Evidence-based pathway plus electronic health records system matches major trials for colorectal cancer treatment outcomes

The BRiTE and BEAT trials have shaped the standard of care in metastatic colorectal cancer (see Table 1). In both studies, progression-free survival for their respective treatment protocols came in at 10 months and 10.8 months. Oncologists at the US Oncology Network believe they have produced equally good PFS results in both trials by using an in-house evidence-based treatment algorithm along with an electronic health records system.

US Oncology purchased the Web-based, cancer care–specific iKnowMed EHR company in 2004. The institution's Level I Pathways are in-house evidence-based guidelines that help oncologists in the network determine which treatment regimen will offer the most clinical effectiveness with the least amount of toxicity. Level I Pathways are developed and updated by a multidisciplinary task force made up of 1,000 community oncologists.

TABLE 1 BRiTE and BEAT trials

BRiTE = Bevacizumab Regimens' investigation of Treatment Effects

Clinical outcomes associated with bevacizumab-containing treatment of metastatic colorectal cancer: The BRiTE observational cohort study,

Oncologist

14:862-870, 2009; ASCO 2008 abstract 4026.

BEAT = Bevacizumab Expanded Access Trial

Safety and efficacy of first-line bevacizumab with FOLFOX, XELOX, FOLFIRI, and fluoropyrimidines in metastatic colorectal cancer: The BEAT study,

Ann Oncol

20:1842-1847, 2009; ASCO 2008 abstract 4025.

Thomas Cartwright, MD, and his colleagues on the US Oncology Pathways Task Force, based in The Woodlands, Tx, compared the PFS of colorectal cancer patients treated at US Oncology Network (USON) sites to the published results of the BRiTE and BEAT registries.

The iKnowMed EHR system holds more than 31,000 colon cancer patient records. The investigators mined this database for patients who received first-line therapy between July 2006 and September 2009. PFS for this population was compared with the results from the two trials. USON patients were excluded if no PFS data were entered, if the calculated PFS was negative, or if the PFS equaled zero (ASCO 2010 abstract 3626).

The average age at diagnosis was 70 years for the 2,746 USON patients who met the inclusion criteria. The most common treatment regimen (1,044) was FOLFOX plus bevacizumab (Avastin).

TABLE 2 Results stratified by age

According to the results, the median PFS for all USON patients was 11.2 months, more than one month beyond the PFS rates reported in BRiTE and BEAT. The authors also stratified the results by age and compared them to BRiTE results (see Table 2). Finally, the USON data were analyzed by Karnofsky performance status (PS), which was not done for BRiTE and BEAT data. The median PFS for USON patients by Karnofsky score were 11.4 months (PS of 80 to 100) and 9.3 months (PS ≤ 70).

The group concluded that their patients achieved a median PFS that was certainly comparable to BRiTE and BEAT patients and that using the EHR allowed for an accurate assessment of outcomes. "This study is a stepping-off point to explore the potential impact on outcomes achieved by using USON's evidence-based Level I Pathways to direct patient care," they wrote. "Future studies may be done to explore PFS by regimen."

Dr. Cartwright told Oncology News International that his group's results may not be reproducible using other EHR systems and more general treatment guidelines such as NCCN or ASCO. Nonetheless, he said the take-home message is that practicing evidence-based medicine will result in the best outcomes for patients.