Potential of supercritical fluid chromatography for endogenous anabolic agents detection in doping control

Code: 21A06AM

The detection of endogenous anabolic agents still represents one of the major challenges in doping control. Gas chromatographic separation in combination with combustion and isotopic ratio mass spectrometry (GC/C/IRMS) is needed to discriminate between intake and natural production of the endogenous steroids. However, the technique requires a very tedious and lengthy sample preparation that makes it inapplicable to screening analysis and limits its use to those samples that have been flagged suspicious in steroidal module of the Athlete Biological Passport (ABP) after GC-MSn analysis. Historically based on gas-chromatography, the analytical approach for anabolic agents has later expanded to include liquid chromatography, and more recently investigations have considered supercritical fluid chromatography (SFC) as well. The introduction in 2012 of the ACQUITY Ultra Performance Convergence Chromatography (UPC²) system, with improved robustness and performance, suggested that a widespread implementation in routine analysis would follow. However, laboratories have been slow in introducing the change, assumedly at least partly because of the limited research being performed and published so far. We intend to perform an in-depth investigation of the potential of UHPSFC combined with mass spectrometry for the analysis of endogenous anabolic agents in urine samples. In particular, this project will investigate the separation and quantification of endogenous steroids both in the free and conjugated form by UHPSFC-MS and evaluate the various factors influencing sensitivity. The optimal conditions for chromatography and MS will be established, such as the use of co-solvents, the effect of the MS parameters, in combination with an appropriate sample preparation procedure to maximise selectivity and sensitivity as well as achieving the required chromatographic separation. This project will provide an alternative or complementary analytical method to current GC/MSn methods traditionally used in anti-doping laboratories. Additionally, we aim to explore the potential use of UHPSFC combined to fraction collection to simplify and automate sample preparation for GC/C/IRMS analysis.

Main findings

We have developed a method based on UHPSFC-MS/MS to measure the urinary steroid profile for the athlete biological passport (ABP). The goal of the project was to understand whether UHPSFC-MS/MS can be considered a valid alternative to GC-MS for the analysis of endogenous steroids. A thorough investigation was performed on the main parameters influencing method performance and after method optimisation a complete validation was carried out. Depending on the steroid, the method is capable of quantification from 0.5 ng/mL up to 10 μg/mL by employing a weighted quadratic regression model. This is in line with current WADA requirements.

To demonstrate the applicability of the method, a comparison with our routine GC-MS method was performed. Authentic urine samples (128) were analysed using both method and the results compared using Passing-Bablok regression analysis. The results showed an excellent correlation and comparable results between the two methods. In our opinion this demonstrate the good performance of the method proposed. Further studies with a much larger sample population would be needed to confirm our results and eventually propose the approach as a valid alternative to the conventional GC-MS.

In a second part of the project, we have investigated the potential of UHPSFC combined with fraction collection for the sample purification prior to GC/C/IRMS analysis. Thanks to the presence of a splitter between the chromatographic column and the mass spectrometer, it is possible to collect fraction while analysing the remaining portion by MS on the same injection. As a proof of principle, we have manually collected the fraction corresponding to the different steroids, then analysed them by GC/C/IRMS. The same positive and negative control samples were also analysed with our routine method, that involves a sample preparation by a double HPLC purification prior to IRMS. Results obtained in terms of peak purity are encouraging and comparable between the routine method and the newly proposed approach. This suggests that further investigations should be pursued to automate the collection (using a fraction collector) and increase the sensitivity, which for the moment is limited by the maximum volume that can be injected on the system without compromising the chromatography.