Gene Expression Based Prognostic Signatures in Lung Cancer Not Ready for Clinical Use

A review of published articles on gene expression based prognostic signatures in lung cancer revealed little evidence that any of the signatures are ready for clinical use, according to researchers at NCI. Serious problems in the design and analysis of the studies were also found.

To assess the progress made toward clinical application of these signatures, Jyothi Subramanian, Ph.D., and Richard Simon of the biometric research branch at the NCI analyzed 16 relevant published studies from 2002 to 2009 that reported on the development of gene expression based prognostic signatures for non-small-cell lung cancer. The review was published online March 16 in the Journal of the National Cancer Institute and is titled “Gene Expression–Based Prognostic Signatures in Lung Cancer: Ready for Clinical Use?”

Dr. Subramanian and Simon evaluated the studies for appropriateness of design, statistical validation of the prognostic signature on independent datasets, presentation of results in an unbiased manner, and demonstration of medical utility for the new signature beyond that obtained using existing treatment guidelines.

They report finding little evidence that any of the signatures are ready for clinical application. Many studies lacked statistical validation of data and reproducibility of the signatures and did not explain their actual medical utility. Also, none were successful in showing clear usefulness for the gene-expression signatures over and above the known risk factors.

Flaws in design and analysis were also discovered. The reviewers write that many studies had a lack of clear specification of therapeutically relevant objectives, inappropriate patient selection, and poor documentation of important prognostic factors.

Dr. Subramanian and Simon thus suggested a set of guidelines to aid the design, analysis, and evaluation of prognostic signature studies. “These guidelines emphasize the importance of focused study planning to address specific medically important questions and the use of unbiased analysis methods to evaluate whether the resulting signatures provide evidence of medical utility beyond standard of care-based prognostic factors.

“We hope that future research in this important field will strive to move away from being another exercise in clinical correlation to one that truly makes an impact on widespread medical practice,” the authors write.