Known for using in silico disease models and virtual patients to predict drug activity, Entelos Inc. expanded from efficacy analyses into safety analyses by acquiring toxicology expert Iconix Biosciences Inc.

"We're about trying to build better predictive capabilities," said Entelos President and CEO James Karis. Drug developers face two key scientific questions - is the drug efficacious and is it safe - and "we think strategically now we have the leading technologies to address these issues," he added.

Foster City, Calif.-based Entelos will buy Mountain View, Calif.-based Iconix for approximately 12.8 million shares up front.

Two months ago, when Entelos (LSE:ENTL) was trading at 55 pence, the deal would have netted investors in privately held Iconix $14.1 million. But due to a recent decline in Entelos' share price, the up-front payment now is valued at $8.3 million based on Wednesday's closing price of 32.5 pence.

Networking capital will be subtracted from the purchase price, once Iconix closes its books. Iconix also may get up to $25 million worth of additional shares one year after the acquisition, if certain financial milestones are met.

Entelos intends to fold Iconix into its technology and services division, although some administrative and operational positions will be consolidated. As of now, Entelos' primary service offering is its PhysioLab technology, which provides a framework for integrating genomics, proteomics, physiological and environmental data relating to a particular disease to better determine how patients might respond to treatments.

The result is virtual patients that represent various subpopulations that may enroll in a clinical trial. Jill Fujisaki, Entelos' vice president of investor relations, explained that those "patients" represent varying severity of the disease, background health issues such as high blood pressure, and differing underlying causes of a disease.

The model "really helps to explain the variability you might see with responders vs. nonresponders," she said.

Entelos has signed more than 20 partnerships regarding its technology, most with large pharmaceutical companies. Although most projects are confidential, Fujisaki offered an example in which Johnson & Johnson was planning a dose-escalation trial that slowly would cycle through six increasing doses of a new diabetes drug in an effort to prevent using too high of a dose and causing hypoglycemia. Entelos' models were able to show that hypoglycemia was not a concern, and J&J was able to conduct the trial far more quickly using only the highest dose.

In another case, when Pfizer was attempting to screen drug candidates to find a new phosphodiesterase inhibitor for asthma, a list of 50 biological effects needed for a desirable drug candidate was slowing down the screening. Entelos was able to show that only four or five of those biological effects would have any impact in clinical trials, speeding up the selection process.

As of now, Entelos offers virtual patient models in asthma, obesity, Type I and Type II diabetes, rheumatoid arthritis and cholesterol metabolism. Models are in development for cardiovascular and skin sensitization applications, and future expansion is planned in metabolic syndrome, chronic obstructive pulmonary disease, osteoporosis and Alzheimer's disease.

Karis said Entelos also has recently become "more involved in toxicology and safety." Earlier this month the company entered a Cooperative Research and Development Agreement with the FDA's Center for Drug Evaluation and Research (CDER) to develop a computer model of drug-induced liver injury. In its toxicology research, Karis said his company got to know Iconix and became convinced its technology was "best in breed."

Iconix is known for its DrugMatrix chemogenomics system, a database of genomic effects of drugs and chemicals, including the Drug Signatures library, which contains hundreds of sets of predictive biomarker genes. Iconix has several big pharma partnerships, and its DrugMatrix system has been installed at CDER to evaluate voluntarily genomic data submissions.

Entelos predicted the Iconix acquisition will be cash-flow neutral on a pro-rata cash basis going forward and will add $1 million to 2007 recognized revenues. Iconix reported $4.9 million in revenues in 2006, while Entelos reported $21.6 million in the same period. While Entelos' revenues are up significantly from the $2.8 million the company booked in 2005, Fujisaki attributed the increase to revenue booked under GAAP rules for multiyear contracts rather than pure demand.

"There has always been an interest," she explained, "but the pharmaceutical industry is cautious. They've been burned by their overzealous adoption of genomics." She added that many biotech companies are struggling to get by and can't spare resources for predictive modeling, while some are so committed to their one or two programs that they'll "plod on" regardless of what the data say.

Karis agreed that there is an adoption process for any new technology. He likened it to the process of drilling for oil. "For years it was someone who felt good about the spot, and now they use geologic analysis."