The recently released guidance on the use of clinical pharmacology data to support biosimilars suggests that in some cases, pharmacokinetic and pharmacodynamic data may be sufficient to complete the clinical evaluation of similarity. However, the bar for this is set very high; requiring a large database (either publically available or generated by the biosimilar sponsor) with detailed information on the pharmacokinetic and pharmacodynamic profiles of the reference product and an armamentarium of scientifically-justified and validated methods to measure these parameters. The pharmacokinetic and pharmacodynamic assays used to establish similarity need to be sensitive, specific, reliable, and robust, able to distinguish a wide range of levels of activity, and highly relevant to the clinical outcome. For example, pharmacokinetic binding assays should be specific to the region of a molecule responsible for the substance’s pharmacological activity and activity assays used to assess pharmacokinetics and/or pharmacodynamics should assess an endpoint highly relevant to the substance’s clinically important pharmacological activity. Moreover, these assays should be conducted in individuals who are demographically similar to the patient population, but healthy so that concomitant drugs or disease do not affect outcomes, yet able to tolerate doses in the steep part of the dose-response curve so that maximal sensitivity to distinguish potential differences is possible. Further, for those indications where activity lags behind administration, assessment of pharmacodynamics should be timed to coincide with clinically relevant activity. Deviations from such ideal conditions are permissible but require scientific justification and validation.
Ultimately, each facet of the pharmacokinetic and pharmacodynamics assessments should be chosen to be maximally sensitive to detect potentially clinically meaningful differences between the biosimilar and the reference product. Should these assays provide a sufficiently wide dynamic range so that the FDA agrees that the results demonstrate that a fingerprint-like similarity between products exists, full clinical evaluation may not be required. If a satisfactorily fingerprint-like similarity is not achieved, yet dissimilarity is not proven, the data obtained through the clinical pharmacology assessments will guide the design of additional targeted animal and/or clinical studies to resolve remaining uncertainty.
Carrie Rabe is a Senior Consultant at The Weinberg Group, the world’s leading food and drug consulting firm. If you have any questions or thoughts on this blog post or others, please reach out to us by email.