OpenLB
Open Library of Bioscience
Advanced Search
The international journal of biostatistics
Mar 28, 2025;

A multivariate Bayesian learning approach for improved detection of doping in athletes using urinary steroid profiles.

Dimitra Eleftheriou, Thomas Piper, Mario Thevis, Tereza Neocleous
Author Information
  1. Dimitra Eleftheriou: Leiden Academic Centre for Drug Research, 4496 Leiden University , Leiden, The Netherlands. ORCID
  2. Thomas Piper: Center for Preventive Doping Research - Institute of Biochemistry, German Sport University Cologne, Cologne, Germany.
  3. Mario Thevis: Center for Preventive Doping Research - Institute of Biochemistry, German Sport University Cologne, Cologne, Germany.
  4. Tereza Neocleous: School of Mathematics and Statistics, University of Glasgow, Glasgow, UK.
PMID: 40150811 DOI: 10.1515/ijb-2024-0019

Abstract

Biomarker analysis of athletes' urinary steroid profiles is crucial for the success of anti-doping efforts. Current statistical analysis methods generate personalised limits for each athlete based on univariate modelling of longitudinal biomarker values from the urinary steroid profile. However, simultaneous modelling of multiple biomarkers has the potential to further enhance abnormality detection. In this study, we propose a multivariate Bayesian adaptive model for longitudinal data analysis, which extends the established single-biomarker model in forensic toxicology. The proposed approach employs Markov chain Monte Carlo sampling methods and addresses the scarcity of confirmed abnormal values through a one-class classification algorithm. By adapting decision boundaries as new measurements are obtained, the model provides robust and personalised detection thresholds for each athlete. We tested the proposed approach on a database of 229 athletes, which includes longitudinal steroid profiles containing samples classified as normal, atypical, or confirmed abnormal. Our results demonstrate improved detection performance, highlighting the potential value of a multivariate approach in doping detection.

Keywords

Bayesian adaptive model anti-doping longitudinal biomarker data multivariate analysis one-class classification urinary steroid profile

References

  1. Kanayama, G, Pope Jr, HG. History and epidemiology of anabolic androgens in athletes and non-athletes. Mol Cell Endocrinol 2018;464:4���13. https://doi.org/10.1016/j.mce.2017.02.039 . [DOI: 10.1016/j.mce.2017.02.039]
  2. World Anti-Doping Agency . 2021 Anti-doping testing figures [Online]. https://www.wada-ama.org/sites/default/files/2023-01/2021_anti-doping_testing_figures_en.pdf [Accessed 19 Feb 2024].
  3. Sottas, PE, Saugy, M, Saudan, C. Endogenous steroid profiling in the athlete biological passport. Endocrinol Metabol Clin 2010;39:59���73. https://doi.org/10.1016/j.ecl.2009.11.003 . [DOI: 10.1016/j.ecl.2009.11.003]
  4. World Anti-Doping Agency . Technical document ��� TD2021EAAS [Online]. https://www.wada-ama.org/sites/default/files/2022-01/td2021eaas_final_eng_v_2.0.pdf [Accessed 19 Feb 2024].
  5. Sottas, PE, Robinson, N, Rabin, O, Saugy, M. The athlete biological passport. Clin Chem 2011;57:969���76. https://doi.org/10.1373/clinchem.2011.162271 . [DOI: 10.1373/clinchem.2011.162271]
  6. Vernec, AR. The athlete biological passport: an integral element of innovative strategies in antidoping. Br J Sports Med 2014;48:817���19. https://doi.org/10.1136/bjsports-2014-093560 . [DOI: 10.1136/bjsports-2014-093560]
  7. World Anti-Doping Agency . WADA technical document ��� TD2021APMU [Online]. https://www.wada-ama.org/sites/default/files/resources/files/td2021apmu_final_eng_0.pdf [Accessed 19 Feb 2024].
  8. Kuuranne, T, Saugy, M, Baume, N. Confounding factors and genetic polymorphism in the evaluation of individual steroid profiling. Br J Sports Med 2014;48:848���55. https://doi.org/10.1136/bjsports-2014-093510 . [DOI: 10.1136/bjsports-2014-093510]
  9. Piper, T, Geyer, H, Haenelt, N, Huelsemann, F, Schaenzer, W, Thevis, M. Current insights into the steroidal module of the athlete biological passport. Int J Sports Med 2021;42:863���78. https://doi.org/10.1055/a-1481-8683 . [DOI: 10.1055/a-1481-8683]
  10. Mareck, U, Geyer, H, Opfermann, G, Thevis, M, Sch��nzer, W. Factors influencing the steroid profile in doping control analysis. J Mass Spectrom 2008;43:877���91. https://doi.org/10.1002/jms.1457 . [DOI: 10.1002/jms.1457]
  11. Harris, EK. Effects of intra-and interindividual variation on the appropriate use of normal ranges. Clin Chem 1974;20:1535���42. https://doi.org/10.1093/clinchem/20.12.1535 . [DOI: 10.1093/clinchem/20.12.1535]
  12. Brooks, RV, Jeremiah, G, Webb, WA, Wheeler, M. Detection of anabolic steroid administration to athletes. J Steroid Biochem 1979;11:913���17. https://doi.org/10.1016/b978-0-08-023796-1.50129-8 . [DOI: 10.1016/b978-0-08-023796-1.50129-8]
  13. Donike, M, Barwald, K, Klostermann, KR, Schanzer, W, Zimmermann, J. Nachweis von exogenem testosteron [detection of exogenous testosterone]. Sport: Leistung und Gesundheit 1983;293���8.
  14. Sottas, PE, Robinson, N, Giraud, S, Taroni, F, Kamber, M, Mangin, P, et al.. Statistical classification of abnormal blood profiles in athletes. Int J Biostat 2006;2. https://doi.org/10.2202/1557-4679.1011 . [DOI: 10.2202/1557-4679.1011]
  15. Sottas, P-E, Saudan, C, Schweizer, C, Baume, N, Mangin, P, Saugy, M. From population-to subject-based limits of T/E ratio to detect testosterone abuse in elite sports. Forensic Sci Int 2008;174:166���72. https://doi.org/10.1016/j.forsciint.2007.04.001 . [DOI: 10.1016/j.forsciint.2007.04.001]
  16. Van Renterghem, P, Van Eenoo, P, Geyer, H, Sch��nzer, W, Delbeke, FT. Reference ranges for urinary concentrations and ratios of endogenous steroids, which can be used as markers for steroid misuse, in a Caucasian population of athletes. Steroids 2010;75:154���63. https://doi.org/10.1016/j.steroids.2009.11.008 . [DOI: 10.1016/j.steroids.2009.11.008]
  17. Sottas, P-E, Baume, N, Saudan, C, Schweizer, C, Kamber, M, Saugy, M. Bayesian detection of abnormal values in longitudinal biomarkers with an application to T/E ratio. Biostatistics 2007;8:285���96. https://doi.org/10.1093/biostatistics/kxl009 . [DOI: 10.1093/biostatistics/kxl009]
  18. Brown, PJ, Kenward, MG, Bassett, EE. Bayesian discrimination with longitudinal data. Biostatistics 2001;2:417���32. https://doi.org/10.1093/biostatistics/2.4.417 . [DOI: 10.1093/biostatistics/2.4.417]
  19. Alladio, E, Caruso, R, Gerace, E, Amante, E, Salomone, A, Vincenti, M. Application of multivariate statistics to the steroidal module of the athlete biological passport: a proof of concept study. Anal Chim Acta 2016;922:19���29. https://doi.org/10.1016/j.aca.2016.03.051 . [DOI: 10.1016/j.aca.2016.03.051]
  20. Amante, E, Pruner, S, Alladio, E, Salomone, A, Vincenti, M, Bro, R. Multivariate interpretation of the urinary steroid profile and training-induced modifications. The case study of a marathon runner. Drug Test Anal 2019;11:1556���65. https://doi.org/10.1002/dta.2676 . [DOI: 10.1002/dta.2676]
  21. Wang, M, Li, Z, Lee, EY, Lewis, MM, Zhang, L, Sterling, NW, et al.. Predicting the multi-domain progression of Parkinson���s disease: a Bayesian multivariate generalized linear mixed-effect model. BMC Med Res Methodol 2017;17:1���10. https://doi.org/10.1186/s12874-017-0415-4 . [DOI: 10.1186/s12874-017-0415-4]
  22. Xia, YM, Tang, NS. Bayesian analysis for mixture of latent variable hidden Markov models with multivariate longitudinal data. Comput Stat Data Anal 2019;132:190���211. https://doi.org/10.1016/j.csda.2018.08.004 . [DOI: 10.1016/j.csda.2018.08.004]
  23. Shamshoian, J, ��ent��rk, D, Jeste, S, Telesca, D. Bayesian analysis of longitudinal and multidimensional functional data. Biostatistics 2022;23:558���73. https://doi.org/10.1093/biostatistics/kxaa041 . [DOI: 10.1093/biostatistics/kxaa041]
  24. World Anti-Doping Agency . WADA technical document ��� TD2021IRMS [Online]. https://www.wada-ama.org/sites/default/files/resources/files/td2021irms_final_eng_0.pdf [Accessed 19 Feb 2024].
  25. R Core Team . The R project for statistical computing [Online]https://www.r-project.org/ [Accessed 20 Oct 2023].
  26. De Figueiredo, M, Saugy, J, Saugy, M, Faiss, R, Salamin, O, Nicoli, R, et al.. A new multimodal paradigm for biomarkers longitudinal monitoring: a clinical application to women steroid profiles in urine and blood. Anal Chim Acta 2023;1267:341389. https://doi.org/10.1016/j.aca.2023.341389 . [DOI: 10.1016/j.aca.2023.341389]
  27. DeGroot, MH. Optimal statistical decisions . Hoboken, New Jersey, United States: John Wiley & Sons; 2005, 82.
  28. Casella, G, George, EI. Explaining the Gibbs sampler. Am Statistician 1992;46:167���74. https://doi.org/10.1080/00031305.1992.10475878 . [DOI: 10.1080/00031305.1992.10475878]
  29. Gelman, A, Carlin, JB, Stern, HS, Dunson, DB, Vehtari, A, Rubin, DB. Bayesian data analysis . Boca Raton, Florida, United States: CRC Press; 2013.
  30. Minter, T. Single-class classification. In: LARS symposia ; 1975:54 p.
  31. Bishop, CM. Novelty detection and neural network validation. IEE Proc Vis Image Signal Process 1994;141:217���22. https://doi.org/10.1049/ip-vis:19941330 .
  32. Khan, SS, Madden, MG. One-class classification: taxonomy of study and review of techniques. Knowl Eng Rev 2014;29:345���74. https://doi.org/10.1017/s026988891300043x . [DOI: 10.1017/s026988891300043x]
  33. Hyndman, RJ. Computing and graphing highest density regions. Am Statistician 1996;50:120���6. https://doi.org/10.1080/00031305.1996.10474359 . [DOI: 10.1080/00031305.1996.10474359]
  34. Rauth, S. Referenzbereiche von urin��ren Steroidkonzentrationen und Steroidquotienten: ein Beitrag zur Interpretation des Steroidprofils in der Routinedopinganalytik [Ph.D. thesis]. Deutsche Sporthochschule K��ln; 1994.
  35. Kicman, AT, Coutts, SB, Walker, CJ, Cowan, DA. Proposed confirmatory procedure for detecting 5 alpha-dihydrotestosterone doping in male athletes. Clin Chem 1995;41:1617���27. https://doi.org/10.1093/clinchem/41.11.1617 . [DOI: 10.1093/clinchem/41.11.1617]
  36. Eleftheriou, D. Bayesian hierarchical modelling for biomarkers with applications to doping detection and prostate cancer prediction [Ph.D. thesis]. University of Glasgow; 2022.
  37. Bates, D, M��chler, M, Bolker, B, Walker, S. Fitting linear mixed-effects models using lme4. J Stat Software 2015;67:1���48. https://doi.org/10.18637/jss.v067.i01 . [DOI: 10.18637/jss.v067.i01]
  38. Kotsiantis, S, Kanellopoulos, D, Pintelas, P. Handling imbalanced datasets: a review. GESTS Int Trans Comput Sci Eng 2006;30:25���36.
  39. Schapire, RE. Empirical inference . Berlin, Germany: Springer; 2013:37���52 pp.
  40. Wilkes, EH, Rumsby, G, Woodward, GM. Using machine learning to aid the interpretation of urine steroid profiles. Clin Chem 2018;64:1586���95. https://doi.org/10.1373/clinchem.2018.292201 . [DOI: 10.1373/clinchem.2018.292201]
Journal Article
Article has an altmetric score of 2

Word Cloud

Created with Highcharts 10.0.0steroiddetectionanalysisurinarylongitudinalmultivariatemodelapproachprofilesBayesiananti-dopingmethodspersonalisedathletemodellingbiomarkervaluesprofilepotentialadaptivedataproposedconfirmedabnormalone-classclassificationathletesimproveddopingBiomarkerathletes'crucialsuccesseffortsCurrentstatisticalgeneratelimitsbasedunivariateHoweversimultaneousmultiplebiomarkersenhanceabnormalitystudyproposeextendsestablishedsingle-biomarkerforensictoxicologyemploysMarkovchainMonteCarlosamplingaddressesscarcityalgorithmadaptingdecisionboundariesnewmeasurementsobtainedprovidesrobustthresholdstesteddatabase229includescontainingsamplesclassifiednormalatypicalresultsdemonstrateperformancehighlightingvaluelearningusing

Similar Articles

Cited By