New research shows that long noncoding RNAs have distinct expression patterns in different cancer types, and may have diagnostic and therapeutic potential as a specific, informative biomarker of cancer.

The evidence that long noncoding RNAs (lncRNA)—transcripts longer than 200 nucleotides that aren't translated into proteins—play a role in cancer development continues to grow. According to a new study, lncRNAs not only are dysregulated in cancer, but have distinct expression patterns in different cancer types.

LncRNAs are plentiful in cells, where their functions include guiding transcription and orchestrating protein-protein interactions. They can also act as tumor suppressors or oncogenes. To better understand how these molecules influence cancer, Lin Zhang, MD, and Chi Van Dang, MD, PhD, of the University of Pennsylvania in Philadelphia, led an in-depth genomic analysis of lncRNA alterations in 5,037 tumor specimens across 13 cancer types, using data from The Cancer Genome Atlas (TCGA) and the Cancer Cell Line Encyclopedia (CCLE).

Zhang, Dang, and their team focused on 7 cancers—including prostate adenocarcinoma and lung squamous carcinoma—with the largest amount of control data from normal tissue. They found that 15% of lncRNAs were more abundant in these tumors than in control tissue, while 11.2% were less abundant. Although some expression pattern changes, such as in the oncogenic lncRNAs PCAT7 and HOTAIR, were shared by multiple cancers, approximately 60% of the dysregulated lncRNAs were cancer type–specific.

The researchers wondered whether DNA methylation, which shuts genes down, might account for the decreased expression of some lncRNAs in these 7 cancer varieties. They determined that 3.9% of lncRNAs had hypermethylated promoter regions and correspondingly reduced expression, suggesting a role for epigenetic silencing in lncRNA dysregulation.

Using data for breast cancer, the researchers next asked whether they could identify clinically relevant cancer-driving lncRNAs. They identified 19 dysregulated lncRNAs that correlated with patient survival, and singled out one, BCAL8, for further investigation. They discovered that suppressing BCAL8 expression significantly reduced the growth of breast cancer cells, as well as tumors in mice. High BCAL8 expression was also associated with poor clinical outcome in ovarian and uterine cancers, the team found.

“This is a landmark paper,” says Timothy Triche, MD, PhD, of the University of Southern California in Los Angeles, who wasn't connected to the study. “It substantiates the fact that noncoding RNAs are not only pervasive and dysregulated [in cancer], but that they are also cancer type–specific.”

Dang, who directs the Abramson Cancer Center, says the study results indicate that “lncRNAs should be considered for diagnostic purposes and potentially for therapeutic purposes.” He and his colleagues have created a searchable database, the Cancer LncRNome Atlas, that incorporates molecular profiles of lncRNAs alongside clinical information from TCGA and the CCLE, to help other scientists investigate these molecules.