Ikonisys Inc.

Two-year-old Ikonisys has developed a cell-based analysis system for detecting rare cells in blood. Its first application is detection of fetal cells in maternal blood, a longstanding ambition of many companies, none of which has succeeded to date. The test would detect the most common birth defects earlier in a pregnancy and more accurately than existing methods. Ikonisys also sees a big opportunity in cancer for early detection of disease and micrometastasis.

Automated image analysis that can detect molecular signals in rare cells

• 5 Science Park

• New Haven, CT 06511

• Phone: [phone]

• Fax: [phone]

• E-Mail/Web Site: [email protected]

• Contact:Petros Tsipouras, MD, Co-Founder, Chairman, and CEO

• Industry Segment:In vitro Diagnostics

• Business:Cell-based image analysis

• Founded:1999

• Founders:Petros Tsipouras, MD; Triantafyllos Tafas, PhD; Michael Kilpatrick, PhD; Pramod Srivastava, PhD; Gamil de Chadrevian

• Employees:18

• Financing to Date:$7.5 million

• Scientific Advisory Board:Joshua Lederberg, PhD (Rockefeller University); Sir Walter Bodmer, PhD (Oxford University); David Ward, PhD (Yale University); Mark Evans, MD (Hahnemann University); Josef Kittler, PhD (University of Surrey, UK); Petros Tsipouras, MD (University of Connecticut); Triantafyllos Tafas, PhD (University of Athens)

The idea for an automated image analyzer that could quickly scan samples came to Triantafyllos Tafas in the early 1980s as he stood in lakes studying microorganisms in drops of water. A hydrobiologist with a specialty in population dynamics and a PhD from the University of Athens in Greece, he was challenged by the inability of existing technologies to analyze large amounts of information, and foresaw the utility of computer-controlled imaging in biology.

At the same time, across the ocean, his friend Petros Tsipouras, MD, was working as a pediatric clinical geneticist at the University of Connecticut Health Center . Over the years, the two men talked, and in 1989, they introduced the first Downs syndrome prenatal screening program to Greece. From this work and other discussions arose a new concept for cellular image analysis, which would keep single cells intact and which could be used to diagnose diseases in humans. It was based on the premise that molecular changes in cells could be indicators of disease and therefore detection of those changes could lead to early diagnosis. Doctors could perform a battery of tests identifying cells, and signals emanating from those cells, in order to obtain more information about patients.

Their ambitions were grand: for their first application they chose one of the toughest challenges in diagnostics, the detection of fetal cells in maternal blood. Many scientists had tried unsuccessfully to find these cells in order to diagnose prenatal abnormalities and birth defects, with the intention of eventually replacing amniocentesis. But the fetal cells were so rare that no one had been able to crack the problem.

Companies had been born around promising methods of doing this, only to fail or radically alter their course. Researchers faced untenable options: they could either work with only a few slides, requiring them to aggressively sort the blood samples in order to find appropriate cells to study, or they could create hundreds of slides, which would allow them to go through samples cell by cell. The former methods destroyed the targeted cells, as well as extraneous material, while the latter was too time consuming and expensive for clinical use. Some scientists used fluorescence-activated cell sorters (FACS) and magnetic-activated cell sorters (MACS) to detect fetal cells and separate them from maternal blood. But that required cells to pass through a narrow gate, which damaged them and rendered them unsuitable for accurate analysis.

In light of their backgrounds, Tafas and Tsipouras chose to pursue a route of fetal cell identification, figuring that if they were successful, other rare cell identification applications should prove to be viable. In this initial application, they wanted to develop a less aggressive yet still effective selection process that could scan large populations of cells without destroying them, and could detect single cells using the lowest possible intensity signal. The maximum number of slides they wanted to scan was 15 to 20 per sample, with plans to reduce that number to less than ten.

In June 1999, they, along with several colleagues, formed Ikonisys Inc. , and began to develop an integrated system for detection and analysis of rare cells, consisting of a fast, highly automated microscope, a camera to capture images, and a variety of molecular signals and probes to help detect cellular characteristics. Their Ikoniscopemicroscope can scan a patient sample in less than one hour, scanning approximately three fields per second. It has a high-speed device allowing it to take focus measurements more than 30 times per second, which is so fast that it can accommodate certain vibrations and therefore is less affected if the microscope moves. Infrared lasers automatically identify the best focus plane, independent of the microscopic image. Laser-based focusing is widely used in industrial applications, but has not been used much in health care. It has to be aimed at reflective surfaces, which are hard to position properly in biological settings. In cellular analysis, for example, the cells lie on slides, underneath slipcovers, which typically aren't reflective. The Ikoniscope utilizes special reflective slip covers a laser prefocusing device that allows the microscope to zero in on the cells of interest.

Ikonisys also has a patented signal production technology that integrates immunohistochemistry (IHC) and fluorescent in situ hybridization (FISH) in one process simultaneously on one slide. To do this, the company has developed a gentler method of fixing IHC samples that doesn't harm cell membranes and render them useless for FISH analysis. This approach can be used with a variety of lab instruments and is also proprietary to Ikonisys. It doesn't require clinical laboratories to make big changes in the way they work, and company executives therefore don't expect resistance to its adoption. The combination of IHC and FISH is useful in prenatal diagnostic applications. IHC can identify the presence of fetal cells, while FISH can be used for chromosomal analysis. Image analysis and the generation of the result report could be carried out on a computer incorporated into the platform. Furthermore, Ikonisys uses markers that it believes are specific to fetal cells and aren't seen in maternal cells, which will be useful in initial detection of the fetal cells—one of the company's key challenges is to make sure the tests are extremely sensitive and specific.

Prenatal detection of Downs syndrome is likely to be the first test on the market, involving the identification of chromosomes X, Y, and 21. Ikonisys plans to start pivotal trials for the test this fall, with an FDA PMA submission in 2003 and approval possibly in 2004. Downs syndrome is the most common chromosomal abnormality. The incidence rises along with the maternal age, reaching nearly 3% in pregnant women older than 40. Older pregnant women typically undergo amniocentesis during the 16th week of pregnancy to screen for birth defects, and many pregnant women, young and old, take the Triple G, a test given in the first trimester that screens for the three most common birth defects, including Downs. But the Triple G isn't terribly accurate, while amniocentesis is associated with a risk of miscarriage. The company expects that the Ikonisys test will both improve the overall accuracy and the side-effect profile of the current prenatal testing protocol.

Ikonisys' founders believe that the ability to do a maternal blood test to determine the presence of Downs syndrome in a fetus earlier in the pregnancy would be of interest to many people. About 2.8 million Downs syndrome tests are currently done each year in the US and 6.4 million are done worldwide, representing a $1 billion opportunity. It would ideally be done as early as five weeks into the pregnancy, at the time of the second OB/gyn visit. Ultimately, once it can demonstrate effective identification of fetal cells in maternal blood, the company hopes to develop screens for other birth defects now identified by amniocentesis.

Cancer is the next logical application, based on the assumption that an instrument which can scan lots of images in a short time is ideally suited to look for rare cells in blood and other specimens and would therefore be useful in early detection and possibly screening for cancer. Using cell-based diagnostics means that scientists don't have to grow genetic material for analysis. The company is studying the use of its system in prostate and breast cancers. One laboratory is considering a collaboration with Ikonisys to study early detection of breast cancer in lymph nodes using the Ikonisys system and epithelial markers; a long-term goal for Ikonisys and its partners would be to find markers that can identify breast and other cancers in blood.

Ikonisys aims to place prototypes of its system in large reference laboratories and is also seeking partnerships with diagnostic instrument companies. The company plans to follow the reagent-rental program concept used by many diagnostic instrument manufacturers, in which customers get instruments at little or no up-front cost, but pay for reagent kits and supplies. On the reimbursement front, Tafas and Tsipouras believe they can command high prices for their tests.

Ikonisys has raised $7.5 million from three rounds of financing from wealthy individuals and institutional investors and is currently seeking $3 to $5 million in a fourth round.

Tafas, the company's chief scientific officer, is a tenured faculty member at the University of Athens and a professor of pediatrics at the University of Connecticut Health Center. Michael Kilpatrick, PhD, VP, biology, received his doctorate from the University of Birmingham in the UK in 1981 for studies on ribonucleic acid of mycoplasma. In 1993, he joined the University of Connecticut Health Center, where he began collaboration with Drs. Tsipouras and Tafas. Bernhard Opitz, SVP, operations, joined the company in 2000 after 20 years in engineering and manufacturing positions at Bayer AG in Germany, Italy and the US. Since 1997, he had been SVP, engineering for Bayer Corp., overseeing Bayer's $800 million annual capital expansion program in the US. He holds an MS in engineering from Technical University in Graz, Austria. Klaus Peters, VP, technology, holds a PhD in biology from the University of Hamburg, specializing in detection and imaging of virus and cell surface antigens. He is an adjunct professor at the University of Connecticut Health Center.—WD