The Clinical Impact of pT3a Lesions in Patients With pT3b Prostate Cancer Undergoing Radical Prostatectomy: A Proposal for a New pT3b Subclassification

Upon approval by the institutional review board, including the request to waive the documentation of patient consent, we assessed consecutive patients who had undergone robot-assisted radical prostatectomy for prostate cancer at our institution between 2009 and 2018. Within our surgical pathology database, we identified a total of 248 men who met the inclusion criteria for pT3b disease. Those who had undergone neoadjuvant therapy before prostatectomy were excluded from analysis.

Gleason score/Grade Group (GG) (primarily in “dominant” nodule[s] in each case) was reevaluated by a senior author on the basis of recommendations by the International Society of Urological Pathology^10,11  as well as the Genitourinary Pathology Society.¹²  In cases with GG2 or GG3 cancer, the presence of a minor tertiary pattern 5 in less than 5% of the tumor was ignored for analysis. We also retrieved clinicopathologic findings, such as age at surgery, preoperative prostate-specific antigen (PSA) value, the statuses of EPE, mBNI, lymphovascular invasion, pN, and surgical margin, and estimated cancer volume, as well as follow-up data. We then histologically reevaluated the presence or absence of EPE/mBNI as well as the extent of EPE, based on the Epstein criteria¹³  (ie, focal [only a few cancer glands] versus nonfocal/established [more than a few glands]). Biochemical recurrence after prostatectomy in patients with no adjuvant therapy (n = 139) was defined as a single PSA level of 0.2 ng/mL or greater, while PSA failure in those undergoing adjuvant treatments, such as hormonal therapy (n = 31), radiotherapy (n = 47), or their combination (n = 31), immediately after prostatectomy (ie, before disease progression), was defined as an increase in PSA value of 2 ng/mL or greater, or at least 50% over nadir, or the introduction of salvage therapy.^14,15  The PSA recurrence in those both with and without adjuvant therapy was considered as disease progression.

Data were analyzed, using the Student t test for continuous variables and the χ² test or Fisher exact test for noncontinuous variables. The survival rate was calculated by the Kaplan-Meier method, and comparison was made by the log-rank test. In addition, the Cox proportional hazards model was used to determine statistical significance of prognostic factors in a multivariate setting. All statistical analyses were performed, using GraphPad Prism version 5 (GraphPad Software) and EZR software,¹⁶  a graphical user interface for R version 4.0.2 (The R Foundation for Statistical Computing). P values less than .05 were considered to be statistically significant.

In a retrospective, blinded manner, we examined a total of 248 radical prostatectomy cases with pT3b disease. Supplemental Table 1 summarizes clinicopathologic features of these patients (see the supplemental digital content containing 3 tables and 6 figures at https://meridian.allenpress.com/aplm in the May 2022 table of contents). Overall, EPE or mBNI was found in 219 (88.3%) or 48 (19.4%) of cases, respectively. Only 1 of 29 EPE⁻ cases (3.4%) showed mBNI, while 47 of 219 EPE⁺ cases (21.5%) had mBNI. Similarly, 47 of 48 mBNI⁺ cases (97.9%) showed EPE, while 172 of 200 mBNI⁻ cases (86.0%) had EPE. Thus, EPE and mBNI were significantly correlated (P = .02).

We next compared cases with versus without EPE (Table 1). The presence of EPE was significantly associated with higher PSA level (P = .02), higher incidence of lymphovascular invasion (P = .02) or positive surgical margin (P = .02), and larger tumor volume (P < .001). There were no significant differences in GG, lymph node metastasis, and the need for adjuvant therapy immediately after prostatectomy (before disease recurrence).

We similarly compared cases with versus without mBNI (Table 2). The presence of mBNI was significantly associated with higher PSA level (P = .03), higher incidence of GG4-5 (P = .006) or GG5 (P = .009), lymphovascular invasion (P = .003), lymph node involvement (P = .02), positive surgical margin (P < .001), or adjuvant therapy (P = .01), and larger tumor volume (P < .001).

We further compared cases between either EPE⁻/mBNI⁺ or EPE⁺/mBNI⁻ and EPE⁻/mBNI⁻ or EPE⁺/mBNI⁺ (Table 3). The presence of both EPE and mBNI was significantly associated with higher PSA level (P = .03), higher incidence of GG4-5 (P = .02) or GG5 (P = .03), lymphovascular invasion (P = .002), positive surgical margin (P < .001) or adjuvant therapy (P = .01), and larger tumor volume (P < .001), compared with EPE or mBNI only. In addition, lymphovascular invasion (P = .02) and larger tumor (P = .002) were significantly more often seen in EPE-only or mBNI-only cases than in EPE⁻/mBNI⁻ cases.

Kaplan-Meier analysis coupled with log-rank test was then performed to assess the impact of EPE and mBNI on the prognosis following radical prostatectomy, with mean and median follow-up of 60.2 and 50.5 months, respectively. We first compared progression-free survival and cancer-specific survival in the entire cohort of patients, according to the status of EPE or mBNI. Extraprostatic extension was significantly associated with a higher risk of disease progression (P < .001; Figure 1, A), but not of cancer-specific mortality (P = .15; Figure 1, B), although none of the EPE⁻ patients died of prostate cancer. In addition, the risks of both progression (P < .001; Figure 1, C) and mortality (P = .01; Figure 1, D) were significantly higher in patients with mBNI than in mBNI⁻ patients. The prognosis was also compared, according to the status of both EPE and mBNI (Figure 2, A and B). The absence of both EPE and mBNI was associated with a significantly lower risk of progression (P = .001; Supplemental Figure 1, A), but not mortality (P = .25; Supplemental Figure 1, B), than was the presence of either EPE or mBNI. Patients with EPE or mBNI only further showed significantly lower risks of progression (P < .001; Supplemental Figure 1, C) and mortality (P = .04; Supplemental Figure 1, D) than did EPE⁺/mBNI⁺ patients.

The prognosis was separately assessed in those without (n = 139) and with (n = 109) adjuvant therapy before disease recurrence. In the nonadjuvant cohort, EPE (P = .001; Supplemental Figure 2, A) was associated with a significantly higher risk of disease progression, but not cancer-specific mortality (P = .26; Supplemental Figure 2, B). The risks of progression (P = .01, Supplemental Figure 2, C) and mortality (P = .09, Supplemental Figure 2, D) were significantly and insignificantly, respectively, higher in those with mBNI. There were significant differences in progression-free survival between EPE⁻/mBNI⁺ or EPE⁺/mBNI⁻ cases and EPE⁻/mBNI⁻ (P = .002) or EPE⁺/mBNI⁺ (P = .02) cases (Supplemental Figure 2, E). However, these significant differences were not seen in cancer-specific mortality (Supplemental Figure 2, F). In the adjuvant cohort, a significant association between EPE and progression-free survival (P = .049; Supplemental Figure 3, A), but not cancer-specific survival (P = .43; Supplemental Figure 3, B), was seen. Similarly, the differences in progression-free survival (P < .001; Supplemental Figure 3, C) and cancer-specific survival (P = .06; Supplemental Figure 3, D) between mBNI⁻ and mBNI⁺ cases were statistically significant and insignificant, respectively. There were also insignificant and significant differences in progression-free survival between cases showing either EPE⁻/mBNI⁺ or EPE⁺/mBNI⁻ and EPE⁻/mBNI⁻ (P = .09) or EPE⁺/mBNI⁺ (P = .001) (Supplemental Figure 3, E). There was an insignificant difference in cancer-specific survival between cases showing EPE or mBNI only versus EPE⁺/mBNI⁺ (P = .09), but not EPE⁻/mBNI⁻ (P = .61) (Supplemental Figure 3, F).

The prognosis was also separately assessed in those without (n = 153) and with (n = 93) lymph node metastasis. In patients with pN0 disease, EPE was associated with the risk of disease progression (P < .001; Supplemental Figure 4, A), but not that of cancer-specific mortality (P = .28; Supplemental Figure 4, B), while the risks of progression (P < .001; Supplemental Figure 4, C) and mortality (P < .001; Supplemental Figure 4, D) were significantly higher in patients with mBNI than in those without mBNI. The differences in progression-free survival between EPE⁺ or mBNI⁺ only and EPE⁻/mBNI⁻ (P = .001) or EPE⁺/mBNI⁺ (P = .005) (Supplemental Figure 4, E), as well as in cancer-specific survival between either EPE⁻/mBNI⁺ or EPE⁺/mBNI⁻ and EPE⁺/mBNI⁺ (P = .001), were statistically significant (Supplemental Figure 4, F). In patients with pN1 disease, there were no significant associations between EPE and progression (P = .32; Supplemental Figure 5, A) or mortality (P = .30; Supplemental Figure 5, B), while mBNI was associated with a significantly lower rate of progression-free survival (P = .01; Supplemental Figure 5, C), but not of mortality (P = .22; Supplemental Figure 5, D). There was a significant difference in progression-free survival (P = .008; Supplemental Figure 5, E), but not in cancer-specific survival (P = .27; Supplemental Figure 5, F), between patients with EPE⁻/mBNI⁺ or EPE⁺/mBNI⁻ and EPE⁺/mBNI⁺.

In EPE⁺/mBNI⁻ cases, focal versus nonfocal EPE was compared (Supplemental Table 2). Nonfocal EPE was significantly associated with higher tumor grade (P < .05), positive surgical margin (P = .04), and larger tumor volume (P = .005). There were no significant differences in PSA level, lymphovascular invasion, lymph node metastasis, or adjuvant therapy. In these cases, the prognostic value of focal versus nonfocal EPE was also assessed. However, there were no significant differences in progression-free survival (P = .27; Supplemental Figure 6, A) or cancer-specific survival (P = .70; Supplemental Figure 6, B) between focal and nonfocal EPE cases.

To further determine if EPE and mBNI were independent predictors of disease progression following radical prostatectomy, multivariate analysis was performed by using the Cox model. In the entire cohort, EPE (hazard ratio [HR] = 6.53; 95% CI, 1.60–26.7; P = .009) and mBNI (HR = 2.33; 95% CI, 1.34–4.07; P = .003) showed significance for the progression. When focal versus nonfocal EPE was also considered, only nonfocal EPE, as well as mBNI, showed significance (Table 4). Additionally, compared with EPE⁻/mBNI⁻, the presence of either EPE or mBNI (HR = 11.7; 95% CI, 1.61–84.6; P = .01) or both EPE and mBNI (HR = 25.9; 95% CI, 3.41–197; P = .002) was associated with the risk of progression (Supplemental Table 3). Moreover, in patients not undergoing adjuvant therapy immediately after prostatectomy, there was an association between the presence of either EPE or mBNI (HR = 11.2; 95% CI, 1.51–83.0; P = .02) or both EPE and mBNI (HR = 30.9; 95% CI, 3.69–258; P = .002) and progression-free survival.

Seminal vesicle invasion by prostate cancer has been considered to be a key prognostic factor. However, pT3b disease is not uniformly associated with a poor prognosis after radical prostatectomy. Meanwhile, the clinical significance of concurrent EPE and mBNI in pT3b prostate cancer remains uncertain. We therefore compared radical prostatectomy findings and long-term oncologic outcomes in a total of 248 men with pT3b prostate cancer, with versus without pT3a lesions. The prognosis was also assessed separately in subgroups of patients: those undergoing or not undergoing adjuvant therapy immediately after radical prostatectomy as well as those with pN0 or pN1 disease.

A previous study in 1132 pT3bN0 cases undergoing radical prostatectomy, including 38.8% of those without EPE, demonstrated strong associations of EPE with higher preoperative PSA level, higher Gleason score, and higher incidence of positive surgical margin.¹⁷  A recent study in 88 patients with pT3b disease also showed a relative risk of 2.75 for lymph node metastasis in patients with EPE (n = 69), compared to those without EPE (n = 19).¹⁸  Consistent with most of these findings, we found in our cohort of patients with pT3bN0 and pT3bN1 disease (plus 2 cases of pT3bNX) that EPE was strongly associated with higher levels of PSA and higher rates of lymphovascular invasion and positive surgical margin, as well as larger estimated tumor volume, although only a subset (ie, 11.7%) of our cases showed no EPE. We additionally found that mBNI was strongly associated with all the clinicopathologic features examined (except for age), including PSA level, GG, lymphovascular invasion, lymph node metastasis, surgical margin status, tumor volume, and delivery of adjuvant therapy. The presence of EPE and mBNI was also significantly correlated. When compared between cases showing either EPE or mBNI only and EPE⁺/mBNI⁺ cases, significant differences in these clinicopathologic features were still seen. Moreover, the tumors showing either EPE or mBNI were found to more often exhibit lymphovascular invasion and to be larger than in EPE⁻/mBNI⁻ cases. Extraprostatic extension and mBNI in pT3b disease, especially when both are present, are thus associated with adverse histopathologic features in prostatectomy specimens.

Several studies have assessed the impact of EPE on the outcomes of pT3b disease following radical prostatectomy.^17–21  Specifically, three^17,18,20  and two^19,21  indicated and failed to indicate, respectively, the prognostic value of EPE, while none of these studies included lymph node–positive cases (a separate cohort of 69 patients with pN0 disease was analyzed in one of the studies described above¹⁸ ). In one of the former studies,¹⁷  EPE was shown to be an independent predictor of systemic progression (HR = 1.56, P = .006) or cancer-specific mortality (HR = 1.71, P = .01). Additionally, in a study involving 105 patients with pT3b disease who underwent radical prostatectomy, the presence of mBNI, which was not considered a pT3a lesion at the time, was also implicated in worse outcomes.²²  In the present study, we assessed the role of EPE and mBNI in the prognosis of patients with pT3b disease who underwent radical prostatectomy as well as that of the clinically relevant subgroups separately. Extraprostatic extension was found to be associated with a significantly higher risk of disease progression not only in the entire cohort of patients where it was an independent prognostic factor, but also in the subgroups, such as non–adjuvant therapy patients, adjuvant therapy patients, and pN0 patients. Furthermore, we found that mBNI was a strong predictor of disease progression in the entire cohort, as an independent factor, and in any of the subgroups examined including the patients with pN1 disease, as well as cancer-specific mortality in the entire cohort or patients with pN0 disease. Importantly, patients with both EPE and mBNI showed a significantly higher risk of disease progression, as an independent factor, as well as cancer-specific mortality, than those with either EPE or mBNI. Additionally, compared with EPE⁻/mBNI⁻, the presence of either EPE or mBNI was found to be an independent predictor of progression. These significant associations were also seen in all 4 (EPE⁺/mBNI⁺) or some (either EPE⁻/mBNI⁺ or EPE⁺/mBNI⁻) subgroups of patients. From these findings, we believe it is logical to propose a novel subclassification of the current pT3b prostate cancer to better predict the prognosis in this context: pT3b1 (SVI alone without EPE or mBNI); pT3b2 (SVI with either EPE or mBNI); and pT3b3 (SVI with both EPE and mBNI).

The incidence of EPE detected in pT3b prostatectomy specimens has been reported to range from 61% to 78%.^17–21,23  Thus, the detection rate in our study (ie, 88%) was slightly higher. In a previous study,²⁴  at least 3 pathways of SVI by prostate cancer were described. These included types I (direct spread through the ejaculatory duct), II (direct spread through EPE), and III (isolated tumor through, for instance, lymphovascular invasion). Accordingly, EPE is not necessarily seen in pT3b disease (eg, type I or III SVI), while many of the EPEs in our cohort were not detected in the regions adjacent to the seminal vesicle. Meanwhile, interobserver variation among pathologists for the diagnosis of EPE in radical prostatectomy specimens was noted. The concordance rate for EPE (57.5%, κ = 0.33) has been reported to indeed be considerably lower than that for SVI (94%, κ = 0.83).²⁵  It may be particularly controversial to make a histopathologic diagnosis of EPE when cancer is not within or adjacent to adipose tissue (but is possibly beyond the condensed smooth muscle of the edge of the prostate).^26,27  Similarly, although the incidence of mBNI in pT3b disease does not appear to have been documented, it is sometimes difficult to distinguish bladder neck muscle fibers from other fibromuscular tissues (in the absence of benign prostatic glandular tissues) in relatively small specimens submitted for histologic assessment.

In pT3a prostate cancer, the extent of EPE (eg, focal versus nonfocal/established)^9,13,27  is associated with survival. We similarly compared radical prostatectomy findings in pT3b cases showing EPE but no mBNI. We found that nonfocal EPE was significantly associated with higher GG, higher incidence of positive surgical margin, and larger tumor volume. However, no prognostic impact of focal versus nonfocal EPE was observed in this subgroup of patients.

There are several limitations in our investigation. First, the present study is subject to potential selection bias due to the retrospective design, although we have analyzed consecutive patients who met the inclusion criteria. Second, we compared only radical prostatectomy cases, and the clinical impact of EPE or mBNI in patients undergoing other treatment options, such as hormonal therapy, was not evaluated. Third, the clinical significance of PSA progression in those who had versus those who had not received adjuvant therapy immediately after prostatectomy might be different, although we additionally performed outcome analysis in each subgroup. Fourth, the region of EPE (eg, positional relationship with SVI) was not examined in this study. Finally, as aforementioned, only a small portion of the present cohort showed EPE⁻/mBNI⁻, while only 1 mBNI⁺ case showed no EPE. Therefore, it may not be possible to adequately compare EPE⁻/mBNI⁺ versus EPE⁻/mBNI⁻ or EPE⁺/mBNI⁺ cases.

In conclusion, the presence of EPE and/or mBNI in pT3b prostate cancer was found to be associated with worse histopathologic features in radical prostatectomy specimens and resultant poorer survival outcomes. These data underscore the necessity of clarifying whether there are pT3a lesions in prostatectomy specimens showing SVI. We also believe our findings provide a logical rationale for a new subclassification that more accurately stratifies the prognosis of the current pT3b prostate cancer. Future prospective studies in larger patient cohorts with prostate cancer showing SVI with or without concurrent EPE/mBNI are warranted to validate our results.