[Diagnostic value of hematological parameters for prostate cancer in patients with gray-zone prostate-specific antigen levels].

OBJECTIVE

To evaluate the diagnostic value of hematological parameters for PCa with prostate-specific antigen （PSA） of 4－10 μg/L and construct a risk-stratification model with these parameters.

METHODS

We retrospectively analyzed the data on the males undergoing the initial prostatic biopsy in the Affiliated Hospital of Xuzhou Medical University with PSA of 4－10 μg/L from March 2010 to April 2021. According to the results of biopsy, we classified the patients into a PCa and a non-PCa group, and compared the hematological parameters between the two groups. We performed univariate and multivariate logistic regression analyses, identified the independent risk factors for PCa, constructed a risk-stratification model for the prediction of PCa and evaluated its efficiency.

RESULTS

A total of 415 cases were included in this study, 107 (25.8%) in the PCa and 308 (74.2%) in the non-PCa group. Compared with the non-PCa males, the PCa patients showed a significantly older age, higher ratios of neutrophil to lymphocyte and platelet to lymphocyte, systemic immune-inflammation index (SII), red blood cell distribution width and cystatin C (CysC) level (all P<0.05), but lower red blood cell count and hemoglobin and free/total PSA (f/tPSA) levels (all P<0.05). Multivariate logistic regression analysis indicated that age, f/tPSA, SII and CysC were independent risk factors for the prediction of PCa (all P<0.05). Five prediction models were constructed based on the above risk factors, and the area under the ROC curve (AUC) of the four-parameter (age+f/tPSA+SII+CysC) model was 0.745 (95% CI: 0.694－0.796), significantly higher than those of the other models (P<0.05). A risk-stratification model (low-, intermediate-, and high-risk) was also constructed based on the total nomogram scores, which showed a comparable performance to that of the Prostate Imaging Reporting and Data System (PI-RADS) for the prediction of PCa (AUC: 0.727 ［95% CI: 0.650－0.804］ vs 0.734 ［95% CI: 0.658－0.811］). However, the prediction rate by the risk-stratification model was evidently higher in the low-risk males than in those with low PI-RADS scores (1－2) (39.4% vs 22.2%).

CONCLUSION

SII and CysC are independent risk factors for the prediction of PCa in patients with gray-zone PSA levels. The risk-stratification model based on age, SII, CysC and f/tPSA is comparable to PI-RADS in the diagnostic efficiency of PCa, with an even higher prediction rate in low-risk patients than in those with low PI-RADS scores, and contributive to precision screening and reduction of excessive biopsies in the diagnosis of PCa with gray-zone PSA.