Simultaneous analysis of serum α2,3-linked sialylation and core-type fucosylation of prostate-specific antigen for the detection of high-grade prostate cancer

Background Altered prostate-specific antigen (PSA) glycosylation patterns can be useful biomarkers in detecting high-grade prostate cancer (HGPC). The microfluidic immunoassay system can analyse α2,3-linked sialylated PSA (α2,3-Sia-PSA) and α1,6-linked fucosylated PSA (α1,6-Fuc-PSA) using different lectins, Mackkia amurensis agglutinin and Pholiota squarrosa lectin, respectively. Here, we investigated the diagnostic value of simultaneous analysis of α2,3-Sia-PSA and α1,6-Fuc-PSA for the detection of HGPC. Methods Men with serum PSA levels of 4–20 ng/mL who underwent prostate biopsy were included. The model to predict HGPC (Gleason grade ≥2) was constructed by multivariate logistic regression analysis, in combination with α2,3-Sia-PSA and α1,6-Fuc-PSA (SF index). Results In the development cohort ( n = 150), the SF index showed good discrimination for HGPC (area under the receiver-operating curve (AUC) 0.842; 95% confidence interval (CI) 0.782–0.903), compared to the single PSA test (AUC 0.632, 95% CI 0.543–0.721), α2,3-Sia-PSA (AUC 0.711, 95% CI 0.629–0.793) and α1,6-Fuc-PSA (AUC 0.738, 95% CI 0.657–0.819). Decision-curve analysis showed the superior benefit of the SF index. In the validation cohort ( n = 57), the SF index showed good discrimination for HGPC (AUC 0.769, 95% CI 0.643–0.895). Conclusions The SF index could differentiate HGPC, providing useful information for decision making for prostate biopsy in men with abnormal PSA levels.