Skip to main content
Premium Trial:

Request an Annual Quote

Bayer Spin-off Teams with Physiomics to Develop Virtual Tumor for Drug Dosage Testing


Bayer Technology Services, a 2,300-employee technology development subsidiary of Bayer AG, has tapped a tiny UK-based software firm as its partner in a new computational systems biology project.

Last week, BTS, based in Leverkusen, Germany, said it had entered into an agreement with Physiomics — a 10-person biosimulation software shop in Oxford — to co-develop technology for clinical response prediction. The companies will integrate PK-Sim, a physiology-based pharmacokinetic simulation tool from BTS, with Physiomics’ SystemCell technology, which models the mammalian cell cycle.

The goal of the open-ended collaboration is to bridge the distance between organism-level modeling and cellular-scale modeling — one of the biggest challenges in computational systems biology, which has struggled for years with reconciling “top-down” and “bottom-up” approaches to modeling biological systems.

Jörg Lippert, a senior scientist in the BTS computational biology group, said that his team had been looking to extend its pharmacokinetic modeling software to predict the cellular response to drug candidates for about two years. After “discovering” Physiomics, he said, the collaboration seemed obvious: “They have detailed models of cellular behavior and the action of molecular interactions within the cells, and together the two technologies are suited to predict the outcome of a pharmacological therapy.”

For Physiomics, the match made sense from the opposite side of the fence. “We have created a mammalian cell cycle model that is designed to be used to assess how much drug you require for a biological effect on cell cycle components,” said John Savin, Physiomics CEO. “We can help a company find the right dose for a drug — but that’s a dose at the tumor site. How much drug do they actually need to give in terms of an oral pill or an [intravenous] administration in order to get sufficient drug at the tumor site?”

That, Savin said, is where the PK-Sim technology comes in, “because Bayer has got this very sophisticated model of the human body. … So you can test the drug out in preclinical testing, and you can get a very good idea before you’ve even entered clinical trials about how it will perform in a human patient.”

Physiomics is predicting “how the drug actually affects the cells,” he said. “And Bayer is providing how much drug actually gets to the right place in the body.”

The companies said they are already working together at several undisclosed customer sites to apply this combined modeling approach to specific pharmaceutical development problems.

Lippert said that the companies have targeted three primary applications for the service: early-stage pre-clinical planning; dosage optimization; and trouble-shooting for failed or failing trials. In silico simulations can help save drug companies time and money at each of these steps, he said.

In the case of pre-clinical planning, for example, the approach can be used to predict the metabolism of a drug within a tumor cell “to decide questions like what type of cancer can be affected by your drug,” Lippert said. Likewise, computer-based simulations of different dosage levels, frequencies, and durations could help eliminate costly animal-based testing. In addition, he said, the approach is applicable to dose individualization and personalized medicine because “you can use the models to estimate the population effects of an application of your drug to a larger population of patients.”

Ultimately, Savin said, the companies see their approach as a step toward in silico drug design. “When you’re building a new aircraft, you don’t build hundreds of prototypes and just see which one crashes the least, which is the way that the pharma industry develops drugs,” he said. “You build models and you design the thing in the computer.”

Convincing the Customer

But despite the theoretical promise of in silico clinical response prediction, both Lippert and Savin acknowledged that the approach is still a hard sell.

“Most companies have very few people — even the big companies only have a few people who really understand this particular science at this time,” Savin said, “and the smaller companies don’t have anybody at all.” Simulation projects are highly specialized service engagements, he said, requiring a deeper commitment than an off-the-shelf software purchase. Many firms are simply unwilling to take that kind of risk with a relatively unproven technology.

Lippert said that BTS and Physiomics are only working with a “small number” of customers so far. “With any emergent technology like this,” he said, “you always have to convince the customer that it’s working and has some value.” Even the PK-Sim component, which is an out-of-the-box package, has a limited customer base because of the level of knowledge required to operate it.

But Savin is optimistic that the combined offering may be of far more interest to pharmaceutical customers than either piece alone. “We’ve found in test marketing that having the two products linked is very powerful,” he said. “People are much more interested in having the combination of things, because we direct this to answer specific questions around the clinical process, for designing trials, [and for] troubleshooting trials that fail for some reason. And those are the key questions that companies — that senior management — has to deal with.”

BTS and Physiomics also face competition from other biosimulation firms like Entelos, Gene Network Sciences, and Genomatica, who are trying to convince the same limited customer base to hire them to answer the same kinds of questions. Savin said that Physiomics’ approach is similar to what the former Physiome Sciences was doing before it was acquired by Predix Pharmaceuticals.

Like Physiome, he said, “we have a bottom-up approach — we focus on the detail of the biological system.” Entelos, he said, “has very much a top-down approach — they’re looking for a much more holistic approach, and yes, they can build very intricate detail into their models. But what Entelos can’t do and what Physiome never did was to integrate those things with a drug delivery model. I think we’re the only company or alliance in the world that has the two aspects of the puzzle.”

Physiomics has another advantage in having the resources of Bayer in its corner. BTS generated revenues of €720 million ($877 million) in 2003. Savin called the collaboration “an excellent launchpad” for his “terribly tiny” company.

Tom Patterson, an Entelos founder, told BioInform that he was not familiar with Physiomics, but said that the companies’ approaches did seem to be somewhat similar at first glance. One feature of Entelos’ simulation platform that appeared to be lacking in Physiomics’ was a means of dealing with “knowledge gaps” in the modeling process, he said.

“It’s good to have competition,” Patterson said, adding that Entelos has been working with pharmaceutical companies on biosimulation products since 1996. “I think we’re still the only people making money from customers — as opposed to SBIR grants — in this space,” he said.

— BT

Filed under

The Scan

Expanded Genetic Testing Uncovers Hereditary Cancer Risk in Significant Subset of Cancer Patients

In Genome Medicine, researchers found pathogenic or likely pathogenic hereditary cancer risk variants in close to 17 percent of the 17,523 patients profiled with expanded germline genetic testing.

Mitochondrial Replacement Therapy Embryos Appear Largely Normal in Single-Cell 'Omics Analyses

Embryos produced with spindle transfer-based mitochondrial replacement had delayed demethylation, but typical aneuploidy and transcriptome features in a PLOS Biology study.

Cancer Patients Report Quality of Life Benefits for Immune Checkpoint Inhibitors

Immune checkpoint inhibitor immunotherapy was linked in JAMA Network Open to enhanced quality of life compared to other treatment types in cancer patients.

Researchers Compare WGS, Exome Sequencing-Based Mendelian Disease Diagnosis

Investigators find a diagnostic edge for whole-genome sequencing, while highlighting the cost advantages and improving diagnostic rate of exome sequencing in EJHG.