This article has been corrected. You may view this article’s correction here.

Finding cancer early can stop it in its tracks. But for many types of cancer, existing diagnostics leave much to be desired. A team of Yale scientists is applying cutting edge methods in protein and genetic research to the problem, hoping that the body’s own efforts to fight tumors can tip doctors off to the threat.

The research lab is using a “protein chip” technique to identify new diagnostic markers for ovarian cancers. With this technology, the team is establishing diagnostic tests that screen tissue and blood for elevated levels of cancerous auto-antibodies.

The researchers are led by Michael Snyder, professor of molecular, cellular and developmental biology, in collaboration with Gil Mor, associate professor of obstetrics and gynecology at the Yale Medical School.

“Our research is based on the hypothesis that cancer patients produce auto-antibodies against antigens that reflect the disease state,” Snyder said. “The lab is exploring whether certain proteins can identify this disease state by binding to antibodies in cancer patients but not to those in healthy women.”

The immune system views cancer as a foreign entity, producing anti-tumor antibodies, or auto-antibodies, not present in healthy individuals. A protein chip, which looks like a microscope slide with a coating that facilitates protein binding, can be used to determine whether certain proteins bind to auto-antibodies at higher levels than to antibodies in healthy control groups, said Li-Yung Kung, a post-doc in Snyder’s research lab.

Kung said the lab uses these chips to select proteins with the greatest difference in binding affinity between cancerous and non-cancerous tissues. The Snyder lab pioneered the use of proteome chips, Kung said, which involves cataloguing the entire set of proteins encoded by an organism’s DNA on a single chip.

Five out of eight proteins tested thus far bind in greater levels in cancerous tissues than in matching non-diseased tissues, Snyder said. The lab has isolated 95 candidate proteins and is currently engaged in testing the remaining proteins, he said.

Identifying new markers of ovarian cancer may help in developing more accurate and sensitive screening tests that are crucial to improving the survival rate of patients, Snyder said.

“This work … has huge implications for diagnostics, because it provides a better way of catching ovarian cancer earlier,” Snyder said. “If you catch it early, it is curable in about 95 percent of cases. In cases of late prognosis, there is a 15 percent survival rate.”

Snyder said he hopes this line of research will eventually allow the group to develop a diagnostic test that screens the blood. While tissue extraction is an invasive, complicated process, drawing blood is relatively simple and easy to do in clinical settings.

The study’s findings may also have applications to a wider range of cancers, Snyder said, since several of the ovarian markers the team has isolated have similar effects in other types of cancerous tissues.

Snyder’s lab is not the first to apply the protein-chip technique to the study of ovarian cancer. Last year, a team at the Barbara Ann Karmanos Cancer Institute in Detroit developed a screening test for the disease based on a similar research technique that involved selecting proteins as biological markers.

The Yale group hopes that its study will allow it to devise therapies specific to proteins in different patients, Kung said.

“One goal is to use these markers to [determine] which women respond better to different treatments, allowing the therapies to be tailored better [to individuals],” he said in an e-mail.

A concurrent project in the Snyder lab aims to explore how genes are expressed in patients with colon cancer relative to healthy individuals. The research is based on the lab’s earlier unpublished finding, which was confirmed by studies conducted at other labs, that approximately 10 percent of the genome varies between any two given individuals. The discovery challenges the existing theory that DNA variation between individuals only occurs on the order of single bases.

This 10 percent difference in DNA might code for proteins that predispose individuals to developing certain kinds of cancer.

“The idea is that large chunks of DNA in cancer patients will be different from healthy individuals,” Snyder said. “We can use this to identify which genes are implicated in the disease.”

Snyder said that the lab — which built the first high-resolution tiling array for the entire human genome — is also applying this technique on a broader scale, hoping to correlate a variety of diseases to the genes that cause them. But this relationship is usually much more complex than a simple linear association, he said, since most diseases are pleiotropic, meaning that single genes can have multiple effects.

Ongoing research may yield valuable insight into genetic patterns that increase the risk of conditions such as diabetes and obesity, Snyder said.

His future goals in this area include exploring the functions of genomic segments that are not transcribed into proteins, which account for 98.5 percent of the human genome.

Snyder said his cousin recently struggled with lung cancer, opening Snyder’s his eyes to the real-world implications of cancer research.

“It is amazing to think that something we invent can be used to cure or prevent cancer in individual cases,” he said.