On the Mark? Identifying Biomarkers of Target Engagement

Retrospective analyses conducted by major pharmaceutical companies on their drug pipelines have revealed that close to one-fifth of Phase II failures due to efficacy did not conclusively demonstrate adequate target exposure. This statistic emphasizes the importance of identifying target engagement biomarkers, which report on drug-target interactions and the cascade of biological changes that occur when a drug is introduced in a living system, in early study phases.

Target engagement biomarkers help to assess on- and off-target effects and elucidate drug mechanism of action, both directly as a measure of target occupancy and indirectly via measurement of how the biochemical pathway downstream of the target is up- or down-regulated. Biomarker readouts can confirm if the drug is in fact on target, and equally importantly, if drug-target binding creates a biological response. For example, a potential target can be negated if drug binding yields no pharmacological effect. When a target is confirmed to be occupied by the drug and biochemical changes along the target-associated pathway result, you can have confidence in the target and the drug’s viability for clinical studies.

Direct target engagement is easier to measure in experimental phases when the target organ itself can be extracted to confirm target occupancy. This becomes much more difficult in human trials. Indirect biomarkers of target engagement are therefore critical translational tools when moving from preclinical to clinical studies.

Understanding effects along the pathway

Pharmacodynamic (PD) biomarkers can be used as indirect biomarkers of target engagement, providing mechanistic evidence of drug effect following drug-target binding. Small molecule biomarkers are particularly useful as PD biomarkers because they are generated by active biological processes occurring in local tissue but can pass through cellular and biological membranes into the central circulation for non-invasive capture.

This means these molecules can be readily measured in blood to read out how biochemistry in the target organ is influenced by a drug occupying a particular target. By comparing molecule levels in patients before and after therapeutic administration (be it a large molecule or small molecule drug, cell therapy, gene therapy, etc.) we can identify those that change in level due to the presence of the drug.

Knowing the target: a case study

A prominent example involving the successful use of a PD biomarker of target engagement comes from the development and eventual approval of Novartis’ sacubitril/valsartan therapy to treat adults with chronic heart failure. Part of its PARADIGM-HF study, a large outcome trial in heart failure and reduced ejection fraction (HFrEF), evaluated how the therapy affected measures of NT-proBNP in patients with chronic heart failure. NT-proBNP is a known and widely used blood biomarker for both the diagnosis and prognosis of heart failure, shown to correlate with heart failure severity: the more elevated the level, the worse the condition.

Baseline NT-proBNP was measured in 2,080 patients; 1,292 had baseline values >1,000 pg/ml and were reassessed at 1 and 8 months. It was found that NT-proBNP levels decreased by 32% in sacubitril-valsartan treated patients 1 month following the start of therapy, and reductions were sustained with therapy through 8 months. Demonstration of pharmacodynamic effect was pivotal in the study, as the FDA label for the approved therapy indicates that the drug “reduces NT-proBNP and is expected to improve cardiovascular outcomes.”

Pathways unknown: where new efficacy insights await

The value of target engagement biomarkers is clear, but their full potential has yet to be explored. The vast majority of drug targets are still lacking well established biomarkers. If you do not have a known biomarker like NT-proBNP to measure, drug-target interactions are much more difficult to evaluate. This is where greater discovery is needed: to identify new, robust biomarkers that can be applied not just to assess target occupancy, but also the downstream effects on target-associated pathways.

Sapient is addressing this unmet need through high-throughput, untargeted small biomarker discovery at scale. Small molecule biomarkers are as diverse as the array of biochemical changes that can occur from a drug binding to its target, and that is why more rapid, large-scale profiling of these markers (the majority of which are still unknown) is needed. Our goal is to provide more non-invasive means to readily test drug-target hypotheses and accelerate decision-making in early study phases. As studies progress, further target validation will require deeper understanding of the disease itself to elucidate how pharmacological effects overlay and influence disease biology. This is where population-level data becomes extremely valuable: to move beyond mechanistic hypotheses and toward a disease-first approach, based in human biology, that builds confidence drug pharmacology will hit its intended target.

At Sapient, we are helping to bridge that gap by building our own Human Biology Database encompassing data from healthy individuals and patients across diverse disease cohorts. This repository is comprised of data from hundreds of thousands of human biosamples that have been assayed via our rLC-MS systems to capture and measure more than 11,000 small molecule biomarkers per sample. We can mine the database for patterns of disease at a statistical scale that differentiates true biochemistry and biochemical effects from noise, and integrate these insights with target engagement data to build greater confidence in drug program viability as the program progresses.