For an initial cost of $14.1m (€10.3m) plus a maximum $25m more - if certain financial milestones are met within a year - the US life sciences company has acquired to acquire Iconix Biosciences, which is based in California.
"We believe the combination of our predictive efficacy models with Iconix's toxicology expertise can create a new paradigm for discovering and developing drugs," said James Karis, CEO of Entelos.
"Drug failures are often due to efficacy or toxicity issues, and while Entelos' predictive disease models address efficacy, Iconix's toxicology databases help to address toxicities."
At the recent Piper Jaffray London Healthcare Conference, UK, Karis explained that the idea for models to predict human responses to drugs came from other industries where it is commonplace to "simulate before you build". In fact, Entelos' founders came from the aerospace and defence industries.
"Currently, it is a trial and error process to try and predict clinical response, but given the high failure rate [of investigational drugs], clearly it doesn't work very well," he said.
Entelos hopes that their 'PhysioLab' models can ensure only the right candidates go to clinical development and so reduce the number of late-stage failures. This, in turn, could drastically cut the average cost of developing a single successful drug - estimated to be around $1.2bn.
The company itself is divided into two parts - Entelos Predictive Sciences which offers its products as part of a service and Entelos Therapeutics, which is the research and development (R&D) and collaboration part of the business. Both Entelos and Iconix have worked with some of the biggest names in pharma: Entelos with names including Pfizer, Johnson & Johnson (J&J) and Novartis; and Iconix boasting alliances with Bristol Myers Squibb (BMS), Abbott Laboratories and Eli Lilly.
Iconix's DrugMatrix system has been installed at the US Food and Drug Administration (FDA) for use by the Center for Drug Evaluation and Research (CDER). There, scientists and reviewers use the technology to perform a range of toxicogenomics applications and is used by the FDA to evaluate voluntarily genomic data submissions.
Karis explained that Entelos' software has also been endorsed by drug regulators: as reported on DrugResearcher.com , the company recently announced an alliance with the FDA to build a model of drug-induced human liver injury to identify patients at risk for developing hepatotoxicity.
The Entelos algorithm uses differential equations to create a mechanistic model of human biology. The resulting large-scale mathematical models can be used in translational medicine, to elucidate the best dose and combination of drugs and which patient subpopulation to use in clinical trials.
As well as the liver model being developed with the FDA, the company has so far developed models for obesity, Type I and II diabetes, cholesterol, cardiovascular/atherosclerosis, skin, haematopoiesis, rheumatoid arthritis, and asthma. They are also developing a model for oncology.
Asked about the future, Karis said that the technology is suitable to create numerous more predictive tools, including ones for metabolic syndrome, hypertension, Alzheimer's, epilepsy and multiple sclerosis.
On the other hand, Iconix's tools are based on its 'DrugMatrix' system, a reference data set of toxicological responses (expression profiles) to drug treatments. These can then be honed into predictive applications.
For every drug that fails for efficacy reasons, there are others that fail due to safety problems. Therefore, although Iconix's systems predict a different drug property, the goal of the two companies remains the same - cheaper, better drugs.