Fuzzy logic-based tumor marker profiles improved sensitivity of the detection of progression in small-cell lung cancer patients.

Tumor markers were used for disease monitoring in small-cell lung cancer patients. The aim of this study was to improve diagnostic efficiency in the detection of tumor progression in small-cell lung cancer patients by using fuzzy logic modeling in combination with a tumor marker panel (NSE, ProGRP, Tumor M2-PK, CYFRA 21-1, and CEA). Thirty-three consecutive small-cell lung cancer patients were included in a prospective study. The changes in blood levels of tumor markers and their analysis by fuzzy logic modeling were compared with the clinical evaluation of response versus non-response to therapy. Clinical monitoring was performed according to the standard criteria of the WHO. Tumor M2-PK was measured in plasma with an ELISA, all other markers were measured in sera. At 90% specificity, clinically detected tumor progression was found by the best single marker, NSE, in 32% of all cases. A fuzzy logic rule-based system employing a tumor marker panel increased the sensitivity significantly (P>0.0001) in small-cell carcinomas to 67% with the threemarker combination NSE/ProGRP/Tumor M2-PK and to 56% with the best two-marker combination ProGRP/Tumor M2-PK, respectively. An improvement of sensitivity was also observed using the two-marker combination of ProGRP/NSE (sensitivity 49%) or NSE/Tumor M2-PK (sensitivity 52%). The fuzzy classifier was able to detect a higher rate of progression in small-cell lung cancer patients compared with the multiple logistic regression analysis using the marker combination NSE/ProGRP/Tumor M2-PK (sensitivity 44%; AUC=0.76). With the fuzzy logic method and different tumor marker panels (NSE, ProGRP and Tumor M2-PK), a new diagnostic tool for the detection of progression in patients with small-cell lung cancer is available.