Abstract

An enzyme immunoassay for serum neuron-specific enolase (NSE) was evaluated with respect to analytical performance and clinical utility and compared with immunohistochemical evaluation of neuroendocrine differentiation. Values obtained agreed well with values obtained using a radioimmunoassay method giving a correlation coefficient of 0.934. Analytical performance of the enzyme immunoassay was good but the diagnostic sensitivity of 82% in extensive and 67% in limited disease was insufficient for serum NSE to be of value in the diagnosis of small cell lung cancer (SCLC). Serum NSE decreased significantly in 11 of 15 patients with SCLC following institution of chemotherapy. Classification of lung cancers into SCLC and non small cell lung cancer (NSCLC) types is largely based on tumour morphology. Neuroendocrine differentiation may not be morphologically evident. Immunohistochemical staining of tumour tissue with markers of neuroendocrine differentiation, i.e. NSE (both monoclonal and polyclonal antibodies) Leu 7, Chromogranin A and P G P 9.5 was performed in both patients with SCLC and NSCLC. 38 per cent of patients with NSCLC had both raised serum NSE and positive NSE (polyclonal) immunoperoxidase staining of lung tissue. A further 35 per cent of patients showed a raised serum NSE or positive immunohistochemistry but not both. The presence of two positive immunoperoxidase markers in lung tissue has been suggested as an indicator of responsiveness to chemotherapy in NSCLC patients. A number of factors may affect immunohistochemical positivity in tissue sections and the additional use of a serum marker may better define chemotherapy responsive groups.