Background
Prostate cancer (PCa) unevenly affects populations worldwide.1–3 Historically, Asia had a lower PCa incidence and mortality compared with Western countries, but this gap is narrowing. This is possibly due to shifts toward Western diets, increased screening, and an aging population.4 PCa incidence and mortality have rapidly increased across East Asia (China, Japan, and Korea), although age-standardized incidence rates remain lower compared with Western countries.5
Poor uptake of population-wide prostate-specific antigen (PSA) screening likely plays a role in the historically lower PCa incidence rate among East Asian men.6–8 Other potential reasons include underreporting, diet, genetics, and lifestyle factors.9,10 Asian Americans are at higher risk of PCa than men living in Asia, but at lower risk than other ancestries,11 suggesting that lifestyle changes and increased screening may contribute to this variation.12 Even the correlation between PSA level and PCa risk seems to differ between Chinese and Western men, with a lower risk in Chinese for similar PSA levels.13
Paradoxically, despite this lower PCa incidence and mortality in Asia, Asian men seem to present with higher grade distribution than men of European ancestry, both on autopsy14 and in screened populations.15–18 Men in Asia are diagnosed at later stages of the disease than in Western countries, with a higher incidence-to-mortality ratio.19 Despite more advanced disease at presentation, Asian ancestry has been associated with better PCa-specific survival20 compared with European men.6,21,22 The underutilization of screening in Asia cannot fully explain the lower reported incidence, nor can it account for the favorable prognosis of PCa in Asian immigrants in Western countries.23 Some studies reflect diagnostic differences, but a proportion is likely due to changes in environmental, dietary, or genetic factors. Germline and genetic analyses have shown disparities in the risk and mutational landscape of the disease between East Asian and Western cohorts.24 For instance, men of Asian ancestry have fewer TMPRSS2::ERG gene fusions.24
Despite well-known differences in disease incidence and mortality by ancestry, most PCa research has disproportionally focused on men of European heritage. Given the paucity of comparative data and the increasing PCa incidence and mortality in Asia, this study sought to determine the proportion of contemporary East Asian men in Asia found at presentation with aggressive PCa (Gleason score [GS] 8–10) on biopsy and the risk of disease progression for unfavorable PCa after radical prostatectomy (RP) compared with non-East Asian men in Western countries.
Methods
Patients
Comparison of GS Distribution on Biopsy
We retrospectively studied men who underwent prostate biopsies at 2 tertiary centers, University Health Network (UHN), Toronto, Canada, and Renji Hospital, Shanghai, China, between 2014 and 2019 (REB UHN 18-5274; Renji 201872). Data from UHN were obtained from the eProstate database and the Princess Margaret Cancer Center PCa Audit Database, and data from Renji Hospital were obtained from their departmental PCa database. In Shanghai, all men were of East Asian ancestry. Given the population diversity in Toronto, ancestry was determined using self-reports or ethnicolr (version 0.9.6), a validated machine learning model that predicts ancestry based on first and last name.25,26 East Asian men were excluded from the Toronto cohort. We reported ethnicity according to an updated guidance using preferentially “ancestry.”27
The 2014 International Society of Urological Pathology (ISUP) grading (GS ≤6, 3 + 4, 4 + 3, 8, and 9–10) was used.28 The primary outcome was the proportion of GS ≥8 disease found on biopsy in men diagnosed in Toronto compared with Shanghai. To account for potential differences in grading between institutions, an expert pathologist in Toronto (T.H. van der Kwast) blindly rereviewed a random sample of 202 biopsies from Shanghai.
Comparison of Clinical Outcomes After RP for GS ≥8 PCa
Retrospective data collection was performed for patients who underwent RP for GS ≥8 PCa at Renji Hospital, Shanghai, China, and UHN, Toronto, Canada, between 2014 and 2019 (REB UHN 19-6200; Renji 201872). This cohort was expanded to include patients from Singapore General Hospital, Sengkang General Hospital in Singapore, and L’Institut Mutualiste Montsouris in Paris, France, for the same study period (Sengkang/SGH 2009/1053/D; Montsouris 19Mar2021). Exclusion criteria included patients with known metastatic disease prior to surgery. Data from Toronto were obtained from the eProstate database and patient’s medical records. Data from Renji Hospital, Singapore General Hospital, Sengkang General Hospital and L’Institut Mutualiste Montsouris were obtained through departmental databases. Only men of East-Asian ancestry in Singapore and Shanghai were included. Patients from Toronto and Paris were all of non-East Asian ancestry.
Statistical Analyses
Comparison of the Gleason Score Distribution on Biopsy
To study the association between ancestry and GS at diagnosis, univariate and multivariable logistic regression analyses were performed. To balance groups and compare the proportion of GS ≥8 disease found on biopsy in men diagnosed in Toronto and Shanghai, men from each center were propensity score matched. The propensity score was calculated using a multivariable regression model using age, prostate volume, and PSA level at presentation (all continuous variables) to predict ancestry. Patients were matched 1:1 using a caliper of 0.20 times the standard deviation of the logit of the propensity score. Two matched cohorts were created: one restricted to patients with PSA <10 ng/mL, and the second with PSA <20 ng/mL. Differences in characteristics by ancestry were assessed using Mann-Whitney U tests and chi-square tests for continuous and categorical characteristics, respectively. To predict diagnosis of GS ≥8 disease, multivariable logistic regression models incorporating age, PSA level, log prostate volume, year of biopsy, and total number of positive cores were fit within each matched cohort.
Comparison of Clinical Outcomes After RP for GS ≥8 PCa
The primary endpoint was metastasis-free survival (MFS). Secondary outcomes included all-cause mortality, biochemical recurrence–free survival (BCR-FS), hormonal therapy–free survival (HT-FS), radiation therapy–free survival (RT-FS), and castration-resistant prostate cancer–free survival (CRPC-FS).29 Metastatic disease was defined as bony, visceral, or lymph node metastases on imaging or pathologic confirmation of metastases. PSA ≥0.2 ng/mL was considered for BCR. The start of follow-up was the RP date. Patients who did not experience the endpoint were censored at date of last follow-up or death. Survival was determined by chart review and search for confirmation of death. For patients who received neoadjuvant hormonal therapy, the biopsy GS was used when GS could not be determined on the RP specimen due to treatment effects.
Propensity score matching was used to reduce the imbalance between RP cohorts from Paris, Singapore, Toronto, and Shanghai. To calculate the propensity score, a logistic regression model predicting ancestry (East Asian vs non-East Asian) was created using age at RP (continuous), GS at RP (8 vs 9–10), neoadjuvant ADT, and preoperative PSA (<10, 10 to <20, ≥20 ng/mL). Preoperative PSA was defined as PSA level at biopsy for patients with neoadjuvant ADT, and PSA pre-RP for patients without. Patients were then matched 1:1 using nearest neighbor matching without replacement, using a caliper of 0.20 times the standard deviation of the logit of the propensity score.
Pretreatment characteristics were compared using chi-square and Mann-Whitney U tests. Actuarial rates of BCR-FS, HT-FS, CRPC-FS, and MFS were calculated using the Kaplan-Meier method, and differences between curves were evaluated using the log-rank test. Multivariable Cox proportional hazards regression was used to compare MFS, HT-FS, BCR-FS, CRPC-FS, and RT-FS in the entire cohorts and after propensity matching. MFS after BCR was also explored as an endpoint. Age at RP (continuous), preoperative PSA level (<10, 10 to <20, ≥20 ng/mL), and pathologic stage (pT2, pT3, pT4) were included as covariates in the models. The proportional hazards assumption was verified by reviewing the Schoenfeld residuals.
Statistical analyses were conducted using R version 4.0.0 (R Foundation for Statistical Computing). All hypothesis tests were 2-sided, and a P value <.05 was considered statistically significant.
Results
Comparison of Biopsy Results Between Men in Shanghai Versus Toronto
A total of 8,722 men who underwent biopsies from Shanghai and Toronto during the study period with no previous diagnosis of PCa were included. Cohort demographics are presented in Table 1. In Toronto, 335 men were of East Asian ancestry and excluded.
Patient Demographics
| Covariate | Full Cohort n (%) |
Shanghai n (%) |
Toronto n (%) |
P Valuea |
|---|---|---|---|---|
| Total, N | 8,387 | 4,905 | 3,482 | |
| Age at biopsy | <.001 | |||
| Mean [SD], y | 66.3 [8.4] | 67.7 [8.1] | 64.5 [8.3] | |
| Prostate volume (mL) | .080 | |||
| Mean [SD] | 48.1 [29.0] | 49.2 [32.3] | 47.4 [25.3] | |
| Missing | 933 | 931 | 2 | |
| Number of biopsy cores | <.001 | |||
| Mean [SD] | 12.2 [1.8] | 11.6 [1.3] | 13.1 [2.0] | |
| Number of positive cores | .005 | |||
| Mean [SD] | 5.6 [3.5] | 5.4 [3.3] | 5.8 [3.6] | |
| PSA level (ng/mL) | <.001 | |||
| Median (Q1–Q3) | 9.4 (6.1–17.1) | 12.4 (8.0–25.0) | 6.6 (4.8–9.7) | |
| Missing | 31 | 25 | 6 | |
| PSA category | <.001 | |||
| <4 ng/mL | 567 (6.8) | 87 (1.8) | 480 (13.8) | |
| 4 to <10 ng/mL | 3,900 (46.7) | 1,733 (35.5) | 2,167 (62.3) | |
| 10 to <20 ng/mL | 2,113 (25.3) | 1,571 (32.2) | 542 (15.6) | |
| ≥20 ng/mL | 1,776 (21.3) | 1,489 (30.5) | 287 (8.3) | |
| Missing | 31 | 25 | 6 | |
| Gleason score on biopsy | <.001 | |||
| <8 | 3,617 (43.1) | 1,605 (32.7) | 2,012 (57.8) | |
| ≥8 | 1,043 (12.4) | 738 (15.0) | 305 (8.8) | |
| Negative | 3,727 (44.4) | 2,562 (52.2) | 1,165 (33.5) | |
| Year (number of biopsies) | <.001 | |||
| 2014 | 982 (11.7) | 501 (10.2) | 481 (13.8) | |
| 2015 | 1,396 (16.6) | 875 (17.8) | 521 (15.0) | |
| 2016 | 1,565 (18.7) | 1,106 (22.5) | 459 (13.2) | |
| 2017 | 1,629 (19.4) | 1,021 (20.8) | 608 (17.5) | |
| 2018 | 1,708 (20.4) | 914 (18.6) | 794 (22.8) | |
| 2019 | 1,107 (13.2) | 488 (9.9) | 619 (17.8) |
Abbreviations: PSA, prostate-specific antigen; Q, quartile.
P values are estimated using the Mann-Whitney U and chi-square tests for continuous and categorical characteristics, respectively.
The mean [SD] number of biopsy samples was 11.6 [1.3] in Shanghai versus 13.1 [2.0] in Toronto (P<.001). PCa was found in 2,343 of 4,905 (48%) East Asians in Shanghai and 2,317 of 3,482 (67%) non-East Asians in Toronto (P<.001). The mean [SD] age at diagnosis was 67.7 [8.1] versus 64.5 [8.3] years and median PSA level was 12.4 (IQR, 8.0–25.0) versus 6.6 (IQR, 4.8–9.7) ng/mL in East Asians and non-East Asians, respectively (both P<.001). The proportion of men with GS ≥8 was higher in East Asian than non-East Asian men (15.0% vs 8.8%; P<.001), whereas the mean [SD] prostate volume was not different (49.2 [32.3] vs 47.4 [25.3] mL; P=.080). A higher proportion of men in Shanghai presented with serum PSA ≥20 ng/mL than in Toronto (30.5% vs 8.3%; P<.001).
The proportion of GS ≥8 was higher in East Asians than non-East Asians for both PSA <10 ng/mL (14.1% vs 6.8%) and 10 to <20 ng/mL (20.2% vs 14.8%; both P<.001), whereas for ≥20 ng/mL, there was no difference between cohorts (46.5% vs 49.6%; P=.40). Annually, the percentages of men who underwent biopsies as well as those with GS ≥8 on biopsy are shown in Table 1 and Supplementary Table S1 in the supplementary materials, available online with this article.
To account for potential grading differences in GS, a central pathology review was blindly performed on 202 biopsies from Shanghai with substantial agreement between pathologists (kappa coefficient, 0.71; concordance, 0.86).
Table 2 reports the characteristics of the matched cohorts in men who underwent prostate biopsy, for both PSA <10 and PSA <20 ng/mL ranges. Age, prostate volume, proportion of men with PSA level <4, 4 to <10, or 10 to <20 ng/mL; year of biopsy; and total number of cores were balanced in matched cohorts (PSA <10, n=1,247 per cohort; PSA <20 ng/mL, n=1,786 men per cohort). Table 3 reports the logistic regression models for these matched cohorts using PSA <10 ng/mL and <20 ng/mL cutoffs. Multivariable analysis using logistic regression analysis (GS <8/≥8) using age, PSA level, prostate volume, year of biopsy, and total number of cores for both PSA <10 ng/mL (odds ratio [OR], 0.95 [95% CI, 0.64–1.42]; P=.80) and for PSA <20 ng/mL (OR, 1.05 [95% CI, 0.77–1.43]; P=.76) showed no difference in the proportion of East Asian and non-East Asian men found with GS ≥8.
Matched Cohort Analysis: Prostate Biopsy When PSA <10 and <20 ng/mL
| PSA <10 ng/mL | PSA <20 ng/mL | |||
|---|---|---|---|---|
| Covariate | Shanghai n (%) |
Toronto n (%) |
Shanghai n (%) |
Toronto n (%) |
| Total, N | 1,247 | 1,247 | 1,786 | 1,786 |
| Age at biopsy | ||||
| Mean [SD], y | 65.5 [7.6] | 65.6 [7.7] | 65.9 [7.8] | 65.9 [7.9] |
| PSA level at biopsy (ng/mL) | ||||
| Median (Q1–Q3) | 6.7 (5.4–8.2) | 6.8 (5.3–8.3) | 8.0 (6.00–10.6) | 7.8 (5.80–10.5) |
| PSA category | ||||
| <4 ng/mL | 61 (4.9) | 102 (8.2) | 61 (3.4) | 99 (5.5) |
| 4 to <10 ng/mL | 1,186 (95.1) | 1,145 (91.8) | 1,198 (67.1) | 1,167 (65.3) |
| 10 to <20 ng/mL | 0 (0.0) | 0 (0.0) | 527 (29.5) | 520 (29.1) |
| ≥20 ng/mL | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) |
| Total Gleason score at biopsy | ||||
| No PCa | 781 (62.6) | 400 (32.1) | 1050 (58.8) | 517 (28.9) |
| 6 | 218 (17.5) | 292 (23.4) | 319 (17.9) | 372 (20.8) |
| 7 | 183 (14.7) | 485 (38.9) | 303 (17.0) | 781 (43.7) |
| 8 | 47 (3.8) | 35 (2.8) | 84 (4.7) | 55 (3.1) |
| 9 | 17 (1.4) | 34 (2.7) | 28 (1.6) | 58 (3.2) |
| 10 | 1 (0.1) | 1 (0.1) | 2 (0.1) | 3 (0.2) |
| Total cores | ||||
| Mean [SD] | 11.9 [0.6] | 13.2 [1.7] | 11.9 [0.6] | 13.2 [1.9] |
| Year | ||||
| 2014 | 130 (10.4) | 128 (10.3) | 202 (11.3) | 200 (11.2) |
| 2015 | 252 (20.2) | 245 (19.6) | 344 (19.3) | 340 (19.0) |
| 2016 | 287 (23.0) | 285 (22.9) | 355 (19.9) | 356 (19.9) |
| 2017 | 245 (19.6) | 256 (20.5) | 354 (19.8) | 354 (19.8) |
| 2018 | 203 (16.3) | 193 (15.5) | 318 (17.8) | 322 (18.0) |
| 2019 | 130 (10.4) | 140 (11.2) | 213 (11.9) | 214 (12.0) |
When matched 1:1 on age, prostate volume, and PSA level (all continuous), a total of 1,247 men from each cohort were matched for PSA <10 ng/mL and 1,786 for PSA <20 ng/mL.
Abbreviations: PCa, prostate cancer; PSA, prostate-specific antigen; Q, quartile.
Logistic Regression Models for Matched Cohort Analysis: PSA <10 and <20 ng/mL
| PSA <10 ng/mL | PSA <20 ng/mL | |||
|---|---|---|---|---|
| Covariate | OR (95% CI) | P Value | OR (95% CI) | P Value |
| Cohort | .80 | .76 | ||
| Non-East Asian | Ref | Ref | ||
| East Asian | 0.95 (0.64–1.42) | 1.05 (0.77–1.43) | ||
| Age | 1.06 (1.03–1.08) | <.001 | 1.06 (1.04–1.08) | <.001 |
| PSA | 1.11 (1.01–1.22) | .032 | 1.13 (1.09–1.17) | <.001 |
| Prostate volume/10 mL | 0.79 (0.71–0.88) | <.001 | 0.81 (0.76–0.87) | <.001 |
| Year | 1.15 (1.02–1.30) | .022 | 1.11 (1.01–1.21) | .027 |
| Total positive cores | 1.05 (0.92–1.19) | .48 | 1.07 (0.97–1.17) | .18 |
Presence of GS ≥8 was modeled in patients with PSA <10 and <20 ng/mL (ie, patients were coded as 1 if they had GS ≥8 and 0 if they had no prostate cancer or GS <8). An OR >1 denotes increased odds of having GS ≥8.
Abbreviations: GS, Gleason score; OR, odds ratio; PSA, prostate-specific antigen.
To address how PCa differs in East Asian men who reside in East Asia versus Toronto and confounding factors of ancestry by region, we performed subanalyses. We examined differences in the proportion of GS 8–10 on biopsy in East Asian men in Toronto compared with Shanghai. There were 335 men of East Asian ancestry in the Toronto cohort versus 4,905 men in the Shanghai cohort. On multivariable logistic regression, no statistically significant difference was observed between Toronto and Shanghai (P=.94; data not shown). When comparing 335 men of East Asian ancestry in Toronto versus 3,482 non-East Asians in Toronto, multivariable logistic regression models showed no difference in the proportion of GS 8–10 between groups (P=.45; data not shown).
Comparison of Clinical Outcome After RP for GS ≥8 PCa
First, in matched cohorts of men who underwent RP in Shanghai and Toronto for GS ≥8 PCa (2014–2019), no difference in 5-year MFS was observed (n=172; P>.05; data not shown). To increase the generalizability of results and statistical power, this cohort was then expanded to include additional sites of East Asians in Singapore and non-East Asians in Paris, for a total of 282 East Asians and 192 non-East Asians. Median follow-up was 46.3 and 56.2 months for East Asians and non-East Asians in this expanded cohort, respectively. Age, PSA level, stage, positive margins, adjuvant or salvage radiotherapy, use of hormonal therapy, and proportion of men found with metastatic disease were different between cohorts (Supplementary Table S2).
After propensity score matching, the 2 groups were well balanced (Figure 1). The 5-year MFS rates were 80% (95% CI, 73%–89%) in East Asian men versus 82% (95% CI, 75%–90%) in matched non-East Asian men (Figure 2A). On multivariable analysis (Supplementary Table S3), ancestry did not reach significance (hazard ratio [HR], 0.97 [95% CI, 0.55–1.70]; P=.92). As expected, MFS was influenced by pathologic stage (Supplementary Table S3).
Distribution of propensity scores by race in the (A) full, unmatched cohort and the (B) matched cohort of patients who underwent radical prostatectomy. Standardized mean differences of <0.1 in the final matched cohort were considered acceptable.
Citation: Journal of the National Comprehensive Cancer Network 2024; 10.6004/jnccn.2024.7061
Kaplan-Meier curves for (A) metastasis-free survival after RP stratified by ancestry and (B) HT-free survival after RP stratified by ancestry. Outcome data were missing for 8 Non-East Asian men. The 72-month survival probabilities in the East Asian and Non-East Asian groups were, respectively, 0.771 (95% CI, 0.679–0.876) and 0.713 (95% CI, 0.617–0.825). The difference between these survival probabilities was not statistically significant (z-test P=.42).
Abbreviations: HT, hormone therapy; RP, radical prostatectomy.
Citation: Journal of the National Comprehensive Cancer Network 2024; 10.6004/jnccn.2024.7061
Kaplan-Meier curves for HT-FS stratified by ancestry showed no difference (P=.42; Figure 2B). HT-FS was influenced by preoperative PSA level and pathologic stage (pT3 and pT4 fared worse; HR, 2.14 [95% CI, 1.48–3.09]; P<.001), but not by ancestry (HR, 1.00 [95% CI, 0.73–1.38]; P=.98; data not shown). Similarly, CRPC-FS from RP did not differ by ancestry after adjusting for age, pathologic stage, and preoperative PSA level (P=.20; Figure 3).
Kaplan-Meier curves for CRPC-free survival stratified by ancestry.
Abbreviations: CRPC, castration-resistant prostate cancer; RP, radical prostatectomy.
Citation: Journal of the National Comprehensive Cancer Network 2024; 10.6004/jnccn.2024.7061
BCR-FS was different between cohorts, (log-rank P=.029; Figure 4). Higher PSA level and stage were independently associated with higher rates of BCR (Supplementary Table S4). MFS after BCR stratified by ancestry was not different (P=.13; data not shown). Significant differences were observed for BCR-FS between those treated or not with RT (all P<.001; Supplementary Figure S1). Finally, neoadjuvant ADT had no statistically significant influence on either BCR-FS or MFS (Supplementary Figure S2).
Kaplan-Meier curves for BCR-free survival stratified by ancestry in matched cohorts. Outcome data were missing for 1 East Asian man. The 60-month BCR-free survival rate was 0.36 (95% CI, 0.28–0.47) for East Asians vs 0.44 (95% CI, 0.36–0.54) for non-East Asians.
Abbreviations: BCR, biochemical recurrence; RP, radical prostatectomy.
Citation: Journal of the National Comprehensive Cancer Network 2024; 10.6004/jnccn.2024.7061
Discussion
PCa cases are projected to increase annually from 1.4 to 2.9 million in 2040 worldwide.8 Similarly, PCa diagnosis and mortality is expected to double in Asia by 2030.10 This could be due to the westernization of many East Asian countries like China, as well as urbanization, dietary changes, and increase in rates of teenage obesity over the last decades.30–34 Despite PCa becoming an emerging health priority in East Asia, our knowledge about this disease remains limited. Furthermore, the Asian population constitutes the fastest growing group in the United States and Canada.35 New approaches to improve early detection, awareness, and research diversification are therefore urgently needed, as discussed in the recent Lancet Commission on PCa.8 This is especially relevant in East Asia, where increasing life expectancy may drive large increases in incidence.8
Consistent with previous reports, this study found that East Asian men present with a higher proportion of GS 8–10 at diagnosis than non-East Asian men.20,21,36 Men from Shanghai presented with higher PSA levels than in Toronto, with 30% of them having PSA ≥20 ng/mL, likely due to differences in access to and delivery of screening. In China, PSA screening is less common, resulting in PCa diagnoses at a later stage and higher PSA levels.
However, at PSA levels typical of early detection programs (PSA <10 or <20 ng/mL), when carefully matched, our findings do not demonstrate a higher proportion of aggressive disease with GS ≥8 in East Asians from Shanghai.
In contrast with other population-based studies,22,23 there was no difference in disease progression after RP for GS 8–10 PCa (MFS) between East Asians and non-East Asians. Asian men have been reported to have a better PCa survival rate, even for unfavorable disease.6,37–39 In a SEER study with 891,100 patients diagnosed between 1988 and 2010, Asian men had a decreased risk (0.67) of PCa-specific mortality, although they were more likely to present with advanced disease.40 Men of European ancestry (n=203,295) had poorer PCa-specific survival compared with Asian men (n=13,497) despite presenting more often with localized disease and less de novo metastasis.41 However, these population-based studies are hampered by the lack of granular clinical information, preventing adequate matching of cohorts on variables such as PSA level at presentation, which holds important prognostic value.
The elevated PSA levels at diagnosis in East Asian patients imply more delayed presentations, likely indicative of variances in access to early detection.
In randomized clinical trials, Asians have also been found to have better outcomes with metastatic PCa41 and a greater likelihood of benefiting from therapy with abiraterone + olaparib.42 However, Asian men treated with docetaxel respond similarly to men of European ancestry.43 A further analysis of ancestry and modifiable factors underpinning survival differences should be explored.
Unlike previous studies, the strengths of our research include a large sample size with detailed clinical information, allowing for propensity score matching. Additionally, the central pathologic review addressed the well-established limitation of grading variability between pathologists.44 To our knowledge, no other study included such pathologic review. Also, and importantly, in contrast to previous studies that did not adjust for PSA level,22,23,37,39 our multivariable analysis adjusted for this key parameter at diagnosis. Lastly, we included East Asian and non-East Asian men from multiple countries in the RP analysis to enhance the generalizability of the results. Surgical outcomes were in line with published series, suggesting consistent surgical quality.
Our study has limitations. As with any retrospective study, differences in patient selection may have introduced unknown confounders. Selection bias and the risk profile of the cohorts could affect the proportion of aggressive PCa between sites. Our results came from tertiary referral centers and may not be generalizable. Shanghai and Singapore are large urbanized modern cities with a strong Western influence. Therefore, our results may not apply to men in rural centers. Additionally, our study results should be viewed in the context that ancestry is inextricably associated with region, which includes numerous unmeasured social determinants of health, quality of medical care, and environmental influences. Therefore, our findings cannot definitively address causation of specific factors, such as ancestry, on the measured oncologic outcomes.
In Toronto, the database did not include self-reported ancestry for all patients, thus software was implemented to determine ancestry, which can introduce errors and bias.
A greater proportion of men in Shanghai presented with elevated PSA level compared with those in Toronto, but body mass index, which may influence PSA level, was not captured. The median follow-up time after RP was longer in non-East Asians (56.2 vs 46.3 months), although this would only reinforce our findings. A small number of events at 8 years after RP limited the ability to detect differences with longer follow-up. However, MFS survival probabilities after RP in East Asians versus non-East Asians at 72 months was not statistically different, and it is unlikely, given the aggressive biology of GS 8–10 PCa, that these outcomes would be significantly different with longer follow-up. Our median follow-up time of <5 years after prostatectomy is also not long enough to make conclusions about survival outcomes.
Our findings after RP cannot be extrapolated to men with more advanced disease because our dataset did not include men with more advanced or metastatic PCa. Finally, PCa epidemiology and genomics across Asia are diverse, and these results may not be generalizable to other Asian populations. Disentangling the respective contributions of biology and differences in access to care among Asian men in various countries is important because this could result in differences in PCa outcomes.45
Conclusions
In contrast to previous reports suggesting that Asian men present with worse PCa grades at diagnosis but have better outcomes, this contemporary study demonstrates that East Asians living in China are equally likely to be found with aggressive PCa on biopsy as matched non-East Asians living in Canada at PSA levels typical of early detection programs. GS ≥8 PCa demonstrated similar aggressiveness after surgery regardless of ancestry or region. The assumption that unfavorable PCa is less aggressive in East Asians should be revisited. Our findings are especially relevant given the projected increases in incidence and mortality rates of PCa, especially in East Asia. Reducing late diagnoses to prevent PCa-related deaths is key, and our study supports the conclusions drawn by the recent Lancet Commission on PCa. Without urgent action for new approaches to improve screening, awareness, and research diversification, these trends will cause global PCa deaths to increase.
References
- 1.↑
Culp MB, Soerjomataram I, Efstathiou JA, et al. Recent global patterns in prostate cancer incidence and mortality rates. Eur Urol 2020;77:38–52.
- 2.↑
Nyame YA, Gulati R, Heijnsdijk EAM, et al. The impact of intensifying prostate cancer screening in black men: a model-based analysis. J Natl Cancer Inst 2021;113:1336–1342.
- 3.↑
Bergengren O, Pekala KR, Matsoukas K, et al. 2022 update on prostate cancer epidemiology and risk factors—a systematic review. Eur Urol 2023;84:191–206.
- 4.↑
Chen W, Sun K, Zheng R, et al. Cancer incidence and mortality in China, 2014. Chin J Cancer Res 2018;30:1–12.
- 5.↑
Kimura T, Egawa S. Epidemiology of prostate cancer in Asian countries. Int J Urol 2018;25:524–531.
- 6.↑
Zhu Y, Mo M, Wei Y, et al. Epidemiology and genomics of prostate cancer in Asian men. Nat Rev Urol 2021;18:282–301.
- 7.↑
Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2021;71:209–249.
- 8.↑
James ND, Tannock I, N’Dow J, et al. The Lancet Commission on prostate cancer: planning for the surge in cases. Lancet 2024;403:1683–1722.
- 9.↑
Taitt HE. Global trends and prostate cancer: a review of incidence, detection, and mortality as influenced by race, ethnicity, and geographic location. Am J Mens Health 2018;12:1807–1823.
- 10.↑
Chen W, Zheng R, Baade PD, et al. Cancer statistics in China, 2015. CA Cancer J Clin 2016;66:115–132.
- 11.↑
Baade PD, Youlden DR, Cramb SM, et al. Epidemiology of prostate cancer in the Asia-Pacific region. Prostate Int 2013;1:47–58.
- 12.↑
Vidal AC, Oyekunle T, Feng T, et al. Asian race and risk of prostate cancer: results from the REDUCE study. Cancer Epidemiol Biomarkers Prev 2020;29:2165–2170.
- 13.↑
Chen R, Sjoberg DD, Huang Y, et al. Prostate specific antigen and prostate cancer in Chinese men undergoing initial prostate biopsies compared with Western cohorts. J Urol 2017;197:90–96.
- 14.↑
Zlotta AR, Egawa S, Pushkar D, et al. Prevalence of prostate cancer on autopsy: cross-sectional study on unscreened Caucasian and Asian men. J Natl Cancer Inst 2013;105:1050–1058.
- 15.↑
Chen R, Ren S, Yiu MK, et al. Prostate cancer in Asia: a collaborative report. Asian J Urol 2014;1:15–29.
- 16.↑
Fowke JH, McLerran DF, Gupta PC, et al. Associations of body mass index, smoking, and alcohol consumption with prostate cancer mortality in the Asia Cohort Consortium. Am J Epidemiol 2015;182:381–389.
- 17.↑
World Health Organization. Ageing and health in China. Accessed March 12, 2024. Available at: https://www.who.int/china/health-topics/ageing
- 18.↑
Uemura H, Ye D, Kanesvaran R, et al. United in Fight against prOstate cancer (UFO) registry: first results from a large, multi-centre, prospective, longitudinal cohort study of advanced prostate cancer in Asia. BJU Int 2020;125:541–552.
- 19.↑
Na R, Jiang H, Kim ST, et al. Outcomes and trends of prostate biopsy for prostate cancer in Chinese men from 2003 to 2011. PLoS One 2012;7:e49914.
- 20.↑
Takahashi H, Epstein JI, Wakui S, et al. Differences in prostate cancer grade, stage, and location in radical prostatectomy specimens from United States and Japan. Prostate 2014;74:321–325.
- 21.↑
Kang DI, Chung JI, Ha HK, et al. Korean prostate cancer patients have worse disease characteristics than their American counterparts. Asian Pac J Cancer Prev 2013;14:6913–6917.
- 22.↑
Zhang C, Zhang C, Wang Q, et al. Differences in stage of cancer at diagnosis, treatment, and survival by race and ethnicity among leading cancer types. JAMA Netw Open 2020;3:e202950.
- 23.↑
Wang C, Kamrava M, King C, Steinberg ML. Racial disparity in prostate cancer-specific mortality for high-risk prostate cancer: a population-based study. Cureus 2017;9:e961.
- 24.↑
Li J, Xu C, Lee HJ, et al. A genomic and epigenomic atlas of prostate cancer in Asian populations. Nature 2020;580:93–99.
- 25.↑
Chintalapati R, Laohaprapanon S, Sood G. Predicting race and ethnicity from the sequence of characters in a name. arXiv. Published online July 8, 2023. Available at: https://arxiv.org/abs/1805.02109
- 26.↑
Kwong JCC, Khondker A, Meng E, et al. Development, multi-institutional external validation, and algorithmic audit of an artificial intelligence-based Side-specific Extra-Prostatic Extension Risk Assessment tool (SEPERA) for patients undergoing radical prostatectomy: a retrospective cohort study. Lancet Digit Health 2023;5:e435–445.
- 27.↑
Flanagin A, Frey T, Christiansen SL. Updated guidance on the reporting of race and ethnicity in medical and science journals. JAMA 2021;326:621–627.
- 28.↑
Epstein JI, Egevad L, Amin MB, et al. The 2014 International Society of Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma: definition of grading patterns and proposal for a new grading system. Am J Surg Pathol 2016;40:244–252.
- 29.↑
Freedland SJ, Howard L, Wallis CJD, et al. Correlation between castration resistant prostate cancer (CRPC) free survival (CRPC-FS) and metastasis free survival (MFS) in men initiating androgen deprivation therapy (ADT) for biochemical recurrence (BCR) after radical prostatectomy (RP): results from the SEARCH database. Ann Oncol 2020;31:S543.
- 30.↑
Gibson C, Kay J. Historical census statistics on the foreign-born population of the United States: 1850 to 2000. United States Census Bureau. Accessed March 12, 2024. Available at: https://www.census.gov/library/working-papers/2006/demo/POP-twps0081.html
- 31.↑
Statistics Canada. Asian Heritage Month…by the numbers. Accessed March 12, 2024. Available at: https://www.statcan.gc.ca/en/dai/smr08/2021/smr08_250
- 32.↑
Dong Y, Jan C, Ma Y, et al. Economic development and the nutritional status of Chinese school-aged children and adolescents from 1995 to 2014: an analysis of five successive national surveys. Lancet Diabetes Endocrinol 2019;7:288–299.
- 33.↑
Tabei T, Taguri M, Sakai N, et al. Does screening for prostate cancer improve cancer-specific mortality in Asian men? Real-world data in Yokosuka City 15 years after introducing PSA-based population screening. Prostate 2020;80:824–830.
- 34.↑
Andriole GL, Crawford ED, Grubb RL, et al. Mortality results from a randomized prostate-cancer screening trial. N Engl J Med 2009;360:1310–1319.
- 35.↑
American Association for Cancer Research. The state of cancer health disparities in 2022. Accessed April 21, 2024. Available at: https://cancerprogressreport.aacr.org/disparities/cdpr22-contents/cdpr22-the-state-of-cancer-health-disparities-in-2022/
- 36.↑
Fedewa SA, Etzioni R, Flanders WD, et al. Association of insurance and race/ethnicity with disease severity among men diagnosed with prostate cancer, National Cancer Database 2004-2006. Cancer Epidemiol Biomarkers Prev 2010;19:2437–2444.
- 37.↑
Robbins AS, Koppie TM, Gomez SL, et al. Differences in prognostic factors and survival among white and Asian men with prostate cancer, California, 1995-2004. Cancer 2007;110:1255–1263.
- 38.↑
Stevenson JKR, Cheung MC, Earle CC, et al. Chinese and South Asian ethnicity, immigration status, and clinical cancer outcomes in the Ontario Cancer System. Cancer 2018;124:1473–1482.
- 39.↑
Stern N, Ly TL, Welk B, et al. Association of race and ethnicity with prostate cancer-specific mortality in Canada. JAMA Netw Open 2021;4:e2136364.
- 40.↑
Chao GF, Krishna N, Aizer AA, et al. Asian Americans and prostate cancer: a nationwide population-based analysis. Urol Oncol 2016;34:233.e7–15.
- 41.↑
Ni X, Luu M, Ma W, et al. Impact of Asian race on prognosis in de novo metastatic prostate cancer. J Natl Compr Canc Netw 2023;21:733–741.e3.
- 42.↑
Saad F, Clarke NW, Oya M, et al. Olaparib plus abiraterone versus placebo plus abiraterone in metastatic castration-resistant prostate cancer (PROpel): final prespecified overall survival results of a randomised, double-blind, phase 3 trial. Lancet Oncol 2023;24:1094–1108.
- 43.↑
Halabi S, Dutta S, Tangen CM, et al. Comparative survival of Asian and White metastatic castration-resistant prostate cancer men treated with docetaxel. JNCI Cancer Spectr 2020;4:pkaa003.
- 44.↑
Flach RN, Egevad L, Eklund M, et al. Use of the ISUP e-learning module improves interrater reliability in prostate cancer grading. J Clin Pathol 2023;77:22–26.
- 45.↑
Nguyen DD, Kwong JCC, Zlotta AR, Wallis CJD. Letter to the editor: disaggregating Asian prostate cancer cohorts can help us improve care for all prostate cancer patients. J Natl Compr Canc Netw 2023;21:xli.
