J Gandhara Med Dent Sci

ORIGINAL ARTICLE

ABSTRACT

OBJECTIVES

This study aimed to evaluate the diagnostic accuracy of mpMRI using PI-

RADS v2 in a prospective study, with histopathology as the reference

standard.

METHODOLOGY

This prospective single-centre diagnostic accuracy study was conducted

between January 2022 and June 2023 in the Department of Radiology,

Hayatabad Medical Complex, Peshawar, Pakistan, in accordance with

STARD guidelines. A total of 127 men (aged 40-85 years) with elevated

prostate-specic antigen (PSA >4 ng/dL) and abnormal digital rectal

examination ndings were enrolled by a consecutive sampling method. The

study was conducted at the Department of Diagnostic Radiology, Hayatabad

Medical Complex, Peshawar, over a period of 6 months. All participants

underwent mpMRI with PI-RADS scoring followed by systematic prostate

biopsy. Clinically signicant prostate cancer was dened as a Gleason score

≥7. Diagnostic accuracy metrics, including sensitivity, specicity, positive

predictive value (PPV), negative predictive value (NPV), and overall

accuracy, were calculated.

RESULTS

Prostate cancer was histopathologically conrmed in 58.3% (74/127) of

patients. Using a predened diagnostic threshold of PI-RADS ≥4 versus <4,

mpMRI demonstrated a sensitivity of 89.2%, specicity of 81.1%, PPV of

86.8%, NPV of 84.3%, and overall accuracy of 85.8%. Performance was

superior in patients with PSA >10 ng/dL (92.5% sensitivity) and for PI-RADS

5 lesions (94.7% concordance with high-grade cancer). False negatives were

predominantly anterior zone tumors (75%), while false positives were

primarily due to prostatitis. Inter-reader agreement was substantial (κ =

0.72).

CONCLUSION

In this study, mpMRI using PI-RADS v2 showed good diagnostic performance

and reproducibility for detecting csPCa in a Pakistani cohort. Limitations

persist in anterior zone assessment, and inammatory mimics were observed,

underscoring the need for careful interpretation and optimized biopsy

strategies.

KEYWORDS: Prostate Cancer, PI-RADS, Multiparametric MRI, Diagnostic

Accuracy, Gleason Score, Prostate Biopsy

How to cite this article

Rahman MU, Wahid G, Jawaid HA,

Sajid A, Khan M, Ali G, et al.

Diagnostic Accuracy of MRI PI-

RADS to Diagnose Prostate Cancer,

Keeping Histopathology as Gold

Standard. J Gandhara Med Dent Sci.

2026;13(1):49-53.

Date of Submission: 21-11-2025

Date Revised: 19-12-2025

Date Acceptance: 19-12-2025

Trainee Medical Ocer ,

Department of Radiology, Hayatabad

Medical Complex, Peshawar

Associate Professor, Department of

Radiology, Hayatabad Medical

Complex, Peshawar

Correspondence

Ghazala Wahid, Associate Professor,

Department of Radiology, Hayatabad

Medical Complex, Peshawar

+92-336-5249949

ghazalawahid3@gmail.com

DIAGNOSTIC ACCURACY OF MRI PI-RADS TO DIAGNOSE PROSTATE CANCER, KEEPING

HISTOPATHOLOGY AS GOLD STANDARD

Muneeb Ur Rahman

, Ghazala Wahid

, Haza Ayesha Jawaid

, Adnan Sajid

, Mehreen Khan

Gohar Ali

, Mahnoor Rehman Khan

INTRODUCTION

Prostate cancer remains a signicant health concern

globally, ranking as the second most common cancer in

men and a major contributor to cancer-related morbidity

and mortality.

In Pakistan, prostate cancer incidence is

rising, yet local diagnostic performance data for

mpMRI remain limited.

Although PI-RADS has been

extensively validated in Western populations, data from

South Asian and other low- and middle-income

countries are comparatively scarce. Despite

histopathology being the denitive method for

diagnosing prostate cancer, its invasive nature poses

various risks, such as infection, bleeding, and

discomfort for the patient.

Consequently, noninvasive

imaging methods, especially multiparametric magnetic

resonance imaging (mpMRI) using the Prostate

Imaging-Reporting and Data System (PI-RADS), have

become increasingly important for identifying and

assessing prostate cancer risk.

The PI-RADS system

standardizes the interpretation of MRI, enhancing

consistency in diagnosis and assisting in the detection

of clinically signicant prostate cancer (Gleason score

≥ 7).

Multiparametric MRI has become integral to

prostate cancer diagnosis by improving the detection of

clinically signicant disease while reducing

https://doi.org/10.37762/jgmds.13-1.787

January - March 2026

J Gandhara Med Dent Sci

unnecessary biopsies. The Prostate Imaging-Reporting

and Data System (PI-RADS) standardizes MRI

acquisition, interpretation, and reporting, facilitating

consistent risk stratication. However, reported

diagnostic accuracy varies across institutions due to

dierences in patient spectrum, MRI protocols, and

reader experience. The results have the potential to

improve patient care by minimizing unnecessary

biopsies and enabling the quick identication of

significant, high-risk prostate cancer. The integration of

MRI-based PI-RADS into standard clinical practice has

greatly improved prostate cancer diagnosis by

enhancing the detection of clinically important tumors

while also lowering the rates of diagnosing indolent or

low-risk conditions.

However, variations in radiologist

experience, MRI scanner protocols, and patient

populations can influence diagnostic performance,

leading to discrepancies in reported sensitivity and

specificity across dierent institutions.

This variability

necessitates localized validation studies to ensure that

PI-RADS maintains high diagnostic accuracy in diverse

healthcare settings, including resource-limited

environments where access to advanced imaging may

be constrained. Despite its advantages, the widespread

adoption of MRI PI-RADS faces challenges, including

cost, accessibility, and interobserver variability in

interpretation.

Recent advancements in articial

intelligence (AI) and machine learning show promise in

augmenting PI-RADS assessments by improving lesion

detection and risk stratication.

However, further

research is needed to standardize imaging protocols,

optimize radiologist training, and validate PI-RADS in

underrepresented populations, including Asian and

Middle Eastern cohorts, where prostate cancer proles

may dier from Western populations.

To address this

local evidence gap, the present study was designed as a

prospective diagnostic accuracy and inter-reader

reliability study of PI-RADS v2 in a Pakistani

population, with additional PSA-stratified performance

analysis. This framing emphasizes external validation

rather than methodological novelty.

METHODOLOGY

This prospective diagnostic accuracy study was

conducted at the Department of Radiology, Hayatabad

Medical Complex, Peshawar, from January 2022 to

June 2023, and was reported in accordance with

STARD recommendations. Ethical approval was

obtained from the Institute bearing Ref. No. HMC-

QAD-F00/1670, Dated: 26-12-2023 and CPSP No -787

Men aged 40-85 years presenting with a prostate-

specific antigen (PSA) level greater than 4 ng/dL and

abnormal digital rectal examination ndings were

enrolled. Patients with a history of prior prostate biopsy

or surgery, use of 5-α-reductase inhibitors within the

preceding six months, recent treatment for prostatitis, or

contraindications to magnetic resonance imaging were

excluded. All included participants were biopsy-naïve.

Multiparametric magnetic resonance imaging (mpMRI)

of the prostate was performed using a 1.5-Tesla scanner

with a phased-array pelvic coil. Imaging followed

Prostate Imaging-Reporting and Data System (PI-

RADS) version 2 recommendations and included axial,

sagittal, and coronal T2-weighted sequences (slice

thickness 3 mm), diusion-weighted imaging with b-

values of 0, 800, and 1400 s/mm² along with apparent

diusion coecient maps, and dynamic contrast -

enhanced imaging using a gadolinium-based contrast

agent. MRI examinations were independently

interpreted by three radiologists with 6-12 years of

experience in prostate MRI, who were blinded to

histopathological ndings but aware of PSA levels and

digital rectal examination status. Prior to study

initiation, readers underwent calibration using standard

PI-RADS v2 training cases. Discrepancies in scoring

were resolved by consensus, and the nal PI-RADS

score was used for analysis. For the primary diagnostic

analysis, a PI-RADS score of ≥4 was considered

positive for clinically signicant prostate cancer, while

PI-RADS 3 lesions were analyzed descriptively but

classified as negative in the primary 2×2 analysis. All

patients subsequently underwent systematic 12-core

transrectal ultrasound-guided prostate biopsy without

MRI-targeted sampling. Histopathological examination

served as the reference standard, with clinically

significant prostate cancer dened as a Gleason score of

≥7. Diagnostic performance was assessed by

calculating sensitivity, specicity, positive predictive

value, negative predictive value, overall accuracy, and

area under the receiver operating characteristic curve,

along with 95% condence intervals. Comparisons

between PSA subgroups were performed using chi-

square tests for proportions, and subgroup analyses

were pre-specied. Inter-reader agreement was

evaluated using Fleiss' kappa for both ordinal PI-RADS

scores and dichotomized categories.

RESULTS

A total of 127 male patients aged 40-85 years were

included in the study. The mean age was 65.4 ± 8.2

years, with a median PSA level of 9.6 ng/dL (IQR: 6.2-

14.1). Histopathological examination conrmed

prostate cancer in 74 patients (58.3%), all of whom had

clinically signicant disease (Gleason score ≥7). The

most frequent Gleason score was 7 (3+4), followed by

7 (4+3), while high-grade disease (Gleason ≥8) was

identified in 17 patients (23.0%). Using a PI-RADS

threshold of ≥4, multiparametric MRI demonstrated

good diagnostic performance for clinically signicant

prostate cancer, with a sensitivity of 89.2% and

Diagnostic Accuracy of MRI PI-Rads to Diagnose Prostate Cancer

January - March 2026

J Gandhara Med Dent Sci

specificity of 81.1%. Diagnostic accuracy was higher in

patients with PSA levels >10 ng/dL than in those with

PSA levels 4-10 ng/dL. Cancer detection increased

progressively with higher PI-RADS categories,

reaching the highest yield in PI-RADS 5 lesions. Inter-

reader agreement among radiologists was substantial,

supporting the reproducibility of PI-RADS scoring in

this cohort.

Table 1: Baseline Clinical Characteristics of Study Participants

(n = 127)

Variable Value

Age (years), mean ± SD 65.4 ± 8.2

PSA (ng/dL), median (IQR) 9.6 (6.2-14.1)

Prostate volume (g), mean ± SD 52.3 ± 18.7

Prostate cancer on histopathology 74 (58.3%)

Clinically signicant cancer (GS ≥7) 74 (58.3%)

Table 2: Diagnostic Performance of PI-RADS v2 (Threshold ≥4)

for Clinically Signicant Prostate Cancer

Parameter Value (95% CI)

Sensitivity 89.2% (83.1-93.6)

Specificity 81.1% (74.3-86.7)

Positive predictive value 86.8% (80.9-91.3)

Negative predictive value 84.3% (77.8-89.4)

Overall diagnostic accuracy 85.8% (81.2-89.6)

Area under the ROC curve 0.87 (0.82-0.92)

Table 3: Cancer Detection Rate According to PI-RADS

Category and PSA Level

Variable csPCa Detected n (%)

PI-RADS 3 (n = 24) 2 (8.3%)

PI-RADS 4 (n = 38) 30 (78.9%)

PI-RADS 5 (n = 38) 36 (94.7%)

PSA 40-10 ng/dL (n = 59) Sensitivity 84.6%

PSA >10 ng/dL (n = 68) Sensitivity 92.5%*

DISCUSSION

This prospective diagnostic accuracy study supports the

clinical utility of mpMRI using PI-RADS v2 for

detecting csPCa in a Pakistani cohort. Our ndings are

consistent with international literature while

highlighting region-specic performance

characteristics. The observed sensitivity of 89.2% (95%

CI: 83.1-93.6%) and specicity of 81.1% (95% CI:

74.3-86.7%) position our results within the spectrum of

recent multicenter validation studies.

While our

sensitivity exceeds that reported by Bangash et al.

(84.85%) , it remains more conservative than Almolla

et al.'s ndings (94.2%).

These variations likely

reect differences in MRI eld strength (1.5T vs 3T),

with mounting evidence suggesting 3T scanners

provide superior lesion characterization.

The

substantial inter-reader agreement (Fleiss' κ=0.72)

reinforces PI-RADS v2’s reliability as a standardized

reporting system,

though the lower agreement for

transition zone lesions (κ=0.65) highlights persistent

interpretation challenges in this anatomically complex

region.

These ndings are particularly relevant given

the increasing adoption of PI-RADS v2.1, which

specifically renes transition zone evaluation criteria.

Our detailed analysis of diagnostic errors revealed

important patterns. The false-negative rate (10.8%) was

dominated by anterior zone tumors (6/8 cases),

confirming this region as a persistent diagnostic blind

spot.

Recent technical advances, including specialized

anterior coil placement and transperineal biopsy

approaches, show promise in addressing this

limitation.

Among false positives (18.9%),

inammatory conditions accounted for the majority of

cases, underscoring the need for advanced imaging

biomarkers to dierentiate malignancy from

prostatitis.

Emerging techniques like restriction

spectrum imaging and quantitative perfusion analysis

demonstrate particular promise in this regard.

The

strong PSA-dependent performance gradient (84.6% vs

92.5% sensitivity for PSA 4-10 vs >10 ng/dL) suggests

MRI may be most eectively deployed after initial PSA

triage.

This aligns with recent cost-eectiveness

analyses favoring MRI after PSA exceeds 10 ng/dL.

Our nding that age and prostate volume minimally

affected accuracy supports MRI's broad applicability

across diverse patient demographics, though emerging

data suggest that adjusted PI-RADS thresholds may

benet patients with very large prostates (>80).

18,19

The

near-perfect concordance (94.7%) between PI-RADS 5

lesions and high-grade cancer on histopathology

supports growing interest in biopsy-free diagnostic

pathways for unequivocal cases. However, the lower

PPV for PI-RADS 4 (78.9%) reinforces the need for

histologic conrmation in these indeterminate cases.

Recent technical advances in MRI-guided biopsy,

particularly the combination of cognitive and software

fusion approaches, may further optimize targeted

sampling.

The optimal biopsy strategy continues to

evolve, with evidence supporting combined systematic

and targeted approaches for comprehensive

assessment.

LIMITATIONS

The study includes a single-center design, the absence

of MRI-targeted biopsy, potential spectrum bias due to

the inclusion of PSA/DRE-positive patients only, and a

lack of PI-RADS v2.1 comparison. These factors limit

generalizability to screening populations.

CONCLUSIONS

In this diagnostic accuracy study, mpMRI using PI-

RADS v2 demonstrated good diagnostic performance

and substantial inter-reader agreement for detecting

clinically signicant prostate cancer in a Pakistani

population. The ndings support its use in similar

clinical settings, while caution is advised, particularly

Diagnostic Accuracy of MRI PI-Rads to Diagnose Prostate Cancer

January - March 2026

J Gandhara Med Dent Sci

nflammatory conditions.for anterior zone lesions and i

CONFLICT OF INTEREST:

None

FUNDING SOURCES: None

REFERENCES

1. Bergengren O, Pekala KR, Matsoukas K, Fainberg J, Mungovan

SF, Bratt O, et al. 2022 update on prostate cancer epidemiology

and risk factors - A systematic review. Eur Urol. 2023

Aug;84(2):191–206.

https://doi.org/10.1016/j.eururo.2023.05.004.

2. Akhtar S, Hassan F, Ahmad S, El-Aendi MA, Khan MI. The

prevalence of prostate cancer in Pakistan: a systematic review

and meta-analysis. Heliyon. 2023 Oct;9(10):e20350.

https://doi.org/10.1016/j.heliyon.2023.e20350.

3. Borghesi M, Ahmed H, Nam R, Schaeer E, Schiavina R,

Taneja S, et al. Complications after systematic, random, and

image-guided prostate biopsy. Eur Urol. 2017 Mar;71(3):353–

65. https://doi.org/10.1016/j.eururo.2016.08.004. PMID:

27542364

4. Turkbey B, Rosenkrantz AB, Haider MA, Padhani AR, Villeirs

G, Macura KJ, et al. Prostate Imaging Reporting and Data

System version 2.1: 2019 update of Prostate Imaging Reporting

and Data System version 2. Eur Urol. 2019 Sep;76(3):340–51.

https://doi.org/10.1016/j.eururo.2019.02.033. PMID: 30824206

5. Ahmed HU, El-Shater Bosaily A, Brown LC, Gabe R, Kaplan

R, Parmar MK, et al. Diagnostic accuracy of multiparametric

MRI and TRUS biopsy in prostate cancer (PROMIS): a paired

validating conrmatory study. Lancet. 2017 Feb

25;389(10071):815–22. https://doi.org/10.1016/S0140-

6736(16)32401-1. PMID: 28110982

6. Kasivisvanathan V, Rannikko AS, Borghi M, Panebianco V,

Mynderse LA, Vaarala MH, et al. MRI-targeted or standard

biopsy for prostate-cancer diagnosis. N Engl J Med. 2018 May

10;378(19):1767–77. https://doi.org/10.1056/NEJMoa1801993.

PMID: 29552975

7. Greer MD, Shih JH, Lay N, Barrett T, Bittencourt L, Borofsky

S, et al. Interreader variability of PI-RADS version 2 in

detecting and assessing prostate cancer lesions at MRI. AJR Am

J Roentgenol. 2019 Jun;212(6):1197–205.

https://doi.org/10.2214/AJR.18.20536. PMID: 30888917

8. Padhani AR, Barentsz J, Villeirs G, Rosenkrantz AB, Margolis

DJ, Turkbey B, et al. PI-RADS Steering Committee: The PI-

RADS multiparametric MRI and MRI-directed biopsy pathway.

Radiology. 2019 Aug;292(2):464–74.

https://doi.org/10.1148/radiol.2019182946. PMID: 31318335

9. Greer MD, Brown AM, Shih JH, Summers RM, Marko J, Law

YM, et al. Accuracy and agreement of PI-RADS v2 for prostate

cancer mpMRI: a multireader study. J Magn Reson Imaging.

2017 Feb;45(2):579–85. https://doi.org/10.1002/jmri.25372.

PMID: 27547857

10. Kaissis G, Ziegelmayer S, Lohöfer F, Algül H, Eiber M,

Weichert W, et al. A machine learning model for the prediction

of survival and tumor subtype in pancreatic ductal

adenocarcinoma from preoperative diusion-weighted imaging.

Eur Radiol Exp. 2019 Dec;3(1):41.

https://doi.org/10.1186/s41747-019-0119-0. PMID: 31749092

11. Bangash M, Khan SS, Naimat M, Alvi SA, Afzal SK, Laique T.

Frequency of signicant prostate cancer (Gleason score ≥7) in

suspected patients having PI-RADS score ≥4: descriptive study.

Pak J Med Health Sci. 2021;15(10):2625.

https://doi.org/10.53350/pjmhs2115102625.

12. Almola R, Eltohamy M, Tantawy H, Desoky E, Tharwat E.

Accuracy of T2, DW-MRI and total PI-RADS v2 in

discriminating a clinically signicant prostatic nodule. Zagazig

Univ Med J. 2022;0(0):0–0.

13. Weinreb JC, Barentsz JO, Choyke PL, Cornud F, Haider MA,

Macura KJ, et al. PI-RADS Prostate Imaging–Reporting and

Data System: 2015, version 2. Eur Urol. 2016 Jan;69(1):16–40.

https://doi.org/10.1016/j.eururo.2015.08.052. PMID: 26427566

14. Cool DW, Zhang X, Romagnoli C, Izawa JI, Romano WM,

Fenster A. Evaluation of MRI-TRUS fusion versus cognitive

registration accuracy for MRI-targeted, TRUS-guided prostate

biopsy. AJR Am J Roentgenol. 2015 Jan;204(1):83–91.

https://doi.org/10.2214/AJR.14.12681. PMID: 25539289

15. Chatterjee A, Bourne RM, Wang S, Devaraj A, Gallan AJ,

Antic T, et al. Diagnosis of prostate cancer with noninvasive

estimation of prostate tissue composition using hybrid

multidimensional MR imaging: a feasibility study. Radiology.

2018 Jun;287(3):863-73.https://doi.org/10.1148/radiol.20181711

30. PMID: 29478306

16. Wegelin O, van Melick HHE, Hooft L, Bosch JLHR, Reitsma

HB, Barentsz JO, et al. Comparing three dierent techniques for

MRI-targeted prostate biopsies: a systematic review of in-bore

vs MRI–TRUS fusion vs cognitive registration. Eur Urol. 2017

Apr;71(4):517–31. https://doi.org/10.1016/j.eururo.2016.08.023.

PMID: 27593531

17. Ahdoot M, Wilbur AR, Reese SE, Lebastchi AH, Mehralivand

S, Gomella PT, et al. MRI-targeted, systematic, and combined

biopsy for prostate cancer diagnosis. N Engl J Med. 2020 Mar

5;382(10):917–28. https://doi.org/10.1056/NEJMoa1910038.

PMID: 32130814

18. van der Leest M, Cornel E, Israël B, Hendriks R, Padhani AR,

Hoogenboom M, et al. Head-to-head comparison of TRUS-

guided prostate biopsy versus multiparametric MRI with

subsequent MRI-guided biopsy in biopsy-naïve men with

elevated PSA: a large prospective multicenter study. Eur Urol.

2019 Apr;75(4):570–8.

https://doi.org/10.1016/j.eururo.2018.11.023. PMID: 30545785

19. Huang C, Cai ZQ, Qiu F, Pu JX, Xi QL, Wei XD, et al. New

model of PI-RADS and adjusted PSA density of peripheral zone

improves the detection rate of initial prostate biopsy: a

diagnostic study. Asian J Androl. 2022 Jan;25(1):126–34.

https://doi.org/10.4103/aja202183. PMID: 35120623

20. Wang YH, Liang C, Zhu FP, Zhou TR, Li J, Wang ZJ, et al.

Improving the understanding of PI-RADS in practice:

characteristics of PI-RADS 4 and 5 lesions with negative

biopsy. Asian J Androl. 2023 Mar;25(2):217–23.

https://doi.org/10.4103/aja202185. PMID: 36944140

21. Deng R, Liu Y, Wang K, Ruan M, Li D, Wu J, et al.

Comparison of MRI articial intelligence-guided cognitive

fusion-targeted biopsy versus routine cognitive fusion-targeted

prostate biopsy: a randomized controlled trial. BMC Med. 2024

Dec;22(1):530. https://doi.org/10.1186/s12916-024-03567-8.

PMID: 39741139

22. Ippoliti S, Fletcher P, Orecchia L, Miano R, Kastner C, Barrett

T. Optimal biopsy approach for detection of clinically

signicant prostate cancer. Br J Radiol.

2021;95(1131):20210413. https://doi.org/10.1259/bjr.20210413.

PMID: 35103556

Diagnostic Accuracy of MRI PI-Rads to Diagnose Prostate Cancer

January - March 2026

J Gandhara Med Dent Sci

Muneeb Ur Rahman - Concept & Design; Data Acquisition;

Drafting Manuscript ; Critical Revision; Final Approval

Ghazala Wahid - Concept & Design; Data Acquisition; Data

Analysis/Interpretation; Drafting Manuscript; Critical

Revision; Final Approval

Hafiza Ayesha Jawaid - Concept & Design; Data Acquisition;

Drafting Manuscript; Critical Revision; Final Approval

Adnan Sajid - Concept & Design; Data Acquisition; Data

Analysis/Interpretation; Drafting Manuscript; Final Approval

Mehreen Khan - Concept & Design; Data Acquisition; Data

Analysis/Interpretation; Drafting Manuscript; Final Approval

Gohar Ali - Concept & Design; Data Acquisition; Data

Analysis/Interpretation; Drafting Manuscript; Final Approval

Mahnoor Rehman Khan - Concept & Design; Data

Acquisition; Drafting Manuscript; Final Approval

AUTHORS CONTRIBUTION

The authors accept responsibility for all aspects of the work

and will ensure that any concerns regarding the accuracy or

integrity of any part are properly investigated and resolved.

LICENSE: JGMDS publishes its articles under a Creative Commons Attribution Non-Commercial Share-Alike license (CC-BY-NC-SA 4.0).

COPYRIGHTS: Authors retain the rights without any restrictions to freely download, print, share and disseminate the article for any lawful purpose.

It includes scholarlynetworks such as Research Gate, Google Scholar, LinkedIn, Academia.edu, Twitter, and other academic or professional networking sites.

Diagnostic Accuracy of MRI PI-Rads to Diagnose Prostate Cancer

January - March 2026