Assessment of Cancer Detection Rate of Mri Guided Prostate Biopsy in Relation to Prostate Imaging Reporting and Data System at Lagos University Teaching Hospital

Introduction

Prostate cancer is the second most common cancer in men worldwide and the fifth most common cause of cancer death in men [1]. prostate cancer incidence increases as men age; as many as 60% of men over 65years of age may be diagnosed with prostate cancer [2]. It is commonly diagnosed in men aged 65 – 74 years with the median age at diagnosis being 66 years [3]. Trans Rectal Ultrasound (TRUS)-guided biopsy is performed to confirm diagnosis following abnormal finding(s) from the screening modalities. The risk of over-diagnosis of clinical insignificant cancers remains worrisome and significant. Clinically significant cancer is defined via imaging or histology as a tumour with a volume greater than 0.5cm³ or a Gleason score of greater or equals to seven/ 6mm in length respectively [4]. Magnetic Resonance Imaging (MRI) is a tool with growing importance in prostate cancer evaluation; the introduction of Multiparametric MRI (mp-MRI) has seen an increase in accuracy for localisation and detection of Prostate Cancer (Pca) [5-8]. Multiparametric MRI involves the use of specific sequences such as T2W, DWI and DCE. Prostate MRI reporting is getting more structured by implementation and improvement of the Prostate Imaging Reporting And Data System (PI-RADS) [9,10]. MRI-Guided Prostate Biopsy (MRGB) has been shown in various studies to improve the quality of a biopsy after a diagnostic MRI and is now becoming the new gold standard [11,12]. A clinical method for targeted prostate biopsy is the targeted TRUS-guided biopsy of Clinical Suspicious Lesions (CSL) on prostate MRI also known as Cognitive Targeted Biopsy (CTB) or visually- registered targeted biopsy. Many studies have shown promising results for cognitive targetting of clinical suspicious lesions [13,14]. This study aims to evaluate the detection rates of prostate cancer for magnetic resonance imaging-guided prostate biopsy related to Prostate Imaging Reporting and Data System score.

Materials and Methods

The study was prospectively conducted in the urology unit of the Lagos University Teaching Hospital (LUTH) between February 2024 to March 2025. The sample size was estimated to be 43 (at attrition rate of 40%) using Cochran’s formula. All male patients who presented for evaluation of their prostatic diseases had DRE and serum PSA assay performed on them. The study protocol was approved by the Lagos University Teaching Hospital Human Research and Ethics Committee (LUTHHREC). Inclusion criteria were the presence of elevated PSA (>4ng/ml) or abnormal prostatic findings on DRE and ultrasonography. Patients with clinical and radiologic features of advanced malignant disease of the prostate were excluded. Fifty patients met the inclusion criteria and gave consent to participate in the study. All the subjects were sent for multiparametric MRI of the prostate (mp-MRI) using a Toshiba Elan Vantage (eS Edition) with a magnetic strength of 1.5T. Suspicious areas and corresponding zones were noted following review of the MRI images. Each patient was worked up for TRUSguided prostate biopsy using unit protocol of preparation viz; stoppage of anticoagulants and anti-platelets, administration of rectal suppository  (bisacodyl 10mg) the night before the procedure and prophylactic intravenous antibiotics (levofloxacin 500mg stat + metronidazole 500mg stat) prior to the procedure. Patients were placed in lateral decubitus position; a preliminary DRE was done by the researcher and findings documented. About 10mls of 2% xylocaine gel was instilled intra-rectally for topical anaesthesia and lubrication 15 minutes before biopsy was commenced. TRUS of prostate was performed with a 7.5MHz trans rectal probe using a digital ultrasound scanner (S22; SonoScape Medical Corp, Guangdong, China). A peri-prostatic nerve block was done using 5ml of 1% lidocaine injection to the apex and both sides of the prostate base. All patients had tissue cores taken from suspicious lesions on pre biopsy multiparametric MRI with the aid of an 18G core biopsy needle mounted on a spring-loaded automated biopsy gun. Patients with negative findings on mp-MRI had systematic biopsies of 18 cores, including sonographic suspicious areas. Relevant information including demographic data, examination findings, indications for biopsy and results of histopathology were obtained using a pro forma. The data were analysed using Statistical Package for the Social Sciences (SPSS) version 25.0. The results are described in statistical indices (sensitivity, specificity, NPV, PPV and accuracy). Receiver Operating Characteristic (ROC) curve and Pearson’s correlation coefficient were used with the P<0.05 considered as significant.

Results

A total of fifty (50) patients were studied with age range of 51 – 83 years and a mean age of 65.90 ± 7.56 years; most patients (22 i.e. 44%) falling within the 61 – 70 years (Figure 1) bracket while only one patient was aged over 80 years(2%). Mean age of patients with benign prostatic hyperplasia is 66.22 ± 6.916 years while the mean age of patients with adenocarcinoma of the prostate was 65.72 ± 8.005 years

Figure 1:  Age Distribution Of Patients Who Had Prostate Biopsy (N=50).

Thirty two of the fifty patients (64%) had prostatic adenocarcinoma while eighteen patients (36%) had benign features on histology of the prostate. The range of the prostate-specific antigen (PSA) was from 0.97 – 20ng/ml. The mean PSA for patients diagnosed with benign prostatic hyperplasia and prostate cancer were 9.8150 ± 4.683ng/ml and 14.2875 ± 4.745ng/ml respectively (Table 1). See Table 1

Table 1: Mean Psa For Patients With Benign And Malignant Prostate Disease.

Out of the thirty-two (32) patients who had prostate cancer, 19 (59.4%) had Gleason score of 7; making it the highest occurring Gleason score of the group (intermediate grade). Nine patients had Gleason score of 6 while three and one patient had Gleason score of 8 and 9 respectively. No patient had Gleason score of 10 (Table 2). See Table 2

Table 2: Distribution Of Gleason Score Of Patients With Prostate Cancer.

Twenty-two patients with prostate malignancy had prostate imaging reporting and data system score (PIRADS) of 5; only 3 patients with PIRADS 5 lesions on MRI had benign features. None of the patients with PIRADS 1 – 3 lesions on MRI had malignant histology (Table 3).

See Table 3

Table 3: Pirads Score Distribution Of Patients.

The mean prostate imaging reporting and data system score (PIRADS) on multiparametric MRI of the prostate of patients who were found histologically to have prostate adenocarcinoma and benign features are 4.63 and 2.83 respectively (Table 4) ; which represents a positive correlation with histological outcome.

See Table 4

Table 4: Mean Pirads Score For Benign And Malignant Prostate Disease.

Multiparametric MRI of the prostate in prostate cancer detection

Ninety six (96) of the 900 prostatic zones had features suggestive of malignancy on MRI (PIRADS 4 & 5) and resulted in a histology of adenocarcinoma (true positive) while 777 prostatic zones had non-malignant feature with benign feature histologically (86.3%), true negative. False negative and false positive accounted for 15 and 12 prostatic zones respectively of the 900 zones (Table 5). See Table 5

Table 5: Multiparametric Mri In Prostate Cancer Detection.

The sensitivity, specificity, positive predictive value, negative predictive value and accuracy of MRI in prostate adenocarcinoma detection were 86.49%, 98.48%, 88.89%, 98.11% and 97.00 respectively (Table 6)

See Table 6

Table 6: Predictive Value Of Mpmri In Prostate Cancer Detection.

Multiparametric magnetic resonance imaging show a high strength in the ability to correctly determine the absence of prostate cancer. This translates to a high specificity of 98.48% for mMRI. The area under the Receiver Operating Characteristic (ROC) curve for mpMRI was 0.933. The mpMRI curve is close to 1 which signifies a high accuracy

See Table 7

Table 7:  Area Under Curve Of Receiver Operating Characteristic For Mpmri Area Under The Curve.

Out of the 32 patients with prostate cancer in this study, 10 had PIRADS 4 lesions while 22 had PIRADS 5 lesions; this gives an overall cancer detection rate for cognitive targeted biopsy to be 64%(32 of 50 patients). Cancer Detection Rate (CDR) was observed to correlate positively with increasing PIRADS score (Table 8). Ten (10) of 12 patients with PIRADS 4 lesions had prostate cancer (CDR of 83.3%) while 22 of 25 patients with PIRADS 5 had prostate cancer (CDR of 88%). See Table 8

Table 8: Cancer Detection Rates For Cognitive Targeted Biopsy.