Re-evaluation of oral contraceptive classifications.

Classification of combination oral contraceptives (OCs) by "generation"--typically based on the time of introduction of new compounds--is problematic. The estrogen and progestin components may be present in differing doses, and their interaction may therefore vary from one formulation to another. In addition, assigning the progestin component itself to a particular generation does not account for the unique characteristics of individual progestins within that group. These issues can be clarified by understanding the evolution of combination OCs in relation to dosage changes and by differentiating the pharmacologic profiles of individual progestins, particularly norgestimate, desogestrel, and gestodene. Although all sex steroids have the same basic structure, relatively minor structural modifications can cause dramatic alterations in biochemical activity. Progestational activity is the desired pharmacologic effect of progestins used in OCs, whereas androgenic activity, which increases the potential for adverse metabolic and physical side effects, is undesired. In vitro assays of the ability of various OC progestins to bind progestin and androgen receptors suggest that the androgen/progestin (A/P) binding ratio--a measure of progestin selectivity--is more favorable for norgestimate than for levonorgestrel, gestodene, or desogestrel. In vivo measurements of interactions between various progestins and human sex hormone binding globulin (SHBG) support the concept that generational classification of OC progestins is misleading. These compounds also differ clinically, as exemplified by differential effects on lipoprotein metabolism. In summary, progestins exhibit individually unique biochemical and clinical properties.