Wakefulness-promoting agent
Armodafinil
The Nuvigil label gives armodafinil's apparent terminal half-life as approximately 15 hours, the same as the app's figure of 15 hours.1
Armodafinil after one dose: 50% at ~15 h, 25% at 30 h, 12.5% at 45 h.
Drawn from the half-life in the app's model, ~15 h. The model's figures
- Compound
- Armodafinil
- Symbol
- ARM
- Family
- Cognitive and sleep peptides
- Parts
- 14
- References
- 11
- Checked
- Last reviewed . Sources are listed below.
Overview
Sources in this part
Armodafinil is the R-enantiomer of modafinil, a wakefulness-promoting agent taken by mouth, and the US label for Nuvigil gives its initial approval year as 2007.1 The label states that Nuvigil is "indicated to improve wakefulness in adult patients with excessive sleepiness associated with obstructive sleep apnea (OSA), narcolepsy, or shift work disorder (SWD)".1 The app lists armodafinil as a wakefulness-promoting agent in the cognitive category, grades its evidence as strong, and gives a typical dose of 150 to 250 mg/day by the oral route.
How it works
Sources in this part
The Nuvigil label states that "The mechanism(s) through which armodafinil promotes wakefulness is unknown".1 It adds that armodafinil and modafinil "bind in vitro to the dopamine transporter and inhibit dopamine reuptake", and describes armodafinil as an indirect dopamine receptor agonist.1 For modafinil, the label links this activity to increased extracellular dopamine in some brain regions of animals, and reports that in mice lacking the dopamine transporter modafinil lacked wake-promoting activity.1 These are in vitro and animal findings, and some were obtained with modafinil rather than armodafinil.1 The app's notes describe dopamine transporter inhibition as a confirmed mechanism, while the label gives the mechanism as unknown, so the page follows the label wording.
Half-life and pharmacokinetics
Values from the Milligram app's model.
- Half-life
- ~15 h
- Bioavailability
- 80%
- Typical dose
- 150–250 mg/day
- Route
- Oral
- Time to peak
- ~2 h
Sources in this part
The table sets the app's figures beside the figures the sources report.
| Parameter | App's figure | Reported in the sources |
|---|---|---|
| Half-life | ~15 h | Apparent terminal half-life of approximately 15 hours on the label;1 mean single-dose terminal half-life of approximately 13 hours in healthy adults (Darwish 2009);2 mean 16.5 hours (SD 4.44) for armodafinil and 14.4 hours for modafinil in 42 patients with treated OSA (Darwish 2010)3 |
| Bioavailability | 80% | The label states that "The absolute oral bioavailability was not determined due to the aqueous insolubility of armodafinil, which precluded intravenous administration"1 |
| Time to peak | ~2 h | Peak plasma concentrations at approximately 2 hours in the fasted state; time to peak may be delayed by approximately 2 to 4 hours in the fed state1 |
| Model | One compartment | The label describes "an apparent monoexponential decline from the peak plasma concentration"1 |
The label states that apparent steady state is reached within 7 days of dosing, and that systemic exposure at steady state is 1.8 times the exposure after a single dose.1 The app lists 80% bioavailability, but the label states that the absolute figure was not determined, so 80% is not a label or trial figure.1
Armodafinil compared with modafinil
Darwish and colleagues (Clinical Drug Investigation, 2009; the first author is listed at the Clinical Pharmacology Department of Cephalon, Inc.) pooled three randomized single-dose studies in healthy adults, with armodafinil at 50, 100, 200, 250, 300 or 400 mg and modafinil at 400 mg, dose-normalised to 200 mg.2 Both drugs had a mean terminal half-life of approximately 13 hours, but plasma concentrations "appeared to decline in a monophasic manner with armodafinil, but in a biphasic manner with modafinil due to the initial rapid elimination of its S-isomer".2 The mean AUC from time zero to the last measurable concentration and to infinity were "33% and 40% higher, respectively, with armodafinil compared with modafinil on a milligram-to-milligram basis".2 The app's notes give the same 33 to 40 percent range; the studies below report larger ratios in patients and at steady state. Darwish and colleagues (Clinical Therapeutics, 2010) compared single and multiple doses of 200 mg in 42 patients with continuous positive airway pressure-treated OSA in an open-label, randomized crossover study at 5 US centers.3 The authors report AUC ratios of armodafinil to modafinil of 1.64 (95% CI 1.60 to 1.68) after a single dose and 1.69 (95% CI 1.65 to 1.72) after multiple doses.3 The multiple-dose Cmax and AUC ratios "did not meet the US Food and Drug Administration (FDA) criteria for bioequivalence".3 The authors report that adverse events were mild to moderate in intensity, and that the most frequently reported while receiving armodafinil or modafinil were headache (29% and 2%, respectively), diarrhea (12% and 5%), nausea (10% and 2%) and dizziness (10% and 5%).3 The label gives a related steady-state comparison: Cmax and AUC of armodafinil were "approximately 37% and 70% higher, respectively" after 200 mg of Nuvigil than after 200 mg of Provigil (modafinil), which it attributes to "the more rapid clearance of the S-enantiomer (elimination half-life approximately 4 hours)".1
Metabolism, interactions and special populations
Darwish and colleagues (Clinical Pharmacokinetics, 2008) summarised three open-label interaction studies in 77 healthy subjects (n = 29, 24 and 24); the first author is listed at Cephalon Inc, and headache and dizziness were the most commonly reported adverse events.4 After at least 22 days of armodafinil at 250 mg/day, the AUC of midazolam fell by approximately 17% (intravenous) and 32% (oral), and oral midazolam Cmax fell by approximately 19%.4 With armodafinil 400 mg given 2 hours before a 40 mg oral omeprazole dose, omeprazole AUC rose by approximately 38%; armodafinil had no effect on caffeine Cmax or AUC.4 The authors conclude that armodafinil "did not induce CYP1A2 but was a moderate inducer of CYP3A4 and a moderate inhibitor of CYP2C19 in healthy subjects", and they add that dosage adjustments "may be required" for substrates of those enzymes.4 The label words the in vitro finding more weakly, as "a weak inducer of CYP3A activity in a concentration-related manner", and reports a 40% increase in omeprazole exposure at armodafinil 400 mg.1 The label states that the effectiveness of hormonal contraceptives "may be reduced when used with NUVIGIL and for one month after discontinuation of therapy".1 In a label-described clinical study, armodafinil exposure was approximately 15% higher in 24 elderly subjects (65 years or older) than in 25 young subjects.1 For hepatic impairment the label gives data for modafinil, not armodafinil: in 9 patients with cirrhosis, oral clearance of modafinil "was decreased by about 60% and the steady state concentration was doubled compared to normal patients".1
Level over 6 weeks
Level, as a share of the peak after one 100 mg dose
As studied: once daily, the label schedule for OSA, narcolepsy and SWD1.
An illustration from a pharmacokinetic model using a half-life of 15 hours. It is not a measured level or a suggested dose.
Human studies
The label's clinical studies section describes four placebo-controlled trials of 12 weeks: two in OSA (395 and 263 patients), one in narcolepsy (196) and one in shift work disorder (254).1
| Study and year | Design | People | Dose and length | Main result |
|---|---|---|---|---|
| Roth and colleagues, Clinical Therapeutics 20065 | Multicenter, double-blind, randomized, placebo-controlled trial at 37 centers in the United States and Canada | 395 enrolled adults aged 18 to 65 with residual excessive sleepiness from OSA/HS despite nCPAP | Armodafinil 150 mg or 250 mg or placebo once daily; 12 weeks | Greater improvement than placebo on MWT sleep latency, CGI-C (72% vs 37% at least minimally improved) and ESS (-5.5 vs -3.3); 4 serious adverse events (1.5%) on armodafinil, judged not or unlikely drug related by the investigator |
| Harsh and colleagues, Current Medical Research and Opinion 20066 | Multicenter, double-blind, randomized, placebo-controlled trial | 196 adults aged 18 to 65 with excessive sleepiness from narcolepsy | Armodafinil 150 mg (n = 65), 250 mg (n = 67) or placebo (n = 64) once daily; 12 weeks | MWT latency rose 1.3, 2.6 and 1.9 minutes (150 mg, 250 mg, combined) against a 1.9 minute decrease on placebo; CGI-C improvement 71% combined vs 33%; common adverse events headache, nausea and dizziness |
| Czeisler and colleagues, Mayo Clinic Proceedings 20097 | Randomized, placebo-controlled trial at 42 sleep research facilities in North America | 254 permanent or rotating night shift workers with SWD (245 received at least one dose) | Armodafinil 150 mg or placebo 30 to 60 minutes before each night shift; 12 weeks | Night-shift MSLT latency 2.3 to 5.3 minutes vs 2.4 to 2.8 on placebo (P<.001); CGI-C improvement 79% vs 59%; 28% of 245 withdrew (30 armodafinil, 38 placebo) |
| Black and colleagues, Journal of Clinical Sleep Medicine 20108 | Open-label extension of four 12-week double-blind studies, 12 months or longer | 743 enrolled patients (474 treated OSA, 113 SWD, 156 narcolepsy) | Armodafinil 50 to 250 mg/day; mean exposure 372.3 days | 57% (420/743) completed 12 months or more; discontinuation for adverse events 13% (95/743) in the first 12 months; blood pressure up 3.6/2.3 mm Hg and heart rate up 6.7 beats per minute |
Deeper analysis: shift work disorder trial (Czeisler 2009)
Czeisler and colleagues reported a 12-week randomized trial of armodafinil 150 mg in shift work disorder in Mayo Clinic Proceedings in 2009.7 Of 747 adults screened, 254 met entry criteria; 245 received at least one dose, and the efficacy analysis included 216 (85%) of 254 patients.7 Mean sleep latency rose from 2.3 (SD 1.6) to 5.3 (SD 5.0) minutes on armodafinil and from 2.4 (SD 1.6) to 2.8 (SD 2.9) minutes on placebo, and 79% of the armodafinil group against 59% of the placebo group were rated improved.7 Secondary results favored armodafinil on diary-reported sleepiness during laboratory nights (P<.001), night shifts at work (P<.001) and the commute home (P=.003), and on memory and attention tests, except that the change in speed of memory at the final visit (armodafinil −257.2 milliseconds, placebo −140.4 milliseconds) was "not statistically significant (P=.09)"; the authors also write that for the commute home the improvement in mistakes, near misses and accidents "was not statistically significant compared with placebo", and that the proportion of placebo patients rated improved on the CGI-C was "higher than what was anticipated".7 Of the 68 withdrawals, adverse events accounted for 7 in the armodafinil group and 4 in the placebo group, and "No patients discontinued participation because of lack of efficacy".7 Headache occurred in 15 of 123 (12%) on armodafinil and 12 of 122 (10%) on placebo, nausea in 9 (7%) and 4 (3%), and anxiety in 6 (5%) and 2 (2%).7 Severe adverse events were reported by the site investigator more often with armodafinil (n=12) than with placebo (n=3), and the one serious adverse event on armodafinil was suicidal ideation in a patient with a history of depression, judged possibly related to the study medication.7 The authors state that "Most patients enrolled were permanent night shift workers" which "may limit the generalizability of these results to individuals working alternative shift schedules", and that "a proportion of patients remained sleepy on objective assessment at the end of treatment".7 Funding and ties: "Cephalon Inc funded the study and developed the protocol in collaboration with the authors", the statistical analysis was conducted by "A qualified statistician (S.A.) employed by the study sponsor", and the analysis was also replicated by an independent academic group.7 The disclosures state that Czeisler holds an endowed professorship at Harvard provided by Cephalon and that Roth "has served on the speakers' bureau for Cephalon and Sanofi".7
Deeper analysis: obstructive sleep apnea trials (Roth 2006)
Roth and colleagues reported a 12-week trial in Clinical Therapeutics in 2006 of armodafinil 150 or 250 mg once daily as adjunctive treatment for residual excessive sleepiness in nCPAP-adherent adults with OSA/HS.5 Of 395 enrolled (133, 131 and 131 in the 150 mg, 250 mg and placebo groups), 392 received at least one dose.5 The record gives the MWT change in the combined armodafinil group as +1.9 minutes (SD 7.3) against 1.7 (SD 8.6) for placebo, with the sign of the placebo value not shown in the PubMed text; the label's table lists the placebo change in this trial as -1.7 minutes.5,1 At the final visit the CGI-C improvement rates were 72% vs 37%, the ESS change was -5.5 (SD 5.0) vs -3.3 (SD 4.7), and the Brief Fatigue Inventory change was -1.2 vs -0.6.5 The authors report "No significant effects on nighttime sleep, as assessed using polysomnography" and that adverse events in both groups were headache, nausea, insomnia, anxiety and dizziness.5 Four serious adverse events (ulcerative colitis, migraine, worsening of Axis II and mood disorder, and duodenal ulcer) occurred in 4 (1.5%) patients on armodafinil and were judged by the investigator not or unlikely to be drug related.5 The PubMed record for this paper lists no funding statement, author disclosures or limitations section, and its publication type is tagged "Research Support, Non-U.S. Gov't"; its author list includes Niebler G and Arora S, and the sister shift work trial paper acknowledges Gwendolyn E. Niebler as a former Cephalon employee and states that its statistician (S.A.) was employed by the study sponsor.5,7 The second OSA trial (263 patients, armodafinil 150 mg against placebo, described on the label) has a ClinicalTrials.gov entry that names Cephalon as lead sponsor; it posts an MWT sleep latency change of 2.3 minutes (SD 7.80, n = 116) on armodafinil against -1.3 (SD 7.08, n = 120) on placebo, and serious adverse events in 0 of 129 on armodafinil and 1 of 130 on placebo.9
Deeper analysis: narcolepsy trial (Harsh 2006)
Harsh and colleagues reported a 12-week multicenter double-blind trial in Current Medical Research and Opinion in 2006, enrolling 196 adults aged 18 to 65 with excessive sleepiness from narcolepsy.6 The MWT changes (0900 to 1500) differed from placebo at p < 0.01 for all three comparisons, and for late-day latency (1500 to 1900) the combined group differed from placebo by 2.8 minutes (p = 0.0358).6 The label adds that although each dose differed from placebo on the MWT, "the magnitude of effect was observed to be greater for the higher dose".1 The PubMed record gives the most common adverse events as headache, nausea and dizziness, without rates, and no discontinuation or serious-event counts.6 The ClinicalTrials.gov entry NCT00078377 lists 196 enrolled and Cephalon as lead sponsor, and its status is completed.10 The posted results list serious adverse events in 0 of 67 on armodafinil 250 mg, 1 of 64 on 150 mg and 0 of 63 on placebo, and 11, 16 and 9 patients not completing the study in the 250 mg, 150 mg and placebo groups.10 The PubMed record for the paper states no funding or limitations.6
Deeper analysis: open-label extension (Black 2010)
Black and colleagues reported an open-label extension in the Journal of Clinical Sleep Medicine in 2010 for patients who completed one of four 12-week double-blind studies, with armodafinil at 50 to 250 mg/day for 12 months or longer.8 Of 743 enrolled, 57% (420/743) completed 12 months or more, and 13% (95/743) discontinued for adverse events in the first 12 months; across the whole study 15% discontinued for adverse events.8 The most common adverse events were headache (25%, 180/731), nasopharyngitis (17%, 123/731) and insomnia (14%, 99/731), which the authors describe as "generally of mild-to-moderate intensity".8 Vital signs rose modestly (blood pressure 3.6/2.3 mm Hg, heart rate 6.7 beats per minute), mostly by month 3, and the authors write that "increased monitoring of blood pressure may be warranted" in patients at risk for cardiovascular disorders, particularly treated OSA.8 One death occurred, in a patient with treated OSA and atherosclerotic heart disease, and was considered unlikely to be related to the study drug by the investigator; 3 patients withdrew for elevated liver function tests.8 Serious adverse events were reported by 8% of the safety population (10% of 468 with treated OSA, 4% of 108 with SWD and 5% of 155 with narcolepsy), 90% of them (83/92) were considered by the investigator not or unlikely to be related to treatment, and hypertension (6%, 42/731) was the most common vascular adverse event.8 The authors state the limitations: "The limitations of this study include its open-label design and use of subjective measures only, both of which could have biased results in favor of overrated efficacy", and that enrolling only completers of earlier studies made patients with a tolerable response more likely to be recruited.8 Funding: "This study was sponsored by Cephalon, Inc., Frazer, PA", and several authors list Cephalon affiliations.8
Doses reported in studies and on labels
| Source | Dose reported | People and comparator | Reference |
|---|---|---|---|
| US label, OSA or narcolepsy | 150 mg to 250 mg orally once a day in the morning | Adults with excessive sleepiness associated with OSA or narcolepsy; the label adds there is "no consistent evidence" that doses up to 250 mg/day in OSA give benefit beyond 150 mg/day | 1 |
| US label, SWD | 150 mg orally once a day, approximately one hour before the start of the work shift | Adults with shift work disorder | 1 |
| US label, hepatic impairment | Reduced dose | Patients with severe hepatic impairment | 1 |
| US label, geriatric patients | "consider lower dose" (the label's wording) | Geriatric patients; in a label-described study armodafinil exposure was approximately 15% higher in elderly subjects | 1 |
| Roth 2006 | 150 mg or 250 mg once daily for 12 weeks | 395 adults with OSA/HS on nCPAP; placebo comparator | 5 |
| Harsh 2006 | 150 mg or 250 mg once daily for 12 weeks | 196 adults with narcolepsy; placebo comparator | 6 |
| Czeisler 2009 | 150 mg, 30 to 60 minutes before each night shift, 12 weeks | 254 night shift workers with SWD; placebo comparator | 7 |
| Black 2010 | 50 to 250 mg/day for 12 months or longer; 250 mg was the most common dose (67% of 731) | 743 patients in an open-label extension; no comparator | 8 |
The app's typical dose of 150 to 250 mg/day matches the label's 150 mg to 250 mg once a day for OSA or narcolepsy.1
Side effects reported
Sources in this part
The label states that Nuvigil "has been evaluated for safety in over 1,100 patients with excessive sleepiness associated with OSA, SWD, and narcolepsy".1 In the pooled placebo-controlled trials the most common adverse reactions (5% or more, and more frequent than on placebo) were headache, nausea, dizziness and insomnia.1 The label's pooled table (645 on Nuvigil, 445 on placebo) lists headache in 17% against 9%.1 In those trials, 44 of 645 patients (7%) on Nuvigil discontinued for an adverse reaction against 16 of 445 (4%) on placebo.1 The label carries warnings for serious rash, and states that "Rare cases of serious or life-threatening rash, including SJS and toxic epidermal necrolysis (TEN), have been reported in adults and children in worldwide postmarketing experience with modafinil and NUVIGIL".1 For DRESS (multi-organ hypersensitivity) the label reports "One fatal case of DRESS that occurred in close temporal association (3 weeks) with the initiation of NUVIGIL treatment" in the postmarketing setting.1 The label also lists postmarketing psychiatric reactions, and in pre-approval trials anxiety, agitation, nervousness and irritability led to discontinuation in 1.2% on Nuvigil against 0.3% on placebo.1 On dependence, the label states that "Physical dependence can occur in patients treated with NUVIGIL", and reports that abuse and misuse have been observed in the postmarketing period.1 For pregnancy, the label states that "Limited available data on armodafinil use in pregnant women are insufficient to inform a drug associated risk of adverse pregnancy outcomes".1
Bloodwork and markers
Sources in this part
The app lists liver enzymes, blood pressure and heart rate as markers to monitor. For liver tests, the label states that "Mean plasma levels of gamma glutamyltransferase (GGT) and alkaline phosphatase (AP) were found to be higher following administration of NUVIGIL, but not placebo", with few patients outside the normal range.1 The label reports no differences in ALT, AST, total protein, albumin or total bilirubin, "although there were rare cases of isolated elevations of AST and/or ALT".1 For blood pressure, the label reports small average increases of 1.2 to 4.3 mmHg in the pre-approval trials of 3 months or less, and a greater share of patients on Nuvigil needing new or increased antihypertensive medication (2.9% against 1.8% on placebo).1 For heart rate, it reports "a small, but consistent, average increase in pulse rate over placebo", varying from 0.9 to 3.5 beats per minute.1 The shift work disorder trial reported that differences in vital sign parameters were not statistically significant at the time points measured.7
Regulatory and anti-doping status
Sources in this part
The US label gives 2007 as the initial approval year and lists armodafinil as "a Schedule IV controlled substance".1 The 2026 WADA Prohibited List places stimulants under S6, prohibited in-competition, and states that "All stimulants, including all optical isomers, e.g. d- and l- where relevant, are prohibited".11 The list names modafinil among the non-specified stimulants in S6.A; the extracted text of the list does not name armodafinil, which the label describes as the R-enantiomer of modafinil.11,1
Storage and handling
Sources in this part
What is not known
Sources in this part
The label states that the mechanism by which armodafinil promotes wakefulness is unknown, and that absolute oral bioavailability was not determined because no intravenous form was possible.1 The label states that the influence of race or ethnicity on armodafinil pharmacokinetics has not been studied, that data specific to armodafinil protein binding are not available, and that safety and effectiveness in pediatric patients have not been established.1 The label gives its hepatic impairment data from modafinil, and states there are no data on armodafinil or its metabolites in human milk.1 Czeisler and colleagues call for a direct comparison of armodafinil and modafinil, and Black and colleagues write that only a double-blind, placebo-controlled study without restriction to earlier completers can "truly discern long-term efficacy and tolerability".7,8 A ClinicalTrials.gov search for "armodafinil" on 16 November 2026 returned 62 registered studies, including trials in malignant glioma, bipolar I depression and traumatic brain injury; their results are not summarised here.
Frequently asked questions
What is the half-life of armodafinil?
How long does armodafinil take to reach its peak level?
The label gives peak plasma concentrations at approximately 2 hours in the fasted state, and the app lists about 2 hours.1
Is armodafinil the same as modafinil?
Armodafinil is the R-enantiomer of modafinil, which is a 1:1 mixture of the R- and S-enantiomers.1 In a crossover study of 42 patients with OSA, the armodafinil to modafinil AUC ratio was 1.64 after a single 200 mg dose, and the authors state the multiple-dose exposures "did not meet the US Food and Drug Administration (FDA) criteria for bioequivalence".3
Is armodafinil banned by WADA?
Is armodafinil a controlled substance?
The US label states that Nuvigil contains armodafinil, "a Schedule IV controlled substance".1
What is armodafinil approved for?
Does armodafinil affect hormonal contraceptives?
The label states that the effectiveness of hormonal contraceptives "may be reduced when used with NUVIGIL and for one month after discontinuation of therapy".1
What are the most common side effects of armodafinil?
Is the absolute bioavailability of armodafinil known?
Guides, tools and comparisons
References
- Cephalon, LLC. Nuvigil (armodafinil) tablets, C-IV. US prescribing information, DailyMed version 32 (Teva Pharmaceuticals as distributor). DailyMed label
- Darwish M, et al. Armodafinil and modafinil have substantially different pharmacokinetic profiles despite having the same terminal half-lives: analysis of data from three randomized, single-dose, pharmacokinetic studies. Clin Drug Investig 2009. PMID 19663523
- Darwish M, et al. Pharmacokinetics of armodafinil and modafinil after single and multiple doses in patients with excessive sleepiness associated with treated obstructive sleep apnea: a randomized, open-label, crossover study. Clin Ther 2010. PMID 21118743
- Darwish M, et al. Interaction profile of armodafinil with medications metabolized by cytochrome P450 enzymes 1A2, 3A4 and 2C19 in healthy subjects. Clin Pharmacokinet 2008. PMID 18076219
- Roth T, et al. Effects of armodafinil in the treatment of residual excessive sleepiness associated with obstructive sleep apnea/hypopnea syndrome: a 12-week, multicenter, double-blind, randomized, placebo-controlled study in nCPAP-adherent adults. Clin Ther 2006. PMID 16861091
- Harsh JR, et al. The efficacy and safety of armodafinil as treatment for adults with excessive sleepiness associated with narcolepsy. Curr Med Res Opin 2006. PMID 16684437
- Czeisler CA, et al. Armodafinil for treatment of excessive sleepiness associated with shift work disorder: a randomized controlled study. Mayo Clin Proc 2009. PMC2770907
- Black JE, et al. The long-term tolerability and efficacy of armodafinil in patients with excessive sleepiness associated with treated obstructive sleep apnea, shift work disorder, or narcolepsy: an open-label extension study. J Clin Sleep Med 2010. PMC2952749
- Cephalon. Armodafinil (CEP-10953) in treatment of excessive sleepiness associated with obstructive sleep apnea/hypopnea (OSA/H) syndrome. ClinicalTrials.gov NCT00079677. NCT00079677
- Cephalon. Safety and efficacy study of armodafinil (CEP-10953) in the treatment of excessive sleepiness associated with narcolepsy. ClinicalTrials.gov NCT00078377. NCT00078377
- World Anti-Doping Agency. World Anti-Doping Code International Standard: Prohibited List 2026. WADA list