- Email: [email protected]
Who Enrolls Onto Clinical Oncology Trials? A Radiation Patterns of Care Study Analysis
Int. J. Radiation Oncology Biol. Phys., Vol. 68, No. 4, pp. 1145–1150, 2007 Copyright © 2007 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/07/$–see front matter
doi:10.1016/j.ijrobp.2007.01.051
CLINICAL INVESTIGATION
Clinical Research
WHO ENROLLS ONTO CLINICAL ONCOLOGY TRIALS? A RADIATION PATTERNS OF CARE STUDY ANALYSIS BENJAMIN MOVSAS, M.D.,* JENNIFER MOUGHAN, M.S.,† JEAN OWEN, PH.D.,† LAWRENCE R. COIA, M.D.,‡ MICHAEL J. ZELEFSKY, M.D.,§ GERALD HANKS, M.D.,¶ 储 AND J. FRANK WILSON, M.D., F.A.C.R. *Department of Radiation Oncology, Henry Ford Health System, Detroit, MI; †American College of Radiology, Philadelphia, PA; Department of Radiation Oncology, Community Medical Center, Toms River, NJ; §Department of Radiation Oncology, Memorial Sloan-Kettering Cancer Center, New York, NY; ¶Department of Radiation Oncology, Fox Chase Cancer Center, Philadelphia, PA; and 储Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, WI
‡
Purpose: To identify factors significantly influencing accrual to clinical protocols by analyzing radiation Patterns of Care Study (PCS) surveys of 3,047 randomly selected radiotherapy (RT) patients. Methods and Materials: Patterns of Care Study surveys from disease sites studied for the periods 1992–1994 and 1996 –1999 (breast cancer, n ⴝ 1,080; prostate cancer, n ⴝ 1,149; esophageal cancer, n ⴝ 818) were analyzed. The PCS is a National Cancer Institute–funded national survey of randomly selected RT institutions in the United States. Patients with nonmetastatic disease who received RT as definitive or adjuvant therapy were randomly selected from eligible patients at each institution. To determine national estimates, individual patient records were weighted by the relative contribution of each institution and patients within each institution. Data regarding participation in clinical trials were recorded. The factors age, gender, race, type of insurance, and practice type of treating institution (academic or not) were studied by univariate and multivariate analyses. Results: Overall, only 2.7% of all patients were accrued to clinical protocols. Of these, 57% were enrolled on institutional review board–approved institutional trials, and 43% on National Cancer Institute collaborative group studies. On multivariate analysis, patients treated at academic facilities (p ⴝ 0.0001) and white patients (vs. African Americans, p ⴝ 0.0002) were significantly more likely to participate in clinical oncology trials. Age, gender, type of cancer, and type of insurance were not predictive. Conclusions: Practice type and race significantly influence enrollment onto clinical oncology trials. This suggests that increased communication and education regarding protocols, particularly focusing on physicians in nonacademic settings and minority patients, will be essential to enhance accrual. © 2007 Elsevier Inc. Clinical trials, Barriers to participation.
The overall participation of adult cancer patients in trials is exceedingly low at approximately 3% (1). This is far less than the greater than 60% of pediatric patients who are recruited onto studies (2). Yet our therapeutic decisions are based in large part on results from clinical oncology trials. In light of the above, it is critical to analyze the factors that influence accrual to adult clinical trials. The purpose of this study was to identify factors that significantly influence accrual to clinical oncology protocols by analyzing the Patterns of Care Study (PCS; now Quality Research in
Radiation Oncology) surveys of more than 3,000 radiotherapy (RT) patients. There is a paucity of data regarding issues specifically related to the accrual of RT patients to clinical trials (3). This study provides an important opportunity to analyze the accrual patterns and challenges in patients receiving RT. Moreover, most prior efforts to address the question of patient accrual to clinical oncology trials have compared the demographics of patients accrued to cancer trials to that of the burden of cancer in the population (4 – 6). The validity of comparing such disparate groups, though, is unclear. Unlike the other analyses, the
Reprint requests to: Benjamin Movsas, M.D., Department of Radiation Oncology, Henry Ford Health System, 2799 West Grand Blvd., Detroit, MI 48202. Tel: (313) 916-5188; Fax: (313) 9163235; E-Mail: [email protected] Supported by Grant No. CA 65435 from the National Cancer Institute, National Institutes of Health. Acknowledgments—The authors thank the radiation oncologists and staff at participating facilities for their support and cooperation, which remain essential to the Patterns of Care Study
(PCS); and PCS Research Associates Thomas Iarocci, M.D., M.S., Debora Grant, R.N., M.S.N., Ann Mann, M.H.S.A., R.T.(T), Andrea Trent, R.N., M.S.N., and Kimberly Roy, R.N., B.S.N., for collecting the survey data, Lisa Morabito for administrative support, and Dolores Malone for database design and administration. Conflict of interest: none. Received Nov 29, 2006 and in revised form Jan 19, 2007. Accepted for publication Jan 21, 2007.
INTRODUCTION
1145
1146
I. J. Radiation Oncology
●
Biology
●
Physics
current study represents surveys whose original purpose, using stratified random sampling, was to assess whether patients did or did not enroll onto clinical trials. METHODS AND MATERIALS Patterns of Care Study surveys from disease sites studied for patients treated during the periods 1992–1994 and 1996 –1999 were analyzed. These included 1,080 patients with breast cancer, 1,149 patients with prostate cancer, and 818 patients with esophageal cancer. The PCS is a National Cancer Institute (NCI)-funded national survey of randomly selected RT institutions in the United States. Patients with nonmetastatic disease who received RT as definitive or adjuvant therapy were randomly selected from eligible patients at each institution. To determine national estimates, the individual patient records were weighted by the relative contribution of each institution and each patient within each institution. Each disease within each stratum had unique weights. Accordingly, a weighted sample size of 133,060 was estimated from the combined surveys and diseases, in which 3,047 patient records were reviewed by trained research associates. Data regarding participation in clinical trials were recorded. The factors age, gender, race, type of insurance, and practice type of treating institution (academic or not) were studied by univariate and multivariate analyses. The PCS is a national survey conducted with a stratified random sample selected from a master list of RT facilities located throughout the United States. Stratified two-stage sampling was used to select institutions and cases for data collection. Radiotherapy facilities were divided into two strata: academic (main teaching hospital of a medical school or an NCI-designated Comprehensive Cancer Center) or not. In the 1992–1994 survey, there were 63 participating institutions, of which 21 were academic and 42 were nonacademic. In the 1996 –1999 survey, there were 59 participating institutions, of which 22 were academic and 37 were nonacademic. Patterns of Care Study research associates performed on-site chart reviews for each participating facility using a survey instrument developed for this purpose by the respective PCS committees. Data collected included demographic, diagnostic, staging, and therapeutic details. To be eligible for participation in these studies, patients had to meet several criteria, including treatment during the periods 1992–1994 or 1996 –1999, nonmetastatic cancer, and the use of RT as definitive or adjuvant therapy. The PCS Executive Committee adopted analysis methods that use SUDAAN statistical software (7), which incorporates the design elements (the two stages of stratification) and weights that reflect the relative contribution of each institution and each patient, in the analysis of this complex survey. The weights were the product of the two following factors: Total number of institutions in the stratum Total number of institutions sampled in the stratum
⫻
Estimated total eligible patients in the sampled institution Total number of eligible sampled patients in the sampled institution All figures calculated using SUDAAN are national estimates for the patient population defined by the eligibility criteria listed above. Three SUDAAN procedures were used to do this analysis:
Volume 68, Number 4, 2007
one for percentages and tests on categoric variables (PROC CROSSTAB), one for descriptive statistics and tests on continuous variables (PROC DESCRIPT), and one for univariate and multivariate logistic regression analyses that calculates odds ratios (PROC RLOGIST). Tests for association were performed using the Pearson chi-square test. Differences were deemed significant if the associated p values were ⬍0.05. A Bonferroni-adjusted p value was used if there were multiple comparisons, to control for experimental error that can result from multiple testing. To permit valid comparisons for types of patients on protocols in both surveys, the 1992–1994 and 1996 –1999 surveys were combined, and the weights and the design were adjusted accordingly. For specific analyses in which the interest was to assess a design factor (e.g., type of radiation therapy facility [academic vs. nonacademic]) as a covariate, the design was necessarily simplified: these analyses did not include these variables as design factors.
RESULTS The distributions of patient characteristics can be seen in Table 1. Overall, only 2.7% of all patients were accrued to clinical protocols, consistent with national estimates. Of these, 57% were enrolled on institutional review board– approved institutional trials and 43% on NCI collaborative group studies. Of the patients accrued to cooperative group trials, 14% of breast patients were on Radiation Therapy Oncology Group (RTOG) trials, whereas 100% of esophagus and prostate patients on cooperative groups were on RTOG trials. The RTOG group trials to which patients were enrolled were RTOG 8610, 9202, 9406, 9408, and 9805 for prostate cancer, and RTOG 9405 for esophageal cancer. For breast cancer, patients were enrolled on a broader array of collaborative group trials, including those from the Cancer and Leukemia Group B, Eastern Cooperative Oncology Group, North Central Cancer Treatment Group, National Surgical Adjuvant Breast and Bowel Project, RTOG, and South West Oncology Group. Although the details of the institutional trials are not known, these institutional trials had a higher rate of enrollment from academic institutions (63%) compared with the cooperative group trials (25%). Table 1 shows the relationship between patient characteristics and enrollment on clinical trial protocols. Of patients treated with RT at an academic practice, 9.3% were accrued to trials, compared with 1.7% at nonacademic sites (p ⫽ 0.002). Yet the vast majority of all patients (86%) were seen at nonacademic sites. White patients enrolled onto studies more than African Americans (2.8% vs. 0.8%, p ⫽ 0.013). Moreover, white patients treated with radiation at an academic practice were accrued to studies more than white patients treated with radiation at nonacademic sites (9.1% vs. 1.9%, p ⫽ 0.01). Similarly, African American patients treated with radiation at an academic practice were accrued to trials more than African American patients treated at nonacademic sites (3.9% vs. 0, p ⫽ 0.001). A higher percentage of patients with esophageal cancer (9.8%) were accrued compared with either breast cancer (2.2%) or prostate cancer (2.0%) (p ⫽ 0.072). Younger age demonstrated a trend toward increased enrollment (p ⫽ 0.05). However,
Who enrolls onto clinical oncology trials?
●
B. MOVSAS et al.
1147
Table 1. Relationship between patient characteristics and enrollment on clinical trial protocols Yes on protocol Characteristic Total Type of protocol IRB-approved institutional trials Collaborative group clinical trials† Age (y)‡ Mean Range Median Gender Male Female Race White African American or Black Other Primary payment method Medicare HMO/Medicare HMO BCBS/other insurance Patient/family/Medicaid/MediCal/Champus/VA/Unknown Type of facility Academic Nonacademic Cancer Breast Esophagus Prostate Survey years 1992–1994 1996–1999
Not on protocol
Unweighted sample size
Weighted sample size
%*
Unweighted sample size
138
3,597
2.7
2,909
100 38†
2,050 1,547
57.0 43.0
59.5 28–78 58
Weighted sample size 129,463
%*
p
97.3
63.2 24–91 66
0.0524 0.4067§
90 48
1,708 1,889
3.3 2.3
1,681 1,228
49,339 80,124
96.7 97.7 0.0534§
115 16 7
3,192 104 301
2.8 0.8 3.7
2,297 429 183
109,404 12,123 7,936
97.2 99.2 96.3 0.1739§
26 13 62 37
1,069 271 1,553 704
1.9 1.8 3.2 5.1
1,213 282 859 555
55,175 14,663 46,652 12,972
98.1 98.2 96.8 94.9 0.0022§
111 27
1,676 1,921
9.3 1.7
1,096 1,813
16,309 113,154
90.7 98.3
31 85 22
1,728 1,012 857
2.2 9.8 2.0
1,049 733 1,127
77,698 9,295 42,470
97.8 90.2 98.0
69 69
1,072‡ 2,525‡
1.5 4.2
1,657 1,252
72,132储 57,331储
98.5 95.8
0.0718§
0.0428§
Abbreviations: IRB ⫽ institutional review board; HMO ⫽ health maintenance organization; BCBS ⫽ Blue Cross/Blue Shield; VA ⫽ Veterans Administration. * Percentages are based on the weighted number of patients. † Collaborative group clinical trials ⫽ 11 patients on National Surgical Adjuvant Breast and Bowel Project trials, 2 patients on South West Oncology Group trials, 4 patients on North Central Cancer Treatment Group trials, 3 patients on Cancer and Leukemia Group B trials, 2 patients on Eastern Cooperative Oncology Group trials, and 16 patients on Radiation Therapy Oncology Group trials. ‡ Excludes 25 patients whose ages are missing. § This p value is from comparing all levels of the variable with whether or not a patient is enrolled on a protocol. 储 These are the weighted totals when the surveys are combined in the analysis.
gender (male 3.3% vs. female 2.3%, p ⫽ 0.41) and type of insurance were not predictive (Medicare 1.9%, health maintenance organization 1.8%, private [e.g., Blue Cross/Blue Shield] 3.2%, other 5.1%, p ⫽ 0.17). Increased accrual over time was observed (1.5% for 1992–1994 vs. 4.2% for 1996 –1999, p ⫽ 0.043). Univariate logistic regression models were performed to further investigate for those variables showing association with enrollment on clinical trial protocols (Table 2). The data in Table 2 further verify that older people are less likely to enroll on trials, as well as African Americans, and patients treated at nonacademic facilities. A patient’s gender and type of insurance had no statistically significant association with enrollment on trials. On multivariate analysis, patients treated at academic facilities (p ⫽ 0.0001) and
white patients (vs. African Americans, p ⫽ 0.0002) were highly associated with enrollment on clinical oncology trials after adjusting for the other variables in the model (Table 3). DISCUSSION In this study, the type of practice (academic vs. not) and race were the most significant factors influencing enrollment onto clinical oncology trials. Similarly, Sateren et al. (4) reported that black men (as well as other minorities) were accrued to clinical trials at lower rates than white cancer patients. Moreover, they found that the number of oncologists and presence of approved cancer programs were significantly associated with increased accrual to clinical trials. Unlike the present study, however, Sateren et al. (4)
1148
I. J. Radiation Oncology
●
Biology
●
Physics
Table 2. Univariate logistic regression analysis identifying variables associated with enrollment on clinical trial protocols Yes on protocol Characteristic Age* Continuous Gender Male Female Race White African American or Black Other Primary payment method Medicare HMO/Medicare HMO BCBS/other insurance Patient/family/Medicaid/ Medi-Cal/Champus/VA/ unknown Type of facility Academic Nonacademic
Odds ratio
95% CI
p
0.98
0.96–1.00
0.0479
Reference 0.68
0.28–1.67
0.3997
Reference 0.29 1.30
0.11–0.76 0.31–5.47
0.0112 0.7188
0.58 0.55 Reference 1.63
0.27–1.26 0.21–1.49
0.1688 0.2411
0.69–3.84
0.2625
Reference 0.17
0.07–0.41
0.0001
Abbreviations: CI ⫽ confidence interval; HMO ⫽ health maintenance organization; BCBS ⫽ Blue Cross/Blue Shield; VA ⫽ Veterans Administration. Survey years and type of cancer were maintained as design factors. * Excludes 25 patients whose ages are missing.
found that patients enrolled onto clinical trials were significantly less likely to be uninsured and more likely to have Medicare insurance. The NCI recognized the need for increased involvement of cancer specialists in clinical trial accrual with the initiation of the Community Clinical Oncology Program (CCOP) in 1983. However, participating physicians had to be affiliated with a hospital (or a group of hospitals/clinics) to successfully compete for a CCOP grant. The Clinical Trial Support Unit is a new program designed to broaden the base of community and clinical investigators. After registering on-line (www.ctsu.org), even individual oncologists who do not have any CCOP or cooperative group affiliation can participate in high-priority cooperative group studies. At the same time, for such individual investigators, access to adequate resources for data collection, management, and research nursing support remains an ongoing challenge. The Eastern Cooperative Oncology Group conducted a study of barriers to African American accrual onto clinical trials (5). As part of the study with the National Medical Association, an organization of approximately 22,000 African American physicians, they found that approximately 70% of National Medical Association physicians cited mistrust of research centers and fear of losing patients as the most important barriers to minority cancer patient referrals. Also cited was lack of information about specific trials, as well as lack of minority
Volume 68, Number 4, 2007
investigators (5). Although back in 1994 the NIH issued guidelines for the inclusion of women and minorities as subjects in clinical research, there seems to be an ongoing challenge regarding minority accrual (8, 9). Hutchins et al. (6) analyzed data from ⬎16,000 cancer patients enrolled on 164 Southwest Oncology Group trials between 1993 and 1996 by gender, race, and age. These rates were compared with the corresponding rates in the U.S. population of patients with cancer. They found that patients aged ⱖ65 years were significantly underrepresented (25% vs. 63%, p ⬍ 0.001) (6). Yee et al. (10) evaluated the enrollment of older patients (age ⱖ65 years) in Canadian Cancer Treatment trials. The Canadian health care system provides reimbursement for health care costs regardless of whether the patient is participating in a clinical trial. Yee et al. found that older patients (age ⱖ65 years) accounted for 22% of trial enrollees, compared with 58% of the Canadian population of cancer, suggesting that age remained a barrier for accrual even though reimbursement was not an issue (10). Likewise, Gross et al. (11) reported that revision of the Medicare program in 2000 (to cover costs for patients in clinical trials) was not associated with a significant increase in the accrual of older patients to cancer trials. Similarly, in our analysis, we found that the type of medical insurance did not significantly impact on the rate of accrual to clinical oncology trials. Although age was significant in our study on univariate analysis, it was not
Table 3. Multivariate logistic regression analysis identifying variables associated with enrollment on clinical trial protocols Yes on protocol Characteristic Age* Continuous Gender Male Female Race White African American or Black Other Primary payment method Medicare HMO/Medicare HMO BCBS/other insurance Patient/family/Medicaid/ Medi-Cal/Champus/VA/ unknown Type of facility Academic Nonacademic
Odds ratio
95% CI
p
0.97
0.93–1.01
0.1164
Reference 0.41
0.13–1.27
0.1209
Reference 0.18 1.16
0.08–0.44 0.28–4.85
0.0002 0.8425
0.88 0.58 Reference 1.08
0.29–2.63 0.23–1.47
0.8137 0.2492
0.46–2.55
0.8615
Reference 0.18
0.07–0.43
0.0001
Abbreviations: CI ⫽ confidence interval; HMO ⫽ health maintenance organization; BCBS ⫽ Blue Cross/Blue Shield; VA ⫽ Veterans Administration. Survey years and type of cancer were maintained as design factors. * Excludes 25 patients whose ages are missing.
Who enrolls onto clinical oncology trials?
independently predictive on multivariate analysis. This suggests that there are more important factors than age alone that influence accrual to trials. Indeed, in our study, the majority of elderly patients were evaluated in nonacademic facilities (90% of patients aged ⱖ65 years were treated at nonacademic centers), a factor that was significant on multivariate analysis. Is it possible that elderly patients are less willing to travel to a distant academic center? Other factors likely play a role, such as underlying comorbidities, as well as patient and physician attitudes (12, 13). For example, Lewis et al. (14) reported that elderly patients were more likely to have medical histories that made them ineligible for clinical trials because of protocol exclusions. Yet Kemeny et al. (15) suggested that the greatest impediment to enrolling older women with breast cancer onto clinical trials was the physicians’ perceptions about age and tolerance of toxicity. There is inadequate evidence to indicate that older cancer patients, who are otherwise favorable, have increased toxicities compared with younger patients. More studies are needed on age-related changes and organ function, the impact of comorbidities, and pharmacokinetic changes of age. By 2030, the number of persons in the United States older than 65 years will double, and the number older than 85 years will quadruple. Although the specialties of the accruing physicians are not known, because this is a radiation PCS it is not too surprising that of the patients with prostrate and esophageal cancer accrued to cooperative group trials, all were enrolled on RTOG trials. The RTOG is an NCI-sponsored cooperative group focusing primarily on radiation-related issues. The fact that only 14% of breast cancer patients were enrolled onto RTOG studies may be owing to the relative paucity of RTOG breast cancer trials available during this study period compared with other cooperative groups. Recently, the RTOG breast committee has increased the number of trials available. This finding may also reflect a greater prevalence of multidisciplinary tumor boards for breast cancer patients and/or the patterns of referral of these patients to other specialties before radiation oncology. Unlike enrollment onto institutional trials, which, as expected, was primarily from academic institutions (63%), 75% of the accrual to cooperative group trials came from nonacademic institutions. These results suggest that, at least
●
B. MOVSAS et al.
1149
for certain disease sites, the radiation oncologist, both in academics and in the community, could serve as a critical “gatekeeper” to help accrue patients to clinical oncology trials. For all disease sites, this process should be done in the context of a multidisciplinary evaluation or tumor board. As a key part of this collaborative team, radiation oncologists should feel empowered to offer relevant clinical oncology trials to their patients. Although it is encouraging that there has been some increase in accrual during the 1990s (1.5% during 1992–1994 vs. 4.2% during 1996 –1999), the reasons underlying this trend are unclear and further improvement is critical. The present study has several inherent limitations. Similar to other studies, it does not include socioeconomic factors (other than insurance) or distance from cancer centers. Other factors that are likely important and have not been studied in most of these analyses include patient attitudes and knowledge base that may influence their willingness to participate (16, 17). Comis et al. (18) recently conducted interviews to better understand the attitudes of American adults toward participation in cancer clinical trials. Interestingly, their findings suggest that the primary barrier to accrual is not the attitude of patients but rather the unavailability of appropriate clinical trials and/or disqualification based on eligibility criteria, as well as the reluctance of some physicians to engage in accrual (18). Moreover, this analysis can not directly determine the eligibility of individual patients for a particular trial and/or whether such trials were actually available at particular institutions. Such limitations similarly apply to other analyses, in which the accrual to cooperative group trials was generally compared with national averages, despite the fact that the validity of such a comparison is unclear (4 – 6). The strength of the present study, unlike the other analyses, is that it represents surveys whose original purpose, using stratified random sampling, was to assess the patterns of care, which included whether patients did or did not enroll onto clinical trials. In conclusion, this study demonstrates that practice type and race significantly influence enrollment on clinical oncology trials. This suggests that increased communication and education regarding protocols, particularly focusing on physicians in nonacademic settings and minority patients, will be essential to enhance accrual.
REFERENCES 1. Taylor KM, Feldstein ML, Skeel RT, et al. Fundamental dilemmas of the randomized clinical trial process: Results of a survey of the 1,737 Eastern Cooperative Oncology Group investigators. J Clin Oncol 1994;12:1796 –1805. 2. Lukens JN. Progress resulting from clinical trials. Solid tumors in childhood cancer. Cancer 1994;74(9 Suppl.):2710 – 2718. 3. Lee JY, Breaux SR. Accrual of radiotherapy patients to clinical trials. Cancer 1983;52:1014 –1016. 4. Sateren WB, Trimble EL, Abrams J, et al. How sociodemographics, presence of oncology specialists, and hospital cancer programs affect accrual to cancer treatment trials. J Clin Oncol 2002;20:2109 –2117.
5. McCaskill-Stevens W, Pinto H, Marcus AC, et al. Recruiting minority cancer patients into cancer clinical trials: A pilot project involving the Eastern Cooperative Oncology Group and the National Medical Association. J Clin Oncol 1999;17: 1029 –1039. 6. Hutchins LF, Unger JM, Crowley JJ, et al. Underrepresentation of patients 65 years of age or older in cancer-treatment trials. N Engl J Med 1999;341:2061–2067. 7. Research Triangle Institute. SUDAAN user’s manual. 8.0 ed. Research Triangle Park, NC: Research Triangle Institute; 2001. 8. Advani AS, Atkeson B, Brown CL, et al. Barriers to the participation of African-American patients with cancer in clinical trials: A pilot study. Cancer 2003;97:1499 –1506.
1150
I. J. Radiation Oncology
●
Biology
●
Physics
9. Adams-Campbell LL, Ahaghotu C, Gaskins M, et al. Enrollment of African Americans onto clinical treatment trials: Study design barriers. J Clin Oncol 2004;22:730 –734. 10. Yee KW, Pater JL, Pho L, et al. Enrollment of older patients in cancer treatment trials in Canada: Why is age a barrier? J Clin Oncol 2003;21:1618 –1623. 11. Gross CP, Wong N, Dubin JA, et al. Enrollment of older persons in cancer trials after the Medicare reimbursement policy change. Arch Intern Med 2005;165:1514 –1520. 12. Townsley CA, Selby R, Siu LL. Systematic review of barriers to the recruitment of older patients with cancer onto clinical trials. J Clin Oncol 2005;23:3112–3124. 13. Townsley CA, Chan KK, Pond GR, et al. Understanding the attitudes of the elderly towards enrollment into cancer clinical trials. BMC Cancer 2006;6:34.
Volume 68, Number 4, 2007
14. Lewis JH, Kilgore ML, Goldman DP, et al. Participation of patients 65 years of age or older in cancer clinical trials. J Clin Oncol 2003;21:1383–1389. 15. Kemeny MM, Peterson BL, Kornblith AB, et al. Barriers to clinical trial participation by older women with breast cancer. J Clin Oncol 2003;21:2268 –2275. 16. Ellis PM, Butow PN, Tattersall MH, et al. Randomized clinical trials in oncology: Understanding and attitudes predict willingness to participate. J Clin Oncol 2001;19:3554 –3561. 17. Mills EJ, Seely D, Rachlis B, et al. Barriers to participation in clinical trials of cancer: A meta-analysis and systematic review of patient-reported factors. Lancet Oncol 2006;7:141–148. 18. Comis RL, Miller JD, Aldige CR, et al. Public attitudes toward participation in cancer clinical trials. J Clin Oncol 2003;21:830 – 835.
doi:10.1016/j.ijrobp.2007.01.051
CLINICAL INVESTIGATION
Clinical Research
WHO ENROLLS ONTO CLINICAL ONCOLOGY TRIALS? A RADIATION PATTERNS OF CARE STUDY ANALYSIS BENJAMIN MOVSAS, M.D.,* JENNIFER MOUGHAN, M.S.,† JEAN OWEN, PH.D.,† LAWRENCE R. COIA, M.D.,‡ MICHAEL J. ZELEFSKY, M.D.,§ GERALD HANKS, M.D.,¶ 储 AND J. FRANK WILSON, M.D., F.A.C.R. *Department of Radiation Oncology, Henry Ford Health System, Detroit, MI; †American College of Radiology, Philadelphia, PA; Department of Radiation Oncology, Community Medical Center, Toms River, NJ; §Department of Radiation Oncology, Memorial Sloan-Kettering Cancer Center, New York, NY; ¶Department of Radiation Oncology, Fox Chase Cancer Center, Philadelphia, PA; and 储Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, WI
‡
Purpose: To identify factors significantly influencing accrual to clinical protocols by analyzing radiation Patterns of Care Study (PCS) surveys of 3,047 randomly selected radiotherapy (RT) patients. Methods and Materials: Patterns of Care Study surveys from disease sites studied for the periods 1992–1994 and 1996 –1999 (breast cancer, n ⴝ 1,080; prostate cancer, n ⴝ 1,149; esophageal cancer, n ⴝ 818) were analyzed. The PCS is a National Cancer Institute–funded national survey of randomly selected RT institutions in the United States. Patients with nonmetastatic disease who received RT as definitive or adjuvant therapy were randomly selected from eligible patients at each institution. To determine national estimates, individual patient records were weighted by the relative contribution of each institution and patients within each institution. Data regarding participation in clinical trials were recorded. The factors age, gender, race, type of insurance, and practice type of treating institution (academic or not) were studied by univariate and multivariate analyses. Results: Overall, only 2.7% of all patients were accrued to clinical protocols. Of these, 57% were enrolled on institutional review board–approved institutional trials, and 43% on National Cancer Institute collaborative group studies. On multivariate analysis, patients treated at academic facilities (p ⴝ 0.0001) and white patients (vs. African Americans, p ⴝ 0.0002) were significantly more likely to participate in clinical oncology trials. Age, gender, type of cancer, and type of insurance were not predictive. Conclusions: Practice type and race significantly influence enrollment onto clinical oncology trials. This suggests that increased communication and education regarding protocols, particularly focusing on physicians in nonacademic settings and minority patients, will be essential to enhance accrual. © 2007 Elsevier Inc. Clinical trials, Barriers to participation.
The overall participation of adult cancer patients in trials is exceedingly low at approximately 3% (1). This is far less than the greater than 60% of pediatric patients who are recruited onto studies (2). Yet our therapeutic decisions are based in large part on results from clinical oncology trials. In light of the above, it is critical to analyze the factors that influence accrual to adult clinical trials. The purpose of this study was to identify factors that significantly influence accrual to clinical oncology protocols by analyzing the Patterns of Care Study (PCS; now Quality Research in
Radiation Oncology) surveys of more than 3,000 radiotherapy (RT) patients. There is a paucity of data regarding issues specifically related to the accrual of RT patients to clinical trials (3). This study provides an important opportunity to analyze the accrual patterns and challenges in patients receiving RT. Moreover, most prior efforts to address the question of patient accrual to clinical oncology trials have compared the demographics of patients accrued to cancer trials to that of the burden of cancer in the population (4 – 6). The validity of comparing such disparate groups, though, is unclear. Unlike the other analyses, the
Reprint requests to: Benjamin Movsas, M.D., Department of Radiation Oncology, Henry Ford Health System, 2799 West Grand Blvd., Detroit, MI 48202. Tel: (313) 916-5188; Fax: (313) 9163235; E-Mail: [email protected] Supported by Grant No. CA 65435 from the National Cancer Institute, National Institutes of Health. Acknowledgments—The authors thank the radiation oncologists and staff at participating facilities for their support and cooperation, which remain essential to the Patterns of Care Study
(PCS); and PCS Research Associates Thomas Iarocci, M.D., M.S., Debora Grant, R.N., M.S.N., Ann Mann, M.H.S.A., R.T.(T), Andrea Trent, R.N., M.S.N., and Kimberly Roy, R.N., B.S.N., for collecting the survey data, Lisa Morabito for administrative support, and Dolores Malone for database design and administration. Conflict of interest: none. Received Nov 29, 2006 and in revised form Jan 19, 2007. Accepted for publication Jan 21, 2007.
INTRODUCTION
1145
1146
I. J. Radiation Oncology
●
Biology
●
Physics
current study represents surveys whose original purpose, using stratified random sampling, was to assess whether patients did or did not enroll onto clinical trials. METHODS AND MATERIALS Patterns of Care Study surveys from disease sites studied for patients treated during the periods 1992–1994 and 1996 –1999 were analyzed. These included 1,080 patients with breast cancer, 1,149 patients with prostate cancer, and 818 patients with esophageal cancer. The PCS is a National Cancer Institute (NCI)-funded national survey of randomly selected RT institutions in the United States. Patients with nonmetastatic disease who received RT as definitive or adjuvant therapy were randomly selected from eligible patients at each institution. To determine national estimates, the individual patient records were weighted by the relative contribution of each institution and each patient within each institution. Each disease within each stratum had unique weights. Accordingly, a weighted sample size of 133,060 was estimated from the combined surveys and diseases, in which 3,047 patient records were reviewed by trained research associates. Data regarding participation in clinical trials were recorded. The factors age, gender, race, type of insurance, and practice type of treating institution (academic or not) were studied by univariate and multivariate analyses. The PCS is a national survey conducted with a stratified random sample selected from a master list of RT facilities located throughout the United States. Stratified two-stage sampling was used to select institutions and cases for data collection. Radiotherapy facilities were divided into two strata: academic (main teaching hospital of a medical school or an NCI-designated Comprehensive Cancer Center) or not. In the 1992–1994 survey, there were 63 participating institutions, of which 21 were academic and 42 were nonacademic. In the 1996 –1999 survey, there were 59 participating institutions, of which 22 were academic and 37 were nonacademic. Patterns of Care Study research associates performed on-site chart reviews for each participating facility using a survey instrument developed for this purpose by the respective PCS committees. Data collected included demographic, diagnostic, staging, and therapeutic details. To be eligible for participation in these studies, patients had to meet several criteria, including treatment during the periods 1992–1994 or 1996 –1999, nonmetastatic cancer, and the use of RT as definitive or adjuvant therapy. The PCS Executive Committee adopted analysis methods that use SUDAAN statistical software (7), which incorporates the design elements (the two stages of stratification) and weights that reflect the relative contribution of each institution and each patient, in the analysis of this complex survey. The weights were the product of the two following factors: Total number of institutions in the stratum Total number of institutions sampled in the stratum
⫻
Estimated total eligible patients in the sampled institution Total number of eligible sampled patients in the sampled institution All figures calculated using SUDAAN are national estimates for the patient population defined by the eligibility criteria listed above. Three SUDAAN procedures were used to do this analysis:
Volume 68, Number 4, 2007
one for percentages and tests on categoric variables (PROC CROSSTAB), one for descriptive statistics and tests on continuous variables (PROC DESCRIPT), and one for univariate and multivariate logistic regression analyses that calculates odds ratios (PROC RLOGIST). Tests for association were performed using the Pearson chi-square test. Differences were deemed significant if the associated p values were ⬍0.05. A Bonferroni-adjusted p value was used if there were multiple comparisons, to control for experimental error that can result from multiple testing. To permit valid comparisons for types of patients on protocols in both surveys, the 1992–1994 and 1996 –1999 surveys were combined, and the weights and the design were adjusted accordingly. For specific analyses in which the interest was to assess a design factor (e.g., type of radiation therapy facility [academic vs. nonacademic]) as a covariate, the design was necessarily simplified: these analyses did not include these variables as design factors.
RESULTS The distributions of patient characteristics can be seen in Table 1. Overall, only 2.7% of all patients were accrued to clinical protocols, consistent with national estimates. Of these, 57% were enrolled on institutional review board– approved institutional trials and 43% on NCI collaborative group studies. Of the patients accrued to cooperative group trials, 14% of breast patients were on Radiation Therapy Oncology Group (RTOG) trials, whereas 100% of esophagus and prostate patients on cooperative groups were on RTOG trials. The RTOG group trials to which patients were enrolled were RTOG 8610, 9202, 9406, 9408, and 9805 for prostate cancer, and RTOG 9405 for esophageal cancer. For breast cancer, patients were enrolled on a broader array of collaborative group trials, including those from the Cancer and Leukemia Group B, Eastern Cooperative Oncology Group, North Central Cancer Treatment Group, National Surgical Adjuvant Breast and Bowel Project, RTOG, and South West Oncology Group. Although the details of the institutional trials are not known, these institutional trials had a higher rate of enrollment from academic institutions (63%) compared with the cooperative group trials (25%). Table 1 shows the relationship between patient characteristics and enrollment on clinical trial protocols. Of patients treated with RT at an academic practice, 9.3% were accrued to trials, compared with 1.7% at nonacademic sites (p ⫽ 0.002). Yet the vast majority of all patients (86%) were seen at nonacademic sites. White patients enrolled onto studies more than African Americans (2.8% vs. 0.8%, p ⫽ 0.013). Moreover, white patients treated with radiation at an academic practice were accrued to studies more than white patients treated with radiation at nonacademic sites (9.1% vs. 1.9%, p ⫽ 0.01). Similarly, African American patients treated with radiation at an academic practice were accrued to trials more than African American patients treated at nonacademic sites (3.9% vs. 0, p ⫽ 0.001). A higher percentage of patients with esophageal cancer (9.8%) were accrued compared with either breast cancer (2.2%) or prostate cancer (2.0%) (p ⫽ 0.072). Younger age demonstrated a trend toward increased enrollment (p ⫽ 0.05). However,
Who enrolls onto clinical oncology trials?
●
B. MOVSAS et al.
1147
Table 1. Relationship between patient characteristics and enrollment on clinical trial protocols Yes on protocol Characteristic Total Type of protocol IRB-approved institutional trials Collaborative group clinical trials† Age (y)‡ Mean Range Median Gender Male Female Race White African American or Black Other Primary payment method Medicare HMO/Medicare HMO BCBS/other insurance Patient/family/Medicaid/MediCal/Champus/VA/Unknown Type of facility Academic Nonacademic Cancer Breast Esophagus Prostate Survey years 1992–1994 1996–1999
Not on protocol
Unweighted sample size
Weighted sample size
%*
Unweighted sample size
138
3,597
2.7
2,909
100 38†
2,050 1,547
57.0 43.0
59.5 28–78 58
Weighted sample size 129,463
%*
p
97.3
63.2 24–91 66
0.0524 0.4067§
90 48
1,708 1,889
3.3 2.3
1,681 1,228
49,339 80,124
96.7 97.7 0.0534§
115 16 7
3,192 104 301
2.8 0.8 3.7
2,297 429 183
109,404 12,123 7,936
97.2 99.2 96.3 0.1739§
26 13 62 37
1,069 271 1,553 704
1.9 1.8 3.2 5.1
1,213 282 859 555
55,175 14,663 46,652 12,972
98.1 98.2 96.8 94.9 0.0022§
111 27
1,676 1,921
9.3 1.7
1,096 1,813
16,309 113,154
90.7 98.3
31 85 22
1,728 1,012 857
2.2 9.8 2.0
1,049 733 1,127
77,698 9,295 42,470
97.8 90.2 98.0
69 69
1,072‡ 2,525‡
1.5 4.2
1,657 1,252
72,132储 57,331储
98.5 95.8
0.0718§
0.0428§
Abbreviations: IRB ⫽ institutional review board; HMO ⫽ health maintenance organization; BCBS ⫽ Blue Cross/Blue Shield; VA ⫽ Veterans Administration. * Percentages are based on the weighted number of patients. † Collaborative group clinical trials ⫽ 11 patients on National Surgical Adjuvant Breast and Bowel Project trials, 2 patients on South West Oncology Group trials, 4 patients on North Central Cancer Treatment Group trials, 3 patients on Cancer and Leukemia Group B trials, 2 patients on Eastern Cooperative Oncology Group trials, and 16 patients on Radiation Therapy Oncology Group trials. ‡ Excludes 25 patients whose ages are missing. § This p value is from comparing all levels of the variable with whether or not a patient is enrolled on a protocol. 储 These are the weighted totals when the surveys are combined in the analysis.
gender (male 3.3% vs. female 2.3%, p ⫽ 0.41) and type of insurance were not predictive (Medicare 1.9%, health maintenance organization 1.8%, private [e.g., Blue Cross/Blue Shield] 3.2%, other 5.1%, p ⫽ 0.17). Increased accrual over time was observed (1.5% for 1992–1994 vs. 4.2% for 1996 –1999, p ⫽ 0.043). Univariate logistic regression models were performed to further investigate for those variables showing association with enrollment on clinical trial protocols (Table 2). The data in Table 2 further verify that older people are less likely to enroll on trials, as well as African Americans, and patients treated at nonacademic facilities. A patient’s gender and type of insurance had no statistically significant association with enrollment on trials. On multivariate analysis, patients treated at academic facilities (p ⫽ 0.0001) and
white patients (vs. African Americans, p ⫽ 0.0002) were highly associated with enrollment on clinical oncology trials after adjusting for the other variables in the model (Table 3). DISCUSSION In this study, the type of practice (academic vs. not) and race were the most significant factors influencing enrollment onto clinical oncology trials. Similarly, Sateren et al. (4) reported that black men (as well as other minorities) were accrued to clinical trials at lower rates than white cancer patients. Moreover, they found that the number of oncologists and presence of approved cancer programs were significantly associated with increased accrual to clinical trials. Unlike the present study, however, Sateren et al. (4)
1148
I. J. Radiation Oncology
●
Biology
●
Physics
Table 2. Univariate logistic regression analysis identifying variables associated with enrollment on clinical trial protocols Yes on protocol Characteristic Age* Continuous Gender Male Female Race White African American or Black Other Primary payment method Medicare HMO/Medicare HMO BCBS/other insurance Patient/family/Medicaid/ Medi-Cal/Champus/VA/ unknown Type of facility Academic Nonacademic
Odds ratio
95% CI
p
0.98
0.96–1.00
0.0479
Reference 0.68
0.28–1.67
0.3997
Reference 0.29 1.30
0.11–0.76 0.31–5.47
0.0112 0.7188
0.58 0.55 Reference 1.63
0.27–1.26 0.21–1.49
0.1688 0.2411
0.69–3.84
0.2625
Reference 0.17
0.07–0.41
0.0001
Abbreviations: CI ⫽ confidence interval; HMO ⫽ health maintenance organization; BCBS ⫽ Blue Cross/Blue Shield; VA ⫽ Veterans Administration. Survey years and type of cancer were maintained as design factors. * Excludes 25 patients whose ages are missing.
found that patients enrolled onto clinical trials were significantly less likely to be uninsured and more likely to have Medicare insurance. The NCI recognized the need for increased involvement of cancer specialists in clinical trial accrual with the initiation of the Community Clinical Oncology Program (CCOP) in 1983. However, participating physicians had to be affiliated with a hospital (or a group of hospitals/clinics) to successfully compete for a CCOP grant. The Clinical Trial Support Unit is a new program designed to broaden the base of community and clinical investigators. After registering on-line (www.ctsu.org), even individual oncologists who do not have any CCOP or cooperative group affiliation can participate in high-priority cooperative group studies. At the same time, for such individual investigators, access to adequate resources for data collection, management, and research nursing support remains an ongoing challenge. The Eastern Cooperative Oncology Group conducted a study of barriers to African American accrual onto clinical trials (5). As part of the study with the National Medical Association, an organization of approximately 22,000 African American physicians, they found that approximately 70% of National Medical Association physicians cited mistrust of research centers and fear of losing patients as the most important barriers to minority cancer patient referrals. Also cited was lack of information about specific trials, as well as lack of minority
Volume 68, Number 4, 2007
investigators (5). Although back in 1994 the NIH issued guidelines for the inclusion of women and minorities as subjects in clinical research, there seems to be an ongoing challenge regarding minority accrual (8, 9). Hutchins et al. (6) analyzed data from ⬎16,000 cancer patients enrolled on 164 Southwest Oncology Group trials between 1993 and 1996 by gender, race, and age. These rates were compared with the corresponding rates in the U.S. population of patients with cancer. They found that patients aged ⱖ65 years were significantly underrepresented (25% vs. 63%, p ⬍ 0.001) (6). Yee et al. (10) evaluated the enrollment of older patients (age ⱖ65 years) in Canadian Cancer Treatment trials. The Canadian health care system provides reimbursement for health care costs regardless of whether the patient is participating in a clinical trial. Yee et al. found that older patients (age ⱖ65 years) accounted for 22% of trial enrollees, compared with 58% of the Canadian population of cancer, suggesting that age remained a barrier for accrual even though reimbursement was not an issue (10). Likewise, Gross et al. (11) reported that revision of the Medicare program in 2000 (to cover costs for patients in clinical trials) was not associated with a significant increase in the accrual of older patients to cancer trials. Similarly, in our analysis, we found that the type of medical insurance did not significantly impact on the rate of accrual to clinical oncology trials. Although age was significant in our study on univariate analysis, it was not
Table 3. Multivariate logistic regression analysis identifying variables associated with enrollment on clinical trial protocols Yes on protocol Characteristic Age* Continuous Gender Male Female Race White African American or Black Other Primary payment method Medicare HMO/Medicare HMO BCBS/other insurance Patient/family/Medicaid/ Medi-Cal/Champus/VA/ unknown Type of facility Academic Nonacademic
Odds ratio
95% CI
p
0.97
0.93–1.01
0.1164
Reference 0.41
0.13–1.27
0.1209
Reference 0.18 1.16
0.08–0.44 0.28–4.85
0.0002 0.8425
0.88 0.58 Reference 1.08
0.29–2.63 0.23–1.47
0.8137 0.2492
0.46–2.55
0.8615
Reference 0.18
0.07–0.43
0.0001
Abbreviations: CI ⫽ confidence interval; HMO ⫽ health maintenance organization; BCBS ⫽ Blue Cross/Blue Shield; VA ⫽ Veterans Administration. Survey years and type of cancer were maintained as design factors. * Excludes 25 patients whose ages are missing.
Who enrolls onto clinical oncology trials?
independently predictive on multivariate analysis. This suggests that there are more important factors than age alone that influence accrual to trials. Indeed, in our study, the majority of elderly patients were evaluated in nonacademic facilities (90% of patients aged ⱖ65 years were treated at nonacademic centers), a factor that was significant on multivariate analysis. Is it possible that elderly patients are less willing to travel to a distant academic center? Other factors likely play a role, such as underlying comorbidities, as well as patient and physician attitudes (12, 13). For example, Lewis et al. (14) reported that elderly patients were more likely to have medical histories that made them ineligible for clinical trials because of protocol exclusions. Yet Kemeny et al. (15) suggested that the greatest impediment to enrolling older women with breast cancer onto clinical trials was the physicians’ perceptions about age and tolerance of toxicity. There is inadequate evidence to indicate that older cancer patients, who are otherwise favorable, have increased toxicities compared with younger patients. More studies are needed on age-related changes and organ function, the impact of comorbidities, and pharmacokinetic changes of age. By 2030, the number of persons in the United States older than 65 years will double, and the number older than 85 years will quadruple. Although the specialties of the accruing physicians are not known, because this is a radiation PCS it is not too surprising that of the patients with prostrate and esophageal cancer accrued to cooperative group trials, all were enrolled on RTOG trials. The RTOG is an NCI-sponsored cooperative group focusing primarily on radiation-related issues. The fact that only 14% of breast cancer patients were enrolled onto RTOG studies may be owing to the relative paucity of RTOG breast cancer trials available during this study period compared with other cooperative groups. Recently, the RTOG breast committee has increased the number of trials available. This finding may also reflect a greater prevalence of multidisciplinary tumor boards for breast cancer patients and/or the patterns of referral of these patients to other specialties before radiation oncology. Unlike enrollment onto institutional trials, which, as expected, was primarily from academic institutions (63%), 75% of the accrual to cooperative group trials came from nonacademic institutions. These results suggest that, at least
●
B. MOVSAS et al.
1149
for certain disease sites, the radiation oncologist, both in academics and in the community, could serve as a critical “gatekeeper” to help accrue patients to clinical oncology trials. For all disease sites, this process should be done in the context of a multidisciplinary evaluation or tumor board. As a key part of this collaborative team, radiation oncologists should feel empowered to offer relevant clinical oncology trials to their patients. Although it is encouraging that there has been some increase in accrual during the 1990s (1.5% during 1992–1994 vs. 4.2% during 1996 –1999), the reasons underlying this trend are unclear and further improvement is critical. The present study has several inherent limitations. Similar to other studies, it does not include socioeconomic factors (other than insurance) or distance from cancer centers. Other factors that are likely important and have not been studied in most of these analyses include patient attitudes and knowledge base that may influence their willingness to participate (16, 17). Comis et al. (18) recently conducted interviews to better understand the attitudes of American adults toward participation in cancer clinical trials. Interestingly, their findings suggest that the primary barrier to accrual is not the attitude of patients but rather the unavailability of appropriate clinical trials and/or disqualification based on eligibility criteria, as well as the reluctance of some physicians to engage in accrual (18). Moreover, this analysis can not directly determine the eligibility of individual patients for a particular trial and/or whether such trials were actually available at particular institutions. Such limitations similarly apply to other analyses, in which the accrual to cooperative group trials was generally compared with national averages, despite the fact that the validity of such a comparison is unclear (4 – 6). The strength of the present study, unlike the other analyses, is that it represents surveys whose original purpose, using stratified random sampling, was to assess the patterns of care, which included whether patients did or did not enroll onto clinical trials. In conclusion, this study demonstrates that practice type and race significantly influence enrollment on clinical oncology trials. This suggests that increased communication and education regarding protocols, particularly focusing on physicians in nonacademic settings and minority patients, will be essential to enhance accrual.
REFERENCES 1. Taylor KM, Feldstein ML, Skeel RT, et al. Fundamental dilemmas of the randomized clinical trial process: Results of a survey of the 1,737 Eastern Cooperative Oncology Group investigators. J Clin Oncol 1994;12:1796 –1805. 2. Lukens JN. Progress resulting from clinical trials. Solid tumors in childhood cancer. Cancer 1994;74(9 Suppl.):2710 – 2718. 3. Lee JY, Breaux SR. Accrual of radiotherapy patients to clinical trials. Cancer 1983;52:1014 –1016. 4. Sateren WB, Trimble EL, Abrams J, et al. How sociodemographics, presence of oncology specialists, and hospital cancer programs affect accrual to cancer treatment trials. J Clin Oncol 2002;20:2109 –2117.
5. McCaskill-Stevens W, Pinto H, Marcus AC, et al. Recruiting minority cancer patients into cancer clinical trials: A pilot project involving the Eastern Cooperative Oncology Group and the National Medical Association. J Clin Oncol 1999;17: 1029 –1039. 6. Hutchins LF, Unger JM, Crowley JJ, et al. Underrepresentation of patients 65 years of age or older in cancer-treatment trials. N Engl J Med 1999;341:2061–2067. 7. Research Triangle Institute. SUDAAN user’s manual. 8.0 ed. Research Triangle Park, NC: Research Triangle Institute; 2001. 8. Advani AS, Atkeson B, Brown CL, et al. Barriers to the participation of African-American patients with cancer in clinical trials: A pilot study. Cancer 2003;97:1499 –1506.
1150
I. J. Radiation Oncology
●
Biology
●
Physics
9. Adams-Campbell LL, Ahaghotu C, Gaskins M, et al. Enrollment of African Americans onto clinical treatment trials: Study design barriers. J Clin Oncol 2004;22:730 –734. 10. Yee KW, Pater JL, Pho L, et al. Enrollment of older patients in cancer treatment trials in Canada: Why is age a barrier? J Clin Oncol 2003;21:1618 –1623. 11. Gross CP, Wong N, Dubin JA, et al. Enrollment of older persons in cancer trials after the Medicare reimbursement policy change. Arch Intern Med 2005;165:1514 –1520. 12. Townsley CA, Selby R, Siu LL. Systematic review of barriers to the recruitment of older patients with cancer onto clinical trials. J Clin Oncol 2005;23:3112–3124. 13. Townsley CA, Chan KK, Pond GR, et al. Understanding the attitudes of the elderly towards enrollment into cancer clinical trials. BMC Cancer 2006;6:34.
Volume 68, Number 4, 2007
14. Lewis JH, Kilgore ML, Goldman DP, et al. Participation of patients 65 years of age or older in cancer clinical trials. J Clin Oncol 2003;21:1383–1389. 15. Kemeny MM, Peterson BL, Kornblith AB, et al. Barriers to clinical trial participation by older women with breast cancer. J Clin Oncol 2003;21:2268 –2275. 16. Ellis PM, Butow PN, Tattersall MH, et al. Randomized clinical trials in oncology: Understanding and attitudes predict willingness to participate. J Clin Oncol 2001;19:3554 –3561. 17. Mills EJ, Seely D, Rachlis B, et al. Barriers to participation in clinical trials of cancer: A meta-analysis and systematic review of patient-reported factors. Lancet Oncol 2006;7:141–148. 18. Comis RL, Miller JD, Aldige CR, et al. Public attitudes toward participation in cancer clinical trials. J Clin Oncol 2003;21:830 – 835.