Normal serum neuron specific enolase (NSE) value after the first cycle of chemotherapy: an early predictor of complete response and survival in patients with small cell lung carcinoma

Cancer. 1998 Mar 15;82(6):1049-55. doi: 10.1002/(sici)1097-0142(19980315)82:6<1049::aid-cncr6>3.0.co;2-9.

Abstract

Background: Serum neuron specific enolase (NSE) is the most sensitive tumor marker of small cell lung carcinoma (SCLC) at diagnosis. Its prognostic value is still debated. Thus, the authors decided to assess the predictive value, in terms of complete response and survival, of serum NSE measured before and after one cycle of chemotherapy in patients with SCLC.

Methods: Sera from 135 patients with histologically proven limited (n = 63) or metastatic (n = 72) SCLC were obtained. Clinical and biologic parameters with a known or suspected prognostic relevance were reviewed. Serum NSE was measured before chemotherapy (D1-NSE) and 28 days after its initiation (D28-NSE). The prognostic value of the parameters under study was evaluated in univariate and multivariate analyses using the Cox proportional hazards model and logistic regression analysis.

Results: The level of serum NSE was raised in 120 patients (88%) prior to therapy. The probability of a normal D28-NSE value was not affected by the baseline D1-NSE value. Disease extension (P = 0.0005), performance status (P = 0.0001), D28-NSE (P = 0.003), and carcinoembryonic antigen (CEA) levels (P = 0.008) were found to be predictive for survival, whereas age, gender, plasma sodium, serum protides, and D1-NSE were not. Median survival and 2-year overall survival were 15.3 months and 21% (95% confidence interval [CI], 13-31%) when D28-NSE was normal and 8.1 months and 15% (95% CI, 8-27%) when it was not (P < 0.03). Only performance status (P = 0.001), disease extension (P = 0.002), and D28-NSE (P = 0.02) were found to be independent prognostic parameters for survival in the multivariate analysis. A simple prognostic index was developed using these 3 variables. Limited disease, a normal D28-NSE value, and a normal CEA value prior to therapy were the only parameters predictive for complete response in the univariate analysis, and D28-NSE (P = 0.01) and disease extension (P = 0.0001) were found to be independent variables in multivariate analysis. A complete response to therapy occurred in 62% with a normal D28-NSE value and in only 34% in the opposite case.

Conclusions: Normal serum D28-NSE is a strong, independent early predictor of both complete response to therapy and survival. This simple tool may be proposed for use in the clinic and in research, in association with an assessment of disease extension and performance status, to predict the outcome of patients with SCLC.

MeSH terms

  • Adult
  • Aged
  • Antineoplastic Agents / therapeutic use*
  • Biomarkers, Tumor / blood*
  • Carcinoembryonic Antigen / blood
  • Carcinoma, Small Cell / drug therapy
  • Carcinoma, Small Cell / enzymology*
  • Carcinoma, Small Cell / pathology
  • Disease Progression
  • Female
  • Humans
  • Logistic Models
  • Lung Neoplasms / drug therapy
  • Lung Neoplasms / enzymology*
  • Lung Neoplasms / pathology
  • Male
  • Middle Aged
  • Phosphopyruvate Hydratase / blood*
  • Predictive Value of Tests
  • Prognosis
  • Retrospective Studies
  • Survival Analysis

Substances

  • Antineoplastic Agents
  • Biomarkers, Tumor
  • Carcinoembryonic Antigen
  • Phosphopyruvate Hydratase