Abstract
Background/Aim: We aimed to evaluate disparities in presentation and treatment of gastric cancer (GC), including time between diagnosis and treatment, based on race, focusing on Japanese patients within the USA. Patients and Methods: The National Cancer Database was queried for patients diagnosed with GC between 2004-2013. Clinical and treatment variables were summarized by race (White, non-Japanese Asian, Japanese). The association between race and overall survival (OS) was evaluated using the log-rank test. Results: A total of 79,481 patients were included. Japanese patients received surgery the earliest after diagnosis in all stages. Regarding radiotherapy, white patients had the shortest waiting time followed by Asian and Japanese patients. Asian patients had better OS at both 3 and 5 years of follow-up. White patients were associated with the lowest OS. Conclusion: Japanese and Asian GC patients have better OS compared to White patients. Moreover, there were disparities in time to both GC diagnosis and treatment, with Japanese patients being sooner diagnosed and surgically treated, which may ultimately impact patient experience.
It is well established that gastric cancer (GC) is one of the most aggressive cancers worldwide, with approximately 990,000 cases and 740,000 deaths annually (1, 2). In the United States of America alone, there are over 26,000 new cases with 10,800 deaths in 2016 (3). One factor that has been of particular interest is the role of race and associated disparities, in the presentation and treatment of GC. Recent literature has shown that race/ethnicity is an important risk factor for GC as well as a strong predictor of overall survival (OS) (4-12). Within 2019 alone, multiple groups have elaborated on the role of race on the presentation and treatment of GC (4, 7, 8, 11, 12).
Regarding the specific role of Asian ancestry on patient OS, recent studies have found that patients with an Asian ancestry had improved OS compared to other races (4-7, 9, 10, 12). A recent study, performed by Rhome et al., has found that GC patients of Asian ancestry treated in the United States of America had improved OS compared to other stage-matched racial groups (12). However, the time between diagnosis and treatment was not addressed in any of the aforementioned studies. Since early initiation of treatment leads to improved OS in cancer (including GC) patients, there is an unmet need for research evaluating any disparities in time between diagnosis and treatment between different races in the USA (13-15). Therefore, our aim was to evaluate disparities in presentation and treatment of GC, including the time between diagnosis and treatment, based on race.
Patients and Methods
Patients. The National Cancer Database (NCDB) was retrospectively queried for patients diagnosed with gastric adenocarcinoma between 2004 and 2013. Patients with histology different from adenocarcinoma and race category not labeled as White or Asian were excluded from the study. As this study utilized a nationwide, de-identified database, it was deemed exempt from our Institutional Review Board.
Statistical analysis. Clinical variables were summarized by race (white, non-Japanese Asian, Japanese) using mean and standard deviation for continuous variables, and using frequencies for categorical variables. Associations with race were evaluated using Wilcoxon rank sum and Chi-square tests. Treatment variables were summarized by race (white, non-Japanese Asian, Japanese) within each pathological stage using mean and standard deviation for continuous variables, and using frequencies for categorical variables. Associations with race were evaluated using Wilcoxon rank sum and Chi-square tests. Survival information was summarized using standard Kaplan-Meier methods. The association between race and OS curves was evaluated using log-rank test. Cox regression model was constructed for survival outcome variables with covariates: race, age, sex, facility type, insurance, chemotherapy, and radiation. All analyses were conducted in SAS v9.4 (Cary, NC, USA) at a significant level of 0.05.
Summary of clinical variables by race.
Results
A total of 79,481 patients were included in the analysis after the application of the inclusion and exclusion criteria. Table I summarizes clinical variables and short-term outcomes for each cohort based on race. Table II shows treatment variables for each race category for pathological stages 0 to 4. When examining the time interval between diagnosis and treatments, statistically significant differences were found between cohorts at each stage for every type of treatment. Except for stage 0, Japanese patients received surgery the earliest after diagnosis at all stages, followed by Asian and white patients (p<0.001). An opposite pattern was found with regard to radiotherapy, where white patients had the shortest waiting time, followed by Asian and Japanese patients (p<0.001 for stages 1, 2, and 3; p=0.003 for stage 4). Regarding chemotherapy, white patients received it the earliest at all stages; however, for stages 1 and 2, Japanese patients received it before Asians, while the opposite occurred in stages 3 and 4 (p<0.001).
Figure 1 shows Kaplan-Meier curves for OS of the overall white, Asian and Japanese cohorts divided by pathological stage, including 3-year OS, 5-year OS, and median survival rates. Regarding the median survival, Japanese patients had better outcomes for stages 3 and 4 compared to Asian ones. White patients were associated with the lowest survival rate for every stage.
The results of the multivariate model used to calculate the hazard ratio (HR) for each variable are shown in Table III. Female gender was found to be significantly associated with better survival rate for stage 0 (HR=0.74, 95%CI=0.57-0.97), stage 1 (HR=0.80, 95%CI=0.75-0.86) and stage 2 GC (HR=0.85, 95%CI=0.79-0.92). Race was significant for each stage except for stage 0. In stage 1 disease, compared to White patients, Asian ones had the best survival outcomes (HR=0.47, 95%CI=0.40-0.56), followed by Japanese (HR=0.71, 95%CI=0.57-0.89). For stages 2, 3, and 4, Japanese patients (2: HR=0.58, 95%CI=0.41-0.81; 3: HR=0.64, 95%CI=0.49-0.83; 4: HR=0.73, 95%CI=0.60-0.89) had the lowest risk of death, followed by Asian patients (2: HR=0.63, 95%CI=0.54-0.73; 3: HR=0.72, 95%CI=0.64-0.80; 4: HR=0.79, 95%CI=0.72-0.87).
Summary of treatment variables by race for pathological stages 0-4.
Parameter estimate of multivariate Cox regression model.
Discussion
Numerous studies have explored the various disparities associated with GC presentation and treatment, based on race/ethnicity, age, socioeconomic factors, etc. However, there is very limited information on the role of any disparities in time from initial diagnosis with GC to treatment (i.e. chemotherapy, radiotherapy, and surgical intervention) based on race. The present study showed that the time between initial diagnosis and surgical intervention was shortest for patients of Japanese and Asian ancestry compared to whites at all stages of GC. Interestingly, the opposite trend was observed for time between initial diagnosis and neoadjuvant therapy (chemotherapy and radiotherapy), where whites received treatment significantly sooner than Asian and Japanese patients.
As noted previously, many studies have concluded that patients of Asian ancestry had higher OS compared to other races, including whites (5, 12, 16). In support, we showed that Asian and Japanese patients had the highest 5-year OS at all stages compared to whites. These differences in OS were more evident at early stages, ranging from 20% in stages 0 and 1 to 5% in stage 4. The consensus of the rationale behind these consistent findings is that there are several differences between Eastern and Western populations (16-19).
Overall survival curves for all stages and summary of survival variables by race.
Unique to these other studies, our findings showed that there was a significant difference in the time between initial diagnosis and treatment for GC between Japanese, Asians, and Whites. Interestingly, for all stages, Japanese patients had the shortest time period between initial diagnosis and surgical intervention (nearly 20 days earlier on average), followed by Asian and White patients. Although it cannot be confirmed from the results of this study, this may indicate a potential connection between Japanese ancestry and the decision to receive surgery sooner, compared to chemo- or radio-therapy. This could be related to the current guidelines for GC that state surgery alone is sufficient for stage I GC, which represents nearly 50% of cases (20-22). Therefore, patients of Japanese ancestry may maintain the mindset of receiving surgery alone. Regarding the impact of time between initial diagnosis and treatment on OS, it has been shown that an increased time period does not significantly impact the patient's OS (23, 24). However, it can potentially impact the overall patient experience. For instance, an increased time period between initial diagnosis and treatment may result in higher levels of stress and anxiety of the patient, which could ultimately lead to a worse overall treatment experience (25, 26).
This study has several limitations. The NCDB data is compiled from over 70% of newly diagnosed cancer cases nationwide, which allows for the inclusion of a large number of diverse patients from across the country in different treatment facilities and settings. However, this analysis is limited by its retrospective nature and related selection biases. Additionally, NCDB does not allow gathering data on disease specific survival, which would provide more accurate information on cancer patients' survival than OS.
Conclusion
In conclusion, the present study confirmed the results of previous studies on OS of Japanese, Asian, and white GC patients, and supported the conclusion that Japanese and Asian patients have better OS compared to white patients. More interestingly and novel, this study highlighted a significant disparity in time between initial diagnosis of GC and treatment for Japanese, Asian, and white patients.
Acknowledgements
This work was supported by Roswell Park Cancer Institute and National Cancer Institute (NCI) grant P30CA016056. We acknowledge and thank the American College of Surgeons Committee on Cancer for providing access to the Participant User File from the National Cancer Data Base. Dr. Gabriel was supported by CTSA Grant Number KL2 TR002379 from the National Center for Advancing Translational Science (NCATS). The contents of the paper are solely the responsibility of the Authors and do not necessarily represent the official views of the NIH.
Footnotes
Authors' Contributions
Drs. Lemini, Jorgenson, Almerey, and Gabriel collected and analyzed the data. Dr. Attwood performed the statistical analysis. Drs. Elli, Colibaseanu, Bagaria, and Gabriel developed the study objectives and design. Drs. Lemini and Gabriel drafted the manuscript. Drs. Elli, Colibaseanu, and Bagaria critically reviewed and edited the manuscript. All Authors reviewed and approved the final manuscript.
Conflicts of Interest
There are no conflicts of interest related to this work for any of the listed Authors.
Disclaimers
The American College of Surgeons Committee on Cancer provided the Participant User File from the National Cancer Data Base, but has not reviewed or validated the results or conclusions of our study.
- Received December 18, 2019.
- Revision received December 27, 2019.
- Accepted January 8, 2020.
- Copyright© 2020, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
