Prognostic Significance of the Proportion of Ductal Component in Ductal Adenocarcinoma of the Prostate

Author's Accepted Manuscript Prognostic significance of the proportion of ductal component in ductal adenocarcinoma of the prostate Won Sik Jang , Su-Jin Shin , Cheol Yong Yoon , Myung Soo Kim , Dong Hyuk Kang , Yong Jin Kang , Won Sik Jeong , Nam Hoon Cho , Young Deuk Choi

PII: DOI: Reference:

S0022-5347(16)31887-0 10.1016/j.juro.2016.11.104 JURO 14229

To appear in:

The Journal of Urology

Please cite this article as: Jang WS, Shin SJ, Yoon CY, Kim MS, Kang DH, Kang YJ, Jeong WS, Cho NH, Choi YD, Prognostic significance of the proportion of ductal component in ductal adenocarcinoma of the prostate, The Journal of Urology® (2017), doi: 10.1016/j.juro.2016.11.104. DISCLAIMER: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our subscribers we are providing this early version of the article. The paper will be copy edited and typeset, and proof will be reviewed before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to The Journal pertain.

Embargo Policy All article content is under embargo until uncorrected proof of the article becomes available online. We will provide journalists and editors with full-text copies of the articles in question prior to the embargo date so that stories can be adequately researched and written. The standard embargo time is 12:01 AM ET on that date. Questions regarding embargo should be directed to [email protected].

1

ACCEPTED MANUSCRIPT

Prognostic significance of the proportion of ductal component in ductal adenocarcinoma of the prostate

RI PT

Authors: Won Sik Jang a, Su-Jin Shin b, Cheol Yong Yoon a, Myung Soo Kim a, Dong Hyuk Kang a, Yong Jin Kang a, Won Sik Jeong a, Nam Hoon Cho b, and Young Deuk Choi a*

SC

Affiliation: a

b

M AN U

Department of Urology, Urological Science Institute, Yonsei University College of Medicine, Seoul, Korea Department of Pathology, Yonsei University College of Medicine, Seoul, Korea

E-mail addresses of the co-authors:

[email protected]

TE D

[email protected]

EP

Word count of text: 2499 (Word count of the abstract: 243)

AC C

*Corresponding author:

Young Deuk Choi M.D., Ph.D. Professor

Department of Urology and Urological Science Institute Yonsei University College of Medicine 50-1, Yonsei-ro, Seodaemun-gu, Seoul 120-752, Korea E-mail: [email protected]; Tel: +82-2-2228-2317; Fax: +82-2-312-2538

Keywords: prostatic neoplasms; prostatectomy; carcinoma, ductal

2

ACCEPTED MANUSCRIPT ABSTRACT PURPOSE: In prostate cancer, ductal adenocarcinoma is mixed with usual acinar adenocarcinoma. However, whether the proportion of ductal component affects oncologic

RI PT

outcomes is currently unknown. Here, we investigated whether the proportion of the ductal component predicts oncologic outcomes in ductal adenocarcinoma.

MATERIALS AND METHODS: We retrospectively reviewed clinical data from 3,038

SC

patients with prostate cancer who underwent radical prostatectomy at our institution between 2005 and 2014. We excluded patients who received neoadjuvant or adjuvant treatment.

M AN U

Patients were stratified based on the proportion of the ductal component. We compared the probability of biochemical recurrence between groups and investigated how the proportion of the ductal component influences biochemical recurrence using Kaplan-Meier estimates and Cox regression models, respectively.

TE D

RESULTS: Of 2,648 patients, 101 (3.8%) had ductal adenocarcinoma and 2,547 (96.2%) had acinar adenocarcinoma. Freedom from biochemical recurrence for patients with ductal adenocarcinoma was significantly lower compared with those with acinar adenocarcinoma (p

EP

<0.001). When ductal cases were stratified by the proportion of the ductal component,

AC C

freedom from biochemical recurrence for the high ductal component (≥30%) group was significantly lower compared that of the low ductal component (<30%) group (p = 0.023). In univariate and multivariate Cox regression analyses, a high ductal component was a significant predictor of biochemical recurrence (p <0.001). CONCLUSIONS: The prognosis for ductal adenocarcinoma can be stratified by the proportion of the ductal component. This marker could potentially be used as a surrogate for poor prognosis or as a determinant for adjuvant therapy.

3

ACCEPTED MANUSCRIPT INTRODUCTION Ductal adenocarcinoma of the prostate (DAP) is a rare, aggressive histologic subtype of prostate cancer. The prevalence of DAP has increased to 5% for mixed ductal and 0.4%–0.8%

RI PT

for pure ductal.1 Histologically, DAP has tall, pseudostratified columnar epithelium with abundant cytoplasm that takes on papillary or cribriform architecture. It can be found either in the central zone or the urethra proximal to verumontanum or the peripheral zone where it is

SC

accompanied by acinar adenocarcinoma of the prostate (AAP).2-5

DAP has characteristics similar to those of advanced AAP. Both spread cancer cells into

M AN U

periurethral ducts, which is an adverse pathological feature.5,6 Recent studies reported that radical prostatectomy (RP) specimens of DAP generally show adverse pathological features, such as high Gleason score, large tumor volume, positive surgical margins (PSM), extraprostatic extension (EPE), seminal vesicle invasion (SVI), and lymph node (LN)

TE D

metastasis.7,8 Epstein et al. recommended that DAP be defined as Gleason pattern 4 because it has a clinical behavior similar to Gleason score 8 (4+4) AAP.9

EP

DAP is often mixed with usual AAP, but whether the proportion of the ductal component (DC) affects oncologic outcomes is unknown.2 We hypothesized that the proportion of the

AC C

DC can predict oncologic outcomes after RP for DAP. Therefore, we compared clinical and pathological characteristics of DAP based on the proportion of the DC and investigated its prognostic significance with respect to freedom from biochemical recurrence (BCR) after RP in patients with DAP.

4

ACCEPTED MANUSCRIPT MATERIALS AND METHODS Patients We obtained approval for these studies from the Institutional Review Board. We

RI PT

retrospectively reviewed clinical data collected from our prostate cancer database. At our institution, 3,038 patients had undergone RP for clinically localized or locally advanced prostate cancer between January 2005 and December 2014. We excluded patients who had

SC

received neoadjuvant or adjuvant treatment (i.e., radiation, androgen deprivation therapy, or

M AN U

both) and those without adequate medical records. In total, 2,648 men were included in the final analysis. Patient characteristics

We reviewed our institutional medical records to obtain patients’ clinical and pathological

TE D

features, including age, preoperative prostate specific antigen (PSA) level, tumor node metastasis (TNM) stage, Gleason score, and margin status of the RP specimen. TNM stage

staging system.9

EP

was determined according to the American Joint Committee on Cancer 7th edition TNM

AC C

Pathological analysis

Pathological analyses of RP specimens were performed as described previously by experienced uropathologists in our institute.10 Briefly, the surface of the resected prostate specimen was coated with India ink, fixed in neutral buffered formalin, and embedded in a paraffin block. Whole-mount step sections were cut transversely at regular intervals from the apex of the prostate to the tips of the seminal vesicles. Each section was examined for SVI, EPE, and PSM. The Gleason score was assigned based on the 2005 International Society of Urological Pathology Modified Gleason System.11 DAP is considered as Gleason pattern 4

5

ACCEPTED MANUSCRIPT while it was regarded as Gleason pattern 5 if there is true comedonecrosis.12 Tumor volume was assessed, and the percentage of the DC in the total tumor was measured using planimetry with a digitizer tablet.13

RI PT

DAP was defined histologically as pseudostratified, tall or columnar epithelium with papillary or cribriform architecture (Fig.1).2,5 We excluded high-grade prostatic intraepithelial neoplasia (HGPIN)-like DAP and intraductal carcinoma of the prostate (IDC-P) both showing

SC

histological similarity to DAP. Histologically it may be difficult to distinguish DAP from HGPIN or IDC-P. In contrast to columnar nuclei and slit-like lumina with true papillary

M AN U

architecture of DAP, HGPIN is composed of round-oval nuclei and punched-out round lumina with micropapillary architecture.14

HGPIN-like DAP shows

a favorable

characteristics similar to a Gleason pattern of 3.15 There is morphological overlap between DAP and IDC-P.2 Distinctive features seen in IDC-P include cuboidal cells, cribriform

TE D

patterns with rounded lumina, and micropapillary tufts without fibrovascular cores.14 Follow-up

EP

Postoperative PSA follow-up was performed at 3-month intervals for the first 2 years and 6month intervals for the next 3 years. PSA follow-ups were recommended annually thereafter.

AC C

BCR was defined as any two consecutive increases in serum PSA ≥0.2 ng/ml following RP. Statistical analysis

We used the Contal and O’Quigley method, based on the log-rank test, to determine the optimal cut-point for the prognostication based on the proportion of DC component in RP specimen.16 Patients were divided into two groups (DC <30% vs. DC ≥30%) according to cut-point by the Contal and O’Quigley method. We compared clinical and pathological characteristics between groups using the Mann–Whitney U test for continuous data and χ2

6

ACCEPTED MANUSCRIPT test for dichotomous variables. We estimated and compared the freedom from BCR between groups using the Kaplan-Meier method with a log-rank test. Univariate and multivariate Cox regression models were built to investigate whether the DC has prognostic relevance in

RI PT

addition to previously well-established prognostic parameters. Patients age and preoperative PSA were examined as continuous variables while T stage, Gleason score, PSM, and LN metastasis were examined as a categorical variable. Significant variables from univariate

SC

analysis were included in the multivariate model. All statistical analyses were conducted using SPSS 18.0 (SPSS, Inc., Chicago, IL, USA) or SAS 9.4 (SAS Institute, Cary, NC, USA).

AC C

EP

TE D

M AN U

Comparisons with p values <0.05 were considered statistically significant.

7

ACCEPTED MANUSCRIPT RESULTS Of 2,648 patients in the final cohort, 101 patients (3.8%) had DAP (12 pure type, 11.9%; 89 mixed type, 88.1%) and 2,547 patients (96.2%) had AAP. The median follow-up from RP was

RI PT

66 months (interquartile range [IQR] 41–85), and the maximum follow-up was 126 months. Table 1 shows the differences in clinical and pathological characteristics between cancer types. Patients in the DAP group were slightly older than patients in the AAP group (median

SC

age 68 vs. 66 years, respectively, p = 0.030) and had a higher median PSA (11.9 vs. 7.7 ng/ml, respectively, p <0.001). Additionally, T stage and Gleason score were higher in DAP patients

M AN U

than those with AAP (p <0.001 for both). Patients with DAP were more likely to have LN metastases (13.9% vs. 4.1% for AAP patients, p <0.001). The PSM rate did not differ significantly between two groups (p = 0.264).

A comparison of clinical and pathological characteristics following stratification into

TE D

patients with a high proportion of the DC (≥30%) and those with a low proportion of the DC (<30%) was detailed in Table 2. There was a significant difference in T stage between low and high DC groups (p = 0.038). Patient age, preoperative PSA, Gleason score, PSM, or LN

EP

metastasis did not differ significantly between two groups.

AC C

The 5-year freedom from BCR rate for patients with DAP was lower than that for patients with AAP (29.5% and 71.6%, respectively). Kaplan-Meier analysis showed that the difference in freedom from BCR between DAP and AAP patients was significant (log-rank test, p <0.001, Fig.2).

When the DAP group was stratified, Kaplan-Meier analysis showed that high DC was significantly associated with decreased freedom from BCR (log-rank test, p = 0.023, Fig.3). Table 3 shows the results of univariate and multivariate Cox regression analyses predicting

8

ACCEPTED MANUSCRIPT BCR following RP. In the multivariate analysis, patient age (hazard ratio [HR] 1.013, p = 0.018), PSA (HR 1.004, p <0.001), T stage (T3a: HR 1.405; ≥T3b: HR 2.972, p <0.001 for both), Gleason score ≥9 (HR 2.237, p <0.001), PSM (HR 2.073, p <0.001), and DC (HR

RI PT

1.013, p <0.001) were independent prognostic factors for BCR. There was no association between LN metastasis and BCR (p = 0.498). In multivariate analysis using DC as a categorical variable, high DC (HR 2.933, p <0.001) was an independent prognostic factor for

AC C

EP

TE D

M AN U

SC

BCR, while low DC was not (p = 0.315).

9

ACCEPTED MANUSCRIPT DISCUSSION DAP was first identified in 1967 by Melicow and Pachter. It was originally thought to have originated from the prostatic utricle as a Mullerian ductal structure.3 Initially, it was referred

RI PT

to as an endometrioid cancer of the prostate due to its histological similarity to uterine endometrial adenocarcinoma. However, immunohistochemical studies demonstrated that these tumors originate from prostatic ducts.17,18,19 Now, DAP is recognized as a prostatic

SC

adenocarcinoma variant by both the Armed Forces Institute of Pathology and the World Health Organization due to its distinct and unique origin and its pathological and clinical

M AN U

features.20

DAP was initially described as growing along the urethra and infiltrating locally but metastasizing rarely.21 Since then, there have been contradictory reports observing advanced stage at RP and demonstrating a high probability of failure following RP.4,7 Recently, several

TE D

studies demonstrated that RP specimens with DAP show adverse pathological features including high Gleason score, large tumor volume, PSM, EPE, SVI, and LN metastasis.2,8 Moreover, Seipel et al. reported that the immunohistochemical expression profile of DAP is

EP

similar to that of AAP, but it shows higher expression of Ki-67, p53, and p16. This

AC C

observation supports the hypothesis that DAP is a biologically aggressive subtype of prostate cancer.22 Thus, DAP is believed to be more aggressive than AAP. Surveillance, Epidemiology, and End Results (SEER) data have shown that DAP has a worse oncological outcome than AAP.23 Morgan et al. reported that DAP is more likely to be high grade and present with distant disease. It has a significantly increased mortality risk in patients with loco-regional disease independent of age, race, grade, clinical T stage, nodal status, or treatment modality. They also reported that DAP was associated with a 30% decrease in geometric mean PSA and more than 2-fold increased odds of PSA <4.0 ng/ml

10

ACCEPTED MANUSCRIPT than those with AAP. In contrast, we identified that DAP group had a higher median PSA than patients in the AAP group. It could be due to higher tumor burden and more adverse pathologic features in patients with DAP. In study of Morgan et al., 62% of DAP group were

RI PT

organ confined disease and some DAP cases were incidentally identified through endoscopic resection of a urethral polyp. Moreover, the results that 19% of DAP in the previous analysis had Gleason score of 6 causes a question about the accuracy of the SEER data.24 Therefore,

SC

we think that one of the reasons for the discrepancy in pre-operative PSA value between the current and previous reports is clinic-pathological heterogeneity between two studies.

M AN U

Meeks et al. reported similar findings.25 In their study, cancer-specific and overall survival for DAP patients was worse than that of AAP patients. When AAP was grouped by Gleason score, oncological outcomes in patients with DAP were similar to those in patients with Gleason score 8 (4+4) tumors. However, these studies were limited because the percentage of the DC

TE D

in each patient was not recorded. Thus, they could not investigate the clinicopathological features and oncological outcome of DAP based on the proportion of the DC. DAP often occurs as a mixed tumor with AAP and accounts for approximately 5% of all

EP

RP cases.1 However, the relationship between the proportion of the DC and oncologic

AC C

outcome has not been fully investigated. Amin et al. reported that mixed DAP (DC ≥10%) with Gleason pattern 3 is more aggressive than AAP with Gleason score 7.2 However, they could not show a significant difference in prognosis following RP. Moreover, AAP with Gleason pattern 4 were not considered; therefore, they could not extend these findings to all Gleason 4 patterns. Samaratunga et al. reported that the presence of DC, large tumor volume, and high Gleason score were independently associated with pT3 disease and investigated whether the proportion of DC is associated with pT3 tumors.26 They did not observe any significant difference among these categories when cases were stratified based on the

11

ACCEPTED MANUSCRIPT percentages of the DC. In this study, freedom from BCR for patients with DAP was worse than that of patients with AAP. These results support those of previous studies, indicating that DAP is more

RI PT

aggressive than AAP. We also validated that the proportion of the DC has prognostic significance in patients with DAP. Our results show that freedom from BCR is significantly lower for men with a high DC compared with those with a low DC. In the multivariate Cox

SC

regression analysis, DC was an independent and significant predictor of BCR.

M AN U

Our study has important clinical implications. First, to the best of our knowledge, this is the first study reporting the prognostic significance of the proportion of the DC in DAP. Second, our study is one of the largest single-center studies to report on DAP. Third, we had an experienced urological pathologist measure the proportion of the DC in DAP by

TE D

histologically reviewing the entire tumor from each RP specimen.

Our study also has several limitations. First, our data were reviewed retrospectively from a single institution; therefore, our results may not be generalizable. Second, our study endpoint

EP

was BCR. We know that BCR is not a surrogate for oncologic outcome and that cancerspecific mortality is a more appropriate endpoint.27,28 However, the median follow-up for our

AC C

data was only 66.0 months. This period is long enough to draw meaningful conclusions about BCR,29 but it is too short to evaluate cancer-specific survival. Because the association between BCR and cancer-specific death has been well documented,30 we believe the endpoint of BCR was sufficient for our research purpose. Lastly, although our study evaluated one of the largest series of DAP cases, the number of pure DAP cases were limited and a comparison of clinical behavior between the pure and mixed DAP could not be performed. To confirm our findings, a larger sample size and longer follow-up are required.

12

ACCEPTED MANUSCRIPT CONCLUSIONS In our study, we demonstrated that the proportion of the DC is an independent prognostic factor for BCR following a RP in patients with DAP. These findings show that the prognosis

RI PT

of DAP could be stratified by the proportion of the DC, indicating that the DC could as a

CONFLICT OF INTEREST STATEMENT

ACKNOWLEDGEMENTS

M AN U

The authors declare no conflicts of interest.

SC

surrogate for poor prognosis or a determinant for adjuvant therapy.

EP

2011-04-02), Korea.

TE D

This study was supported by a grant from the Korean Foundation for Cancer Research (CB-

AC C

FIGURE LEGENDS

Fig.1 Ductal adenocarcinoma of the prostate composed of tall columnar cells with pseudostratified nuclei and displaying cytological atypia Fig.2 Kaplan-Meier curves for freedom from biochemical recurrence in patients with acinar adenocarcinoma of the prostate (AAP) and ductal adenocarcinoma of the prostate (DAP) Fig.3 Kaplan-Meier curves for freedom from biochemical recurrence in patients with acinar adenocarcinoma of the prostate (AAP) and ductal adenocarcinoma of the prostate (DAP) stratified based on the proportion of the ductal component (DC)

13

ACCEPTED MANUSCRIPT REFERENCES 1.

Epstein JI, Partin AW, Potter SR et al: Adenocarcinoma of the prostate invading the se minal vesicle: prognostic stratification based on pathologic parameters. Urology 2000; 56: 283. Amin A and Epstein JI: Pathologic stage of prostatic ductal adenocarcinoma at radical

RI PT

2.

prostatectomy: effect of percentage of the ductal component and associated grade of acinar adenocarcinoma. Am J Surg Pathol 2011; 35: 615.

Melicow MM and Pachter MR: Endometrial carcinoma of proxtatic utricle (uterus masc ulinus). Cancer 1967; 20: 1715.

4.

Christensen WN, Steinberg G, Walsh PC et al: Prostatic duct adenocarcinoma. Findings at radical prostatectomy. Cancer 1991; 67: 2118.

5.

SC

3.

Aydin H, Zhang J, Samaratunga H et al: Ductal adenocarcinoma of the prostate diagn

M AN U

osed on transurethral biopsy or resection is not always indicative of aggressive disease : implications for clinical management. BJU Int 2010; 105: 476. 6.

Bock BJ and Bostwick DG: Does prostatic ductal adenocarcinoma exist? Am J Surg Pat hol 1999; 23: 781.

7.

Bostwick DG, Kindrachuk RW and Rouse RV: Prostatic adenocarcinoma with endometri oid features. Clinical, pathologic, and ultrastructural findings. Am J Surg Pathol 1985; 9 : 595.

Brinker DA, Potter SR and Epstein JI: Ductal adenocarcinoma of the prostate diagnose

TE D

8.

d on needle biopsy: correlation with clinical and radical prostatectomy findings and pr ogression. Am J Surg Pathol 1999; 23: 1471. 9.

Edge SB BD, Compton CC, Fritz AG, Greene FL, Trotti A: AJCC cancer staging manual ( 7th ed). New York, NY: Springer 2010.

Koo KC, Tuliao P, Komninos C et al: Prognostic impact of time to undetectable prosta

EP

10.

te-specific antigen in patients with positive surgical margins following radical prostatec 11.

AC C

tomy. Ann Surg Oncol 2015; 22: 693. Epstein JI, Allsbrook WC, Jr., Amin MB et al: The 2005 International Society of Urologi

cal Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma . Am J Surg Pathol 2005; 29: 1228.

12.

Shah RB: Current perspectives on the Gleason grading of prostate cancer. Arch Pathol Lab Med 2009; 133: 1810.

13.

Ruijter ET, Miller GJ, Aalders TW et al: Rapid microwave-stimulated fixation of entire p rostatectomy specimens. Biomed-II MPC Study Group. J Pathol 1997; 183: 369.

14.

Guo CC and Epstein JI: Intraductal carcinoma of the prostate on needle biopsy: Histol ogic features and clinical significance. Mod Pathol 2006; 19: 1528.

15.

Tavora F and Epstein JI: High-grade prostatic intraepithelial neoplasialike ductal adenoc arcinoma of the prostate: a clinicopathologic study of 28 cases. Am J Surg Pathol 200

14

ACCEPTED MANUSCRIPT 8; 32: 1060. 16.

Contal C and O'Quigley J: An application of changepoint methods in studying the effe ct of age on survival in breast cancer. Computational Statistics & Data Analysis 1999; 30: 253.

17.

Myers RH: Classical and modern regression with applications, 2nd ed. Boston: PWS-KE NT 1990; p viii. Greene LF, Farrow GM, Ravits JM et al: Prostatic adenocarcinoma of ductal origin. J Ur ol 1979; 121: 303.

19.

Eade TN, Al-Saleem T, Horwitz EM et al: Role of radiotherapy in ductal (endometrioid) carcinoma of the prostate. Cancer 2007; 109: 2011.

20.

Liu T, Wang Y, Zhou R et al: The update of prostatic ductal adenocarcinoma. Chin J C

SC

ancer Res 2016; 28: 50. 21.

Tannenbaum M: Endometrial tumors and/or associated carcinomas of prostate. Urology

M AN U

1975; 6: 372. 22.

RI PT

18.

Seipel AH, Samaratunga H, Delahunt B et al: Immunohistochemical profile of ductal ad enocarcinoma of the prostate. Virchows Arch 2014; 465: 559.

23.

Morgan TM, Welty CJ, Vakar-Lopez F et al: Ductal adenocarcinoma of the prostate: inc reased mortality risk and decreased serum prostate specific antigen. J Urol 2010; 184: 2303.

24.

Kim A, Kwon T, You D et al: Clinicopathological features of prostate ductal carcinoma: 2015; 30: 385.

25.

TE D

matching analysis and comparison with prostate acinar carcinoma. J Korean Med Sci Meeks JJ, Zhao LC, Cashy J et al: Incidence and outcomes of ductal carcinoma of the prostate in the USA: analysis of data from the Surveillance, Epidemiology, and End Re sults program. BJU Int 2012; 109: 831.

Samaratunga H, Duffy D, Yaxley J et al: Any proportion of ductal adenocarcinoma in r

EP

26.

adical prostatectomy specimens predicts extraprostatic extension. Hum Pathol 2010; 41: 27.

AC C

281.

Stephenson AJ, Kattan MW, Eastham JA et al: Defining biochemical recurrence of prost

ate cancer after radical prostatectomy: a proposal for a standardized definition. J Clin Oncol 2006; 24: 3973.

28.

Imnadze M, Sjoberg DD and Vickers AJ: Adverse Pathologic Features at Radical Prostat

ectomy: Effect of Preoperative Risk on Oncologic Outcomes. Eur Urol 2016; 69: 143.

29.

Johansson JE, Andren O, Andersson SO et al: Natural history of early, localized prostat e cancer. Jama 2004; 291: 2713.

30.

Freedland SJ, Humphreys EB, Mangold LA et al: Risk of prostate cancer-specific mortali ty following biochemical recurrence after radical prostatectomy. JAMA 2005; 294: 433.

ACCEPTED MANUSCRIPT Table 1. Baseline clinical and pathological characteristics Total (n = 2,648) 66 61–70 7.8 5.3–13.4

Age, years, median IQR PSA, ng/ml, median IQR T stage, n (%) T2 T3a ≥T3b Gleason score, n (%) ≤8 ≥9 PSM, n (%) No Yes LN metastasis, n (%) No Yes

Acinar (n = 2,547) 66 61–70 7.7 5.2–13.1

Ductal (n = 101) 68 61–72 11.9 7.4–26.7

p value 0.030 <0.001

RI PT

Variable

<0.001

2,310 (90.7) 237 (9.3)

M AN U

2,383 (90.0) 265 (10.0)

1,174 (46.1) 1,102 (43.3) 271 (10.6)

1,362 (51.4) 1,286 (48.6) 2,530 (95.5) 118 (4.5)

27 (26.7) 47 (46.6) 27 (26.7)

SC

1,201 (45.4) 1,149 (43.3) 298 (11.3)

1,316 (51.7) 1,231 (48.3) 2,443 (95.9) 104 (4.1)

73 (72.3) 28 (27.7)

TE D EP

0.264

46 (45.5) 55 (54.5) <0.001 87 (86.1) 14 (13.9)

IQR = interquartile range; PSA = prostate specific antigen; PSM = positive surgical margin; LN = lymph node.

AC C

<0.001

ACCEPTED MANUSCRIPT Table 2. Comparison of clinical and pathological characteristics of patients with ductal adenocarcinoma of the prostate stratified by the proportion of the ductal component

Total (n = 101) 68 61–72 11.9 7.4–26.7

Age, years, median IQR PSA, ng/ml, median IQR T stage, n (%) T2 T3a ≥T3b Gleason score, n (%) 8 ≥9 PSM, n (%) No Yes LN metastasis, n (%) No Yes

DC < 30% (n = 22) 68 59–72 8.0 6.2–19.1

DC ≥ 30% (n = 79) 68 62–73 14.4 8.1–28.0

p value 0.364

RI PT

Variable

0.139 0.038

46 (45.5) 55 (54.5)

TE D

87 (86.1) 14 (13.9)

18 (22.7) 42 (53.2) 19 (24.1)

SC

73 (72.3) 28 (27.7)

9 (40.9) 5 (22.7) 8 (36.4)

16 (72.7) 6 (27.3)

M AN U

27 (26.7) 47 (46.6) 27 (26.7)

13 (59.1) 9 (40.9) 19 (86.4) 3 (13.6)

57 (72.2) 22 (27.8) 0.226 33 (41.8) 46 (58.2) >0.999 68 (86.1) 11 (13.9)

EP

IQR = interquartile range; PSA = prostate specific antigen; PSM = positive surgical margin; LN = lymph node; DC = ductal component.

AC C

>0.999

ACCEPTED MANUSCRIPT

Table 3. Univariate and multivariate analyses of factors associated with biochemical recurrence

RI PT

Multivariate b

p value

HR (95% CI)

p value

HR (95% CI)

p value

1.024 (1.013–1.035) 1.005 (1.005–1.006)

<0.001 <0.001

1.013 (1.002–1.024) 1.004 (1.003–1.005)

0.018 <0.001

1.013 (1.002–1.024) 1.004 (1.003–1.005)

0.022 <0.001

1 (Ref) 2.014 (1.686–2.405) 6.326 (5.167–7.735)

<0.001 <0.001

1 (Ref) 1.405 (1.163–1.697) 2.972 (2.360–3.743)

<0.001 <0.001

1 (Ref) 1.406 (1.164–1.698) 2.959 (2.349–3.729)

<0.001 <0.001

1 (Ref) 5.145 (4.300–6.156)

<0.001

1 (Ref) 2.237 (1.846–2.711)

<0.001

1 (Ref) 2.188 (1.803–2.654)

<0.001

1 (Ref) 2.073 (1.743–2.465)

<0.001

1 (Ref) 2.072 (1.742–2.464)

<0.001

1 (Ref) 1.102 (0.833–1.458)

0.498

1 (Ref) 1.085 (0.821–1.434)

0.568

1.013 (1.009–1.017)

<0.001

-

1 (Ref) 1.659 (1.243–2.215)

M AN U

<0.001

TE D

1 (Ref) 2.952 (2.516–3.464)

SC

HR (95% CI)

<0.001

EP

Age PSA T stage T2 T3a ≥T3b Gleason score ≤8 ≥9 PSM No Yes LN metastasis No Yes DC Continuous variable Categorical variable AAP DAP (DC <30%) DAP (DC ≥30%)

Multivariate a

Univariate

1.018 (1.014–1.022)

<0.001

1 (Ref) 2.066 (1.028–4.152) 4.595 (3.464–6.094)

0.041 <0.001

AC C

Variable

-

1 (Ref) 1.435 (0.709–2.903) 2.933 (2.199–3.913)

0.315 <0.001

PSA = prostate specific antigen; PSM = positive surgical margin; LN = lymph node; DC = ductal component; AAP = acinar adenocarcinoma of the prostate; DAP = ductal adenocarcinoma of the prostate; HR = hazard ratio; CI = confidence interval. a Multivariate analyses using DC as a continuous variable b Multivariate analyses using DC as a categorical variable

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT STANDARD ABBREVIATIONS ductal adenocarcinoma of the prostate acinar adenocarcinoma of the prostate ductal component radical prostatectomy positive surgical margins extraprostatic extension seminal vesicle invasion lymph node biochemical recurrence prostate specific antigen high-grade prostatic intraepithelial neoplasia intraductal carcinoma of the prostate hazard ratio confidence interval

AC C

EP

TE D

M AN U

SC

RI PT

DAP AAP DC RP PSM EPE SVI LN BCR PSA HGPIN IDC-P HR CI