UT Southwestern Medical Center researchers have developed a computer model that can predict patients’ susceptibility to lung cancers utilizing bioinformatic technology, a new study says. The software was created after analyzing different sub-types and micro-environments from cancer cell data.
According to UT Southwestern, the current algorithms were based on images of cancer-tissue slides by pathologists used to distinguish between benign and malignant tumors. Dr. Andy Xiao, associate professor of clinical sciences and bioinformatics at UT Southwestern and senior author of the study, says this proof-of-concept study demonstrates the feasibility of the approach.
“This computational approach should someday make it possible for doctors to tailor the treatment of individual patients based on risk predicted by computer algorithms; for instance, choosing to treat patients at higher risk more aggressively,” Xiao said in a statement.
Researchers studied two sub-types of non-small cell lung cancer–the most common cause of lung cancer deaths. They analyzed 3,206 slides of cancer tissue from 523 patients with adenocarcinoma and from 511 patients with squamous cell carcinoma.
Xiao says “pathological examination of cancer-tissue slides from individual patients is a routine part of clinical practice for lung cancer diagnosis and prognosis.” When a pathologist matches an image slide with his or her memory of certain cancer-related features, the process can be time-consuming, subjective, and possibly introduces variation to results.
“Our computerized analysis used a pathological imaging scanner and a desktop computer, so it would be applicable to most clinical settings,” Xiao said. “The team is exploring the possibility of patenting the process.”
The computer analysis also includes features of the tissue surrounding a tumor; this micro-environment is believed to play an essential role in cancer, according to Xiao.
So far, the team has developed objective and quantitative computational approaches to measure and analyze differences in the structure and form of the cancer cells and their micro-environment as depicted in the patient images of tissue samples.
Moving forward, one challenge the team must overcome is translating features the computers identify into language pathologists understand. Before this computerized image analysis can be implemented, it needs to be tested in a study evaluating the effects of different image resolutions, specimen sizes, and types of samples.
“Nevertheless, we can envision a time when such pathology-based image analysis could be integrated with molecular analyses and clinical tests to guide doctors’ decision-making in the age of ‘precision medicine,’” Xiao said.
Researchers from UT Arlington, UT M.D. Anderson Cancer Center, and the University of Florida also participated in this study. The research was supported by the National Cancer Institute.