dc.description.abstract | Abstract
Background: The World Anti-Doping Agency (WADA) bans the use of recombinant human erythropoietin (rHuEpo) in sports, challenging to detect with the Athlete Biological Passport (ABP) that monitors haematological data longitudinally. Since 2009, the ABP has identified potential doping trends, but the individual variability in transcriptomic signatures related to rHuEpo, high altitude, and exercise among non-doping individuals is still uncertain. This study seeks to create a biobank of non-doping samples to improve ABP's detection capabilities and establish transcriptomic reference ranges to reduce false doping results.
Methods: Four blood and urine samples were collected from 108 university students based in Eldoret, Kenya (~2100 above sea level) and Kisumu, Kenya (~1000 m above sea level) with 4 to 6 weeks between each collection. The students included Eldoret males (21±2 years), Eldoret females (22±2 years), Kisumu males (22±2) and Kisumu females (22±2 years). Blood was collected into a K2EDTA and a Tempus™ Blood RNA Tube for haematological and transcriptomic analysis, respectively. Haematological variables used as blood doping markers in the ABP include Red Blood Cells (RBC), Haematocrit (HCT), Haemoglobin (HGB), Mean Corpuscular Haemoglobin (MCH), Mean Corpuscular Haemoglobin Concentration (MCHC), Mean Corpuscular Volume (MCV), Number of Reticulocytes (RET#) and Reticulocytes percentage (RET%). The Off-score was calculated for each sample using the formula: Hgb x 10 - 60(√RET%). The “clean” status of athletes was assessed using an ABP style model, created in MATLAB (version 6.1.0 with Statistics Toolbox version 3.0). Cut-off was applied with an adaptive Bayesian model to calculate individualized upper and lower limits for these variables, incorporating factors such as mean subject variance, between-subject variance, sex, and baseline data. This method aimed to distinguish between drug-free samples, which stayed within these personalized limits and suspicious samples which deviated significantly. Statistical analysis of haematological variables such as HGB, RET% and OFF-score, crucial for doping detection, were performed using R (R Studio, Version 1.2.5042, ABPS package, Vienna, Austria).
Results: Males from both Eldoret and Kisumu consistently exhibited higher (p<0.05) haematological variables than their female counterparts. However, female participants from both Kisumu and Eldoret showed a significantly higher (p<0.05) RET% compared to males. None of the participants from Eldoret exceeded the Bayesian cut-off for any haematological variable. Participants from Kisumu exceeded the cut-offs at only three time points for both sexes, OFF-score values, for females HGB values. Sixty participants surpassed the ABPS cut-off. Transcriptomic analysis has not yet been conducted, but results are anticipated by July 2024.
Conclusions: The blood samples collected in this study offer invaluable insights into the haematological reference values for healthy, non-doping Kenyan student-athletes and serve as the critical establishment of a control group. This foundational step is crucial for the next phase of this research, which involves developing transcriptomic tests designed to improve the detection of rHuEpo doping. | |