In anticipation of launching its flagship Drug-Sensitivity Predictor, the Medical Prognosis Institute over the past month has secured funding for clinical trials, concluded licensing deals with software companies, and hired additional staff, according to a company official.
CEO and President Jesper Drejet told BioArray News last week that the firm is “definitely closer” to commercializing the technology, and that these preparatory steps will support the debut.
“The worst-case scenario is [that the company launches the product] two years from now, but we have several other interesting projects going on that could shortcut this time frame,” he said.
Drejet said that MPI has developed the DSP to guide the treatment of cancer patients by assessing their genetic makeup to determine how they can be expected to respond to different therapies. The DSP uses a custom Affymetrix gene-expression array to identify a patient’s gene-expression signature, while MPI developed the software that uses those results to identify the best possible treatment course.
“Often we see that one hospital has one standard treatment for a certain [type of] cancer, while another hospital has a different standard treatment,” Drejet said. “This technology allows you to treat on a patient-by-patient basis knowing exactly which treatment will be effective. It will give oncologists an indispensable parameter for choosing treatment strategy for patients,” he said.
According to Chief Scientific Officer Steen Knudsen, the DSP is based on a “new algorithm that starts with cancer cell lines and measures their sensitivity to specific drugs.” Based on that model, MPI devised an expression-profiling assay that signifies the sensitivity for each drug. Using that expression profile MPI claims it can then predict the sensitivity of a patient tumor to a drug.
Knudsen said that the firm plans to sell kits that contain the chip, the software, and the reagents. However, the software will link to MPI’s central server in Denmark, and the company will perform the data analysis internally, he said. Last week, MPI signed a deal with software-developer Insightful that will enable MPI to use the company’s S-PLUS 8 Enterprise Server for statistical data analysis and predictive analytics.
“The software for doing this [analysis] is quite complicated,” said Knudsen. “We need a stable software platform that can serve both the purpose of getting US Food and Drug Administration and European approval for a clinical device and also can be used as a server to facilitate the prediction in any hospital lab across the world,” he said.
According to Knudsen, the DSP can hypothetically produce a prognosis for treating all cancers with all drug compounds. Therefore, the firm will have to submit data for every possible cancer as it makes its way through the regulatory processes in the US and EU.
“We expect to have to submit [data for] every cancer to the FDA,” Knudsen said. “We are conducting clinical trials for specific cancers to validate that it works for a specific cancer and specific drugs [and] so far it looks like it works for all cancers and all drugs,” he said. “We are doing clinical trials on lymphoma and leukemia at present. The expectation is to submit in 2010.”
“The market is also huge, because every company that has a drug in the pipeline is concerned about the response rate.”
Drejet said that MPI has obtained DKK 20 million ($4.2 million) in public grants from the Danish Strategic Research Council to support these clinical trials. Some of that funding is specifically for lung cancer research, while some will go to a B-cell lymphoma study where MPI is testing its DSP in a phase III clinical trial, Drejet said.
According to Drejet, the clinical trials will take place in Denmark at a number of hospitals including facilities in Copenhagen, Gentofte, Aarhus, Aalborg, and Odense. For the US arm of various clinical trials, MPI is collaborating with Roswell Park Cancer Center in Buffalo, NY, and the University of Alabama.
“We expect the first results within 24 months,” Drejet said of the clinical trials. “We have [done] phase I, II, and retrospective studies that have validated the technology,” he explained. “For FDA approval, we also have to validate our retrospective data in prospective studies, which we are doing now.”
The intensity of these new rounds of clinical trials has encouraged MPI to add personnel. Two weeks ago MPI hired as senior scientist Jesper Dahlgaard, who has eight years experience working with human DNA microarrays and clinical trials.
Dahlgaard is charged with managing MPI’s clinical trials as well as research collaborations and research contracts. Wiktor Mazin, who has expertise in developing classification algorithms for gene-expression data, was hired as head of bioinformatics.
Because of the nature of the DSP platform and its purported ability to test for all cancers, Knudsen said it is difficult to name one single rival that could challenge MPI in the market. More likely the company will have to compete against firms that have developed prognostic tools on an indication-by-indication basis.
“The competition depends on how narrowly you define it,” said Knudsen. “Just about every biotech and pharma company is trying to find biomarkers that predict sensitivity,” he said. One firm that is developing a somewhat similar tool is Iris Biotechnologies, a Santa Clara, Calif.-based diagnostics company.
Iris CEO Simon Chin told BioArray News
last month that the firm has developed a prognostic tool for breast cancer as well as a software tool called BioWindows designed to help oncologists make treatment decisions (see BAN 3/11/2008
While BioWindows will first be included with a breast cancer test that the company expects to submit to the FDA this year, Iris envisions the software tool as being useful in guiding therapies for other cancers and diseases. For instance, Iris has in its pipeline a CardioChip for detecting and treating heart disease, and a NeuroChip, designed to diagnose degenerative neurological disorders such as Alzheimer’s disease and Parkinson’s disease.
According to Drejet, MPI will be able to cast a wider net in the marketplace by partnering with biotechs or pharmas in drug co-development projects that make use of the DSP. He said that because the DSP can be used to predict tumor response to any compound, not just drugs on the market, it could play a pivotal role in the drug-development process.
“Companies developing drugs can use our Drug-Sensitivity Predictor for their clinical trials,” Drejet said. “The failure rate for drug development is extremely high,” he said. “These would be potentially rescuable if failure is not due to toxicity. We will identify subpopulations within phase III clinical populations where the drug is effective,” he added.
Drejet said that the DSP could also be “extremely useful” at the preclinical level because it can be used to determine whether a given molecule will be likely effective against certain kinds of cancer. It can also be used for pre-selecting a clinical population for a trial, and can also be used to assess toxicity and dose regimens for patients, he said.
Beyond drug rescue, MPI last month decided to open itself up for collaborations with external partners for any indication. “We are open to letting other discovery scientists get access to the technology,” Drejet said. “Certainly we would like to help others achieve similar amazing results like those we have been getting in the cancer field.”
While the firm uses Affy arrays, it touts its software and lab expertise in this new offering. “These arrays have enormous amounts of information,” said Drejet. “We are truly experts in purifying RNA and extracting the maximum amount of information you can get from a tissue or blood sample on a chip for any disease. MPI is just focusing on cancer because that is where there is the greatest unmet need.”