ADULT UROLOGY
MUCINOUS ADENOCARCINOMA OF THE PROSTATE DOES NOT CONFER POOR PROGNOSIS BRIAN R. LANE, CRISTINA MAGI-GALLUZZI, ALWYN M. REUTHER, HOWARD S. LEVIN, MING ZHOU, AND ERIC A. KLEIN
ABSTRACT Objectives. To report a series of patients with mucinous (colloid) adenocarcinoma (MC) at prostatectomy who were treated at a single institution from 1987 to 2005. MC is a rare form of prostate cancer reported in some cases to have a more aggressive clinical course than conventional adenocarcinoma (AC). Methods. Radical prostatectomy specimens with mucinous features were identified from a database of 3613 consecutive patients. Each case was reviewed again by a single pathologist who confirmed the diagnosis of MC in 14 patients. MC was defined by the presence of pools of extracellular mucin in more than 25% of the tumor. Eighteen additional cases were identified in which the mucinous component occupied only a small portion of the tumor and were referred to as AC with focal mucin (AFM). The biochemical and overall survival of 26 patients with MC or AFM who had completed ⱖ6 months of follow-up was analyzed using Kaplan-Meier estimates. Results. No patients with MC or AFM died of disease, and 11 (91.7%) of 12 patients with MC and 9 (64.3%) of 14 patients with AFM were clinically and biochemically free of disease. No significant difference was found in biochemical recurrence or overall survival between those with MC or AFM and a matched group of patients with AC. Conclusions. We report what we believe to be the largest published series of cases of MC (n ⫽ 14) with a median overall follow-up of 6.4 years. MC appears to behave clinically in a similar fashion to AC, with no statistically significant difference in biochemical failure or survival. UROLOGY 68: 825–830, 2006. © 2006 Elsevier Inc.
M
ucinous (colloid) adenocarcinoma (MC) of the prostate is a rare variant that has been traditionally thought to carry a poorer prognosis than conventional adenocarcinoma (AC).1 The diagnosis of MC is made at prostatectomy and is conferred when at least 25% of the resected tumor contains lakes of extracellular mucin.2,3 Case reports have indicated that patients with MC have variable outcomes in terms of hormone responsiveness and recurrence.4 –15 Few data exist to document the clinical course of a series of patients with this form of prostate cancer. We report a single-center clinical experience with patients diagnosed with MC, describing clinical and From the Glickman Urological Institute and Department of Pathology, Cleveland Clinic Foundation, Cleveland, Ohio Reprint requests: Eric A. Klein, M.D., Glickman Urological Institute, Cleveland Clinic Foundation, 9500 Euclid Avenue, A100, Cleveland, OH 44195. E-mail:
[email protected] Submitted: December 5, 2005, accepted (with revisions): April 25, 2006 © 2006 ELSEVIER INC. ALL RIGHTS RESERVED
histologic features and biochemical relapse-free and overall survival. MATERIAL AND METHODS PATIENT SELECTION From January 1987 to June 2005, 3613 consecutive patients with prostate cancer who underwent radical prostatectomy were included in an institutional review board-approved computerized patient registry. Of these patients, 2572 did not receive adjuvant therapy and had complete data, and thus were used for comparison in this study. The review of this database identified 32 patients with mucinous features noted on the radical prostatectomy pathologic report. The recorded clinical information included initial prostate-specific antigen (iPSA) level, biopsy Gleason score, preoperative tumor stage, neoadjuvant treatment, surgical or pathologic Gleason score (pGS), extraprostatic extension, margin status, seminal vesicle involvement, nodal status, all postoperative PSA values, and follow-up information. Patients with AC who had received adjuvant hormonal therapy or adjuvant radiotherapy were excluded from this study, and none of the patients with MC or AC with focal mucin (AFM) received this therapy. Patients with MC (n ⫽ 14) or AFM (n ⫽ 18) were matched according to treatment year within a 3-year interval, neoadju0090-4295/06/$32.00 doi:10.1016/j.urology.2006.04.028 825
TABLE I. Characteristic
Clinical characteristics MC (n ⴝ 14)
AFM (n ⴝ 18)
AC (n ⴝ 2572)
Age (yrs) Median 59 57 62 Range 45–70 46–66 39–79 Initial PSA (ng/mL) Median 6.9 7.2 6.3 Range 0.1–24.1 2.9–21.6 0.1–1823 Clinical T stage T1-T2a 13 (92.9) 16 (88.9) 2254 (87.6) T2b-T3 1 (7.1) 2 (11.1) 318 (12.4) Biopsy Gleason score ⱕ6 8 (57.1) 13 (72.2) 1768 (68.7) ⱖ7 6 (42.9) 5 (27.8) 804 (31.3) Neoadjuvant therapy No 8 (57.1) 16 (88.9) 2221 (86.4) Yes 6 (42.9) 2 (11.1) 351 (13.6) Surgical Gleason score ⱕ6 1 (12.5) 6 (37.5) 838 (37.7) ⱖ7 7 (87.5) 10 (62.5) 1383 (62.3) Organ-confined disease Yes 8 (57.1) 12 (66.7) 1318 (51.2) No 6 (42.9) 6 (33.3) 1254 (48.8) KEY: MC ⫽ mucinous adenocarcinoma; AFM ⫽ adenocarcinoma with focal mucinous areas; PSA ⫽ prostate-specific antigen. Data presented as numbers, with percentages in parentheses, unless otherwise noted.
FIGURE 1. Histopathologic section of (A) MC and (B) prostate adenocarcinoma with focal mucin (AFM) at 10⫻ magnification.
vant hormone status, pGS, and organ-confined disease status with patients with AC at a 1:2 ratio for statistical analyses of biochemical recurrence-free survival (bRFS) and overall survival. Chi-square tests, Fisher’s exact tests, and unpaired t tests were used to compare groups. Biochemical failure (BF) and overall survival were determined using survival regression analysis. BF was defined as two consecutive serum PSA levels of 0.2 ng/mL or a single PSA level greater than 0.3 ng/mL.
PATHOLOGIC ANALYSIS Histologic sections of the radical prostatectomy specimens harboring prostate carcinoma identified as “mucinous” or with AFM on initial pathologic report were reviewed again by a single pathologist (C.M.-G.). Using criteria developed for mucinous carcinomas in other organs, the diagnosis of MC of the prostate was made only when greater than 25% of the tumor contained lakes of extracellular mucin (Fig. 1A).2,3 Specimens with mucinous areas occupying only a small portion of the tumor (Fig. 1B) were termed AFM.3 As with other unusual subtypes of prostate cancer, the pGS was assigned only to the conventional AC component.3
RESULTS Of the 32 patients included in this study, 14 were classified as having MC and 18 as having AFM, 826
representing 0.39% and 0.50% of 3613 prostatectomy specimens assessed. The clinical information regarding the patients with a pathologic diagnosis of MC, AFM, or AC is summarized in Table I. In patients with MC, the pGS was 6 (n ⫽ 1), 7 (n ⫽ 4), 8 (n ⫽ 3), and not assessable because of neoadjuvant hormonal therapy (n ⫽ 6). Non-organ-confined disease was present in 12 patients (37.5%) with MC or AFM, including 12 with extraprostatic extension, 6 with positive surgical margins, and 4 with seminal vesicle involvement. Pelvic lymph node dissection was performed in 50% of patients with MC or AFM, none of whom had nodal disease. The comparison of the clinical parameters at presentation of the 14 patients with MC, 18 with AFM, 32 with MC or AFM, and entire cohort with AC (n ⫽ 2572) revealed no statistically significant differences among these groups in terms of iPSA level, clinical T stage, biopsy Gleason score, pGS, or nonorgan-confined disease. A greater percentage of patients with MC had received neoadjuvant hormonal therapy (42.9% with MC, 11.1% with AFM, and 13.6% with AC), although no correlation between neoadjuvant therapy and bRFS was observed. Two (25%) of 8 patients who received neoadjuvant hormonal therapy had BF compared with 4 (22.2%) of 18 who did not (P ⫽ NS). In contrast, 2 (33.3%) of 6 who developed biochemical recurrence had received hormonal therapy compared UROLOGY 68 (4), 2006
with 6 (30%) of 20 who were biochemically free of disease. The median follow-up was 6.4 years (range 0.2 to 11.2) for the 14 patients with MC and 1.3 years (range 0.3 to 12.6) for the 18 patients with AFM. Of the 32 patients with MC or AFM, 26 had at least two postoperative PSA values and biochemical follow-up of at least 6 months. These 26 patients constituted the population used for analysis of biochemical and overall survival, with a median follow-up of 3.6 years (range 0.6 to 12.6). The remaining 6 patients were alive at last follow-up and had an undetectable initial postoperative PSA measurement. Of the 26 patients, 6 (23.1%) had BF, with a median time to failure of 4.2 years (range 0.1 to 9.8). Of these 6 patients, 1 had MC (BF at 6.2 years) and 5 had AFM. Only 1 patient died during the study period, 10.2 years after prostatectomy for MC without evidence of recurrence. Of the 6 patients with non-organ-confined MC, 6 had extraprostatic extension, 4 had positive surgical margins, and 1 also had seminal vesicle involvement; none of these patients had BF at a median of 8.3 years. For overall survival, the Kaplan-Meier estimates showed no statistically significant differences among those with MC, AFM, or AC. The 5-year overall survival rate was 100%, 100%, and 96.6% for those with MC, AFM, and AC, respectively. The 5-year bRFS rate was 100%, 69.2%, and 77.8% for those with MC, AFM, and AC, respectively. These trends were consistent throughout the follow-up period of 12.6 years, but small patient numbers made the comparison at later points less reliable. The overall survival and bRFS rates at 5 and 11 years after prostatectomy are summarized in Fig. 2. To control for the small number of patients diagnosed with MC or AFM and the comparatively large population of patients with AC and the differences observed between these groups with regard to age and neoadjuvant hormone status, a group of patients with AC was selected for matching with those with MC and AFM from the larger pool at a rate of 2:1. The controls with AC were matched according to treatment year within a 3-year interval, neoadjuvant hormone status, organ-confined disease status, and pGS. The clinical characteristics of the patients used in this analysis are listed in Table II. The median age was 60.5 years (range 50 to 70), 57.5 years (range 46 to 66), and 59.0 years (range 45 to 71) in the MC, AFM, and AC groups, respectively. To evaluate overall survival in those with MC or AFM and matched controls, a Kaplan-Meier curve was constructed (Fig. 2A). The P value for overall survival was not calculable because no patients in the AFM group died. The 5-year overall survival rate was 100%, 100%, and 93.8%, for those with MC, AFM, and AC, respectively, similar to the UROLOGY 68 (4), 2006
rates in the analysis that included all those with AC (Fig. 2A). On univariate analysis, no variables were predictive of survival. For bRFS, the Kaplan-Meier curves showed no statistically significant differences among the MC, AFM, and AC groups (P ⫽ 0.17, Fig. 2B). The 5-year bRFS rate was 100%, 69.2%, and 68.0% for those with MC, AFM, and AC, respectively. On univariate analysis, only iPSA as a continuous variable (P ⫽ 0.0263) and biopsy Gleason score (P ⫽ 0.0435) were significant; age, T stage, and pathologic diagnosis (MC, AFM, or AC) were not. In the multivariate Cox model, only iPSA was significant (P ⫽ 0.0263, relative risk ⫽ 1.053, 95% confidence interval 1.006 to 1.102). COMMENT The differentiation of normal prostatic epithelial cells into prostate cancer involves numerous cytologic and genetic changes.16 Although mucinous changes may occur during prostate carcinogenesis,8,17,18 MC of the prostate represents a distinct pathologic entity from “conventional” prostatic AC. Prominent mucin production is associated with a poor prognosis in mucinous variants of adenocarcinoma of the colon19 and a good prognosis in breast cancer variants.20,21 Thus far, the clinical course of patients with MC of the prostate has not been well characterized. Several case reports have described the aggressive clinical nature of this rare variant.5–9,13,15 Other reports have indicated that some patients with MC experience a more benign clinical course.10 –12,14 To better understand the behavior of MC of the prostate, we evaluated our experience with mucinproducing carcinomas of the prostate during the past 18 years. We report what we believe to be the largest published series of mucinous adenocarcinomas of the prostate, totaling 14 patients with a median follow-up of 6.4 years. Although some degree of intracellular mucin production is detected in between 43% and 61% of conventional AC of the prostate, MC of the prostate is a rare entity.2,3,8,17,22 The distinction between these entities lies in the production of extracellular mucin, resulting in pools. To diagnose MC of the prostate, at least 25% of the tumor must show extracellular mucin production.1–3 On biopsy, the presence of extracellular mucin may not be reflective of the entire tumor and is referred to as “prostate adenocarcinoma with mucinous features.” Therefore, the diagnosis of MC can only be made at prostatectomy.3 By these criteria, MC represented 0.39% of prostatectomy specimens reviewed at our institution, yielding a prevalence comparable to that previously reported.3,6 Eighteen additional patients were identified as having mucinous features in less than 25% of the tumor; these cases were referred to as “prostatic adenocarci827
FIGURE 2. Kaplan-Meier curves depicting (A) overall survival and (B) bRFS in patients with MC (black circles), AFM (crosses), and matched controls with AC (black triangles). Number of patients at risk indicated at yearly intervals. Insets indicate percentage of patients surviving, 95% confidence intervals, and number of patients at risk at 5 and 11 years in MC, AFM, and AC groups and total population of 2572 patients with AC (unmatched AC).
828
UROLOGY 68 (4), 2006
TABLE II.
Comparison of clinical characteristics of patients with MC, AFM, and case-matched controls with AC
Characteristic Age (yr) Median Range Initial PSA (ng/mL) Median Range Clinical T-stage T1-T2a T2b-T3 Biopsy Gleason score ⱕ6 ⱖ7 Neoadjuvant therapy No Yes Surgical Gleason score ⱕ6 ⱖ7 Organ-confined disease Yes No
MC (n ⴝ 12)
AFM (n ⴝ 14)
AC (n ⴝ 52)
60.5 50.0–70.0
57.5 46.0–66.0
8.7 0.1–24.1
P Values MC vs. AFM
MC vs. AC
AFM vs. AC
59.0 45.0–71.0
0.17*
0.76*
0.18*
6.6 2.9–21.6
7.2 0.9–37.0
0.57*
0.88*
0.66*
0.99†
0.99†
0.99†
11 (91.7) 1 (8.3)
12 (85.7) 2 (14.3)
45 (86.5) 7 (13.5) 0.99†
0.72†
0.50†
8 (66.7) 4 (33.3)
9 (64.3) 5 (35.7)
39 (75.0) 13 (25.0) 0.09†
0.21‡
0.31†
0.61†
0.99†
0.73†
0.69†
0.63‡
0.76†
6 (50.0) 6 (50.0)
12 (85.7) 2 (14.3)
36 (69.2) 16 (30.8)
1 (16.7) 5 (83.3)
4 (33.3) 8 (66.7)
10 (27.8) 26 (72.2)
6 (50.0) 6 (50.0)
9 (64.3) 5 (35.7)
30 (57.7) 22 (42.3)
Abbreviations as in Table I. Data presented as numbers, with percentages in parentheses, unless otherwise noted. * Unpaired t test. † Fisher’s exact test. ‡ Chi-square test.
noma with focal mucin (AFM)” and represented an additional 0.50% of prostatectomy specimens. Compared with patients with MC, fewer patients with AFM had received neoadjuvant hormonal therapy and more had BF. In both regards, patients with AFM as a group were more similar to patients with AC. The standard pathologic distinction between MC and AFM may therefore be relevant clinically, even though no statistically significant differences between these two groups of patients were detected in the present study. Obviously, the small number of patients with MC or AFM made this comparison, as well as comparisons between these groups and patients with AC, difficult. Despite the limitations of this study, it is clear that MC is not associated with a poorer prognosis compared with AC (or AFM). No statistically significant differences were found among patients with MC, AFM, or AC with regard to organ-confined disease status or overall survival. With regard to disease progression, only 1 (8.3%) of 12 patients with MC who were followed up for longer than 6 months had experienced BF compared with 5 (35.7%) of 14 patients with AFM. The 5-year bRFS rate for those with MC, AFM, and AC was 100%, 69.2%, and 68.0%, respectively. No statistically significant difference was found in the three Kaplan-Meier curves, with the MC curve demonstrating the greatest proportion of UROLOGY 68 (4), 2006
patients without BF. We believe that MC is not associated with a worse prognosis than AC. Although only 1 patient with MC developed PSA recurrence, perhaps reflecting that MC is truly a less aggressive type of tumor than AC, this low biochemical failure rate may have been related to the small total number of patients with MC in this study, the greater percentage of patients receiving neoadjuvant hormonal therapy in the MC group, or a lack of sensitivity of PSA to detect MC recurrence. We believe the diagnosis of MC of the prostate should continue to be made as currently practiced.2,3 Differentiating between MC and AFM on the basis of the percentage of tumor demonstrating extracellular mucin production with a cutoff of 25% appears to separate “true” cases of MC from those that might be more accurately classified as conventional AC. Whole genome scanning is being used to identify genotypes associated with recurrence and metastasis in prostate cancer.16,23,24 Similar molecular studies may help to determine the genotype of MC and other rare tumors of the prostate, distinguishing those from AC. CONCLUSIONS We report what we believe to be the largest published series of cases of MC of the prostate (n ⫽ 14) 829
with a median overall follow-up of 6.4 years. The recurrence-free survival rate was 100% at 5 years, with a single BF evident 6.2 years after prostatectomy. In our study, the pathologic diagnosis of MC of the prostate, limited to tumors in which greater than 25% of the lesion contained pools of extracellular mucin, was not associated with a poor prognosis. REFERENCES 1. Bostwick DG, Qian J, and Schlesinger C: Contemporary pathology of prostate cancer. Urol Clin North Am 30: 181– 207, 2003. 2. Epstein JI, Algaba F, Allsbrook WC Jr, et al: Acinar adenocarcinoma, in Eble JN, Sauter G, Epstein JI, et al (Eds): World Health Organization Classification of Tumors: Pathology and Genetics of Tumors of the Urinary System and Male Genital Organs. Lyon, International Agency for Research on Cancer (IARC) Press, 2004, pp 162–192. 3. Epstein JI, and Yang XJ: Prostate Biopsy Interpretation, 3rd ed, Philadelphia, Lippincott Williams & Wilkins, 2002. 4. Elbadawi A, Craig W, Linke CA, et al: Prostatic mucinous carcinoma. Urology 13: 658 – 666, 1979. 5. Hsueh Y, and Tsung SH: Prostatic mucinous adenocarcinoma. Urology 24: 626 – 627, 1984. 6. Epstein JI, and Lieberman PH: Mucinous adenocarcinoma of the prostate gland. Am J Surg Pathol 9: 299 –308, 1985. 7. Manne RK, and Haddad FS: Mucinous adenocarcinoma of prostate. Urology 33: 247–249, 1989. 8. Ro JY, Grignon DJ, Ayala AG, et al: Mucinous adenocarcinoma of the prostate: histochemical and immunohistochemical studies. Hum Pathol 21: 593– 600, 1990. 9. Ishibashi Y, Kishida T, Yamaguchi T, et al: Two cases of mucinous adenocarcinoma of the prostate. Hinyokika Kiyo 38: 463– 467, 1992. 10. Watarai Y, Demura T, Togashi M, et al: Mucinous adenocarcinoma of the prostate: a case report and analysis of the literature. Nippon Hinyokika Gakkai Zasshi 85: 1276 –1279, 1994. 11. Olivas TP, and Brady TW: Mucinous adenocarcinoma of the prostate: a report of a case of long-term survival. Urology 47: 256 –258, 1996.
830
12. Teichman JM: Mucinous adenocarcinoma of the prostate: a report of a case of long-term survival. Urology 48: 660, 1996. 13. Saito S, and Iwaki H: Mucin-producing carcinoma of the prostate: review of 88 cases. Urology 54: 141–144, 1999. 14. Rhee AC, Olgac S, Ohori M, et al: Mucinous adenocarcinoma of the prostate: a case report of long-term disease-free survival and a review of the literature. Urology 63: 779 –780, 2004. 15. Larrousse C, Brasseur P, and Sukkarieh F: Port-site metastasis following laparoscopic radical prostatectomy for mucinous adenocarcinoma of the prostate. J Radiol 86: 337–339, 2005. 16. Dhanasekaran SM, Barrette TR, Ghosh D, et al: Delineation of prognostic biomarkers in prostate cancer. Nature 412: 822– 826, 2001. 17. Pinder SE, and McMahon RF: Mucins in prostatic carcinoma. Histopathology 16: 43– 46, 1990. 18. Legrier ME, de Pinieux G, Boye K, et al: Mucinous differentiation features associated with hormonal escape in a human prostate cancer xenograft. Br J Cancer 90: 720 –727, 2004. 19. Nakamori S, Ota DM, Cleary KR, et al: MUC1 mucin expression as a marker of progression and metastasis of human colorectal carcinoma. Gastroenterology 106: 353–361, 1994. 20. McGuckin MA, Walsh MD, Hohn BG, et al: Prognostic significance of MUC1 epithelial mucin expression in breast cancer. Hum Pathol 26: 432– 439, 1995. 21. Komenaka IK, El-Tamer MB, Troxel A, et al: Pure mucinous carcinoma of the breast. Am J Surg 187: 528 –532, 2004. 22. Murphy WM, Soloway MS, and Barrows GH: Pathologic changes associated with androgen deprivation therapy for prostate cancer. Cancer 68: 821– 828, 1991. 23. Kumar-Sinha C, and Chinnaiyan AM: Molecular markers to identify patients at risk for recurrence after primary treatment for prostate cancer. Urology 62(suppl 1): 19 –35, 2003. 24. Paris PL, Andaya A, Fridlyand J, et al: Whole genome scanning identifies genotypes associated with recurrence and metastasis in prostate tumors. Hum Mol Genet 13: 1303–1313, 2004.
UROLOGY 68 (4), 2006
