Treatment patterns and costs in second-line (2L) metastatic colorectal cancer (mCRC).

Abstract

544 Background: Administrative claims databases are an important data source to examine treatment (tx) patterns, costs and outcomes, but are subject to limitations. This study included a chart review to assess a claims-based algorithm for identifying 2L mCRC patients (pts). Tx patterns and costs were compared between mCRC pts treated with bevacizumab (BEV) or cetuximab (CET). Methods: 2LmCRC pts were identified from a US commercial insurance claims database (1/2008-6/2011). Initial identification required pts to have ≥2 claims ≥30 d apart for colon or rectal cancer and metastatic (met) disease. Continuous enrollment for ≥6 mo before and after the date of 1st met claim, ≥2 lines of therapy (LOT), and 2L tx with BEV or anti-EGFR was required. 2L was defined as addition of any new agent ≥28 d after start of 1st LOT. Charts from 92 pts were abstracted and used to refine and corroborate the algorithm. Generalized Linear Models (GLM) were used to assess differences in healthcare costs between patients on BEV or CET during 2L. Results: The match rate of claims identified mCRC pts treated with multiple lines of therapy confirmed by chart abstraction ranged from 85% to 97%. Applying the final algorithm to the overall claims data resulted in the identification of 569 pts: 450 pts receiving BEV and 119 pts receiving CET. A total of 38 panitumumab patients were excluded from analysis due to small sample size. Pt characteristics were similar; mean age was 61 years and 58% were males. In 2L, BEV was commonly used with 5FU or capecitabine-based regimens (81%) and CET with irinotecan-based regimens (73%). BEV pts had significantly lower all-cause healthcare costs (adjusted difference: –$12,239, p=0.02), and medical costs (–$13,672, p=0.01) during 2L. While on second-line therapy, BEV pts also had lower average monthly all-cause and medical costs than CET pts. Sensitivity analysis using variations of the case finding algorithm yield similar results. Conclusions: The use of BEV during second line treatment of mCRC was found to be associated with significantly lower healthcare costs relative to the use of CET from a large US healthcare claims database.