Multivariate logistic regression analysis of the clinical factors influencing locally advanced prostate cancer.

ABSTRACT

Background: Locally advanced prostate cancer (PCa) carries a high risk of recurrence and metastasis after surgery, and the prognosis is poor. We explored the risk factors for locally advanced PCa among clinical factors (neutrophil lymphocyte ratio, lymphocyte monocyte ratio) and indicators of systemic inflammation [prostate-specific antigen (PSA) level, Gleason score, body mass index (BMI)] through retrospective evaluation of patients with PCa diagnosed at our center. The pathologic T stage was a key indicator of locally advanced PCa. Methods: Data from patients with pathologically confirmed PCa at our center from 1 January 2015 to 1 May 2020 were collected in strict accordance with inclusion and exclusion criteria. Clinical data were collected and the relationship between the indicators and the pathologic T stage was explored. First, Spearman rank correlation analysis was used to find the correlates of the pathologic T stage. Then, logistic ordered multiple regression analysis was used to identify independent risk factors. Finally, receiver operating characteristic (ROC) curves were used to assess the diagnostic accuracy for the T stage of PCa. Results: After rigorous screening, the data of 177 patients were obtained. Spearman correlation analysis showed that BMI, the PSA level, Gleason score, hypertension, N stage, and M stage were significantly correlated with the T stage (P<0.05), suggesting that these factors may be involved in locally advanced PCa. Analyses of ROC curves showed that the PSA level [area under the ROC curve (AUC) =0.802] had greater value than BMI (0.675) for the diagnosis of the pathologic T stage PCa, and that a combination of BMI and PSA (combined AUC =0.822) could improve locally advanced PCa diagnosis. Conclusions: BMI and PSA are independent risk factors for locally advanced PCa. They may play a key part in locally advanced PCa.