CombiMatrix last week said that it expects to have a microRNA-based cancer-screening test on the US market around the middle of next year, about five months after announcing that it had begun developing the product (see RNAi News, 10/9/2008).
And while the company recently released data showing that it can use miRNA signatures in serum to accurately distinguish healthy individuals from cancer patients, more work will need to be done before the test can meet a key goal of differentiating different types of cancer.
But it remains to be seen whether CombiMatrix can fund that work; the company this week warned that it only has enough funds to continue operations into the fall if it is unable to raise additional capital.
According to CombiMatrix CEO Amit Kumar, the so-called Comprehensive Cancer Array is designed to confirm the presence of cancer by analyzing miRNA-expression patterns in a patient's blood and identify the organ system in which the malignancy is growing.
Initially, he said, the company expects to market the test as a screen for cancers of the prostate, colon, ovaries, breast, and lung, but additional cancers may be added to the panel "within a few years."
Like Rosetta Genomics and Exiqon, two other companies developing cancer tests based on miRNA signatures, CombiMatrix plans to market the CCA through an in-house Clinical Laboratory Improvement Amendments-certified laboratory, which will require regulatory approval from state health agencies.
Unlike those other companies, however, CombiMatrix's test won't be designed to diagnose a disease, but only to indicate whether a patient is a candidate for additional cancer testing.
Kumar likened the CCA to existing prostate cancer screens that measure blood levels of prostate-specific antigen and are routinely given to men at risk for the disease.
"What we're doing … is developing a PSA test across multiple cancers," he said. Should a patient's test indicate the presence of cancer, "a physician will recommend a subsequent workup … [to] do the final diagnosis."
Looking to tap into a market that it estimates could be as high as several billion dollars annually in the US alone, CombiMatrix hopes that the CCA will become a go-to test for "all individuals [aged] over 40, those with a history of cancer in the family, as well as those who have been tested to have a high genetic risk of contracting cancer," Kumar said in a statement.
And also as with the PSA test, the non-invasiveness of the CCA is expected to help drive its adoption, Kumar told RNAi News this week.
Of the roughly four million PSA tests performed each year in the US, about 1 million will indicate the presence of cancer, he said. Still, only about 200,000 of those hits correctly reflect the presence of prostate cancer.
"Even though it has a lot of false positives, people do [the PSA test] because it's simple — you just have to do a blood draw," Kumar said.
[ pagebreak ]
Still, because the CCA only screens for cancer, it is not expected to fetch the kind of premium that miRNA-based diagnostics are getting. For instance, Rosetta Genomics is selling its first miRNA diagnostics for more than $3,000 each (see RNAi News, 12/31/2008). These tests are designed to differentiate lung cancer from mesothelioma, determine the source of cancers of unknown primary origin, and differentiate squamous from non-squamous non-small cell lung cancer.
CombiMatrix, meanwhile, expects the CCA to sell for between $250 and $300 per test, Kumar said.
"Because it's a screening test, and because we anticipate the market opportunity for this is tens of millions of individuals per year, we have to bring it in for a modest cost," he noted. "Our current array-based tests cost anywhere from $1,000 to $2,000, [but] that's just way too high for a screening test."
He added that the CCA may initially cost "slightly higher" than anticipated, but that the price is expected to come down to the $250-$300 range once CombiMatrix has scaled up operations related to the test.
Last week, CombiMatrix also released data demonstrating that the CCA is capable of detecting cancer using serum miRNA signatures, although the company did not provide data showing that the screen could distinguish between cancer types.
The data were presented at Cambridge Healthtech Institute's 16th International Molecular Medicine Tri-Conference held in San Francisco last month, CombiMatrix said.
In the study, a microarray containing all human miRNA sequences contained in the Wellcome Trust Sanger Institute's miRBase was used to analyze serum samples from 36 individuals, either healthy or with prostate, colon, ovarian, breast, or lung cancer.
According to a poster presented by CombiMatrix at the CHI event, each microarray contained three probes for each of 547 human miRNA sequences: antisense to wildtype; double-mutated antisense; and sense-negative control probes, which were included if the corresponding sequence was not found in the miRNA database. In addition, the company included as controls four sheep, three C. elegans, and two human sequences.
Company researchers were able to show that enough miRNAs were present in one milliliter of serum to detect miRNA-expression patterns without the need for amplification techniques, the poster states. Further, the investigators were able to use the patterns to "discriminate correctly between normal and cancer-patient samples."
CombiMatrix did not, however, present data showing that the CCA could distinguish one type of cancer from another — a major hurdle that needs to be overcome if the test is to be successful, Kumar noted.
With the CCA, the company aims to be able to answer three questions, he said: Does an individual have a cancerous growth of some sort in his or her body; in what organ system is that growth; and at what stage is the malignancy?
"The data that we presented demonstrates that we have a very high confidence that we can answer the first question," he said. "The second question … we didn't demonstrate … completely, but we did demonstrate that there are approaches to making those calls."
Specifically, the CombiMatrix team was able to distinguish colon cancer from other cancers using K*, an instance-based learner, and Bayes network classifiers. However, they were unable to distinguish prostate cancer from the overall cancer sample pool.
Still, "by extending this work with more samples, it will be possible to discover more sets of attributes that will allow accurate classification of other diseases," the poster notes.
[ pagebreak ]
As for the CCA's ability to determine how far a cancer has advanced, "we're going to have to do some additional analysis, which we are going to be doing in the next few months," Kumar said.
CombiMatrix hopes that its test will be able to identify the presence of cancer in its earliest stages, but "that's the question we haven't answered yet," and which may not be fully answered until after the CCA has reached the market, he added.
Getting the test to market will require it be evaluated in additional patients with a variety of different diseases, Kumar explained. "One of the things we want to confirm is that patients who have certain infectious diseases [or] patients who have [had] surgeries don't look … like cancer patients" in terms of their miRNA-expression profiles.
Ultimately, CombiMatrix will need two datasets: one required under CLIA regulations so that the test can be officially launched, and one that can be used in marketing so that the test can gain acceptance by physicians, he said.
The company, he noted, is capable of conducting this work on its own. But when it comes to actual commercialization, a partner will likely be needed.
"We're a small company, and the market opportunity for this is incredibly large, so we'd be looking for a number of partners," Kumar said. "We've already had discussions with some [potential partners], and we'll see where those go."
However, it is unclear whether CombiMatrix has the financial resources to fully develop the CCA.
This week, the company reported that its fourth-quarter 2008 revenues decreased 37 percent, and that it would need to raise additional funds since the $9.1 million in cash, cash equivalents, and marketable securities it had at the end of 2008 would only support operations until September.
Total revenues in the quarter fell to $1.2 million from $1.9 million in the fourth quarter of 2007. The firm's net loss for the quarter widened to $4.1 million, or $0.66 per share, from a net loss of $3.5 million, or $0.58 per share, for the same period a year earlier.
CombiMatrix's R&D spending rose 18 percent to $1.3 million from $1.1 million in the year-ago period, and its SG&A spending decreased slightly to $2.4 million from $2.5 million year over year.
In a statement, Kumar called 2008 a "transformative year" for the firm as it "launched a record number of products in our diagnostics business, including the introduction of our first prostate cancer array-based test."
However, CombiMatrix warned that for it to "continue as a going concern beyond this point and ultimately to achieve profitability, we will be required to obtain capital from external sources, increase revenues and reduce operating costs.
"As a result, the anticipation that we will be required to obtain additional financing in the foreseeable future raises substantial doubt about our ability to continue as a going concern beyond September 2009," it said.