Therapeutic Equivalence Between Branded and Generic WARFArin Tablets in Brazil (WARFA)

Therapeutic Equivalence Between Branded and Generic Warfarin Sodium Tablets in Adult Patients With Atrial Fibrillation in Brazil - Crossover Randomized Controlled Equivalence Trial

The purpose of this study is to assess whether the switch from branded to generic warfarin or between different generic warfarin tablets may cause fluctuation in the results of coagulation tests (International Normalized Rate, acronym INR) in patients, thus predisposing them to unnecessary risks.

[Changes to the protocol made until the stage of the statistical analysis plan, before data analysis and also before unblinding.] Objective 1.1 Main Objective We aim to determine, through mean INR difference between formulations, whether there is therapeutic equivalency between the branded warfarin sodium (Marevan®) and the generic formulations in atrial fibrillation patients in Brazil. 1.2 Specific Objective In order to achieve the objective, we aim to assess with each warfarin formulation : Delta INR [new primary outcome]; mean INR [mean prothrombin time will not be analyzed since it will convey the same information as the mean INR]; mean dosage needed for anticoagulation [new outcome]; clinical events (the frequency of thromboembolic events, bleedings, deaths and adverse events); the time in therapeutic range; and compliance with treatment. Hypothesis Our hypothesis is that the delta INR of patients when using the generic tablets of warfarin sodium does not differ by more than 0.49 (two-sided) from the mean INR obtained when using the branded formulation, thereby demonstrating therapeutic equivalence. We will also assess this same outcome to assess therapeutic equivalence between the generic formulations to each other. Type of Study This study is a crossover randomized controlled equivalence trial. It comprises four phases, each one one-month long, being the first one a run-in period. Patients will use a different warfarin sodium tablet formulation in each one of the phases 2, 3 and 4: either the branded Marevan® (União Química/Farmoquímica, Brasil), or two generic drugs, (manufactured by União Química Farmacêutica Nacional S/A or Laboratório Teuto Brasileiro S/A, Brasil) purchased from Brazilian drugstores (not directly from the manufacturer). Every patient will use each one of the three formulations in a previously determined sequence (A, B, C, D, E or F) to which he was assigned by randomization. Since patients included in the study will already be using warfarin and it is not ethically acceptable to interrupt this treatment, we planned a run-in period equivalent to seven warfarin half-lives, long enough to washout any previous warfarin treatment. Then, in this first phase, patients will start treatment with the same warfarin formulation that will be used in the second phase thus avoiding the carry-over effect. This period will also be used to select patients compliant with treatment, accepting those that during the run-in have at least one of the three (including the baseline) INR results within 2.0 and 3.0 and a difference of INR results at the 3th and 4th week ≤ ±0.8. Methods An independent investigator (initials TFCP) will use a computer random number generator to allocate patients to one of the six sequences of treatment [update: we opted to apply randomization through numbered, opaque, sealed envelopes in order to facilitate allocation concealment]. The principal investigator (initials CGF) will separate, repack and dispense drugs for patients in opaque drug containers of identical appearance [update: the step of repacking the drugs in opaque containers was made by two investigators external to the study, BR and MFST, instead of the principal investigator]. This process will be based on the initials of the patient and the alphanumeric code, maintaining the allocation concealment since TFCP will be responsible for assigning to the repacked drugs the patient initials and the alphanumeric code correlated with the formulation of warfarin utilized. TFCP will also be accountable to conceal the allocation until the statistical analysis at the end of the study [update: BR and MFST created the coding and maintained its concealment]. Due to the features of the drug containers, physicians and the principal investigator (that assess patients and collect outcomes) will be blinded. Patients may not be blinded depending on the appearance of the manufactured tablets, but the main outcome will be a laboratory result (INR). Sample Size Forty-eight patients (eight in each group) is the minimum necessary number of individuals to detect a clinically significant fluctuation of 0.49 in the mean INR, with a 90% power at a 5% level of significance, if one were present. This sample size was estimated considering the Pocock's statistical method for quantitative outcomes and an achieved mean INR and standard deviation with the branded warfarin of 2.45 and 0.29 respectively [update: later we considered a more conservative standard deviation of 0.34, which resulted in a sample size of 11 patients per group]. To compensate possible withdrawals or exclusions of patients we plan to recruit 60 patients. Statistical Analysis For the INR, PT and TTR outcomes paired t-test will be used at a two-sided 5% level of significance. These analysis will be per protocol, i.e., data from patients that during the follow-up start treatment with substances or drugs that interact either in a moderate or major way or are contraindicated when used with warfarin (Annex I) will be excluded [update: multiple t-tests would not be adequate because nominal significance levels would not reflect the actual alfa, and thus we decided to apply multilevel mixed-effects linear regression. We have new definitions for per protocol analyses.]. Binary outcomes (thromboembolic or bleeding events) will be analysed by Mcnemar test at a 5% level of significance using intention to treat i.e. considering missing data as adverse outcomes. In this case sensibility analysis will be conducted [update: we decided for just presenting the events recorded during the trial without any hypothesis testing due to cross-over design limitations and lower power for these outcomes]. Exploratory analysis for subgroup of patients are not intended. Annex I (obtained from the online database Micromedex ® 2.0 in September 13, 2013) abciximab acarbose acemetacin acenocoumarol acetaminophen agrimony alclofenac alefacept allopurinol aloe aminoglutethimide amiodarone amitriptyline amobarbital sodium amoxapine amoxicillin ampicillin trihydrate amprenavir angelica anise antithyroid agents apazone apixaban aprepitant aprobarbital argatroban armodafinil arnica asafetida aspirin astragalus atazanavir atenolol atovaquone avocado azathioprine azithromycin bee pollen benoxaprofen benzbromarone betamethasone bicalutamide bilberry bismuth subsalicylate bivalirudin black cohosh extract black currant black haw black tea bladderwrack boceprevir bogbean boldo borage bosentan bromelain bromfenac buchu bufexamac butabarbital butalbital capecitabine capsaicin carbamazepine carbenicillin disodium carboplatin carprofen cassia cat's claw cefadroxil cefamandole cefazolin sodium cefdinir cefepime cefixime cefoperazone cefotaxime cefotetan cefpodoxime ceftazidime ceftibuten ceftizoxime ceftriaxone celecoxib celery cephalexin cephalothin sodium cephapirin chamomile chaparral chitosan chloral hydrate chloramphenicol chlordiazepoxide chlorotrianisene chlorpromazine hydrochloride cholestyramine chondroitin cimetidine cinchona ciprofloxacin cisapride monohydrate cisplatin citalopram clarithromycin clofibrate clomipramine hydrochloride clopidogrel clove clove oil cloxacillin benzathine coenzyme Q10 colesevelam contraceptives, combination cortisone cranberry juice curcumin cyclophosphamide cyclosporine, modified dabigatran dabrafenib dalteparin danaparoid danazol dandelion dapsone darunavir deferasirox delavirdine demeclocycline desipramine desvenlafaxine devil's claw dexamethasone dexlansoprazole dexmethylphenidate diazoxide diclofenac dicloxacillin diethylstilbestrol diflunisal dipyridamole dipyrone disopyramide disulfiram dong quai dothiepin doxepin doxorubicin doxycycline calcium dronedarone droxicam duloxetine enoxacin enoxaparin enteral nutrition enzalutamide eptifibatide erlotinib erythromycin acistrate escitalopram esomeprazole eterobarb ethacrynic acid ethanol ethchlorvynol ethotoin etodolac etoposide etravirine etretinate evening primrose oil exenatide ezetimibe felbamate fenbufen fenofibrate fenofibric acid fenoprofen fenugreek feverfew fish oil floctafenine flosequinan fluconazole fludrocortisone acetate flufenamic acid fluorouracil fluoxetine fluoxymesterone flurbiprofen flutamide fluvastatin fluvoxamine fosamprenavir fosaprepitant garlic gatifloxacin gefitinib gemcitabine gemfibrozil gemifloxacin ginger ginkgo ginseng, siberian glimepiride glipizide glucagon glucosamine glutethimide glyburide goldenseal green tea griseofulvin guggul halothane heparin calcium heptabarbital hexobarbital sodium high protein food horse chestnut horseradish hydrocortisone ibritumomab ibuprofen ifosfamide imatinib imipramine hydrochloride indomethacin indoprofen infliximab influenza virus vaccine ipriflavone isoniazid isoxicam itraconazole ivacaftor ivermectin kava ketoconazole ketoprofen ketorolac tromethamine lactulose lansoprazole leflunomide lepirudin levamisole levofloxacin licorice lopinavir lornoxicam lovastatin lycium marijuana meadowsweet mechlorethamine meclofenamate mefenamic acid melatonin meloxicam menthol mephobarbital mercaptopurine mesalamine mesna methandrostenolone methicillin methotrexate methyl salicylate methylphenidate methylprednisolone acetate methyltestosterone metronidazole miconazole mifepristone milnacipran minocycline hydrochloride mistletoe mitotane moricizine hydrochloride motherwort moxalactam moxifloxacin nabumetone nafcillin nalidixic acid nandrolone naproxen nelfinavir neomycin nettle extract nevirapine niacin nilutamide nimesulide norfloxacin nortriptyline hydrochloride noscapine ofloxacin omega-3-acid ethyl esters omeprazole onion oil orlistat oseltamivir oxacillin oxandrolone oxaprozin oxymetholone oxyphenbutazone oxytetracycline pantoprazole papaya paramethasone paroxetine parsley passionflower pau d'arco penicillin G penicillin V benzathine pentosan polysulfate sodium pentoxifylline phenindione phenobarbital phenprocoumon phenylbutazone phenytoin phytonadione piperacillin piracetam pirazolac piroxicam pirprofen policosanol pomegranate poplar posaconazole prasugrel prednisolone prednisone prickly ash primidone procarbazine proguanil propafenone hydrochloride propoxyphene propranolol propyphenazone proquazone protriptyline pumpkin seed quassia quetiapine quinestrol quinidine quinine rabeprazole sodium raloxifene ranitidine red clover rifabutin rifampin rifapentine rifaximin rilonacept ritonavir rivaroxaban rofecoxib romidepsin ropinirole rosuvastatin roxithromycin salicylates saquinavir sarsaparilla saw palmetto extract secobarbital senega sertraline simvastatin sitaxsentan skullcap sorafenib soybean spironolactone st john's wort stanozolol sucralfate sulfamethoxazole sulfasalazine sulfinpyrazone sulfisoxazole sulindac sulofenur suprofen sweet woodruff tamarind tamoxifen tan shen teduglutide telaprevir telithromycin tenidap sodium tenoxicam terbinafine teriflunomide testosterone tetracycline thyroid hormones tiaprofenic acid tibolone ticarcillin ticlopidine tigecycline tinidazole tinzaparin tirofiban tocilizumab tolmetin tolterodine tartrate tonka toremifene citrate torsemide tramadol trastuzumab treprostinil triamcinolone acetonide trimipramine valdecoxib valproic acid vancomycin vemurafenib venlafaxine vilazodone vincristine vindesine vitamin A vitamin E vitamin K voriconazole vorinostat wild lettuce willow wintergreen yarrow zafirlukast zileuton zomepirac sodium zotepine

Therapeutic Equivalency, Warfarin, Drugs, Generic, Anticoagulants, Anticoagulants [Pharmacological Action], Coumarins

Phase 1 (run-in): Marevan® Phase 2: Marevan® Phase 3: generic warfarin #1 Phase 4: generic warfarin #2

Phase 1 (run-in): generic warfarin #1 Phase 2: generic warfarin #1 Phase 3: Marevan® Phase 4: generic warfarin #2

Phase 1 (run-in): generic warfarin #1 Phase 2: generic warfarin #1 Phase 3: generic warfarin #2 Phase 4: Marevan®

Phase 1 (run-in): Marevan® Phase 2: Marevan® Phase 3: generic warfarin #2 Phase 4: generic warfarin #1

Phase 1 (run-in): generic warfarin #2 Phase 2: generic warfarin #2 Phase 3: Marevan® Phase 4: generic warfarin #1

Phase 1 (run-in): generic warfarin #2 Phase 2: generic warfarin #2 Phase 3: generic warfarin #1 Phase 4: Marevan®

Branded warfarin sodium tablets (manufacturer União Química Farmacêutica Nacional S/A, Brazil) by mouth; dose adjusted according to INR and medical judgement.

Generic warfarin sodium tablets, manufacturer (União Química Farmacêutica Nacional S/A, Brazil), by mouth; dose adjusted according to INR and medical judgement.

Generic warfarin sodium tablets, manufacturer (Laboratório Teuto Brasileiro S/A, Brazil), by mouth; dose adjusted according to INR and medical judgement.

INR: International normalized ratio. INR will be assessed twice a month during the 16 weeks of study. Delta INR will be calculated by the difference, in absolute numbers (i.e. in module), between the two valid INR measurements due for each patient with each formulation at each period. Difference between the delta INRs will be the outcome. We will compare generic to branded warfarin (Marevan®) and the generic formulations to each other.

INR: International normalized ratio. INR will be assessed twice a month during the 16 weeks of study. We will calculate the mean INR for each formulation of warfarin; difference between the mean INRs will be the outcome. We will compare generic to branded warfarin (Marevan®) and the generic formulations to each other.

The warfarin dosage will be calculated as the sum of the dosage taken in the week before the study visit, as referred by the patient. We will calculate the mean dosage for each formulation of warfarin; difference between the mean dosage will be the outcome. We will compare generic to branded warfarin (Marevan®) and the generic formulations to each other.

Includes ischemic stroke (differential diagnosis with hemorrhagic stroke by tomography) and thromboembolism of viscera or extremities (diagnosed by acute symptoms and relevant diagnostic tests).

Classified as major or minor bleeding events. Major: intracranial hemorrhage, fatal bleeding, bleeding resulting in hemoglobin loss equal to or greater than 2.0 g / L, hemorrhage requiring transfusion, bleeding in sensitive areas such as the retina or pericardium. Minor: all other bleeding.

TTR calculated by Rosendaal Method also known as linear interpolation method. We will calculate the mean TTR for each formulation of warfarin; difference between the mean TTRs will be the outcome. We will compare generic to branded warfarin (Marevan®) and the generic formulations to each other.

Inclusion Criteria: diagnosis of nonvalvular atrial fibrillation atrial flutter by electrocardiogram and echocardiography; CHA2DS2VASc score equal to or greater than 1; already in use of warfarin for at least 2 months; during the run-in, at least one of the three (including the baseline) INR results within 2.0 and 3.0 and a difference of INR results at the 3th and 4th week ≤ ±0.8 signing of Informed Consent Form. Exclusion Criteria: patients with serious contraindications to the use of anticoagulants (significant bleeding, known sensitivity to warfarin); women of childbearing age, pregnant or breastfeeding; patients with thrombocytopenia; patients with hepatic or renal impairment; patients with a history of bleeding episodes due to congenital deficiency of coagulation factors; patients enrolled in another trial; patients initiating treatment with drugs with major interactions or which are contraindicated when used concomitantly with warfarin, according to our list (Annex I).

Gomes Freitas C, Walsh M, Coutinho EL, Vincenzo de Paola AA, Atallah AN. Examining therapeutic equivalence between branded and generic warfarin in Brazil: The WARFA crossover randomized controlled trial. PLoS One. 2021 Apr 1;16(4):e0248567. doi: 10.1371/journal.pone.0248567. eCollection 2021.

Freitas CG, Walsh M, Atallah AN. Design and rationale for the WARFA trial: a randomized controlled cross-over trial testing the therapeutic equivalence of branded and generic warfarin in atrial fibrillation patients in Brazil. BMC Cardiovasc Disord. 2017 Jun 7;17(1):148. doi: 10.1186/s12872-017-0584-4.