A 7 1 9 -A 8 1 3 chemotherapy. Methods: We performed a systematic literature search of the EMBASE, MEDLINE, Cochrane Library, Google Scholar, ABI/Inform, and the Web of Science using such search terms "filgrastim," "pegfilgrastim," "cost analysis," and "economic evaluation." Studies were limited to primary research in patients with solid tumor cancer, specifically, studies comparing filgrastim with pegfilgrastim and resulting in full manuscripts. Identified studies were evaluated by the Drummond checklist 6 and characterized by study perspectives, time horizon, data sources, and funding. Results: Six studies fulfilled the inclusion criteria. Most studies modeled hypothetical cohorts of women aged 30-80 years with breast cancer (Stages I-III) from a payer's perspective. The median Drummond score was 9 of 10 (range, 8-9). Methodological and reporting variations were common. Key assumptions were made about FN-related deaths during chemotherapy, hospitalization and outpatient management, chemotherapy costs, and data sources. All six studies were funded by the drug manufacturer. Pegfilgrastim was found to be cost-saving compared to 11-day filgrastim. However, when compared to 6-day filgrastim, the choice of intervention depends on the decision-maker's willingnessto-pay. ConClusions: Variations in methodology, reporting, and assumptions made comparisons between studies difficult and may explain in part the observed results reported in EEs. Studies independent of industry sponsor are needed to make conclusive interpretations. PCN22objeCtives: Cost-effectiveness analysis and cost-utility analysis were adopted to evaluate the tertiary breast cancer screening and diagnosis system in Guangdong province. Methods: Using data from Guangdong project to evaluate the validity and reliability of screening strategies. The intervention group received tertiary screening and diagnosis system, while control group received routine screening. The actual cost, detection rate and cost-effectiveness ratio were calculated. The Markov simulation model was constructed based on the natural history of breast cancer with TreeAge Pro 2011. The model was running over thirty years (each cycle represents one year). The sensitivity analysis was performed for incidence of breast cancer and health state utility. Results: The intervention group involved 26224 females while the control group involved 24282. The detection rate of breast cancer (1/10 million) was 91.54 and 28.86. The percentage of early stage breast cancer was 45.83% and 28.57%, respectively. The highest detection rate was found in women aged from 45 to 65. In order to detection one case of breast cancer, the number need to invite for screening program was 1595. Cost-effectiveness analysis was 6152.37yuan per detection rate of breast cancer (1/10 million). During the following 30 years, comparing to the control group, the tertiary breast cancer screening and diagnosis system for 100 thousand women will reduce 61 cases of breast cancer, and save 557.00 LYs, 649.05 QALYs. With the discount ra...