Beleaguered by declining revenues and impatient investors, Ciphergen Biosystems last week said it will announce a pivotal diagnostics deal before the end of June.
In addition, the company, which last month said it would "broadly" restructure its sales staff, finally disclosed that the restructuring includes turning a dozen field scientists in North America and Europe into salespeople.
Announcing its first-quarter earnings, Ciphergen said that revenues declined by around 50 percent year over year which it had predicted in April and blamed the decrease on lackluster sales and on a 15-percent reduction in overall headcount, including sales people, that has been ongoing since Dec. 31, 2004.
Ciphergen said that total revenues for the three months ended March 31 shrank to $6.6 million from $13.3 million during the first quarter of 2004. The numbers were in line with the company's revised revenue forecast released last month (see ProteoMonitor 4/8/2005).
"Our belief is that [the] perturbation [in sales staff] was one of the main reasons we had the reduction in Q1," said Ciphergen CEO Bill Rich during an earnings call. "Most of these sales were not lost; they've fallen into Q2, and we do have a backlog of orders for Q2."
With $32 million left in cash and cash investments, Ciphergen has recently taken steps to reduce its operating expenses. For example, the company has reduced its number of employees by approximately 15 percent since Dec. 31, 2004.
One of the big stories for Ciphergen this week is its restructured sales force, which coincided with first-quarter sales and marketing expenses that fell to $5.3 million from $6 million year over year. As part of the reorganization, Ciphergen this month turned a number of technical staffers into sales people. Specifically, the company turned into salespeople between six and eight US field scientists, one Canadian field scientist, and three European field scientists.
"These younger field scientists are very enthusiastic about getting more involved in the sales process, and they can translate their enthusiasm directly into sales," Rich said during the call.
In addition, Ciphergen reduced its research and development expenses from $5.8 million during the first quarter of 2004 to $3.5 million during Q1 of this year. It also reduced its general and administrative expenses slightly from $3.7 million during Q1 of 2004 to $3.5 million during Q1 of this year.
Looking to the second quarter, Ciphergen's chief financial officer Matthew Hogan said the company is "continuing to work hard on forming a commercialization partnership," to market its diagnostic product or products.
"We believe [the partnership] will happen in the second quarter, but I can't go into the details," said Hogan. "We believe it will be a significant announcement for Ciphergen."
The Ovarian Cancer Diagnostic
The other big story for Ciphergen is its diagnostic program. The product Ciphergen officials talked about most during the company's earnings call last week was a diagnostic for ovarian cancer. Last August, researchers at Johns Hopkins University published a paper in Cancer Research describing a panel of three protein biomarkers that, when combined with the standard CA125 ovarian cancer biomarker, could detect early-stage ovarian cancer in 74 percent of women study patients who had the disease (see PM 8/20/2004).
Since then, researchers collaborating with Ciphergen have been further testing the protein biomarker panel in a validation study involving 1,500 samples (see PM 11/5/2004). Preliminary analysis of 436 samples in the validation study showed that the biomarker panel combined with CA 125 scored between .88 and .91 out of a perfect score of one, in terms of specificity and sensitivity.
Currently, Ciphergen and its collaborators are developing prospective clinical trials for its ovarian cancer test that may be used to support a submission for approval from the US Food and Drug Administration. The company recently penned a collaboration with MD Anderson Cancer Center (see PM 2/18/2005) in which MD Anderson will provide Ciphergen with ovarian clinical samples to be analyzed using Ciphergen's DeepProteome and PatternTrack suite of new proteomic tools.
But Ciphergen faces tough competition on the ovarian cancer front. Last June, Correlogic, in collaboration with the North Shore-Long Island Jewish Research Institute in Huntington, NY, launched a two-year validation study for its OvaCheck proteomics pattern-based ovarian cancer test (see PM 6/25/2004).
Correlogic's test, developed in collaboration with LabCorp and Quest, is based on a study led by Lance Liotta and Emanuel Petricoin, which was published in the February 2002 issue of Lancet. The study, which involved 50 women with ovarian cancer and 66 unaffected women or women with non-malignant disorders, showed that a proteomic pattern was able to discriminate ovarian cancer with 100 percent sensitivity and 95 percent specificity.
Correlogic's ovarian cancer test differs from Ciphergen's in that is based on a proteomic pattern created by five proteins which have not been identified. Ciphergen's ovarian cancer is based on a panel of three protein biomarkers that have been identified and characterized. The three proteins are apoplipoprotein A1, a truncated form of transthyretin, and a cleavage fragment of inter-alpha-trypsin inhibitor heavy chain H4.
Aside from Correlogic, Ciphergen also faces competition on the ovarian cancer front from PerkinElmer and Predictive Diagnostics, which, like Correlogic, are developing pattern-based assays for ovarian cancer.
PerkinElmer entered the ovarian cancer competition in May 2004 with its collaboration with the NCI-FDA Clinical Proteomics program, then led by Liotta and Petricoin (see PM 6/11/2004). The company recently collaborated with Predictive Diagnostics to discover a panel of biomarkers for diagnosing Alzheimer's disease, and it is likely that the two companies will collaborate in developing an ovarian cancer diagnostic.
In terms of groups that are developing ovarian cancer diagnostics based on identified protein biomarkers, Ciphergen is dealing with rivals or potential collaborators from various government institutions and academic groups.
Last week, the Nevada Cancer Institute announced that a group of its researchers, led by David Ward, the deputy director of the institute, has developed a blood test for ovarian cancer that showed 95 percent sensitivity and 95 percent specificity in a study involving 200 healthy women and women with ovarian cancer. The test, reported in this week's issue of the Proceedings of the National Academy of Science, is based on a panel of four protein biomarkers: leptin, prolactin, osteopontin and insulin-like growth factor-II.
Ward said he and his research group have no plans yet to commercialize their test, largely because its sensitivity and specificity are not high enough for it to be used as a general population screening test.
The researchers plan to improve the sensitivity and specificity of their test by evaluating other proteins they have identified but not yet tested in blind studies. Ward said his group is open to collaborating with others who have identified additional ovarian cancer markers.
"We are willing to collaborate… [with] both academic institutions and commercial enterprises in order to generate an optimal test as quickly as possible, as this, hopefully, will save lives," he said.
Asked how the various ovarian cancer tests under development measure up, Andrew Berchuck, a professor of gynecological oncology at Duke University said that all of the tests being developed have a long way to go before they can be used as a screen for early stage ovarian cancer. A test that is suitable for ovarian cancer screening should have a specificity of at least 99.5 percent, Berchuck said. At that level of specificity, a test would give 10 false positive results for every true positive.
"The main thing is to really have a good handle on how often the test shows up positive in women who don't have cancer," Berchuck said. "If you have too aggressive of a screening test, you could end up doing surgery on people who don't need it, and you could cause more harm than good."
Berchuck said that the only thing that can be said when comparing the various ovarian cancer tests being developed is that the tests are all different in terms of exact proteomic technologies.
"In terms of which is most promising for use one day as a screening test I don't think you can compare at this point," he said. "I think they're all a long way off."
Ciphergen did not respond to telephone calls seeking comment for this article.