Association between Marital Status and Prognosis in Patients with Prostate Cancer: A Meta-Analysis of Observational Studies Zhenlang Guo1, Chiming Gu1, Siyi Li1, Shu Gan1, Yuan Li1, Songtao Xiang1, Leiliang Gong2, Shusheng Wang1* Purpose: The impact of marital status on the prognosis amongst patients diagnosed with prostate cancer remains controversial. Thus, a meta-analysis was performed to determine whether marital status can influence the progno- sis in patients with prostate cancer. Materials and Methods: Literature search of the MEDLINE, PsycINFO, Embase and Cochrane Library databases was conducted to identify eligible studies published before April 2020. Multivariate adjusted risk estimates and corresponding 95% confidence intervals (CIs) were extracted and calculated using the random effects model. Results: A total of 11 observational studies comprising 1,457,799 patients diagnosed with prostate cancer were identified. Results indicated that unmarried status (separated, divorced, widowed or never married) was associated with an increased risk of all-cause mortality (hazard ratio, HR = 1.39, 95% CI: 1.30–1.50; P < .001; I2 = 92.2%) compared with married status, especially for divorced and never-married patients. Similarly, being unmarried had an elevated risk of cancer-specific mortality (HR = 1.29, 95% CI: 1.17–1.41; P < .001; I2 = 82.5%) in patients with prostate cancer. A significant difference was also observed between unmarried status and shorter overall survival (HR = 1.37, 95% CI: 1.20–1.56; P < .001; I2 = 94.5%). Conclusion: Results demonstrated that unmarried status is associated with a worse prognosis regarding mortality and survival in patients diagnosed with prostate cancer, particularly in divorced and never-married patients. Hence, further research should explore the potential mechanisms which can benefit the development of novel, more per- sonalized management methods for unmarried patients with prostate cancer. Keywords: marital status; prostate cancer; prognosis; meta-analysis INTRODUCTION Prostate cancer has become a major health problem and a leading cause of morbidity and mortality in men worldwide.(1,2) In the United States in 2018, pa- tients newly diagnosed with prostate cancer reached 164,690, while 26,730 patients with prostate cancer died despite the high overall survival (OS) for the dis- ease.(3) However, only a few risk factors for prostate cancer have been identified, including age, family his- tory, race, and certain genetic polymorphisms, thereby limiting the prevention of prostate cancer.(4,5) Interest- ingly, some research demonstrated that being unmar- ried (never married, separated, widowed or divorced) is associated with shorter survival and higher mortality for several malignancies compared with married status.(6–10) Nonetheless, the impact of marital status on the progno- sis amongst patients diagnosed with prostate cancer is still inconclusive. Several studies revealed that unmarried men have consistently been found to be associated with worse prognosis in patients with prostate cancer,(11,12) where- as other studies reported conflicting results.(13,14) Spe- 1Department of Urology, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, China. 2Department of mechanical engineering, National University of Singapore, Kent Ridge, Singapore. *Correspondence: Department of Urology, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine , Guangzhou, 510120, P.R.China. Tel: +86 13512704335. E-mail: shushengwanggzy@163.com. Received April 2020 & Accepted November 2020 cifically, Nepple et al.(13) reported that never-married men diagnosed with prostate cancer have an elevated risk of all-cause mortality (ACM) compared with those who were married, but this significant association was not observed amongst divorced or widowed men with prostate cancer. In light of these different findings re- ported in previous literature, we performed a systematic review and meta-analysis to explore how married sta- tus influences the prognosis (survival, mortality, etc.) in patients with any form of diagnosed prostate cancer. MATERIAL AND METHODS This systematic review and meta-analysis was per- formed in accordance with the Cochrane Collaboration criterion.(15) Our study adhered to the Preferred Report- ing Items for Systematic Reviews and Meta-Analyses statement guidelines(16) and Meta-analysis of Observa- tional Studies in Epidemiology (MOOSE) guidelines. (17) Search strategy Literature search of the MEDLINE (via PubMed), Psy- Urology Journal/Vol 18 No. 4/ July-August 2021/ pp. 371-379. [DOI: 10.22037/uj.v16i7.6197] REVIEW cINFO, Embase and Cochrane Library databases was conducted to identify eligible studies from database in- ception up to April 2020 using a combination of Med- ical Subject Headings (MeSH) and non-MeSH terms, including ‘marital status’, ‘marriage’, ‘married’, ‘un- married’, ‘divorced’ or ‘widowed’ and ‘prostate can- cer’, ‘prostate carcinoma’, ‘prostate tumor’ or ‘prostate neoplasms’ and ‘prognosis’, ‘survival’ or ‘mortality’, with no language, region or publication status restric- tion. Furthermore, important citation database such as Scopus was searched, and manual searches of reference lists were also performed in relevant original and review articles for additional eligible studies. The main search was carried out independently by the senior investigator (ZL. Guo). Disagreement was resolved by consulting another investigator (SS. Wang) who was not involved in the initial search procedure. Eligibility criteria All published eligible studies were included if they met the following inclusion criteria; (1) men diagnosed with any stage of prostate cancer only; (2) married sta- Marital status and prognosis in prostate cancer-Guo et al. Figure 1. Flow diagram of literature searches according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement. Figure 2. Marital status and all-cause mortality in patients with prostate cancer. Abbreviations: CI, confidence interval; HR, Hazard Ratio Vol 18 No 4 July-August 2021 372 tus is defined as married and/or living with a partner or family, while unmarried status is defined as widowed, divorced or living alone; (3) original trials regarding the impact of marital status on prognosis (survival or mortality) in patients with any form of diagnosed prostate cancer; (4) studies using an observational study design (i.e. prospective or retrospective cohort, cross-sectional or case-control study); and (5) studies reporting sufficient data on risk estimates (hazard ratio, HR; odds ratio, OR; relative risk, RR) with associated 95% confidence intervals (CIs) or sufficient raw data for calculation. For studies from the same population, only the largest studies with the longest follow-up pe- riod were retained. In addition, certain articles, such as case series, case reports, and review articles were ex- cluded. Disagreement was resolved through discussion amongst the investigators. Data extraction and methodological quality assessment Two investigators (ZL. Guo and CM. Gu) independent- ly extracted data from the eligible studies by using an predefined data extraction form. The following data were extracted: first author, study design, country, ba- Review 373 Figure 3. Marital status and cancer-specific mortality in patients with prostate cancer. Abbreviations: CI, confidence interval; HR, Hazard Ratio Figure 4. Marital status and overall survival in patients with prostate cancer. Abbreviations: CI, confidence interval; HR, Hazard Ratio Marital status and prognosis in prostate cancer-Guo et al. sic characteristics (i.e. sample size, age, and follow-up), definitive therapy (i.e. radical prostatectomy, androgen deprivation therapy, and radiation therapy), marital sta- tus, adjusted confounders, and risk estimates (HR, OR or RR) with associated 95% CIs or sufficient raw data. If the information reported in the eligible studies were insufficient, we contacted the primary authors to obtain and verify the data. First author Study Country Total participants Age, y, Follow-up, y, Treatments Marital Outcomes Adjustments year design (married, %) Mean (Range) Mean (Range) status Abdollah F Retro- USA 163,697 (83.1%) 63 (35-90) NA RP Married CSM Age, race, 2011. (21) spective SDW ACM socioeconomic status, population- Never married tumor grade, based cohort and year of surgery Aizer AA Retro- USA 190,648 (76.7%) 63 ± (SD: 12) 3.1 ± (0.1-5.9) Mitox- Married CSM Demographic factors 2013. (22) spective antrone Unmarried (age, race, income, population- education, and urban based cohort or rural residence), tumor stage, nodal stage, and whether definitive treatment administered Du KL 2012. (23) Prospective USA 3,570 (76%) 69.2 (41-48) NA Radiation Married OS Age, clinical stage, cohort therapy SDW Karnofsky Performance Score (KPS), Gleason Score, Prostate Specific Antigen (PSA), Biologic Effective Dose (BED), and type of treatment received Gomez SL 2016. (11) Retro- USA 178,586 (75.4%) NA NA NA Married ACM Cancer site, spective Unmarried race/ethnicity, population- and treatment based cohort Huang TB 2018. (24) Retro- USA 95,846 (81.6%) NA 6.5 ± (SD: 1.9) RP Married OS Age, ethnicity, spective SDW CSS grade, stage, population- Gleason scores, based cohort and sequence number Khan S 2019. (12) Retro- USA 3,579 (86.8%) 60.4 NA RP Married ACM Age, race, comorbidity spective Unmarried status, PSA, and cohort CSM biopsy Gleason grade Knipper S 2019. (25) Retro- USA 433,197 (75.4%) 65.4 (59-71) NA RP External Married CSM Prostatic-specific spective beam Brachy Unmarried antigen value, population- therapy age at diagnosis, based cohort year of diagnosis, treatment, clinical tumor stage, and race in all groups Lai H 1999. (26) Retro- USA 261,070 (70%) 65.4 ± (SD: 13.6) NA NA Married SDW Age, race, spective OS and treatment population- based cohort Nepple KG Retro- USA 3,596 (86.9%) NA NA RP Married CSM PSA, clinical stage, 2012. (13) spective Divorced ACM and biopsy Gleason cohort Widowed grade, comorbidity, Never married ethnicity, age, and marital status at time of treatment Schiffmann Retro- Germany 8,088 (91.1%) 63.5 (35.8–79.8) 4 ± (3.1) RP Married ACM PSA, biopsy Gleason J 2015. (14) spective Unmarried score, number of biopsy cohort cores taken, number of positive biopsy cores, clinical tumor stage Tyson MD Retro- USA 115,922 (78%) NA NA NA Married CSM Age, AJCC stage, 2013. (27) spective Single ACM tumor grade, and race population- Divorced based cohort Widowed Separated Table 1. Characteristics of the included studies. Abbreviations: ACM, all-cause mortality; AJCC, American Joint Committee on Cancer; CSM, cancer-specific mortality; CSS, can- cer-specific survival; NA, not applicable; OS, overall survival; SD, standard deviation; SDW, separated/divorced/widowed; RP, Radical prostatectomy; y, year. Marital status and prognosis in prostate cancer-Guo et al. Vol 18 No 4 July-August 2021 374 The methodological quality and risk of bias assess- ment were performed by two investigators according to the Newcastle–Ottawa scale (NOS),(18) which con- sists of nine items that evaluate the representativeness of included studies. Each item was assessed as either ‘unclear’, ‘yes’ or ‘no’, which corresponded to ‘0’, ‘1’ or ‘0’ in accordance with the information reported by the studies. The total score ranged from 0 to 9, and a study was categorised as follows: a score of 8–9 was considered high quality, a score of 6–7 was considered moderate quality, and a score of ≤ 5 was considered low quality. Any disagreements were settled through a discussion amongst the authors. Statistical analyses In general, the total risk estimates (HR and RR with associated 95% CIs) extracted from the included studies were calculated via Stata version 15.0 (serial number: 10699393; StataCorp Wyb). HR and RR with associ- ated 95% CIs were calculated through inverse variance using random or fixed effects models. For consistent definitions, HR with associated 95% CIs was used as a common measure because marital status and prostate cancer-related survival or mortality were considered as rare events. The differences amongst the various measures of risk estimates could be generally ignored. Hence, the RRs extracted from the included studies could be considered approximations of HRs.(19) I2 was used to assess heterogeneity across studies, with I2 val- ues of 0%, 25%, 50% and 75% representing no, low, moderate and high heterogeneity, respectively.(15) Spe- cifically, a severe heterogeneity of I2 ≥ 50% warrants the use of random effects models. Otherwise, a fixed effects model should be used.(15) Statistical significance was set at P < .05. Moreover, weight estimation was conducted in the meta-analysis according to the valid- ity or risk of bias for included studies. Note that if all the weights are the same then the weighted average is equal to the mean intervention effect. The bigger the weight given to the study, the more it will contribute to the weighted average. The weights are therefore chosen to reflect the amount of information that each study contains. In the presence of heterogeneity, a ran- dom-effects meta-analysis weights the studies relative- ly more equally than a fixed-effect analysis. Subgroup analysis based on different treatments was performed to explore the possible origins of heterogeneity. Sensi- tivity analysis could assess the quality and consistency of the results through omitting each study individually. In addition, meta-regression analysis was conducted to explore the possible sources of heterogeneity in several variables, and restricted maximum likelihood was used in the analysis. However, the application of Egger(20) and Begg–Mazumdar(21) tests was limited because of the low number of studies evaluated. RESULTS Study identification and selection The search process and study selection are described in Figure 1. In general, 569 articles were identified through the initial assessment, and 459 articles were retrieved after duplicates were removed. Next, 420 ar- ticles were removed after title/abstract evaluation from the remaining articles. Finally, 39 articles were evaluat- ed on the basis of the full text, and 28 were excluded for the following reasons: no prostate cancer (6 articles), no marital status (4 articles), no prognostic assessment (14 articles), and not sufficient data for extraction (4 articles). Therefore, 11 articles(11–14, 22–28)comprising 1,457,799 patients diagnosed with prostate cancer were identified for systematic review and meta-analysis ac- cording to the eligibility criteria. Study characteristics and methodological quality The basic characteristics of studies included in this sys- tematic review and meta-analysis are described in Table 1. These studies (1 prospective cohort(24) and 10 retro- spective cohort studies(11–14, 22, 23, 25–28)) were published between 1999 and 2019. There were 10 studies(11–13, 22–28) from the United States and 1 from Germany(14). The sample sizes also varied between 3,570 and 433,197 pa- tients with prostate cancer who were treated with vari- ous definitive therapies, including radical prostatectomy (RP), radiation therapy, external beam, brachytherapy, and mitoxantrone. The follow-up duration ranged from 3.1 years to 6.5 years. All the included studies reported risk estimates adjusted for confounding factors. Overall, the methodological quality and risk bias assess- ment of the included studies(11–14, 22–28) was performed according to the NOS. Five articles(14, 22–24, 26) acquired 8 or 9 points and were considered as high quality, eight articles(11–13, 25, 28) acquired 6 or 7 points and were con- Review 375 Table 2. Results of subgroup analyses. Overall ACM risk Studies, N Participants,N HR (95% CI) p value p of heterogeneity I2 (%) 6 910,639 1.39 (1.30–1.50) < 0.001 < 0.001 92.2 Different definite therapies RP 4 178,960 1.37 (1.25–1.51) < 0.001 0.006 67.2 Other therapies 2 178,586 1.42 (1.25–1.60) < 0.001 < 0.001 98.8 Overall CSM risk Studies, N Participants,N HR (95% CI) p value p of heterogeneity I2 (%) 6 473,468 1.29 (1.17–1.41) < 0.001 < 0.001 82.5 Different definite therapies RP 3 170,872 1.34 (1.05–1.71) 0.017 0.002 73.3 Mitoxantrone 1 190,648 1.35 (1.23–1.49) < 0.001 NA NA RP External beam Brachy therapy 1 433,197 1.19 (1.15–1.24) < 0.001 NA NA Other therapies 1 115,922 1.40 (1.33–1.47) < 0.001 NA NA Abbreviations: ACM, all-cause mortality; CI, confidence interval; CSM, cancer-specific mortality; HR, Hazard Ratio; NA, not applica- ble; RP, Radical prostatectomy. Marital status and prognosis in prostate cancer-Guo et al. sidered as moderate quality and one study(27) scored 5 points and was considered as low quality. Marital status and mortality in patients with prostate cancer Six studies(12,13,22,23,26,28) comprising 910,639 patients di- agnosed with prostate cancer used the random effects model (Figure 2) and demonstrated that unmarried status (separated, divorced, widowed or never married) was associated with an increased risk of ACM (HR = 1.39, 95% CI: 1.30–1.50; P < .001; I2 = 92.2%) com- pared with married status. Specifically, both divorced and never-married men had an excess risk of ACM (di- vorced men, HR = 1.45, 95% CI: 1.39–1.51; P < .001; never-married men, HR = 1.46, 95% CI: 1.05–2.03; P < .001), whereas this significant association was not observed amongst widowed patients (HR = 1.34, 95% CI: 0.83–2.16; P = .23) owing to the limited number of studies included. In the subgroup analyses stratified ac- cording to different definitive therapies, unmarried men who were treated with RP was associated with a higher risk of ACM (HR = 1.37, 95% CI: 1.25–1.51; P < .001) compared with other treatments (Table 2). In six studies(11–14,22,28) comprising 473,468 patients di- agnosed with prostate cancer, unmarried men had an elevated risk of cancer-specific mortality (CSM) (HR = 1.29, 95% CI: 1.17–1.41; P < .001; I2 = 82.5%) than those married men using a random effects model (Fig- ure 3). However, this significant association was not observed in never-married (HR = 1.03, 95% CI: 0.91– 1.17; P = .645) or widowed men (HR = 1.13, 95% CI: 0.15–8.36; P = .905), except for divorced patients (HR = 1.32, 95% CI: 1.22–1.43; P < .001) with prostate can- cer. When stratified by different treatments, the results were significant and consistent (Table 2). The results of meta-regression analyses regarding the heterogeneity amongst studies for ACM and CSM re- vealed that definitive therapy (ACM, P = .711; CSM, P = .798) could not result in heterogeneity amongst the included studies. Therefore, the other important confounding factors such as age should be fully ex- plored in future relevant studies. Moreover, the adjust- ed R-squared values from -11.64% to -8.97% because the regression line is worse than using a horizontal line, which indicated that the regressors slightly contributed to the explanation of the response variables (Table 3). When any study was omitted in turn, the stability of the results by sensitivity analysis did not show any signifi- cant change for ACM and CSM (Table 4). Marital status and survival in patients with prostate cancer Three studies(24,25,27) comprising 360,486 patients diag- nosed with prostate cancer reported risk estimates of overall survival (OS) and marital status. A significant difference was observed between unmarried status and shorter OS (HR = 1.37, 95% CI: 1.20–1.56; P < .001; I2 = 94.5%) through a random effects model (Figure 4). However, meta-regression and subgroup analyses were limited because of the small number of studies included. Notably, the results of sensitivity analysis revealed that the stability of meta-analysis had no sig- nificant change after each study was omitted in turn. For cancer-specific survival (CSS), only one retrospec- tive population-based cohort reporting the risk of CSS and the impact of marital status was included. Huang et al.(25) found that divorced and never-married men were significantly associated with shorter CSS (HR = 1.61, 95% CI: 1.34–1.93; P < .001 and HR = 1.20, 95% CI: 1.00–1.40; P < .001, respectively). By contrast, this as- sociation was not observed amongst widowed men (HR = 1.13, 95% CI: 0.81–1.58; P > .05). DISCUSSION Main findings The systematic review and meta-analysis identified 11 studies comprising 1,457,799 patients diagnosed with prostate cancer regarding the association between mar- ital status and prognosis in prostate cancer. We found that higher mortality and shorter survival in patients diagnosed with prostate cancer are associated with un- married status, particularly in divorced and never-mar- ried patients, than in married men. However, this sig- nificant association does not seem to be validated in widowed populations because of the limited number of relevant studies evaluated. Note that the risk estimates Table 3. Results of meta-regression. Covariates Exp(b) Standard error t P > |t| 95% CI R-squared ACM Treatment 1.40693 1.242582 0.39 0.711 0.174297 11.35678 -8.97% CSM Treatment 0.7745458 0.7430051 -0.27 0.798 0.0801535 7.484653 -11.64% Abbreviations: ACM, all-cause mortality; CI, confidence interval; CSM, cancer-specific mortality. Study omitted HR 95% CI ACM Abdollah F 2011. [21] 1.42 1.31 1.54 Abdollah F 2011. [21] 1.38 1.26 1.50 Gomez SL 2016. [11] 1.41 1.30 1.53 Khan S 2019. [12] 1.39 1.29 1.50 Nepple KG 2012. [13] 1.38 1.28 1.48 Nepple KG 2012. [13] 1.40 1.30 1.50 Nepple KG 2012. [13] 1.39 1.29 1.50 Schiffmann J 2015. [14] 1.40 1.30 1.50 Tyson MD 2013. [27] 1.36 1.28 1.43 Combined 1.39 1.30 1.50 CSM Aizer AA 2013. [22] 1.27 1.14 1.42 Abdollah F 2011. [21] 1.34 1.22 1.47 Abdollah F 2011. [21] 1.28 1.14 1.44 Khan S 2019. [12] 1.28 1.16 1.40 Knipper S 2019. [25] 1.31 1.18 1.46 Nepple KG 2012. [13] 1.28 1.17 1.41 Nepple KG 2012. [13] 1.29 1.17 1.42 Nepple KG 2012. [13] 1.27 1.16 1.39 Tyson MD 2013. [27] 1.25 1.13 1.38 Combined 1.29 1.17 1.41 Abbreviations: ACM, all-cause mortality; CI, confidence inter- val; CSM, cancer-specific mortality; HR, Hazard Ratio. Table 4. Results of sensitivity analyses. Marital status and prognosis in prostate cancer-Guo et al. Vol 18 No 4 July-August 2021 376 extracted from all included articles were based on the adjustment of confounding factors. Finally, sensitivity analysis revealed that the stability of the results had no significant change after each study was omitted in turn, and the meta-regression could not identify the potential confounding factors that might affect the level of heter- ogeneity between studies. Most of the included studies revealed that unmarried men have consistently been found to be associated with worse prognosis in patients with prostate cancer, whereas few studies reported conflicting results.(14, 22) A retrospective cohort comprising 8,088 patients with prostate cancer conducted by Schiffmann et al.(14) in Germany failed to reveal a significant association be- tween unmarried men and ACM (HR = 0.80, 95% CI: 0.40–1.70; P = .6). Apart from that, Abdollah et al.(22) demonstrated that divorced patients were statistically associated with an increased risk of CSM (HR = 1.32, 95% CI: 1.20–1.40; P < .001) compared with married men. By contrast, this significant association was not observed in never-married patients (HR = 1.03, 95% CI: 0.91–1.17; P > .05). In the subgroup analyses strat- ified by different definitive therapies, unmarried men who have been treated with RP were associated with higher risks of ACM and CSM compared with those who underwent other treatments. However, of all in- cluded studies, three did not report definitive treatments for patients with prostate cancer.(11,27,28) Comparison with another previous study One systematic review that assessed a similar topic was published by Buja et al.(29). Several differences between Buja et al. and the current work should be noted. First- ly, the previous review included only 3 articles that in- volved 372,412 patients with prostate cancer and mar- ital status. By comparison, our meta-analysis involved 11 studies comprising 1,457,799 patients diagnosed with prostate cancer. With the added statistical power of 8 studies and at least 1,085,387 cases, our meta-anal- ysis, which was inconsistent with the results of Buja et al., was the latest and the most comprehensive review to date. Secondly, the association between marital status and prognosis (i.e. survival and mortality) was evalu- ated amongst patients with prostate cancer treated by any therapy in line with the predefined inclusion cri- teria. Sensitivity analysis and meta-regression failed to identify confounding factors that might affect the level of heterogeneity between studies, thereby reinforcing the main findings. Implications for clinical practice Some researchers recently suggested that the choice of treatment therapy for prostate cancer in married men may be different from that of unmarried men. For mar- ried men, their spouse can encourage them to choose a treatment, such as RP or radiation therapy.(30,31) The association between marriage and treatment for prostate cancer has been confirmed and positively correlated. However, researchers did not evaluate survival or mor- tality as an endpoint.(30) Overall, unmarried status rep- resents an important determinant of a worse prognosis. Note that the detrimental influence of unmarried status remains consistent when stratified by different thera- pies. Hence, the negative impact of unmarried status on mortality or survival may still be explained through the later diagnosis or treatments in these unmarried patients. Lifestyle choice is also an important risk factor for the prognosis between unmarried and married men. For in- Review 377 stance, unmarried men are more likely to abuse tobacco or alcohol than do married men.(33) Similarly, married men are more likely to avoid unhealthy habits post-di- agnosis because of their responsibility to their spouse and family.(33) Therefore, the possible mechanisms by which marriage may potentially influence the prognosis in men with prostate cancer can be shown as follows.(1) Patients with prostate cancer may receive more psycho- logical support from their spouse or the society after di- agnosis, which can improve their likelihood of survival. (2) The spouse may affect the postoperative compliance (such as follow-up) and reception of adjuvant or sec- ondary treatment, such as androgen deprivation therapy and adjuvant or salvage radiation therapy.(3) The lack of physical activity is more common in men with insuffi- cient emotional support, which may be associated with higher mortality(34–39). Nevertheless, our understanding on the association between marital status and different stages of prostate cancer remains unclear because of the lack of studies that examine such a relationship. Thus, further research is warranted to investigate the person- alised intervention and management methods for un- married patients with prostate cancer. This systematic review and meta-analysis demonstrat- ed several crucial strengths in multiple ways. Firstly, our study comprehensively investigated the association between marital status and prognosis amongst patients with prostate cancer, and subgroup analyses stratified according to definitive therapy were conducted to deter- mine whether this variable moderated such an associa- tion and the level of heterogeneity between the studies. Secondly, multivariate-adjusted risk estimates were ap- plied to minimise other relevant risk factors that might affect the overall results. Finally, sensitivity analysis and meta-regression validated the rationality and relia- bility of the results for our study. Several limitations of this study should be noted. First- ly, most of the studies used retrospective cohort design, which has the disadvantages of missing data and risk of bias. Secondly, the number of articles included in this study was limited, especially in the subgroup, which may lead to unreliable results and might not reflect the comprehensiveness of the overall results. Lastly, sig- nificant heterogeneity was observed and the random effects model was applied in the pooled analysis. How- ever, subgroup analysis and meta-regression analyses failed to explore the potential factors leading to signif- icant heterogeneity. Therefore, other important factors should be adequately studied in further high-quality re- searches regarding this topic. CONCLUSIONS Existing evidence indicates that unmarried status is associated with a worse prognosis regarding mortality and survival in patients diagnosed with prostate cancer, particularly in divorced and never-married patients. Hence, further research should explore the potential mechanisms which can benefit the development of nov- el, more personalised management methods for unmar- ried patients with prostate cancer, who was considered as representatives of high-risk groups. ACKNOWLEDGEMENTS This study was supported by grants from the Science and Technology Research Project of Guangdong Provincial Hospital of Chinese Medicine (No. YN2019ML05). Marital status and prognosis in prostate cancer-Guo et al. Vol 18 No 4 July-August 2021 378 REFERENCES 1. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer. 2010;127:2893–917. 2. Buja A, Lange JH, Perissinotto E, et al. Cancer incidence among male military and civil pilots and flight attendants: an analysis on published data. Toxicol Ind Health. 2005;21:273–82. 3. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. CA Cancer J Clin. 2018;68:7– 30. 4. Eeles RA, Olama AA, Benlloch S, et al. Identification of 23 new prostate cancer susceptibility loci using the iCOGS custom genotyping array. Nat Genet. 2013;45:385– 91, 391e1–2. 5. Culp SH, Schellhammer PF, Williams MB. Might men diagnosed with metastatic prostate cancer benefit from definitive treatment of the primary tumor? A SEER-based study. Eur Urol. 2014;65:1058–66. 6. Chang SM, Barker FG. Marital status, treatment, and survival in patients with glioblastoma multiforme: a population based study. Cancer. 2005;104(9):1975–84. 7. Datta GD, Neville BA, Kawachi I, Datta NS, Earle CC. Marital status and survival following bladder cancer. J Epidemiol Community Health. 2009;63:807–13. 8. Hellenthal NJ, Chamie K, Ramirez ML, deVere White RW. Sociodemographic factors associated with nephrectomy in patients with metastatic renal cell carcinoma. J Urol. 2009;181:1013–8; discussion 1018–9. 9. Kravdal O. The impact of marital status on cancer survival. Soc Sci Med. 2001;52:357– 68. 10. Nelles JL, Joseph SA, Konety BR. The impact of marriage on bladder cancer mortality. Urol Oncol. 2009;27:263–7. 11. Gomez SL, Hurley S, Canchola AJ, et al. Effects of marital status and economic resources on survival after cancer: A population-based study. Cancer. 2016;122:1618–25. 12. Khan S, Nepple KG, Kibel AS, et al. The association of marital status and mortality among men with early-stage prostate cancer treated with radical prostatectomy: insight into post-prostatectomy survival strategies. Cancer Causes Control. 2019;30:871–6. 13. Nepple KG, Kibel AS, Sandhu GS, et al. Impact of marital status on prostate cancer- specific mortality and overall mortality after radical prostatectomy. Journal of Clinical Oncology. 2012;30. 14. Schiffmann J, Beyer B, Tennstedt P, et al. Oncological outcome after radical prostatectomy: Marital status does not make a difference. Int J Urol. 2015;22:484–9. 15. Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration. 2011. 16. Moher D, Shamseer L, Clarke M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev. 2015;4:1. 17. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA. 2000;283:2008–12. 18. Wells GA , Shea B, O’Connell D. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in meta- analysis. Ottawa Hospital Research Institute website. http:// www.ohri.ca/programs/ clinical_epidemiology/oxford.asp. 2014. 19. Greenland S. Quantitative methods in the review of epidemiologic literature. Epidemiol Rev. 1987;9:1–30. 20. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315:629– 34. 21. Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics. 1994;50:1088– 101. 22. Abdollah F, Sun M, Thuret R, et al. The effect of marital status on stage and survival of prostate cancer patients treated with radical prostatectomy: a population-based study. Cancer Causes Control. 2011;22:1085–95. 23. Aizer AA, Chen MH, McCarthy EP, et al. Marital status and survival in patients with cancer. J Clin Oncol. 2013;31:3869–76. 24. Du KL, Bae K, Movsas B, Yan Y, Bryan C, Bruner DW. Impact of marital status and race on outcomes of patients enrolled in Radiation Therapy Oncology Group prostate cancer trials. Support Care Cancer. 2012;20:1317– 25. 25. Huang TB, Zhou GC, Dong CP, et al. Marital status independently predicts prostate cancer survival in men who underwent radical prostatectomy: An analysis of 95,846 individuals. Oncol Lett. 2018;15:4737–44. 26. Knipper S, Preisser F, Mazzone E, et al. Contemporary analysis of the effect of marital status on survival of prostate cancer patients across all stages: A population-based study. Urol Oncol. 2019;37:702–10. 27. Lai H, Lai S, Krongrad A, Trapido E, Page JB, McCoy CB. The effect of marital status on survival in late-stage cancer patients: an analysis based on surveillance, epidemiology, and end results (SEER) data, in the United States. Int J Behav Med. 1999;6:150–76. 28. Tyson MD, Andrews PE, Etzioni DA, et al. Marital status and prostate cancer outcomes. Can J Urol. 2013;20:6702–6. 29. Buja A, Lago L, Lago S, Vinelli A, Zanardo C, Baldo V. Marital status and stage of cancer at diagnosis: A systematic review. Eur J Cancer Care (Engl). 2018;27:e12755. 30. Denberg TD, Beaty BL, Kim FJ, Steiner JF. Marriage and ethnicity predict treatment in localized prostate carcinoma. Cancer. 2005;103:1819–25. Marital status and prognosis in prostate cancer-Guo et al. 31. Denberg TD, Glodé LM, Steiner JF, et al. Trends and predictors of aggressive therapy for clinical locally advanced prostate carcinoma. BJU Int. 2006;98:335–40. 32. Rosengren A, Wedel H, Wilhelmsen L. Marital status and mortality in middle-aged Swedish men. Am J Epidemiol. 1989;129:54–64. 33. Umberson D. Gender, marital status and the social control of health behavior. Soc Sci Med. 1992;34:907–17. 34. Watt RG, Heilmann A, Sabbah W, et al. Social relationships and health related behaviors among older US adults. BMC Public Health. 2014;14:533. 35. O'Shaughnessy PK, Laws TA, Esterman AJ. The prostate cancer journey: results of an online survey of men and their partners. Cancer Nurs. 2015;38:E1–E12. 36. Bellardita L, Rancati T, Alvisi MF, et al. Predictors of health-related quality of life and adjustment to prostate cancer during active surveillance. Eur Urol. 2013;64:30–6. 37. Chamie K, Kwan L, Connor SE, Zavala M, Labo J, Litwin MS. The impact of social networks and partnership status on treatment choice in men with localized prostate cancer. BJU Int. 2012;109:1006–12. 38. Forsythe LP, Alfano CM, Kent EE, et al. Social support, self-efficacy for decision-making, and follow-up care use in long-term cancer survivors. Psycho-oncology. 2014;23:788–96. 39. Holt-Lunstad J, Smith TB, Layton JB. Social relationships and mortality risk: a meta- analytic review. PLoS Med. 2010;7:e1000316. Marital status and prognosis in prostate cancer-Guo et al. Review 379