Bladder neck invasion is an independent predictor of prostate-specific antigen recurrence

Authors

  • Christopher K. Poulos M.D.,

    1. Department of Pathology and Laboratory Medicine, Indiana University School of Medicine, Indianapolis, Indiana
    Search for more papers by this author
  • Michael O. Koch M.D.,

    1. Department of Urology, Indiana University School of Medicine, Indianapolis, Indiana
    Search for more papers by this author
  • John N. Eble M.D.,

    1. Department of Pathology and Laboratory Medicine, Indiana University School of Medicine, Indianapolis, Indiana
    Search for more papers by this author
  • Joanne K. Daggy M.S.,

    1. Division of Biostatistics, Indiana University School of Medicine, Indianapolis, Indiana
    Search for more papers by this author
  • Liang Cheng M.D.

    Corresponding author
    1. Department of Pathology and Laboratory Medicine, Indiana University School of Medicine, Indianapolis, Indiana
    2. Department of Urology, Indiana University School of Medicine, Indianapolis, Indiana
    • Department of Pathology and Laboratory Medicine, Indiana University Medical Center, University Hospital 3465, 550 North University Boulevard, Indianapolis, IN 46202
    Search for more papers by this author
    • Fax: (317) 274-5346


Abstract

BACKGROUND

The 1997 TNM staging system for prostatic carcinoma and the 2002 revision thereof classified prostatic carcinoma with bladder neck involvement classified as pT4 disease. This classification is based on the belief that tumors that invade surrounding structures are more aggressive and warrant higher staging than tumors that do not invade surrounding structures. Recent reports in the literature suggested that microscopic involvement of the bladder neck does not carry independent prognostic significance. Therefore, resection specimens with bladder neck involvement should not be classified as pT4. The current study prospectively examined the prognostic significance of bladder neck involvement by prostatic carcinoma.

METHODS

The authors analyzed the totally embedded and whole-mounted radical prostatectomy specimens from 364 consecutive patients. The mean patient age was 66 years (range, 41–77 years). The bladder neck, which had been coned from the specimen, was cut in a perpendicular fashion. Involvement of the bladder neck was defined as the presence of neoplastic cells within the smooth muscle bundles of the coned bladder neck. The data were prospectively collected. Bladder neck involvement was analyzed in relation to age, preoperative prostate-specific antigen (PSA) level, prostate weight, Gleason score, final pathologic classification, tumor volume, surgical margin status, the presence of high-grade prostate intraepithelial neoplasm, multifocality, seminal vesicle invasion, extraprostatic extension, perineural invasion, and PSA recurrence.

RESULTS

Bladder neck involvement was found in 22 (6%) of 364 patients. Univariate results indicated that bladder neck involvement versus no bladder neck involvement was significantly associated with preoperative PSA (P < 0.001), higher pathologic classification (P < 0.001), larger tumor volume (P< 0.001), extraprostatic extension (P < 0.001), positive surgical margins (P < 0.001), and PSA recurrence (P = 0.003). In a multivariate logistic regression model controlling for pathologic classification, Gleason score, and surgical margin status, bladder neck involvement was an independent predictor of PSA recurrence (P = 0.04). The adjusted odds ratio for bladder neck involvement was 3.3 (95% confidence interval, 1.04–10.03).

CONCLUSIONS

In the current study, bladder neck involvement was an independent predictor of early PSA recurrence. The data demonstrated the importance of continued assessment of bladder neck invasion and supported the placement of tumors with bladder neck involvement in a stage that recognizes the prognostic implications of such involvement. Cancer 2004. © 2004 American Cancer Society.

Accurate and uniform staging of tumors is vital for prediction of tumor behavior, treatment selection, evaluation of response to established and experimental treatment, and exchange of information and data among institutions. In the 1997 and 2002 TNM classification systems,1, 2 prostate carcinoma with bladder neck involvement was classified within a unified T4 category. Tumors that are fixed or invade adjacent structures, such as the rectum, the external sphincter, levator muscles, or pelvic sidewall musculature, are also classified as T4. Because most T4 tumors are not excised due to extensive local invasion, incidental tumor infiltration of the bladder neck is the most common occasion in which resection specimens are designated as pT4. The prognostic significance of bladder neck involvement by prostate carcinoma is uncertain, and some recent reports in the literature3–5 indicate that it may not carry as worse a prognosis as tumors that invade other structures. Using a whole-mount series from 364 patients, the current study prospectively examined the importance of bladder neck involvement in patients with prostate carcinoma treated by radical prostatectomy.

MATERIALS AND METHODS

Radical Prostatectomy Specimens

The current study prospectively analyzed the radical prostatectomy specimens from 364 patients who underwent surgery for prostate carcinoma at the Indiana University Hospital between 1999 and 2003. The mean patient age was 66 years (range, 41–77 years). All had clinically localized prostate carcinoma. None had received hormonal or radiotherapy before surgery. The radical prostatectomy specimens were examined as previously described.6–10 The prostate specimens were weighed, measured, inked, and fixed in 10% neutral buffered formalin. After fixation, the apex and base were coned from the specimen. The cones were then serially sectioned at 3–5-mm increments in a radial fashion, perpendicular to the inked margin. The seminal vesicles were sectioned parallel to their junction with the prostate and entirely submitted for examination. The remaining prostate specimen was serially sectioned perpendicular to the long axis from the apex of the prostate to the base, and whole-mount sections were prepared.

The prostatectomy specimens were graded and staged by a single urologic pathologist (L.C.), and all pathologic data, including the presence or absence of bladder neck invasion, were collected prospectively. The Gleason grading system11 was used and the final score was obtained by summing the primary and secondary Gleason pattern grades based on assessment of the entire specimen. The 1997 TNM system was used for pathologic staging.1, 2 Surgical margins were considered positive when carcinoma cells were in contact with the inked margin.12 Involvement of the bladder neck was defined as the presence of neoplastic cells within the smooth muscle cells of the coned bladder neck, as previously described (Fig. 1).3 The volume of carcinoma in the entire prostate was determined by the grid method and was the sum of the volumes of individual foci of tumor.8, 13–16

Figure 1.

(A,B) Examples of bladder neck involvement by prostatic adenocarcinoma. Prostatic carcinoma cells infiltrate the smooth muscle bundles of the bladder neck, which is characteristic of such invasion.

This research was approved by the Indiana University institutional review board.

Patient Follow-Up

The patients were followed at 3-month intervals during Year 1 and at 6-month intervals during Year 2. Thereafter, follow-up was adjusted according to the clinical situation but at least annually. The patients were followed for 1.5–48 months (mean, 14 months; median, 12 months). Prostate-specific antigen (PSA) recurrence was defined as a PSA level ≥ 0.1 ng/mL.17 The time of recurrence was the point at which the PSA level first reached this level. Of the 364 patients, 45 (12%) were defined as having recurrence by this criterion.

Statistical Analysis

Bladder neck involvement was analyzed in relation to age, preoperative PSA level, prostate weight, Gleason score, final pathologic classification, tumor volume, surgical margin status, the presence of high-grade prostatic intraepithelial neoplasia (PIN), multifocality, seminal vesicle invasion, extraprostatic extension, perineural invasion, and PSA recurrence. The Fisher exact test was used to test for an association between bladder neck involvement and categorical variables. The Wilcoxon rank sum test was used to test for differences in continuous variables grouped by bladder neck invasion status. Logistic regression was also used to determine the effect of bladder neck involvement on PSA recurrence.

RESULTS

We analyzed radical prostatectomy specimens from 364 consecutive patients. Of these patients, 22 (6%) had bladder neck involvement and 45 (12%) had PSA recurrence. In addition, 287 (79%) patients had perineural invasion, 86 (24%) had other surgical margins that were positive, 273 (75%) had T2 tumors, and 91 (25%) had T3 tumors. The patients and radical prostatectomy characteristics were categorized by bladder neck involvement and summarized in Table 1. PSA recurrence occurred in 37 (11%) of the 342 patients without bladder neck involvement, whereas 8 (36%) of the 22 patients with bladder neck involvement had PSA recurrence. Univariate results indicated that bladder neck involvement versus no involvement was significantly associated with preoperative PSA level (P < 0.001), higher pathologic classification (P < 0.001), larger tumor volume (P < 0.001), extraprostatic extension (P < 0.001), positive surgical margins (P < 0.001), and PSA recurrence (P = 0.003). Bladder neck involvement was not associated with prostate weight, Gleason score, presence of high-grade PIN, perineural invasion, seminal vesicle invasion, or multifocality (Table 1).

Table 1. Patient Characteristics by Presence or Absence of Bladder Neck Invasion
CharacteristicsBladder neck invasionP valuea
Present (n = 22) (%)Absent (n = 342) (%)
  • PSA: prostate-specific antigen; PIN: prostatic intraepithelial neoplasia.

  • a

    Median (range) and Wilcoxon rank sum test or number (%) and the Fisher exact test were reported.

Mean age59 (47–70 yrs)61 (41–77 yrs)0.21
Mean preoperative PSA level9.2 (4.9–150 ng/mL)6 (0.3–58 ng/mL)0.0001
Mean prostate weight38 (20–63 g)38 (14–149 g)0.89
Pathologic classification  0.70
 T28 (36)265 (78)< 0.0001
 T314 (64)77 (23) 
Mean tumor volume5 (0.5–38 cm3)1.6 (0.03–13.6 cm3)< 0.0001
Mean largest tumor dimension2.8 (0.7–4.4 cm)1.6 (0–4.1 cm)< 0.0001
Mean Gleason score7 (5–9)7 (4–9)0.87
Lymph node metastasis1 (5)5 (2) 
Extraprostatic extension14 (64)74 (22)< 0.0001
Seminal vesical invasion1 (5)17 (5)> 0.999
Surgical margins15 (68)71 (21)< 0.0001
Multifocality19 (86)291 (85)> 0.999
Perineural invasion21 (96)266 (78)0.06
High-grade PIN22 (100)338 (99)> 0.999
PSA recurrence8 (36)37 (11)0.003

In univariate logistic regression predicting PSA recurrence using only bladder neck involvement, the odds of PSA recurrence in subjects with bladder neck involvement was approximately 5 times higher than in subjects without bladder neck involvement. (odds ratio [OR] = 4.7, 95% confidence interval [CI] = 1.8–11.8, P = 0.003). In a multivariate model, controlling for pathologic classification, Gleason score, and surgical margin status, bladder neck involvement was still a significant independent predictor of PSA recurrence (Table 2). PSA recurrence was approximately 3 times as likely for subjects with bladder neck involvement adjusting for other covariates in the model (adjusted OR = 3.3, P = 0.04) (Table 2).

Table 2. Multivariate Logistic Regression Predicting PSA Recurrence (n = 364)
VariableAdjusted odds ratio (95% CI)P valuea
  • PSA: prostate-specific antigen; 95% CI: 95% confidence interval.

  • a

    The 95% confidence-intervals are the profile-likelihood type, and the P values are from likelihood ratio tests.

Pathologic classification (T2 vs. T3)2.01 (0.91–4.37)0.08
Gleason score2.39 (1.69–3.46)< 0.0001
Bladder neck involvement3.28 (1.04–10.03)0.04
Surgical margins2.41 (1.14–5.05)0.02

DISCUSSION

The decision to place tumors with bladder neck involvement in the unified pT4 category is controversial as some recent reports in the literature3–5 suggest that microscopic involvement of the bladder neck does not carry independent prognostic significance. Therefore, resection specimens with bladder neck involvement should not be classified as pT4. Other studies suggested that patients with bladder neck margin involvement had worse clinical outcome compared with patients without bladder neck margin involvement.18, 19 The difficulty regarding the significance of bladder neck involvement is compounded by some discrepancies as to the precise definition of bladder neck involvement,3, 4 with some reports that seek to discuss bladder neck involvement actually reporting what others would describe as bladder neck margin status.

The current study seeks to analyze the effect of bladder neck involvement on PSA recurrence. We found that bladder neck involvement versus no bladder neck involvement was significantly associated with preoperative PSA level, higher pathologic classification, larger tumor volume, and positive surgical margins. In multivariate logistic regression controlling for pathologic classification, Gleason score, and surgical margin status, bladder neck involvement was an independent predictor of PSA recurrence (P = 0.04, adjusted OR = 3.3; 95% CI = 1.04–10.03). PSA recurrence was more than 3 times as likely for patients with bladder neck involvement adjusting for other covariates in the model (adjusted OR = 3.3, P = 0.04) (Table 2). Our findings support placing tumors with bladder neck involvement in a stage that recognizes the prognostic implications of such involvement.

Several studies have specifically examined the importance of bladder neck involvement with respect to disease progression and PSA recurrence (Table 3).3–5 The incidence of bladder neck involvement by prostate carcinoma in the current study (6%) was similar to the findings of other studies (range, 2.8–8.7%). The results of these studies, however, are in disagreement with our study in regards to the significance of bladder neck involvement for patient outcome. Yossepowitch et al.3, 4 have evaluated the importance of bladder neck involvement in two separate studies. The authors found that microscopic bladder neck involvement was not an independent predictor of disease progression.4 Similarly, Dash et al.5 examined the significance of bladder neck involvement in predicting PSA recurrence and, despite finding the bladder neck to be the site-specific margin with the highest PSA recurrence, their multivariate model found that bladder neck involvement was not a significant independent predictor of PSA recurrence (P = 0.5). Even though they did not find bladder neck involvement to be a significant predictor of progression, Dash et al. found that the group with bladder neck involvement had a larger proportion of high-grade tumors and extraprostatic extension. The authors suggested that a tumor that has escaped the prostate into the muscle of the bladder neck would carry a worse prognosis.

Table 3. Literature Review on the Significance of BN Invasion by Prostatic Adenocarcinoma
StudyNo. of PatientsMethod of processing BNWhole-mount processingIncidence of BN invasion (%)Results
  1. BN: bladder neck; PSA: prostate-specific antigen.

Yossepowitch et al., 20003286Perpendicularly sectioned BN coneNo25 (8.7)BN invasion does not connote a higher progression risk
Dash et al., 200251123Shave biopsy of bladder neckNo60 (5)BN invasion was not a significant independent predictor of PSA recurrence (P = 0.5)
Yossepowitch et al., 200242571 (after exclusions)Margin obtained by sampling soft tissue at the junction of the rough prostatic capsule and smooth bladder neck or most proximal area of the specimen corresponding to the bladder neckNo72 (2.8)BN invasion has no independent ability to predict progression
Current study364Perpendicularly sectioned BN coneYes22 (6)BN invasion is an independent predictor of early PSA recurrence (P = 0.04)

The discrepancies among different studies may be attributed to study design, different patient populations, specimen handling, sample size, variation in statistical methods, and the definition of bladder neck involvement. In the current study, we utilized the whole-mount processing technique and pathologic data were prospectively collected. Our study and the first study by Yossepowitch et al.3 used a coned and perpendicularly sectioned bladder neck margin to allow distinction of bladder neck involvement and bladder neck margin positivity, whereas the study by Dash et al.5 and the second study by Yossepowitch et al.4 each relied on differing methods (Table 3). In particular, Dash et al. utilized a shave margin, in which tumor cells at the inked margin were defined as representing bladder neck involvement, and this may have rendered the distinction of bladder neck involvement from bladder neck margin positivity difficult.

Öbek et al.18 compared the apex/urethra, posterior, anterior, lateral, posteriolateral, and bladder neck margins and found, using multivariate analysis, that bladder neck margin positivity was a significant (P = 0.003) independent predictor of disease progression. The authors suggested that tumors with extracapsular extension and positive bladder neck margins may represent a more metastatic phenotype. Similarly, in a study of 2712 patients, Blute et al.19 evaluated multiple surgical margins and their impact on PSA failure. They found that the bladder neck was, after adjusting for Gleason grade, PSA, and DNA ploidy, the only positive margin site that was a significant predictor of PSA recurrence. In that study,19 the 5-year recurrence-free survival rates were 85% and 56%, respectively, for tumors without and with a positive bladder neck margin. These studies do not distinguish bladder neck margin positivity from bladder neck involvement without a positive margin.

This is the first study that has demonstrated the independent prognostic significance of bladder neck invasion. The main limitations of the current study are the relatively small sample size and short follow-up. Therefore, our findings should be considered preliminary. The number of outcome events are small, limiting statistical power. Nonetheless, we found that bladder neck involvement is a significant independent predictor of PSA recurrence in patients with prostate carcinoma treated by radical prostatectomy. Identification of patients at early treatment failure is critically important for assessing prognosis and treatment plans, despite short follow-up. The data gathered support continued assessment of the bladder neck invasion and classification of tumors with bladder neck involvement in a category that highlights their independent prognostic significance.

In conclusion, bladder neck involvement was found in 6% of radical prostatectomy specimens. Bladder neck involvement versus no involvement was significantly associated with preoperative PSA, higher pathologic classification, larger tumor volume, and positive surgical margins. In multivariate analysis, bladder neck involvement is an independent predictor of early PSA recurrence. Our data demonstrate the importance of continued assessment of bladder neck invasion and support the placement of tumors with bladder neck involvement in a stage that recognizes the prognostic implications of such involvement.

Ancillary