Inform Genomics said this week that it has signed a collaborative agreement with Tesaro, an oncology-focused biopharmaceutical company, that allows the latter to use Inform's proprietary analysis platform to evaluate the risk of chemotherapy-induced nausea and vomiting in patients who receive certain moderately emetogenic chemotherapy regimens and standard antiemetic therapies.
Inform specifically said that Tesaro will use its Oncology Preferences and Risk of Toxicity, or OnPART, platform, which uses proprietary statistical methods to help oncologists make better treatment decisions for patients who will receive chemotherapy for breast, colorectal, lung, or ovarian cancer. The platform assesses patients' genomic risk for six common therapy-related side-effects — diarrhea, nausea and vomiting, oral mucositis, fatigue, cognitive dysfunction, and peripheral neuropathy.
Inform is also developing a separate second tool, dubbed Transplant, which relies on the same statistical methods and will be able to predict the risk of developing oral mucositis — or mouth sores — in patients about to undergo high-dose chemotherapy regimens prior to stem cell transplants for blood-based cancers.
These products are firsts for privately held, Boston-based Inform, which officially opened its doors in 2012 with the goal of creating products that could eventually become "the standard of care for oncologists in evaluating individual patient risks for side effects of chemotherapy," President and CEO Ed Rubenstein said. Its founders also believed from the outset that "if we were successful in developing those products and if we could demonstrate there was biological validity, not just statistical validity, of the predictions then we would be in a position to also transform drug discovery and development," he added.
He also said that the partnership with Tesaro is a way for Inform to "demonstrate the value of its technology" as it continues to develop and prepare the product for the market. Furthermore, he described the company's underlying informatics platform as a key component of its offerings stating that "[Its] the intellectual property behind all of our discoveries" and is "extremely important for all our activities."
Inform has just completed the first phase of development for both products and will launch a second round of development after it raises sufficient capital, according to Rubenstein. He said that the company plans to have commercially ready products within the next 24 months. They expect the market for both products to include providers, patients, and payers who are looking for improved methods of customizing cancer care and improving outcomes.
The first phase of OnPART development involved a single-center study at the West Clinic in Memphis, Tenn., that analyzed data from 384 patients with breast, colorectal, lung, and ovarian cancer who received three cycles of chemotherapy. The patients were treated with standard chemotherapy regimens: dose-dense doxorubicin, cyclophosphamide, and paclitaxel for breast cancer; 5-fluoururacil- and oxaliplatin-based regimens for colorectal cancer; and carboplatin-plus-paclitaxel-based regimens for lung and ovarian cancer. Patients also filled out questionnaires to provide self-reported records of their symptoms.
For the Transplant product's first phase of development, Inform analyzed data from 153 patients from the Dana-Farber Cancer Institute that had been treated for Hodgkin or non-Hodgkin lymphoma or multiple myeloma, according to a paper describing the study that was presented earlier this year at the American Society of Clinical Oncology conference.
In both studies, the researchers collected saliva samples from patients and analyzed extracted DNA using Illumina BeadChip arrays, which detect about 2.5 million single nucleotide polymorphisms per sample. These SNPs along with the self reported data were then analyzed using the company's proprietary Bayesian-based analysis platform, Rubenstein told BioInform.
The platform, which runs on Amazon Web Services, has a mechanism for filtering out variants that are not linked to the list of six side effects — it uses a statistical inference method called non-informative prior distribution for this step — as well as algorithms to classify variants that are believed to be tied to adverse side effects as having a severe, moderate, mild, or no effect.
By combining treatment regimen information with variant data as well as self-reported symptoms, the company was able to generate SNP network models for each side effect targeted by both products and link them to treatment regimens, Rubenstein said. The company was also able to map the variants in these networks back to the genes in which they occur, enabling further exploration of the biological pathways in which those genes, he said.
Inform Genomics is working on raising capital to fund a next set of studies to further develop and refine its offerings, Rubenstein said. This phase of development, he said, will focus on building the "evidence package that we would need for a successful commercial launch."