST has increased in accordance with updated treatment guidelines. Diagnosis of GIST has evolved with increased use of KIT testing. The duration of targeted therapy in the adjuvant therapy setting is similar in community and academic practices.

10061 Background: Imatinib (IM) is approved for adjuvant (ADJ) treatment (tx) of adult patients (pts) following complete, gross resection of GIST. In the IM registrational study, IM 400 mg was administered for one year. However, patient characteristics, tumor characteristics, dose, and duration of therapy are not known outside the context of a clinical trial. Methods: We used a data collection tool designed to provide insights into practice patterns surrounding the diagnosis and management of GIST pts (reGISTry) to explore nonclinical trial ADJ dosing in community and academic practices. Results: Since 11/04, 1,053 pts have entered the reGISTry. ADJ IM tx were reported as early as 2001 (n=1) with steady increases in 2007 (n=64) and 2008 (n=43). Most pts (56%) received ADJ therapy for over one year. Pts treated in the community seemed to receive a longer duration of therapy than those at academic centers (424d [2-1,970] vs. 348d [16-1,762]), but not statistically significant (p=0.21). Pts with small intestine primaries appeared to have a slightly longer duration of ADJ tx (433[3-1525] vs 378[2-1970] p=0.43), but no correlation with tumor size or mitotic count. Most pts (87%, 181/208) received IM 400 mg/d as the initial dose. 17.3% of ADJ IM dose changes were due to disease progression. Mitotic count was only reported in 33.6% of pts receiving ADJ tx. Conclusions: We found that most pts with primary GIST receive IM therapy at a dose of 400mg daily for more than one year. Duration of therapy did not appear to closely correlate with tumor size, site, or mitotic rate. Most pts receiving ADJ tx were treated without a known mitotic rate. Adjuvant imatinib in the tx of most pts with GIST does not appear to be based on standard, published, risk factors. Variable Adjuvant therapy N (%) Total = 208 No adjuvant therapy N (%) 1°localized GIST Tumor size >10 cm 73 (35.0) 162/658 (24.6) 2-5 cm 48 (23.0) 201/658 (30.5) Tumor location Stomach 109 (52.4) 397/660* (60.1) Small intestine 64 (30.8) 166/660* (25.1) Mitotic count >10/HPF 37 (17.8) 240/658 (11.9) 6-10/HPF 32 (15.4) 47/658 (7.1) * 2 pts did not have mitotic count or tumor size data available. Author Disclosure Employment or Leadership Position Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Novartis Novartis Novartis Novartis

10072 Background: Imatinib (IM) has been studied in over 2,500 patients (pts) in phase II and III trials in both the adjuvant and metastatic settings. Standardized dosing using 400 and 800 mg has proven to be efficacious. Careful dose escalation from 400 mg/d to 800 mg/d is hypothesized to improve the tolerability of the higher dose. Methods: We used a data collection tool designed to provide insights into practice patterns surrounding the diagnosis and management of GIST pts (reGISTry) to explore nonclinical trial dosing as initial dose or dose modification in community and academic practices. Results: Since 11/04, 1,053 pts have been enrolled. Out of 613 pts receiving IM, nonstandard dosing was observed in 142 instances (29.9%). Primary reasons for nonstandard dose changes were toxicity and disease progression. In patients with reported Exon 9 mutations, the mean IM dose administered was 489.5 mg. Three patients with Exon 9 mutations alternated IM and S through the course of their treatment. Nonstandard dosing was reported in both the adjuvant and metastatic/unresectable settings as seen below. Conclusions: Despite the demonstrated benefit of standardized doses, non-standard dosing schemas were observed. One analysis by Demetri et al (J Clin Oncol 2009) demonstrated that patients with IM Cmin below 1,100 ng/mL showed a shorter time to progression and lower rate of clinical benefit. Observed pts in the reGISTry had only a slightly higher mean dose (489.5 mg). Dosing as observed in the reGISTry may lead to suboptimal trough levels due to the pharmacokinetic profile of IM. Further education in the importance of dose intensity and side effect management may ultimately improve clinical outcomes. Nonstandard dosing in reGISTry Dose (mg) Schedule Either 50, 100, 150, 200, 250, 275, 300, 350, 500, or 700 QD 100 BID 100 TID 200 or 400 Two days on/one day off 200 or 400 QOD 200/300; 200/400 Alternating daily 200 or 300 Two days on; one day off 300/200/0 or 300/300/0 Repeating cycle 400 Three times weekly 400 M-F 400 Four times weekly 400/800 400 M – F; 800 S, S Metastatic/unresectable Nonstandard doses (N = 141) Adjuvant nonstandard doses (N = 42) Most common 200 mg QD Most common 300 mg QD 60/141 (42.6%) 11/42 (26.2%) Author Disclosure Employment or Leadership Position Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Novartis Novartis Novartis Novartis

10557 Background: This observational reGISTry, initiated 11/04, characterizes the evolving patterns of care for GIST, such as the recent use and approval of Imatinib mesylate (IM) in adjuvant GIST. Methods: Data from consented pts (e.g. demographics, clinical characteristics, therapy, outcomes) are entered onto a web-based database. Updated analyses are performed every 6 months (mo), with data from unique sites being compared to the aggregate. Results: 792 pts enrolled from 121 centers, 55% from community practices. 79% were diagnosed with localized tumor, 87% of which received surgery as primary treatment. 94% had c-kit testing and 4% had mutational analysis performed at any time (1.4% of the pts in the community; 7.8% in Universities). 59% of pts had mutations in KIT Exon 11, 12% in Exon 9, 3% in Exon 17, 9% in PDGFRA Exon 18 and 18% had no detectable mutations. 13% of pts from Universities were enrolled in clinical trials vs 3% in the community. 78% of pts receiving IM at any time started at 400mg qd and 70% of pts receiving Sunitinib malate (SU) started at 50mg (4 wks on, 2 wks off). 6.6% of all pts received neoadjuvant IM, for a median of 4.3 mo for those pts that have completed (81%). 120 pts (15%) received adjuvant IM (13% of the pts in the community; 17% in Universities). Prior to Jun07 (ACOSOG Z9001 adjuvant IM positive results released) 14% of eligible pts received adjuvant IM vs 29% after Jun07. Median duration of adjuvant IM was 361 days for those pts that have completed (43%). Conclusions: reGISTry is a useful tool for measuring evolving pt management patterns in GIST capturing treatment variations from standard guidelines and differentiating management in Universities from that occurring in the community. Mutational analysis and clinical trial participation are still infrequent. The starting dose of IM and SU remains 400mg and 50mg, respectively, for most pts. The use of adjuvant IM has increased after Jun 07, suggesting that prescribing habits may have been influenced by evolving study data in these pts. [Table: see text] [Table: see text]