Yale researchers have developed a new technology that may help doctors diagnose and treat cancer and other diseases by better identifying patterns of protein expression.

Three investigators from the School of Medicine published their findings in the November issue of Nature Medicine. In their study, they revealed an automated process through which tissue samples can be examined more accurately and quickly.

The digital imaging technology will be particularly useful in testing patients’ response to new drugs that require more sophisticated identification of proteins, said David Rimm, a pathology professor and the study’s lead researcher.

Rimm said the new technology may be used to determine whether the benefits of cancer treatments outweigh possible side effects.

Traditionally, pathologists have identified patterns by “reading” arrays of proteins, often requiring researchers to examine samples for many hours at a time. In addition, existing methods of identifying patterns of protein expression are often inexact, even when conducted with highly advanced microscopes, researchers wrote.

In their study, Rimm and his co-investigators Gina Chung and Robert Camp used “tissue microarrays” — slides containing tissue samples taken from hundreds of different tumors. The researchers found that an automated analysis more accurately identified proteins in samples of both breast and colon cancer tissue.

The researchers found that the automated analysis of the protein beta-catenin in colon cancer samples was better at identifying differences between tumors than traditional methods of analysis. The distinctions detected by the automated analysis helped explain differences in survival rates among patients, researchers wrote.

Rimm said his team’s methods of analysis may be used to develop cancer medications that will provide more specific treatments than traditional chemotherapy.

“In the future, a lot of the drugs in the pipeline will require us to know the levels of the targets,” Rimm said.

The investigators also suggested that their technology could shorten the time necessary to analyze cancerous tissue to two to five hours. The process typically takes many hours for two or more pathologists.

“Aside from being more accurate and more robust, automated analysis can be performed continuously and results tabulated immediately,” the researchers wrote. “We estimate that a fully integrated tissue microarray reader could be 30 to 50 times faster than pathologist-based scoring.”

Histometrix, Inc., a company currently being established by the University, will develop and market the technology.

Robert Curtis, president and chief executive officer of Histometrix, said the company was currently in discussions with pharmaceutical companies about providing research and development services. He said the technology might be put to commercial use in as soon as three to six months.

“It could be used to identify new targets for drug discovery or to validate targets for drug discovery,” Curtis said. “It can actually be used to find patients that are most likely to respond to drugs in clinical trials.”