Examining therapeutic equivalence between branded and generic warfarin in Brazil: The WARFA crossover randomized controlled trial

• Published in: PloS one 2021-04, Vol.16 (4), p.e0248567
• Main Authors: Gomes Freitas, Carolina, Walsh, Michael, Coutinho, Enia Lucia, Vincenzo de Paola, Angelo Amato, Atallah, Álvaro Nagib
• Format: Article
• Language: English
• Subjects: Aged; Anticoagulants; Anticoagulants - therapeutic use; Atrial Fibrillation - drug therapy; Atrial Fibrillation - pathology; Bleeding; Blood Coagulation - drug effects; Brazil; Breast feeding; Cardiac arrhythmia; Cardiology; Clinical trials; Coagulation; Comparative analysis; Cross-Over Studies; Drug dosages; Drug Monitoring - methods; Drugs, Generic - therapeutic use; Editing; Electrophysiology; Emergency medical care; Emergency medical services; Equivalence; Female; Fibrillation; Flutter; Funding; Generic drugs; Health risks; Heart; Heart valves; Humans; Informed consent; International Normalized Ratio - methods; Laboratories; Male; Medical research; Medicine; Medicine and Health Sciences; Middle Aged; Patients; People and places; Proprietary drugs; Randomization; Renal failure; Research and Analysis Methods; Research methods; Statistical analysis; Testing; Therapeutic Equivalency; Thrombocytopenia; Treatment Outcome; Warfarin; Warfarin - therapeutic use
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0248567

To determine whether the generic and branded warfarins used as anticoagulants in Brazil are therapeutic equivalents based on their international normalized ratio (INR) results. This crossover randomized controlled trial had four periods. We used the branded Marevan and two generic versions of warfarin sodium tablets, manufactured by União Química and Teuto laboratories, all purchased from retail drugstores. Eligible participants were outpatients from an anticoagulation clinic at a university hospital in São Paulo, Brazil. They had atrial fibrillation or flutter and had been using warfarin for at least 2 months with an INR therapeutic range of 2.0-3.0. Randomization was by numbered, opaque, sealed envelopes. Healthcare personnel and outcome assessors were blinded to treatments, but patients were not. The primary outcome was the variability in the INR (ΔINR) and secondary outcomes included mean INR. We accepted formulations as equivalent if the 95% confidence interval (CI) of the comparison of ΔINR between branded and generic formulations was within the limit of ±0.49. One hundred patients were recruited and randomized to six sequences of treatment (four sequences with n = 17 and two sequences with n = 16). União Química generic warfarin had equivalent variability in the INR to Marevan (ΔINR +0.09 [95% CI -0.29 to +0.46], n = 84). Comparison between Teuto generic warfarin and Marevan was inconclusive (ΔINR +0.29 [95% CI -0.09 to +0.68], n = 84). Marevan and União Química warfarin had equivalent therapeutic effectiveness and both could be confidently used for anticoagulation. The comparison between Marevan and TW was inconclusive and does not warrant a statement of equivalence. Our methods are especially important for comparing generic and branded drugs that raise concerns and may be subject of future investigations by regulatory agents. ClinicalTrials.gov NCT02017197.