|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 24, No 3 (January 20), 2006: pp. 361-369 © 2006 American Society of Clinical Oncology. DOI: 10.1200/JCO.2005.02.3127 The Use of Radiation As a Component of Breast Conservation Therapy in National Comprehensive Cancer Network CentersFrom the Department of Radiation Oncology and Medical Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, TX; Department of Oncology, City of Hope Hospital, Los Angeles, CA; Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA; Department of Breast and Soft Tissue Surgery, Roswell Park Cancer Institute, Buffalo, NY; and Department of Oncology, Fox Chase Cancer Center, Philadelphia, PA Address reprint requests to Thomas A. Buchholz, MD, Department of Radiation Oncology, The University of Texas M.D. Anderson Cancer Center, 1515 Holcolmbe Blvd, Unit 1202, Houston, TX 77030; e-mail: tbuchhol{at}mdanderson.org
Purpose Benchmark data regarding quality measures of breast cancer management are needed. We investigated rates of radiation use after breast conservation therapy (BCT) for patients treated for ductal carcinoma-in-situ (DCIS) or invasive breast cancer at National Comprehensive Cancer Network (NCCN) centers. Patients and Methods We studied 3,333 consecutive patients treated between 1997 and 2002 with BCT for DCIS (n = 587) or for stage I or II breast cancer (n = 2,746) in eight NCCN centers. Results The overall rate of radiation therapy use was 91%, with a lower frequency of radiation use in DCIS versus invasive breast cancers (82% v 94%; odds ratio [OR] = 0.31; P < .0001). In a multivariable analysis of the patients with DCIS, the only factor significantly associated with lower rates of radiation use was low/intermediate grade (OR = 0.19; P = .0003). For patients with invasive breast cancer, significant factors were presence of comorbidity (OR = 0.53; P = .0005), tubular histology (OR = 0.39; P = .02), type of health insurance (P = .0072), and the NCCN institution (P = .0005). The model also showed lower rates of radiation use in patients with stage II disease who did not receive systemic therapy (OR = 0.01; P = .0001), younger patients who did not receive systemic therapy (P = .003); and older patients with stage I disease (P < .0001). Conclusion Radiation use as a component of BCT was high for patients seen at NCCN centers; however, there was variability in practice patterns noted across institutions. Radiation was most commonly omitted in patients with favorable disease characteristics, patients with comorbidities, and patients who also did not receive guideline-recommended systemic treatment.
Phase III randomized trials have demonstrated that radiation use after breast conservation surgery significantly reduces the probability of local recurrence for patients with ductal carcinoma-in-situ (DCIS) or early-stage invasive breast cancer.1-5 Despite this high-quality evidence, prior studies have suggested that underuse of radiation after breast-conserving surgery is common.6-13 However, these studies have relied largely on tumor registry data, and there is mounting evidence that radiation is systematically undercaptured by registries.14,15 Linked Surveillance, Epidemiology, and End Results (SEER) –Medicare data provide more reliable estimates of true radiation use, but only among women older than age 65.16-20 Furthermore, these types of data sources do not contain detailed information on many of the clinical characteristics that may be driving decisions about whether to use radiation. Finally, much of the data on radiation use predates the final results of the randomized trials, and therefore may not reflect current patterns of care. In this article, we studied the rates and determinants of radiation use as a component of breast conservation therapy (BCT) among patients treated at National Comprehensive Cancer Network (NCCN) centers. The NCCN, a nonprofit alliance of cancer centers, has produced comprehensive guidelines for the care of all of the common cancers. The NCCN breast cancer guidelines recommend that radiation be used as a component of BCT for all patients with invasive breast cancer.21 The guidelines also recommend radiation after BCT for all patients with DCIS except those with low-grade lesions and tumors less than 0.5 cm. In such patients, radiation is considered optional.21 Since 1997, the NCCN Outcomes Project has been collecting prospective data on all newly diagnosed breast cancer patients treated in participating centers. One goal of the project is to track concordance with guidelines. As a result, we collect extensive information on patient and clinical factors that might influence treatment choice. Data sources include not only tumor registries, but also inpatient and outpatient medical record abstractions. Furthermore, the data are subject to rigorous quality assurance, including on-site audits against source documentation. Therefore, an analysis of rates and determinants of radiation use in this data set provides insights into treatment use in the current era, is minimally affected by under-reporting biases, and permits a comprehensive analysis of the association of radiation use with patient, disease, and treatment characteristics.
Patient Characteristics The study cohort consisted of women receiving their primary breast cancer care at one of the eight NCCN institutions participating in the NCCN Breast Cancer Outcomes Project between July 1997 and June 2002: City of Hope National Medical Center, Duarte, CA; Dana-Farber Cancer Institute, Boston MA; Fox Chase Cancer Center, Philadelphia, PA; The University of Texas M.D. Anderson Cancer Center, Houston TX; Roswell Park Cancer Institute, Buffalo, NY; University of Michigan Comprehensive Cancer Center, Ann Arbor, MI; Arthur G. James Cancer Hospital and Richard Solove Research Institute at the Ohio State University, Columbus, OH; and H. Lee Moffitt Cancer Center and Research Institute at the University of South Florida, Tampa, FL. Each center is an academic comprehensive cancer center with a closed, staff-model practice. The majority of surgical and medical oncologists treating breast cancer at these institutions devote most or all of their clinical effort to breast cancer care. The institutional review boards at each center approved the data collection process, data transmission methods, and data storage protocols. At institutions where the institutional review board required project-specific signed informed consent for data collection, only patients who provided consent were included in this analysis. Women were eligible for this study if they had newly diagnosed DCIS, or stage I or II (American Joint Committee on Cancer, fifth edition) breast cancer and presented for care at one of the eight participating institutions between July 1, 1997, and June 25, 2002. Patients presenting for second opinions, those receiving no primary therapy at the NCCN institution, and patients referred solely for bone marrow transplantation were not eligible for entry into the database. Women with synchronous bilateral breast cancers and those with a bilateral breast cancer diagnosed within 90 days of the index primary were excluded. For other women with a second primary breast cancer, only the first episode of breast cancer was considered in the analysis. Of the 7,848 eligible patients, we then identified 4,728 patients who had received BCT as their definitive surgery for DCIS (n = 742) or a stage I or II breast cancer (n = 3,986). To ensure that patients whose radiation was deferred until after adjuvant systemic therapy were identified, we restricted the cohort to patients with at least 150 days of follow-up from initial presentation for DCIS, and 365 days of follow-up from initial presentation for invasive breast cancer. The resulting sample included 3,374 patients (623 DCIS patients and 2,751 stage I and II patients). Patients who transferred their care out of the study institution, underwent a bone marrow transplantation, or developed a new cancer other than breast cancer within these same intervals were also excluded. Finally, we excluded 41 patients (36 DCIS and five stage I/II) enrolled onto clinical trials investigating the use of radiation after BCT. After these exclusions, our sample population consisted of 3,333 patients who had received BCT: 587 for DCIS and 2,746 for invasive breast cancer.
Data Sources Race, ethnicity, educational status at presentation, and employment status at diagnosis were collected by patient survey. Income was assessed at the level of zip code by linking 2000 Census data on median household income (US Census 2000, Summary File 3) to zip code of residence at first presentation. Sixteen patients could not be matched on the basis of zip code. To determine distance to the nearest radiation therapy facility, we identified the latitude and longitude of the 1,197 hospitals listed in the 2000 American Hospital Association Annual Survey of Hospitals24 as providing radiation therapy services. For 15 hospitals, location was not available in the 2000 American Hospital Association data set, so we determined latitude and longitude from the US Census Bureau Web site (http://www.census.gov/cgi-bin/gazetteer). Latitude and longitude of the residences of all US patients were derived from their zip code using ZIPList5 (Geocode Z5LLDOC.TXT, 1995 to 2002; www.zipinfo.com). Distance from patients' residence to the nearest radiation facility was then calculated. Data regarding radiation dose, dose per fraction, and treatment fields were not recorded in the database and therefore were not available for analysis. Rigorous data quality assurance processes were in place to validate the accuracy of the data used in this study, including initial and follow-up data management training; on-line edit checking during web-based data entry; programmed logic checks against the pooled data repository; routine quality assurance reports to the centers for rectification by the data managers; and on-site audits of a random sample of source documents against the submitted data, conducted within the first few months of data collection and repeated annually.
Definition of Receipt of Therapy
Statistical Analysis Age at diagnosis, median household income, and minimum distance to a radiation therapy facility were considered as categoric variables because each had a nonlinear association with the receipt of radiation therapy. Age was categorized a priori into the following groups: younger than 50, 50 to 70, and older than 70; income and distance were categorized into quartiles identified from the data. Separate multivariable logistic regression models were constructed for DCIS and invasive disease. For both models, we included NCCN institution, stage of disease (for the model with stage I and II patients), age at diagnosis, and all potential explanatory variables that had an association with radiation use at a significance level of at least P = .20 in the univariate analyses. The multivariable logistic regression analyses evaluated each variable both independently and in its associations with other variables. The final models include variables or combinations of variables that were considered significant at a level of P = .05. The P values presented are from the overall logistic regression results along with the odds ratio (OR) and 95% CI for the categories of interest.
The overall use of radiation after BCT for the entire study cohort of 3,333 patients was 91%. Among patients with invasive breast cancer, 94% received radiation as a component of BCT; in contrast, only 82% of DCIS patients received radiation (P < .0001). The NCCN guidelines identify a single subset of DCIS patients for whom radiation after BCT may be considered optional: women with low-grade lesions and tumors less than 0.5 cm. Of the 62 patients in this subgroup, 71% received radiation; in comparison, 83% of the remaining DCIS patients (n = 525) were treated with radiation. The proportion of DCIS patients receiving radiation according to patient, tumor, and treatment factors is listed in Tables 1 and 2. In univariate analyses, patients with low/intermediate histologic grade and those in the lowest income quartile were significantly less likely to receive radiation. In addition, the NCCN center of treatment was a statistically significant factor associated with radiation use. Other clinical factors, socioeconomic factors, and distance to nearest radiation therapy facility were not associated with radiation use. In multivariate analysis, the only pathology factor statistically significantly associated with radiation for DCIS was histologic grade (Table 3).
Among patients with invasive disease, a number of factors were significantly associated with radiation use in univariate analyses (Table 4). Age older than 70, presence of at least one comorbid condition, education less than a high school, unemployed status, and income in the lower two quartiles were all associated with a lower likelihood of receiving radiation. Patients with Medicare plus managed care insurance, Medicare plus supplemental insurance, and Medicare alone were all less likely to receive radiation than those with managed care insurance. However, those receiving Medicaid/indigent care or patients paying for all care out of pocket had high rates of radiation use. Finally, patients who did not receive systemic therapy were also less likely to receive radiation therapy. Other variables that showed potential as predictors of radiation use (P  .20) were histology, histologic grade, surgical margins, and the NCCN institution where treatment was performed.
In multivariate logistic regression analysis, the following characteristics were found to be independently associated with radiation use in invasive disease: tubular histology (OR = 0.39; 95% CI, 0.18 to 0.86; P = .02), comorbidity score of 1 or higher (OR = 0.53; 95% CI, 0.37 to 0.75; P = .0005), health insurance type (P = .0072), and NCCN institution (P = .0005; Table 5). As depicted in Figure 1, significant interactions were seen between receipt of systemic therapy, stage, and age group. The use of radiation was high in the patients younger than 70 who received systemic therapy (top two lines of the graph). In contrast, radiation use was less common in the patients who were not treated with systemic therapy. This was most dramatic in younger patients with stage II disease (the bottom line of the graph). Patients with stage I disease who were older than 70 years were also less likely to receive radiation.
The data from this study indicate that the use of radiation as a component of BCT is high in patients treated in NCCN centers. The NCCN guideline recommending radiation for all women with stage I or II breast cancer was followed in 94% of women. Women treated with BCT for DCIS were less likely to receive radiation, although rates were still high (82% overall and 83% among those for whom the guidelines recommend it). The lower rate of radiation use in patients with DCIS was likely due to patient and provider biases. However, in defining the use of radiation, we selected different intervals in the period between presentation and when radiation was required to start for patients with DCIS (150 days) versus those with invasive disease (365 days) to account for the use of adjuvant chemotherapy. Potentially, if a number of patients received radiation between 150 and 365 days, these differences may have artificially lowered the true rate of radiation use in patients with DCIS. This study provides important data concerning a quality measure in breast cancer care in selected institutions recognized for their expertise in cancer care. These data can serve as benchmarks for comparisons to similar sets of data from other communities that administer cancer care. In general, the data from this study suggest that radiation after BCT use in NCCN centers is much higher than the rates reported from national registries and SEER. However, this is the first study to report rates of radiation use in a large sample of patients using data specifically abstracted from medical records for the purpose of outcome studies. A strength of this study is that all of the data were prospectively entered into a database and were subject to rigorous quality assurance testing to validate accuracy. This is of particular importance for the end point of this study because radiation use is an outpatient treatment modality and as such is prone to being under-reported in survey studies and in national databases. For example, Malin et al14 compared the tumor registry data of 304 breast cancer patients with data abstracted from the patients' medical records. Although the tumor registry was highly accurate for some parameters, such as the type of surgical procedure, there was poor accuracy for outpatient-based parameters. Specifically, with respect to radiation use, the tumor registry records were accurate in only 72% of the patients.14 Even considering the potential for under-reporting biases in the previously published articles investigating radiation use, it is likely that the rates of radiation use as a component of BCT we observed are still higher than those reported by others. For example, Bland et al,6 using 1995 data from the National Cancer Database (NCD), reported radiation use rates of 45% for patients treated with BCT for DCIS. Later, Winchester et al7 again examined the DCIS data from the NCD and found that radiation use after BCT was only 38% in 1985 but increased to a 44% to 54% range between 1991 and 1993. More recently, Baxter et al8 reported similar data from the SEER Program. In this article, the rate of radiation use after BCT for women with DCIS was 45% in 1992 and 54% in 1999. Others who have examined the SEER data regarding radiation use after BCT for DCIS have reported similar results.9,10 Finally, a study using linked SEER-Medicare data that included 1,725 patients age 65 or older treated between 1991 and 1996 with BCT for DCIS found that only 39% received radiation.16 One limitation in comparing these data to those of our study is that our study evaluated patients in a more recent time period (1997 to 2002). In addition, these studies did not use the timing criteria and same selection criteria for defining radiation use. For example, a patient who developed metastatic disease during adjuvant chemotherapy may have been coded as not receiving radiation in some databases but would have been excluded from our study set. In general, studies examining the rates of radiation use after BCT for women with invasive disease have reported higher rates than the studies examining women treated for DCIS. However, the 94% radiation use rate we report for patients with invasive disease is again higher than rates in the published literature. Nattinger et al11 reported data from the SEER Program on 47,278 patients who underwent BCT for invasive breast cancer between the years of 1983 to 1995, and found that only 75% received radiation. Previously, Farrow et al,12 using SEER data from 1985 and 1986, reported similar rates of radiation use and found that these rates varied from 60% to 81% across geographic areas of the United States. Using 1995 data from the NCD, Bland et al6 reported a rate of radiation use after BCT of 72% for women with stage I or II breast cancer. White et al13 more recently published an updated study that combined NCD data with data from the American College of Radiology Patterns of Care Studies. This analysis evaluated 6,990 patients with invasive breast cancer treated in 1994 and found that radiation was used in 86% of patients. In summary, the data from this study indicate that radiation use as a component of BCT is more common across NCCN centers than across those centers that contribute to SEER and other national databases. We believe that the data from the NCCN regarding this quality indicator can serve as a benchmark for other non-NCCN institutions and third-party payer programs. In our analyses, we found a number of factors correlated with radiation use after BCT. For patients treated for DCIS, radiation use was less common with low-grade disease. The NCCN guidelines state that lumpectomy with or without radiation is an appropriate option for patients with low-grade disease whose tumors are less than 0.5 cm. In addition, the lower use of radiation in low-grade disease may reflect a perception, supported by single-institution retrospective data, that patients with low- or intermediate-grade DCIS and those with wide negative surgical margins have a relatively low risk of recurrence after surgery alone.24 Previously published articles that used SEER data to assess radiation use in DCIS have shown an association of radiation use with age, race, tumor size, geographic location, and year of diagnosis.8,10,16 Unlike our study, a shortcoming of these reports is that they did not have available data concerning grade and margin status.
For patients with invasive disease, we identified a number of factors that correlated independently with radiation use. Patients with comorbidities were treated less frequently with radiation and rates of radiation use were lower in patients older than 70 with stage I disease. Previously published data from the NCD also found that age affected radiation use, with the use of radiation decreasing to 79% for women
Another factor we found to correlate with radiation use after breast-conserving surgery for invasive disease was insurance type. Specifically, we found that patients with Medicare insurance, with or without supplemental policies, were treated less frequently with radiation compared with those with other insurance types. Although Medicare insurance is correlated with increased age, our multivariable model found each of these variables to correlate independently with radiation use. Two previous studies found similar results. Specifically, women age Patient compliance may also affect radiation use. Specifically, we found low rates of radiation use in younger patients, particularly those with stage II disease who did not receive systemic treatment. It would be strongly recommended for young patients with stage II disease to receive systemic therapy, so it is likely that patient preferences entered into the decisions regarding systemic and radiation treatments in these patients. However, we did not have direct information about whether these patients were referred for systemic and radiation treatments and chose to pursue a recommended course of treatment. Finally, for both patients with DCIS and patients with invasive disease, the NCCN center of treatment was an independent factor that correlated with radiation use. As listed in Table 1, there was a 39% difference in radiation use between the centers with the lowest and highest percentage for patients with DCIS, and a 10% difference for patients with invasive breast cancer. These data indicate that individual physician biases likely have a significant impact on radiation use patterns and are consistent with the geographic variation in the use of radiation in patients with DCIS previously reported in a SEER database study.8,12 In conclusion, we found that the use of radiation as a component of BCT is high within the NCCN centers of this study. Variables associated with lower radiation use suggest that radiation is less common in patients with poor compliance, when physicians perceive risk of recurrence is low, and when patients are thought to have a competing risk of death from other illnesses. Outcome data such as we report are important as quality measures and allow for identification of practices in which compliance with published treatment guidelines were less frequent. Finally, these data can serve as important benchmarks for national practice standards.
The authors indicated no potential conflicts of interest.
Supported in part by the Nellie B. Conally Breast Cancer Research Fund, The University of Texas M.D. Anderson Cancer Center, and by Grant No. P50CA89393 to the Dana-Farber Cancer Institute from the National Cancer Institute. Presented at the 2004 American Radium Society Meeting, May 1-5, 2004, Houston, TX, and the 2004 NCCN Meeting, March 12, 2004, Hollywood, FL. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Fisher B, Anderson S, Bryant J, et al: Twenty-year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast cancer. N Engl J Med 347:1233-1241, 2002 2. Fisher B, Bryant J, Dignam JJ, et al: Tamoxifen, radiation therapy, or both for prevention of ipsilateral breast tumor recurrence after lumpectomy in women with invasive breast cancers of one centimeter or less. J Clin Oncol 20:4141-4149, 2002 3. Fisher B, Land S, Mamounas E, et al: Prevention of invasive breast cancer in women with ductal carcinoma in situ: An update of the national surgical adjuvant breast and bowel project experience. Semin Oncol 28:400-418, 2001[CrossRef][Medline] 4. Julien JP, Bijker N, Fentiman IS, et al: Radiotherapy in breast-conserving treatment for ductal carcinoma in situ: First results of the EORTC randomised phase III trial 10853—EORTC Breast Cancer Cooperative Group and EORTC Radiotherapy Group. Lancet 355:528-533, 2000[CrossRef][Medline] 5. Houghton J, George WD, Cuzick J, et al: Radiotherapy and tamoxifen in women with completely excised ductal carcinoma in situ of the breast in the UK, Australia, and New Zealand: Randomised controlled trial. Lancet 362:95-102, 2003[CrossRef][Medline] 6. Bland KI, Menck HR, Scott-Conner CE, et al: The National Cancer Data Base 10-year survey of breast carcinoma treatment at hospitals in the United States. Cancer 83:1262-1273, 1998[CrossRef][Medline] 7. Winchester DP, Menck HR, Osteen RT, et al: Treatment trends for ductal carcinoma in situ of the breast. Ann Surg Oncol 2:207-213, 1995[CrossRef][Medline] 8. Baxter NN, Virnig BA, Durham SB, et al: Trends in the treatment of ductal carcinoma in situ of the breast. J Natl Cancer Inst 96:443-448, 2004 9. Ernster VL, Barclay J, Kerlikowske K, et al: Incidence of and treatment for ductal carcinoma in situ of the breast. JAMA 275:913-918, 1996 10. Schootman M, Aft R: Rural-urban differences in radiation therapy for ductal carcinoma in-situ of the breast. Breast Cancer Res Treat 68:117-125, 2001[CrossRef][Medline] 11. Nattinger AB, Hoffmann RG, Kneusel RT, et al: Relation between appropriateness of primary therapy for early-stage breast carcinoma and increased use of breast-conserving surgery. Lancet 356:1148-1153, 2000[CrossRef][Medline] 12. Farrow DC, Hunt WC, Samet JM: Geographic variation in the treatment of localized breast cancer. N Engl J Med 326:1097-1101, 1992[Abstract] 13. White J, Morrow M, Moughan J, et al: Compliance with breast-conservation standards for patients with early-stage breast carcinoma. Cancer 97:893-904, 2003[CrossRef][Medline] 14. Malin JL, Kahn KL, Adams J, et al: Validity of cancer registry data for measuring the quality of breast cancer care. J Natl Cancer Inst 94:835-844, 2002 15. Bickell NA, Chassin MR: Determining the quality of breast cancer care: Do tumor registries measure up? Ann Intern Med 132:705-710, 2000 16. Gold HT, Dick AW: Variation in treatment for ductal carcinoma in situ in elderly women. Med Care 42:267-275, 2004[CrossRef][Medline] 17. Ballard-Barbash R, Potosky AL, Harlan LC, et al: Factors associated with surgical and radiation therapy for early stage breast cancer in older women. J Natl Cancer Inst 88:716-726, 1996 18. Du X, Freeman JL, Goodwin JS: Information on radiation therapy treatment in patients with breast cancer: The advantages of the linked Medicare and SEER data. J Clin Epidemiol 52:463-470, 1999[CrossRef][Medline] 19. Du X, Freeman JL, Nattinger AB, et al: Survival of women after breast conserving surgery for early stage breast cancer. Breast Cancer Res Treat 72:23-31, 2002[CrossRef][Medline] 20. Potosky AL, Merrill RM, Riley GF, et al: Breast cancer survival and treatment in health maintenance organization and fee-for-service settings. J Natl Cancer Inst 89:1683-1691, 1997 21. Carlson RW, Edge SB, Theriault RL, et al: NCCN: Breast cancer. Cancer Control 8(6 suppl 2):54-61, 2001[Medline] 22. Charlson ME, Pompei P, Ales KL, et al: A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis 40:373-383, 1987[CrossRef][Medline] 23. Katz JN, Chang LC, Sangha O, et al: Can comorbidity be measured by questionnaire rather than medical record review? Med Care 34:73-84, 1996[CrossRef][Medline] 24. American Hospital Association: Annual Survey of Hospitals: Hospital Statistics. Chicago, IL, American Hospital Association, 2000 25. Silverstein MJ, Lagios MD, Craig PH, et al: A prognostic index for ductal carcinoma in situ of the breast. Cancer 77:2267-2274, 1996[CrossRef][Medline] 26. Riley GF, Potosky AL, Klabunde CN, et al: Stage at diagnosis and treatment patterns among older women with breast cancer: An HMO and fee-for-service comparison. JAMA 281:720-726, 1999 Submitted April 8, 2005; accepted October 31, 2005.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2006 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
70 years of age.
