Addressing Common Roadblocks In Drug Development: Keynotes At 2016 Discovery On Target

By Dana Barberio

September 28, 2016 | With the plethora of potential drug targets, there is tremendous unlocked potential for new therapeutic drugs. Some of the key aspects of drug development—preclinical target validation and prioritization, as well as the need for new therapeutic modalities—were addressed with novel solutions in compelling plenary keynote talks at the 2016 Discovery on Target Conference* in Boston last week.

A Collaborative Bioinformatics Platform for Identifying and Prioritizing Targets

“85% or more of new drug candidates don’t make it to market,” said Jeffrey Barrett, the founding director of Open Targets and Group Leader at Wellcome Trust Sanger Institute. “Even more worrying is that in the Phase 3 approval stage, when we get to the most expensive part of drug development, there are many failures due to lack of efficacy.”

Open Targets is a public-private initiative that provides part of the solution, providing genome-scale experiments and analysis to build evidence on the validity of therapeutic targets. “GWAS, a study design pioneered 10 years ago, has become one of the workhouses in understanding the genetic basis of complex diseases,” said Barrett. A genome-wide association study (GWAS) involves rapidly scanning markers across the complete genomes of tens of thousands of individuals to find genetic variations associated with a particular disease.

“From the time GWAS first started being used in 2006, we’ve identified over 200 independent associations between genetic variants and risk of inflammatory bowel disease (IBD). So the question is, what do these variants mean biologically and how do we feed them into target discovery?” There are many GWAS studies published but the “scam behind most of these studies is that most of the time we have no earthly idea which… variant is causal, which gene it effects, and certainly not what the mechanism is relating the variant to the gene.” This was the incentive behind founding Open Targets—a collaboration of the European Bioinformatics Institute, GSK, Wellcome Trust Sanger Institute, and Biogen—with the goal of systematically using genome-scale experiments and analysis to identify and prioritize targets.

“We believe that… understanding the underlying biological mechanisms of complex disease and prioritizing potential targets should be done in a pre-competitive way,” he said. “It’s too complicated a problem and requires too much expertise for every company to try to build (this expertise) in-house.”

Open Targets works with data from a variety of sources: GWAS, gene expression, protein expression and function, as well as literature mining for existing drug-target interactions. They apply statistical models that bring all of these elements together in a systematic way. They also do their own GWAS and CRISPR gene editing screens/studies, results of which are published and placed in public databases. Areas of focus for Open Targets are cancer and neurodegeneration, IBD, respiratory disease, inflammation, and immunity. Their bioinformatics platform, targetvalidation.org, is available to any researchers seeking evidence of an association between a target and disease.

Another plenary keynote speaker, Aaron Day-Williams, the Biogen Scientific Lead at Open Targets and the Associate Director and Head of Statistical Genetics at Biogen, discussed Open Targets’ methods of transitioning GWAS results into actual targets. From GWAS data alone, it’s often not known what the causal gene is, or the direction of the effect on the disease.

Using the statistical framework set up at Open Targets that utilizes a variety of data sources and methodologies, they were able to deduce that there are many coding variants across the ABCA7 gene that strongly associate with Alzheimer’s Disease. Next, using beta-amyloid as a biomarker for Alzheimer’s, the group applied statistical evidence to link the genetics, the biomarker and the disease, said Day-Williams. This gene-variant evidence and variant-disease evidence for ABCA7 and Alzheimer’s Disease can be examined in detail on their publicly-available target validation platform, targetvalidation.org.

Enhancing Cell Uptake for Better Targetability

Gregory Verdine, professor at Harvard University and Harvard Medical School, in the final keynote presentation, discussed his uniquely engineered cell-penetrating mini-proteins that combine the best attributes of both biologics and small molecules.

“The fact is that the vast majority of human proteins are not targetable by small molecules,” said Gregory Verdine, professor at Harvard University and Harvard Medical School in his keynote presentation. No matter how good you are at drug discovery, you slam right into the druggability problem.” Over the last 15 years, Verdine has been looking at modalities to fill the gap between the two.

The inherent issue with small molecules is their small contact area and limited adhesive powers, and as such they require engulfment by the things they target to maximize utilization, said Verdine. “Antibodies and small molecules are not going to go away, but over the next decade there will be an explosion of new modalities, because demand has risen and companies and academics are rising up to the challenge,” predicted Verdine.

“Our future lies in the exploitation of endocytic vesicle cell trafficking,” in which molecules are engulfed by a cell and actively transported. “It’s one of the most underdeveloped areas of biology relative to how important it is to the future of medicine. The cell-penetrating mini-proteins are just one class that we think will play a role in this area. They combine the accessibility of small molecules… with the broad targeting ability of biologics,” he said.

The mini-proteins have been engineered by the Harvard researchers with an all carbon “alpha helix stapling system,” which adds a long serum half-life. The engineered peptides get into the cell cytoplasm through endocytic vesicle cell trafficking, escaping endosomes (avoiding degradation downstream), and freely crossing the nuclear membrane. They have been shown to target transcription factors. The first engineered ‘stapled’ peptide is now in Phase I clinical trials. At Harvard, they continue to evaluate targets and optimize the chemistry for second generation ‘stapled’ peptides.

* Discovery on Target hosted by Cambridge Healthtech Institute, September 19-22, Boston, Mass.