Immunoassay of neuron-specific enolase (NSE) and serum fragments of cytokeratin 19 (CYFRA 21.1) as tumor markers in small cell lung cancer: Clinical evaluation and biological hypothesis

NSE is a biochemical marker for small cell lung cancer (SCLC) diagnosis and management. CYFRA 21.1 is a newly developed immunoassay to detect the serum fragments of cytokeratin 19 which are also expressed in SCLC with or without neurofilaments. The aim of this study was to evaluate the diagnostic performance and prognostic role of the two markers in SCLC and their contribution to chemotherapy monitoring and patient follow-up. We studied 62 patients with pathologically proven SCLC:28 with limited disease (LC) and 34 with extensive disease (ED), and 100 patients with non-malignant pulmonary disease. Immunoradiometric assays (IRMA) were employed to test NSE and CYFRA 21.1 in patients and control subjects. For each patient subset results were expressed as median and interquartile distribution range. NSE and CYFRA 21.1 sensitivity was 0.52 (33/62) and 0.56 (35/62), respectively. In the group of patients with LD, NSE and CYFRA 21.1 sensitivity was 0.42 (12/28) and 0.54 (15/28) and in patients with ED, NSE and CYFRA 21.1 were positive in 0.62 (21/34) and 0.59 (20/34), of cases, respectively. Combining the two markers, a sensitivity of 0.78 (22/28) in LD, 0.82 (28/34) in ED and a global sensitivity of 0.80 (50/62) was obtained. Only NSE was significantly linked to the extension of disease (Mann-Whitney U test p = 0.002) while CYFRA 21.1 did not correlate. The analysis of survival and the evaluation of the two markers at diagnosis showed CYFRA 21.1 to be strongly linked to the patients' outcome, independently of both clinical prognostic factors and NSE levels (log rank and Cox's model). The markers' performance during chemotherapy was tested in a group of 33 patients with at least one marker above cut-off. NSE can be considered a reliable marker of tumor mass modifications under chemotherapy, while CYFRA 21.1 expression seems to be relatively independent of tumor volume modifications. An applicable model of biomarkers in SCLC could be the concurrent assay of NSE and CYFRA 21.1 in pre-therapeutic assessment and therapy planning. CYFRA 21.1 does not play an important role during therapy monitoring and follow-up; in these phases NSA alone may be employed.