Urologic Oncology: Seminars and Original Investigations 33 (2015) 70.e1–70.e7

Original article

Metastatic small cell carcinoma of the prostate: Population-based analysis of patient characteristics and treatment paradigms Andrew Cohen, M.D., Kyle A. Richards, M.D., Sanjay Patel, M.D., Adam Weiner, B.S., Scott E. Eggener, M.D., Russell Z. Szmulewitz, M.D.* The University of Chicago Medicine, Chicago, IL Received 23 May 2014; accepted 10 June 2014

Abstract Introduction: Small cell carcinoma of the prostate is a rare malignancy comprising o1% of prostate cancers. Little is known about population-based treatment patterns for metastatic small cell carcinoma of the prostate. We evaluated clinical characteristics, treatment patterns, and survival outcomes. Methods: Using the National Cancer Database, we identified patients between 1998 and 2011 diagnosed with pure small cell carcinoma of the prostate as their only malignancy who presented with nodal involvement or distant metastasis. Results: Treatment information was available for 379 patients. Of them, 122 (32.5%) underwent chemotherapy (CT) alone, 25 (6.7%) received hormonal therapy (androgen-deprivation therapy) alone, 10 (2.7%) underwent radiation therapy alone, 3 (1%) underwent radical prostatectomy, and 167 (44.4%) underwent combination therapy. The 1- and 3-year survival rates were 35.3% and 4.4%, respectively. Those receiving any CT as part of their treatment had a median survival of 9.3 vs. 3.2 months for those not receiving it (P o 0.001). Those receiving CT, androgen-deprivation therapy, and radiation had a median survival of 15.1 vs. 7 months for those receiving CT alone (P o 0.001). On multivariable analysis (controlling for age, Charlson comorbidity index, extent of metastasis, prostate-specific antigen level, and type of treatment), older age (hazard ratio [HR] ¼ 3.87; 95% CI: 1.41–9.34; P ¼ 0.007) and distant metastatic disease (HR ¼ 7.17; 95% CI: 1.62– 31.8; P ¼ 0.010) increased risk of death, whereas receipt of CT (HR ¼ 0.15; 95% CI: 0.05–0.44; P ¼ 0.001) decreased risk of death. Conclusion: Men presenting with metastatic small cell carcinoma of the prostate have poor overall survival. Older patients and those presenting with distant metastases have an increased risk of death. It appears that patients receiving CT experience a modest survival benefit. The role of hormonal therapy in this population remains unclear. r 2014 Elsevier Inc. All rights reserved.

Keywords: Small cell carcinoma; Prostate cancer; Survival; Treatment

1. Introduction Small cell carcinoma is a rare variant of prostate cancer, and studies have reported metastasis at presentation ranging from 33% to 75% [1–3]. Although several case reports and single series have been published, very little is known about population-based treatment patterns for metastatic small cell carcinoma of the prostate [1,4,5]. There have been several single-arm phase II prospective clinical trials for metastatic small cell prostate cancer; however, the studies are limited by * Corresponding author. Tel.: þ1-773-702-7609. E-mail address: [email protected] (R. Szmulewitz).

http://dx.doi.org/10.1016/j.urolonc.2014.06.006 1078-1439/r 2014 Elsevier Inc. All rights reserved.

size, lack of randomization against a control arm, and heterogeneity regarding patient inclusion characteristics [6]. Few large multicenter population studies exist, and because of the low prevalence of disease, no guidelines have been proposed regarding treatment. Currently, patients undergo hormone therapy, radiation, systemic chemotherapy (CT), surgery, or multimodal combination therapy [6]. CT regimens are those commonly practiced in the treatment of small cell carcinoma of the lung (SCLC), with the most commonly used first-line regimen in extensive-stage SCLC being cisplatin or carboplatin with etoposide [7]. Small series of men with small cell carcinoma of the prostate have been treated with Adriamycin, vincristine, doxorubicin and cyclophosphamide, or etoposide and cisplatin with or without doxorubicin [8,9].

70.e2

A. Cohen et al. / Urologic Oncology: Seminars and Original Investigations 33 (2015) 70.e1–70.e7

As level 1 evidence to guide the management of metastatic small cell prostate cancer is lacking, the preferred therapeutic strategy for those with metastatic disease remains unknown. To our knowledge, no nationwide retrospective study to date has focused solely on men with metastatic small cell prostate cancer. Given the rarity of metastatic small cell prostate carcinoma, we sought to use the National Cancer Database (NCDB), which captures 70% of all newly diagnosed cancers in the United States each year, to better describe the patient characteristics, treatment modalities, and survival rates for patients with the disease [10]. The goal of this study was to capitalize on the size and clinical annotation of the NCDB to investigate any associations between clinical characteristics and treatment strategies and effect on overall survival. 2. Methods 2.1. Data source Institutional review board exemption was acquired as no patient, physician, or hospital identifiers were examined. The NCDB is a hospital-based cancer registry that captures approximately 70% of all cancer diagnoses in the United States yearly from more than 1,400 hospitals. The NCDB includes data on patient demographics, socioeconomic status, clinical and pathologic staging, treatment course, comorbidities, vital status based on patient records and death registry updates, and hospital level data. The disease course and therapy of each patient diagnosed and treated at a participating NCDB institution are coded and reported based on the American College of Surgeons' Facility Oncology Registry Data Standards (http://www.facs.org/cancer/coc/fordsmanual.html). 2.2. Inclusion criteria We identified all patients diagnosed with prostate cancer between 1998 and 2011. Of these men, we included those who were diagnosed with pure small cell carcinoma of the prostate (International Classification of Diseases (ICD)-O-3 morphologic codes 8002 and 8041–8044). We limited our study to men with prostate cancer as their first and only malignancy and those who presented with metastatic disease either via regional (American Joint Committee on Cancer [AJCC] category cN1) or distant metastasis (AJCC category cM1).

preexisting comorbidities. Clinical stage was coded according to the AJCC Cancer Staging Manual edition in use during the year in which the case was diagnosed. Patients were initially categorized by treatment type, including radical prostatectomy (RP), CT, hormonal therapy (androgen-deprivation therapy [ADT]), radiation therapy (RT), or multimodality therapy. The NCDB recorded PSA levels starting in 2004, as per collaborative staging stipulations, as the highest values before the diagnostic procedure or, if that value is unavailable, the earliest pretreatment but postdiagnostic value [12]. Gleason scores were not analyzed because a 2005 modification of the Gleason grading system determined that small cell carcinoma of the prostate should not be assigned a Gleason score [13]. The NCDB defines the patient's setting as their hometown population. These are categorized as metropolitan— more than 250,000 residents, urban—more than 2,500, and rural—fewer than 2,500. Hospital variables include geographic location and program type. Community programs were defined as those where more than 100 but fewer than 500 patients with cancer were treated per year. Comprehensive programs were defined as those where more than 500 cancer cases per year were treated. Academic programs were defined as those where more than 500 cancer cases per year were treated and graduate medical education was provided in at least 4 areas. Facility geographic location was categorized into 4 categories, including the Northeast, Southeast, Midwest, and West. 2.4. Statistical analysis Analysis was performed with Stata version 12 (Stata, College Station, TX). The Fisher exact, chi-square, and Mann-Whitney U tests were used for comparison of categorical variables. Kaplan-Meier survival analysis and the stratified log-rank test were used to compare overall survival stratified by extent of metastases (regional vs. distant) and various treatment modalities. Overall survival was estimated as time from diagnosis to death. Patients known to be alive were censored at the last follow-up evaluation or the last contact by correspondence. A multivariable Cox proportional hazards model was performed to identify independent predictors of overall survival. A 2-sided P o 0.05 was considered statistically significant. 3. Results

2.3. Study variables

3.1. Patient and disease variables

Patient demographic and clinical characteristics included age, race, cancer facility (type and location), metropolitan/ urban/rural, home county, cancer characteristics (prostatespecific antigen [PSA] level and AJCC clinical stage), and Deyo-Charlson comorbidity index (CCI) [11]. Race was categorized as non-Hispanic white, black, or other/unknown. The CCI was calculated based on the reporting of up to 6

Of the 1,774,062 cases of prostate cancers noted between 1998 and 2011, 896 (0.05%) were diagnosed with pure small cell carcinoma. Of those, 684 (76.3%) were diagnosed as only/first malignancy and 387 (43.2%) were diagnosed as metastatic upon presentation. Of the 387 patients in the cohort, the median age at diagnosis was 70 years, and 88.9% of the patients were white (Table 1). PSA data were

A. Cohen et al. / Urologic Oncology: Seminars and Original Investigations 33 (2015) 70.e1–70.e7

70.e3

Table 1 Continued

Table 1 Patient, hospital, and clinical characteristics Characteristic

No. (n ¼ 387)

Percentage

Age at diagnosis, y Median (IQR) o65 65–75 475

70 127 142 118

(61–78) 32.8 36.7 30.5

Race White Black Other Unknown

344 33 8 2

88.9 8.5 2.1 0.5

CCI 0 1 41 Unknown

226 49 17 95

58.4 12.7 4.4 24.6

Region Northeast South/Southeast Midwest West

101 96 104 86

26.1 24.8 26.9 22.2

Setting Metro Urban Rural Unknown

309 54 9 15

79.8 14.0 2.3 3.9

Program type Community Comprehensive Academic Other

47 216 115 9

12.1 55.8 29.7 2.3

Diagnosis year 1998–1999 2000–2003 2004–2007 2008–2011

37 87 97 166

9.6 22.5 25.1 42.9

PSA, ng/ml Median (IQR) o10 10–20 420 Unknown

4.1 123 23 41 200

(1.4–16) 31.8 5.9 10.6 51.7

Clinical T category cT1 cT2 cT3 cT4 Unknown

35 56 53 131 112

9.0 14.5 13.7 33.9 28.9

Local treatment RP RT No local Rxa

3 138 246

0.8 35.7 63.6

Systemic treatment CT No CT ADT

281 103 114

72.6 26.6 29.5

Characteristic No ADT Unknown Multimodal therapy Radiation and chemotherapy Radiation and ADT Radiation/chemotherapy/ADT Chemotherapy and ADT Extent of metastasis Lymph node only Distant metastasis Unknown

No. (n ¼ 387)

Percentage

267 6

69.0 1.5

79 15 31 42

21.0 4.0 8.2 11.2

57 327 3

14.7 84.5 0.8

IQR ¼ interquartile range. a No patient had RP þ RT.

available in 187 (48.3%) at the time of diagnosis. The median PSA level was 4.1 ng/ml (interquartile range: 1.4– 16), and while most were unfortunately unknown, 31.8% were less than 10 ng/ml. The patients were distributed evenly across regions, but 79.8% lived in metropolitan areas (4250,000 residents). Most of them (55.8%) were treated in designated comprehensive centers. There were increasing numbers of patients diagnosed throughout the study period with a plurality of patients diagnosed between 2008 and 2011 (42.9%). Nearly all patients presented with distant metastasis (84.5%), however, 14.7% of patients presented with regional metastasis alone. 3.2. Treatment modalities Of the patients within this cohort, 122 (32.5%) underwent CT alone, 25 (6.7%) received hormonal therapy (ADT) alone, 10 (2.7%) underwent RT alone, and 3 (1%) underwent RP. A plurality of patients, 167 (44.4%), underwent multimodal therapy, and a relatively large number of patients, 52 (13.7%), received no therapy for their disease. Of the patients who underwent multimodal therapy, combination CT and radiation was the most frequently used combination, used in 79 (21.0% of total cohort), with radiation and ADT in 15 (4.0% of total cohort), CT and ADT in 42 (11.2% of total cohort), and radiation, CT, and ADT in 31 (8.2% of total cohort). On comparing patients who received CT as part of their treatment with patients who did not receive it, patients who received CT were more likely to be younger (37.4% vs. 21.4% age o 65, P o 0.002) with a trend toward lower PSA levels (70.2% vs. 51.1% PSA o 10, P ¼ 0.055). Otherwise, there was no difference in CCI or the extent of metastasis between patients who received CT and those who did not receive it (Table 2). Of those men receiving RT, 69 (48.9%) received radiation to the prostate/pelvis, 28 (19.9%) received palliative radiation to the bone, and 9 (6.4%) received whole-brain radiotherapy. The

70.e4

A. Cohen et al. / Urologic Oncology: Seminars and Original Investigations 33 (2015) 70.e1–70.e7

Table 2 Characteristics of patients classified by treatment with chemotherapy

Age at diagnosis, y o65 65–75 475 CCI 0 1 41 PSA, ng/ml o10 10–20 420 Extent of metastasis Lymph node only Distant metastasis

Chemotherapy (n ¼ 281)

No chemotherapy (n ¼ 103)

n

%

n

%

105 103 73

37.4 36.7 25.9

22 38 43

21.4 36.9 41.7

176 35 11

79.3 15.8 5.0

47 14 6

70.2 20.9 9.0

99 16 26

70.2 11.4 18.4

23 7 15

51.1 15.6 33.3

P value

Median (mo) 0.002

0.25

0.055

0.7 43 238

15.3 84.7

14 89

Table 3 Median, 12-, and 36-mo survival based on extent of metastasis at presentation and treatment received

13.6 86.4

median time from diagnosis to radiation treatment was short at 41 days (interquartile range: 19–119), likely because of the percentage of patients receiving radiotherapy to the pelvis/ prostate around the time of diagnosis. 3.3. Overall survival The 1- and 3-year overall survival rates were 35.3% (95% CI: 27.6–41.0) and 4.4% (95% CI: 2.1–8.1), respectively (Table 3 and Fig. 1A). The median survival for the entire cohort was 7.6 (95% CI: 6.4–9.3) months. Those patients with nodal metastasis only had a median survival of 17.2 (95% CI: 7.9–20.7) vs. 7.0 (95% CI: 5.2–8.3) months for those with distant metastases (Table 3 and Fig. 1B, P ¼ 0.02). Those patients within the database who received any CT as part of their treatment had a significantly prolonged median survival of 9.5 (95% CI: 7.6–11.9) vs. 3.4 (95% CI: 2.3–5.2) months for those who did not receive it (Table 3 and Fig. 2A, P o 0.001). Those who received ADT as part of their treatment had a median survival of 10.9 (95% CI: 7.2–13.5) vs. 6.6 (95% CI: 5.1–8.2) months for those treated without it (Table 3 and Fig. 2B, P ¼ 0.1). Patients who received XRT as part of their treatment had a longer median survival of 10.9 (95% CI: 7.5–13.5) vs. 6.4 (95% CI: 4.5–7.9) months for those who did not receive it (Table 3 and Fig. 2C, P ¼ 0.04). Similarly, those who received RT in combination with CT seemed to have the longest median survival at 12.9 (95% CI: 8.4–15.1) months. Patients receiving neither therapy had the poorest survival at 3.2 (95% CI: 2.3–5.1) months. The median survival after radiotherapy alone (4.2 mo [95% CI: 1.2–11.6]) and CT alone (7.7 mo [95% CI: 6.6–10.4]) is in between with respect to overall survival (P ¼ 0.0002) (Fig. 2D). Our Cox proportional hazards model demonstrated that age 475 years (hazard ratio [HR] ¼ 3.87; 95% CI:

Overall Extent of metastasis Nodal only Distant metastasis Chemotherapy Yes No Radiation therapy Yes No Hormone therapy Yes No Radiation þ chemotherapy No chemotherapy or radiation Radiation alone Chemotherapy alone a

12 mo

36 mo

%

%

95% CI

95% CI

7.6

35.3 27.6–41.0 4.4 2.1–8.1

17.2 7.0

61.9 38.1–78.8 6.4 1.6–26.1 28.9 21.7–36.5 3.7 1.4–7.8

9.5 3.4

40.7 32.0–49.1 5.8 2.6–11.0 a 21.5 12.5–32.2 a

10.9 6.4

45.5 33.2–56.9 6.3 2.0–14.0 28.4 20.8–36.4 1.7 1.1–7.9

10.9 6.6 12.9 3.2

46.0 28.2 51.0 22.0

4.2 7.7

33.5–57.7 4.9 1.3–12.4 20.7–36.2 1.5 0.7–6.7 36.6–63.6 6.1 1.6–15.1 11.8–34.2 b b

a 20.0 4.9–42.4 a 33.5 23.0–44.3 2.4 1.1–10.6

Only 1 patient lived past 36 mo in these groups. No patient lived to 36 mo in this group.

b

1.41–9.34; P ¼ 0.007) and presence of distant metastases (HR ¼ 7.17; 95% CI: 1.62–31.8; P ¼ 0.01) increased risk of death, whereas treatment with CT was associated with a decreased risk of death (HR ¼ 0.15; 95% CI: 0.05–0.44; P ¼ 0.001; Table 4). ADT, PSA levels, and RT were not independent predictors of survival.

4. Discussion Metastatic small cell carcinoma of the prostate is a rare malignancy with a dismal prognosis. The current limited literature focuses on largely retrospective, single-institution series ranging from 17 to 191 subjects [2,3,14,15]. To our knowledge, we present the largest series of patients with metastatic small cell carcinoma of the prostate. The prevalence of pure small cell carcinoma of the prostate is low in our cohort (0.05%) when compared with that in other studies (0.5%–2%); however, this may represent exclusion of concurrent adenocarcinoma histology [16]. Within our data set, the plurality of patients (42.9%) was diagnosed in the last 3 years of the 13-year analysis. This may be because of a growing appreciation of small cell and neuroendocrine differentiation in the prostate cancer field as illustrated by recent review articles on the topic [6,22]. Our analysis is similar to that by Deorah et al., who reported a median survival of 7 (95% CI: 6–10) months for 89 patients presenting with distant metastasis using Surveillance, Epidemiology, and End Results database data. Smaller studies describe more variable survivorship, such as a series of 27 patients with survival ranging from 2 to 90 months [3]. The

A. Cohen et al. / Urologic Oncology: Seminars and Original Investigations 33 (2015) 70.e1–70.e7

Fig. 1. The Kaplan-Meier curve of overall survival. (Color version of figure is available online.)

heterogeneity in reported survival stems not only from smaller sample sizes, with potential bias introduction, but also more likely from the fact that the diagnosis of small cell carcinoma may be a term used to describe multiple different neuroendocrine forms of prostate cancer. Neuroendocrine prostate cancer was recently subdivided into a cohort of histologies, representing the spectrum of neuroendocrine differentiation [17]. Given the low incidence of pure small cell carcinoma of the prostate, there are limited data to inform clinical management. PSA level does not appear to be a screening tool or a prognostic factor. This is concordant with small cell carcinoma being thought of as androgen independent, with KLK3 (PSA) being an androgen receptor–regulated gene. Because of the different biology, PSA levels typically remain low despite metastatic disease [6]. Other serum biomarkers have been indentified for their potential to inform clinical decision making. Spiess et al. [1] found serum lactate dehydrogenase and albumin levels at the time of diagnosis to be predictive for inferior progression-free survival in 83 patients. Neuroendocrine secretory products, such as chromogranin A, and neurone-specific enolase have been associated with poor prognosis, tumor progression, and androgen independence [18]. The serum carcinoembryonic antigen concentration

70.e5

strongly predicted overall survival in a population with castrate-resistant prostate cancer with features characteristic of small cell prostate carcinoma [19]. These variables are not present within the NCDB, but might warrant additional prospective evaluation for their prognostic significance. In our study population, 73% received CT, which was independently associated with improved overall survival. Our study methodologies are limited in that the specific chemotherapies employed are not available within the database. Most commonly practiced regimens are based on those used in extensive-stage SCLC and include platinum (carboplatin or cisplatin) and etoposide or docetaxel and carboplatin [19]. There are several reported single-arm prospective clinical trials of combination CT for small cell prostate cancer. A recent 120patient investigation of “anaplastic” prostate cancer, including histologic small cell carcinoma, studied patients treated with carboplatin and docetaxel followed by cisplatin and etoposide. This study demonstrated clinical benefit to CT in that most patients were without progression after the first course of therapy and reported a median overall survival of 16 months [19]. In another single-arm phase II study, 36 patients with metastatic small cell prostate cancer were treated with combination doxorubicin, etoposide, and cisplatin. Although there was a 61% partial response rate, the authors noted high toxicity and believe the regimen failed to improve outcomes when compared with historical controls [20]. Given the rarity of pure prostatic small cell carcinoma, the likelihood of successfully completing a randomized prospective trial in this population is low. Current prospective trials of systemic therapy for this disease often include mixed histologies (e.g., adenocarcinoma with neuroendocrine features). Whether neuroendocrinedifferentiated prostate adenocarcinoma is a distinct entity in comparison with pure small cell carcinoma with separate biology, such as up-regulation of Aurora kinase, is unclear [17,21]. It is known that neuroendocrine-differentiated prostate cancers have limited responsiveness to androgen signaling inhibition [22]. Our analysis revealed an improvement in overall survival in patients receiving combination CT and RT. Although this did not persist on multivariable analysis, potentially because of small sample size, it is possible that RT offers an additional benefit to CT. It is also possible that patients who received more treatment were more fit at baseline, and thus the benefit of multimodal therapy may be because of selection bias inherent in such retrospective analysis. It is possible that RT confers a degree of local control, as the target of therapy in most cases was the prostate or the pelvis or both. The prospective data to support the use of multimodal therapy are limited. Case reports have demonstrated variable success with multimodal hormonal, CT, and radiation treatment [23]. In fact, there are several reports of rare long-term success with concomitant hormonal and CT in cases of single patients with mixed adenocarcinoma and small cell cancer of the prostate. For example, 1 reported patient who received cisplatin, etoposide, hormonal therapy, and 73-Gy of radiation to the prostate survived more than 36 months with complete

70.e6

A. Cohen et al. / Urologic Oncology: Seminars and Original Investigations 33 (2015) 70.e1–70.e7

Fig. 2. The Kaplan-Meier curve of overall survival based on treatment rendered. (Color version of figure is available online.)

remission [24]. There is another report of survival of 70 months using combination radiation and CT (albeit in a patient initially diagnosed with adenocarcinoma treated with androgen-deprivation therapy and later found to have small cell carcinoma) [20]. Such case reports are clearly limited, and on a population level, it remains unknown what treatment modalities are associated with the best long-term overall and disease-specific survival. RP was performed in only 3 patients in our population, as local surgical control and cytoreductive surgery do not currently have a role in the management of patients with metastatic small cell prostate cancer. Androgen-deprivation therapy did not significantly affect survival in our population. It is interesting to note that 30% of our patient population received androgen therapy alone or in combination with other therapy. Given the different pathophysiology of small cell carcinoma vs. adenocarcinoma of the prostate, it is not surprising that hormone therapy has a limited effect on survival [3,6,22]. It is unclear whether treatment with ADT represents a delay to adopt other treatment modalities or simply represents an attempt to use all available means for a very aggressive disease. Whether androgen-deprivation therapy has a role in mixed-histology

prostate cancer harboring both conventional adenocarcinoma and small cell carcinoma components is not known as these patients were excluded from our analysis. Given the available data, CT remains the foundation for treatment of those with metastatic small cell prostate cancer, and clinicians should encourage its use in all appropriate patients. Further study should be undertaken to specify appropriate treatment protocols. Understanding the underlying biology of aggressive prostate cancer variants, such as small cell prostate cancer, is a significant unmet need in the field [24]. Palliative care should likely be involved early, especially for those patients with distant metastasis who have a median survival of 7 months. We acknowledge the limitations of our data set, including the lack of randomization and confounding by treatment indication. We also do not have information about sequencing of treatment. Other drawbacks include those inherent to using the NCDB, an administrative data set, in a retrospective analysis. For example, the NCDB lacks specific information on specific CT drugs or number of cycles. Similar limitations in the RT data exist. Although patients were coded as having pure small cell carcinoma in the data set, there was no central

A. Cohen et al. / Urologic Oncology: Seminars and Original Investigations 33 (2015) 70.e1–70.e7 Table 4 Adjusted Cox proportional hazards model for the entire cohort with overall survival as primary outcome Characteristic Age at diagnosis, y o65 65–75 475 CCI 0 1 41 Extent of metastasis Node only Distant metastasis PSA Chemo No Yes Hormone No Yes Radiation No Yes

HR

95% CI

P value

REF 0.80 3.87

0.32–2.01 1.41–9.34

0.6 0.007

REF 3.11 5.52

1.14–8.52 0.474–64.4

0.027 0.17

REF 7.17 0.99

1.62–31.8 0.97–1.00

0.01 0.16

REF 0.15

0.05–0.44

0.001

REF 0.62

0.27–1.44

0.27

REF 1.73

0.85–2.54

0.13

REF ¼ reference.

path review and some may have mixed components. As noted, it is possible that patients who received aggressive therapy (such as CT) were the most fit. Although unlikely, it is possible that the choice of practitioners to use such aggressive therapy is itself a prognostic indicator. Without further clinical data, such as a patient's candidacy for CT, the true benefit of the therapy itself is not known. 5. Conclusion Small cell carcinoma of the prostate is an aggressive disease with poor overall survival. Patients who are eligible and receiving CT appear to have a modest survival benefit. Combination therapy with radiation and CT appears to confer an improvement in survival, suggesting a potential role for local control. Androgen deprivation does not appear to offer a benefit and likely should not be used given its detrimental effect on quality of life. Acknowledgments The authors thank Dr. Greer Gay for directing the NCDB database and assisting our analysis. References [1] Spiess PE, et al. Treatment outcomes of small cell carcinoma of the prostate—A single-center study. Cancer 2007;110:1729–37. [2] Mackey JR, Au HJ, Hugh J, Venner P. Genitourinary small cell carcinoma: determination of clinical and therapeutic factors associated with survival. J Urol 1998;159(5):1624–9.

70.e7

[3] Oesterling JE, Hauzeur CG, Farrow GM. Small cell anaplastic carcinoma of the prostate: a clinical, pathological and immunohistological study of 27 patients. J Urol 1992;147(3):804–7. [4] Dixit S, Coup A, Hunt C, Coombs L. Small cell cancer of the prostate. Urology 2012;80:58–60. [5] Anker CJ, Dechet C, Isaac JC, Akerley W, Shrieve DC. Diagnosis in oncology—small-cell carcinoma of the prostate. J Clin Oncol 2008; 26:1168–71. [6] Nadal R, Schweizer M, Kryvenko ON, Epstein JI, Eisenberger MA. Small cell carcinoma of the prostate. Nat Rev Urol 2014;11: 213–9. [7] National Comprehensive Cancer Network. Small cell cancer (Version 2.2014). 〈Available at: http://www.nccn.org/professionals/physician_ gls/pdf/small.pdf〉 [accessed 05.03.2014]. [8] Katou M, Soga N, Onishi T, Arima K, Sugimura Y. Small cell carcinoma of the prostate treated with amrubicin. Int J Clin Oncol 2008;13:169–72. [9] Amato RJ, Logothetis CJ, Hallinan R, Ro JY, Sella A, Dexeus FH. Chemotherapy for small cell carcinoma of prostatic origin. J Urol 1992;147:935–7. [10] Winchester DP, Stewart AK, Bura C, et al. The National Cancer Data Base: a clinical surveillance and quality improvement tool. J Surg Oncol 2004;85:1. [11] Berglund A, Garmo H, Tishelman C, et al. Comorbidity, treatment and mortality: a population based cohort study of prostate cancer in PCBaSe Sweden. J Urol 2011;185:833. [12] Brawer MK, Beatie J, Wener MH, et al. Screening for prostatic carcinoma with prostate specific antigen: results of the second year. J Urol 1993;150:106. [13] Epstein JI, Allsbrook WC Jr, Amin MB, et al. The 2005 International Society of Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma. Am J Surg Pathol 2005; 29:1228. [14] Deorah S, Rao MB, Raman R, Gaitonde K, Donovan JF. Survival of patients with small cell carcinoma of the prostate during 1973–2003: a population-based study. BJU Int 2012;109:824–30. [15] Pervez F, et al. Genitourinary small-cell carcinoma: a singleinstitution experience. Curr Oncol 2013;20:258–64. [16] Palmgren JS, Karavadia SS, Wakefield MR. Unusual and underappreciated: small cell carcinoma of the prostate. Semin Oncol 2007;34:22–9. [17] Epstein JI, et al. Proposed morphologic classification of prostate cancer with neuroendocrine differentiation. Am J Surg Pathol 2014;38 (6):756–67. [18] Abrahamsson PA. Neuroendocrine cells in tumour growth of the prostate. Endocr Relat Cancer 1999;6:503–19. [19] Aparicio AM, et al. Platinum-based chemotherapy for variant castrate-resistant prostate cancer. Clin Cancer Res 2013;19: 3621–3630. [20] Papandreou CN, et al. Results of a phase II study with doxorubicin, etoposide, and cisplatin in patients with fully characterized small-cell carcinoma of the prostate. J Clin Oncol 2002;20: 3072–3080. [21] Beltran H, et al. Molecular characterization of neuroendocrine prostate cancer and identification of new drug targets. Cancer Discov 2011;1:487–95. [22] Aggarwal R, Zhang T, Small EJ, Armstrong AJ. Neuroendocrine prostate cancer: subtypes, biology, and clinical outcomes. J Natl Compr Canc Netw 2014;12:719–26. [23] Uemura K-I, et al. A useful treatment for patients with advanced mixed-type small cell neuroendocrine carcinoma of the prostate: a case report. Oncol Lett 2013;5:793–6. [24] Brammer JE, Lulla P, Lynch GR. Complete remission in a patient with metastatic mixed adenocarcinoma/extrapulmonary small cell carcinoma of the prostate Int. J Clin Oncol 2011;16: 722–725.