Researchers Identify Predictive Factors for Lung Cancer Recurrence

News

Researchers Identify Predictive Factors for Lung Cancer Recurrence with an 83% Accuracy Rate

Researchers Identify Predictive Factors for Lung Cancer Recurrence with an 83% Accuracy Rate

A new study led by New York University’s Langone Medical Center and its Perlmutter Cancer Center has found that genetic information collected from seemingly healthy tissue near lung tumors may be more accurate in predicting cancer recurrence after treatment than analyzing the tumors themselves.

Healthy tissue RNA predicts lung cancer recurrence with an accuracy rate of 83%.

Researchers Identify Predictive Factors for Lung Cancer Recurrence with an 83% Accuracy Rate

This research involved 147 early-stage lung cancer patients and has the potential to revolutionize early cancer treatment and biomarker identification.

According to data from the Centers for Disease Control and Prevention, lung adenocarcinoma, a cancer that forms in the lung alveolar epithelial cells, accounts for approximately one-third of all lung cancers in the United States. Most patients can be cured if tumors are surgically removed in the early stages of the disease. However, about 30% of cases see the residual cancer cells regrow, potentially leading to fatal outcomes. As a result, experts have long been searching for biomarkers or predictive factors for recurrence to encourage more aggressive early treatments for patients.

Research Method and Findings

The study involved 147 male and female patients who had received treatment for early-stage lung cancer. It explored the practical value of the transcriptome, which is the entire set of RNA molecules that tells cells what proteins to make. Analyzing RNA collected from visibly healthy tissue near tumor cells accurately predicted cancer recurrence in 83% of cases, compared to the tumor’s RNA, which had a reference value in only 63% of cases.

“Our research findings suggest that gene expression patterns in ostensibly healthy tissue can serve as an effective biomarker to predict early-stage lung cancer recurrence,” said Dr. Dolgalev, an assistant professor in the Department of Medicine at NYU Grossman School of Medicine and a member of the Perlmutter Cancer Center. This investigation is the largest-scale study to date comparing the genetic material near tumors and its ability to predict recurrence.

Researchers Identify Predictive Factors for Lung Cancer Recurrence with an 83% Accuracy Rate

Advanced Analysis Techniques and Significance

In this study, the research team collected nearly 300 tumor and healthy tissue samples from lung cancer patients. They sequenced the RNA of each sample and inputted this data, along with information on whether a recurrence occurred within five years after surgery, into an artificial intelligence algorithm. The program used a technique called “machine learning” to create mathematical models for estimating recurrence risk.

The study found that gene expression in seemingly normal lung tissue near tumors, related to inflammation or enhanced immune system activity, was particularly useful for prediction. The authors of the research report suggest that this defensive response should not be present in truly healthy tissue and may serve as an early warning signal for the disease.

Dr. Huabo Zhou, a co-first author of the study, a bioinformatician at NYU Grossman School of Medicine, and a member of the Perlmutter Cancer Center, stated, “Our research findings suggest that apparently normal tissue near the tumor may not be healthy. Instead, escaping tumor cells may trigger this unexpected immune response in their neighbors.”

Dr. Aristotelis Tsirigos, a co-first author of the study and a cancer biologist, added, “Immunotherapy can enhance the body’s immune defenses, potentially helping prevent and treat cancer before it is detected through traditional methods.”

Dr. Zirigos, a professor in the Department of Pathology at NYU Grossman School of Medicine and a member of the Perlmutter Cancer Center, emphasized that the investigation was conducted in reverse, using known cases of disease recurrence to train the computer program.

Dr. Zirigos, director of the Laboratory of Applied Bioinformatics at NYU Langone’s Lagni School, noted that the research team’s next step is to use the program for prospective assessment of recurrence risk in newly treated early-stage lung cancer patients.