Development of in vitro-in vivo correlation for extended-release niacin after administration of hypromellose-based matrix formulations to healthy volunteers.

Development of in vitro-in vivo correlations (IVIVCs) for extended-release (ER) products is commonly pursued during pharmaceutical development to increase product understanding, set release specifications, and support biowaivers. This manuscript details the development of Level C and Level A IVIVCs for ER formulations of niacin, a highly variable and extensively metabolized compound. Three ER formulations were screened in a cross-over study against immediate-release niacin. A Multiple Level C IVIVC was established for both niacin and its primary metabolite nicotinuric acid (NUA) as well as total niacin metabolites urinary excretion. For NUA, but not for niacin, Level A IVIVC models with acceptable prediction errors were achievable via a modified IVIVC rather than a traditional deconvolution/convolution approach. Hence, this is in contradiction with current regulatory guidelines that suggest that when a Multiple Level C IVIVC is established, Level A models should also be readily achievable. We demonstrate that for a highly variable, highly metabolized compound such as niacin, development of a Level A IVIVC model fully validated according to agency guidelines may be challenging. However, Multiple Level C models are achievable and could be used to guide release specifications and formulation/manufacturing changes.