IS DRIED BLOOD SPOT TESTING A SUITABLE METHOD TO CREATE THE ATHLETE BIOLOGICAL PASSPORT?

Master: Forensic Science Supervisor: Wim Best Co-assessor: Ate Kloosterman 19-01-2018

IS DRIED BLOOD SPOT TESTING A

SUITABLE METHOD TO CREATE THE

ATHLETE BIOLOGICAL PASSPORT?

Ingrid Jense - 10378944

Content

Abstract...2

Introduction...3

Research question...4

The Athlete Biological Passport...5

The haematological module...5

The steroidal module...6

ABP: The advantages...7

ABP: The limitations and solutions...7

Dried Blood Spot Sampling...11

DBS: What can be measured?...11

DBS: The advantages...11

DBS: The limitations and solutions...12

Discussion...15

Conclusion...17

Search strategy...18

References...19

Abstract

The major objective of this study is to determine whether dried blood spot (DBS) sampling can be used to create the athlete biological passport (ABP). DBS sampling is less invasive and time consuming compared to venous sampling and therefore can lead to cost reductions and, hence, could allow for a greater number of samples to be collected. This study is necessary since the past has shown that doping misuse is still a recurrent issue in sports, indicating that more effective anti-doping systems are called for.

Data for this study were collected through the assessment of literature on DBS and ABP. First the content of the ABP is described and its advantages, limitations and the corresponding solutions. Secondly analytes are described which can be found in DBS, their advantages, limitations and the new developments in DBS sampling.

The results suggest that the ABP in the present layout cannot exclusively be based on DBS sampling. However, when new parameters are implemented as expression level of specific genes, the immature reticulocyte percentage (IRC%) and insulin-like growth factor (IGF-I) levels, DBS sampling could well be beneficial compared to the current venous blood and urine sample taking. Nevertheless, more research is needed in order to determine whether the levels of the promising biomarkers are high enough in DBS to provide results of good quality. Moreover, it should be evaluated whether developing new protocols and new analysing techniques are cost-effective and will lead to a more effective anti-doping system.

Introduction

For a long time the world of sports and the associated fight against doping has been a forensic challenge and it seems to become even more challenging every year. With the recent statements of the World Anti-Doping Agency (WADA) the fight against doping reached a new level all together. “WADA believes that the IOC has taken an informed decision to sanction Russia for its involvement in institutionalized manipulation of the doping control process before, during and after the 2014 Sochi Olympic Games,” said Sir Craig Reedie, President, of the WADA (32). This statement was made upon the decision of the International Olympic Committee (IOC) to ban Russia from the 2018 Olympic Winter Games in South Korea. WADA is an organization initiated by IOC to promote, coordinate and monitor the fight against drugs in sports (35). Last year, the fight against doping has been in the news a lot. The world became aware of this institutionalized doping program in Russia when WADA published the McLauren Investigation report last year (30). This report made clear that the Sochi laboratory was working with an unprecedented sample swapping scheme. Positive urine samples were swapped via a hole in the wall of the lab with previously gathered and preserved clean urine samples. After this revelation, involved Russian athletes were stripped down from their medals. By banning Russia from the Olympics, the IOC wants to make a statement. It has chosen to protect clean athletes, to end this destructive period for the sport and to stimulate a more effective anti-doping system.

That a more effective anti-doping system is needed is also reflected in the annually revised list of Prohibited Substances and Methods. September last year WADA published the new list that has come into force on January 1st _{2018. This list defines for all sports which methods and substances are} prohibited, both out-of- and in-competition, and which substances are banned in a particular sport (33). The fact that this list is revised annually indicates that the doping market is continually changing and still growing. Laboratories are constantly striving to incorporate new substances into existing initial testing procedures and developing new analytical procedures for detection. This makes the fight against doping a refined cat-and-mouse game that requires constant improvement of technology, especially in the case of direct detection. Direct detection of forbidden substances relies on the fact that most of these substances are different from the normal components of the human body (17). However, because of the growing list and the expanding spectrum of analytes tested, routine doping controls becomes a demanding task, especially for drugs that are not on the market yet, or designer drugs. For these new compounds no data are available about their metabolism, disposition, and elimination. To detect these substances and their metabolites research is required as well as the development of new methods. In this sense the implementation of the Athlete Biological Passport (ABP) was a big step forward. Here doping is detected in an indirect way. Through the analysis of biomarkers the effect of doping is measured rather than the doping substance itself. The biomarkers are measured and used to establish an athlete’s biological profile and in this way circumvent the limitations of direct detection.

The matrices used for the measurements of the ABP are blood and urine, both having their advantages and limitations. However, there are other possible matrices that could alter testing strategies, like testing dried blood spots (DBS), oral fluid or breath. Because these methods are less invasive and time consuming, the use of these matrices could decrease the costs and allow for a greater number of samples to be collected leading to a more efficient and effective anti-doping system.

Aiming for a more effective anti-doping system is forensically very relevant. In the past athletes have been accused of the use of doping on the basis of mistakenly interpreted research results, which is, of course a forensic nightmare. Looking only at the obtained data without knowing the context can lead to false accusations. The background and whereabouts of the athlete have to be taken into account. In addition, more frequent sample taking provides the forensic experts with a more solid basis for their expert opinion. The introduction of DBS is in this respect a contribution as it takes away constraints of venous blood and urine collection and analysis. More frequent sample taking is also beneficial in relation to doping substances or methods that have a short detection window. Abnormal fluctuations in the measured values now have a higher probability to be detected. The fact that the obstruction of doping control is still reported, indicates that new sampling methods, which are less prone to manipulation, are called for.

Research question

The aim of this thesis is to answer the question whether it is possible to develop a DBS sampling method that can be used to create the ABP and to improve the effectiveness of the fight against doping. The advantages and limitations of both the ABP and DBS sampling are explored and new developments in these fields are described. Based on the limitations and possibilities of both methods a conclusion is drawn.

The Athlete Biological Passport

The athlete biological passport (ABP) consists of a haematological and steroidal module. The former monitors the haematological parameters of athletes. Steroid profiling is done in the latter, which involves longitudinal monitoring of five urinary steroid ratios. When variations in an athlete’s biological profile are found to be incompatible with physiological or medical conditions, there is an indication of doping violation. The presence of the doping agent or the doping method used may not be directly detectable, but their effects on the haematological or steroid profile will remain for a much longer period (1).

The haematological module

One major challenge for direct detection is blood doping, in which the aim is to increase the capacity of the body to transport oxygen. To increase this capacity, blood transfusions or erythropoiesis stimulating agents (ESA), such as erythropoietin (EPO) are used (23). For homologous blood transfusion, where blood of a donor is used, flow cytometry can be used to directly detect the doping. However, for autologous blood transfusion, where stored blood of the athlete is used, there is no validated method to directly detect the doping (19). Recombinant human erythropoietin (rhEPO) is also used. The intake of this substance has an advantage for athletes as it avoids the need for blood transfusions. Since rhEPO has the same structure as endogenous EPO, detection of rhEPO is very difficult and, moreover, it is excreted in very low concentrations in the urine (23).

The haematological module is developed to detect this use of substances and methods for the enhancement of oxygen transport. Biomarkers are monitored over time and analysed using mathematical models in order to identify suspicious patterns. The haematological module consists of the following biomarkers (31):

- Haemoglobin (HGB), the mass of haemoglobin in a red blood cell. This is the oxygen-carrier

and iron-containing protein in red blood cells. As mentioned before the aim of blood doping is to increase the oxygen transport through increasing haemoglobin.

- Haematocrit (HCT), the volume percentage of red blood cells in blood (37). An increase of

HCT leads to an increase in oxygen transport.

- Red blood cell count (RBC), the number of red blood cells in the sample. Here the same

applies as for HCT and HGB when oxygen transport is increased.

- Reticulocytes percentage (RET%), the percentage of reticulocytes (RET) in the total of red

blood cells. RETs are young red blood cells. When RET% is high, it means that red blood cell production of the marrow has increased, indicating blood doping (18).

- Absolute number of reticulocytes (RET#)

- Mean corpuscular volume (MCV), the average size of red blood cells.

- Mean corpuscular haemoglobin (MCH), the average mass of haemoglobin in red blood cells. - Mean corpuscular haemoglobin concentration (MCHC), the average concentration of

haemoglobin in red blood cells.

- Red cell distribution width, standard deviation (RDW-SD).

- Immature reticulocyte fraction (IRF), fraction immature reticulocytes (IRC) of all the

reticulocytes. IRCs are the first to response to red blood cell production signals and therefore a very sensitive marker to detect blood doping (6).

Two other markers are calculated:

- OFF-hr Score (OFFS), combination of HGB and RET%. This parameter is sensitive for the ‘wash

out’ phase, the period of the cessation of rhEPO injections. During this phase lower values are measured but by calculating the OFF-score rhEPO misuse can still be detected (24).

- Abnormal Blood Profile Score (ABPS), combination of HCT, HGB, RBC, RET%, MCV, MCH, and

MCHC

The steroidal module

The use of testosterone can be detected by examining the level of selected endogenous steroid hormones in urine at a specific time. However, the most sensitive criterion applied in control, the T/E ratio, is found to be subjected to large inter-individual variability and therefore applicable to monitor over time. To improve the detection of T misuse, the ‘steroidal module’ was implemented in the ABP. Now the T/E ratio and other metabolites ratios are monitored over time and by making each individual its own reference the interindividual variability is eliminated (20).

The aim of the steroidal module is to identify endogenous anabolic-androgenic steroids (EAAS) and other anabolic agents. This module consists of the following biomarkers (31):

- Testosterone (T), hormone produced naturally in the body and is associated with

masculinization, virilization and protein building. AAS are affiliated with the hormone testosterone to promote muscle mass and strength.

- Epitestosterone (E), testosterone’s inactive epimer. This hormone produced naturally in the

body and the production is not affected by exogenous administration of testosterone.

- Testosterone/Epitestosterone ratio (T/E). Since E is not affected by administration of

exogenous T, this ratio is a good marker for T administration.

- Androsterone (A), metabolite of testosterone - Etiocholanolone (Etio), metabolite of testosterone

- 5α-androstane-3α,17β-diol (5αAdiol), metabolite of testosterone - 5β-androstane-3 α,17β-diol (5β Adiol), metabolite of testosterone

Further ratios that are analysed are:

- A/T - A/Etio

- 5αAdiol/5βAdiol - 5αAdiol/E

When the steroidal profile is assessed as atypical, the sample is reanalysed. When this gives the same result as the initial test, GC/C/IRMS (Gas Chromatograph/Carbon/Isotope Ratio Mass Spectrometry) is performed as confirmatory assay. This assay measures the differences in 13_C/12_{C ratio of} testosterone metabolites because pharmaceutical and natural testosterone have a different 13_C content. GC/C/IRMS is done because individuals can for instance produce naturally elevated T/E. Thus, the T/E ratio is used as an indicative test (25).

For the previous described markers, the intra-individual variability is lower than the interindividual variability. This is the reason why each athlete has its own reference range for the markers. These ranges are adapted to indicate a 99 or 99.9% specificity for abnormality of a given result. Each analysis result is mathematically compared to the reference values to determine whether there is a notable variation in the ABP for an athlete (See Appendix A). To determine this, Bayesian statistics are used. The profile also has to be evaluated by an expert. Together with the calculation, the expert

decides whether the findings are indicative of doping or have another origin. Based on this decision, the Athlete Passport Management Unit (APMU) can recommend specific doping tests in order to directly detect forbidden substances in athletes (18).

ABP: The advantages

As previously described, one of the major advantages of the ABP, compared to the conventional methods, is that it detects the effect of doping on the biomarkers instead of directly detecting the prohibited substances. In other words, it does not have to cope with the growing and changing production of new doping substances. There are no data available on metabolism and distribution for every new doping substance, so the possibility of targeted detection is limited. Moreover, it is possible that enzyme activity differs per person, which can lead to unexpected elimination profiles (26). The complex characterization of the metabolic products and elucidation of the biotransformation pathways is not needed for ABP. Another thing is that it is difficult but inevitable for direct detection to develop reference materials to be used in the identification and quantification of prohibited substances (2). This challenge is circumvented in the ABP since the athlete delivers his own reference values.

As mentioned before, when a suspicious profile is found, specific target tests can be performed in order to find the substance used. These resulting specific target tests show the advantage of the ABP as, apparently, this has led to a number of positive findings in recent years (18). In 2008 and 2009, ABP data caused 250% increase in the number of positive rhEPO cases based on targeted testing. Moreover the number of positive cases was still about 300% higher in 2011 than it was before the introduction of ABP (17). The ABP can be used as a tool in the fight against doping by sanctioning athletes directly on the basis of abnormalities in their profile. Besides that, the ABP can identify potential doping patterns and in this way future anti-doping tests can be planned. This identification of suspected patterns makes it possible to target suspected athletes. The ABP can also help in the prevention of doping use as it turned out that the doping use of athletes has changed since the introduction of the ABP (23).

ABP: The limitations and solutions

Endocrine module

WADA is currently working on further development of the ABP by developing an endocrine module which aims to detect the abuse of growth factors such as growth hormones (34). Growth hormones are often used in combination with anabolic steroids and insulin. This abuse can be detected by longitudinal monitoring of growth hormone dependent markers, Insulin-like growth factor (IGF-I) and type 3 procollagen (P-III-np) (22). However, precise knowledge about the intra-individual variations is still lacking today, so longitudinal studies are required to implement these biomarkers in an endocrine module of the ABP (18).

Plasma volume

In the literature on the ABP, there are some recurrent limitations. For instance, plasma volume fluctuations, which can be caused by changes in posture, exercise, training, altitude exposure, season as well as storage conditions. Some athletes try to influence volume dependent markers, such as haematocrit values and OFF score values, by plasma volume expansion leading to haemodilution or

hyper hydration (6)(27). Therefore the sensitivity of the haematological module could be improved by techniques to estimate the degree of such plasma volume variations or plasma volume independent markers (17).

According to Cox & Eichner (6) could the percentage of immature reticulocytes of the total red blood cells (IRC%), complement the haematological module since it is independent of plasma volume fluctuations. Next to this, IRCs are the first ones that response to changes in red blood cell production and therefore may increase the sensitivity of the test. In addition, a study of Lobigs et al. (11) identified some markers from a serum sample, that could, when included in the ABP model, account for around 67% of the plasma volume variance. The set of markers they used was [Hb], transferrin, creatinine, calcium, platelets, low-density lipoprotein, albumin and total protein. With this correction 66% of the haemoglobin concentration variance is explained and the number of atypical ABP findings can be reduced (11).

Altitude

Unusual environmental conditions as hypoxia of altitude, can alter the haematological profile of the athlete. This makes altitude a distorting factor for the blood markers. When experts evaluate the blood profile of an athlete, the time of altitude exposure, degree of altitude and time of sample collection compared to hypoxia, must be taken into account in order to prevent false positives. Moreover, athletes could attribute their suspicious profile to altitude (21). Thus, the way in which the altitude factor is taken into account needs to be standardized and the evaluating experts need to consider different hypoxic exposure protocols.

Instability of urine

Due to two circumstances the efficacy of the steroidal module is still in question. One is the instability of urine samples, that is influenced by several factors like bacterial contamination, medication or enzyme induction-inhibition. Another reason is polymorphism of the UGT2B17 gene, that can influence the T/E ratio. This gene encodes a protein responsible for glucuronidation (20). Individuals with a genotype causing an insensitive UGT2B17 enzyme, almost never exceed the individual thresholds. So this makes detection of testosterone misuse very difficult for people with this genotype (16).

A possible solution for this problem is suggested by Ponzetto et al (16) in a study in which endogenous steroids and metabolites are profiled in blood. In this study 14 steroid hormones were quantified in serum samples after testosterone administration. The study showed that after testosterone administration, individual threshold levels eventually were exceeded for the selected steroid hormones. This was especially the case for testosterone and dihydrotestosterone and these hormones were detectable up to 96 hours post administration. Moreover, no difference was found between different genotypes for UGT2B17, which indicates that steroid profiling in serum could complement the urinary steroid profile, especially in the case related to polymorphism. However, before implementing steroid profiling of blood samples in routine anti-doping controls, more studies are necessary. Those studies need to evaluate the presence of eventual confounding factors in blood, such as the use of masking agents or the influence of alcohol. In addition, other polymorphisms need to be evaluated (16).

‘Omics’ approach

There is much literature suggesting that the ‘omics’ approach could prove to be useful. Variations in genomics, proteomics, transcriptomics and metabolomics, reveal distinct changes after doping and, when implemented, could improve the use of the ABP. It is suggested that it could even help to distinguish alterations in the blood profile caused by altitude or by blood doping (36). This is also a long-term goal of the WADA, as it wants to make use of advances in the analytical chemistry. It wants to develop a large panel of biomarkers of doping and a better understanding of the ‘omics’ field (34). Omics’ is defined by Horgan et al. (9): “Omic technologies are primarily aimed at the universal detection of genes (genomics), mRNA (transcriptomics), proteins (proteomics) and metabolites (metabolomics) in a specific biological sample.”

Doping substances or methods have been recognized to influence mRNA expression. This can occur because the transcriptome is subject to environmental variations (19). Transcriptome is defined by Horgan et al. (9): “The transcriptome is the total mRNA in a cell or organism and the template for protein synthesis in a process called translation. The transcriptome reflects the genes that are actively expressed at any given moment.” It is measured by a microarray technology that can assess the transcription state of the cells and in this way characterise the gene expression signature of that particular stimulus (9).

In a study funded by WADA in which it was tried to determine the ‘molecular signature’ of rhEPO, it was found that many gene transcripts were differently expressed during rhEPO administration up to 4 weeks after administering the doping (14). The challenge of using transcriptomics is that the gene response to doping must be significantly different from the response to allowed training regimes and pathological conditions in order to ascertain the use of doping (2).

Since some athletes claim that the altitude exposure explains their variation in blood markers, there is a need to develop a method which can distinguish altitude training from blood doping. For this purpose a study was designed in which 13 previously identified genes showed changes in expression after rhEPO administration, but showed no changes after altitude training. This suggests that transcriptomics could help distinguishing altitude and rhEPO (36).

Transcriptomics could also prove to be useful relating to another limitation recurrently mentioned in literature, microdosing. There are athletes that take micro doses of rhEPO to avoid abnormal fluctuations in the blood parameters. In this way they reduce the sensitivity of the ABP detection and minimize their risk of being caught. Interestingly, there are studies that show 5 genes that remained differentially expressed after micro dosing of rhEPO. These data suggest that the use of transcriptomics may provide a sufficiently sensitive method to detect micro doses of rhEPO (36). Next to gene transcripts, the ability of circulating microRNAs (miRNAs) to serve as biomarkers of the ABP is also a recurrent topic in the literature about the ABP .The transcriptome also consists of miRNAs small non-coding RNAs, and they play a crucial role in gene expression regulation. Cell-free miRNA is stable and detectable in blood plasma or serum and can be used as specific and sensitive marker of various pathophysiological processes (19). miRNA has the potential to serve as biomarker for detection of EPO, recombinant human growth hormone or autologous blood transfusion. The levels of circulating miRNA are less affected by environmental factors like inadequate storage of the blood samples. However, it is also mentioned that little is known about the effects of short- and long-term exercise or altitude training on the miRNA levels in blood (10).

It appears that transcriptomics could also prove to be useful to detect testosterone misuse by blood testing instead of urine. In a study of Salamin et al. (20) a potential new biomarker is suggested for the detection of testosterone abuse. In this study the effect of testosterone administration on a panel

circulating miRNAs was examined. They discovered that administering testosterone triggers an increase of the miRNA-122 level in comparison to a control period without treatment of testosterone. This indicates that this could be a potential biomarker to be included in the steroidal module. Another advantage of using miRNA-122, next to the independency of UGTB17 polymorphism, is the independency of blood-cell count. However, miRNA-122 should be combined with other biomarkers since this single biomarker is not powerful enough on its own (20). These results support the idea of a transcriptomic based approach for both modules of the ABP in the future.

Iron metabolism

Next to transcriptomics, markers of iron metabolism have been studied extensively. The fact that blood donation and blood reinfusion alter iron metabolism, indicate that quantification of proteins involved may improve the ABP (19).

Limitations Solutions

Growth hormones not detectable Endocrine module

Volume dependent markers New parameter IRC%, set of specific markers

Microdosing Gene expression Traning at altitude Gene expression

Instability of urine markers Steroid profiling in blood, miRNAs

Dried Blood Spot Sampling

DBS:

What can be measured?

Dried blood spot (DBS) sampling is used since the 1960s for testing new-borns with a simple heel prick. Capillary blood can also be obtained from a finger prick. Today DBS sampling is used for more applications such as therapeutic drug monitoring, pharmacokinetic studies in drug development, toxicology, detecting drugs of abuse, and viral disease management (5). With respect to doping, several studies showed that different prohibited substances can be detected directly by using the DBS sampling method, such as the small molecule anabolic agents, beta-2 agonists, beta-blockers, diuretics, stimulants, and cannabinoids (28). It is also demonstrated that DBS can be used to detect larger molecules like peptides and proteins (5) (13) (26). An important biomarker for the detection of growth hormone abuse, IGF-I, can also be identified in DBS (6) (26). All the steroid hormones should be measurable in blood. However, since DBS sampling uses a very small volume of blood, it could be that the level of some steroid hormones is not measurable in DBS samples. T he steroid hormones that can be determined in DBS according to literature, are corticosterone, deoxycorticosterone, progesterone, 17-hydroxyprogesterone, 11-deoxycortisol, 21-deoxycortisol, androstenedione, testosterone, dihydrotestosterone and cortisol. However even for some of these steroids the use of DBS sampling methods is not distinctive enough (29).

In addition, important markers of the ABP can be traced using DBS, such as haemoglobin and haematocrit (4) (6). And recently a promising method was developed to measure other markers of the ABP. In a study of Cox et al. (6) a DBS method was developed which is able to measure immature reticulocyte cells (IRC) and red blood cells (RBC). This is done by the quantification of cell-specific cluster of differentiation (CD) proteins. CD proteins are cell surface proteins and include protein markers that are used to characterize cell types and disease states in blood. IRCs are counted by the measurement of CD71 and RBCs are counted by the measurement of CD233 (6).

The literature about mRNA and miRNA in DBS is limited. A study of Ponnusamy et al. (15) showed that it is possible to extract a sufficient quantity of good quality miRNAs from a single DBS. And a study of Matsubara et al. (12) showed that mRNA is present in DBS and can be used for further analysis.

DBS: The advantages

Comparison to venous blood sampling

Often mentioned advantages of DBS sampling compared to venous blood sampling are: less invasiveness, fast sample collection, easier storage and transport, and enhanced stability of target analytes leading to the cost and time effectiveness. In addition, unannounced tests out of competition can be executed more easily since the presence of a physician or nurse is not needed and it takes less time. As such, doping controls can be much more effective since the testing intervals can be shorter (13). And the more frequent athletes can be tested, the lower the required retrospectivity is for the DBS method. This makes sense since retrospectivity is very important for anti-doping organizations when new methods are developed (26). Obviously it is important that new methods cover the largest possible part of the spectrum of prohibited substances, so more research can be rewarding.

Comparison to urine sampling

In comparison with urine, using blood as a matrix has several advantages. One of them is the limited detection window in urine. It could also be that relevant levels for a doping violation are not even reached in urine. This was shown in a study of Thomas et al. (28) in which they measured concentrations in urine and blood after oral administration of a prohibited dose of pseudoephedrine. The maximum value in blood was reached after 2-4h whereas in urine it was reached after 7-9h. In addition, the relevant level for doping violations in urine was not reached. This indicates that urine testing might miss the maximum value that was present in the circulation during competition. This limitation of urine testing is especially relevant for measurements of steroids. However the same limitation could apply in blood testing, where it is also possible that that relevant analytes are not secreted in blood and therefore not detected (26). For the athlete blood sampling can be advantageous since the urine sample collection sometimes takes several hours when spontaneous urination is not possible. In addition, some athletes have difficulties when they have to urinate under supervision. This also can lead to a scenario where an athlete has to wait hours in order to get enough urine in the sample cup. From this perspective is DBS sampling even more advantageous because the blood is obtained in a minimally invasive way.

This urination under supervision is needed to prevent manipulation of the urine sample. The presence of DNA in blood and in the DBS, makes it possible to identify the donor when this is called for. It could be argued that this is also possible in urine, but the storage temperature and duration were found to be critical parameters for success in this DNA analysis (26). Another advantage of blood compared to urine concerning analytical issues, is related to bacterial contamination or polymorphism as described previously.

DBS sampling compared to urine DBS sampling compared to blood Less invasive Less invasive

Fast Fast

Easier to store and transport Easier to store and transport No supervised urination No physician or nurse needed  Athletes can be tested more frequently

Table 2 The summarized benefits of DBS sampling compared with urine and venous blood sampling

DBS: The limitations and solutions

Sample handling

Against these advantages of DBS, there are limitations to be mentioned. For instance, the sample volume being small also brings a disadvantage because the analytical methods that are used, consequently have to be very sensitive (13). Although it is more easy to obtain the DBS sample, the sample handling may be more demanding. This is because the DBS has to be cut out and the analytes have to be extracted followed by centrifugation, aliquot transfer, drying and dissolution. Automation of this elaborate process has its limitations. The requirements for automation can be quite expensive. Moreover the used robotic sample handler has to be able to adapt to different forms, since a dried blood spot is not perfectly round shaped or the volume of the spot differs. This automatic adaptation could be a challenge when automatization of sample handling is implemented (37).

Haematocrit effect

Another, recurrent, limitation of DBS described in literature is the haematocrit effect. Haematocrit is the percentage of red blood cells in blood. The amount of red blood cells have an effect on the viscosity of blood causing uneven distribution of analytes within the spot on the DBS paper. As such, haematocrit can have an influence on the accuracy and precision of the results of DBS testing (37). When the haematocrit level is increased, the surface of the spot becomes smaller. As such, the analyte concentration in the spot becomes higher. When the haematocrit level is low, the surface of the spot increases causing a lower concentration of the analyte in the spot (38).

One way to avoid the haematocrit effect is by analysing the whole DBS. Youhnovski et al. (38) developed a method for this purpose, a pre-cut dried blood spot (PCDBS). This PCDBS shows no variation when blood with different haematocrit value is applied. Moreover, due to the fact that the spot doesn’t have to be cut out anymore, sample carryover from the mechanical punching of conventional DBS is now avoided (38). However, analysing the whole spot results in another critical parameter: the accurate volumetric application, because a fixed volume needs to be deposited on the card. This problem can be solved by accurate sampling of blood volumes prior to spotting, or by the technique of volumetric absorptive micro sampling (27).

Another approach suggested in literature, is the measurement of haematocrit by means of a marker (37). For this purpose potassium is used. Potassium corresponds with the red blood cell fraction since it is primarily located intracellularly. Furthermore, it is a stable component of which the levels are under physiological control. Multiple studies showed a good correlation between the potassium concentration and the haematocrit. A drawback of this study is that the tests were not performed with capillary blood and it is a destructive method (3). Recently a new non-destructive method is developed to determine the haematocrit in DBS. In this case reflectance spectroscopy is used. By using the reflectance at a single wavelength the total haemoglobin content is measured. This study showed that this method was able to measure the caffeine concentration in blood by DBS sampling, after applying a haematocrit dependent correction factor (figure 1) (4).

The spreading of the blood is not only dependant on the amount of red blood cells but also on the chemical and physical properties of the analyte in combination with the DBS paper (37).

Figure 1 Plotted are the % difference of caffeine concentrations measured in DBS and whole blood. The black spots indicating the values before Hct correction and the white spots indicating the values

after Hct correction. The solid line represents the 0% difference and the dotted line represents the 15% difference (4).

Insulin-like growth factor

As mentioned before, the possibility to determine IGF-1 in DBS is an advantage of DBS sampling. In a study of Cox et al. (8) a method was developed to quantify IGF-1 in DBS. They also looked at the stability of the samples at room temperature. They remained stable for 8 days. Furthermore there was a good correlation between the concentration of IGF-1 measured in DBS and measured after venous sampling. Although the sample recovery was very low, it was still sufficient to achieve the lower limits of quantification (LLOQ), the lowest standard curve points that can be used for quantification, in order to measure IGF-1 in any athlete. This observation suggests that the low sample recovery could be a limitation of protein detection in DBS. Because proteins are large molecules, they tend to stick to the surface of the sample paper. The haematocrit effect on the IGF-1 measurements was also studied. Here they only noticed an effect of haematocrit levels higher than 60%. This was not considered to be a problem since normal haematocrit levels are between 30 and 50% (8).

New ABP parameter

In the study of Cox et al. (6) they looked at the longitudinal stability of the concentration of CD71, the measure for IRCs, in DBS. Over an 8-week period an intra individual variation was measured between 17.1-38.7%. This suggested that CD71 could be a new biomarker for the indication of blood doping. A follow up study confirmed this suggestion (7). In order to count IRC and RBC to calculate IRC% as mentioned previously, CD71 and CD233 were measured. This was done in an autologous blood transfusion study in which 15 subjects received blood and 11 subjects received saline. The CD71/CD233 ratio, representing IRC%, was measured after 5, 6, 13 and 20 days and showed a statistically significant difference between the blood and saline group. Moreover, compared with Ret %, the CD71/CD233 ratio showed a larger decrease from the baseline. In addition, they also compared the number of subjects that was tested positive for autologous blood transfusion using Ret %, CD71/CD233 ratio or the OFF-score. Remarkably, using the OFF-score only 1 subject was tested positive, using the Ret% 3 subjects were tested positive and using the CD71/CD233 ratio 7 subjects were tested positive. So the DBS method could be very promising in order to improve the sensitivity of ABP for testing autologous blood transfusion (7).

Limitations Solutions

Increased sample handling Automatization

Hematocrit effect Potassium marker, reflectance spectroscopy

Small samples Sensitive analysing methods Not all current ABP biomarkers are

detectable in DBS

IGF-1 and new parameter IRC% can be determined in DBS

Table 3 Summarized limitations and solutions of DBS sampling

-Discussion

The fight against doping will benefit from the ABP when implemented in an efficient way. Depending on the branch of sport and how the competitions are organised throughout the year, the approach may vary. Some sports only have one major event per year and the rest of the year the athlete is preparing for this one event, while other sports have competitions spread over a very long season. An easier way of sample taking could help here. Especially in the case of blood sampling. When the sites of collection are located a large distance from the nearest antidoping lab, the labile blood samples must be shipped under very specific conditions and within a specific time frame. As described before, these are all limitations of venous blood sampling. In addition, the scheduling of a physician or nurse could also hinder the unannounced out-of-competition testing, and venous blood collection is invasive. As such the question arises whether DBS sampling could improve the use of antidoping tests that require blood.

In order to answer this question the principles of ABP and DBS as well as the advantages, limitations and new developments of both techniques have been looked into. The recent ABP consists of the haematological and steroidal module, but the WADA wants to develop a third endocrine module. The aim of this module in development is to detect growth hormone abuse by monitoring IGF-I and P-III-NP. Before the endocrine module is implanted in the ABP longitudinal studies of IGF-I and P-III-NP should be done. DBS could be used for the endocrine module since there is already a method developed that monitors IGF-I in DBS. However, it is mentioned that the recovery rates are very low. When implementing DBS sampling for endocrine module it is important that the recovery rate is consistent. In order to make the detection of growth hormone abuse by DBS sampling more sensitive, a method should be developed to measure P-III-NP.

In case of the steroidal module, the possibility of sampling blood instead of urine first has to be evaluated. Detection of testosterone through the urinary steroid profile faces several obstacles. Some are polymorphism and contaminations which were found not to be or less presented in blood. In addition, the less invasiveness and shorter duration of the testing procedure are also in favour of DBS. However, in order to use DBS sampling to create the steroidal module it must be possible to detect steroids in DBS. Among the current biomarkers of the steroidal module, only the measurement of testosterone in DBS has been mentioned in literature. In order to use DBS for the current steroidal, the presence of the other markers in DBS needs to be established in future research and other sorts of polymorphisms needs to be evaluated. On the other hand, there are other potential biomarkers in blood to detect steroid misuse described in literature. One of them is miRNA-122 which level is altered after testosterone administration and is not dependent of UGTB17 polymorphism and blood cell count. miRNAs are present in DBS, but it is important to know whether this specific miRNA can be determined in DBS.

On the other hand, the use of DBS for the haematological module should be easier to implement since the used biomarkers are already measured in blood. However, it is possible that current markers are not stable in DBS. According to literature it is possible to measure haemoglobin and haematocrit in DBS and indirectly immature reticulocytes and red blood cells. With these markers the OFF-score can be calculated. Nevertheless, there are many limitations involved in the current markers that could be circumvented by introducing new markers. Doping substances or methods have been recognized to influence mRNA expression. Studies to a transcriptomic approach showed that there are genes that remained differently expressed after micro dosing, which is a considerable limitation of the current haematological module. Circulating miRNA has also the potential to serve as biomarker for

this module, like for the steroidal. However the influence of altitude on miRNA levels need to be evaluated. It can be concluded that, it is important to investigate whether these specific markers can be determined in DBS . As such further research regarding the transcriptomics approach could reveal the promising prospects.

It could also be interesting to investigate the use of CD proteins for the ABP further as they are measurable in DBS. Especially regarding the effect of growth hormone and steroid abuse on different CD proteins. When all modules of the ABP have CD proteins as biomarkers, the DBS sampling method to investigate CD proteins could be easier implemented in the ABP. Moreover this would simplify the analysis of the spots, as only CD proteins have to be measured, and make the fight against doping more efficient.

The use of DBS testing for doping controls has been extensively studied. However, there are still some practical aspects to be considered. The existing methods and protocols nowadays used for blood analysis are probably not capable to accommodate DBS analysis. A whole new protocol needs to be established. This is not impossible of course but it should be evident that all the markers are detectable in DBS to begin with. Furthermore, some markers of the ABP are plasma volume dependent, so when new biomarkers are developed they should be independent of the plasma volume.

In addition, one could argue that the concentrations of biomarkers or blood values differ when measured after venous sampling or after capillary sampling. However, this should not be a problem when these measurements are used for the ABP since the ABP makes use of intra-individual variation of these values. When switching from urine to blood, one could point out that there is a possibility that relevant analytes are not present in blood and therefore not detectable using DBS. Again, this should not be a problem when DBS sampling is used for creating the ABP, since indirect detection is used. In studies describing the use of DBS for different purposes, different DBS cards were used. It is possible that for each card there is a different way to extract the analytes. So, when implementing DBS sampling, it is important to take into account the different techniques that are needed to analyse the various biomarkers. Hence the different analysing methods of each biomarker need to be studied to see whether there is an overlap. In order to make the analysis efficient for creating the ABP, it would be best when only a couple of analysing techniques are used. Also the different characteristics of the used DBS papers need to be looked into.

More research into the use of DBS for creating the ABP could be rewarding. When both modules require blood samples, the collection process is simplified, and no additional sample material needs to be taken from the athlete. However, the current ABP is not able to replace traditional doping control. According to Wada the fight against doping relies on several strategies, not only on the ABP. For now, the fight against doping is more effective when both strategies are combined. However, when new substances or modifications of prohibited substances are developed, it is difficult to directly detect the misuse. Hence the ABP is used as an indicative method to complement the conventional testing approach. Maybe in the future when the ABP is enhanced, the ABP on its own will be effective enough, which certainly would be more efficient.

Conclusion

There is a lot of research on the development of the ABP and on the use of DBS for doping testing. Nevertheless, literature describing the use of DBS sampling for creating the ABP is rather limited. Compared with direct detection the indirect detection of doping has the benefit of not having to cope with the growing list of different substances used. One of the goals of WADA is developing a endocrine module. For this module it would be possible to use DBS sampling for IGF-I measurements. For the other marker P-III-NP a DBS sampling methods still needs to be developed. An limitation of the current used parameters of the haematological module are the dependency on plasma volume or hypoxia of altitude, although there are some studies describing possible new parameters that are independent of these aspects. The ‘omics’ approach has produced some promising markers some of which are also detectable in DBS. When DBS sampling is used to create the ABP, steroid profiling in blood needs to be possible. There are several advantages of steroid profiling in blood compared to testing urine. However, it is not known yet whether all steroids can be detected in DBS. Compared to the traditional venous blood collection, DBS sampling has many benefits including cost and time effectiveness and less invasiveness. Because DBS samples can be taken more frequently, the ABP can be created in a more efficient way thereby making it a useful matrix for anti-doping testing. In conclusion, the development of DBS sampling as a technique to sample blood is challenging as well as promising. The development will take time since most of the variables need to be evaluated, such as the haematocrit, blood volume and the chromatographic effect. Furthermore, the interferences origination from the filter paper that is used, can complicate the development. As a conclusion: DBS sampling cannot be used for the ABP as it is now because not all current biomarkers are detectable after DBS sampling. More research is needed in order to use DBS sampling for creating the ABP. However there are already some current markers detectable in DBS (table 4). Moreover, when new promising markers are implemented in the modules, DBS sampling could provide a more efficient creation of the ABP and therefore a more effective approach of the fight against doping.

Parameters of current ABP Possible new parameters for ABP

Testosterone IRC%

Haemoglobin mRNA

Haematocrit miRNA

Immature reticulocytes (indirect) IGF-I Red blood cells (indirect) CD proteins

Search strategy

Search 1 pubmed

The used search terms: athlete biological passport[Title/Abstract] AND blood AND (biomarker OR hematology OR hematological OR steroid)

This resulted in 40 articles that were scanned by abstract and title. After this 22 articles were saved and fully read in order to decide whether they could be used for the thesis.

Search 2 google scholar

The used search terms: "athlete biological passport" AND biomarker AND hematological AND steroid This resulted in 135 articles. To make the search smaller the word ‘omics’ was added. This was a recurrent term in the previous found articles. This resulted in 40 articles that were scanned by abstract and title. After this 10 articles were saved of which 3 overlapped with previous found articles. The resulting articles were saved and fully read in order to decide whether they could be used for the thesis.

Search 3 UvA library

The used search terms: subject contains blood and title contains biological passport.

This resulted in 22 articles that were scanned by abstract and title after this 3 articles were saved and fully read in order to decide whether they could be used for the thesis.

Search 4 pubmed

The used search terms: (dried blood[Title/Abstract]) AND doping

This resulted in 18 articles that were scanned by abstract and title. After this 6 articles were saved and fully read in order to decide whether they could be used for the thesis.

Search 5 pubmed

The used search terms: (dried blood spot[Title/Abstract]) AND steroid

This resulted in 51 articles that were scanned by abstract and title. After this 2 articles were saved and fully read in order to decide whether they could be used for the thesis.

Search 6 pubmed

The used search terms: (dried blood spot[Title/Abstract]) AND hematology

This resulted in 23 articles that were scanned by abstract and title. After this 2 articles were saved and fully read in order to decide whether they could be used for the thesis.

Search 7 web of science

Web of science is used because there can be specifically searched for reviews. The used search terms: (TI=(dried blood spot)) AND LANGUAGE: (English) AND DOCUMENT TYPES: (Review)

This resulted in 39 articles that were scanned by abstract and title. After this 6 articles were saved and fully read in order to decide whether they could be used for the thesis.

Other articles are found through the references of selected articles. Some articles are only used for definitions.

References

1.

2. Bowers, L., & Bigard, X. (2017). Achievements and Challenges in Anti-Doping Research. Medicine and Sport Science, 77-90.

3. Capiau, S., Stove, V., Lambert, W., & Stove, C. (2012). Prediction of the Hematocrit of Dried Blood Spots via Potassium Measurement on a Routine Clinical Chemistry Analyzer. Anatytical Chemistry, 404-410.

4. Capiau, S., Wilk, L., de Kesel, P., Aalders, M., & Stove, C. (2018). Correction for the Hematocrit Bias in Dried Blood Spot Analysis Using a Non-destructive, Single-wavelength Reflectance-based Hematocrit Prediction Method. Analytical Chemistry, 1-18.

5. Chambers, A. G., Percy, A. J., Yang, J., & Borchers, C. H. (2015). Multiple Reaction Monitoring Enables Precise Quantification of 97 Proteins in Dried Blood Spots. Molecular and Cellular Proteomics, 3094-3104.

6. Cox, H. D., & Eichner, D. (2017). Mass Spectrometry Method to Measure Membrane Proteins in Dried Blood Spots for the Detection of Blood Doping Practices in Sport. Analytical

Chemistry.

7. Cox, H. D., Miller, G. D., Lai, A., Cushman, D., & Eichner, D. (2017). Detectin of Augologous Blood Transfusions using a Novel Dried Blood Spot Method. Drug Testing and Analysis. 8. Cox, H. D., Rampton, J., & Eichner, D. (2013). Quantification of insulin-like growth factor-1 in

dried blood spots for detection of growth hormone abuse in sport. Analytical and Bioanalytical Chemistry, 1949-1958.

9. Horgan, R. P., & Kenny, L. C. (2011). 'Omic' technologies: genomics, transcriptomics, proteomics, and metabolomics. The Obstetrician & Gyneacologist, 189-195.

10. Leuenberger, N., & Saugy, M. (2015). Circulating microRNAs: The Future of Biomarkers in Anti-doping Field. microRNA:Medical Evidence, 401-408.

11. Lobigs, L. M., Sottas, P.-E., Bourdon, P. C., Nikolovski, Z., El-Gingo, M., Varamenti, E., . . . Schumacher, Y. O. (2017). A step towards removing plasma volume variance from the Athlete's Biological Passport: The use of biomarkers to describe vascular volumes from a simple blood test. Drug Testing and Analysis, 1-7.

12. Matsubara, Y., Ikeda, H., Endo, H., & Narisawa, K. (1998). Dried blood spot on filter paper as a source of mRNA. Nucleid Acids Research.

13. Möller, I., Thomas, A., Geyer, H., Schänzer, W., & Thevis, M. (2012). Development and

validation of a mass spectrometric detection method of peginesatide in dried blood spots for drug testing. Analytical and Bioanalytical Chemistry, 2715-2724.

14. Pitsiladis, Y. P., Durussel, J., & Rabin, O. (2014). An integrative 'Omics' solution to the detection of recombinant human erythropoietin and blood doping. British Journal of Sports medicine, 1-6.

15. Ponnusamy, V., Kapellou, O., Yip, E., Evanson, J., Wong, L.-F., Michael-Titus, A., . . . Shah, D. K. (2016). A study of microRNAs from dried blood spots in newborns after perinatal asphyxia: a simple and feasible biosampling method. Nature, 799-805.

16. Ponzetto, F., Mehl, F., Boccard, J., Baume, N., Rudaz, S., Saugy, M., & Nicoli, R. (2016).

Longitudinal monitoring of endogenous steroids in human serum by UHPLC-MS/MS as a tool to detect testosterone abuse in sports. Analytical and Bioanalytical Chemistry, 705-719. 17. Pottgiesser, T., & Schumacher, Y. O. (2013). Current strategies of blood doping detection.

Analytical and Bioanalytical Chemistry, 9625-9639.

18. Robinson, N., Sottas, P.-E., & Schumacher, Y. O. (2017). The Athlete Biologcial Passport: How to personalize Anti-Doping Testing across an Athlete's Career? Medicine and Sport Science, 107-118.

19. Salamin, O., De Angelis, S., Tissot, J.-D., Saugy, M., & Leuenberger, N. (2016). Autologous Blood Transfusion in Sports: Emerging Biomarkers. Transfusion Medicine Reviews, 109-115. 20. Salamin, O., Jaggi, L., Baume, N., Robinson, N., Saugy, M., & Leuenberger, N. (2016).

Circulating microRNA-122 as Potential Biomarker for Detection of Testosterone Abuse. PLoS ONE, 1-13.

21. Sanchis-Gomar, F., Pareja-Galaeano, H., Brioche, T., Martinez-Bello, V., & Lippi, G. (2014). Altitude exposure in sports: the Athlete Biolgoical Passport standpoint. Drug Testing and Analysis, 190-103.

22. Saugy, M., Lundby, C., & Robinson, N. (2014). Monitoring of biolgical markers indicative of doping: the athlete biolocial passport. British Journal of Sports Medicine, 1-8.

23. Schumacher, Y. O., Saugy, M., Pottgiesser, T., & Robinson, N. (2012). Detection of EPO doping and blood doping: the heamatological module of the Athlete Biological Passport. Drug Testing and Analysis, 846-853.

24. Sottas, P. E., Robinson, N., Giraud, S., Taroni, F., Kamber, M., Mangin, P., & Saugy, M. (2006). Statistical Classification of Abnormal Blood Profiles in Athletes. The International Journal of Biostatistics.

25. Sottas, P. E., Saugy, M., & Saudan, C. (2010). Endogenous steroid profiling in the athlete biological passport. Endocrinology and metabolism clinics of North America, 59-73. 26. Thevis, M., Geyer, H., Tretzel, L., & Schänzer, W. (2016). Sports drug testing using

complementary matrices: Advantages and limitations. Journal of Pharmaceutical and Biomedical Analysis, 220-230.

27. Thevis, M., Kuuranne, T., Geyer, H., & Schänzer, W. (2016). Annual banned-substance review: analtical approaches in human sports drug testing. Drug Testing and Analysis, 6-29.

28. Thomas, A., Geyer, H., Guddat, S., Schänzer, W., & Thevis , M. (2011). Dried blood spots (DBS) for doping control analysis. Drug Testing and Analysis, 806-813.

29. Velghe, S., Capiau, S., & Stove, C. P. (2016). Opening the toolbox of alternative samplin strategies in clinical routine: A key-role for (LC-)MS/MS. Trends in Analytical Chemistry, 61-73.

30. WADA. (2016, July 2016). Doping control process. Retrieved from World Anti-Doping Agency: https://www.wada-ama.org/en/resources/doping-control-process/mclaren-independent-investigation-report-part-i

31. WADA. (2017, January 1). Athlete Biological Passport. Retrieved from Wolrd Anti-Doping Agency: https://www.wada-ama.org/en/resources/athlete-biological-passport/athlete-biological-passport-abp-operating-guidelines

32. WADA. (2017, December 5). News. Retrieved from World Anti-Doping Agency:

https://www.wada-ama.org/en/media/news/2017-12/wada-statement-regarding-the-iocs-decision-concerning-russia

33. WADA. (2017, September 29). News. Retrieved from World Anti-Doping Agency: https://www.wada-ama.org/en/media/news/2017-09/wada-publishes-2018-list-of-prohibited-substances-and-methods

34. WADA. (2018). Athlete Biological Passport . Retrieved from World Anti-Doping Agency: https://www.wada-ama.org/en/questions-answers/athlete-biological-passport

35. WADA. (2018). Who we are. Retrieved from World Anti-Doping Agency: https://www.wada-ama.org/en/who-we-are

36. Wang, G., Karanikolou, A., Verdouka, I., Friedmann, T., & Pitsiladis, Y. (2017). Next Generation "Omics" Appoaches in the "Fight" against Blood Doping. Medicine and Sport Science, 119-128.

37. Wilhelm, A. J., den Burger, J. C., & Swart, E. L. (2014). Therapeutic Drug Monitoring by Dried Blood Spot: Progress to Date and Future Directions. Clinical Pharmacokinetics, 961-971. 38. Youhnovski, N., Bergeron, A., Furtado, M., & Garofolo, F. (2011). Pre-cut dried blood spot

(PCDBS): an alternative to dried blood spot (DBS) technique to overcome hematocrit impact. Rapid Communications in Mass Spectrometry, 2951-2958.

Appendix

These are illustrations of an individual’s ABP of the parameters [Hb], OFF-score and ret%. The red line shows the individual’s threshold limit values. The blue lines shows the scores of the measured values of the parameters over time. (a) shows normal values, (b) shows values of a doped individua