a Swiss Childhood Cancer Registry, Institute of Social and Preventive Medicine (ISPM), University of Bern, Switzerland
b Department of Paediatrics, Inselspital, Bern University Hospital, University of Bern, Switzerland
c Department of Cardiology, Inselspital, Bern University Hospital, University of Bern, Switzerland
d Department of Paediatrics, University Children’s Hospital, Basel, Switzerland.
BACKGROUND AND AIMS
Cardiovascular diseases (CVD) increase late morbidity and mortality in survivors of acute lymphoblastic leukaemia (ALL). We compared the risk of CVD in ALL survivors to siblings, examined time trends, quantified treatment-related risks, and investigated whether risk extends beyond patients treated with anthracyclines and chest radiotherapy.
The Swiss Childhood Cancer Survivor Study assessed CVD by patient questionnaire in 5-year ALL survivors diagnosed between 1976 and 2005 and their siblings. Participants were asked whether a physician had ever told them that they had hypertension, arrhythmia, heart failure, myocardial infarction, angina pectoris, stroke, thrombosis or valvular problems. We investigated treatment-related risk factors for CVD using multivariable logistic regression, adjusting for demographic and socioeconomic factors, BMI, smoking, diabetes mellitus, alcohol consumption and physical activity.
We contacted 707 survivors and 1299 siblings, 511 (72%) and 709 (55%) of whom responded, respectively. Survivors had a higher risk of developing CVD than siblings (odds ratio [OR] 1.9, 95% confidence interval 1.3–2.8), in particular heart failure (OR 13.9, 1.8–107.4). Compared to patients treated 1976–85, the risk of CVD was 1.4 (0.7–2.8) for those treated 1985–1994 and 1.5 (0.6–3.7) for those treated 1995–2005. The overall CVD risks after anthracycline treatment (OR 3.1, 2.0–4.7), haematopoietic stem cell transplantation (OR 8.0, 2.4–26.9) or relapse (OR 4.1, 1.9–8.8) were increased compared to those of siblings, while the CVD risks of survivors treated without anthracycline or chest radiotherapy were similar (OR 1.0; 0.5–2.0).
Despite attempts to reduce cardiotoxicity in childhood cancer treatment, CVD risks in ALL survivors treated more recently do not seem to have declined.
Trial registration number
Keywords: cardiovascular, late effects, leukaemia, cardiotoxic, heart failure, anthracycline, Swiss Childhood Cancer Registry, Europe
Acute lymphoblastic leukaemia (ALL) is the most common malignancy in childhood. Its incidence has increased in recent decades . Better treatments have improved five-year survival rates to 85 percent [1, 2]. A consequence of these effective treatments is damage to normal tissues, which may result in the long-term dysfunction of many organs. The cardiovascular system is particularly vulnerable to cancer treatment, and cardiovascular diseases (CVD) are recognised late effects of cancer treatment . Survivors of childhood ALL have a lifelong increased risk for cardiovascular morbidity and mortality compared to the general population, mainly caused by chemotherapy with anthracyclines or, to a lesser extent, by direct or scattered chest radiotherapy . ALL survivors also have an increased prevalence of insulin resistance and dyslipidaemia , and differ from their siblings in health behaviours relevant for CVD, such as physical activity, smoking and alcohol consumption [6–9].
In recent decades, cumulative doses of potentially cardiotoxic treatments for ALL have been reduced. Today, low-risk patients receive lower anthracycline doses, and craniospinal radiotherapy is only given to children who relapse [3, 10]. A recent American study reported a reduction of cardiac mortality among ALL survivors , but there are no studies of time trends in non-lethal cardiovascular late effects after childhood cancer. Previous studies of cardiovascular morbidity were mostly restricted to survivors diagnosed before 1990, who were treated with protocols that are no longer in use [4, 11–14]. More recent work has suggested that cardiovascular risk may extend beyond survivors who had anthracyclines or chest radiotherapy . Follow-up care after the end of treatment has been intensified and now includes screening for early CVD.
In the Swiss Childhood Cancer Survivor Study (SCCSS), we assessed physician-diagnosed CVD with patient questionnaires. This study compares the risk of CVD reported by ALL survivors to that of their siblings, and changes in risk based on the calendar period of diagnosis. It also examines treatment-related risk factors for CVD in order to determine whether risk increases are confined to patients exposed to cardiotoxic anthracyclines and chest radiotherapy.
Material and methods
The SCCSS is a population-based, long-term follow-up study of all childhood cancer patients registered in the Swiss Childhood Cancer Registry (SCCR) who have survived for at least five years after cancer diagnosis. Children in the study were diagnosed between 1976 and 2005 and before the age of 20 with leukaemia, lymphoma, central nervous system tumours, malignant solid tumours or Langerhans cell histiocytosis . Between 2007 and 2012, we sent questionnaires (in German, French or Italian) to all survivors. The questionnaire is based on those used in American and British childhood cancer survivor (CCS) studies [17, 18]. It assessed quality of life, health outcomes, current medication and health service use, psychological distress, health behaviour and socioeconomic information. This study includes only participants who were diagnosed with ALL before age 16 years, had survived more than five years after diagnosis, and were 16 years or older and alive at the time of the survey. Those who did not respond to the first mailing received a second copy of the questionnaire. If they again did not respond, we contacted them by phone. Siblings were contacted as a control group. If survivors agreed, we sent a similar questionnaire without cancer-related questions to their siblings. More detailed information on the study design has been published elsewhere .
Ethics approval was granted by the Ethics Committee of the Canton of Bern to the SCCR and SCCSS (KEK-BE: 166/2014) and the SCCSS is registered at ClinicalTrials.gov (identifier: NCT03297034).
We reported the results according to the STROBE guidelines .
The questionnaire included a separate section on CVD, as in the American and British CCS studies (see figure S1 in appendix 1) [17, 18]. We asked survivors and siblings whether they had ever been diagnosed with arterial hypertension, arrhythmia, heart failure or cardiac dysfunction, myocardial infarction, angina pectoris, stroke, deep vein thrombosis or pulmonary embolism, or valvular problems. Arteriosclerosis, though mentioned in the questionnaire, was not analysed as CVD. Reporting at least one CVD was the primary outcome for the analysis. Respondents could use a free text field to describe additional CVD, including pericarditis. Two authors, CEK and JNC, independently recoded all problems mentioned in the free text fields into the CVD categories mentioned and resolved any discrepancies. We coded missing data in outcomes as “no” based on the assumption that people who did not tick “yes” did not have a clinically relevant disease.
We obtained detailed information on sociodemographic characteristics, diagnoses and treatments from the SCCR (see appendix 2). These included gender, nationality, cancer diagnosis, year and age at cancer diagnosis, age at survey, chemotherapy, clinical study participation, treatment protocol, radiotherapy, surgery, haematopoietic stem cell transplantation (HSCT) and history of relapse. Information on follow-up, sociodemographic and behavioural cardiovascular risk factors came from the SCCSS questionnaire.
For better comparison to survivors, we weighted the sibling responses for all analyses according to age, gender, language region and migration background, as previously described [8, 21] (supplementary table S1 in appendix 3).
First (analysis 1), we compared the prevalence of CVD reported by ALL survivors and siblings using standardised χ2 tests. We calculated odds ratios and their 95% confidence intervals (CI) for each outcome, comparing survivors to siblings using weighted univariable logistic regressions.
We then investigated the effects of cancer related variables on our main outcome (≥1 CVD), again using weighted logistic regressions. In the first of these analyses (analysis 2), we investigated whether CVD risks among survivors differed by period of cancer diagnosis (1976–1984, 1985–1994, 1995–2005). This regression model was further adjusted for participation in follow-up care in order to examine any potential effects of detection or recall bias. Survivors attending follow-up care might have been better and/or more recently screened for cardiac problems and therefore report more CVD.
Next (analysis 3), we investigated differences in CVD risk according to treatment-related risk factors (cancer diagnosis, age at diagnosis, history of relapse, chemotherapy, radiotherapy, haematopoietic stem cell transplantation, follow-up care and the different cardiotoxic treatment combinations described above). This analysis included both survivors and siblings (reference group).
Regression models of analyses 2 and 3 were adjusted for recognised cardiovascular risk factors (age, gender, migration background, language region, parents’ highest education, BMI, smoking status, diabetes mellitus, alcohol consumption and physical activity). Given this large number of covariates, we followed an approach proposed by Arbogast et al. and Miettinen to reduce dimensionality [22, 23]. In this approach, adjustment for multiple cardiovascular risk factors is done using a summary disease risk score (DRS). To generate the DRS, the outcome is regressed on all the potential confounders and exposures of interest. The results from this intermediate regression are shown in table S2 (appendix 3). For each participant and for each exposure of interest, the DRS was calculated as the log odds of the outcome predicted from this fitted model while treating the individual as unexposed. In the same way, a DRS was calculated for each participant by treating the individual as simultaneously unexposed to all exposures of interest.
All analyses were performed using the statistical software package Stata (Version 13, Stata Corporation, Austin, Texas).
Characteristics of ALL survivors and siblings
Fifty percent (n = 258) of the 511 ALL survivors were male. Twenty-six percent of survivors were aged 16–20, 49% 21–30, and 25% over 30 at the time of the survey. Cardiovascular risk factors differed between ALL survivors and siblings: ALL survivors had more diabetes mellitus, consumed less alcohol and were less physically active (table S1 in appendix 3). Responders were more often between 21 and 30 years old at survey, female, and diagnosed in earlier time periods than nonresponders. They did not differ by history of relapse or cancer treatment (table S3).
Cardiotoxic treatment has changed in recent decades (table 1). All ALL survivors had received chemotherapy, including 62% (315/511) with anthracyclines. Anthracycline use increased from 42% in 1976–1984, to 71% in 1985–1994, and decreased again to 63% in those diagnosed 1995–2005. Overall, 30% (151/511) of survivors received radiotherapy, and 5% (23/511) radiotherapy to the chest. This last proportion decreased from 9% in 1976–1984 to 4% in 1985–1994 and 1% in 1995–2005. HSCT was performed in 5% (23/511) of all survivors, with no significant change over time.
|ALL Survivors (n = 511)||1976–1984|
(n = 135)
(n = 245)
(n = 131)
|Age at questionnaire (years)||16–20||0||0||47||19||88||67||<0.001|
|41 or more||18||13||0||0||0||0|
|Age at diagnosis (years)||0–4||69||51||140||57||35||27||<0.001|
|10 or more||18||13||39||16||55||42|
|History of relapse||Yes||29||22||26||11||15||11||0.009|
|Chest RT§ <20 Gy||7||5||7||3||0||0|
|Chest RT§ 20–39 Gy||5||3||4||2||1||1|
|Combination of therapies¶||Other CT†, no or other RT‡||77||57||69||28||48||37||<0.001|
|Anthracyclines, no or other RT‡||46||34||165||67||82||63|
|Anthracyclines, chest RT§||11||8||10||4||1||1|
|Still in follow-up care‖||Yes||14||11||44||19||77||63||<0.001|
Risk of CVD in ALL survivors
Compared to siblings, ALL survivors had an increased risk of CVD. Fourteen percent of 511 ALL survivors reported at least one CVD compared to 8% of siblings (OR 1.9, 95% CI 1.3–2.8; p = 0.002) (table 2). Evidence of an increased risk among survivors was strongest for cardiac problems (OR 2.3, CI 1.4–3.8), especially high for heart failure (OR 13.9, CI 1.8–107.4).
(n = 511)
(n = 709)
|Multiple (≥2) CVD||12||2.4||8||1.1||2.1||0.8–5.6||0.142|
|Cardiac problems only‖||47||9.2||29||4.2||2.3||1.4–3.8||0.001|
|Heart failure or cardiomyopathy**||12||2.4||1||0.2||13.9||1.8–107.4||0.012|
|Venous thrombosis/pulmonary embolism**||5||1.0||5||0.7||1.5||0.4–4.9||0.532|
We found no evidence for a time trend in self-reported CVD (fig. 1). As figure 1 illustrates, compared to survivors diagnosed 1976–84, the relative odds of CVD were 1.4 (0.7–2.8) for those diagnosed 1985–94 and 1.5 (0.6–3.7) for those diagnosed 1995–2005 (p = 0.567, Model A). Odds ratios remained qualitatively similar when we adjusted for the attendance of follow-up care (Model B): 1.3 (0.7–2.6) for those diagnosed 1985–1994 and 1.1 (0.4–2.9) for those diagnosed 1995–2005.
Treatment-related risk factors for CVD
Adjusting for the baseline DRS, we found increased risks for developing CVD compared to siblings in ALL survivors treated recently (OR 4.1, CI 2.1–7.9), with a history of relapse (OR 4.1, CI 1.9–8.8), those treated with anthracyclines (OR 3.1, CI 2.0–4.7), and those who had a HSCT (OR 8.0, CI 2.4–26.9) (table 3).
|No. total||No. CVD||%* CVD||OR†||95% CI||p-value‡|
|Diagnosis of ALL||511||73||14.3||2.2||1.5–3.4||<0.001|
|Period of cancer diagnosis||1976–1984||135||19||14.1||1.2||0.6–2.4||0.018|
|Age at diagnosis (years)||0–4||244||29||11.9||2.1||1.3–3.6||0.362|
|10 or more||112||23||20.5||3.0||1.6–5.6|
|History of relapse||No||441||56||12.7||1.9||1.2–3.0||0.047|
|Chemotherapy||Other chemotherapeutic agents§||196||19||9.7||1.2||0.6–2.3||0.005|
|Chest RT‖ 1–19 Gy||14||1||7.1||0.8||0.1–7.5|
|Chest RT‖ 20 or more Gy||10||3||30||8.5||1.4–51.6|
We observed an increased risk of CVD in those treated with both anthracyclines and chest radiotherapy (OR 2.5, CI 0.5–13.7) and in those treated with anthracyclines alone (OR 2.8, CI 1.8–4.5) (fig. 2). ALL survivors treated with chemotherapy other than anthracyclines had a risk similar to that of their siblings (OR 1.0, CI 0.5–2.0).
Swiss ALL survivors had a nearly two-fold greater risk for CVD than siblings overall. Risk was most pronounced for heart failure, and CVD risk did not decrease in survivors treated more recently. Only patients treated with anthracyclines, radiotherapy to the chest or HSCT were at risk for CVD, with little evidence of increased risks of other regimens.
Drawing upon the national SCCSS, these results are representative of all Swiss ALL survivors, conditional on being 5-year survivors and alive. Because our study had nationwide coverage, we believe our results are broadly generalisable for ALL patients from developed countries. Patients in Switzerland are treated according to international protocols used in Europe and North America. Of particular relevance is our inclusion of recently treated patients. A further strength of this study is its systematic use of the disease risk score, which allowed us to adjust for many cardiovascular risk factors simultaneously to obtain more precise effect estimates .
The main limitation of the study is the low number of cases, caused by the young age of participants. This leads to large confidence intervals. A further limitation is that cardiovascular conditions were self-reported without external validation. Some participants could actually have hypertension but not be aware of it because their blood pressure has never been measured. As a sensitivity analysis, we excluded ALL survivors who reported only hypertension from the analysis of risk factors. The directions and strength of the associations did not change (period of cancer 1976–1984, OR 1.7, CI 0.7–3.7; 1985–1994, OR 2.9, CI 1.7–5.1; 1995–2005, OR 3.9, CI 1.9–8.1). However, period of cancer diagnosis was no longer significantly associated (p = 0.22) (see table S5 in appendix 3). We interpret this as a lack of statistical power due to the lower numbers of cardiac outcomes in the analysis. Although this could lead to either under- or overestimation of CVD , good agreement is possible between self-reported CVD and medical records . We did not include deceased cases; however, few patients in our population died from a cardiac cause . We could not retrieve the anthracycline status for 6% of the patients, and their treatment was classified as other chemotherapeutic agent. This might have introduced bias, leading to an underestimation of the effect of anthracyclines and an overestimation of the risk in those exposed to other chemotherapy. We could not grade severity of CVD according to the Common Terminology Criteria for Adverse Events (CTCAE) with the available information .
Overall, the prevalence of CVD in our ALL-survivor cohort is similar to that reported from questionnaire surveys in other childhood cancer survivors such as the assessment of CVD in a large American childhood cancer survivor study that used the same questions . Respective prevalences in that study and ours were broadly comparable: 0.7% and 0.2% for myocardial infarction, 1.3% and 0.2% for pericardial disease, 1.7% and 2.4% for cardiomyopathy, and 1.6% and 1.6% for valvular diseases. Prevalence of CVD in siblings was comparable in both. Hazard ratios (odds ratios) for CVD, comparing ALL survivors with siblings in this American (and our) study were 4.2 (13.9, 95% CI 1.8–107.4) for congestive heart failure, 3.3 (0.6, 95% CI 0.0–6.5) for myocardial infarction, and 2.6 (2.6, 95% CI 0.8–8.0) for pericardial disease and valvular abnormalities.
A Dutch study that assessed CVD during follow-up consultations and by questionnaire reported a prevalence of 3.7% for cardiac events (congestive heart failure, cardiac ischemia, valvular disease, cardiac arrhythmia, pericarditis) in a population of 1,362 CCS with a median age of 29.1 years at study . The reason why this prevalence is lower than our figure of 14.3% might be because the Dutch study only included CVD with a CTCAE grade higher than 2, whereas we might also have included less severe manifestations. Both that study and ours contrast with a recent study that assessed CVD by physical examinations and echocardiography, which reported a prevalence of 7.4% for cardiomyopathy, 3.8% for coronary artery disease, 28% for valvular disease and 4.4% for conduction/rhythm disorders . However, only high-risk patients treated with anthracyclines or chest radiotherapy were included in the study, and subclinical disease was also detected by these examinations.
Our study supports observations that HSCT and anthracycline therapy are risk factors for CVD . We found no evidence suggesting that anthracycline and chest radiotherapy have a synergistic effect, which is consistent with no interaction between radiation dose and anthracycline exposure in relation to cardiac deaths , although our study was underpowered for assessing such interactions. Our results do not support the hypothesis that all cancer treatments, in addition to anthracyclines and chest radiotherapy, lead to an increased risk of CVD .
Over the past 20 years, anthracycline doses have been reduced in low-risk ALL patients, and craniospinal radiotherapy is now limited to patients with relapse. The American childhood cancer survivor study reported a decrease in the incidence of cardiac deaths from 0.6 in 1970–74 to 0.1 in 1990–1994 . It is perhaps surprising, then, that although cardiac mortality has declined, we observed no relative reduction of risk for CVD in recently treated ALL survivors compared to ALL survivors treated earlier. We offer two possible explanations for this finding. First, if the incidence of CVD has remained constant while follow-up care has improved, then we might detect CVD earlier and offer better treatment options, and decreased cardiac mortality ensues. Second, a small group of high-risk ALL-patients has received even higher doses of anthracyclines. Treatment of this small, high-risk group may have offset an otherwise small decline to produce an incidence of CVD that appears little changed. Engagement in more intense follow-up may have led to the detection of more CVD in recent years. In Model B (fig. 1), we adjusted for follow-up care. However, we cannot fully rule out possible residual confounding.
New studies should include clinical assessment of CVD and more recently treated patients, and also assess both morbidity in relation to mortality of CVD to avoid survivor bias. In the clinic, closer, prospective follow-up of those patients at risk may improve detection of subclinical CVD, thus permitting earlier intervention that could reduce the severity of outcomes.
Our national study of ALL survivors demonstrates the higher risk of CVD in ALL survivors treated with known cardiotoxic treatment regimens. Survivors treated with other regimens did not seem to have increased risk compared to siblings. Despite attempts to reduce cardiotoxicity in cancer treatment regimens during the past decades, we found no decrease in CVD risk over time.
We thank all childhood cancer survivors and families for participating in our survey. We thank the study team of the SCCSS (Rahel Kuonen, Grit Sommer, Erika Brantschen-Berclaz, Jana Remliger, Annette Weiss, Annette Schneeberger, Laura Wengenroth, Corina Rueegg, Cornelia Rebholz, Christina Schindera, Nicolas Waespe and Maria Otth), the data managers of the SPOG (Claudia Anderegg, Pamela Balestra, Nadine Beusch, Rosa-Emma Garcia, Franziska Hochreutener, Friedgard Julmy, Nadia Lanz, Rodolfo Lo Piccolo, Heike Markiewicz, Annette Reinberg, Renate Siegenthaler and Verena Stahel) and the team of the SCCR (Verena Pfeiffer, Katharina Flandera, Shelagh Redmond, Meltem Altun, Parvinder Singh, Vera Mitter, Elisabeth Kiraly, Marlen Spring, Christina Krenger and Priska Wölfli). Finally, we thank Christopher Ritter for his editorial assistance.
JN Caccia and EM Hau equally contributed to this work.
This study was supported by Swiss Cancer Research (grant no: 02783-02-2011), the Swiss Cancer League (grant no: 01605-10-2004, 01869-02-2006, 2215-02-2008, 3412-02-2014, 3886-02-2016), the Bernese Cancer League, the Lung League Bern and the Stiftung zur Krebsbekämpfung. The work of the Swiss Childhood Cancer Registry is supported by the Swiss Paediatric Oncology Group (www.spog.ch), Schweizerische Konferenz der kantonalen Gesundheitsdirektorinnen und -direktoren (www.gdk-cds.ch), Swiss Cancer Research (www.krebsforschung.ch), Kinderkrebshilfe Schweiz (www.kinderkrebshilfe.ch), the Federal Office of Public Health (FOPH) and the Institute of Cancer Epidemiology and Registration (www.nicer.org). The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.
Potential competing interests
None of the authors report any conflict of interest related to the study. The commercial funders of the Swiss Childhood Cancer Registry support the daily running of the registry and have not had and will not have any role in the design, conduct, interpretation, or publication of the Swiss Childhood Cancer Registry itself as well as the related research projects.
Claudia E. Kuehni, MD MSc, Institute of Social and Preventive Medicine, University of Bern, Mittelstrasse 43, CH-3012 Bern, claudia.kuehni[at]ispm.unibe.ch
1 Gatta G, Botta L, Rossi S, Aareleid T, Bielska-Lasota M, Clavel J, et al.; EUROCARE Working Group. Childhood cancer survival in Europe 1999-2007: results of EUROCARE-5--a population-based study. Lancet Oncol. 2014;15(1):35–47. doi:. http://dx.doi.org/10.1016/S1470-2045(13)70548-5 PubMed
2 Kuehni CE, Michel G, Sturdy M, Redmond S, Zwahlen M, von der Weid NX. Swiss Childhood Cancer Registry - Annual Report 2004. Bern: Swiss Childhood Cancer Registry; 2005.
3 Lipshultz SE, Adams MJ, Colan SD, Constine LS, Herman EH, Hsu DT, et al.; American Heart Association Congenital Heart Defects Committee of the Council on Cardiovascular Disease in the Young, Council on Basic Cardiovascular Sciences, Council on Cardiovascular and Stroke Nursing, Council on Cardiovascular Radiolo. Long-term cardiovascular toxicity in children, adolescents, and young adults who receive cancer therapy: pathophysiology, course, monitoring, management, prevention, and research directions: a scientific statement from the American Heart Association. Circulation. 2013;128(17):1927–95. doi:. http://dx.doi.org/10.1161/CIR.0b013e3182a88099 PubMed
4 Geenen MM, Bakker PJ, Kremer LC, Kastelein JJ, van Leeuwen FE. Increased prevalence of risk factors for cardiovascular disease in long-term survivors of acute lymphoblastic leukemia and Wilms tumor treated with radiotherapy. Pediatr Blood Cancer. 2010;55(4):690–7. doi:. http://dx.doi.org/10.1002/pbc.22518 PubMed
5 Oeffinger KC, Adams-Huet B, Victor RG, Church TS, Snell PG, Dunn AL, et al.Insulin resistance and risk factors for cardiovascular disease in young adult survivors of childhood acute lymphoblastic leukemia. J Clin Oncol. 2009;27(22):3698–704. doi:. http://dx.doi.org/10.1200/JCO.2008.19.7251 PubMed
6 Rebholz CE, Rueegg CS, Michel G, Ammann RA, von der Weid NX, Kuehni CE, et al.; Swiss Paediatric Oncology Group (SPOG). Clustering of health behaviours in adult survivors of childhood cancer and the general population. Br J Cancer. 2012;107(2):234–42. doi:. http://dx.doi.org/10.1038/bjc.2012.250 PubMed
7 Lown EA, Mertens AC, Korcha RA, Leisenring W, Hudson MM, Greenfield TK, et al.Prevalence and predictors of risky and heavy alcohol consumption among adult siblings of childhood cancer survivors. Psychooncology. 2013;22(5):1134–43. doi:. http://dx.doi.org/10.1002/pon.3121 PubMed
8 Rueegg CS, von der Weid NX, Rebholz CE, Michel G, Zwahlen M, Grotzer M, et al.; Swiss Paediatric Oncology Group (SPOG). Daily physical activities and sports in adult survivors of childhood cancer and healthy controls: a population-based questionnaire survey. PLoS One. 2012;7(4):e34930. doi:. http://dx.doi.org/10.1371/journal.pone.0034930 PubMed
9 Frobisher C, Winter DL, Lancashire ER, Reulen RC, Taylor AJ, Eiser C, et al.; British Childhood Cancer Survivor Study. Extent of smoking and age at initiation of smoking among adult survivors of childhood cancer in Britain. J Natl Cancer Inst. 2008;100(15):1068–81. doi:. http://dx.doi.org/10.1093/jnci/djn210 PubMed
10 Armstrong GT, Chen Y, Yasui Y, Leisenring W, Gibson TM, Mertens AC, et al.Reduction in late mortality among 5-year survivors of childhood cancer. N Engl J Med. 2016;374(9):833–42. doi:. http://dx.doi.org/10.1056/NEJMoa1510795 PubMed
11 Oeffinger KC, Mertens AC, Sklar CA, Kawashima T, Hudson MM, Meadows AT, et al.; Childhood Cancer Survivor Study. Chronic health conditions in adult survivors of childhood cancer. N Engl J Med. 2006;355(15):1572–82. doi:. http://dx.doi.org/10.1056/NEJMsa060185 PubMed
12 Mulrooney DA, Yeazel MW, Kawashima T, Mertens AC, Mitby P, Stovall M, et al.Cardiac outcomes in a cohort of adult survivors of childhood and adolescent cancer: retrospective analysis of the Childhood Cancer Survivor Study cohort. BMJ. 2009;339:b4606. doi:. http://dx.doi.org/10.1136/bmj.b4606 PubMed
13 Bowers DC, Liu Y, Leisenring W, McNeil E, Stovall M, Gurney JG, et al.Late-occurring stroke among long-term survivors of childhood leukemia and brain tumors: a report from the Childhood Cancer Survivor Study. J Clin Oncol. 2006;24(33):5277–82. doi:. http://dx.doi.org/10.1200/JCO.2006.07.2884 PubMed
14 van der Pal HJ, van Dalen EC, van Delden E, van Dijk IW, Kok WE, Geskus RB, et al.High risk of symptomatic cardiac events in childhood cancer survivors. J Clin Oncol. 2012;30(13):1429–37. doi:. http://dx.doi.org/10.1200/JCO.2010.33.4730 PubMed
15 Lipshultz SE, Landy DC, Lopez-Mitnik G, Lipsitz SR, Hinkle AS, Constine LS, et al.Cardiovascular status of childhood cancer survivors exposed and unexposed to cardiotoxic therapy. J Clin Oncol. 2012;30(10):1050–7. doi:. http://dx.doi.org/10.1200/JCO.2010.33.7907 PubMed
16 Michel G, von der Weid NX, Zwahlen M, Redmond S, Strippoli MP, Kuehni CE; Swiss Paediatric Oncology Group (SPOG). Incidence of childhood cancer in Switzerland: the Swiss Childhood Cancer Registry. Pediatr Blood Cancer. 2008;50(1):46–51. doi:. http://dx.doi.org/10.1002/pbc.21129 PubMed
17 Hawkins MM, Lancashire ER, Winter DL, Frobisher C, Reulen RC, Taylor AJ, et al.The British Childhood Cancer Survivor Study: Objectives, methods, population structure, response rates and initial descriptive information. Pediatr Blood Cancer. 2008;50(5):1018–25. doi:. http://dx.doi.org/10.1002/pbc.21335 PubMed
18 Robison LL, Mertens AC, Boice JD, Breslow NE, Donaldson SS, Green DM, et al.Study design and cohort characteristics of the Childhood Cancer Survivor Study: a multi-institutional collaborative project. Med Pediatr Oncol. 2002;38(4):229–39. doi:. http://dx.doi.org/10.1002/mpo.1316 PubMed
19 Kuehni CE, Rueegg CS, Michel G, Rebholz CE, Strippoli MP, Niggli FK, et al.; Swiss Paediatric Oncology Group (SPOG). Cohort profile: the Swiss childhood cancer survivor study. Int J Epidemiol. 2012;41(6):1553–64. doi:. http://dx.doi.org/10.1093/ije/dyr142 PubMed
20 von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. Int J Surg. 2014;12(12):1495–9. doi:. http://dx.doi.org/10.1016/j.ijsu.2014.07.013 PubMed
21 Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res. 2011;46(3):399–424. doi:. http://dx.doi.org/10.1080/00273171.2011.568786 PubMed
22 Arbogast PG, Kaltenbach L, Ding H, Ray WA. Adjustment for multiple cardiovascular risk factors using a summary risk score. Epidemiology. 2008;19(1):30–7. doi:. http://dx.doi.org/10.1097/EDE.0b013e31815be000 PubMed
23 Miettinen OS. Stratification by a multivariate confounder score. Am J Epidemiol. 1976;104(6):609–20. doi:. http://dx.doi.org/10.1093/oxfordjournals.aje.a112339 PubMed
24 Casagranda L, Trombert-Paviot B, Faure-Conter C, Bertrand Y, Plantaz D, Berger C. Self-reported and record-collected late effects in long-term survivors of childhood cancer: a population-based cohort study of the childhood cancer registry of the Rhône-Alpes region (ARCERRA). Pediatr Hematol Oncol. 2013;30(3):195–207. doi:. http://dx.doi.org/10.3109/08880018.2013.772682 PubMed
25 Louie AD, Robison LL, Bogue M, Hyde S, Forman SJ, Bhatia S. Validation of self-reported complications by bone marrow transplantation survivors. Bone Marrow Transplant. 2000;25(11):1191–6. doi:. http://dx.doi.org/10.1038/sj.bmt.1702419 PubMed
26 Schindler M, Spycher BD, Ammann RA, Ansari M, Michel G, Kuehni CE; Swiss Paediatric Oncology Group (SPOG). Cause-specific long-term mortality in survivors of childhood cancer in Switzerland: A population-based study. Int J Cancer. 2016;139(2):322–33. doi:. http://dx.doi.org/10.1002/ijc.30080 PubMed
27 Trotti A, Colevas AD, Setser A, Rusch V, Jaques D, Budach V, et al.CTCAE v3.0: development of a comprehensive grading system for the adverse effects of cancer treatment. Semin Radiat Oncol. 2003;13(3):176–81. doi:. http://dx.doi.org/10.1016/S1053-4296(03)00031-6 PubMed
28 Mulrooney DA, Armstrong GT, Huang S, Ness KK, Ehrhardt MJ, Joshi VM, et al.Cardiac outcomes in adult survivors of childhood cancer exposed to cardiotoxic therapy: a cross-sectional study. Ann Intern Med. 2016;164(2):93–101. doi:. http://dx.doi.org/10.7326/M15-0424 PubMed
29 Tukenova M, Guibout C, Oberlin O, Doyon F, Mousannif A, Haddy N, et al.Role of cancer treatment in long-term overall and cardiovascular mortality after childhood cancer. J Clin Oncol. 2010;28(8):1308–15. doi:. http://dx.doi.org/10.1200/JCO.2008.20.2267 PubMed
Questions on cardiovascular problems in the Swiss Childhood Cancer Survivor Study
The questions on cardiovascular problems in the Swiss Childhood Cancer Survivor Study are shown in figure S1.
Appendix 2: Explanatory variables from the Swiss Childhood Cancer Survivor Study (SCCSS)
We obtained detailed information on sociodemographic characteristics, diagnoses and treatments from the SCCR. These included gender, nationality, cancer diagnosis, year and age at cancer diagnosis, age at survey, chemotherapy (yes/no), clinical study participation (yes/no, study protocol, treatment arm), treatment protocol, radiotherapy (yes/no, area, dose), surgery (yes/no, area, type), haematopoietic stem cell transplantation (yes, including autologous or allogeneic/no), and history of relapse (yes/no). We determined anthracycline and other chemotherapeutic agent use (yes/no) through information on clinical study participation and treatment protocol from the SCCR. Other chemotherapeutic agents included any chemotherapy not containing anthracyclines and chemotherapy with unknown details. We classified radiotherapy into four categories: No radiotherapy, Other radiotherapy, Chest radiotherapy 1-19 Gray (Gy) and Chest radiotherapy 20 Gy or more. Chest radiotherapy included total body irradiation, mantle field irradiation or irradiation to the thorax, mediastinum, or thoracic spine. Cardiotoxic treatment was divided into other chemotherapeutic agents and/or other radiotherapy, other chemotherapeutic agents and chest radiotherapy, anthracyclines and other radiotherapy, and anthracyclines and chest radiotherapy. We also categorised period of diagnosis (1976-84, 1985-94, ≥1995), age at diagnosis (0-4, 5-9, ≥10 years) and age at questionnaire (16-20, 21-30, 31-40, or 41 or more years).
Information on follow-up, sociodemographic and behavioural cardiovascular risk factors came from the SCCSS questionnaire. We used data on parents’ highest education (primary, secondary, tertiary), migration background (yes/no), smoking status (current/ever), height (cm) and weight (kg), diabetes mellitus (yes/no), alcohol consumption (Occasionally/never, >=1 standard drink weekly, not daily, 1 standard drink daily and >1 standard drink daily) and physical activity (yes/no). We asked survivors if they were still under regular follow-up care for their former childhood cancer (still in follow-up care/not in follow-up care). We coded migration status as yes if a participant was born abroad, had no Swiss citizenship at birth, or had at least one parent with no Swiss citizenship. We defined participants as physically active if they reported being engaged in any fitness training or sports. We calculated the body mass index (BMI) for each participant (kg/m2) and divided it into four categories: underweight, BMI <18; normal weight, BMI 18-24.9; overweight, BMI 25-30; and obese, BMI >30.
Appendix 3: Supplementary data
(n = 511)
(n = 709)
|Age at questionnaire (years)|
|‒ 41 or more||18||(3.5)||75||(10.6)||<0.001||(3.5)||n.a.a|
|‒ No migration background||466||(91.2)||679||(95.8)||(76.7)|
|‒ Migration background||45||(8.8)||26||(4.2)||<0.001||(23.3)||n.a.a|
|‒ French or Italian||127||(24.9)||118||(16.6)||<0.001||(23.2)||n.a.a|
|Parents' highest educationf|
|‒ Primary education||55||(10.8)||62||(8.7)||(7.3)|
|Body mass index (kg/m2)|
|‒ Underweight (<18)||26||(5.4)||16||(2.3)||(2.9)|
|‒ Normal weight (18–24.9)||336||(69.1)||499||(71.2)||(75.0)|
|‒ Overweight (25–30)||102||(21.0)||149||(21.3)||(18.6)|
|‒ Obese (>30)||22||(4.5)||37||(5.3)||0.043||(3.5)||0.098|
|Alcohol consumption(standard drink)|
|‒ ≥1 weekly, not daily||261||(53.7)||422||(60.4)||(66.0)|
(n = 1122)
|nb||(%)c||Odds ratio||(95% CI)|
|Age at questionnaire (years)|
|‒ 41 or more||15||(16)||7.9||(2.9–21.4)|
|‒ No migration background||105||(11)||1.0|
|‒ Migration background||33||(15)||1.9||(1.0–3.4)|
|‒ French or Italian||23||(9)||0.7||(0.3–1.2)|
|Parents' highest educationf|
|‒ Primary education||97||(12)||1.0|
|Body mass index (kg/m2)|
|‒ Underweight (<18)||6||(14)||1.1||(0.5–2.5)|
|‒ Normal weight (18–24.9)||88||(11)||1.0|
|‒ Overweight (25–30)||28||(11)||1.1||(0.6–2.0)|
|‒ Obese (>30)||12||(20)||1.7||(0.7–4.0)|
|‒ Yes||87||(11)||1.0||omitted d|
|Alcohol consumption (standard drink)|
|‒ ≥1 weekly, not daily||70||(10)||1.0||(0.6–1.7)|
|Year at diagnosis|
|‒ 1995–2005||18||(14)||1.0||omitted d|
|Age at diagnosis|
|‒ ≥10||23||(21)||1.0||omitted d|
|‒ No history of relapse||56||(13)||0.6||(0.2–1.4)|
|‒ Relapse||17||(24)||1.0||omitted d|
|‒ Other chemotherapy||19||(10)||0.6||(0.3–1.0)|
|‒ Anthracycline||54||(17)||1.0||omitted d|
|‒ No RT||48||(13)||0.7||(0.1–4.9)|
|‒ Other RT||21||(16)||0.7||(0.1–4.8)|
|‒ RT heart <20 Gy||1||(7)||0.1||(0.0–2.2)|
|‒ RT heart 20–39 Gy||3||(30)||1.0||omitted d|
|‒ No HSCT||66||(14)||0.3||(0.1–1.4)|
|‒ HSCT||7||(29)||1.0||omitted d|
|‒ No follow–up||48||(14)||0.7||(0.3–1.4)|
|‒ Follow–up||24||(18)||1.0||omitted d|
|Age at questionnaire (years)||0.043|
|‒ 41 or more||18||(4)||8||(4)|
|‒ French or Italian||127||(25)||127||(27)|
|Cancer related characteristics|
|Age at diagnosis (years)|
|‒ 10 or more||112||(22)||39||(20)|
|Year at diagnosis||0.005|
|History of relapse||0.104|
|No FU||% a||FU||%||No FU||%||FU||%||No FU||%||FU||%|
|Age at questionnaire (years)|
|40 or more||16||13.4||2||14.3|
|French or Italian||25||21.0||1||7.1||47||24.6||9||20.5||18||40.0||22||28.6|
|Cancer related characteristics|
|Age at diagnosis (years)|
|10 or more||18||15.1||0||0.0||29||15.2||10||22.7||18||40.0||34||44.2|
|History of relapse|
|No. total||No. CVD||%a CHD||ORb||95% CI||p-valuec|
|Diagnosis of ALL||511||54||10.5||2.6||1.6–4.3||0.013|
|Period of cancer diagnosis||0.219|
|Age at diagnosis (years)||0.431|
|10 or more||112||16||14.3||3.3||1.7–6.8|
|History of relapse||0.078|
|Other chemotherapeutic agentsd||196||13||6.6||1.3||0.6–2.9|
|Chest RTf 1–19 Gy||14||1||7.1||1.6||0.2–13.6|
|Chest RTf 20 or more Gy||10||3||30.0||13.5||2.5–72.6|
Published under the copyright license CC BY-NC-SA: This license allows reusers to distribute, remix, adapt, and build upon the material in any medium or format for noncommercial purposes only, and only so long as attribution is given to the creator. If you remix, adapt, or build upon the material, you must license the modified material under identical terms.