Risk assessment is a continuous process beginning right from discovery through the life of the drug. While we collaborate closely with our discovery, non-clinical and clinical pharmacology colleagues in understanding safety issues, we, as physicians, constantly bring in the clinical perspective. We present here a few examples of how we undertake a safety review for a program in development.
Early in development, we look for potential adverse effects based on the physico-chemical characteristics of the drug. For example, a lipid soluble molecule is likely to cross the blood brain barrier. We proactively look for grey (seizures, drowsiness, coma) and white matter (spasticity) symptoms. A large molecule is likely to be excreted by the hepatobiliary system (not renally excreted) and liver dysfunction can be a potential contraindication.
We next look at the primary pharmacology of the drug which relates to its activity on the desired therapeutic target. An understanding of the downstream pathways from the target is critical to building a safety profile for the drug. An on-target effect may be associated with both therapeutic and adverse effects. A beta agonist intended as a bronchodilator might adversely cause tachycardia. The secondary pharmacology of the drug relates to its off-target (usually adverse) effects. A low selectivity index (ratio of the off-target to on-target half-maximal inhibitory concentration for 50% [IC 50]) decreases the therapeutic window for the drug and could be potentially toxic. Recently, we postulated (based on the primary and secondary pharmacology) that the safety profile of an orally bioavailable tyrosine kinase inhibitor would not be unlike that of a monoclonal antibody, and we have not been proved wrong yet.
Mortality is the most feared adverse outcome. Mortality can only result from respiratory or cardiac failure/arrest or brain death. Safety pharmacology studies evaluate the effect of therapeutic and supra-therapeutic doses on the respiratory, cardiovascular, and nervous systems. In most situations, drugs with a safety pharmacology signal are often killed before they get to us. In one instance, after prolonged non-clinical exposure, central nervous system (CNS) toxicity was noted (subacute myelo-optic neuropathy) while a phase 2 study was ongoing. The clinical study was put on a regulatory hold. Our safety diligence and exposure response analysis could not rule out the possibility that it was not a drug effect. The program was killed.
Non-clinical toxicology and toxicokinetics can provide crucial information. We are wary when PK parameters in humans begin to approach those seen at no-observed-adverse-effect-level (NOAEL). On the other hand, in a recent development program for a topical where systemic concentrations were three orders of magnitude lower than the NOAEL concentrations, we were able to convince the biopharma not to have a data safety monitoring board, and regulators that safety lab monitoring was not required for an open label extension.
In early clinical development, we strongly recommend a Sponsor unblinded trial even as the patient and investigator remain blinded. This is the most efficient way to assess safety. During a blinded phase 2 and 3 study, we might prospectively identify adverse events of interest or examine Standard MedDRA queries (SMQs). We suspect a drug related adverse event (AE) if the overall blinded AE rate exceeds the background rate for the disease. For lab data, we plot baseline versus max and last values to detect signals.
For gauging the risk for embryo-fetal toxicity and contraception recommendations, we typically follow the guidelines of the Heads of Medicines Agencies (https://tinyurl.com/jpzj9272). In previous blogs, we have covered post-marketing risk assessment from maternal exposure and causality assessment of birth defects.
As always, we have learnt a lot along the way, and needless to say, love both safety due diligence and risk management.
For more blogs, please visit: https://www.rxmd.com/insights