Stroke As a Late Treatment Effect of Hodgkin's Disease: A Report From the Childhood Cancer Survivor Study
http://www.100md.com
《临床肿瘤学》
The University of Texas Southwestern Medical Center at Dallas, Dallas
The M.D. Anderson Cancer Center, Houston, TX
US Food and Drug Administration, Rockville, MD
Fred Hutchinson Cancer Research Center, Seattle, WA
University of Minnesota, Minneapolis, MN
St Jude Children's Research Hospital, Memphis, TN
Stanford University School of Medicine, Stanford, CA
Children's National Medical Center, Washington, DC, and the Childhood Cancer Survivor Study (see online Appendix)
ABSTRACT
PURPOSE: The objectives of this report are to examine the incidence of and risk factors for stroke among childhood Hodgkin's disease (HD) survivors.
PATIENTS AND METHODS: The Childhood Cancer Survivor Study is a multi-institutional cohort study of more than 5-year cancer survivors diagnosed between 1970 and 1986 and a sibling comparison group. Incidence rates of stroke among HD survivors (n = 1,926) and siblings (n = 3,846) were calculated and compared. Cox proportional hazards models were used to estimate the hazard ratios, reported as relative risks (RR), of developing stroke between HD survivors and siblings.
RESULTS: Nine siblings reported a stroke, for an incidence of 8.00 per 100,000 person-years (95% CI, 3.85 to 14.43 per 100,000 person-years). Twenty-four HD survivors reported a stroke. The incidence of late-occurring stroke among HD survivors was 83.6 per 100,000 person-years (95% CI, 54.5 to 121.7 per 100,000 person-years). The RR of stroke among HD survivors was 4.32 (95% CI, 2.01 to 9.29; P = .0002). All 24 survivors received mantle radiation exposure (median dose, 40 Gy). The incidence of late-occurring stroke among HD survivors treated with mantle radiation was 109.8 per 100,000 person-years (95% CI, 70.8 to 161.1 per 100,000 person-years). The RR of late-occurring stroke among HD survivors treated with mantle radiation was 5.62 (95% CI, 2.59 to 12.25; P < .0001).
CONCLUSION: Survivors of childhood HD are at increased risk of stroke. Mantle radiation exposure is strongly associated with subsequent stroke. Potential mechanisms may include carotid artery disease or cardiac valvular disease.
INTRODUCTION
The annual incidence rate of Hodgkin's disease (HD) in the United States is 12.1 per million children aged 19 years. As a result, an estimated 850 to 900 children and adolescents in the United States will be diagnosed with HD each year.1 With current therapies, a substantial majority of children with HD will become long-term survivors of their cancer.1,2 During the last two decades, several late effects that may not manifest until many years after diagnosis have been shown to occur at an increased frequency among survivors of childhood HD, including cardiovascular and pulmonary disease, endocrinopathies, and secondary cancers (including breast, thyroid, and colorectal cancers).2-7 Therefore, studies are needed to better understand the incidence of and risk factors for late effects among adult survivors of childhood HD, with the ultimate objective of developing effective strategies to reduce the frequency and severity of adverse effects.
Stroke is the third leading cause of death in the United States and a major cause of severe and permanent disability among those who survive the event. The age- and sex-adjusted 1995 overall incidence rate for occurrence of stroke in the United States was 259 per 100,000 population.8 In middle-aged and older adults, large-vessel atherosclerosis is the primary risk factor for stroke, generally after years of hypertension, insulin resistance, or dyslipidemia. In contrast, strokes are rare events in young adults, with a reported incidence between 10 and 23 per 100,000 per year in individuals aged 18 to 44 years.9-12
Stroke has been reported among survivors of cancers that occur during adulthood, including HD, after therapeutic radiation administered to the neck region, presumably secondary to an acceleration of the atherosclerotic process.7,13 However, an association between mantle or neck radiation therapy and stroke has not been established among individuals exposed to radiation at younger ages. To date, only cranial radiation exposure in childhood or adolescence has been associated with an increased risk of stroke.14,15 The purpose of this study was to determine whether treatment for childhood HD is associated with an increased risk of stroke in young adulthood.
PATIENTS AND METHODS
Patient Selection and Contact
This report from the Childhood Cancer Survivor Study (CCSS) was restricted to individuals who met the following eligibility criteria: diagnosis of HD; diagnosis and initial treatment at one of 26 collaborating CCSS institutions; diagnosis date between January 1, 1970, and December 31, 1986; age younger than 21 years at diagnosis; and survived at least 5 years past diagnosis. Of the 22,125 childhood cancer survivors identified by the collaborating institutions, 3,017 (13.6%) could not be located and were considered lost to follow-up. Among the 19,108 patients located, 14,372 (75.2%) consented to participate and completed a self-report questionnaire. A random sample of 3,846 siblings of participating cancer survivors were also recruited for the study. This analysis included 1,926 HD survivors and 3,846 siblings of childhood cancer survivors.
The CCSS protocols and documents were reviewed for assurances of participant protection and confidentiality and approved by the investigational review board at each collaborating institution. Each participant volunteered informed consent to participate in the study and provided consent for release of medical records. Medical record abstraction, according to a structured protocol, was conducted at each CCSS center and included detailed clinical information about cancer type and treatments received. Baseline data were collected for members of the study cohort using a 24-page questionnaire on enrollment onto the cohort. The questionnaire was designed to capture a wide range of information, including demographic characteristics, health habits, frequency and age of diagnosed medical conditions, surgical procedures, recurrent disease, and second cancers. A follow-up 16-page questionnaire was administered in 2000 to 2002. The baseline and follow-up surveys used in data collection are available for review at http://www.cancer.umn.edu/ccss. A detailed description of the CCSS study design, methods, and cohort characteristics, including comparison of respondents and nonrespondents, is provided elsewhere.16
Statistical Analysis
This analysis had the following three objectives: (1) to estimate the incidence rate of stroke among HD survivors 5 years after diagnosis of HD (henceforth called a late occurrence); (2) to compare the risk of stroke between survivors of HD and siblings of childhood cancer survivors; and (3) to assess the effects of specific treatments on the risk of late-occurring stroke.
In the CCSS study, the occurrence of stroke and age at the first occurrence of stroke were self-reported. The specific type of stroke (eg, thrombotic, embolic, or hemorrhage) was not reported. We defined the date of stroke occurrence as the date that is exactly one-half year past the birthday of the age when the patient had a stroke.
Stroke incidence rates were calculated as the number of patients with an event divided by the number of person-years at risk for stroke. For survivors, person-years at risk were computed beginning on the date 5 years from the primary cancer diagnosis to the date of the first stroke, death, or time of study. For siblings, person-years at risk were calculated from birth to date of the first stroke, death, or time of study.
Potential risk factors examined for an association with stroke included age at time of study, sex, race, vital status, age at diagnosis, interval from diagnosis of cancer, treatment (radiation therapy and chemotherapy, including carmustine, bleomycin, lomustine, chlorambucil, cyclophosphamide, doxorubicin, dacarbazine, nitrogen mustard, corticosteroid [prednisone, dexamethasone, hydroxycortisone, or methylprednisolone], procarbazine, vinblastine, vincristine, and etoposide), history of splenectomy, and selected comorbid conditions (hypertension and diabetes mellitus). Radiation therapy treatment volumes included the neck, chest, abdomen, and pelvis. For the purposes of this article, patients who were treated with supradiaphragmatic radiation volumes including both the neck and chest were considered to have received mantle radiation. Mantle radiation included a standard mantle, modified mantle (without the axillae), or lymph nodes in the mediastinum and neck (with or without the axillae).
Cox proportional hazards models were used to estimate the age-, sex-, and race-adjusted hazard ratio of stroke between HD survivors and CCSS siblings using age as the time axis in the Cox models. Effects of mantle radiation in HD survivors were evaluated using a time-dependent covariate that indicated the starting time of receiving mantle radiation; two hazard ratios for HD survivors, relative to the siblings, were estimated, one for the person-time at risk before mantle radiation and one for the person-time at risk after mantle radiation. Hazard ratios are reported as relative risks (RR). Variances were adjusted for interfamily correlations using sandwich SE estimates.
To evaluate the late effects of specific chemotherapies (eg, doxorubicin and corticosteroids), we added indicator variables for the use of these chemotherapy agents to Cox proportional hazards models. Standard asymptotic inference methods for Cox regression based on the partial likelihood were used to construct 95% CIs and to calculate two-sided significance tests.
RESULTS
Of the 1,926 survivors of childhood HD who were diagnosed between January 1, 1970, and December 31, 1986, 880 (45.7%) were female (Table 1). The mean age at diagnosis was 13.8 years (standard deviation [SD] = 4.3 years), and the interval from diagnosis until time of study was 19.5 years (SD = 5.9 years). The mean age at time of study was 33.8 years (SD = 7.1 years). At time of study, 1,708 participants (89.7%) were alive. Stroke was reported by 26 HD survivors; 24 strokes occurred 5 years after diagnosis of HD, one occurred during therapy, and one occurred without an age at stroke reported (Fig 1). The patients with the stroke during therapy and without a reported age are not considered further in this report. The median interval from diagnosis to stroke was 17.5 years. The median age at time of stroke was 33 years (range, 21 to 45 years). The rate of late-occurring stroke was 83.6 per 100,000 person-years (95% CI, 54.5 to 121.7 per 100,000 person-years).
The comparison population consisted of 3,846 siblings of childhood cancer survivors who were recruited by CCSS and used for analysis of RRs for stroke and risk factors associated with stroke. Of this sibling comparison group, 2,000 (52%) were female. The mean age at time of study was 28.8 years (SD = 9.3 years), which was significantly younger than the HD survivors (P < .0001). Nine siblings of childhood cancer survivors reported a stroke, for a rate of 8.00 per 100,000 person-years (95% CI, 3.85 to 14.43 per 100,000 person-years). The RR of stroke for HD survivors compared with sibling controls was 4.32 (95% CI, 2.01 to 9.29; P = .0002; Table 2) after adjusting for age, sex, and race.
All 24 survivors with late-occurring strokes received radiation therapy to the chest and neck (median dose = 40 Gy; Table 3). The rate of stroke for HD survivors who were treated with mantle radiation was 109.8 per 100,000 person-years (95% CI, 70.8 to 161.0 per 100,000 person-years). Compared with siblings, the RR of late-occurring stroke in HD survivors treated with mantle radiation was 5.62 (95% CI, 2.59 to 12.25; P = < .0001) after adjusting for age at study, sex, and race.
At baseline enrollment onto the CCSS cohort (1995 to 1996), study participants were asked if they had ever smoked, which was defined as smoking more than 100 cigarettes in their lifetime. However, the follow-up contact with participants (2000 to 2002) did not include questions regarding tobacco use. Recognizing this limitation, at baseline enrollment, 15 (63%) of 24 HD survivors who reported a stroke also reported having smoked. In contrast, only 34% (461 of 1,335 patients; available data) of HD survivors without a stroke reported having ever smoked (2 test, P = .0042). The odds ratio for stroke in HD survivors who reported ever smoking compared with survivors who did not report ever smoking was 3.37 (95% CI, 1.21 to 10.77; P = .026). History of splenectomy, treatment with chemotherapy (including doxorubicin), hypertension, diabetes mellitus, and oral contraceptive use by women were not associated with an increased risk of stroke. Of the 24 5-year survivors of childhood HD with late-occurring stroke, 12 reported a cardiac problem, including arrhythmia, valve disease, or congestive heart failure.
Only two of 24 5-year survivors of childhood HD with late-occurring strokes (patients 6 and 7) died. In both cases, these patients' death certificates reported cerebrovascular disease to be at least a contributing cause of death. Because of the study's design, none of the siblings who completed questionnaires died.
DISCUSSION
This study demonstrates a significantly increased risk of late-occurring stroke among young adult survivors of childhood HD. The incidence of late-occurring stroke was 83.6 per 100,000 person-years among HD survivors and 109.8 per 100,000 person-years among patients treated with mantle radiation therapy. Compared with the siblings, the RR of late-occurring stroke among HD survivors was 4.32, and among survivors who were treated with mantle radiation, the RR was 5.62. The incidence of stroke in the sibling comparison group was 8.00 per 100,000 person-years, which is a rate similar to the expected range of 10 to 23 per 100,000 person-years in adults aged 18 to 44 years in the general population.9-12
Two previous studies examined the occurrence of stroke and carotid artery disease after treatment for HD. Both of these studies were largely confined to HD survivors who had received their radiation therapy during their adult years, when comorbid conditions, such as diabetes mellitus and hypertension, can contribute to the vascular disease. Hull et al7 retrospectively evaluated cardiovascular outcomes in 415 2-year survivors of HD who had a median age at HD diagnosis of 25 years (range, 4 to 75 years). They identified 10 patients with strokes that occurred an average of 5.6 years after HD diagnosis.7 For these 10 patients, the median age at time of radiation therapy for HD was 51 years. They also identified 30 patients with clinically asymptomatic carotid and/or subclavian arterial disease (> 40% stenosis by ultrasound or angiography) that was diagnosed at a median of 21 years after HD therapy. The median age of HD therapy for this latter group was 20 years. Hypertension and diabetes were associated with having either symptomatic or asymptomatic carotid and/or subclavian artery disease. The authors concluded that, although vascular disease had been previously reported after high doses of radiation therapy for primary carcinomas of the head and neck, patients with HD, who usually receive lower doses of radiation, are also at an increased risk of radiation-associated atherosclerotic arterial disease. In a study in the early 1980s, Elerding et al17 retrospectively evaluated 910 5-year survivors of cancer who had received neck irradiation, including 247 survivors of HD with a mean age of 28 years at HD diagnosis. The frequency of stroke among HD survivors was not significantly higher than an older comparison population. Elerding et al17 prospectively evaluated 118 5-year or longer survivors who had been treated with radiation therapy to the neck and identified turbulent carotid artery blood flow by carotid phonoangiogram in 17 (22%) of 77 patients studied, suggesting that HD survivors were at increased risk of carotid artery disease. The authors attributed the carotid artery disease to the radiation therapy they received as treatment for their malignancy.
High-dose radiation therapy to the neck has been associated with premature atherosclerosis of the carotid artery and stroke in survivors treated for head and neck cancer. The study by Dorresteijn et al13 of adult survivors of head and neck cancer described a RR of 10.1 for carotid artery occlusive disease and stroke among adult patients treated with radiation therapy for head and neck cancer. Radiation-induced vascular disease is believed to be a result of premature atherosclerotic changes in the arterial wall, including endothelial cell damage, subintimal fibrosis and hypertrophy, thinning and fragmentation of the elastic membrane, and thickening and fibrosis of the muscularis.18 Atherosclerotic injury to the carotid arteries and stroke after radiation therapy have been reported in a variety of adults with malignancies.13,17,19-24
Extrapolating from these studies, one might speculate that the mechanism of stroke in young adult survivors of HD may be a radiation-induced injury to the carotid artery. Alternatively, stroke may also be cardioembolic in etiology, emanating from valvular disease after mantle radiation. Radiation-induced injury to the heart can include both wall motion abnormalities and valvular disease. Twelve of 24 HD survivors in our series who experienced a stroke reported some type of cardiac problem, including valvular disease or congestive heart failure. In the Hull et al7 study of 415 2-year survivors of HD described earlier, 6.2% developed valvular disease at a median of 22 years after treatment. Valvular lesions were more likely to be left sided, with the predominant dysfunction more likely to be stenosis rather than insufficiency.
Our study did not find that chemotherapeutic agents or splenectomy were associated with stroke risk. For example, high cumulative doses of anthracycline chemotherapy are associated with dysrhythmias, cardiomyopathy, and congestive heart failure.25 Stroke, as a result of cerebral embolization of a mural thrombus, has been reported in a patient with doxorubicin-induced cardiomyopathy and congestive heart failure.26 Another potential treatment-related variable examined as a risk factor for stroke included splenectomy. Schilling27 reported an increased frequency of coronary or carotid artery surgery among men with hereditary spherocytosis treated with splenectomy. Likewise, Robinette and Fraumeni24 identified increased deaths as a result of ischemic heart disease among 745 ex-servicemen who had splenectomy for trauma. Potential reasons for this increased risk of stroke and myocardial infarction include increased serum cholesterol concentrations,27 elevated hemoglobin concentrations,28 and increased platelet concentrations after splenectomy.18
Limitations of this report include our inability to validate the accuracy or completeness of self-reported stroke among our study participants. However, the accuracy of self-reported stroke would be expected to be similar in both groups. The sibling comparison group was somewhat younger (mean age, 29 years) than the HD survivors (mean age, 34 years). Supporting the use of this sibling comparison group, the incidence of stroke in the sibling population was in the expected range for young adults in the general population, where stroke in young adulthood is generally not the result of progressive large-vessel atherosclerosis. Also, it is important to recognize that this study only examined the risk of stroke and did not ask about transient ischemic attacks or asymptomatic carotid arterial disease; thus, the overall risk in HD survivors may be underestimated. Additionally, although the patients enrolled onto this study have been observed for a mean of 19.5 years after diagnosis of HD, we are unable to comment on whether the risk for late-occurring stroke will increase as the survivors continue to age. Finally, 60% of the 1,926 HD survivors in this study received a maximum RT dose between 35 and 44.99 Gy, and 79% received a maximum dose between 25 and 54.99 Gy. Current multimodality treatment regimens for HD use more restrained doses and limited radiation fields for children and adolescents with HD compared with those used for patients in this report. It remains to be determined whether stroke and cerebrovascular disease will continue to be observed among HD survivors treated with contemporary therapy, and this needs to be studied further in the future.
Cigarette smoking may be related to an increased risk of stroke among HD survivors. Smoking is undoubtedly recognized as a powerful and independent risk factor for stroke.29,30 Although this report identified a highly significant association between smoking and stroke among HD survivors, we acknowledge that this report has several limitations. Specifically, at the time of baseline enrollment, when study participants were asked about their smoking history, 129 HD survivors (6.7%) were still 21 years of age or younger and, therefore, younger than the usual age when most individuals start smoking. As a result, the proportion of HD survivors who may have started smoking during the 5 years between baseline enrollment and follow-up contact is unknown. Furthermore, we recognize that the optimal method of determining a potential association between smoking with stroke and mantle radiation would be through a proportional hazards model with smoking as a time-dependent covariate. However, because of the limitations in the smoking data, as noted earlier, this was not an option. Instead, a logistic regression, adjusted for age, sex, and race, was performed. As a result, we recognize that there may be a potential association between smoking and subsequent stroke among HD survivors and encourage future studies examining this issue to include details regarding participants' smoking histories in the study design.
In summary, this report by the CCSS is the first study to identify an increased risk of late-occurring stroke in 5-year survivors of childhood HD. Challenges for the future include identifying appropriate methods of screening survivors of HD to identify those most likely to benefit from intervention strategies. Potentially effective screening methods may include interviewing patients regarding symptoms, including amaurosis fugax or transient ischemic attacks, detailed neurologic examinations, determinations of carotid artery intima media thickness, and the identification of genetic predispositions that may predict subsequent stroke among young patients who are treated with radiation therapy.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
NOTES
Supported by grant No. 5U01-CA-55727-05 from the Department of Health and Human Services and funding to the University of Minnesota from the Children's Cancer Research Fund.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Ries LAG, Smith MA, Gurney JG, et al: Cancer Incidence and Survival Among Children and Adolescents: United States SEER Program 1975-1995. Bethesda, MD, National Cancer Institute, Surveillance, Epidemiology, and End Results Program, NIH publication 99-4649, 1999
Marina NM, Greenwald CA, Fairclough DL, et al: Serial pulmonary function studies in children treated for newly diagnosed Hodgkin's disease with mantle radiotherapy plus cycles of doxorubicin, bleomycin, vinblastine, and dacarbazine. Cancer 75:1706-1711, 1995
Bhatia S, Yasui Y, Robison LL, et al: High risk of subsequent neoplasms continues with extended follow-up of childhood Hodgkin's disease: Report from the Late Effects Study Group. J Clin Oncol 21:4386-4394, 2003
Mefferd JM, Donaldson SS, Link MP: Pediatric Hodgkin's disease: Pulmonary, cardiac, and thyroid function following combined modality therapy. Int J Radiat Oncol Biol Phys 16:679-685, 1989
Hancock SL, Tucker MA, Hoppe RT: Factors affecting late mortality from heart disease after treatment of Hodgkin's disease. JAMA 270:1949-1955, 1993
Green DM, Gingell RL, Pearce J, et al: The effect of mediastinal irradiation on cardiac function of patients treated during childhood and adolescence for Hodgkin's disease. J Clin Oncol 5:239-245, 1987
Hull MC, Morris CG, Pepine CJ, et al: Valvular dysfunction and carotid, subclavian, and coronary artery disease in survivors of Hodgkin lymphoma treated with radiation therapy. JAMA 290:2831-2837, 2003
Williams GR, Jiang JG, Matchar DB, et al: Incidence and occurrence of total (first-ever and recurrent) stroke. Stroke 30:2523-2528, 1999
Kristensen B, Malm J, Carlberg B, et al: Epidemiology and etiology of ischemic stroke in young adults aged 18 to 44 years in Northern Sweden. Stroke 28:1702-1709, 1997
Marini C, Totaro R, De Santis F, et al: Stroke in young adults in the community-based L'Aquila Registry: Incidence and prognosis. Stroke 32:52-56, 2001
Naess H, Nyland HI, Thomassen L, et al: Incidence and short-term outcome of cerebral infarction in young adults in Western Norway. Stroke 33:2105-2108, 2002
Jacobs BS, Boden-Albala B, Lin IF, et al: Stroke in the young in the Northern Manhattan Stroke Study. Stroke 33:2789-2793, 2002
Dorresteijn LDA, Kappelle AC, Boogerd W, et al: Increased risk of ischemic stroke after radiotherapy on the neck in patients younger than 60 years. J Clin Oncol 20:282-288, 2002
Bowers DC, Mulne AF, Reisch JS, et al: Nonperioperative strokes in children with central nervous system tumors. Cancer 94:1094-1101, 2002
Gurney JG, Kadan-Lottick NS, Packer RJ, et al: Endocrine and cardiovascular late effects among adult survivors of childhood brain tumors: Childhood Cancer Survivor Study. Cancer 97:663-673, 2003
Robison LL, Mertens AC, Boice JD, et al: Study design and cohort characteristics of the Childhood Cancer Survivor Study: A multi-institutional collaborative project. Med Pediatr Oncol 38:229-239, 2002
Elerding SC, Fernandez RN, Grotta JC, et al: Carotid artery disease following cervical irradiation. Ann Surg 194:609-615, 1981
Conomy JP, Kellermeyer RW: Delayed cerebrovascular consequences of therapeutic radiation: A clinicopathologic study of a stroke associated with radiation-related carotid arteriopathy. Cancer 36:1702-1708, 1975
Call GK, Bray PF, Smoker WRK: Carotid thrombosis following neck irradiation. Int J Radiat Oncol Biol Phys 18:635-640, 1990
Lam WMM, Leung SF, So NMC: Incidence of carotid stenosis in nasopharyngeal carcinoma patients after radiotherapy. Cancer 92:2357-2363, 2001
Cheng SWK, Wu LLH, Ting ACW: Irradiation induced extracranial carotid stenosis in patients with head and neck malignancies. Am J Surg 178:323-328, 1999
Silverberg GD, Britt KH, Goffinet DR: Radiation induced carotid artery disease. Cancer 41:130-137, 1978
Levinson SA, Close MB, Ehrenfeld WK, et al: Carotid artery occlusive disease following external cervical irradiation. Arch Surg 107:395-397, 1973
Robinette CD, Fraumeni JF: Splenectomy and subsequent mortality in veterans of the 1939-45 war. Lancet 8029:127-129, 1977
Swain SM, Whaley FS, Ewer MS: Congestive heart failure in patients treated with doxorubicin: A retrospective analysis of three trials. Cancer 97:2869-2879, 2003
Schachter S, Freeman R: Transient ischemic attack and adriamycin cardiomyopathy. Neurology 32:1380-1381, 1982
Schilling RF: Spherocytosis, splenectomy, strokes, and heart attacks. Lancet 350:1677-1678, 1997
Kannel WB: Current status of the epidemiology of brain infarction associated with occlusive arterial disease. Stroke 2:295-318, 1971
Shinton R, Beevers G: Meta-analysis of relation between cigarette smoking and stroke. BMJ 298:789-794, 1989
Straus SE, Majumdar SR, McAlister FA: New evidence for stroke prevention: Scientific review. JAMA 288:1388-1395, 2002(Daniel C. Bowers, D. Eliz)
The M.D. Anderson Cancer Center, Houston, TX
US Food and Drug Administration, Rockville, MD
Fred Hutchinson Cancer Research Center, Seattle, WA
University of Minnesota, Minneapolis, MN
St Jude Children's Research Hospital, Memphis, TN
Stanford University School of Medicine, Stanford, CA
Children's National Medical Center, Washington, DC, and the Childhood Cancer Survivor Study (see online Appendix)
ABSTRACT
PURPOSE: The objectives of this report are to examine the incidence of and risk factors for stroke among childhood Hodgkin's disease (HD) survivors.
PATIENTS AND METHODS: The Childhood Cancer Survivor Study is a multi-institutional cohort study of more than 5-year cancer survivors diagnosed between 1970 and 1986 and a sibling comparison group. Incidence rates of stroke among HD survivors (n = 1,926) and siblings (n = 3,846) were calculated and compared. Cox proportional hazards models were used to estimate the hazard ratios, reported as relative risks (RR), of developing stroke between HD survivors and siblings.
RESULTS: Nine siblings reported a stroke, for an incidence of 8.00 per 100,000 person-years (95% CI, 3.85 to 14.43 per 100,000 person-years). Twenty-four HD survivors reported a stroke. The incidence of late-occurring stroke among HD survivors was 83.6 per 100,000 person-years (95% CI, 54.5 to 121.7 per 100,000 person-years). The RR of stroke among HD survivors was 4.32 (95% CI, 2.01 to 9.29; P = .0002). All 24 survivors received mantle radiation exposure (median dose, 40 Gy). The incidence of late-occurring stroke among HD survivors treated with mantle radiation was 109.8 per 100,000 person-years (95% CI, 70.8 to 161.1 per 100,000 person-years). The RR of late-occurring stroke among HD survivors treated with mantle radiation was 5.62 (95% CI, 2.59 to 12.25; P < .0001).
CONCLUSION: Survivors of childhood HD are at increased risk of stroke. Mantle radiation exposure is strongly associated with subsequent stroke. Potential mechanisms may include carotid artery disease or cardiac valvular disease.
INTRODUCTION
The annual incidence rate of Hodgkin's disease (HD) in the United States is 12.1 per million children aged 19 years. As a result, an estimated 850 to 900 children and adolescents in the United States will be diagnosed with HD each year.1 With current therapies, a substantial majority of children with HD will become long-term survivors of their cancer.1,2 During the last two decades, several late effects that may not manifest until many years after diagnosis have been shown to occur at an increased frequency among survivors of childhood HD, including cardiovascular and pulmonary disease, endocrinopathies, and secondary cancers (including breast, thyroid, and colorectal cancers).2-7 Therefore, studies are needed to better understand the incidence of and risk factors for late effects among adult survivors of childhood HD, with the ultimate objective of developing effective strategies to reduce the frequency and severity of adverse effects.
Stroke is the third leading cause of death in the United States and a major cause of severe and permanent disability among those who survive the event. The age- and sex-adjusted 1995 overall incidence rate for occurrence of stroke in the United States was 259 per 100,000 population.8 In middle-aged and older adults, large-vessel atherosclerosis is the primary risk factor for stroke, generally after years of hypertension, insulin resistance, or dyslipidemia. In contrast, strokes are rare events in young adults, with a reported incidence between 10 and 23 per 100,000 per year in individuals aged 18 to 44 years.9-12
Stroke has been reported among survivors of cancers that occur during adulthood, including HD, after therapeutic radiation administered to the neck region, presumably secondary to an acceleration of the atherosclerotic process.7,13 However, an association between mantle or neck radiation therapy and stroke has not been established among individuals exposed to radiation at younger ages. To date, only cranial radiation exposure in childhood or adolescence has been associated with an increased risk of stroke.14,15 The purpose of this study was to determine whether treatment for childhood HD is associated with an increased risk of stroke in young adulthood.
PATIENTS AND METHODS
Patient Selection and Contact
This report from the Childhood Cancer Survivor Study (CCSS) was restricted to individuals who met the following eligibility criteria: diagnosis of HD; diagnosis and initial treatment at one of 26 collaborating CCSS institutions; diagnosis date between January 1, 1970, and December 31, 1986; age younger than 21 years at diagnosis; and survived at least 5 years past diagnosis. Of the 22,125 childhood cancer survivors identified by the collaborating institutions, 3,017 (13.6%) could not be located and were considered lost to follow-up. Among the 19,108 patients located, 14,372 (75.2%) consented to participate and completed a self-report questionnaire. A random sample of 3,846 siblings of participating cancer survivors were also recruited for the study. This analysis included 1,926 HD survivors and 3,846 siblings of childhood cancer survivors.
The CCSS protocols and documents were reviewed for assurances of participant protection and confidentiality and approved by the investigational review board at each collaborating institution. Each participant volunteered informed consent to participate in the study and provided consent for release of medical records. Medical record abstraction, according to a structured protocol, was conducted at each CCSS center and included detailed clinical information about cancer type and treatments received. Baseline data were collected for members of the study cohort using a 24-page questionnaire on enrollment onto the cohort. The questionnaire was designed to capture a wide range of information, including demographic characteristics, health habits, frequency and age of diagnosed medical conditions, surgical procedures, recurrent disease, and second cancers. A follow-up 16-page questionnaire was administered in 2000 to 2002. The baseline and follow-up surveys used in data collection are available for review at http://www.cancer.umn.edu/ccss. A detailed description of the CCSS study design, methods, and cohort characteristics, including comparison of respondents and nonrespondents, is provided elsewhere.16
Statistical Analysis
This analysis had the following three objectives: (1) to estimate the incidence rate of stroke among HD survivors 5 years after diagnosis of HD (henceforth called a late occurrence); (2) to compare the risk of stroke between survivors of HD and siblings of childhood cancer survivors; and (3) to assess the effects of specific treatments on the risk of late-occurring stroke.
In the CCSS study, the occurrence of stroke and age at the first occurrence of stroke were self-reported. The specific type of stroke (eg, thrombotic, embolic, or hemorrhage) was not reported. We defined the date of stroke occurrence as the date that is exactly one-half year past the birthday of the age when the patient had a stroke.
Stroke incidence rates were calculated as the number of patients with an event divided by the number of person-years at risk for stroke. For survivors, person-years at risk were computed beginning on the date 5 years from the primary cancer diagnosis to the date of the first stroke, death, or time of study. For siblings, person-years at risk were calculated from birth to date of the first stroke, death, or time of study.
Potential risk factors examined for an association with stroke included age at time of study, sex, race, vital status, age at diagnosis, interval from diagnosis of cancer, treatment (radiation therapy and chemotherapy, including carmustine, bleomycin, lomustine, chlorambucil, cyclophosphamide, doxorubicin, dacarbazine, nitrogen mustard, corticosteroid [prednisone, dexamethasone, hydroxycortisone, or methylprednisolone], procarbazine, vinblastine, vincristine, and etoposide), history of splenectomy, and selected comorbid conditions (hypertension and diabetes mellitus). Radiation therapy treatment volumes included the neck, chest, abdomen, and pelvis. For the purposes of this article, patients who were treated with supradiaphragmatic radiation volumes including both the neck and chest were considered to have received mantle radiation. Mantle radiation included a standard mantle, modified mantle (without the axillae), or lymph nodes in the mediastinum and neck (with or without the axillae).
Cox proportional hazards models were used to estimate the age-, sex-, and race-adjusted hazard ratio of stroke between HD survivors and CCSS siblings using age as the time axis in the Cox models. Effects of mantle radiation in HD survivors were evaluated using a time-dependent covariate that indicated the starting time of receiving mantle radiation; two hazard ratios for HD survivors, relative to the siblings, were estimated, one for the person-time at risk before mantle radiation and one for the person-time at risk after mantle radiation. Hazard ratios are reported as relative risks (RR). Variances were adjusted for interfamily correlations using sandwich SE estimates.
To evaluate the late effects of specific chemotherapies (eg, doxorubicin and corticosteroids), we added indicator variables for the use of these chemotherapy agents to Cox proportional hazards models. Standard asymptotic inference methods for Cox regression based on the partial likelihood were used to construct 95% CIs and to calculate two-sided significance tests.
RESULTS
Of the 1,926 survivors of childhood HD who were diagnosed between January 1, 1970, and December 31, 1986, 880 (45.7%) were female (Table 1). The mean age at diagnosis was 13.8 years (standard deviation [SD] = 4.3 years), and the interval from diagnosis until time of study was 19.5 years (SD = 5.9 years). The mean age at time of study was 33.8 years (SD = 7.1 years). At time of study, 1,708 participants (89.7%) were alive. Stroke was reported by 26 HD survivors; 24 strokes occurred 5 years after diagnosis of HD, one occurred during therapy, and one occurred without an age at stroke reported (Fig 1). The patients with the stroke during therapy and without a reported age are not considered further in this report. The median interval from diagnosis to stroke was 17.5 years. The median age at time of stroke was 33 years (range, 21 to 45 years). The rate of late-occurring stroke was 83.6 per 100,000 person-years (95% CI, 54.5 to 121.7 per 100,000 person-years).
The comparison population consisted of 3,846 siblings of childhood cancer survivors who were recruited by CCSS and used for analysis of RRs for stroke and risk factors associated with stroke. Of this sibling comparison group, 2,000 (52%) were female. The mean age at time of study was 28.8 years (SD = 9.3 years), which was significantly younger than the HD survivors (P < .0001). Nine siblings of childhood cancer survivors reported a stroke, for a rate of 8.00 per 100,000 person-years (95% CI, 3.85 to 14.43 per 100,000 person-years). The RR of stroke for HD survivors compared with sibling controls was 4.32 (95% CI, 2.01 to 9.29; P = .0002; Table 2) after adjusting for age, sex, and race.
All 24 survivors with late-occurring strokes received radiation therapy to the chest and neck (median dose = 40 Gy; Table 3). The rate of stroke for HD survivors who were treated with mantle radiation was 109.8 per 100,000 person-years (95% CI, 70.8 to 161.0 per 100,000 person-years). Compared with siblings, the RR of late-occurring stroke in HD survivors treated with mantle radiation was 5.62 (95% CI, 2.59 to 12.25; P = < .0001) after adjusting for age at study, sex, and race.
At baseline enrollment onto the CCSS cohort (1995 to 1996), study participants were asked if they had ever smoked, which was defined as smoking more than 100 cigarettes in their lifetime. However, the follow-up contact with participants (2000 to 2002) did not include questions regarding tobacco use. Recognizing this limitation, at baseline enrollment, 15 (63%) of 24 HD survivors who reported a stroke also reported having smoked. In contrast, only 34% (461 of 1,335 patients; available data) of HD survivors without a stroke reported having ever smoked (2 test, P = .0042). The odds ratio for stroke in HD survivors who reported ever smoking compared with survivors who did not report ever smoking was 3.37 (95% CI, 1.21 to 10.77; P = .026). History of splenectomy, treatment with chemotherapy (including doxorubicin), hypertension, diabetes mellitus, and oral contraceptive use by women were not associated with an increased risk of stroke. Of the 24 5-year survivors of childhood HD with late-occurring stroke, 12 reported a cardiac problem, including arrhythmia, valve disease, or congestive heart failure.
Only two of 24 5-year survivors of childhood HD with late-occurring strokes (patients 6 and 7) died. In both cases, these patients' death certificates reported cerebrovascular disease to be at least a contributing cause of death. Because of the study's design, none of the siblings who completed questionnaires died.
DISCUSSION
This study demonstrates a significantly increased risk of late-occurring stroke among young adult survivors of childhood HD. The incidence of late-occurring stroke was 83.6 per 100,000 person-years among HD survivors and 109.8 per 100,000 person-years among patients treated with mantle radiation therapy. Compared with the siblings, the RR of late-occurring stroke among HD survivors was 4.32, and among survivors who were treated with mantle radiation, the RR was 5.62. The incidence of stroke in the sibling comparison group was 8.00 per 100,000 person-years, which is a rate similar to the expected range of 10 to 23 per 100,000 person-years in adults aged 18 to 44 years in the general population.9-12
Two previous studies examined the occurrence of stroke and carotid artery disease after treatment for HD. Both of these studies were largely confined to HD survivors who had received their radiation therapy during their adult years, when comorbid conditions, such as diabetes mellitus and hypertension, can contribute to the vascular disease. Hull et al7 retrospectively evaluated cardiovascular outcomes in 415 2-year survivors of HD who had a median age at HD diagnosis of 25 years (range, 4 to 75 years). They identified 10 patients with strokes that occurred an average of 5.6 years after HD diagnosis.7 For these 10 patients, the median age at time of radiation therapy for HD was 51 years. They also identified 30 patients with clinically asymptomatic carotid and/or subclavian arterial disease (> 40% stenosis by ultrasound or angiography) that was diagnosed at a median of 21 years after HD therapy. The median age of HD therapy for this latter group was 20 years. Hypertension and diabetes were associated with having either symptomatic or asymptomatic carotid and/or subclavian artery disease. The authors concluded that, although vascular disease had been previously reported after high doses of radiation therapy for primary carcinomas of the head and neck, patients with HD, who usually receive lower doses of radiation, are also at an increased risk of radiation-associated atherosclerotic arterial disease. In a study in the early 1980s, Elerding et al17 retrospectively evaluated 910 5-year survivors of cancer who had received neck irradiation, including 247 survivors of HD with a mean age of 28 years at HD diagnosis. The frequency of stroke among HD survivors was not significantly higher than an older comparison population. Elerding et al17 prospectively evaluated 118 5-year or longer survivors who had been treated with radiation therapy to the neck and identified turbulent carotid artery blood flow by carotid phonoangiogram in 17 (22%) of 77 patients studied, suggesting that HD survivors were at increased risk of carotid artery disease. The authors attributed the carotid artery disease to the radiation therapy they received as treatment for their malignancy.
High-dose radiation therapy to the neck has been associated with premature atherosclerosis of the carotid artery and stroke in survivors treated for head and neck cancer. The study by Dorresteijn et al13 of adult survivors of head and neck cancer described a RR of 10.1 for carotid artery occlusive disease and stroke among adult patients treated with radiation therapy for head and neck cancer. Radiation-induced vascular disease is believed to be a result of premature atherosclerotic changes in the arterial wall, including endothelial cell damage, subintimal fibrosis and hypertrophy, thinning and fragmentation of the elastic membrane, and thickening and fibrosis of the muscularis.18 Atherosclerotic injury to the carotid arteries and stroke after radiation therapy have been reported in a variety of adults with malignancies.13,17,19-24
Extrapolating from these studies, one might speculate that the mechanism of stroke in young adult survivors of HD may be a radiation-induced injury to the carotid artery. Alternatively, stroke may also be cardioembolic in etiology, emanating from valvular disease after mantle radiation. Radiation-induced injury to the heart can include both wall motion abnormalities and valvular disease. Twelve of 24 HD survivors in our series who experienced a stroke reported some type of cardiac problem, including valvular disease or congestive heart failure. In the Hull et al7 study of 415 2-year survivors of HD described earlier, 6.2% developed valvular disease at a median of 22 years after treatment. Valvular lesions were more likely to be left sided, with the predominant dysfunction more likely to be stenosis rather than insufficiency.
Our study did not find that chemotherapeutic agents or splenectomy were associated with stroke risk. For example, high cumulative doses of anthracycline chemotherapy are associated with dysrhythmias, cardiomyopathy, and congestive heart failure.25 Stroke, as a result of cerebral embolization of a mural thrombus, has been reported in a patient with doxorubicin-induced cardiomyopathy and congestive heart failure.26 Another potential treatment-related variable examined as a risk factor for stroke included splenectomy. Schilling27 reported an increased frequency of coronary or carotid artery surgery among men with hereditary spherocytosis treated with splenectomy. Likewise, Robinette and Fraumeni24 identified increased deaths as a result of ischemic heart disease among 745 ex-servicemen who had splenectomy for trauma. Potential reasons for this increased risk of stroke and myocardial infarction include increased serum cholesterol concentrations,27 elevated hemoglobin concentrations,28 and increased platelet concentrations after splenectomy.18
Limitations of this report include our inability to validate the accuracy or completeness of self-reported stroke among our study participants. However, the accuracy of self-reported stroke would be expected to be similar in both groups. The sibling comparison group was somewhat younger (mean age, 29 years) than the HD survivors (mean age, 34 years). Supporting the use of this sibling comparison group, the incidence of stroke in the sibling population was in the expected range for young adults in the general population, where stroke in young adulthood is generally not the result of progressive large-vessel atherosclerosis. Also, it is important to recognize that this study only examined the risk of stroke and did not ask about transient ischemic attacks or asymptomatic carotid arterial disease; thus, the overall risk in HD survivors may be underestimated. Additionally, although the patients enrolled onto this study have been observed for a mean of 19.5 years after diagnosis of HD, we are unable to comment on whether the risk for late-occurring stroke will increase as the survivors continue to age. Finally, 60% of the 1,926 HD survivors in this study received a maximum RT dose between 35 and 44.99 Gy, and 79% received a maximum dose between 25 and 54.99 Gy. Current multimodality treatment regimens for HD use more restrained doses and limited radiation fields for children and adolescents with HD compared with those used for patients in this report. It remains to be determined whether stroke and cerebrovascular disease will continue to be observed among HD survivors treated with contemporary therapy, and this needs to be studied further in the future.
Cigarette smoking may be related to an increased risk of stroke among HD survivors. Smoking is undoubtedly recognized as a powerful and independent risk factor for stroke.29,30 Although this report identified a highly significant association between smoking and stroke among HD survivors, we acknowledge that this report has several limitations. Specifically, at the time of baseline enrollment, when study participants were asked about their smoking history, 129 HD survivors (6.7%) were still 21 years of age or younger and, therefore, younger than the usual age when most individuals start smoking. As a result, the proportion of HD survivors who may have started smoking during the 5 years between baseline enrollment and follow-up contact is unknown. Furthermore, we recognize that the optimal method of determining a potential association between smoking with stroke and mantle radiation would be through a proportional hazards model with smoking as a time-dependent covariate. However, because of the limitations in the smoking data, as noted earlier, this was not an option. Instead, a logistic regression, adjusted for age, sex, and race, was performed. As a result, we recognize that there may be a potential association between smoking and subsequent stroke among HD survivors and encourage future studies examining this issue to include details regarding participants' smoking histories in the study design.
In summary, this report by the CCSS is the first study to identify an increased risk of late-occurring stroke in 5-year survivors of childhood HD. Challenges for the future include identifying appropriate methods of screening survivors of HD to identify those most likely to benefit from intervention strategies. Potentially effective screening methods may include interviewing patients regarding symptoms, including amaurosis fugax or transient ischemic attacks, detailed neurologic examinations, determinations of carotid artery intima media thickness, and the identification of genetic predispositions that may predict subsequent stroke among young patients who are treated with radiation therapy.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
NOTES
Supported by grant No. 5U01-CA-55727-05 from the Department of Health and Human Services and funding to the University of Minnesota from the Children's Cancer Research Fund.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Ries LAG, Smith MA, Gurney JG, et al: Cancer Incidence and Survival Among Children and Adolescents: United States SEER Program 1975-1995. Bethesda, MD, National Cancer Institute, Surveillance, Epidemiology, and End Results Program, NIH publication 99-4649, 1999
Marina NM, Greenwald CA, Fairclough DL, et al: Serial pulmonary function studies in children treated for newly diagnosed Hodgkin's disease with mantle radiotherapy plus cycles of doxorubicin, bleomycin, vinblastine, and dacarbazine. Cancer 75:1706-1711, 1995
Bhatia S, Yasui Y, Robison LL, et al: High risk of subsequent neoplasms continues with extended follow-up of childhood Hodgkin's disease: Report from the Late Effects Study Group. J Clin Oncol 21:4386-4394, 2003
Mefferd JM, Donaldson SS, Link MP: Pediatric Hodgkin's disease: Pulmonary, cardiac, and thyroid function following combined modality therapy. Int J Radiat Oncol Biol Phys 16:679-685, 1989
Hancock SL, Tucker MA, Hoppe RT: Factors affecting late mortality from heart disease after treatment of Hodgkin's disease. JAMA 270:1949-1955, 1993
Green DM, Gingell RL, Pearce J, et al: The effect of mediastinal irradiation on cardiac function of patients treated during childhood and adolescence for Hodgkin's disease. J Clin Oncol 5:239-245, 1987
Hull MC, Morris CG, Pepine CJ, et al: Valvular dysfunction and carotid, subclavian, and coronary artery disease in survivors of Hodgkin lymphoma treated with radiation therapy. JAMA 290:2831-2837, 2003
Williams GR, Jiang JG, Matchar DB, et al: Incidence and occurrence of total (first-ever and recurrent) stroke. Stroke 30:2523-2528, 1999
Kristensen B, Malm J, Carlberg B, et al: Epidemiology and etiology of ischemic stroke in young adults aged 18 to 44 years in Northern Sweden. Stroke 28:1702-1709, 1997
Marini C, Totaro R, De Santis F, et al: Stroke in young adults in the community-based L'Aquila Registry: Incidence and prognosis. Stroke 32:52-56, 2001
Naess H, Nyland HI, Thomassen L, et al: Incidence and short-term outcome of cerebral infarction in young adults in Western Norway. Stroke 33:2105-2108, 2002
Jacobs BS, Boden-Albala B, Lin IF, et al: Stroke in the young in the Northern Manhattan Stroke Study. Stroke 33:2789-2793, 2002
Dorresteijn LDA, Kappelle AC, Boogerd W, et al: Increased risk of ischemic stroke after radiotherapy on the neck in patients younger than 60 years. J Clin Oncol 20:282-288, 2002
Bowers DC, Mulne AF, Reisch JS, et al: Nonperioperative strokes in children with central nervous system tumors. Cancer 94:1094-1101, 2002
Gurney JG, Kadan-Lottick NS, Packer RJ, et al: Endocrine and cardiovascular late effects among adult survivors of childhood brain tumors: Childhood Cancer Survivor Study. Cancer 97:663-673, 2003
Robison LL, Mertens AC, Boice JD, et al: Study design and cohort characteristics of the Childhood Cancer Survivor Study: A multi-institutional collaborative project. Med Pediatr Oncol 38:229-239, 2002
Elerding SC, Fernandez RN, Grotta JC, et al: Carotid artery disease following cervical irradiation. Ann Surg 194:609-615, 1981
Conomy JP, Kellermeyer RW: Delayed cerebrovascular consequences of therapeutic radiation: A clinicopathologic study of a stroke associated with radiation-related carotid arteriopathy. Cancer 36:1702-1708, 1975
Call GK, Bray PF, Smoker WRK: Carotid thrombosis following neck irradiation. Int J Radiat Oncol Biol Phys 18:635-640, 1990
Lam WMM, Leung SF, So NMC: Incidence of carotid stenosis in nasopharyngeal carcinoma patients after radiotherapy. Cancer 92:2357-2363, 2001
Cheng SWK, Wu LLH, Ting ACW: Irradiation induced extracranial carotid stenosis in patients with head and neck malignancies. Am J Surg 178:323-328, 1999
Silverberg GD, Britt KH, Goffinet DR: Radiation induced carotid artery disease. Cancer 41:130-137, 1978
Levinson SA, Close MB, Ehrenfeld WK, et al: Carotid artery occlusive disease following external cervical irradiation. Arch Surg 107:395-397, 1973
Robinette CD, Fraumeni JF: Splenectomy and subsequent mortality in veterans of the 1939-45 war. Lancet 8029:127-129, 1977
Swain SM, Whaley FS, Ewer MS: Congestive heart failure in patients treated with doxorubicin: A retrospective analysis of three trials. Cancer 97:2869-2879, 2003
Schachter S, Freeman R: Transient ischemic attack and adriamycin cardiomyopathy. Neurology 32:1380-1381, 1982
Schilling RF: Spherocytosis, splenectomy, strokes, and heart attacks. Lancet 350:1677-1678, 1997
Kannel WB: Current status of the epidemiology of brain infarction associated with occlusive arterial disease. Stroke 2:295-318, 1971
Shinton R, Beevers G: Meta-analysis of relation between cigarette smoking and stroke. BMJ 298:789-794, 1989
Straus SE, Majumdar SR, McAlister FA: New evidence for stroke prevention: Scientific review. JAMA 288:1388-1395, 2002(Daniel C. Bowers, D. Eliz)