RESULTS
All included cases had histopathologically proven PCa. The mean value
for age was calculated as 65.25 ± 6.45 (minimum = 48, maximum = 79).
According to histopathological results, there were 17 (15.5%) ISUP
grade 1, 40 (36.3%) ISUP grade 2, 18 (16.4%) ISUP grade 3, 19 (17.3%)
ISUP grade 4, and 16 (14.5%) ISUP grade 5 tumors. The Gleason
score-based numbers were as follows: 17 (15.5%) Gleason 6, 58 (52.7%)
Gleason 7, 19 (17.3%) Gleason 8, 14 (12.7%) Gleason 9, and 2 (1.8%)
Gleason 10 tumors. The stage was T2 in 58 cases (52.7%) and T3 in 52
cases (47.3%). The histopathological evaluation revealed 52 cases
(47.3%) with EPE and 58 cases (52,7%) without EPE (Table 2).
The prostate mpMRI findings for the cases were categorized as PI-RADS 3,
4, and 5. For Observer 1, 20% of the cases had PI-RADS 3 lesions,
35.5% had PI-RADS 4, and 44.5% had PI-RADS 5 lesions. Similarly, the
values were 21%, 35.5%, and 45.5% respectively for Observer 2.
Furthermore, interobserver agreement was good for PI-RADS, LL, and the
TCL, whereas it was moderate for the presence of EPE(Table 3).
For Observer 1, sensitivity was 40.4%, specificity was 96.6%, positive
predictive value was 91.3%, and negative predictive value was 64.3%
for the detection of EPE. For Observer 2, sensitivity, specificity,
positive predictive value, and negative predictive value were 40.4%,
84.5%, 70%, and 61.3%, respectively. Prostate mpMRI demonstrated low
sensitivity but high specificity in the detection of EPE, which
hadsignificant value (p <0.0001 for Observer 1 and p = 0.003
for Observer 2).
EPE was present in 9% of PI-RADS 3, in 36% of PI-RADS 4, and in 73%
of PI-RADS 5 cases for Observer 1, whereas these rates were 10%, 36%,
and 72%, respectively, for Observer 2. The incidence of EPE increased
as the PI-RADS score increased for both observers (p <0.0001
for both observers).
For both observers, there was a significant positive correlation between
LLs measured on MRI examination.Rho correlation coefficient values were
0.72 for Observer 1 and 0.57 for Observer 2 (p <0.0001).
The TCL demonstrated a significant difference in cases with and without
EPE (Table 4). The TCL is a significant parameter for EPE, with
relatively high sensitivity and low specificity for both observers.
Furthermore, the LL had significant results, with moderate sensitivity
and specificity for measurements on the radiological assessment of both
observers and measurements made from the pathology specimen (Table 5).
In the histopathological examination, a cutoff value of 22 mm was
observed for the LL concerning EPE, with a sensitivity of 63% and a
specificity of 75%. Moreover, the mean value determined forObserver 1
for TCL was 18 mm,while it was 14.5 mm for Observer 2. The cutoff value,
which had the highest sensitivity according to the Youden index, was 9.5
mm. The mean values of LL were 22 and 19 mm, respectively. For both
observers, the cut-off value for high sensitivity was determined as 14.5
mm.