See referenced original article on pages 2282–2290, this issue.
Racial disparities in prostate cancer care: Is adherence to National Comprehensive Cancer Network guidelines good enough for our patients?
Version of Record online: 10 APR 2013
Copyright © 2013 American Cancer Society
Volume 119, Issue 12, pages 2209–2211, 15 June 2013
How to Cite
Master, V. A. and Moses, K. A. (2013), Racial disparities in prostate cancer care: Is adherence to National Comprehensive Cancer Network guidelines good enough for our patients?. Cancer, 119: 2209–2211. doi: 10.1002/cncr.28006
- Issue online: 4 JUN 2013
- Version of Record online: 10 APR 2013
- Manuscript Accepted: 23 JAN 2013
- Manuscript Revised: 15 JAN 2013
- Manuscript Received: 8 JAN 2013
Disparities in incidence, treatment, and outcome of prostate cancer in African American (AA) men have been well documented. Data from the American Cancer Society indicate that AA men have a 2.4-fold likelihood of prostate cancer death compared with Caucasian men. Although several groups have postulated that differences in presentation and survival may be caused by genetic factors, sociodemographic factors, and/or comorbidity, it has been demonstrated that these differences mostly can be obviated when controlling for treatment received and sociodemographic factors. Even among patients who receive definitive treatment for prostate cancer, there are significant, persistent differences in the type of treatment received. Multiple reports from population-based data sets have revealed persistent differences in the receipt of treatment and the overuse of noncurative therapies in AA men. Therefore, the identification of disparities in treatment for prostate cancer remains an important public health issue.
In this issue of Cancer, Ellis et al performed a well designed study using the North Carolina Health Care Access Project database to determine the receipt of National Comprehensive Cancer Network (NCCN) guideline-based therapy for newly diagnosed prostate cancer according to race and risk classification. Those investigators identified 341 AA men and 436 Caucasian men and stratified them into 5 risk groups: namely low, intermediate, high, very high, and lymph node involvement/metastatic. AA men were younger, had higher prostate-specific antigen levels and Gleason grade, and were more likely to be uninsured, to be unmarried, and to have lower levels of educational attainment. It is noteworthy that there were no significant differences in clinical staging, Charlson comorbidity index scores, or perceptions of mistrust of health care providers; however, AA men were more likely to report lower perceived access to care. Consistent with previous studies, AA men were more likely to receive radiation treatment and androgen-deprivation therapy (ADT) and were less likely to undergo surgery. Although the majority of men received guideline-based treatment (83.5%), and there were no overall differences according to race, AA men who had intermediate-risk disease were significantly less likely to receive guideline-based therapy (75.3% of AA men vs 85.9% of Caucasian men). In addition, men who had lymph node involvement or metastatic disease had significantly lower odds of receiving guideline-based therapy compared with men who had with low-risk disease. To assess the full meaning of the reported analyses, it is helpful to examine the data based on risk classification.
When comparing the type of treatment received, 17.8% of AA men received radiation therapy (RT) or brachytherapy compared with 7.2% of Caucasian men: a 2.5 fold difference. Conversely, fewer AA men underwent radical prostatectomy (RP) compared with Caucasian men. Approximately 11% of men received expectant management as primary treatment; whereas almost 12% of men received non-NCCN guideline-based therapy or no documented therapy. Differences in the receipt of various treatments may be because of physician and patient preferences, although potential biases also exist. What is clear from multiple national databases is that the disparity in receipt of RP and RT between AA and Caucasian men is persistent over several decades. It has been demonstrated that AA race, being unmarried, and having public/no insurance are associated with receiving nondefinitive treatment; and, among those who do receive definitive therapy, AA race and being unmarried are associated with receiving RT over RP. Overall survival, however, was associated with undergoing RP, younger age, and being married, among other factors. An additional consideration is whether all men with low-risk disease even require aggressive treatment. The 2010 NCCN guidelines addressed this aspect of prostate cancer management by including active surveillance as a treatment option for men with low-risk disease, updating the recommendations presented in 2007. The uptake of active surveillance among patients with low-risk disease and their physicians remains low, and the factors that impact its use should be further studied.
Ellis and colleagues demonstrate a significant difference in the receipt of NCCN guideline-based treatment for men with intermediate-risk cancer. The most striking finding from this group, as has been demonstrated over decades of population-based research, is that AA men are less likely to undergo RP and are more likely to receive RT with or without ADT. Whether this difference is clinically relevant is possibly best illustrated by the finding that stage migration has occurred in AA men and white men, whereas the mortality gap remains the same. Therefore, the penetrance of prostate-specific antigen screening has not trumped the effect of disparate treatment. The results from the study cannot clarify this issue, because there were only 31 patients in this subgroup, making it difficult to assess the results, even if the entire group is statistically significant. Clearly, however, the factors that predict receipt of treatment need to be further evaluated. In an abstract presented at the American Urological Association in 2012 by Underwood et al, the authors demonstrated that, among AA men, 16% believed exposure of cancer to “air” (surgery) spreads the disease, and an additional 33% stated that they did not know whether surgery was a risk for disease spread. To our knowledge, there are no physician-based studies that identify provider treatment-specific recommendations that vary by race. However, there is fertile ground for future investigations investigating the role of patient and provider in treatment decision-making.
High Risk, Very High Risk, Lymph Node Involvement, and Metastatic Disease
With small numbers of patients, it is difficult to reach a definitive conclusion regarding the role of treatment received within these groups. Anywhere from 8% to 73% of men received non-NCCN guideline-based therapy, indicating wide variation in treatment recommended and received. This may be because of physician training and comfort with dealing with more advanced disease. Another hypothesis, particularly for patients with lymph node involvement/metastatic disease, is that some clinicians will attempt aggressive therapy (RP with or without RT with or without ADT) in an attempt to obtain a favorable response in a small percentage of otherwise young, healthy patients. However, this approach has to be tempered with the high morbidity that is likely to be incurred for a very low chance of achieving a durable cure.
It may be argued that men with intermediate-risk prostate cancer are the most likely to derive an oncologic benefit from curative therapy that would outweigh the risk of decreased quality of life observed with the over treatment of men with low-risk cancer or the increased likelihood of morbidity from multimodality treatment or ADT necessary for men with high-risk cancer. Thus, if AA men with intermediate disease are less likely to receive appropriate care, then there could be a disparity in adverse prostate cancer outcomes, such as biochemical recurrence-free survival and cancer-specific survival. Furthermore, it also may be argued that men with metastatic disease have an even greater impetus to receive appropriate therapy, because their survival is potentially measured in months if not treated in a timely and correct manner. It is distressing, then, that groups that have been identified as intermediate-risk or high-risk for mortality are less likely to receive appropriate therapy.
Conclusions and Future Directions
Persistent differences in the type of care received by AA and Caucasian men merit continued attention. In addition, studies are needed regarding differences in actual receipt of treatment, regardless of modality. This is particularly true for patients like those in the study by Ellis et al, who expressed low levels of perceived access to care and may receive lower quality care as a result of this reduced access. These studies need to be sufficiently powered to detect meaningful differences, because smaller studies like that by Ellis and colleagues, although elegantly analyzed, may be skewed in subgroup analysis. Additional analyses to investigate the influence, exposure to culturally competent training, and practice experience of providers may be of value. Although it may be reassuring that the majority of men in their cohort received NCCN guideline-based treatment, these results should be interpreted in the context of the guidelines being based primarily on low-level evidence or expert opinion. The authors of the 2010 NCCN guidelines went so far as to state that prostate cancer is a complex disease that has many controversial aspects of management and a dearth of sound data to support treatment recommendations. Optimal treatment for prostate cancer is a moving target, with many physician-specific and patient-specific factors playing a significant role. Studies like the effort by Ellis et al will continue to shed light on areas of disparity and identify strategies for improving the care of men with prostate cancer.
No specific funding was disclosed.
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.
Viraj A. Master, MD, PhD1 and Kelvin A. Moses, MD, PhD2
1Department of Urology, Winship Cancer Institute, Emory University, Atlanta, Georgia; 2Division of Urology, Georgia Health Sciences University Cancer Center, Atlanta, Georgia
- 3Obese African-Americans with prostate cancer (T1c and a prostate-specific antigen, PSA, level of <10 ng/mL) have higher-risk pathological features and a greater risk of PSA recurrence than non-African-Americans. BJU Int. 2010;106:1157–1160., , , , .
- 15Air spreads cancer: so don't cut me open [abstract]. Paper presented at: Annual Meeting of the American Urological Association; May 19-23, 2012; Atlanta, GA. Abstract 1644., , .