A new test that measures the activity of 14 genes in tumor tissue in may help doctors identify which early, non-small-cell lung cancers are the most dangerous, according to new research published on Friday.

Two independent clinical trials were conducted to measure the test’s efficacy, one study consisted of 433 people with early-stage lung cancer in northern California, and another study involved 1,006 people with early-stage cancer in China. Both of the studies found that the test could accurately predict the risk of death within five years of surgery to remove a lung cancer was low, intermediate or high based on genetic “fingerprints”.

Lung cancer is the most common cause of cancer death both domestically and globally. More people die from lung cancer than from breast, colon and prostate cancers combined, and unlike stage I breast cancer, which has a five year survival rate of 88 percent, stage I non-small-cell lung cancer has a five-year survival rate of 45 percent, according to the American Cancer society.

Lung cancer often goes undetected in its early stages, when it is the most treatable, because the early-stage disease does not usually cause symptoms. Researchers said that the test would also better match patients to an effective therapy to improve the survival rate for lung cancer.

University of California, San Francisco and China Clinical Trials Consortium researchers said that the analysis begins with a piece of the patient’s cancerous tissue that has been preserved, after which RNA is extracted from the tissue to reveal the activity level of genes within tissues. Experts say that the 14 specific genes are the determined and compared to normal lung levels.

Eleven of the genes are linked to lung cancer biology and the other three are common genes used as standardized measurements for cancer genes.

Researchers hope that doctors can use the test to determine which patients may benefit from more aggressive treatment for their lung cancers.

Pinpoint Genomics developed an algorithm for calculating risk of death after examining tissue from patients who all went under surgical procedure to treat non-squamous, non-small cell lung cancer and found that the 14 gene levels correlated with the clinical outcomes of the patients.

“The algorithm very accurately differentiated patients with high, intermediate or low risks of death in these larger cohorts of patients, even for patients with stage II and stage III lung cancer,” the researchers wrote.

Patients with tumors the test measured to be a high-risk profile had a high rate of early death, and about half of those patients were not alive five years after their diagnoses.

About 42 percent of patients with tumors tested as intermediate based on their genetic profile were not alive five years later their diagnoses and 25 to 30 percent of patients with low-risk tumors had not survived five years later.

"This has the potential to help hundreds of thousands of people every year survive longer." Co-author David Jablons, the Ada Distinguished Professor in Thoracic Oncology said in a statement released Thursday.

The researchers’ results are published in The Lancet.