Causes of Death Among Patients With Metastatic Prostate Cancer in the US From 2000 to 2016

Ahmed O Elmehrath; Ahmed M Afifi; Muneer J Al-Husseini; Anas M Saad; Nathaniel Wilson; Kyrillus S Shohdy; Patrick Pilie; Mohamad Bassam Sonbol; Omar Alhalabi

doi:10.1001/jamanetworkopen.2021.19568

. 2021 Aug 5;4(8):e2119568. doi: 10.1001/jamanetworkopen.2021.19568

Causes of Death Among Patients With Metastatic Prostate Cancer in the US From 2000 to 2016

Ahmed O Elmehrath ¹, Ahmed M Afifi ^2,³, Muneer J Al-Husseini ⁴, Anas M Saad ^3,⁵, Nathaniel Wilson ⁶, Kyrillus S Shohdy ⁷, Patrick Pilie ⁸, Mohamad Bassam Sonbol ⁹, Omar Alhalabi ^8,^✉

¹Faculty of Medicine, Cairo University, Cairo, Egypt

²University of Kentucky College of Medicine, Lexington

³Faculty of Medicine, Ain Shams University, Cairo, Egypt

⁴Department of Medicine, Ascension St John Hospital, Detroit, Michigan

⁵Cleveland Clinic Foundation, Cleveland, Ohio

⁶Department of Internal Medicine, McGovern Medical School, The University of Texas Health Science Center at Houston

⁷Division of Hematology and Medical Oncology, Department of Medicine, Weill Cornell Medicine, New York, New York

⁸Department of Genitourinary Medical Oncology, Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center, Houston, Texas

⁹Mayo Clinic Cancer Center, Division of Hematology and Oncology, Mayo Clinic, Phoenix, Arizona

Accepted for Publication: May 23, 2021.

Published: August 5, 2021. doi:10.1001/jamanetworkopen.2021.19568

^✉

Corresponding Author: Omar Alhalabi, MD, Department of Genitourinary Medical Oncology, Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center, 1155 Pressler St, Unit 1374, CPB7.3538, Houston, TX 77030 (oalhalabi@mdanderson.org).

Author Contributions: Mr Elmehrath and Dr Afifi contributed equally to this manuscript. Mr Elmehrath and Dr Saad had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Elmehrath, Afifi, Al-Husseini, Saad, Shohdy, Sonbol, Alhalabi.

Acquisition, analysis, or interpretation of data: Elmehrath, Afifi, Saad, Wilson, Pilie, Sonbol, Alhalabi.

Drafting of the manuscript: Elmehrath, Al-Husseini, Saad, Shohdy, Sonbol, Alhalabi.

Critical revision of the manuscript for important intellectual content: Elmehrath, Afifi, Wilson, Shohdy, Pilie, Sonbol, Alhalabi.

Statistical analysis: Elmehrath, Afifi, Al-Husseini, Saad, Shohdy.

Administrative, technical, or material support: Saad, Pilie.

Supervision: Al-Husseini, Shohdy, Sonbol, Alhalabi.

Conflict of Interest Disclosures: Dr Pilie reported receiving grant funding from the Prostate Cancer Foundation unrelated to this work. Dr Alhalabi reported receiving grant funding from the American Society of Clinical Oncology Conquer Cancer Foundation unrelated to this work. No other disclosures were reported.

^✉

Corresponding author.

PMCID: PMC8343467 PMID: 34351403

This cohort study uses data from the Surveillance, Epidemiology, and End Results Program to identify the most common causes of death among US men diagnosed with metastatic prostate cancer from 2000 to 2016.

Key Points

Question

What are the most common causes of death in patients diagnosed with metastatic prostate cancer (PC)?

Findings

In this cohort study of 26 168 patients with metastatic PC, 16 732 died during the follow-up period. Of these deaths, 77.8% were from PC, 5.5% were from other cancers, and 16.7% were from noncancer causes, including cardiovascular diseases, chronic obstructive pulmonary disease, and cerebrovascular diseases.

Meaning

These findings suggest that therapy and follow-up should be tailored to the needs of each patient with metastatic PC and counseling should be provided regarding future health risks.

Abstract

Importance

Owing to improved survival among US patients with prostate cancer (PC), patients tend to live long enough after a PC diagnosis for non–cancer-related comorbidities to be associated with their overall survival. Although studies have investigated causes of death among patients with localized PC, data are lacking regarding causes of death among patients with metastatic PC.

Objective

To assess causes of death among US patients with metastatic PC from 2000 to 2016.

Design, Setting, and Participants

This retrospective cohort study used data from the Surveillance, Epidemiology, and End Results Program database to analyze a sample of 26 168 US men who received a diagnosis of metastatic PC from January 1, 2000, to December 31, 2016. Data were analyzed from February 2 to July 28, 2020.

Exposure

Diagnosis of metastatic PC.

Main Outcomes and Measures

Standardized mortality ratios (SMRs) for different causes of death were calculated by dividing the observed number of deaths from each cause of death by the expected number of deaths in the age-matched US male population for the same period, adjusting for age and race/ethnicity.

Results

Of 26 168 patients with metastatic PC included in the analysis, 48.9% were aged 50 to 70 years (mean age at diagnosis, 70.83 years); 74.5% were White individuals, and 72.7% received a diagnosis of stage M1b metastatic PC. A total of 16 732 patients (63.9%) died during the follow-up period. The mean age at death was 74.13 years. Most deaths (59.0%) occurred within the latency period of 2 years after diagnosis of metastatic PC, whereas 31.6% occurred 2 to 5 years after diagnosis and 9.4% occurred more than 5 years after diagnosis. Of the total deaths, 13 011 (77.8%) were from PC, 924 (5.5%) were from other cancers, and 2797 (16.7%) were from noncancer causes. During all latency periods, the most common noncancer causes of death were cardiovascular diseases (SMR, 1.34; 95% CI, 1.26-1.42), chronic obstructive pulmonary disease (SMR, 1.19; 95% CI, 1.03-1.36), and cerebrovascular diseases (SMR, 1.31; 95% CI, 1.13-1.50).

Conclusions and Relevance

In this cohort study, deaths from noncancer causes, including cardiovascular disease, constituted a substantial number of deaths among men with metastatic PC. Therapy and follow-up should be tailored to the needs of each patient with metastatic PC, and counseling regarding future health risks should be provided.

Introduction

Prostate cancer (PC) is the most frequently diagnosed cancer among men in the US, with approximately 191 930 newly diagnosed cases per year, and the PC mortality rate is the second highest cancer mortality rate among men, with an estimated 33 330 deaths per year in the US.¹ There has been a recent decrease in the incidence of PC overall in the US,² owing in part to decreased screening in accordance with the US Preventive Services Task Force recommendations.³ However, the incidence of metastatic PC has increased in the past decade.^4,5 Between 1991 and 2017, deaths associated with PC decreased, but the trend seems to have steadied recently.^1,6,7 The decreased mortality is likely associated with the advancement in systemic therapies and multidisciplinary treatment strategies.^{8,9,10,11,12,13}

Owing to the improved survival among patients with PC in the US,¹⁴ patients tend to live long enough after a PC diagnosis for non–cancer-related comorbidities to be associated with their overall survival.^{15,16,17,18,19} Because of the high prevalence of PC and its consequences for public health, many studies from different countries^20,21,22,23 have evaluated and reported causes of death (CODs) after PC diagnosis, with some studies^20,21,22,23 reporting varying trends in competing CODs after PC diagnosis. A Swedish study²¹ reported that noncancer CODs (primarily cardiovascular disease) were the most common COD (31%) among patients with low-risk PC, followed by other cancers (30%) and PC (18%). On the contrary, a study from Korea²² showed that PC accounted for 46.3% of deaths among patients with PC, followed by other cancers (35.4%) and cardiovascular disease (6.6%). However, both of those studies analyzed the CODs among patients with localized PC or among older patients with PC.

In this study, we used data from the Surveillance, Epidemiology, and End Results (SEER) Program database to conduct a long-term population-based analysis of noncancer CODs after a diagnosis of metastatic PC. We analyzed the data with respect to different demographic and tumor-related factors to investigate whether there were any associations between certain factors and specific CODs.

Methods

Study Design and Data Source

In this retrospective cohort study,²⁴ we used SEER*Stat software, version 8.3.5,²⁵ to access the 2018 version of the SEER 18 registries, which included 27.8% of the general US population from January 1, 2000, to December 31, 2016.²⁶ Data were analyzed from February 2 to July 28, 2020. SEER data are anonymized, and use of the data is considered as non–human participant research. Thus, institutional review board approval and informed consent were not needed. The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Study Population

We included US men with histologically proven metastatic prostatic adenocarcinoma diagnosed from January 1, 2000, to December 31, 2016. Cancer registry entries are linked to vital statistics to ascertain entries in death records, in which vital status was ascertained until December 31, 2016. For patient selection, we used the site recode World Health Organization International Classification of Diseases for Oncology, 3rd Edition variable to select prostate; the histology recode–broad grouping variable to select 8140-8398 adenomas and adenocarcinomas; a variable from the American Joint Committee on Cancer (AJCC) Cancer Staging Manual, 6th edition (SEER*Stat variable, “Stage - 6th edition. Derived AJCC M, 6th ed [2004-2015]”) to select metastatic disease for patients diagnosed before 2015; and a variable from the Cancer Staging Manual, 7th edition (SEER*Stat variable, “Stage - 7th edition. Derived SEER Combined M [2016+]”) to select metastatic disease for patients diagnosed after 2015. To assess metastatic disease, we selected cases with M1a, M1b, or M1c stage according to the AJCC Cancer Staging Manual, 6th edition and the derived AJCC M, 6th ed (2004-2015) variable, and we selected cases with c1A, c1B, c1C, p1A, p1B, or p1C stage according to the AJCC Cancer Staging Manual, 7th edition and the derived SEER combined M (2016+) variable. To minimize the risk of selection bias, we included all eligible men with metastatic PC documented in the SEER registries.

Outcomes

For included men with metastatic PC, we inspected noncancer CODs with respect to the following variables: age at diagnosis, race, and treatment (surgery, radiotherapy, and chemotherapy). With regard to the prognosis of metastatic PC, a recent study by Siegel et al¹⁴ using US Cancer Statistics registries (2001-2017) reported a 5-year survival rate of approximately 30% among patients with metastatic PC. Therefore, to identify the most clinically relevant COD among patients with metastatic PC and inform clinically relevant decisions regarding follow-up, we focused on periods of up to 5 years after diagnosis and more than 5 years after diagnosis. In addition, we stratified CODs within the initial 2 years after diagnosis because we observed that 59% of deaths had already occurred by 2 years. Thus, we classified CODs by latency period as less than 2 years, 2 to 5 years, and more than 5 years after diagnosis of metastatic PC. The CODs were obtained using the SEER Cause of Death Recode, which is based on death certificates and, since 1999, has been based on International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) codes.²⁷ Under the category other infectious and parasitic diseases, we included tuberculosis, syphilis, and other bacterial, viral, and parasitic diseases. Examples of other CODs include stomach and duodenal ulcers, homicide, and legal intervention. Treatment-related deaths were included as adverse-event CODs. The definition of each COD in the study according to ICD-10 is provided in eTable 1 in the Supplement.

Statistical Analysis

Standardized mortality ratios (SMRs) were calculated for each COD after a diagnosis of metastatic PC as the observed to expected ratio, in which observed represented patients with metastatic PC who died of a specific COD and expected represented patients in a demographically similar population who were expected to die of the same COD. Because COD is likely to differ by age and race/ethnicity, consideration needed to be given to this through adjustment or stratification. Thus, the SMR was calculated by dividing the observed number of deaths by the expected number of deaths in a demographically similar population, adjusting for age and race/ethnicity as demographic variables. The SMRs in this study represented the change in risk of a specific COD after a diagnosis of metastatic PC compared with the risk among the general population of men in the US. Data are presented in 3 groups based on the latency from time of diagnosis: less than 2 years, 2 to 5 years, and more than 5 years. Mortality rates in the general US male population (observed deaths divided by expected deaths) were gathered by the National Center for Health Statistics between 1969 and 2017 and were retrieved using SEER*Stat software, version 8.3.5.²⁶

We calculated SMRs with 95% CIs using SEER*Stat software, version 8.3.5. A significant increase in risk was defined as observed deaths attributed to a specific COD after a diagnosis of metastatic PC being greater than expected deaths from that COD with P < .05. All statistical tests were 2-sided.

Results

Baseline Characteristics

A total of 26 168 patients with a diagnosis of metastatic PC were included in the analysis; 48.9% were aged 50 to 70 years; 74.5% were White individuals, and 72.7% had received a diagnosis of stage M1b metastatic PC. The mean age at diagnosis was 70.83 years. Overall survival analysis of the included patient cohort showed a median survival of 29 months (interquartile range, 13-63 months), a 1-year survival rate of 77.5%, and a 5-year survival rate of 26.4%. Of included patients, 16 732 (63.9%) died during the follow-up period; the mean age at death was 74.13 years. Most deaths (59.0%) occurred within 2 years after diagnosis, whereas 31.6% occurred from 2 to 5 years and 9.4% occurred after 5 years (Table 1).

Table 1. Baseline Characteristics of Patients With Metastatic Prostate Cancer and of Those Who Died According to the Time of Death After Diagnosis.

Characteristic	Diagnosed cases, No.	Deaths, No.	Age at death, mean (SD), y	Deaths by time after diagnosis, No. (%)
Characteristic	Diagnosed cases, No.	Deaths, No.	Age at death, mean (SD), y	<2 y	2-5 y	>5 y
All patients	26 168	16 732	74.13	9869 (59.0)	5290 (31.6)	1573 (9.4)
Age at diagnosis, y
<50	625	384	49.24	190 (49.5)	156 (40.6)	38 (9.9)
50-70	12 797	7393	64.94	3965 (53.6)	2616 (35.4)	812 (11.0)
>70	12 746	8955	82.79	5714 (63.8)	2518 (28.1)	723 (8.1)
Race
White	19 486	12 592	74.96	7361 (58.5)	4036 (32.1)	1195 (9.5)
Black	4989	3246	70.52	2004 (61.7)	960 (29.6)	282 (8.7)
American Indian or Alaska Native	162	104	73.51	67 (64.4)	28 (26.9)	9 (8.7)
Asian or Pacific Islander	1531	790	75.87	437 (55.3)	266 (33.7)	87 (11.0)
Cancer stage
M1a	1604	794	73.09	349 (44.0)	342 (43.1)	103 (13.0)
M1b	19 017	12 004	74.40	6903 (57.5)	3945 (32.9)	1156 (9.6)
M1c	5547	3934	73.54	2617 (66.5)	1003 (25.5)	314 (8.0)
Treatment
Cancer-directed surgery	2949	1826	75.74	1057 (57.9)	577 (31.6)	192 (10.5)
Radiotherapy	6108	3793	71.29	2296 (60.5)	1152 (30.4)	345 (9.1)
Chemotherapy	2780	1290	67.36	828 (64.2)	377 (29.2)	85 (6.6)

Open in a new tab

Of the total deaths, 13 011 (77.8%) were from PC, 924 (5.5%) were from other cancers, and 2797 (16.7%) were from noncancer causes (Table 2). During all latency periods, the most common noncancer CODs were cardiovascular diseases (SMR, 1.34; 95% CI, 1.26-1.42), chronic obstructive pulmonary disease (SMR, 1.19; 95% CI, 1.03-1.36), and cerebrovascular diseases (SMR, 1.31; 95% CI, 1.13-1.50). The Figure shows the proportions of different CODs according to time of death after PC diagnosis.

Table 2. Observed Deaths and SMRs for Causes of Death After Diagnosis of Metastatic Prostate Cancer.

Cause of death	Deaths by time after diagnosis						Total deaths
	<2 y		2-5 y		>5 y		Total deaths
	Observed, No. (%)	SMR (95% CI)^a	Observed, No. (%)	SMR (95% CI)^a	Observed, No. (%)	SMR (95% CI)^a	Observed, No. (%)	SMR (95% CI)^a
All	9869 (100)	6.43 (6.30-6.56)^b	5290 (100)	6.07 (5.90-6.23)^b	1573 (100)	3.63 (3.45-3.81)^b	16 732 (100)	5.89 (5.80-5.98)^b
Prostate cancer	7792 (79.0)	NA	4171 (78.8)	NA	1048 (66.6)	NA	13 011 (77.8)	NA
Other cancers	527 (5.3)	1.68 (1.54-1.82)^b	271 (5.1)	1.52 (1.35-1.72)^b	126 (8.0)	1.50 (1.25-1.78)^b	924 (5.5)	1.60 (1.50-1.71)^b
Noncancer causes^c	1550 (15.7)	1.32 (1.26-1.39)^b	848 (16.0)	1.27 (1.19-1.36)^b	399 (25.4)	1.19 (1.08-1.31)^b	2797 (16.7)	1.29 (1.24-1.33)^b
Septicemia	69 (4.5)	3.00 (2.34-3.80)^b	31 (3.7)	2.37 (1.61-3.36)^b	8 (2.0)	1.21 (0.52-2.38)	108 (3.9)	2.53 (2.08-3.05)^b
Infectious and parasitic diseases including HIV infection	21 (1.4)	1.55 (0.96-2.38)	9 (1.1)	1.20 (0.55-2.28)	3 (0.8)	0.87 (0.18-2.54)	33 (1.2)	1.35 (0.93-1.90)
Diabetes	58 (3.7)	1.23 (0.93-1.59)	34 (4.0)	1.27 (0.88-1.77)	10 (2.5)	0.75 (0.36-1.38)	102 (3.6)	1.17 (0.95-1.42)
Alzheimer disease^d	27 (1.7)	0.57 (0.37-0.83)^b	21 (2.5)	0.76 (0.47-1.15)	16 (4.0)	1.03 (0.59-1.67)	64 (2.3)	0.70 (0.54-0.90)^b
Cardiovascular diseases	653 (42.1)	1.40 (1.29-1.51)^b	335 (39.5)	1.28 (1.14-1.42)^b	159 (39.8)	1.23 (1.05-1.44)^b	1147 (41.0)	1.34 (1.26-1.42)^b
Cerebrovascular diseases	107 (6.9)	1.30 (1.07-1.58)^b	55 (6.5)	1.19 (0.90-1.55)	36 (9.0)	1.56 (1.10-2.17)^b	198 (7.1)	1.31 (1.13-1.50)^b
Pneumonia and influenza	51 (3.3)	1.28 (0.96-1.69)	30 (3.5)	1.34 (0.91-1.92)	10 (2.5)	0.90 (0.43-1.66)	91 (3.3)	1.24 (1.00-1.53)^b
COPD and associated conditions	99 (6.4)	1.05 (0.86-1.28)	72 (8.5)	1.34 (1.05-1.69)^b	36 (9.0)	1.35 (0.95-1.88)	207 (7.4)	1.19 (1.03-1.36)^b
Chronic liver disease and cirrhosis	19 (1.2)	1.46 (0.88-2.28)	4 (0.5)	0.56 (0.15-1.42)	2 (0.5)	0.63 (0.08-2.28)	25 (0.9)	1.07 (0.69-1.58)
Nephritis, nephrotic syndrome, and nephrosis	36 (2.3)	1.00 (0.70-1.39)	19 (2.2)	0.93 (0.56-1.45)	16 (4.0)	1.54 (0.88-2.50)	71 (2.5)	1.06 (0.83-1.34)
Accidents and adverse effects of medications	72 (4.6)	1.70 (1.33-2.14)^b	37 (4.4)	1.55 (1.09-2.13)^b	13 (3.3)	1.10 (0.59-1.89)	122 (4.4)	1.56 (1.30-1.86)^b
Suicide and self-inflicted injury	30 (1.9)	2.97 (2.00-4.24)^b	19 (2.2)	3.42 (2.06-5.34)^b	5 (1.3)	2.01 (0.65-4.68)	54 (1.9)	2.97 (2.23-3.88)^b
Other	308 (19.9)	1.20 (1.07-1.34)^b	182 (21.5)	1.22 (1.05-1.41)^b	85 (21.3)	1.08 (0.86-1.33)	575 (20.6)	1.18 (1.09-1.28)^b

Open in a new tab

Abbreviations: COPD, chronic obstructive pulmonary disease; NA, not applicable; SMR, standardized mortality ratio.

^{^a}

P < .05.

^{^b}

The SMR was calculated by dividing the observed number of deaths from each cause of death by the expected number of deaths in the age-matched US male population for the same period, adjusting for age and race/ethnicity.

^{^c}

Percentages for individual noncancer causes of death reflect the percentage of total noncancer causes of death. Percentages may not sum to 100 owing to rounding.

^{^d}

Based on International Classification of Diseases, Ninth Revision; International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes.

CODs Within 2 Years After Metastatic PC Diagnosis

A total of 9869 deaths (59.0% of all deaths) occurred within 2 years after diagnosis of metastatic PC; 7792 (79.0%) patients died of PC, 527 (5.3%) died of nonprostate cancers, and 1550 (15.7%) died of noncancer causes. The most common noncancer COD was cardiovascular disease (653 deaths [42.1%]), followed by cerebrovascular disease (107 [6.9%]) and chronic obstructive pulmonary disease (COPD) (99 [6.4%]).

The overall risk of death among patients with metastatic PC within 2 years after diagnosis was higher than that in the general US male population (SMR, 6.43; 95% CI, 6.30-6.56), as was the risk of death from cardiovascular disease (SMR, 1.40, 95% CI, 1.29-1.51) (Table 2). In general, trends in CODs within 2 years after diagnosis of metastatic PC were similar across various demographic and tumor-related subgroups (eTables 2-14 in the Supplement).

Men younger than 50 years at the time of diagnosis of metastatic PC (eTable 2 in the Supplement) had an increased overall risk of death within 2 years after their diagnosis compared with men younger than 50 years without metastatic PC (SMR, 42.53; 95% CI, 36.70-49.03). In subgroups by cancer stage (eTables 9-11 in the Supplement), patients with PC with visceral involvement (stage M1c) had the greatest risk of death within 2 years after diagnosis (SMR, 9.03; 95% CI, 8.69-9.38) (eTable 11 in the Supplement).

Non–PC-related causes accounted for 21 of 190 deaths (11.1%) among men younger than 50 years who died within 2 years after metastatic PC diagnosis. In contrast, non–PC-related causes accounted for 2056 of 9679 deaths (21.2%) within 2 years after metastatic PC diagnosis among men 50 years or older (Fisher exact test P < .001). Non–PC-related causes among men with metastatic PC who were older than 50 years included cardiovascular disease (651 deaths [31.2%]), cerebrovascular disease (107 [5.2%]), and COPD (99 [4.8%]).

CODs From 2 to 5 Years After Metastatic PC Diagnosis

A total of 5290 men with metastatic PC died from 2 to 5 years after their cancer diagnosis, of whom 4171 (78.8%) died of PC, 271 (5.1%) died of nonprostate cancers, and 848 (16.0%) died of noncancer causes. The most common noncancer COD was cardiovascular disease (335 patients [39.5%]), followed by COPD (72 [8.5%]) and cerebrovascular diseases (55 [6.5%]) (Table 2).

Among men with metastatic PC, the overall risk of death from 2 to 5 years after their cancer diagnosis was significantly greater than that in the general US male population (SMR, 6.07; 95% CI, 5.90-6.23). With regard to noncancer CODs, rates of death from septicemia, suicide, accidents, COPD, and cerebrovascular diseases were significantly increased among men diagnosed with metastatic PC compared with the general US male population (Table 2).

In general, CODs from 2 to 5 years after diagnosis of metastatic PC followed similar trends across various demographic and tumor-related subgroups (eTables 2-14 in the Supplement). All age groups (eTables 2-4 in the Supplement) had a significantly higher overall risk of death from 2 to 5 years after diagnosis. Of note, White patients (SMR, 3.04; 95% CI, 2.26-4.01) and Asian or Pacific Islander patients (SMR, 5.47; 95% CI, 1.13-16) had an increased risk of suicide, but Black patients (SMR, 0.89; 95% CI, 0.02-4.94) and American Indian or Alaska Native patients (SMR, 0.00; 95% CI, 0.00-82.06) did not have an increased risk of suicide (eTables 5-8 in the Supplement).

CODs More Than 5 Years After Metastatic PC Diagnosis

A total of 1573 men with metastatic PC died more than 5 years after their cancer diagnosis, of whom 1048 (66.6%) died of metastatic PC, 126 (8.0%) died of nonprostate cancers, and 399 (25.4%) died of noncancer causes. The most common noncancer COD was cardiovascular disease (159 patients [39.8%]), followed by cerebrovascular disease (36 [9.0%]) and COPD (36 [9.0%]) (Table 2). Among men with metastatic PC, the overall risk of death more than 5 years after diagnosis was significantly greater than that in the general US male population (SMR, 3.63; 95% CI, 3.45-3.81) (Table 2).

Discussion

This cohort study showed that most deaths (59.0%) from metastatic PC occurred within 2 years after diagnosis among US men with metastatic PC diagnosed from 2000 to 2016. Non–PC-related causes accounted for a greater percentage of deaths among older patients (aged ≥50 years) compared with younger patients (aged <50 years). Furthermore, we observed that the number of deaths from non–PC-related causes increased in association with increasing latency period after the diagnosis of metastatic PC, and cardiovascular and cerebrovascular diseases were the most common causes of non–PC-related deaths. Overall survival analysis of the patient cohort showed 1-year survival of 77.5% and 5-year survival of 26.4%; these rates are comparable with those in a study using US Cancer Statistics,¹⁴ which reported 1-year and 5-year survival for distant-stage PC of approximately 75% and approximately 30%, respectively.

Cardiovascular diseases were the most common cause of non–PC-related deaths in all latency periods. In a previous SEER-based analysis,²⁸ death from cardiac disease was found to be more common in patients with cancer than in the general population. Furthermore, another study²⁹ showed that men with PC and no prior cardiac disease had greater risk of heart failure after androgen deprivation therapy (ADT). Low testosterone bioavailability may be associated with increased risk of atherosclerosis and ischemic heart diseases.^30,31,32 Another study³³ suggested an association between cardiotoxic effects of ADT and myocardial infarction regardless of medical history in general. This finding highlights the importance of multidisciplinary care for such patients and the role of primary care physicians in optimizing cardiovascular risk prevention and providing early referrals to cardiologists; tailoring the approach of ADT to each patient’s needs may be associated with improved survival, especially for patients with factors associated with cardiovascular disease.

Cerebrovascular disease was among the most prevalent noncancer CODs in our analysis during all examined latency periods and in all subgroups. A previous meta-analysis³⁴ showed that gonadotropin-releasing hormone analogues, either alone or with oral antiandrogen, or orchiectomy was significantly associated with stroke in patients with PC. Other studies also showed a similar association between stroke and ADT.^35,36 This finding may be attributable to the association between low serum testosterone levels and many factors associated with stroke, including low-density lipoprotein cholesterol level, triglyceride levels, endothelial dysfunction, and proinflammatory factors.^{30,31,32,37,38,39,40} Of interest, opposing results were also reported in a study⁴¹ of a large cohort that showed an association between ADT and decreased risk of stroke (adjusted hazard ratio, 0.88; 95% CI, 0.81-0.96; P = .001). Thus, the association of ADT with stroke in patients with PC remains controversial. Of note, in this study, other CODs included a combination of less common CODs, such as stomach and duodenal ulcers, homicide, and legal intervention (eTable 1 in the Supplement).

Prostate cancer is particularly associated with the development of other primary cancers such as lung, colon, and thyroid cancers.⁴² The results of this study suggest that men with metastatic PC may have a greater risk of death from other cancers than from PC. This finding may be important in counseling patients about their risk of developing other cancers, especially with data indicating that patients with multiple primary malignancies are enriched for germline mutations.⁴³

Limitations

This study has limitations. The findings may be limited by the inherent bias in retrospective analyses, which have some restrictions regarding the adjusted factors. In addition, there is potential reporting bias in death certificates, leading to misclassification of CODs.^44,45 However, mortality data provided by the National Vital Statistics System and National Center for Health Statistics follow standardized and systematic data collection procedures to ensure accuracy of CODs recorded in SEER,^46,47 and the use of death certificates recorded in SEER has been previously validated.^48,49 In addition, the SEER database does not allow differentiation between castration-resistant and castration-sensitive PC; thus, our analysis may include a heterogeneous sample of 2 groups with distinct prognoses. Furthermore, some CODs may have been underreported or reported under other CODs in the registry. Moreover, for older patients, there might be misclassification of COD because of the various comorbidities.⁵⁰ We analyzed the data for different demographic and tumor characteristics but did not include data on existing health conditions, treatment details, or ADT exposure because they are not available in the SEER database. Prospective studies assessing the association of strategies to reduce risk of cardiovascular disease, including a personalized ADT approach, with mortality among patients older than 50 years who have metastatic PC are needed to confirm the findings.

Conclusions

In this cohort study, a large number of deaths among patients with metastatic PC were attributable to noncancer causes, especially among patients older than 50 years at diagnosis. Furthermore, cardiac diseases, COPD, and cerebrovascular diseases were among the most common causes of noncancer deaths. These findings may provide insight into how men with metastatic PC should be counseled regarding future health risks and highlight the importance of multidisciplinary care for such patients.

Supplement.

eTable 1. Definition of each cause of death and corresponding codes in the ICD-10 of Diseases and Related Health

eTable 2. Standardized mortality ratios (SMRs) for each cause of death following metastatic prostate cancer diagnosis in patients younger than 50 years

eTable 3. Standardized mortality ratios (SMRs) for each cause of death following metastatic prostate cancer diagnosis in patients aged 50-70 years

eTable 4. Standardized mortality ratios (SMRs) for each cause of death following metastatic prostate cancer diagnosis in patients older than 70 years

eTable 5. Standardized mortality ratios (SMRs) for each cause of death following metastatic prostate cancer diagnosis in White patients

eTable 6. Standardized mortality ratios (SMRs) for each cause of death following metastatic prostate cancer diagnosis in Black patients

eTable 7. Standardized mortality ratios (SMRs) for each cause of death following metastatic prostate cancer diagnosis in Asian or Pacific Islander patients

eTable 8. Standardized mortality ratios (SMRs) for each cause of death following metastatic prostate cancer diagnosis in Indian American or Alaska Native patients

eTable 9. Standardized mortality ratios (SMRs) for each cause of death following stage M1a prostate cancer diagnosis

eTable 10. Standardized mortality ratios (SMRs) for each cause of death following stage M1b prostate cancer diagnosis

eTable 11. Standardized mortality ratios (SMRs) for each cause of death following stage M1c prostate cancer diagnosis

eTable 12. Standardized mortality ratios (SMRs) for each cause of death following metastatic prostate cancer diagnosis in patients who underwent cancer-directed surgery

eTable 13. Standardized mortality ratios (SMRs) for each cause of death following metastatic prostate cancer diagnosis in patients who underwent radiotherapy

eTable 14. Standardized mortality ratios (SMRs) for each cause of death following metastatic prostate cancer diagnosis in patients who underwent chemotherapy

Click here for additional data file.^{(613.4KB, pdf)}

References

1.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin. 2020;70(1):7-30. doi: 10.3322/caac.21590 [DOI] [PubMed] [Google Scholar]
2.Culp MB, Soerjomataram I, Efstathiou JA, Bray F, Jemal A. Recent global patterns in prostate cancer incidence and mortality rates. Eur Urol. 2020;77(1):38-52. doi: 10.1016/j.eururo.2019.08.005 [DOI] [PubMed] [Google Scholar]
3.Li J, Berkowitz Z, Hall IJ. Decrease in prostate cancer testing following the US Preventive Services Task Force (USPSTF) recommendations. J Am Board Fam Med. 2015;28(4):491-493. doi: 10.3122/jabfm.2015.04.150062 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Weiner AB, Matulewicz RS, Eggener SE, Schaeffer EM. Increasing incidence of metastatic prostate cancer in the United States (2004-2013). Prostate Cancer Prostatic Dis. 2016;19(4):395-397. doi: 10.1038/pcan.2016.30 [DOI] [PubMed] [Google Scholar]
5.Kelly SP, Anderson WF, Rosenberg PS, Cook MB. Past, current, and future incidence rates and burden of metastatic prostate cancer in the United States. Eur Urol Focus. 2018;4(1):121-127. doi: 10.1016/j.euf.2017.10.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Howlader N, Noone AM, Krapcho M, et al, eds. SEER Cancer Statistics Review (CSR) 1975-2016. National Cancer Institute. Updated April 9, 2020. Accessed January 15, 2020. https://seer.cancer.gov/csr/1975_2016/
7.Parimi S, Bondy S, Aparicio M, et al. Presenting stage and risk group in men dying of prostate cancer. Curr Oncol. 2020;27(6):e547-e551. doi: 10.3747/co.27.6385 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.de Bono JS, Logothetis CJ, Molina A, et al. ; COU-AA-301 Investigators . Abiraterone and increased survival in metastatic prostate cancer. N Engl J Med. 2011;364(21):1995-2005. doi: 10.1056/NEJMoa1014618 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Fizazi K, Tran N, Fein L, et al. Abiraterone acetate plus prednisone in patients with newly diagnosed high-risk metastatic castration-sensitive prostate cancer (LATITUDE): final overall survival analysis of a randomised, double-blind, phase 3 trial. Lancet Oncol. 2019;20(5):686-700. doi: 10.1016/S1470-2045(19)30082-8 [DOI] [PubMed] [Google Scholar]
10.Gravis G, Fizazi K, Joly F, et al. Androgen-deprivation therapy alone or with docetaxel in non-castrate metastatic prostate cancer (GETUG-AFU 15): a randomised, open-label, phase 3 trial. Lancet Oncol. 2013;14(2):149-158. doi: 10.1016/S1470-2045(12)70560-0 [DOI] [PubMed] [Google Scholar]
11.Sweeney CJ, Chen Y-H, Carducci M, et al. Chemohormonal therapy in metastatic hormone-sensitive prostate cancer. N Engl J Med. 2015;373(8):737-746. doi: 10.1056/NEJMoa1503747 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Kyriakopoulos CE, Chen Y-H, Carducci MA, et al. Chemohormonal therapy in metastatic hormone-sensitive prostate cancer: long-term survival analysis of the randomized phase III E3805 CHAARTED trial. J Clin Oncol. 2018;36(11):1080-1087. doi: 10.1200/JCO.2017.75.3657 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Helgstrand JT, Røder MA, Klemann N, et al. Trends in incidence and 5-year mortality in men with newly diagnosed, metastatic prostate cancer—a population-based analysis of 2 national cohorts. Cancer. 2018;124(14):2931-2938. doi: 10.1002/cncr.31384 [DOI] [PubMed] [Google Scholar]
14.Siegel DA, O’Neil ME, Richards TB, Dowling NF, Weir HK. Prostate cancer incidence and survival, by stage and race/ethnicity—United States, 2001-2017. MMWR Morb Mortal Wkly Rep. 2020;69(41):1473-1480. doi: 10.15585/mmwr.mm6941a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Fowler JEJ Jr, Terrell FL, Renfroe DL. Co-morbidities and survival of men with localized prostate cancer treated with surgery or radiation therapy. J Urol. 1996;156(5):1714-1718. doi: 10.1016/S0022-5347(01)65489-2 [DOI] [PubMed] [Google Scholar]
16.Satariano WA, Ragland KE, Van Den Eeden SK. Cause of death in men diagnosed with prostate carcinoma. Cancer. 1998;83(6):1180-1188. doi: [DOI] [PubMed] [Google Scholar]
17.Lu-Yao G, Stukel TA, Yao S-L. Changing patterns in competing causes of death in men with prostate cancer: a population based study. J Urol. 2004;171(6 Pt 1):2285-2290. doi: 10.1097/01.ju.0000127740.96006.1a [DOI] [PubMed] [Google Scholar]
18.Newschaffer CJ, Otani K, McDonald MK, Penberthy LT. Causes of death in elderly prostate cancer patients and in a comparison nonprostate cancer cohort. J Natl Cancer Inst. 2000;92(8):613-621. doi: 10.1093/jnci/92.8.613 [DOI] [PubMed] [Google Scholar]
19.Ketchandji M, Kuo YF, Shahinian VB, Goodwin JS. Cause of death in older men after the diagnosis of prostate cancer. J Am Geriatr Soc. 2009;57(1):24-30. doi: 10.1111/j.1532-5415.2008.02091.x [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Albertsen PC, Hanley JA, Gleason DF, Barry MJ. Competing risk analysis of men aged 55 to 74 years at diagnosis managed conservatively for clinically localized prostate cancer. JAMA. 1998;280(11):975-980. doi: 10.1001/jama.280.11.975 [DOI] [PubMed] [Google Scholar]
21.Van Hemelrijck M, Folkvaljon Y, Adolfsson J, et al. Causes of death in men with localized prostate cancer: a nationwide, population-based study. BJU Int. 2016;117(3):507-514. doi: 10.1111/bju.13059 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Park J, Suh B, Shin DW, Hong JH, Ahn H. Cause of death in Korean men with prostate cancer: an analysis of time trends in a nationwide cohort. J Korean Med Sci. 2016;31(11):1802-1807. doi: 10.3346/jkms.2016.31.11.1802 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Nabid A, Carrier N, Vigneault E, et al. Causes of death in intermediate- and high-risk prostate cancer treated with radiotherapy with or without androgen deprivation therapy: analysis from two phase III trials. J Clin Oncol. 2016;34(2 suppl):34. doi: 10.1200/jco.2016.32.2_suppl.34 [DOI] [Google Scholar]
24.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative . The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. Int J Surg. 2014;12(12):1495-1499. doi: 10.1016/j.ijsu.2014.07.013 [DOI] [PubMed] [Google Scholar]
25.SEER*Stat. Version 8.3.5. National Cancer Institute Surveillance, Epidemiology, and End Results Program; 2018. Accessed February 2, 2020. https://www.seer.cancer.gov/seerstat
26.National Cancer Institute Surveillance, Epidemiology, and End Results (SEER) Program. SEER*Stat Database: Incidence—SEER 18 regs excluding AK custom data (with additional treatment fields), Nov 2018 sub (2000-2016) for SMRs—linked to county attributes—total U.S., 1969-2017 counties, National Cancer Institute, DCCPS, Surveillance Research Program. Released April 2019, based on the November 2018 submission. Accessed February 2, 2020. http://www.seer.cancer.gov
27.National Cancer Institute Surveillance, Epidemiology, and End Results (SEER) Program. SEER Cause of Death Recode 1969+ (03/01/2018)—SEER Data Reporting Tools. Accessed February 2, 2020. https://seer.cancer.gov/codrecode/1969_d03012018/index.html
28.Gad MM, Saad AM, Al-Husseini MJ, et al. Temporal trends, ethnic determinants, and short-term and long-term risk of cardiac death in cancer patients: a cohort study. Cardiovasc Pathol. 2019;43:107147. doi: 10.1016/j.carpath.2019.08.001 [DOI] [PubMed] [Google Scholar]
29.Haque R, UlcickasYood M, Xu X, et al. Cardiovascular disease risk and androgen deprivation therapy in patients with localised prostate cancer: a prospective cohort study. Br J Cancer. 2017;117(8):1233-1240. doi: 10.1038/bjc.2017.280 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Hak AE, Witteman JC, de Jong FH, Geerlings MI, Hofman A, Pols HA. Low levels of endogenous androgens increase the risk of atherosclerosis in elderly men: the Rotterdam study. J Clin Endocrinol Metab. 2002;87(8):3632-3639. doi: 10.1210/jcem.87.8.8762 [DOI] [PubMed] [Google Scholar]
31.Jones RD, Nettleship JE, Kapoor D, Jones HT, Channer KS. Testosterone and atherosclerosis in aging men: purported association and clinical implications. Am J Cardiovasc Drugs. 2005;5(3):141-154. doi: 10.2165/00129784-200505030-00001 [DOI] [PubMed] [Google Scholar]
32.Khaw KT, Dowsett M, Folkerd E, et al. Endogenous testosterone and mortality due to all causes, cardiovascular disease, and cancer in men: European Prospective Investigation into Cancer in Norfolk (EPIC-Norfolk) prospective population study. Circulation. 2007;116(23):2694-2701. doi: 10.1161/CIRCULATIONAHA.107.719005 [DOI] [PubMed] [Google Scholar]
33.Keating NL, O’Malley AJ, Freedland SJ, Smith MR. Does comorbidity influence the risk of myocardial infarction or diabetes during androgen-deprivation therapy for prostate cancer? Eur Urol. 2013;64(1):159-166. doi: 10.1016/j.eururo.2012.04.035 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Meng F, Zhu S, Zhao J, et al. Stroke related to androgen deprivation therapy for prostate cancer: a meta-analysis and systematic review. BMC Cancer. 2016;16:180. doi: 10.1186/s12885-016-2221-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Azoulay L, Yin H, Benayoun S, Renoux C, Boivin JF, Suissa S. Androgen-deprivation therapy and the risk of stroke in patients with prostate cancer. Eur Urol. 2011;60(6):1244-1250. doi: 10.1016/j.eururo.2011.08.041 [DOI] [PubMed] [Google Scholar]
36.Van Hemelrijck M, Garmo H, Holmberg L, et al. Absolute and relative risk of cardiovascular disease in men with prostate cancer: results from the population-based PCBaSe Sweden. J Clin Oncol. 2010;28(21):3448-3456. doi: 10.1200/JCO.2010.29.1567 [DOI] [PubMed] [Google Scholar]
37.Whitsel EA, Boyko EJ, Matsumoto AM, Anawalt BD, Siscovick DS. Intramuscular testosterone esters and plasma lipids in hypogonadal men: a meta-analysis. Am J Med. 2001;111(4):261-269. doi: 10.1016/S0002-9343(01)00833-6 [DOI] [PubMed] [Google Scholar]
38.Traish AM, Saad F, Feeley RJ, Guay A. The dark side of testosterone deficiency: III. Cardiovascular disease. J Androl. 2009;30(5):477-494. doi: 10.2164/jandrol.108.007245 [DOI] [PubMed] [Google Scholar]
39.Laughlin GA, Barrett-Connor E, Bergstrom J. Low serum testosterone and mortality in older men. J Clin Endocrinol Metab. 2008;93(1):68-75. doi: 10.1210/jc.2007-1792 [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Isidori AM, Giannetta E, Greco EA, et al. Effects of testosterone on body composition, bone metabolism and serum lipid profile in middle-aged men: a meta-analysis. Clin Endocrinol (Oxf). 2005;63(3):280-293. doi: 10.1111/j.1365-2265.2005.02339.x [DOI] [PubMed] [Google Scholar]
41.Alibhai SM, Duong-Hua M, Sutradhar R, et al. Impact of androgen deprivation therapy on cardiovascular disease and diabetes. J Clin Oncol. 2009;27(21):3452-3458. doi: 10.1200/JCO.2008.20.0923 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Van Hemelrijck M, Feller A, Garmo H, et al. Incidence of second malignancies for prostate cancer. PLoS One. 2014;9(7):e102596. doi: 10.1371/journal.pone.0102596 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Pilié PG, Johnson AM, Hanson KL, et al. Germline genetic variants in men with prostate cancer and one or more additional cancers. Cancer. 2017;123(20):3925-3932. doi: 10.1002/cncr.30817 [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Park HS, Lloyd S, Decker RH, Wilson LD, Yu JB. Limitations and biases of the Surveillance, Epidemiology, and End Results database. Curr Probl Cancer. 2012;36(4):216-224. doi: 10.1016/j.currproblcancer.2012.03.011 [DOI] [PubMed] [Google Scholar]
45.Sun M, Trinh QDA. A Surveillance, Epidemiology and End Results (SEER) database malfunction: perceptions, pitfalls and verities. BJU Int. 2016;117(4):551-552. doi: 10.1111/bju.13226 [DOI] [PubMed] [Google Scholar]
46.Yang K, Zheng Y, Peng J, et al. Incidence of death from unintentional injury among patients with cancer in the United States. JAMA Netw Open. 2020;3(2):e1921647-e1921647. doi: 10.1001/jamanetworkopen.2019.21647 [DOI] [PMC free article] [PubMed] [Google Scholar]
47.German RR, Fink AK, Heron M, et al. ; Accuracy of Cancer Mortality Study Group . The accuracy of cancer mortality statistics based on death certificates in the United States. Cancer Epidemiol. 2011;35(2):126-131. doi: 10.1016/j.canep.2010.09.005 [DOI] [PubMed] [Google Scholar]
48.Lund JL, Harlan LC, Yabroff KR, Warren JL. Should cause of death from the death certificate be used to examine cancer-specific survival? a study of patients with distant stage disease. Cancer Invest. 2010;28(7):758-764. doi: 10.3109/07357901003630959 [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Hu CY, Xing Y, Cormier JN, Chang GJ. Assessing the utility of cancer-registry-processed cause of death in calculating cancer-specific survival. Cancer. 2013;119(10):1900-1907. doi: 10.1002/cncr.27968 [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Mieno MN, Tanaka N, Arai T, et al. Accuracy of death certificates and assessment of factors for misclassification of underlying cause of death. J Epidemiol. 2016;26(4):191-198. doi: 10.2188/jea.JE20150010 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials