Keywords
prostate carcinoma - PIRADS - mpMRI of prostate - prostate specific antigen
Prostate carcinoma (PCa) is the second leading cause of carcinoma-related death in
men worldwide with an incidence of 7.3%.[1] Among the Asian countries, the standardized incidence rates of prostate carcinoma
is lowest in Nepal (1.5/100,000), after Bhutan.[2] Pre-biopsy magnetic resonance imaging (MRI) of prostate is the next step in the
evaluation of patients having abnormal digital rectal examination and/or elevated
prostate specific antigen (PSA). Pre-biopsy MRI has negative predictive value of 82.4%
for diagnosis of overall PCa and 88.1% for diagnosis of clinically significant prostate
carcinoma (csPCa).[3] The prevalence of “any” PCa in PIRADS 3 lesion is 25% (95% CI: 22 to 29%) and csPCa
is 11% (95% CI: 8 to 14%).[4]
Prostate Imaging-Reporting and Data System (PIRADS) published by the European Society
of Urogenital Radiology (ESUR) is a common terminology devised to standardize the
acquisition, interpretation, and reporting of prostate multiparametric MRI (mpMRI)
data that was originally adopted from breast imaging assessment in 2012 (PIRADS v1)
and later updated in 2015 as PIRADS v2 and 2019 PIRADS v2.1.[5] The PIRADS v2 scoring system depends on diffusion-weighted image (DWI) “scores”
in peripheral zone and T2-weighted (T2W) “scores” in transitional zone lesion.[6] PIRADS v2 uses a five-point scale, ranging from 1 to 5. PIRADS categories 1 and
2 have a very low and low risk of csPCa and further biopsy is not recommended. There
is a high or very high likelihood of clinically significant carcinoma in PIRADS 4
and 5 categories and prostate biopsy is warranted.[7]
[8] PIRADS 3 lesion is called equivocal or indeterminate due to lack of universally
agreed definition and highly variable risk of prostate carcinoma (9–46% depending
on clinical profile) posing challenge in definitive management. A direct impact on
the diagnostic accuracy of prostate carcinoma was observed when they were treated
as positive, negative, or excluded.[9] If we consider PIRADS 3 as a “negative,” a quarter of the prostate carcinoma diagnosis
would get missed out. On the other hand, if we consider it as a positive disease,
there is a risk of overdiagnosis and overtreatment. Studies also showed that the risk
of clinically significant PCa among PIRADS 3 was low and most of it could be kept
in active surveillance.[9]
[10] The primary aim of clinical and radiological evaluation is to improve prostate carcinoma
detection rate and avoid unwanted biopsies.[11] There are a number of risk-based models/calculators to guide the optimal management
of the PIRADS 3 prostate lesion without compromising the disease outcomes. Age of
the patient, PSA kinetics, family history, prior biopsy, and various biomarkers are
in use to predict the risk of prostate carcinoma in equivocal cases.[12]
The purpose of this study is to explore the clinicopathological characteristics of
PIRADS 3 lesions and find out the detection rate of csPCa in PIRADS 3 lesion by transrectal
ultrasound (TRUS) guided needle biopsy of prostate.
Materials and Methods
It is a retrospective study done under urology unit under the Department of Surgery,
Nepalese Army Institute of Health Sciences, Nepal, after approval from institutional
review committee (IRC-NAIHS Reg No. 1306). The clinical, radiological, and pathological
data of the patients diagnosed with PIRADS 3 prostate lesion on mpMRI of prostate
from October 2019 to December 2024 were analyzed.
Patients who underwent TRUS guided biopsy of prostate for suspected prostate carcinoma
were reviewed and the data were collected using study proforma. The inclusion criteria
were patients with PIRADS 3 lesion who underwent TRUS guided prostate biopsy for the
first time (biopsy naïve). The exclusion criteria were PIRADS 4 and 5 lesions, finger-guided
prostate biopsy, nonsystematic prostate biopsy, prior biopsy, and metastatic prostate
carcinoma.
The data collected were independent variables, namely, patient demographics, clinical
symptoms, digital rectal examination, prostate volume, serum PSA, PSA density (PSAD),
and size and location of lesion on mpMRI of prostate, and the dependent variables,
namely, histopathological details of the TRUS guided needle biopsy of prostate. The
data collected in study proforma were entered into SPSS 2.0 for analysis. The carcinoma
detection rate of csPCa—ISUP grade group ≥2 in PIRADS 3 lesion—and its relationship
with PSA density (cut off at 0.1 ng.mL/cc) was measured.[13]
Distribution of the study variables was measured in mean/median ± standard deviation.
The difference in measurement were interpreted as statistically significant when p-value was <0.05.
Results
All the patients underwent mpMRI of prostate (3 tesla) for suspected prostate carcinoma
based on digital rectal examination and serum PSA value. High-resolution T2-weighted
image (T2WI), diffusion-weighted image (DWI), and diffusion contrast enhancement images
(DCEI) were obtained and the prostate mpMRI was reported according to PIRADS v2 protocol.[5]
Patients with PIRADS ≥3 underwent systematic as well as cognitive targeted biopsy
after confirming sterile urine and normal coagulation profile. Preprocedural 30 to
60 mL glycerin bowel suppositories were given to empty the rectum, and a single dose
of 60 mg intramuscular Gentamicin was given for antibiotic prophylaxis. Digital rectal
examination was performed. TRUS guided peri-prostatic nerve block was done with 1%
lignocaine (10–20 mL) for pain management. The TRUS guided needle biopsy was done
with 20 cm/18 G core needle biopsy gun. All core biopsy samples were processed and
reported as per the 2019 International Society of Urological Pathology (ISUP) protocols.[13]
Among the cohort of 178 patients subjected to TRUS guided prostate biopsy during the
study period, 34 (19%) of them had PIRADS 3 lesions. The mean age of patients was
67 ± 8 years. A total of 20/34 (59%) had abnormal DRE findings. The prostate size,
PSA value, and PSAD of the study population are listed in [Table 1].
Table 1
Clinicopathological characteristics of PIRADS 3 lesion
|
Variables
|
|
Biopsy
|
|
p-Value
|
|
Histopathology for PCa
(
n
, %)
|
Negative
(20, 59)
|
isPCa
(7, 20.5)
|
csPCa
(7, 20.5)
|
0.03
|
|
Age (years), Mean ± SD
|
65 ± 9.8
|
67 ± 7
|
71 ± 5.2
|
NS
|
|
Prostate volume (cc), Mean ± SD
|
45.7 ± 26.7
|
43 ± 15
|
40 ± 25
|
NS
|
|
Index lesion maximum diameter (mm), Mean
|
8.0
|
10.2
|
18.4
|
|
|
PSA (ng/mL)
|
|
|
|
|
|
Mean
range
|
8.9
4.2–12.4
|
12.6
6–19.6
|
14.2
9–26.1
|
NS
|
|
Mean PSA density (ng/mL/cc)
|
0.06
|
0.21
|
0.35
|
0.01
|
|
Location of lesion (PZ/TZ)
|
18/2
|
6/1
|
6/1
|
NS
|
Abbreviations: csPCa, clinically significant prostate carcinoma; isPCa, insignificant
prostate carcinoma; PCa, prostate carcinoma; PIRADS 3, Prostate Imaging-Reporting
and Data System category 3; PSA, prostate specific antigen; PZ, peripheral zones;
TZ, transitional zones.
Out of 34 cases, 14 (41%) with PIRADS 3 prostate lesion had proven malignancy on histopathological
examination and all were acinar adenocarcinoma. A total of 20.5% (7 out of 34) had
clinically significant PCa (ISUP grade group ≥2). The mean PSA density in PCa was
>0.2 ng/mL/cc. The mean PSAD in insignificant prostate carcinoma (isPCa) was 0.21 ng/ml/cc
and csPCa was 0.35 ng/mL/cc. Majority (85%) of the csPCa had prostate lesion ≥10 mm
in size. Low PSA density index PIRADS 3 prostate lesions at transitional zone had
lower rate of carcinoma. With cut-off PSA density ≥0.15 ng/mL/cc, the sensitivity
and specificity of mpMRI to detect csPCa was 85.7%.
Discussion
The mpMRI utilized three different sequences, namely, T2-weighted (T2W), diffusion-weighted
image (DWI), and dynamic contrast enhancement image (DCEI), and evaluated a total
of 39 different sectors (38 in prostate, 1 in seminal vesicle, and 1 in urethra) for
suspected prostate carcinoma. “Prostate Imaging-Reporting and Data System” (PIRADS)
is based on “DWI score” of peripheral zones (PZ) lesions and “T2W score” of transitional
zones (TZ) lesions.[6] The PIRADS 3 lesion appears heterogenous with ill-defined margins on T2 and appears
isointense or mildly hyperintense on high b-value DWI and focal mild/moderate hypointense
on apparent diffusion coefficient (ADC). The PZ DWI score 3, which appears negative
for DCE (focal and early enhancement), is given final PIRADS category 3, and if positive
(enhancement) given final PIRADS category 4. The TZ score 3 on T2W that has DWI score
4 or less is given final PIRADS category of 3, and the TZ score 3 that has DWI score
5 is given final PIRADS category 4.[6] Multiparametric MRI has been widely used for diagnosis, surveillance, and staging
of patients with prostate carcinoma that has negative predictive value of 80 to 90%.[7] According to the “PRECISION Trail” MRI-guided prostate biopsies are superior to
standard biopsy with adjusted difference of 12% (95% CI, 4–20; p = 0.005), and prebiopsy MRI can avoid prostate biopsy in 28% of the cases.[14] The diagnostic accuracy is limited by the experience of the radiologist. The current
modification of PIRADS into v2 and v2.1 is expected to improve inter-reader variability
and simplify PIRADS assessment for better accuracy in terms of detection, localization,
characterization, and risk stratification of the index prostate lesion.[15]
About three-quarters of all the MRI for suspected prostate carcinoma patients who
have PIRADS ≥3 are likely to undergo invasive procedures. The prevalence of csPCa
(ISUP ≥2) in PIRADS 3 biopsy naïve patients is approximately 21% in comparison to
39 and 73% in PIRADS 4 and PIRADS 5 lesions, respectively.[10] The high rate (61–71%) of benign pathology in PIRADS 3 lesion is commonly due to
benign hyperplasia, inflammation, or fibrosis resulting in false-positive MRI findings.[10]
[16] MRI of prostate often shows no difference in DCE and T2W scores if prostate tumors
are sparse or normal.[17] Presence of small infiltrative gland with large intervening stroma, sparse or more
densely packed tumor cells in PIRADS 3 lesion tends to have clinically insignificant
prostate carcinoma with Gleason grade of 1 to 3.[18] Tumor aggressiveness and the size of the tumor also impact the tumor visibility,
detection, and interpretation on prostate MRI. Around 94 to 95% of the tumor volume
>0.5 cc is correctly classified by PIRADS. Sparse malignant tissue with volume <0.5 cc
is mostly undetectable on MRI and has a high chance of missing it in biopsies.[19] Besides, the interobserver variability among radiologists for PIRADS ≤3 is higher
than PIRADS ≥4 (k −0.552), thus impacting the case detection rate in PIRADS 3 lesions.[17]
The prevalence csPCa in PIRADS 3 is 11% (95% CI: 8 to 14%),[4] which are mostly located in the peripheral zone (18.5%) and less frequently in the
transition zone (6%).[11] In a multicenter study, the risk of unfavorable prostate carcinoma (ISUP GG 3–5
and/or pT3–4) was 2.3-fold with PSAD between 0.15 and 0.20 ng/mL/cc, and the risk
was 5.8-fold for PSAD >0.20 ng/mL/cc.[20] The higher rate (56.7%) of unfavorable disease in radical prostatectomy in this
series of PIRADS 3 disease might be due to inclusion of patients with higher PSA densities,
median 0.17 ng/mL/cc (IQR: 0.12–0.26), higher number of csPCa (ISUP GG ≥2) in needle
biopsy (72%), and prior biopsy patients (33.3%).[20]
In our study, population PSA density, location, and size of index prostate lesions
are important predictors of malignancies in PIRADS 3 lesions. Studies have shown the
disease progression rate of 20 to 78% even in low-risk disease (ISUP 1), directly
depending on the size of the index prostate lesion.[21] It is mentioned that the chance of csPCa with cut-off at index lesion size of 5
to 7 mm is very low and could be monitored with serial MRI only.[21]
[22] The carcinoma detection rate in PIRADS 3 is high (>80%) with PSA threshold of ≥4 ng/mL,
and with high PSA density (>0.1 ng/mL/cc).[23]
[24] The PCa in our study group had high PSA density of >0.2 ng/mL/cc (EAU high-risk
group).[12] All lesions with size ≥10 mm resulted in higher incidence of PCa among PIRADS 3
in our case series.
Change in the current practice guidelines to manage PIRADS 3 needs more research evidence.
Shifting the approach of target biopsy for PIRADS ≥4 has better interobserver agreement
but could miss substantial portion (24%) of the csPCa in biopsy naïve patients. On
the other hand, shifting the definition of csPCa to ISUP ≥3 decreases the prevalence
of csPCa by half to one-third at more risk of disease progression.[10] Current scientific evidence is not sufficient to replace prostate biopsy with clinical
and serial MRI for management of PIRADS 3 lesions, especially due to risk of disease
progression, and high false positive rate of MRI.[9]
There are a number of risk assessment models to estimate the risk of PCa in PIRADS
3 lesions, and serum PSA kinetics is the strongest one. Serum PSA, percentage of free
PSA, PSA density, age, race, and family history, prior MRI, and prior biopsy are important
clinical risk predictors.[12]
[25] A recent study by Deniffel and colleagues showed that at the threshold of <10%,
normalized ADC (0.81) and PSA density (0.08 ng.nL/cc) outperform other risk models—MRI-ERSCPC
risk calculator, lesion volume, and Radtke risk model—in the diagnosis of PCa.[26] There are a number of blood-, urine-, or tissue-based biomarkers under evaluation
to diagnose the indolent PCa.[25] EAU guidelines mentioned that Prostate Health Index (PHI) test, four kallikrein
(4K) test, and Stockholm3 test have potential role in reducing the number of mpMRI
and/or prostate biopsies.[12] 4K score and family history are considered strong predictors of csPCa on biopsy
even when the mpMRI is negative.[27] The Proclarix test is considered a better predictor than PSAD and ERSPC MRI model,
and could detect 100% of csPCa, and reduce biopsies by 21.3% and detection rate of
insignificant tumor by 15.4%.[28]
EAU guidelines outline combining MRI and PSA density pathway to avoid redundant prostate
biopsy and detect csPCa. The risk-adapted matrix table for prostate biopsy decision
protocol does not recommend biopsy if PSA density is <0.1 ng.mL/cc in PIRADS 3 lesion
where the risk of csPCa is in the range of 5 to 10%.[12] Similarly, “AUA/SUO Early Detection of Prostate Carcinoma Algorithm” recommends
risk-based and adjunctive biomarker–based risk stratification before proceeding with
prostate biopsy. Prostate MRI is recommended before repeat biopsy and is optional
for initial biopsy.[25] A single assessment tool is not sufficient. Multivariate risk modeling combining
clinical, biochemical, and radiological data could be helpful to plan the management
of PIRADS 3 prostate lesion.[10]
Our study is limited due to retrospective design, single-center study with small sample
size. Although the prevalence of PCa in Nepal is one of the lowest among Asian countries,
we have noticed rise in the incidence of the suspected prostate carcinoma due to wider
use of clinical screening and prostate MRI. MRI-based evaluation and ultrasound-guided
prostate biopsy are not routine practice in Nepal until recently, primarily due to
scarce resources and expertise.[29]
[30] The population-based screening showed cancer detection rate of 0.73 based on finger-guided
tru-cut biopsy.[29] In a prospective study done in 2020 in Nepal, the carcinoma detection rate of TRUS
guided prostate biopsy was 0.80.[30] We recommend more rigorous clinicopathological evaluation and assessment of the
risk predictors for the optimal cost-effective management of the PIRADS 3 prostate
lesion in the limited resources settings.
Conclusion
The overall incidence of prostate carcinoma in PIRADS 3 lesion is 41%, out of which
20% is csPCa. Prostate biopsy, when based on combined MRI characterization, and clinical
and biochemical parameters, is likely to outweigh the harm of not doing biopsies.
Bibliographical Record
Bikash Bikram Thapa, Bina Basnet, Gaurav Karki, Narayan Thapa. Clinicopathological
Characteristics of PIRADS 3 Prostate Lesion: A Retrospective Study. Surg J (N Y) 2025;
11: a26828456.
DOI: 10.1055/a-2682-8456
