Variability of serum CYFRA 21 − 1 and its susceptibility to clinical characteristics in individuals without cancer: a 4-year retrospective analysis

Abstract

Abstract Background CYFRA 21 − 1 is a useful marker for diagnosing and monitoring lung cancer. However, its stability remains unclear. Moreover, while its applicability to screening is now being investigated, CYFRA 21 − 1 levels in individuals without cancer, who are targets for cancer screening, have not yet been the focus of research. Therefore, the present study investigated variability in and the factors increasing serum CYFRA 21 − 1 levels. Methods This retrospective study recruited 951 individuals undergoing annual medical examinations for six years. We used data obtained in the first four years. Variability in serum CYFRA 21 − 1 levels over a period of four years were investigated. CYFRA 21 − 1 was categorized as normal (≤ 3.5 ng/ml) or elevated (> 3.5 ng/ml). The rate of an elevated level in one visit and the transition from an elevated to normal level between visits were visualized. A multiple logistic regression model was used to study the relationships between the frequency of elevated CYFRA 21 − 1 levels and clinical characteristics, such as age, sex, body mass index, weight changes, and the smoking status. Results Approximately 5% of subjects had elevated CYFRA 21 − 1 levels once in five tests over four years, while 15% had elevated CYFRA 21 − 1 levels once or more. Among subjects with elevated CYFRA 21 − 1 levels in one blood test, between 63 and 72% had normal levels in the next test. The median CYFRA 21 − 1 level in subjects with elevations in one blood test significantly decreased in the next test at all four time points. The frequency of elevated CYFRA 21 − 1 levels was associated with an older age [odds ratio (OR) = 6.99, 95% confidence interval (CI) = 3.01–16.2], current heavy smoking (OR = 3.46, 95% CI = 1.52–7.9), and weight loss (OR = 1.86, 95% CI = 1.07–3.24). Conclusions Variability in and the factors increasing serum CYFRA 21 − 1 levels beyond the cut-off value need to be considered when interpretating CYFRA 21 − 1 test results. The future application of CYFRA 21 − 1 to lung cancer screening may require more than a single measurement.