Misclassification of Gleason grade and tumor stage in Asian‐American patients with low‐risk prostate cancer

To investigate the misclassification rates of Asian‐American patients with low‐risk prostate cancer (PCa). Patients diagnosed with low‐risk PCa treated with radical prostatectomy between 2010 and 2015 in the Surveillance, Epidemiology, and End Results database were included in this study. Then, basic characteristics and pathological outcomes of the enrolled patients were retrospectively extracted. We compared the rates of upgrading and/or upstaging between Asian‐American patients and White/Black patients. Moreover, temporal trend analyses were performed to explore the changes in upgrading and upstaging rates in each race over time. Finally, logistic regression models were constructed to explore the role of Asian race in upgrading and upstaging and to screen out potential risk factors for predicting upgrading and upstaging in Asian‐American patients. Asian‐Americans had a significantly higher rate of upgrading than Whites (P < .001), while no statistical difference was found in the comparison of upstaging rate (P = .536). Moreover, Asian‐Americans were more likely to upgrade to diseases with higher ISUP grade than Whites (P = .010). The rate of upgrading increased significantly over time in White and Black patients, but not in Asian‐American patients. Finally, race seemed to be an independent risk factor for predicting upgrading, while the racial differences seemed to be more pronounced between White and Black patients. Asian‐American patients had a significantly higher rate of upgrading than White patients. Moreover, Asian‐American patients were more likely to upgrade to diseases with higher ISUP grade. Further risk assessment before clinical decision for low‐risk PCa patients with the help of significant clinical variables is required.

clinical practice guidelines, the most recommended/preferred treatment for patients with low-risk (PSA < 10 ng/mL, Gleason score <7 and clinical T1-2a) PCa is active surveillance (AS), rather than invasive radical prostatectomy (RP) or radiotherapy (www.nccn.org/patients, https:// uroweb.org/). In recent years, many studies have discussed the best indication of AS for PCa. Clearly, it is essential and difficult to develop strategies to balance disease progression and over treatment. Hassan et al. (3) demonstrated that 3.9% of pure Gleason score 3 + 3 patients experienced focal extra-prostatic extension at RP, while Gleason score 6 was rarely associated with seminal vesicle invasion and lymph node metastasis. Sato et al. (4) found that Gleason score 3 + 4 patients with the highest Gleason pattern 4 < 5% had a similar risk of adverse pathology and biochemical recurrence with Gleason score 3 + 3 patients.
However, most references included in the clinical guidelines were from European and American countries, and most of the enrolled patients were White. To have a better understanding of the racial disparities in low-risk PCa, many studies (5,6) have investigated the role of Black race in AS. However, the lack of high-quality, well-designed clinical trials in Asian countries makes it difficult to explore the misclassification rates of Gleason grade and tumor stage for low-risk patients. Additionally, the risk of upgrading and upstaging after RP is an important evaluation criterion for assessing whether some patients should be treated with AS. Therefore, we performed this study to explore the misclassification of Gleason grade and tumor stage in Asian patients who met the AS criteria through the Surveillance, Epidemiology, and End Results (SEER) database, to develop the bestpersonalized treatment strategy for these patients.

| MATERIALS AND METHODS
In this study, all data were extracted from the SEER database utilizing the SEER*Stat software (Version 8.3.6; NCI). As one of the largest tumor databases in the world, the SEER registry currently covers 30% of the US population. We signed the data agreement and accessed this database with the username of 10977-Nov2019. Considering the publicity of the data itself, our study was exempt from Institutional Review Board (IRB) approval.

| Patient identification
Patients diagnosed with low-risk PCa from 2010 to 2015 were retrospectively identified from the SEER 18 database. All patients should meet the following criteria: (1) diagnosed with prostate adenocarcinoma with positive pathology, (2) type of reporting source was not from death certificate/autopsy only, (3) patients were treated using RP. Furthermore, the exclusion criteria were as follows: (1) patients with unknown data on some important variables, such as prostate-specific antigen (PSA) values, Gleason score on needle core biopsy/transurethral resection of prostate (TURP) and RP, clinical T stage and race, (2) patients were treated with neoadjuvant radiotherapy or chemotherapy prior to RP, (3) patients with nonstandard records in some variables.

| Upgrading and upstaging
In this study, upgrading was defined as at least 1 ISUP grade increase from needle biopsy/TURP ISUP grade to the final pathology at RP, and upstaging was defined as a non-organ confined tumor (≥ pT3 or pN1) at RP. In addition, upgrading and/or upstaging was a combination of both.
First, our study made comparisons in the rate of upgrading, upstaging, and upgrading and/or upstaging between Asian race and White/ Black race, and further analyzed the specific categories of upgrading in each group. Then, we investigated the potential interaction between upgrading and upstaging. Similarly, these comparisons were conducted within different Asian populations. Subsequently, we analyzed the temporal trends of upgrading and upstaging rates, and then calculated the average annual percent change (AAPC) of each race. Additionally, a multivariate logistic regression model was developed to explore the role of Asian race in upgrading and upstaging. Finally, uni-and multivariate logistic regression analyses were performed to screen out potential risk factors for predicting upgrading and upstaging in Asian populations.

| Statistical analyses
Categorical variables were presented in the form of n (%), and chisquare test or Fisher's exact test was used to test the statistical significance of proportion differences. Continuous variables that disobey the normal distribution were displayed in the form of median (IQR), and Mann-Whitney U or Kruskal-Wallis test was applied for the statistical significance of differences. Temporal trends of upgrading/upstaging rates were examined and AAPCs were calculated using Joinpoint software (version 4.9.0.0, https://surveillance.cancer.gov/ joinpoint/). SPSS 23.0 software (SPSS Inc., Chicago, IL, USA) was utilized for all statistical analyses, and two-sided P < .05 was statistically significant.

| Basic characteristics
The flowchart of patient selection is shown in Figure 1. As given in
In upgraded patients, Asian-American patients were more likely to upgrade to diseases with higher ISUP grade than White patients (P = .010). Moreover, no significant differences were detected in the In Asian population (Table S2), Japanese patients had the highest rate of upgrading (66.00%), while patients from South Asian had the lowest upgrading rate (46.09%). Meanwhile, the rate of upgrading and/or upstaging was 69.00% in Japanese patients, which was significantly higher than that in patients from South Asian (47.66%). However, no significant differences were detected regarding the proportions of upgrading with concomitant upstaging (all P > .05).

| Multivariate regression analyses
Finally, a multivariate logistic regression model was developed to investigate the effect of Asian Race on upgrading and/or upstaging in patients with low risk PCa. As given in Table 4, race (P = .015) seemed to be an independent risk factor for predicting upgrading in patients with low risk PCa after being adjusted by age at diagnosis, PSA, clinical T stage, number or biopsy cores, number of positive cores and core ratio. However, the racial differences seemed to be more pronounced between White and Black patients, rather than between Asian-American and White/Black patients. Older age at diagnosis, the higher PSA, the smaller number of biopsy core and larger number of positive cores could predict a higher risk of upgrading and upstaging.
Univariate logistic regression analyses revealed that race, age at diagnosis, PSA, number of biopsy cores, number of positive cores and core ratio were predictive factors for upgrading in Asian-American patients. Finally, multivariate logistic regression analyses showed that Japanese, higher PSA, smaller number of biopsy cores and larger number of positive cores were tightly associated with higher risk of F I G U R E 1 Flowchart of patient selection.
upgrading (Table S3). Similarly, higher PSA and larger number of positive cores were closely related to a higher risk of upstaging in Asian-American patients (Table S4).

| DISCUSSION
Low risk PCa is often indolent, and the affected patients are more likely to die from other causes than the tumor itself. In addition, invasive RP may bring unnecessary adverse events (urinary or bowel dysfunction, impotence) to these patients and reduce their quality of life (7,8). Hence, as is recommended as the preferred treatment for low- In our study, patients had a higher rate of upgrading no matter in which race (ranging from 45.18% to 51.25%), which meant that upgrading should be expected in 1 in 2 individuals with low-risk PCa.
Moreover, Asian-American patients were more likely to upgrade to ISUP grade 3-5 diseases than White patients. The rate of upgrading to ISUP 3-5 diseases was significant higher in Asian-American patients than that in White and Black patients. Clearly, the prognosis and tumor characteristics of ISUP grade 3-5 diseases are significantly different from that of ISUP grade 1-2 diseases. Previous study (18) found that PCa patients with ISUP grade 1-2 (99.1%-99.5%) had better 8-year PCa-specific mortality than those with ISUP grade 3-5 (85.8%-97.4%, P < .001) at biopsy. Although only 77 Asian-American patients were upgraded to ISUP 3-5, this also accounted for about 10% of the total Asian population. One in ten people upgraded to PCa with Gleason score of 4 + 3 or higher, which did not include some patients who did not upgrade to ISUP 3-5 but upstage to pT3+ or pN1. Therefore, many low-risk PCa Asian-American patients will  (23) and so on. In Asian patients, Japanese race had a significant higher risk of upgrading than other races. However, race seemed to have no significant impact on upstaging.
In recent years, the imaging technique of PCa detection has undergone tremendous changes, mainly including transrectal ultrasound (TRUS), multi-parameter magnetic resonance imaging (mpMRI) and prostate specific membrane antigen (PSMA). It was well-known that significant discrepancies were observed in results among different imaging techniques, which were used to guide the biopsy. Wegelin et al. (24) found that MRI-guided biopsy had increased detection rates of clinically significant PCa (csPCa) than TRUS-guided biopsy.

| CONCLUSION
Patients with low-risk PCa had a high risk of misclassification regardless of race. Asian-American patients had a significantly higher rate of upgrading to ISUP 3-5 diseases than White and Black patients. Moreover, Asian-American patients were more likely to upgrade to diseases with higher ISUP grades. In addition, Japanese patients had the highest risk of upgrading among Asian patients. Further risk assessment before clinical decision for these patients with the help of significant clinical variables is required.

CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest.

ETHICS STATEMENT
Ethical review and approval were waived for this study due to this study using de-identified data from the National Cancer Institute's Surveillance, Epidemiology, and End Results program.

INFORMED CONSENT STATEMENT
This study used de-identified data from the National Cancer Institute's Surveillance, Epidemiology, and End Results program, and informed consent was not required.

DATA AVAILABILITY STATEMENT
Please contact the author for data requests.