Prognostic value and predictive model construction for patients undergoing laparoscopic radical prostatectomy based
Hao Wang1, Pu-Shen Yang1, Yi-Rui Wei1
1Department of Urology, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China.
Objective:
To explore the prognostic value of preoperative hematological indicators for prostate cancer (PCa) patients with laparoscopic radical prostatectomy (LRP) and construct a nomogram prediction model based on hematological indicators and clinicopathological characteristics.
Method:
PCa patients who underwent LRP in Beijing Chaoyang Hospital from January 2017 to December 2022 were retrospectively analyzed. Clinicopathological data and blood indicators, including the neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), lymphocyte-to-monocyte ratio (LMR), red blood cell distribution width (RDW), prognosis nutritional index were compared between non-recurrence and recurrence groups. The NPL-IRS score was inflammatory response system score based on the cut-off values NLR, PLR, LMR. Kaplan-Meier analysis was used to calculate the prognostic survival curve. Univariable and multivariable Cox regression risk models were used to identify independent risk factors. A nomogram prediction model was developed, and its accuracy was evaluated and validated through receiver operating characteristic (ROC) curve, C-index, and calibration curve. Internal validation was conducted using Bootstrap method, and the model was also evaluated through external validation.
Results:
The number of PCa patients in the training set and external validation set was 210 and 110, respectively. A higher NLR, PLR, RDW, and NPL-IRS score but lower LMR and prognosis nutritional index levels were related to a poor recurrence-free survival (RFS). In training set, the area under the curve (AUC) of the NLR, PLR, LMR, NPL-IRS score, prognosis nutritional index, and RDW were 0.735, 0.710, 0.719, 0.768, 0.728, and 0.599, respectively. Prostate specific antigen density (PSAD), prognosis nutritional index, NPL-IRS score, Gleason score (GS), and positive surgical margin (PSM) were independent risk factors. A new nomogram model was constructed based on these parameters to predict one-year, three-year, and five-year RFS with the AUC of 0.828, 0.867, and 0.892, which could provide an additional clinical net benefit. In external validation set, the AUCs were 0.847, 0.894, and 0.906, respectively.
Conclusions:
Higher preoperative NLR, PLR, and RDW or lower LMR and prognosis nutritional index are associated with poorer RFS. The nomogram prediction model based on preoperative PSAD, prognosis nutritional index, NPL-IRS score, GS, and PSM provides important postoperative treatment guidance.
