A Simple Prostate Cancer Scoring System Based On The Prostate Cancer Related Parameters

Objective:A prediction model for simple diagnosis of prostate cancer is established to evaluate the possibility of prostate cancer and increase the diagnostic rate of prostate cancer.It provides a more reliable basis for biopsy in suspicious prostate cancer patients.Methods:From September 2010 to September 2017,122 male（PSA≤20ng/ml）who underwent guilding prostate biopsy with transrectal ultrasound in The First Hospital of Jilin University were received retrospectively.69 patients with prostate cancer were defined as prostate cancer group,and 53 non-prostate cancer patients were defined as non prostate cancer group.The significance and value of DRE,PSA,f/t PSA,PSAD and MRI in diagnosis of prostate cancer were analyzed for two groups.Based on the above indicators,the prediction model was established by SPSS software,and the sensitivity,specificity and critical value were analyzed through the ROC curve.Results:The prediction model of this experiment was a variant of prostate cancer as a variable.Age,BMI,PSA,f/tPSA,PSAD,MRI,DRE were used as the variables.The results showed that PSA,f/tPSA,PSAD,MRI and DRE had significant effects on the results of prostate biopsy,and statistically significant（P<0.05）.The predictive model of prostate cancer is as follows:The risk of prostate cancer was calculated with tPSA>11.12ng/ml,f/tPSA<0.20,PSAD>0.25ng/ml/ml as 1 points（all median）^[1],DRE positive,and MRI positive for 2 points.We found that the scores were 0,1,2,3,4,5,6,and 7.When the patient’s score was 3⁷,the detection rate of prostate cancer was higher than 0².The area（AUC）under the ROC curve of the model is 0.771.The P value is the probability of the occurrence of prostate cancer,the range of which is 0¹.The critical value（cutoff）of P was 3,the sensitivity was 0.493,the specificity was 0.925,and the correct index was 0.417,which indicated that the diagnostic value of the scoring system was high.Based on this scoring system,we found a significant increase in the detection rate of prostate cancer:therefore,in clinical work,when we find a patient’s score is greater than 3 points,we can consider that the probability of prostate cancer is significantly increased.Conclusions:1.When PSA≤20ng/ml,using this scoring model is helpful to evaluate the possibility of prostate cancer,it is a diagnostic tool with high accurate rate of screening for prostate cancer.It can improve the specificity of the diagnosis in prostate cancer and provide reliable evidence for the clinical prostatic biopcy.2.This scoring model for predicting the probability of P is 3,when P is larger than 3 there is a strong indication for prostate biopsy.