A new study from the National Institutes of Health is being published in the journal of Clinical Oncology and demonstrates that elevated levels of C-Reactive Protein (CRP), a biomarker of chronic inflammation, is associated with lung cancer. Specifically, the study matches serum from 592 patients with lung cancer to 670 control patients who participated in a large cancer screening study known as the Prostate, Lung, Colorectal, and Ovary (PLCO) Trial and did not prove to have lung cancer; it also compared the forms of common genetic variants of CRP known as single nucleotide polymorphisms (SNPs, and pronounced "snips": see here for further discussion) of CRP from DNA material of 378 lung cancer patients to that of 447 controls.

Investigators divided serum CRP levels into quarters from lowest to highest and looked at whether there were associations of higher levels of CRP with the presence or future development of lung cancer. In fact, they found that elevated levels were associated with an almost exactly two-fold greater (odds ratio 1.98, P trend <0.001 for comparison of highest to lowest quartile) risk of lung cancer. The association was seen in most lung cancer subtypes, though not really in adenocarcinoma patients:

(click on image to enlarge)

Interestingly, the association of lung cancer risk with elevated CRP levels was most prominent 2-5 years before a cancer diagnosis. The association was seen in current or former smokers but not in never-smokers with lung cancer.

The work on different CRP SNPs comparing patients with lung cancer to controls didn't reveal any significant differences in these "isoforms" and risk of lung cancer.

The next question is what this means. The authors note that this implicates inflammation as a cause of lung cancer, and I think that may well be, but I would suggest that it's also possible that part of the early process of cancer development is that it may cause some degree of inflammation reflected in elevated CRP levels. I'll also say that a doubling of lung cancer risk still doesn't mean that lung cancer is now "common" in people with elevated CRP levels or reassuringly rare in someone who doesn't have elevated serum CRP levels. Also, elevated CRP levels are associated with many other medical problems, ranging from heart disease to diabetes to other cancers.

Overall, I'd say that this represents interesting but still early work. Though I don't see clear implications for managing lung cancer or segregating patients for screening or not, but it offers a lead for the possibility of estimating risk and perhaps even a future mechanism of helping to treat/modulate the cancer process in people with established lung cancer.