Implementation of the HIF activator IOX-2 in routine doping controls – Pilot study data

Early in 2020, racehorse doping cases revolved around the hypoxia-inducible factor (HIF) activator IOX-2. While the composition of IOX-2 has also been known and monitored in human doping controls for several years, the testing capability of routine sports drug testing methods was revisited for this newly surfaced doping agent. IOX-2 and the analytically well-established HIF activator roxadustat (FG-4592) share identical precursor/product ion pairs, enabling their co-detection in existing initial testing procedures in routine doping controls for the intact unconjugated analytes. In addition, hydroxylated IOX-2 and the corresponding glucuronic acid conjugates were identified as major metabolites in a microdose elimination study, contributing to enhanced initial testing and confirmation procedures.


| INTRODUCTION
Hypoxia-inducible factor (HIF) activators are prohibited in sports both in-and out-of-competition. 1 A variety of drug candidates has been closely monitored in sports drug testing since 2012, 2 Figure 1B), and the testing capability of human routine sports drug testing methods was revisited for this newly surfaced doping agent.
By association (and elemental composition), the co-detection of IOX-2 and roxadustat as intact and unconjugated analyte using identical precursor/product ion pairs was utilized (Figures 2 and 3) and since 2015, no adverse analytical findings have been recorded. 4 However, to date no metabolic biotransformation reactions have been assessed.
The aim of this pilot study was the implementation of IOX-2 into an existing initial test method to enable detection at the lowest possible additional workload for the laboratory. Furthermore, a microdose elimination study was performed to identify the compound's major metabolites and to allow a first estimation of urinary excretion profiles and detection times.

| Sample preparation and instrumentation
Analytical parameters of IOX-2 using established routine doping control methods 5 were determined from reference substance analyses. Urine sample analysis was conducted by means of fortifying an aliquot of 95 μL of urine with 5 μL of an internal standard working solution, and 10 μL was subsequently injected into the liquid chromatographic-mass spectrometric instrument.
For chromatographic separation, a Vanquish UHPLC System (Thermo Scientific, Bremen, Germany) equipped with a Nucleodur C18 Pyramide analytical column (2 × 50 mm, 1.8 μm particle size; Macherey-Nagel, Düren, Germany) and a guard column (2 × 4 mm) of the same material was used. The mobile phases were composed of A: 0.1% formic acid and B: acetonitrile. The LC gradient (total run time:

| Microdose elimination study
With written consent, a microdose elimination study was performed with one healthy male volunteer (64 years, 81 kg, 170 cm), who was orally administered 1 mg of IOX-2. Urine samples were collected before and up to 91 h post-administration, and specimens were subjected to routine doping control analytical approaches (diluteand-inject) as well as dedicated product ion scan experiments on in silico predicted phase I and phase II metabolites.

| Method validation
The method for the semi-quantitative determination of IOX-2 was comprehensively characterized with regard to the following parameters according to World Anti-Doping Agency (WADA) guidelines. 6 Selectivity: Ten different blank urine samples obtained from healthy volunteers were tested for the presence of interfering Carryover: The carryover was assessed by analyzing one sample with a concentration set at 400% of the MRPL (8.0 ng/mL) followed by the injection of one blank sample.

Limit of detection (LOD):
The LOD is the lowest concentration of an analyte that can be detected in 95% of representative samples (ie 95% detection rate). Here, in each case six urine samples were fortified with 0.02-2.00 ng/mL of IOX-2 and analyzed using the ITP.
Robustness: The robustness of the approach was determined by analyzing six different urine specimens spiked at 50% of the minimum required performance level (MRPL) of the compound (1.0 ng/mL). The coefficients of variation (%CV) were calculated on the basis of the ISTD-normalized peak areas as well as relative retention times (rRT).
Robustness was estimated using the CP.

Limit of identification (LOI):
The LOI is the lowest concentration of an analyte that meets the WADA Technical document TD IDCR 7 criteria in 95% of representative samples (ie 95% identification rate or 5% false-negative rate). Here, in each case six urine samples were fortified with 0.02-2.00 ng/mL of IOX-2 and analyzed using the CP.   (Figure 3). In addition to the unmodified intact substance of IOX-2, its hydroxylated analog and the corresponding glucuronic acid conjugates were also detected in post-administration urine samples (Figure 3), supporting both initial testing and confirmation procedures in routine doping controls.

| IOX-2 excretion profile and detection window
As demonstrated in Figure 4, the urinary excretion profiles of the intact drug, as well as the identified major metabolites follow a similar excretion profile with peak concentrations between 4 and 6 hours after application of the substance. All metabolites including the glucuronide. In none of the cases examined, were IOX-2 or any of the described metabolites found.

| Method validation
The method employed for a semi-quantitative determination of IOX-2 after direct injection of native urine specimens treated with ISTD was comprehensively characterized in accordance with WADA criteria. The results of the method validation are summarized in Table 1. According to the fact that the aforementioned IOX-2 metabolites are not commercially available, most of the described method validation parameters (robustness, LOD, LOI, and linearity) were estimated with IOX-2 itself, while selectivity and carryover was done for all the investigated metabolites. The assay is characterized by a high selectivity with almost zero biological noise in the blank urine specimens for IOX-2 and all investigated IOX-2 metabolites. The approach was found to be linear from 0.25 to 125 ng/mL (R 2 > 0.99) with an LOD (ITP) of 0.6 ng/mL and an LOI (CP) of 0.4 ng/mL. Furthermore, the assay demonstrates adequate robustness (CV% rRT: 0.1; CV% area ratio: 13.3) and carryover (< 1%). Although the compound has a high (mis)use potential in sports, retrospective monitoring for IOX-2 and its metabolites indicates that the drug is not likely to have been widely abused over the time period of assessment. However, further studies are indicated in order to complement the pattern of urinary metabolites and their utility in terms of enhanced retrospectivity.