BACKGROUND Chronic inflammation is known to be one of the main steps in carcinogenesis. Identification of those with chronic inflammation may help identify subjects at risk of cancer. Previous studies have reported low albumin-to-globulin ratio (AGR) to be associated with increased cancer mortality in cancer patients, but there has been no study based on healthy populations.

PATIENTS AND METHODS Our retrospective cohort study involved 26 974 generally healthy adults aged 30 or older who visited Seoul National University Hospital Health Promotion Center for self-referred health checkup. National medical service claims data were used to determine cancer incidence, and Korean death registry data was used to determine mortality. Median follow-up time for survival was 5.9 years (interquartile range 4.1 years).

RESULTS Compared with subjects with AGR ≥ 1.5, subjects with 1.1 > AGR ≥ 1.0 and 1.0 > AGR showed adjusted hazard ratio (aHR) 2.69 (95% confidence interval, CI, 1.54-4.72) and aHR 6.71 (95% CI 3.56-12.66) for all-cause mortality, aHR 2.95 (95% CI 1.42-6.11) and aHR 4.38 (95% CI 1.57-12.25) for cancer mortality, and aHR 2.07 (95% CI 1.28-3.36) and aHR 3.99 (95% CI 2.10-7.58) for cancer incidence, respectively. When cancer incidence events after 2 years from baseline were separately analyzed, subjects with 1.1 > AGR ≥ 1.0 and 1.0 > AGR were associated with aHR 1.88 (95% CI 1.01-3.48) and aHR 2.55 (95% CI 1.03-7.11) for cancer incidence, respectively. Cancer events were increased in all types of cancer, but especially in liver and hematologic malignancies.

CONCLUSIONS Low AGR is a risk factor for cancer incidence and mortality, both short- and long terms, in a generally healthy screened population. The results of this study need to be replicated in larger studies, along with the determination of the sensitivity and other diagnostic values of low AGR.