Selecting and validating biologic markers for drug development.

Biochemical and clinical markers are critical for efficient development of new molecular entities. Biologic markers of drug effect, sometimes referred to as "surrogate" markers, are used when such clinical outcome measures as survival are substantially delayed relative to predictive biochemical changes or clinical effects of the new molecular entity. Biologic markers have generally been used for early-phase decision-making studies and accelerated regulatory approvals for much-needed drugs to treat cancer and acquired immune deficiency syndrome (AIDS). The rationale for these two uses of biologic markers is different and therefore the foundation required for establishing and validating each may be different. The theoretical foundation required for a marker that will be used to justify the regulatory approval of a treatment for a life-threatening disease should be greater than the required for an early decision-making study with an angiotensin II antagonist that will be used to treat mild to moderate hypertension. Use of CD4 counts as "surrogate markers" for prolonged survival was inappropriate. In contrast, changes in angiotensin-II concentrations and other renin-angiotensin system biochemical markers, observed for the first time in a study in humans, with a purported angiotensin-II receptor antagonist indicate that the new molecular entity is working as hoped. This is a good decision-making tool, because theory indicates that these changes should lead to reduced blood pressure, which is a predictive "surrogate" for reduction in subsequent cardiovascular events. Surrogate, biologic markers should be used only if they have a rational theoretical basis, are proven in preclinical or clinical experience, and are measured with validated methods. Different validation-acceptance criteria for decision-making markers compared with markers used for regulatory approval must be prospectively acknowledged and delineated.