Researchers Discover a Unique Molecular Profile for Lung Cancer

A team of researchers has found that the expression pattern of certain microRNAs, or miRNAs, may predict tumor aggressiveness in some patients with lung cancer.

These findings indicate that miRNAs may represent a new class of diagnostic and prognostic tools for lung cancer.

The study is a collaboration among researchers at The Ohio State University Comprehensive Cancer Center, Columbus, Ohio; the Jikei University School of Medicine, Tokyo, Japan; National Cancer Center Research Institute, Tokyo, Japan; and the Center for Cancer Research at the National Cancer Institute (NCI). The study results appear in the March 13, 2006, issue of "Cancer Cell".

The researchers identified two miRNAs - "has-mir-155" and "has-let-7a-2" - that could be used as prognostic indicators in patients with adenocarcinoma, a malignancy of the mucous glands in the lungs.

High levels of "has-mir-155" or low levels of "has-let-7a-2" were associated with poor prognosis.

Specifically, overexpression of "has-mir-155", or the signalling of the miRNA to change the amount of a protein produced, was the most significant indicator of this prognosis, independent of tumor stage.

Although these miRNAs have been identified in other cancers, this is the first evidence linking "has-mir-155" to lung cancer.

A tumor with an overexpression of "has-mir-155" or reduced expression of "has-let-7a-2" would indicate the need for aggressive chemotherapy or radiation treatments.

Other tumors that do not show high "has-mir-155" or low "has-let-7a-2" levels are less aggressive, and those patients might not require more therapy.

"This study is significant because it provides another tool for studying prognosis that is independent of tumor stage," said Curtis Harris, M.D., chief of the Laboratory of Human Carcinogenesis at NCI and co-leader of this study.

"Following surgery, 50 to 60 percent of patients with stage I lung cancer will develop metastatic disease within five years. This may indicate that there are micrometastases that have not been detected by imaging, scanning or pathology."

Harris noted that, "In the future, we can use miRNAs and other biological predictors to select patients who may need more aggressive treatment versus those who may not. Additional studies confirming these results are the next step before incorporating miRNA analysis into routine clinical practice."

"miRNAs are going to be important biomarkers of not only diagnosis and prognosis, but therapy, as well," said Harris.

"The next step is to identify the genes that the miRNAs are affecting; they could be used as potential targets for developing novel therapies."

In the study, a total of 104 pairs of primary tumor tissues and corresponding noncancerous lung tissues were examined.

The tumor and corresponding normal tissues were obtained from the same patient to eliminate genetic differences between tumor and normal tissues.

Researchers focused primarily on the more common adenocarcinomas in their analysis; adenocarcinomas and squamous cell carcinomas.

Unlike adenocarcinomas, squamous cell carcinomas form in the cells lining the internal surfaces in the lung. These two tumor types are the most common lung cancers and are referred to as non-small cell lung cancers (NSCLC).

In the study, adenocarcinomas of the lung comprised 63 percent of the tumors studied and squamous cell carcinomas comprised 37 percent.

Patterns of miRNA expression in each tumor and normal tissue pair were studied by microarray analysis. A microarray allows the measurement of hundreds of miRNAs simultaneously in a single sample.

Five miRNAs displayed different expression levels in tumor tissues versus their corresponding noncancerous controls and, thus, were selected for further study.

Upon statistical analysis, data showed that patients with high "has-mir-155" or low "has-let-7a-2" had poorer survival than patients showing low "has-mir-155" or high "has-let-7a-2" expression. The difference in the prognosis of these two groups was highly statistically significant.

After examining tissue from lung cancer patients, and following each patient to see how long they lived, researchers found that miRNA expression patterns were independent of tumor stage.

When the scientists combined all clinical and molecular factors, they found that a high level of "has-mir-155" or a low level of "has-let-7a-2" was the most significant prognostic factor for an unfavorable patient outcome.

"This is promising research and is the first study to link these miRNAs to lung cancer," said Carlo Croce, M.D., chair of Molecular Virology, Immunology and Medical Genetics at The Ohio State University and study co-leader.

"We are proposing that "has-mir-155" may be acting like an oncogene in lung cancer." Oncogenes control cell growth and, when mutated, can contribute to abnormal cell growth.

"miRNAs control the expression of a number of genes," noted Croce, "so if the miRNA is altered, this could lead to the alteration of a number of genes affecting malignant tumor growth."

"Although these results are encouraging, further testing is required to demonstrate the validity of using these markers for predicting patient outcomes."

Other types of cancers also have been shown to express specific miRNAs; however, the exact role that miRNAs play in the development of human cancers is unknown.

A recent paper appearing in the "Proceedings of the National Academy of Science" supports the observation that cancer-related genes are regulated by miRNAs in solid tumors.