P-0056 Treatment Patterns and Outcomes in Metastatic Gastric Cancer in The Community Oncology Setting

      This paper is only available as a PDF. To read, Please Download here.



      There were an estimated 21,520 U.S. cases and 10,340 U.S. deaths from all gastric cancer in 2011, with increasing incidence of cancer of the cardia and gastro-esophageal junction. The five year survival of metastatic gastric cancer (mGC) is less than 5%, and long remissions are rare. NCCN guidelines recommend several platinum-based chemotherapy regimens as first line treatment of mGC. Chemotherapy plus trastuzumab is recommended for patients with HER2 positive disease. This study examined treatment patterns, outcomes, and HER2 testing over time in mGC patients treated in community oncology practices.


      This study was a retrospective review of medical records of patients diagnosed with mGC 1/1/2004 to 7/1/2011, and treated at one of 11 community oncology practices. Patients were≥18 years old, with no concurrent other malignancy. Primary study endpoints were treatment patterns, progression-free survival (PFS), and overall survival (OS). Kaplan-Meier survival analysis and Cox regression were used to examine PFS and OS across regimen groups. Covariates considered for Cox regression models were age, gender, race, stage at initial diagnosis, performance status, liver metastasis, lung metastasis, number of comorbidities, and tumor grade.


      Patients (N=243) were 59.7% Caucasian and 27.2% African American, 60.9% male, with mean age of 63.9 (±13.3) years. Most (70.4%) were initially diagnosed at stage IV; 18.1% had impaired performance status; 28.8%, 20.6%, 10.7% had cancers of gastro-esophageal junction, corpus, and cardia respectively. Of patients alive and under care, by year, HER2 testing had been done on: ≤2005: 0%, 2006: 2.6%, 2007: 3.3%, 2008: 7.5%, 2009: 18.42%, 2010: 41.8%, 2011: 48.1%. Median PFS overall by line was: first line: 4.4 months (N=243); second line: 3.2 months (N=100); third line: 3.0 months (N=38). First line regimen groups, with PFS (in months) by group, were Gp1: docetaxel+platinum±fluoropyrimidine (6.4 months; N=75); Gp2: epirubicin+platinum±fluoropyrimidine (6.7; N=22); Gp3: carboplatin+paclitaxel (3.8; N=21); Gp4: oxaliplatin+fluoropyrimidine (4.4; N=28); Gp5: Other platinum combinations (5.0; N=19); Gp6: Non-platinum regimens (3.7; N=27); Gp7: No systemic therapy (2.1; N=51). Overall log rank test p <.0001; excluding Gp7, p=.43. Second line groups with PFS were Gp1 (3.4 months; N=15), Gp4 (3.9; N=25), Gp6 (2.7; N=46) and Gp8: Irinotecan-based regimens (3.0; N=14). Overall log rank test p=.68. Third line groups with PFS were Gp5 (3.0 months; N=13), Gp6 (2.8; N=15) and Gp8 (3.6; N=10). Overall log rank test p=.15. Median OS was 8.0 months overall, 2.3 months in Gp7, and did not significantly vary by other first line regimen groups (log rank test p=.74). Cox regression showed impaired performance status predicted shorter PFS in all lines and shorter OS. Trastuzumab was included in regimens of 7/10 HER2 positive patients in first line, 5/7 in second line, and 2/2 in third line.


      Patients with mGC in community settings have PFS and OS similar to clinical trials. We were not able to detect statistically significant differences in outcomes across treatment regimens due to modest sample size. Treatment patterns reflect increasing testing of HER2 overexpression.