|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2007.11.8877 on January 28 2008 © 2008 American Society of Clinical Oncology.
Late Cardiac Effects of Adjuvant Chemotherapy in Breast Cancer Survivors Treated on Southwest Oncology Group Protocol S8897
From the University of California, Los Angeles, Los Angeles; The Angeles Clinic and Research Institute, Santa Monica, CA; Southwest Oncology Group Statistical Center, Seattle, WA; University of Arkansas for Medical Science, Little Rock, AR; Wichita Community Clinical Oncology Program, Wichita, KS; Greenville Community Clinical Oncology Program, Greenville, SC; Ozarks Regional Community Clinical Oncology Program, Springfield, MO; and Loyola University Stritch School of Medicine, Maywood, IL Corresponding author: Patricia A. Ganz, MD, Schools of Medicine and Public Health, Division of Cancer Prevention & Control Research, Jonsson Comprehensive Cancer Center, University of California, Los Angeles, 650 Charles Young Dr South, Room A2-125 CHS, Los Angeles, CA 90095-6900; e-mail: pganz{at}ucla.edu
Purpose The late cardiac effects of adjuvant anthracycline therapy in survivors of early-stage breast cancer have had limited study. Subclinical and clinical cardiac late effects may contribute to added comorbidity over time. Patients and Methods We recruited patients treated on Southwest Oncology Group (SWOG) protocol S8897 who had been randomly assigned to adjuvant chemotherapy with or without doxorubicin. Left ventricular ejection fraction (LVEF) was evaluated at 5 to 8 years and 10 to 13 years after treatment randomization. Cardiac risk factors and events were reported by clinicians annually between the two assessments. Results A total of 180 breast cancer survivors from a potential sample of 1,176 patients were entered, 163 patients at 5 to 8 years and 17 additional patients at 10 to 13 years, with 93 longitudinal assessments of LVEF. There was no significant difference in the proportion of women with an LVEF less than 50% at 5 to 8 (cyclophosphamide, doxorubicin, and fluorouracil [CAF] v cyclophosphamide, methotrexate, and fluorouracil [CMF]: 5% v 7%; P = .68) or 10 to 13 years (CAF v CMF: 3% v 0%; P = .16); however, in an exploratory analysis, the mean LVEF in the doxorubicin group was statistically significantly lower in the 5- to 8-year sample (64.8% v 61.4%; P = .01) but not in the 10- to 13-year sample. In the longitudinal analysis, there was no significant deterioration in LVEF. Conclusion Women enrolled onto an adjuvant chemotherapy treatment clinical trial for breast cancer were successfully recruited to participate in a research study of the late effects of treatment, although many SWOG institutions and potentially eligible patients chose not to participate. In this selected sample, with up to 13 years of follow-up, exposure to doxorubicin did not increase the likelihood of adverse cardiac effects.
Breast cancer accounts for almost one fourth of the more than 10.5 million cancer survivors in the United States.1 Breast cancer survivors will continue to increase in numbers due to the widespread dissemination of mammography and adjuvant therapies.2 Gains in survival are not without some human cost, which may be manifested by short- and long-term sequelae, including pain, fatigue, depression, vasomotor symptoms, loss of fertility, and decreased physical functioning.3-6 Despite the growing number of breast cancer survivors, a recent systematic review of late cardiac effects of cancer treatment identified only two clinical studies that measured cardiac function in long-term breast cancer survivors exposed to anthracycline adjuvant therapy.7 Information from clinical trials is lacking because systematic long-term follow-up data collection is usually limited to mortality and breast cancer recurrence. Population studies are limited due to a paucity of data on dose or patient characteristics. In this study, we attempted to overcome these limitations by linking long-term cardiac outcomes to prior treatment exposure in a cohort of women with early-stage breast cancer. Anthracyclines play a central role in contemporary adjuvant treatment regimens.8 However, anthracyclines have dose-limiting cardiac toxicity and can result in congestive heart failure and cardiomyopathy years after administration.9 In the late 1980s, anthracycline therapy was cautiously extended to the adjuvant treatment of stage I patients, and Southwest Oncology Group (SWOG) protocol S8897 (Intergroup Protocol INT-0102) was a clinical trial designed to examine the value of an anthracycline-containing regimen in this setting.10 In 1993, about 5 years after the initiation of S8897, we designed SWOG protocol S9342, Late Cardiac Effects of Adjuvant CMF (cyclophosphamide, methotrexate, and fluorouracil) versus CAF (cyclophosphamide, doxorubicin, and fluorouracil) in Women with Node Negative Breast Cancer Treated on S8897, to examine late cardiac outcomes in this patient population with favorable prognosis. We also evaluated the feasibility of recruitment of long-term cancer survivors treated within the cooperative group setting for a late-effects study.
Treatment Trial Results for S8897 The SWOG-coordinated protocol S8897 (INT-0102) enrolled women from multiple cooperative groups between July 1989 and February 1993. Women at high risk for breast cancer recurrence were randomly assigned to either CMF for six cycles, or CAF for six cycles, with or without tamoxifen for 5 years (study schema [Fig A1] and treatment schedule are listed in the Appendix, online only). Women in the CAF arm were expected to receive a total of 360 mg/m2 doxorubicin, the exposure of interest in S9342. INT-0102 showed no significant difference in disease-free survival for CAF compared with CMF. However, patients treated with CAF had slightly better 10-year overall survival than patients treated with CMF (85% v 82%; P = .03, planned one-sided test), with tamoxifen benefit only in hormone receptor–positive patients.10 Median follow-up for patients last known alive was 10.8 years. Cardiac effects were more severe in the CAF arm during the first year of follow-up (11 severe cardiac effects in CAF arms, including one fatal event, v four severe cardiac effects in the CMF arms). Cardiac events after 1 year were reported before relapse in 53 patients in the CMF arms and 63 patients in the CAF arms. Events included cardiomyopathy, cardiomegaly, or congestive heart failure in 22 patients in CMF arms and in 27 patients in CAF arms.10 There were 57 deaths without relapse in the CMF arms and 72 deaths without relapse in the CAF arms.
Design, Patient Eligibility, and Recruitment Procedures Eligibility for S9342 included registration on S8897 and completion of at least one cycle of assigned chemotherapy; being currently disease free; having no intervening treatment with chemotherapy or radiation for another malignancy; random assignment to treatment on S8897 between 5 and 8 years earlier; and treatment at a SWOG-affiliated institution. Lack of funding for the MUGA scans delayed study initiation until February 1997. Initially, we had planned to enroll eligible women at 5 years after random assignment, with a baseline MUGA scan and cardiac history, followed by interval annual cardiac history evaluations, and then a follow-up MUGA scan at 10 years. Delay in study activation required expansion of the baseline period to include women from 5 to 8 years after random assignment, and the follow-up MUGA assessment time to include any woman between 10 and 13 years after registration, whether or not she had enrolled in the baseline assessment. Patients registered in the 5- to 8-year window consented to have MUGA scans on study entry and at 10 years after randomization to S8897. The protocol required annual investigator-reported cardiac history assessments. Patients registered in the 10- to 13-year window were required to have a MUGA scan and a cardiac history assessment. The extramural funding for this study also supported a recruitment coordinator at the SWOG Statistical Center to help support the logistical challenges in recruiting and enrolling long-term survivors in this study. The coordinator informed each of the SWOG site investigators of the potential list of patients from S8897 who were potentially eligible for S9342, and also encouraged each of the sites to obtain institutional review board (IRB) approval for the study. A special form was developed and used to track reasons for refusal to participate once a woman had been approached. These data were available to examine barriers to recruitment. The study protocol was approved by the IRB at each participating SWOG institution, and all women provided written informed consent for participation.
Statistical Methods and Analysis Plan Subsequent exploratory analyses examined differences in the mean LVEF% between the two treatment arms at both time points. Linear regression was used to examine differences in mean LVEF% adjusted for several factors (receipt of tamoxifen, age at registration to S9342, hypertension, smoking status, diabetes, and time between random assignment to S8897 and registration to S9342). Cross-sectional analyses only included individuals with both MUGA scan and cardiac history data. For individuals with complete MUGA and cardiac history data from both the 5- to 8-year window and at 10 to 13 years, a change score analysis was conducted on the raw change in resting LVEF over time using linear regression analysis adjusted for chemotherapy arm and time between the two MUGA scans. P values were obtained from Wald tests of regression coefficients for the treatment arm. Patient characteristics and cardiac history data were analyzed by participation pattern (to evaluate the cohort profile), by follow-up pattern (to ascertain whether patients with different follow-up patterns represented quantifiably different cohorts), and by treatment arm.
Accrual and Data Submission There were 157 SWOG institutions with 1,176 patients potentially available to enter S9342. (Sixty institutions obtained IRB approval for the study, representing 662 potentially eligible patients.) An estimated 954 women of the 1,176 available from S8897 met eligibility (had not died or had not experienced progressive/recurrent cancer) to enroll onto S9342 in the 5- to 8-year window; 518 women were estimated to be eligible in the 10- to 13-year window. Figure 1 describes the flow of potentially available patients. In the subset of potentially eligible patients with known reasons for nonparticipation in S9342 (n = 302), the main reasons for nonparticipation were institutional and administrative impediments (eg, unable to contact patient, not enough time to contact patient, other institutional reason; 52%) and patient refusal (28%).
One hundred eighty women (from 50 institutions) were registered onto S9342, with 163 initial registrations occurring at the 5- to 8-year window and 17 at the 10- to 13-year window (Fig 1). Patients were defined as assessable if they had an on-time MUGA scan (Table 1). At 5 to 8 years, there were 156 assessable patients (16% of the estimated 954 women eligible to enroll), 76 receiving CMF and 80 receiving CAF. At 10 to 13 years, there were 110 (93 repeat scans and 17 first scans) assessable patients (21% of the estimated 518 women eligible to enroll), with 52 receiving CMF and 58 receiving CAF. The accrual goal of 70 patients per treatment arm was met for the 5- to 8-year window but not for the 10- to 13-year window. The mean time between random assignment to S8897 and the 5- to 8-year MUGA scan was 6.7 years (standard deviation [SD] = 1.0), and was 10.7 years (SD = 0.8) for the 10- to 13-year MUGA scan.
For exploratory regression analyses, assessable patients were required to have available cardiac histories. Both MUGA scans and cardiac histories were available for 141 women at 5 to 8 years and for 83 women at 10 to 13 years. Ninety-three women had MUGA scans available at both times for the longitudinal analysis.
Patient Characteristics and Cardiac History
We further examined attrition over the course of the study as follows: for the 5- to 8-year analysis, patients were divided into the 93 patients with both 5- to 8-year and 10- to 13-year MUGA and the 63 with only 5- to 8-year MUGA data. Similarly, for the 10- to 13-year analysis, patients were divided into the 93 patients with both scans and the 17 patients with only the 10- to 13-year MUGA. Table 1 lists selected characteristics by follow-up type for each analysis (5- to 8-year analysis and 10- to 13-year analysis). Table 1 also lists an additional comparison by chemotherapy treatment arm. There is little indication that patterns of follow-up or treatment received were associated with baseline medical or demographic characteristics. Table 3 lists cardiac history data obtained within 90 days of the MUGA scan by follow-up type and treatment arm. In the comparisons by follow-up type, there was little indication that patterns of follow-up were associated with cardiac characteristics. In addition, with the exception of angina in the 5- to 8-year sample (6% in CAF patients and 0% in CMF patients), there were no significant differences in MUGA scan or cardiac history by treatment arm for either the 5- to 8-year or 10- to 13-year samples.
Finally, in an assessment of low LVEF% by follow-up type (data not shown), there was no evidence that the proportion of patients with LVEF less than 50% at 5 to 8 years was higher among patients who dropped out before 10 to 13 years than among patients who remained on study through the 10- to 13-year assessment (proportion with LVEF < 50% at 5 to 8 years compared with 10 to 13 years, 3% v 8%, respectively; P = .22). Thus, once on study, the 10- to 13-year assessment was not biased toward a healthier sample.
Primary Analysis
Exploratory Analyses For the 5- to 8-year window, mean LVEF was 61.4% (SD = 7.3) in the CAF arm and 64.8% (SD = 9.5) in the CMF arm (P = .01; Table 4). For the 10- to 13-year window, mean LVEF was 63.4% (SD = 7.0) in the CAF arm and 62.8% (SD = 7.2) in the CMF arm (P = .64). Cross-sectional box plots for resting LVEF% by MUGA scan time and treatment arm are shown in Figure 2. Results were similar for the adjusted analyses using linear regression with mean LVEF% as the outcome. In the adjusted analyses, fewer patients were included (due to the requirement of complete data for the adjustment variables), influencing mean LVEF % estimates. For the longitudinal analysis, the average increase in resting LVEF% from 5 to 8 versus 10 to 13 years was 5.4% (SE = 3.4) in the CAF group and 3.0% (SE = 3.6) in the CMF group (P = .23).
We have demonstrated that it is possible to recruit long-term breast cancer survivors initially enrolled onto a cooperative group trial to participate in a late cardiac effects study; however, institutional participation was not complete and many patients who were approached for the study declined to participate. Although chemotherapy treatment was assigned at random to these long-term breast cancer survivors, those who participated in S9342 are a selected observational subgroup from the original treatment trial. Participation also required being alive without breast cancer recurrence, and we are missing important information on those women who were not available or were ineligible to participate in this follow-up study. Thus, this retrospectively designed study likely underestimates the true incidence of cardiac complications after treatment exposure, and these limitations must be considered as we interpret our findings. We found no significant differences between CMF and CAF arms in the proportions of patients with a resting LVEF less than 50% at 5 to 8 years or at 10 to 13 years. An exploratory analysis of the mean LVEF% at 5 to 8 years showed that CAF treatment resulted in a decrease in LVEF% compared with CMF treatment. This result was statistically significant in both the univariate and multivariate analyses. At 10 to 13 years, mean LVEF% values were similar for both CAF- and CMF-treated patients in both the univariate and multivariate settings; however, this result is underpowered. An exploratory longitudinal analysis showed no difference by treatment arm in the change in LVEF% over time. S9342 was originally designed with a dichotomous end point, given that baseline LVEF data were not available on patients. It was hypothesized that only an LVEF score less than 50% would warrant clinical attention as an adverse outcome in women who were otherwise healthy, and there were no medical therapies used at that time for decreased LVEF. The primary study analysis reported here followed the original protocol plan; however, when the proportion of patients with LVEF less than 50% was lower than expected, an analysis using the continuous LVEF% outcomes was undertaken to evaluate clinically meaningful differences in LVEF% between treatment arms. Although the 5- to 8-year regression analyses showed statistically significant differences favoring the CMF group, this result is exploratory (ie, does not reflect the primary study design), and the clinical significance of the 3.4% difference in mean LVEF is uncertain. A prospective study is required to assess whether declines of this magnitude are meaningful, and whether or not their interaction with other prognostic factors over the long term would be important. The nature of this design requires the analysis of possible selection factors that could influence the results. In particular, patients with different follow-up patterns could represent fundamentally different cohorts with respect to baseline demographic, biologic, and cardiac history characteristics, which in turn could be associated with the primary end point. However, we found minimal evidence of differences in these characteristics by follow-up pattern, providing justification for combining patients with different follow-up types. In addition, we did not find any significant differences in patient characteristics or time on treatment between potentially eligible patients who either enrolled or did not enroll onto S9342, indicating that this cohort, with respect to known factors, is adequately representative of all patients potentially available for enrollment. Prospective assessment and monitoring of cardiac function is included in many adjuvant therapy trials today, especially when additional cardiotoxic drugs are added to anthracycline-containing regimens. For example, high rates of cardiac dysfunction after standard therapy with doxorubicin (240 mg/m2) and cyclophosphamide were noted before initiation of trastuzumab therapy (ie, 6.7% had a decline of 16% or fell below the lower limit of normal).11,12 Unfortunately, patient-reported outcomes were not collected in S9342. The follow-up of women in current treatment trials could provide additional insight into the long-term effects of adjuvant therapy on cardiac function, fatigue, and physical function, also taking into account potential changes from left-sided breast irradiation.13,14 In S8897, doxorubicin was administered in split dosage on days 1 and 8 of each cycle, which may have modified the degree of cardiac toxicity observed in these breast cancer survivors. Within the limitations of the study design, we did not observe major deficits in cardiac function for women who received CAF chemotherapy as part of standard adjuvant therapy in S8897. Results are similar to those of other reports, with similar designs, that have examined late cardiac effects in women exposed to anthracyclines as part of adjuvant therapy.15,16 Modern therapies for congestive heart failure management have improved, such that outcomes in these patients may be much better.17 Nevertheless, population-based studies in older breast cancer survivors show a substantial increase in the risk of cardiomyopathy with anthracycline-based chemotherapy, suggesting the need for continued attention to this issue as adjuvant therapy is extended to older breast cancer patients.18,19
The author(s) indicated no potential conflicts of interest.
Conception and design: Patricia A. Ganz, Carol M. Moinpour, Laura F. Hutchins, Silvana Martino, Kathy S. Albain Financial support: Patricia A. Ganz, Kathy S. Albain Administrative support: Silvana Martino Provision of study materials or patients: Shaker R. Dakhil, Jeffrey K. Giguere, J. Wendall Goodwin, Kathy S. Albain Collection and assembly of data: Michael A. Hussey, Shaker R. Dakhil, Jeffrey K. Giguere, J. Wendall Goodwin, Silvana Martino, Kathy S. Albain Data analysis and interpretation: Patricia A. Ganz, Michael A. Hussey, Carol M. Moinpour, Joseph M. Unger, Silvana Martino, Kathy S. Albain Manuscript writing: Patricia A. Ganz, Michael A. Hussey, Carol M. Moinpour, Joseph M. Unger, Kathy S. Albain Final approval of manuscript: Patricia A. Ganz, Michael A. Hussey, Carol M. Moinpour, Joseph M. Unger, Laura F. Hutchins, Shaker R. Dakhil, Jeffrey K. Giguere, J. Wendall Goodwin, Silvana Martino, Kathy S. Albain
We thank the breast cancer patients who were willing to participate in this study of the late effects of adjuvant treatment of breast cancer. This study could also not have been conducted without the dedication of the SWOG investigators and study coordinators who enrolled these women onto S9342 and ensured completion of the MUGA scans and cardiac history forms.
published online ahead of print at www.jco.org on January 28, 2008. Supported in part by the following Public Health Service Cooperative Agreement grants awarded by the National Cancer Institute, Department of Health and Human Services: Grants No. CA38926, CA32102, CA35431, CA67663, CA45560, CA37981, CA12644, CA22433, CA35176, CA58416, CA35261, CA58658, CA46113, CA58882, CA13612, CA46441, CA46282, CA04919, CA35090, CA35178, CA35192, CA58723, CA27057, CA16385, CA35281, CA58686, CA52654, CA58861, CA20319, CA35119, CA76447, CA42777, CA45377, CA67575, CA86780, CA12213, CA35996, CA28862, CA35128, and CA35262. Supplemental funding for the Cardiac Evaluation Sub-study was provided by the National Cancer Institute and the Susan G. Komen Foundation. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Cancer survivorship–United States, 1971-2001. MMWR Morb Mortal Wkly Rep 53:526-529, 2004[Medline] 2. Berry DA, Cronin KA, Plevritis SK, et al: Effect of screening and adjuvant therapy on mortality from breast cancer. N Engl J Med 353:1784-1792, 2005 3. Bower JE, Ganz PA, Desmond KA, et al: Fatigue in long-term breast carcinoma survivors. Cancer 106:751-758, 2006[CrossRef][Medline] 4. Ganz PA, Desmond KA, Leedham B, et al: Quality of life in long-term, disease-free survivors of breast cancer: A follow-up study. J Natl Cancer Inst 94:39-49, 2002 5. Ganz PA, Greendale GA, Petersen L, et al: Breast cancer in younger women: Reproductive and late health effects of treatment. J Clin Oncol 21:4184-4193, 2003 6. Ganz PA, Kwan L, Stanton AL, et al: Quality of life at the end of primary treatment of breast cancer: First results from the moving beyond cancer randomized trial. J Natl Cancer Inst 96:376-387, 2004 7. Carver JR, Shapiro CL, Ng A, et al: American Society of Clinical Oncology clinical evidence review on the ongoing care of adult cancer survivors: Cardiac and pulmonary late effects. J Clin Oncol 25:3991-4008, 2007 8. Hamilton A, Hortobagyi G: Chemotherapy: What progress in the last 5 years? J Clin Oncol 23:1760-1775, 2005 9. Floyd JD, Nguyen DT, Lobins RL, et al: Cardiotoxicity of cancer therapy. J Clin Oncol 23:7685-7696, 2005 10. Hutchins LF, Green SJ, Ravdin PM, et al: Randomized, controlled trial of cyclophosphamide, methotrexate, and fluorouracil versus cyclophosphamide, doxorubicin, and fluorouracil with and without tamoxifen for high-risk, node-negative breast cancer: Treatment results of Intergroup Protocol INT-0102. J Clin Oncol 23:8313-8321, 2005 11. Romond EH, Perez EA, Bryant J, et al: Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med 353:1673-1684, 2005 12. Hayes DF, Picard MH: Heart of darkness: The downside of trastuzumab. J Clin Oncol 24:4056-4058, 2006 13. Correa CR, Litt HI, Hwang WT, et al: Coronary artery findings after left-sided compared with right-sided radiation treatment for early-stage breast cancer. J Clin Oncol 25:3031-3037, 2007 14. Harris EE, Correa C, Hwang WT, et al: Late cardiac mortality and morbidity in early-stage breast cancer patients after breast-conservation treatment. J Clin Oncol 24:4100-4106, 2006 15. Zambetti M, Moliterni A, Materazzo C, et al: Long-term cardiac sequelae in operable breast cancer patients given adjuvant chemotherapy with or without doxorubicin and breast irradiation. J Clin Oncol 19:37-43, 2001 16. Bonneterre J, Roche H, Kerbrat P, et al: Long-term cardiac follow-up in relapse-free patients after six courses of fluorouracil, epirubicin, and cyclophosphamide, with either 50 or 100 mg of epirubicin, as adjuvant therapy for node-positive breast cancer: French Adjuvant Study Group. J Clin Oncol 22:3070-3079, 2004 17. Tallaj JA, Franco V, Rayburn BK, et al: Response of doxorubicin-induced cardiomyopathy to the current management strategy of heart failure. J Heart Lung Transplant 24:2196-2201, 2005[CrossRef][Medline] 18. Doyle JJ, Neugut AI, Jacobson JS, et al: Chemotherapy and cardiotoxicity in older breast cancer patients: A population-based study. J Clin Oncol 23:8597-8605, 2005 19. Giordano SH, Duan Z, Kuo YF, et al: Use and outcomes of adjuvant chemotherapy in older women with breast cancer. J Clin Oncol 24:2750-2756, 2006 Submitted March 25, 2007; accepted October 23, 2007.
Related Editorial
Related Correspondence
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2008 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
