Thrombotic thrombocytopenic purpura (TTP) is a condition caused by deficiency of ADAMTS13 resulting in accumulation of ultra large Von Willebrand factor multimers (ULVWF), leading to micro thrombi in multiple organs. The varying susceptibilities of blood group antigens to ADAMTS13 have been demonstrated. A and B antigens are protective of VWF; and VWF purified from blood group O individuals has been shown to be cleaved faster by ADAMTS13 compared to VWF from blood group AB individuals. We proposed that there may be a difference in the incidence of blood groups in TTP patients compared with the general population. We felt this to be important for a life-threatening disease with poorly understood epidemiology. We report a retrospective analysis of 74 patients presenting from 1993 to 2008 with idiopathic TTP. We studied the incidence across various blood groups and also estimated the recurrence and mortality in each group. The incidence of various blood groups were as follows: O 36%, A 36%, B 25%, and AB 2%, compared with expected frequencies in the Detroit area: O 44%, A 33% B 20%, and AB 3%. There was a trend of lower than expected frequency of blood group O. There were 24 recurrences and 14 deaths, uniform across blood groups. We hypothesized that there may be an association between blood groups and the risk of TTP; however the differences in our study were not statistically significant. Recurrence and disease specific mortality did not appear to be impacted by blood group.

e14163 Background: Metastatic colorectal disease is generally incurable and treatment is palliative with the intent to balance toxicity with quality of life. Coin 3 trial showed that pre-chemotherapy platelet counts > 400,000 per μL were associated with poor survival when intermittent chemotherapy was used, whereas patients with lower platelet counts did not have any significant difference between the intermittent and continuous chemotherapy arms. Methods: We reviewed retrospectively 775 stage IV colorectal cancer patients at Karmanos Cancer Center over a 10 year period to see if high platelet count was associated with a poor outcome irrespective of treatment. Our analysis included 480 patients with adenocarcinoma who had not received chemotherapy prior to referral, and where information on the baseline platelet count, race, and age was available. We also analyzed the impact of race, age and bevacizumab use. We used Cox regression model for analysis. Results: Among the patients 48.3% were African American (AA) and 51.7% were Caucasians (C). 34.4 % had had PLT > 400,000 per μL. For those with lower platelet counts the median survival was 26.2 months in the C and 14.1 months in the AA groups respectively. Patients with platelet above 400,000 had a median survival of 15.2 months for C and 12.6 months for AA. Cox regression analysis, showed hazard ratios for outcome of death were; 1.16(1.07-1.26) p<0.001 for age (per 10 yrs), 1.60(1.31-1.94) [AA versus C(ref)] p< 0.001 for race and 1.35(1.10-1.65)[>400 versus <] p<0.004 for platelet count. In subset analysis, 296(61.7%) patients who received chemotherapy had data regarding use of bevacizumab (B). Among the 31.7% who received B, the median survival was 25.6 months compared to14.1 months in the no B arm. A Cox regression model using B as a stratification variable showed that the impact of race {hazard ratio = 1.32 (1.02-1.69) p =0.03} and platelet count {hazard ratio = 1.27(0.97-1.65) p =0.08} were much less. Conclusions: Pre-chemotherapy Platelet count< 400,000, C race and younger age are associated with improved survival. Use of bevacizumab may mitigate the impact of these factors.