EDITOR'S CHOICE IN CELL & MOLECULAR BIOLOGY
D. Aran et al., “Widespread parainflammation in human cancer,” Genome Biol, 17:145, 2016.
Researchers in Israel were trying to understand why a particular mouse model was so unusually cancer-prone once it suffered a mutation in the tumor suppressor gene p53. They noticed that the animal’s epithelial tissue showed signs of low-grade inflammation, undetectable by normal hallmarks such as white blood cell recruitment. Intriguingly, treating the mutated mouse with a nonsteroidal anti-inflammatory drug (NSAID) reduced its risk of cancer—an effect observed in some human studies as well.
To explore what was driving this so-called parainflammation (PI), the team recently collaborated with Dvir Aran of the University of California, San Francisco, and colleagues to identify its genetic signature, and found an upregulation of 40 genes. The researchers then analyzed gene expression data from more than 800 human cancer cell lines and from about 6,500 patient tumor samples, including 18 cancer types, to uncover whether any harbored this signature.
Striking a match
By scoring gene expression data based on their similarity to the PI signature, the team found four cancers—including lung and pancreatic adenocarcinomas—with notably higher tallies. Mutations in p53 were also associated with a higher PI score. The prognoses for these cancers were dramatically worse, suggesting PI might exacerbate cancer progression.
When the team treated mouse adenoma organoids and two human cancer cell lines with an NSAID, the expression of PI genes fell by more than half. The findings could translate to the clinic, by helping to identify patients who would benefit from NSAID treatment, says Michael Karin, who studies inflammation in colorectal cancer at UC San Diego. “Right now, nobody is stratifying the patients,” he says.