Fresh off a CHF 2.6 million ($2.85 million) Series A funding round, Swiss proteomics firm ProteoMedix is moving to validate blood-based protein biomarkers for diagnostic, prognostic, and companion diagnostic tests for prostate cancer.
The company, which spun out of the lab of Swiss Federal Institute of Technology researcher Ruedi Aebersold in 2010, is currently developing assays to be used in the validation process and aims to bring its first test to market by 2015, co-founder and CEO Ralph Schiess told ProteoMonitor.
The Series A funding, which ProteoMedix closed in January, will go primarily toward developing ELISAs to validate four proteins to be used in a diagnostic test for prostate cancer, he said. The round was led by Swiss venture capital firm Altos Venture and included private investors from the BioValley Business Angels Club BioBAC and Zurcher Kantonalbank.
The diagnostic is intended as an improvement over prostate-specific antigen, currently the standard for prostate cancer screening. PSA testing suffers from poor specificity, leading to large numbers of false positives and unnecessary biopsies.
The ProteoMedix panel, Schiess said, improves the accuracy of PSA testing by including measurements of four cancer-specific proteins. In a 143-patient study published last year in the Proceedings of the National Academy of Sciences, the company's panel detected prostate cancer with an accuracy of 84 percent, compared to 70 percent for PSA alone (PM 2/11/2011). The most significant improvement was in specificity, with the ProteoMedix panel offering 79 percent specificity compared to 45 percent for PSA alone.
According to a company presentation, such an increase in accuracy could reduce unnecessary biopsies by up to 75 percent, saving $2 billion annually in the US alone.
The firm plans to position the panel as a follow-up test for patients with positive PSA results, Schiess said. "In the future you would still use PSA as a routine screening test, but since the false positive rate [for PSA] is very high, you would use our test as a second-line test."
Researchers in Aebersold's lab discovered the prostate cancer markers via label-free mass spec on a Thermo Scientific LTQ-FT instrument. The lab also performed some initial validation work using selected-reaction monitoring mass spec. Such platforms, though, remain impractical for large validation efforts or clinical implementation, Schiess said, explaining why ProteoMedix was converting to an ELISA format.
"It's simply not feasible right now to have a diagnostic test that is widely used on mass spec," he said. "No platform is really adapted to usage in large sample cohorts. Right now the throughput isn't there for large validation sets or standard clinical use."
ProteoMedix has secured access to several hundred prostate cancer samples that it will use for validation. It is also launching new prospective studies as part of its validation effort.
The company plans to pursue CE marking for the test, Schiess said, noting that initially "the ideal partners would be local diagnostic laboratories that do routine testing for urologists." Using data from such testing, the company would then possibly pursue a deal with a "big diagnostic company like Roche, Abbott, or Siemens," he said.
In addition to the diagnostic test, ProteoMedix is working on a prostate cancer prognostic that could distinguish between aggressive and slow-moving forms of the disease.
"That's actually a very important issue right now," Schiess said. "Even if you have a better [diagnostic] test, still sometimes you don't know what to do with a person with prostate cancer. It's very invasive to remove the whole prostate, and there are lots of side effects, so one might think of doing surveillance instead of surgery, and so it's very important to know which cancers are aggressive and need treatment right away and which ones could be monitored by non-invasive tests."
On this front, the company will face competition from several genomic assays for prostate cancer prognosis. Myriad Genetics launched its Prolaris prostate cancer recurrence test in 2010 and Genomic Health is developing a similar test that it plans to launch in 2013. Both tests use gene expression measurements to assess the risk of cancer recurrence.
ProteoMedix is also exploring use of several markers as a companion diagnostic for identifying patients likely to respond to mTOR inhibitors. At the 2011 American Society of Clinical Oncology annual meeting, ProteoMedix researchers presented data from a 37-subject multicenter phase II trial identifying protein markers predictive of prostate cancer patient response to Novartis' mTOR inhibitor everolimus, marketed as Afinitor. According to the presentation, two proteins in combination predicted whether patients reached progression-free survival at 12 weeks with an accuracy of 85 percent.
ProteoMedix is currently testing the markers with additional inhibitors and has spoken to several pharma companies about this work, Schiess said, but it isn't involved in any formal collaborations.
These companion diagnostic markers, he noted, emerged from the experimental design of Aebersold's original discovery work, which used mouse models containing PTEN-knockout mutations – which are involved in roughly 80 percent of prostate cancers – to identify potential markers.
"We found markers that not only correlate with cancer, but also very specifically with the activity of [the PTEN] pathway," he said. So we have markers that specifically show which patients have an overactivation of that pathway and which don't. So if you have a drug … that is targeting this pathway up or downstream, [these markers] could predict" which patients will respond.
Moving forward, ProteoMedix plans to maintain a focus on biomarker validation, as opposed to discovery, Schiess said, pointing to its deal with ETH Zurich to license the prostate cancer markers as a guide.
"Right now there is a lot of discovery ongoing, and there are a lot of markers out there," he said. "We believe these markers work and so our goal as a company is to further develop [them]."
He noted, though, that the company might in the future license additional markers from Aebersold's lab, which, he said, is currently looking into several other diseases including pancreatic, ovarian, and colorectal cancer.
"[Aebersold] is a co-founder and [while] we don't have an exclusive right to in-license these markers, one [reason] ProteoMedix was founded was to bring these markers to clinical applications," Schiess said. "Therefore we would certainly be interested in developing further markers [from Aebersold's lab] It's not that we have some sort of legal situation that says markers [from Aebersold's lab] automatically go to ProteoMedix, but we are very close."
Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.