Impact of obesity on response to first line treatment in primary immune thrombocytopenia - data from Latin America population

Abstract

Abstract Introduction Although Immune thrombocytopenia (ITP) natural history and treatment have been well characterized, the determinants of response to first line therapy are not well defined. In recent years, obesity has emerged as a possible adverse prognostic factor in primary ITP, with a few publications, reporting lower response to corticosteroid therapy and lower treatment-free survival for obese patients, when compared to normal weight counterparts. However, this prognostic impact needs further validation in different patient cohorts. Methods We conducted a retrospective study of adult patients with primary ITP treated at two tertiary centers in Brazil (University of São Paulo and University of Campinas). Inclusion criteria were primary ITP, appropriate information on first line treatment and available BMI data. Electronic medical records were reviewed to assess response to first-line therapy, stratified by body mass index (BMI) categories (normal weight, overweight, and obese). Secondary outcomes included duration of response, number of treatment lines, and platelet counts during follow-up. Results Of 340 screened patients, 94 met inclusion criteria. Women comprised 60% of the patients and median age at diagnosis was 45 years (IQR 27 – 59). Forty-one percent of patients were classified as normal weight, while 27% and 32% were classified as overweight and obese, respectively. Prednisone was the corticosteroid of choice in 64% of cases. Rates of overall response to initial treatment were similar across groups (normal weight 79%, overweight 68%, obese 83%, P = 0.37). Rates of complete response (52%, 48% and 47%, respectively) and partial response (29%, 20% and 39%, respectively) were also similar across groups (P = 0.57). Median platelet counts at diagnosis for normal weight (10x109/L, IQR 3 – 20), overweight (8x109/L, IQR 2 – 22) and obese (10x109/L, IQR 4 – 24) were similar. The platelet counts were also similar in follow-up, with median platelet counts for normal weight (58x109/L, IQR 12 – 151), overweight (93x109/L, IQR 22 – 169) and obese (64x109/L, IQR 53 – 135) at 3 months after treatment. In the obese group, the rate of loss of response was 68%, opposed to 51% and 47% among normal and overweight, respectively, but this finding was not statistically significant (P = 0.32). Moreover, we observed no difference in the median duration of response across groups (P = 0.20). Discussion In our cohort of primary ITP patients undergoing first-line corticosteroid treatment, we did not find a difference in outcomes when comparing normal weight to obese patients. These findings contrast with previous reports suggesting a negative prognostic impact of elevated BMI. As was the case with other works on the subject, our study was limited by its retrospective design. Moreover, our data is from reference centers, with most patients starting the follow-up in our services after having received first line treatment. This has impacted the size of our patient cohort, what limits its power to detect differences across groups. Nonetheless, our findings question the prognostic impact of BMI in the outcomes of primary ITP, highlighting the need for further research on the topic.