Recombinant human erythropoietin (rHuEpo), has been commercially available since 1988, and it can be used by athletes in aerobic sports to increase Hb concentration, oxygen transport, aerobic power, and performance (4,13,18). Like endogenous erythropoietin (Epo), rHuEpo exerts a specific receptor-mediated effect on committed erythroid stem cells, inducing these target cells to proliferate and develop into mature erythrocytes (17,21). Since 1990, rHuEpo has been officially included on the International Olympic Committee (IOC) list of products considered to be illegal drugs. However, only one analytical technique is yet available to detect its misuse by athletes (29). Several factors thwart the measurement of circulating rHuEpo as an indicator of erythropoietin misuse, including its short plasma half-life, the close to perfect homology between endogenous and recombinant Epo, and the late clinical manifestation of its effect. Indeed, the erythropoietin-induced effects persist at significant values several days after the last intake of the hormone, when the erythropoietin blood and urine concentrations are in the range of normal values (16,18). To prevent rHuEpo abuse, the International Cycling Union (IUC) has established a Hct threshold of 50% and the International Skiing Federation (FIS) has established a Hb threshold of 185 g·L−1.
Recently, Gareau et al. (14,15) showed that rHuEpo induces a delayed increase in serum soluble transferrin receptors (sTfr) and a delayed decrease in ferritin (fr) concentration. These authors expressed sTfr in relationship to fr, thus giving a serum sTfr/fr index. They concluded that erythropoietin doping in athletes could be investigated using an algorithm, where Hct, sTfr, and fr could serve when dealing with blood samples. In a recent report, Bressolle et al. (7) used a population pharmacodynamic approach to relate serum erythropoietin concentrations to the effect of rHuEpo on sTfr, fr, and the sTfr/fr ratio; the concentration-effect relationship was best described using the sigmoid Emax model (7). From the results, sTfr seemed to be the most effective marker. A method based on the measurement of these markers from blood samples was thus proposed to determine whether the observed values were related to rHuEpo, administration.
Prolonged exercise is often associated with hemoconcentration; moreover, iron supplementation is frequent among athletes, which makes fr an unsuitable marker. Thus, to take into account a possible hemoconcentration, we decided to correct the sTfr concentrations by the serum protein concentrations by computing the (sTfr × 103)/serum protein ratio.
In contrast to a previous study (26), this study reproduced a situation of rHuEpo abuse in nine healthy volunteer athletes. RHuEpo was administered subcutaneously (50 IU·kg−1·d−1) for 26 d. Because of its action on bone marrow, our primary objective was to evaluate the effects of daily rHuEpo administrations on hematological parameters to quantify the erythropoietic response. Moreover, because it is unknown whether highly trained athletes can improve their maximum oxygen uptake (O2)max and ventilatory threshold (VT) with rHuEpo, we examined the changes in the physiological test results of these athletes after treatment; throughout the study period, the athletes followed their usual training program. The secondary objective of this study was to evaluate the appropriateness of a statistical method to determine whether the observed values of the sTfR/serum protein ratio are related to rHuEpo intake.
Subjects. Nine trained healthy volunteers (2 women and 7 men) participated in this study after giving informed written consent. The study protocol was in accordance with legal requirements and the Declaration of Helsinki, and it was approved by the Regional Ethics Committee. The subjects were (mean ± SD) 24.3 ± 3.1 yr and weighed 72.8 ± 7.5 kg. All subjects had been involved in regular training for several years, and during the study period they were allowed to continue their normal physical activity (training of 15-25 h·wk−1). None of them took part in an official competition during this period. Each subject underwent clinical and physical examinations one week before admission to the study. The inclusion criteria were age (from 18 to 30 yr) and hematological and iron status within the normal range. There is also no deficiency in folic acid or vitamin B12.
To determine the reference mean baseline values of the (sTfR × 103)/serum protein ratio in the largest possible population of subjects, 224 subjects were included in an additional group: 176 athletes (146 men and 30 women) participating in the French mountain-biking championship, 15 internationally ranked (wrestlers) after 8 d of training at 1800-2000 m, and 33 subjects living at or above 3000 m. These subjects were enrolled in the study according to the same inclusion and exclusion criteria as the nine subjects.
Drug administration. Recombinant human Epo alpha (Eprex®, Issy-les-Moulinaux, France) was supplied by Cilag AG, France. The drug was provided as a sterile buffered solution in a 1-mL ampoule containing an activity of 10,000 units·mL−1, which corresponds approximately to 8.4 ng rHuEpo·mL−1. The drug was stored at 4°C and was protected from light exposure. Before administration of rHuEpo, the solution was equilibrated to room temperature.
Each subject received a daily subcutaneous injection (50 IU·kg−1) of rHuEpo for 26 d. The treatment was stopped if the Hct level rose above 50%. The injections were administered in the morning, always at the same time for each subject. To assess tolerance, all subjects were instructed to report any abnormal events.
At the time of rHuEpo injection, each athlete also received an oral dose of 200 mg of iron sulfate (Ferograd®, Abbott Laboratories, Rungis, France).
Sampling. To determine erythropoietin concentrations and biochemical parameters (sTfr and serum proteins), blood samples (5 mL) were drawn from a forearm vein before drug administration (day 0), during treatment on days 10, 14, 17, 21, and 24 and posttreatment on days 1, 3, 7, 14, and 25. To exclude any possible diurnal influences on the determined erythropoietin concentrations (19,28), all specimens were drawn in the morning, at the same time, for each subject. The samples were allowed to clot at room temperature and serum was immediately separated by centrifugation (900 × g at 4°C for 15 min). The serum was divided into aliquots and transferred to polypropylene tubes. The tubes were labeled, capped, and rapidly deep-frozen and stored at −20°C until assayed. The stability of erythropoietin handled in this manner has been demonstrated (1,12).
Hematological parameters (reticulocytes, Hb and Hct) were assessed at the same sampling times. Blood samples were drawn into EDTA-coated tubes (2 mL). Analyses were then immediately performed using an automated cytoanalyzer and standard laboratory procedures.
Sample analysis. Serum samples were analyzed for erythropoietin concentrations by using an immunoradiometric assay based on a sandwich technique using two monoclonal antierythropoietin antibodies (mouse) raised against erythropoietin (125I-EPO COATRIA® kit, BioMérieux, Lyon, France). All assays were carried out in duplicate. The detection limit was 0.8 mIU·mL−1. The interassay and intraassay coefficients of variation were 2.6 and 4.2%, respectively. The assay cannot distinguish between endogenous Epo and rHuEpo (26).
The ELISA technique (Ramco Laboratories, Inc., Houston, TX) was used to assay sTfr serum concentrations (7,15,26).
Assessment of physiological responses by athletes. Each subject performed an incremental exercise cycle test to volitional exhaustion on an electromagnetic ergometer (Ergo-Metrics 900, Ergoline D 7474, Bitz, Germany). The incremental protocol started with a 3-min warm-up at 60 W. The workload was then increased by 30 W every min until volitional exhaustion. During the exercise test, subjects wore a nose clip and breathed through a mouthpiece attached to a pneumotachograph (Fleich no. 3). The respired gas was continuously sampled by an automated system (CPX, Medical Graphics, St. Paul, MN) for breath-by-breath determination of metabolic and ventilatory variables. An IBM computer (Personal Computer AT, RDI, Langlade, France) was used to calculate an average value for each of the variables for any time period: O2, CO2, E, and respiratory exchange ratio (RER). Heart rate (HR) data were automatically recorded and calculated during exercise using an integrated electrocardiograph. The use of a noninvasive method to determine the VT has been described by Beaver et al. (3). This method is based on the breakdown point in the oxygen uptake/carbon dioxide output relationship, which corresponds to the VT expressed in mL·min−1·kg−1. The breakdown point was determined graphically. The reading was effected independently by two experienced reviewers. In the rare case of discordance, a third reviewer was used to reach consensus.
The tests were performed 2-3 d before rHuEpo administration to determine baseline values and the first week after the last drug intake.
Statistical analysis. All data are tabulated as mean ± SD.
A Student t-test was performed, on erythropoietin concentrations and biological parameters, during and after the treatment, to detect significant differences with baseline values (day 0, reference value).
Because the sample was small and the distribution was not normal, the comparison between physiological test data before and after rHuEpo administration was performed using a nonparametric method of Wilcoxon rank statistic.
A P-value of less than 0.05 was considered to be the limit of significance. The statistical analyses were carried out using the Pk-fit computer package (24). The Bartlett test was used to check variance homogeneity.
Mean baseline (± SD) sTfR/serum protein values with the corresponding 99.9% standard confidence intervals (CI) were computed from both the basal values measured in the subjects entering this study (9 subjects), and the values measured in the additional group of 176 subjects. A criterion based on the use of the t-statistics was defined to assess whether the ratio value measured in a new individual should be considered outside the expected baseline reference range. Because we showed in a earlier paper that sTfr values are not normally distributed (7), this statistical test was performed after log-transformation of the data. The criterion is based on the calculation of the minimal difference between the computed mean baseline and a new measurement value. The same statistical approach was carried out in the group of subjects at high altitude (48 subjects). The use of the t-statistics supplies a decisional criterion to assess whether such a difference should be considered to be an acceptable random fluctuation around the baseline value at a given probability level of 0.01.
Subjects. The physical characteristics of the nine athletes receiving rHuEpo are given in Table 1. At days 18, 19, and 21 of treatment, respectively, three athletes had an Hct value above 50% and consequently did not receive the final doses of rHuEpo. Individual body weight did not change over the study period. After repeated subcutaneous administration of rHuEpo, no adverse effect was reported by the subjects. There was no change in blood pressure response, and no thrombolitic event was observed.
Erythropoietin concentrations and biological parameters. Table 2 summarizes the mean values for Hct, Hb, and reticulocytes measured at day 0 before drug administration (reference values); at days 10, 14, 17, 21, and 24 during rHuEpo injection; and on days 1, 3, 7, 14, and 25 after the last drug injection. In comparison with day 0, a significant increase in reticulocytes was seen during treatment, from day 10 to 24. Moreover, for 7 d after treatment, reticulocytes were significantly higher than the baseline value, whereas 14 and 25 d after the last rHuEpo injection, reticulocytes were significantly lower than at day 0. Similarly, when compared with values at day 0, Hct was significantly increased from the last day of treatment up to 14 d after the last drug injection, and Hb was increased from day 1 to 7 posttreatment (Table 2).
A significant increase in serum erythropoietin concentrations was observed during treatment. From posttreatment day 2 to 4, these concentrations did not differ from the baseline values, whereas at days 7, 14, and 21 posttreatment, they were significantly lower than at day 0. During treatment and up to 14 d after the last drug injection, sTfr and the sTfr/serum protein ratio were statistically higher than at day 0. There was no statistically significant difference for serum proteins at any sampling time for the duration of the study. Results are given in Table 3.
Physiological test results. Metabolic and cardiorespiratory values obtained during cycle ergometry are presented in Table 4. O2max and power output at O2 max were significantly higher after 26 d of rHuEpo administration than at baseline. Maximal ventilation at O2max (Emax) and RER were not significantly different. The maximal heart rate of 177 beats·min−1 at the beginning of our study was significantly higher than the value of 168 beats·min−1 after 26 d of rHuEpo administration.
Compared with the values measured at baseline, the VT measured after rHuEpo administration occurred at a statistically significant higher level of oxygen uptake. When oxygen uptake measured at the VT was expressed as a percentage of O2max, the values obtained were also significantly higher.
Indirect detection of rHuEpo injections. In a recent study (7), we used the t-test to define a maximal acceptable threshold value of 10 μg·mL−1 for sTfR. From the results obtained in the present study, in which athletes received daily administration (50 IU·kg−1) of rHuEpo, we can see that during treatment, the sTfR levels were higher than 10 μg·mL−1 on day 10 for four of the nine subjects, and from days 14 to the last day of treatment for all the subjects. After the end of treatment, this marker remained higher than 10 μg·mL−1 for all subjects between days 1 and 7, and for five subjects on day 14. Twenty-five days after the last drug administration, sTfr values were lower than 10 μg·mL−1 for all subjects.
For the sTfr/serum protein ratio, the results are given in Table 5. The maximal acceptable difference between this marker and its baseline value was defined using the same t-test. The threshold value for sTfr/serum protein was 135 for the group of 185 athletes living at low altitude and 153 for the group of 48 subjects at high altitude. We therefore assumed that a ratio higher than 153 would indicate rHuEpo abuse in a healthy subject. The sTfR/serum protein ratios were above 153 on day 10 of treatment for four of the nine subjects, on day 14 for 8 subjects and from days 17 to 24 for all subjects. At the end of treatment, this marker remained above this threshold value for all subjects between days 1 and 7, for four subjects on day 14, and for only one subject on day 25. Three situations are possible: 1) this ratio is within the reference confidence intervals (i.e., 70-98), in which case we do not have evidence of the presence of doping factors; 2) this ratio is outside the reference confidence intervals but lower than the threshold value, in which case we can consider this value as highly improbable (<1/1000) in a normal population of athletes; and 3) this ratio is greater than the reference threshold of 153; in which case, we can assume the presence of a doping factor with a low probability of error (lower than 1%).
Few investigations have been conducted in healthy subjects after repeated subcutaneous administration of rHuEpo (7,23,26). A dose-dependent increase in hematological parameters has been reported under treatment with rHuEpo (2,5,6,8). To have a doping effect in athletes, with increases in RBC. Hb and Hct, repeated doses of rHuEpo would be necessary. Hence, in the present study, daily doses of 50 IU·kg−1 of rHuEpo alpha were administered subcutaneously for 26 d to mimic a situation of abuse of rHuEpo in athletes.
The recombinant hormone was well tolerated during our study. Hypertension, seizures, and vascular thrombolitic events, related to the increase in Hct, appear to be the major adverse effects in patients with chronic renal failure (22,25,27). In our study, such side-effects were not observed.
The results of this study indicate that repeated doses (50 IU·kg−1) of rHuEpo, administered subcutaneously, to normal iron-replete athletes, stimulated erythropoiesis. A significant increase in reticulocytosis was observed during treatment and up to 7 d after the last drug administration. Likewise, during the same period, the values of Hb and Hct remained significantly increased compared with baseline values. A negative-feedback loop of reticulocyte production occurred 14 d after end of treatment. These results agree with those of other investigators (9,11).
Serum erythropoietin concentrations were statistically different from baseline up to 1 d after the last injection. These concentrations, which were significantly lower 7 d posttreatment than at day 0, may be explained by a negative feedback loop of endogenous erythropoietin production.
We also evaluated the athletes' responses to rHuEpo administration, in terms of oxygen uptake and cardiorespiratory values during incremental cycle ergometry. The protocol used is optimal for cardiopulmonary assessment (10). The subjects were already familiarized with this protocol, so we were able to assume that they made maximal effort. During the study period, athletes followed their usual training program. After this period, significant improvements in O2max and O2 at VT were found. In contrast, HRmax was significantly decreased. All athletes improved their aerobic power. Because we observed an increase in oxygen uptake at VT expressed in percentage of O2max, we conclude that all subjects had obtained a significantly higher level of aerobic capacity. Two factors could have improved the oxygen uptake of each athlete: the continuation of their usual training program and/or the administration of rHuEpo. We were unable to evaluate precisely the part due to rHuEpo administration for several reasons: 1) our population of nine athletes was not homogeneous in terms of sporting activity, because the principal aim of the study was to find an indirect way to detect erythropoietin abusers; and 2) a control group with the same training program was not included.
Kohrt et al. (20) determined how the physiological test results of triathletes change after a 2-month period of training. Five subjects of the present study (5 triathletes) were of varying levels of ability, as evidenced by O2 max values at the beginning of the study (range = 55-68 mL·min−1·kg−1). This range is comparable to that of the triathletes population in the study of Kohrt et al. (range = 50.9-67.2 mL·min−1·kg−1) (20). As shown in Table 6, the training volumes (km·wk−1) in the two studies were similar. After 2-month intervals through a triathlon season, Kohrt et al. (20) found an increase of 4% in O2max (53.4 ± 15 to 55.5 ± 1.5 mL·min−1·kg−1). In the present study, after 26 d of similar training volume in km·wk−1, we found an increase of 8.5% in O2max (59 ± 6 to 65 ± 5 mL min·−1·kg−1). It therefore seems quite likely that rHuEpo made a substantial contribution to the improvement in maximal energetic capacity of the subjects of this study.
A significant decrease in HRmax was observed after rHuEpo administration. Such a decrease was not found by Kohrt et al. (20) after a 2-month training period. This decrease thus might be an effect of rHuEpo administration, which increased Hct and blood viscosity. To our knowledge, the effect of blood viscosity on HR has not been well defined; it therefore remains difficult to interpret this result.
Studies on homogeneous populations of athletes in terms of sporting activity and training program, matched to a control group, are necessary to assess the effect of rHuEpo on performance and cardiopulmonary responses.
In conclusion, in the absence of an erythropoietin plasma concentration greater than the baseline value, or in absence of rHuEpo detectable with the appropriate analytical method (23), the decisional rule above discussed requires the measurement of the sTfR/serum protein ratio from a blood sample. Ratio values above 153 indicate probable rHuEpo intake with a low probability of error (<1%). However, as previously reported (7), to confirm this finding, we recommend a second blood sample obtained 1 wk later to quantify changes in sTfR levels. In summary, an increased Hct, with sTfr above 10 μg·mL−1 and sTfr/serum proteins above 153, indicates the probable intake of rHuEpo. This approach, which is both quick and simple, may constitute a first step in doping control during competition.
1. Annable, L., P. M. Cotes, and M. V. Musset. The Second International Reference Preparation of Epo, human, urinary, for bioassay. Bull. W.H.O.
2. Ashai, N. I., E. P. Paganini, and J. M. Wilson. Intravenous versus subcutaneous dosing of Epoetin: a review of the literature. Am. J. Kidney Dis.
22(Suppl. 1):23-31, 1993.
3. Beaver, W. L., K. Wasserman, and B. J. Whipp. A new method for detecting anaerobic threshold by gas exchange. J. Appl. Physiol.
4. Berglund, B., and B. Ekblom. Effect of recombinant human erythropoietin
treatment on blood pressure and some haematological parameters in healthy males. J. Int. Med.
5. Besarab, A. Optimizing epoetin therapy in end-stage renal disease: the case of subcutaneous administration. Am. J. Kidney Dis.
22(Suppl. 1):13-22, 1993.
6. Bommer, J., M. Kugel, W. Schoeppe, et al. Dose-related effects of recombinant human erythropoietin
on erythropoiesis. Contrib. Nephrol.
7. Bressolle, F., M. Audran, R. Gareau, R. D. Baynes, A. Duvallet, and R. Gomeni. Population pharmacodynamics for monitoring epoetin in athletes
. Clin. Drug Invest.
8. Brockmöller, J., J. Köchling, W. Weber, M. Looby, I. Roots, and H. H. Neumayer. The pharmacokinetics and pharmacodynamics of recombinant human erythropoietin
in haemodialysis patients. Br. J. Clin. Pharmacol.
9. Brugnara, C., G. M. Colella, J. Cremins, et al. Effects of subcutaneous recombinant human erythropoietin
in normal subjects: development of decreased reticulocyte Hb content and iron-deficient erythropoiesis. J. Lab. Clin. Med.
10. Buchfuhrer, M. J., J. E. Hansen, and T. E. Robinson. Optimizing the exercise protocol for cardiopulmonary assessment. J. Appl. Physiol.
11. Casoni, I., G. Ricci, E. Ballarin, et al. Hematological indices of erythropoietin administration in athletes
. Int. J. Sports Med.
12. Eckardt, K. U., A. Kurtz, P. Hirth, P. Scigalla, L. Wieczorek, and C. Bauer. Evaluation of the stability of human erythropoietin in samples for radioimmunoassay. Klin. Wochensch.
13. Ekblom, B., and B. Berglund. Effect of erythropoietin administration on maximal aerobic power in man. Scand. J. Med. Sci. Sports
14. Gareau, R., G. R. Brisson, C. Ayotte, M. Audran, and J. L. Chanal. A possible approach to erythropoietin doping detection. In: Blood Samples in Doping Control: Proceedings of the Second International Symposium on Drugs in Sports,
P. Hemmersbach and K. I. Birkeland (Eds.). 1993 Aug 29-31: Oslo: Pensumtjenest, 1994, pp. 141-151.
15. Gareau, R., M. Audran, R. D. Baynes, et al. Erythropoietin abuse in athletes
16. Imai, N., A. Kawamura, M. Higuchi, et al. Physicochemical and biological comparison of recombinant human erythropoietin
with human urinary erythropoietin. J. Biochem.
17. Jelkman, W. Biology of erythropoietin. Clin. Invest.
18. Kicman, A. T., and D. A. Cowan. Peptides hormones and sport: misuse and detection. Br. Med. Bull.
19. Klausen, T., F. Dela, E. Hippe, and H. Galbo. Diurnal variations of serum erythropoietin in trained and untrained subjects. Eur. J. Appl. Physiol. Occup.
20. Kohrt, W. M., J. S. O'Connor, and J. S. Skinner. Longitudinal assessment of responses by triathletes to swimming, cycling, and running. Med. Sci. Sports Exerc.
21. Koury, M. J., and M. C. Bondurant. The molecular mechanism of erythropoietin action. Eur. J. Biochem.
22. Markham, A., and H. M. Bryson. Epoetin alfa: a review of its pharmacodynamic and pharmacokinetic properties and therapeutic use in nonrenal applications. Drugs
23. McMahon, F. G., R. Vargas, M. Ryan, et al. Pharmacokinetics and effects of recombinant human erythropoietin
after intravenous and subcutaneous injections in healthy volunteers. Blood
24. Pk-fit, version 1.0, R.D.P.P., Montpellier, France.
25. Singbart, G. Adverse events of erythropoietin in long-term and in acute/short-term treatment. Clin. Invest.
26. Souillard, A., M. Audran, F. Bressolle, R. Gareau, A. Duvallet, and J. L. Chanal. Recombinant human erythropoietin
and pharmacodynamic parameters in athletes
: interest of blood sampling for doping control. Br. J. Clin. Pharmacol.
27. Veys, N., and S. Ringoir. The subcutaneous administration route of epoetin: advantages, pain at the injection site and patient acceptance. Int. J. Artif. Organs
28. Wide, L., C. Bengtsson, and G. Birgegard. Circadian rhythm of erythropoietin in human serum. Br. J. Haematol.
29. Wide, L., C. Bengtsson, B. Berglund, and B. Ekblom. Detection in blood and urine of recombinant erythropoietin administered to healthy men. Med. Sci. Sports Exerc.