The Prognostic Significance of Circulating Neuroendocrine Markers Chromogranin A, Pro-Gastrin-Releasing Peptide and Neuron-Specific Enolase in Patients with Advanced Non-Small-Cell Lung Cancer

Background: Chromogranin A (CGA), Pro-gastrin-releasing peptide (ProGRP) and neuron-specific enolase (NSE) are known as immunohistochemical tissue markers closely associated with neuroendocrine differentiation in non-small-cell lung carcinoma (NSCLC). The aim of the present study was to assess the value of serum levels of these markers in predicting response to chemotherapy and survival of patients with unresectable NSCLC. Methods: The study included 67 patients with advanced NSCLC treated with chemotherapy. Before treatment, serum levels of CGA, ProGRP and NSE were measured with commercial kits. Results: No association was found between serum NSE and age, gender, histology, performance status or extent of the disease. Distribution of serum CGA differed significantly according to gender and histology, with higher levels being found in men (p = 0.01) and in squamous cell carcinoma (p = 0.01). Serum ProGRP levels correlated with disease extent, being higher in patients with metastatic disease (M1) than in those with locoregional disease (M0; p = 0.02). The association of NSE, CGA and ProGRP levels with response to chemotherapy was not significant. While NSE had no impact on survival, the median survival was shorter for patients with elevated serum CGA and longer for patients with high ProGRP levels. Association with survival was significant when the Classification and Regression Tree (CART)-derived or median cutoff points were explored. On inclusion in multivariate Cox models, both CGA and ProGRP retained significance with high levels showing an opposite effect on survival [CART-derived cutoff points: CGA, relative risk (RR) –4.0; p < 0.001, and ProGRP, RR –0.4; p = 0.006, and median cutoff points: CGA, RR –1.8; p = 0.04, and ProGRP, RR –0.5; p = 0.03]. The combined use of CGA, ProGRP and NSE allowed for definition of two sets of patients with significantly different median survival times (25.2 vs. 8.8 months, p = 0.0001). Conclusions: In the circulation, CGA and Pro-GRP appear to bear important information related to the prognosis for NSCLC patients before chemotherapy. While a high CGA before treatment was found as an unfavorable prognostic determinant, a high ProGRP conferred a survival advantage. The combined use of serum CGA, ProGRP and NSE may supply additional information to prognosis.