Advertisement

Cardiac function in childhood cancer survivors treated with vincristine: Echocardiographic results from the DCCSS LATER 2 CARD study

  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Remy Merkx
    Correspondence
    Corresponding authors at: P.O. Box 9101, 6500 HB Nijmegen, the Netherlands.
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Department of Medical Imaging/Radiology, Medical UltraSound Imaging Centre, Radboud Institute for Health Sciences, Radboud university medical center, Nijmegen, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    E. (Lieke) A.M. Feijen
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Princess Máxima Center for Pediatric Oncology, Utrecht, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Jan M. Leerink
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Department of Clinical and Experimental Cardiology, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Esmée C. de Baat
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Princess Máxima Center for Pediatric Oncology, Utrecht, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Louise Bellersen
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Department of Cardiology, Radboud university medical center, Nijmegen, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Elvira C. van Dalen
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Princess Máxima Center for Pediatric Oncology, Utrecht, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Eline van Dulmen-den Broeder
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Department of Pediatric Oncology, Amsterdam UMC, VU University, Amsterdam, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Margriet van der Heiden-van der Loo
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Trial- and Data Center, Princess Máxima Center for Pediatric Oncology, Utrecht, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Marry M. van den Heuvel-Eibrink
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Princess Máxima Center for Pediatric Oncology, Utrecht, the Netherlands

    Department of Pediatric Oncology, Sophia Children's Hospital, Erasmus Medical Center, Rotterdam, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Chris L. de Korte
    Correspondence
    Corresponding authors at: P.O. Box 9101, 6500 HB Nijmegen, the Netherlands.
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Department of Medical Imaging/Radiology, Medical UltraSound Imaging Centre, Radboud Institute for Health Sciences, Radboud university medical center, Nijmegen, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Jacqueline Loonen
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Department of Haematology, Radboud university medical center, Nijmegen, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Marloes Louwerens
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Department of Internal Medicine, Leiden University Medical Center, Leiden, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Cécile M. Ronckers
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Princess Máxima Center for Pediatric Oncology, Utrecht, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Arco J. Teske
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Department of Cardiology, University Medical Center Utrecht, Utrecht, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Wim J.E. Tissing
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Princess Máxima Center for Pediatric Oncology, Utrecht, the Netherlands

    Department of Pediatric Oncology, University of Groningen, University Medical Center Groningen, Groningen, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Andrica C.H. de Vries
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Princess Máxima Center for Pediatric Oncology, Utrecht, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Annelies M.C. Mavinkurve-Groothuis
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Princess Máxima Center for Pediatric Oncology, Utrecht, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Helena J.H. van der Pal
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Princess Máxima Center for Pediatric Oncology, Utrecht, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Gert Weijers
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Department of Medical Imaging/Radiology, Medical UltraSound Imaging Centre, Radboud Institute for Health Sciences, Radboud university medical center, Nijmegen, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Wouter E.M. Kok
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Department of Clinical and Experimental Cardiology, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Leontien C.M. Kremer
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Princess Máxima Center for Pediatric Oncology, Utrecht, the Netherlands
    Search for articles by this author
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Livia Kapusta
    Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
    Affiliations
    Department of Pediatric Cardiology, Amalia Children's Hospital, Radboud University Medical Center, Nijmegen, the Netherlands.

    Department of Paediatrics, Pediatric Cardiology Unit, Tel Aviv Sourasky Medical Center, Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel.
    Search for articles by this author
  • on behalf of the Dutch LATER Study Group
  • Author Footnotes
    # All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation
Open AccessPublished:July 31, 2022DOI:https://doi.org/10.1016/j.ijcard.2022.07.049

      Highlights

      • Echocardiographic GLS measurement may detect cardiac dysfunction in an early stage
      • Vincristine-treated childhood cancer survivors show reduced GLS
      • No dose-response relation was present between vincristine dose and LVEF or GLS
      • Diastolic blood pressure and body mass index contributed to lower GLS
      • Whether these survivors are at risk for cardiovascular events should be elucidated

      Abstract

      Background

      Anthracyclines and radiotherapy involving the heart region are cardiotoxic, but the potential cardiotoxicity of vincristine remains unknown. We assessed cardiac function in vincristine-treated >5-year childhood cancer survivors (CCS).

      Methods and results

      We cross-sectionally compared echocardiograms of 101 vincristine-treated CCS (median age 35 years [range: 17–53], median vincristine dose 63 mg/m2) from the national Dutch Childhood Cancer Survivor Study, LATER cohort, to 101 age- and sex-matched controls. CCS treated with anthracyclines, radiotherapy involving the heart region, cyclophosphamide or ifosfamide were excluded. Twelve CCS (14%) versus four controls (4%; p 0.034) had a decreased left ventricular ejection fraction (LVEF; men <52%, women <54%). Mean LVEF was 58.4% versus 59.7% (p 0.050). Global longitudinal strain (GLS) was abnormal in nineteen (24%) CCS versus eight controls (9%; p 0.011). Mean GLS was 19.0% versus 20.1% (p 0.001). No ≥grade 2 diastolic dysfunction was detected. In multivariable logistic regression analysis CCS had higher risk of abnormal GLS (OR 3.55, p 0.012), but not abnormal LVEF (OR 3.07, p 0.065), than controls. Blood pressure and smoking history contributed to variation in LVEF, whereas obesity and diastolic blood pressure contributed to variation in GLS. Cumulative vincristine dose was not associated with either abnormal LVEF or abnormal GLS in multivariable models corrected for age and sex (OR per 50 mg/m2: 0.88, p 0.85 and 1.14, p 0.82, respectively).

      Conclusions

      Vincristine-treated long-term CCS showed an abnormal GLS more frequently than controls. Their risk for future clinical cardiac events and the role of risk factor modification should be further elucidated.

      Graphical abstract

      Keywords

      1. Introduction

      Cardiotoxicity is an important side-effect of treatments for childhood cancer. Heart failure due to cardiotoxic treatments can become overt in childhood cancer survivors (CCS) even after decades. Anthracyclines and radiotherapy involving the heart region are the main treatment-related risk factors [
      • Leerink J.M.
      • de Baat E.C.
      • Feijen E.A.M.
      • Bellersen L.
      • Dalen E.C.
      • Grotenhuis H.B.
      • et al.
      Cardiac disease in childhood cancer survivors.
      ]. Mortality due to heart failure in CCS is six-fold higher, compared to the general population [
      • Fidler M.M.
      • Reulen R.C.
      • Henson K.
      • Kelly J.
      • Cutter D.
      • Levitt G.A.
      • et al.
      Population-based long-term cardiac-specific mortality among 34 489 five-year survivors of childhood Cancer in Great Britain.
      ]. Therefore, a surveillance guideline recommends regular and life-long echocardiography for CCS at risk, to detect left ventricular (LV) systolic dysfunction at an asymptomatic stage [
      • Armenian S.H.
      • Hudson M.M.
      • Mulder R.L.
      • Chen M.H.
      • Constine L.S.
      • Dwyer M.
      • et al.
      Recommendations for cardiomyopathy surveillance for survivors of childhood cancer: a report from the international late effects of childhood Cancer guideline harmonization group.
      ].
      Vincristine, a tubulin-binding drug from the vinca-alkaloid group, is known for its dose-limiting neurotoxicity [
      • Madsen M.L.
      • Due H.
      • Ejskjær N.
      • Jensen P.
      • Madsen J.
      • Dybkær K.
      Aspects of vincristine-induced neuropathy in hematologic malignancies: a systematic review.
      ], but cardiotoxicity due to vincristine has been topic of debate. Coronary vascular events after administration of vinca-alkaloids have been reported in case reports on adult patients, often with known cardiovascular risk [
      • Gros R.
      • Hugon V.
      • Thouret J.M.
      • Peigne V.
      Coronary spasm after an injection of vincristine.
      ,
      • Taniguchi T.
      • Nakamura T.
      • Sawada T.
      Arterial stiffness, endothelial dysfunction and recurrent angina post-chemotherapy.
      ,
      • Stoter G.
      • Koopman A.
      • Vendrik C.P.
      • Struyvenberg A.
      • Sleyfer D.T.
      • Willemse P.H.
      • et al.
      Ten-year survival and late sequelae in testicular cancer patients treated with cisplatin, vinblastine, and bleomycin.
      ,
      • Roy A.
      • Khanna N.
      • Senguttuvan N.B.
      Rituximab-vincristine chemotherapy-induced acute anterior wall myocardial infarction with cardiogenic shock.
      ,
      • Goli A.K.
      • Osman M.N.
      • Koduri M.
      • Byrd R.P.
      • Roy T.M.
      A case report of vinorelbine monotherapy-related acute bronchospasm and non-ST elevation acute coronary syndrome.
      ]. In long-term CCS, however, vincristine was not shown to entail higher risk of late self-reported myocardial infarction, when adjusted for cardiac radiation dose [
      • Mulrooney D.A.
      • Yeazel M.W.
      • Kawashima T.
      • Mertens A.C.
      • 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.
      ]. Our previous work showed that CCS who developed clinical heart failure more often had received vincristine than CCS without heart failure, but vincristine dose was not associated with incident heart failure after correction for anthracyclines and radiotherapy involving the heart region [
      • Feijen E.
      • Font-Gonzalez A.
      • Van der Pal H.J.H.
      • Kok W.E.M.
      • Geskus R.B.
      • Ronckers C.M.
      • et al.
      Risk and temporal changes of heart failure among 5-year childhood Cancer survivors: a DCOG-LATER study.
      ]. A single-centre Dutch echocardiography study in CCS showed a near-significant association between vincristine exposure and reduced fractional shortening. However, this study included high-risk CCS treated with either anthracyclines, radiotherapy involving the heart region, or the potentially cardiotoxic high-dose cyclophosphamide or ifosfamide, which may have attenuated the risk specific to vincristine in multivariable analysis [
      • van der Pal H.J.H.
      • van Dalen E.C.
      • Hauptmann M.
      • Kok W.E.M.
      • Caron H.N.
      • van den Bos C.
      • et al.
      Cardiac function in 5-year survivors of childhood cancer: a long-term follow-up study.
      ].
      Since vincristine is still frequently used to treat childhood cancer, we investigated the association between vincristine treatment and LV dysfunction on echocardiography, in the cardiology study of the Dutch Childhood Cancer Survivor Study, LATER cohort (1963–2001) part 2; clinical visit & questionnaire study (DCCSS LATER 2 CARD). Since we aimed to evaluate the effects of vincristine, we excluded CCS who received other potentially cardiotoxic therapy as well.

      2. Material and methods

      2.1 Study population

      A detailed description of the DCCSS LATER 2 CARD study cohort and methodology has been published [
      • Leerink J.M.
      • Feijen E.A.M.
      • van der Pal H.J.H.
      • Kok W.E.M.
      • Mavinkurve-Groothuis A.M.C.
      • Kapusta L.
      • et al.
      Diagnostic tools for early detection of cardiac dysfunction in childhood cancer survivors: methodological aspects of the Dutch late effects after childhood cancer (LATER) cardiology study.
      ]. In brief, a nationwide cohort of ≥5-year CCS treated with potentially cardiotoxic therapies from 1963 to 2001 before the age of 18 years, along with an unexposed sibling cohort, were prospectively recruited between 2016 and 2020 for a cross-sectional outpatient clinic evaluation, including echocardiography. The current study focuses on CCS treated with vincristine as the only potentially cardiotoxic agent. CCS who also received anthracyclines, radiotherapy involving the heart region, cyclophosphamide, or ifosfamide were excluded. Since CCS in this exploratory study-arm do not undergo regular echocardiographic surveillance, we included an arbitrary maximum of 100 CCS. This number was slightly exceeded due to concurrent invitation at multiple centres. To account for the background cardiovascular risk, for each CCS, the ‘nearest’ control subject in terms of age and sex was selected from the LATER CARD sibling cohort in a 1:1 ratio, using propensity score matching. As these were not necessarily all siblings of the vincristine-treated CCS included in the current study, they are further referred to as untreated control group and no paired statistics were performed. Participants with congenital heart disease were excluded. All participants gave their informed consent for use of the study and historical data and the medical ethic boards of all centres approved the study protocol according to the declaration of Helsinki.

      2.2 Data collection

      Cancer diagnosis and treatment history, including cumulative vincristine dose, were previously obtained retrospectively and available from our central registry. The number of vincristine administrations was noted when registered.
      Before their visit, participants completed questionnaires on medical history, cardiovascular risk factors and medication use. Chronic cardiovascular medication use was coded according to the Anatomical Therapeutic Chemical classification. Self-reported history of heart failure, myocardial infarction, hypertension and diabetes were validated against use of appropriate medication. Use of lipid lowering medication was also noted. Being overweight was defined as a body mass index >25.
      For 78 CCS and 86 controls that participated in the DCCSS LATER study part 1, self-reported smoking history was available at a mean of, respectively, 4.3 and 3.5 years before the current participation. New participants at the current study completed more extensive questionnaires including smoking history. A participant was considered to ever have smoked when having smoked ≥1 cigarette/week for ≥1 year. At the outpatient clinic, length, weight and resting blood pressure were recorded.
      A detailed echocardiography protocol was followed to ensure image quality and frame rate [
      • Merkx R.
      • Leerink J.M.
      • Feijen E.A.M.
      • Kremer L.C.M.
      • de Baat E.C.
      • Bellersen L.
      • et al.
      Echocardiography protocol for early detection of cardiac dysfunction in childhood cancer survivors in the multicenter DCCSS LATER 2 CARD study: design, feasibility, and reproducibility.
      ]. Two physicians (RM, JL) centrally performed standard structural and functional measurements, including biplane ejection fraction (LVEF) and diastolic function measurements [
      • Lang R.M.
      • Badano L.P.
      • Mor-Avi V.
      • Afilalo J.
      • Armstrong A.
      • Ernande L.
      • et al.
      Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
      ,
      • Nagueh S.F.
      • Smiseth O.A.
      • Appleton C.P.
      • Byrd 3rd, B.F.
      • Dokainish H.
      • Edvardsen T.
      • et al.
      Recommendations for the evaluation of left ventricular diastolic function by echocardiography: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
      ]. Strain analyses were performed separately in vendor-independent software (2D CPA 1.4, TomTec Imaging Systems, Germany). End-systolic midwall global longitudinal strain (GLS) was calculated from three apical views [
      • Merkx R.
      • Leerink J.M.
      • Feijen E.A.M.
      • Kremer L.C.M.
      • de Baat E.C.
      • Bellersen L.
      • et al.
      Echocardiography protocol for early detection of cardiac dysfunction in childhood cancer survivors in the multicenter DCCSS LATER 2 CARD study: design, feasibility, and reproducibility.
      ,
      • Sugimoto T.
      • Dulgheru R.
      • Bernard A.
      • Ilardi F.
      • Contu L.
      • Addetia K.
      • et al.
      Echocardiographic reference ranges for normal left ventricular 2D strain: results from the EACVI NORRE study.
      ,
      • Voigt J.U.
      • Pedrizzetti G.
      • Lysyansky P.
      • Marwick T.H.
      • Houle H.
      • Baumann R.
      • et al.
      Definitions for a common standard for 2D speckle tracking echocardiography: consensus document of the EACVI/ASE/industry task force to standardize deformation imaging.
      ]. GLS-rate represents the steepest slope of the GLS curve during systole. Global circumferential strain was obtained from the mid-ventricular parasternal short axis view (mid-GCS) and averaged over six segments. Reproducibility and feasibility measures of the primary outcome measurements were previously published. Measurement feasibility was slightly lower in CCS than controls and generally in accordance with the literature [
      • Merkx R.
      • Leerink J.M.
      • Feijen E.A.M.
      • Kremer L.C.M.
      • de Baat E.C.
      • Bellersen L.
      • et al.
      Echocardiography protocol for early detection of cardiac dysfunction in childhood cancer survivors in the multicenter DCCSS LATER 2 CARD study: design, feasibility, and reproducibility.
      ].
      Qualitative references regarding GLS and mid-GCS concern the absolute values (i.e. -18% is lower and worse than −20%). Positive correlations indicate worsening (lower absolute) strain value.

      2.3 Outcomes

      Primary outcomes were the prevalence of an LVEF below normal (men <52%, women <54%), GLS below software-specific cut-off values, or relevant LV diastolic dysfunction defined as ≥grade 2 [
      • Lang R.M.
      • Badano L.P.
      • Mor-Avi V.
      • Afilalo J.
      • Armstrong A.
      • Ernande L.
      • et al.
      Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
      ,
      • Nagueh S.F.
      • Smiseth O.A.
      • Appleton C.P.
      • Byrd 3rd, B.F.
      • Dokainish H.
      • Edvardsen T.
      • et al.
      Recommendations for the evaluation of left ventricular diastolic function by echocardiography: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
      ,
      • Sugimoto T.
      • Dulgheru R.
      • Bernard A.
      • Ilardi F.
      • Contu L.
      • Addetia K.
      • et al.
      Echocardiographic reference ranges for normal left ventricular 2D strain: results from the EACVI NORRE study.
      ]. Secondary outcomes were LVEF and GLS as continuous values, additional measurements of LV systolic function (lateral mitral annular plane systolic excursion, tissue doppler lateral and septal s', GLS-rate, mid-GCS), diastolic function (E/A ratio, average E/e’ ratio, septal and lateral e’ and tricuspid regurgitation gradient), and structural measurements of the LV and left atrium.

      2.4 Statistical analysis

      Continuous variables are presented as mean (±SD) or median [range] where appropriate, and categorical variables as frequencies. Inference tests on matching variables (a good match defined as standardized mean difference < 0.1) and echocardiographic measurements between CCS and controls were performed with the Student's t-test or Kruskall-Wallis test for continuous variables and Pearson chi-square of Fisher's exact test for categorical variables. Non-linearity of dose-response relationships between vincristine and cardiac function measurements was tested with restricted cubic splines with three knots.
      Multivariable linear and logistic regression analyses were performed in CCS and controls combined to identify whether being a vincristine-treated CCS was an independent risk factor for cardiac dysfunction. These models always included age at echocardiography and sex, together with covariates with a p-value <0.2 in univariable analysis in either CCS or controls, as an exploratory selection criterion. Subsequently, we assessed the differential contribution of covariates in CCS versus controls with interaction terms in the linear models, by stepwise entering interactions with the most significant covariates and omitting any non-significant interactions. Of note, in multivariable analysis comparing risk of cardiac dysfunction in CCS versus controls, time variables regarding cancer diagnosis (age at diagnosis and time since diagnosis) were not considered confounders, since controls did not have cancer.
      Two-sided p-values <0.05 were considered statistically significant. Since echocardiographic measurements are inter-correlated, we did not correct our secondary outcomes for multiple testing. All statistical tests beyond the primary outcomes should be regarded as exploratory. Analyses were performed in R (version 3.5.3, R Foundation, Vienna, Austria).

      3. Results

      3.1 Participants

      The inclusion flowchart is shown in Fig. 1. Of 130 vincristine-treated CCS that visited the outpatient clinic, 101 CCS were included, along with 101 matched controls. Baseline demographics and clinical data are shown in Table 1 and details of non-participants in Supporting Table 1. Median age at diagnosis was 4.0 [0.2–15.8] years and median time since diagnosis 31 [16–48] years. Dominant cancer type was leukaemia (n = 72; 71%). Median cumulative vincristine dose was 63 [10–138] mg/m2. The number of vincristine administrations was available for 51 CCS. No participants reported to be diagnosed with or treated for heart failure or myocardial infarction. Included CCS had a higher mean blood pressure and heart rate than controls, and six CCS and no controls were treated for hypertension.
      Fig. 1
      Fig. 1Inclusion flowchart.
      * Neither anthracyclines, radiotherapy involving the heart region, cyclophosphamide nor ifosfamide
      †CHD were either known before cancer diagnosis (CCS), reported in LATER questionnaire and confirmed or detected at current echocardiography.
      AR = aortic regurgitation; ASD = atrial septal defect; AVSD = atrioventricular septal defect; BAV = bicuspid aortic valve; CCS = childhood cancer survivors; CHD = congenital heart disease.
      Table 1Demography and baseline characteristics of vincristine-treated survivors and controls.
      Vincristine-treated CCS (n = 101)Unexposed controls (n = 101)SMD
      Demography, diagnosis and treatment history
       Male/ female sex (n (%))48/53(48/53)44/57(44/56)0.080
       Age at cancer diagnosis, years (median [range])4.0[0.2–15.8]
      <5 (n (%))59(58)
      5–926(26)
      10–1415(15)
      15–181(1)
       Incidence year (median [range])1987[1970–2001]
      1970–1979 (n (%))17(17)
      1980–198940(40)
      1990–199925(25)
      ≥200019(19)
       Primary cancer diagnosis (n (%))
       Leukaemia72(71)
       Lymphoma / reticuloendothelial6(6)
       Central nervous system, other intracranial, intraspinal5(5)
       Renal16(16)
       Soft tissue2(2)
       Cumulative vincristine dose, mg/m2 (median [range])63[10–138]
      <50 (n (%))37(37)
      50–10061(60)
      ≥1003(3)
       Time since cancer diagnosis, years (median [range])31[16–48]
      10–19 (n (%))31(31)
      20–2915(15)
      30–3942(42)
      ≥4013(13)
       Age at echocardiography, years (median [range])35[17–53]35[17–56]0.085
      15–24 (n (%))25(25)18(18)
      25–3424(24)33(33)
      35–4435(35)35(35)
      ≥4517(17)15(15)
      Questionnaire data
       Self-reported diagnosis of heart failure
      Validated with appropriate medication use.
      (n (%))
      00
       Self-reported diagnosis of myocardial infarction
      Validated with appropriate medication use.
      (n (%))
      00
       Self-reported diagnosis of hypertension
      Validated with appropriate medication use.
      (n (%))
      4(4)0
       Self-reported diagnosis of diabetes
      Validated with appropriate medication use.
      (n (%))
      2(2)0
       Use of lipid lowering medication (n (%))3(3)0
       Ever smoked >1 year
      Composite of LATER study part I and II. Valid for 102 CCS and 88 controls.
      (n (%))
      28(30)28(35)
      Outpatient clinic data
       Body mass index, kg/m2 (median [range])25[17–42]25[19–39]
       Systolic blood pressure, mmHg (mean (SD))126(16)118(14)
       Diastolic blood pressure, mmHg (mean (SD))75(12)71(10)
       Heart rate at echocardiography, bpm (mean (SD))68(11)62(11)
      CCS = childhood cancer survivors; SD = standard deviation; SMD = standardized mean difference (for matching variables only). No p-values are reported in Table 1 following the STROBE recommendations.
      a Validated with appropriate medication use.
      b Composite of LATER study part I and II. Valid for 102 CCS and 88 controls.

      3.2 Left ventricular systolic and diastolic function

      Echocardiographic measurements are summarized in Table 2. LVEF could be measured in 88 CCS (87%) and 97 (96%) controls. Twelve CCS (14%) had an abnormal LVEF, compared to four controls (4%; p 0.034), with a lowest measured LVEF of 47% in CCS and 51% in controls. The mean LVEF was 58.4 (±4.7)% in CCS versus 59.7 (±3.7)% in controls (p 0.050).
      Table 2Echocardiographic measurements of vincristine-treated survivors and controls.
      Vincristine-treated CCSUnexposed controlsp-value
      Primary outcomes
       Abnormal LV ejection fraction
      <52% for men, <54% for women.
      12(14)4(4)0.034
       Abnormal global longitudinal strain
      Age-, sex- and vendor specific cut-off values.
      19(24)8(9)0.011
       Diastolic dysfunction ≥grade II001
      Structural measurements
       Interventricular septum thickness, mm7.8(1.5)7.4(1.2)0.069
       LV end-diastolic diameter index, mm/m224(2)25(3)0.256
       LV end-diastolic volume index, ml/m250(9)53(11)0.041
       LV mass index, g/m265(14)64(13)0.655
       LA volume index, ml/m223(7)25(6)0.075
      Functional measurements
       LV ejection fraction, %58.4(4.7)59.7(3.7)0.050
       Global longitudinal strain, %−19.0(2.4)−20.1(2.2)0.001
       Mid-ventricular global circumferential strain, %−21.9(2.8)−22.0(3.2)0.813
       Global longitudinal strain rate, 1/s−0.93(0.12)−0.95(0.12)0.384
       Mitral annular systolic plane excursion, mm16(3)17(3)0.005
       Tissue Doppler LV septal s', cm/s8(1)9(1)0.061
       Tissue Doppler LV lateral s', cm/s10(2)11(2)0.344
       Diastolic dysfunction grade I, normal LV ejection fraction1(1)00.466
       Mitral inflow E/A ratio1.6(0.5)1.6(0.4)0.878
       Mitral average E/e’ ratio5.9(1.3)5.4(1.5)0.020
       Tissue Doppler LV lateral e’, cm/s15(4)16(4)0.110
       Tissue Doppler LV septal e’, cm/s12(3)13(3)0.003
       TR gradient, mmHg14(5)15(6)0.534
      Continuous data are mean (standard deviation), proportions are n (%).
      CCS = childhood cancer survivors; LA = left atrium; LV = left ventricle; TR = tricuspid regurgitation.
      Bold values indicate statistical significance (p < 0.05).
      a <52% for men, <54% for women.
      b Age-, sex- and vendor specific cut-off values.
      GLS was measured in 79 (78%) CCS and 92 (91%) controls. Abnormal GLS was encountered in nineteen CCS (24%) and eight controls (9%; p 0.011) and mean GLS values were − 19.0 (±2.4) % versus −20.1 (±2.2)%, respectively (p 0.001). Mild differences were also found in other conventional measurements of LV systolic function, but not in mid-GCS and GLS-rate. Mean GLS was lower in CCS compared to controls in four out of six cardiac walls (Supporting Table 2).
      No participants showed echocardiographic findings compatible with ≥grade 2 diastolic dysfunction.

      3.3 Associations of LV systolic dysfunction

      Table 3 shows the multivariable linear and logistic associations of LVEF and GLS. The univariable analyses underlying our variable selection are shown in Supporting Table 3.
      Table 3Multivariable associations of systolic function measurements.
      Linear regression
      All variables in the linear models were tested for interactions with being a CCS by stepwise addition starting with the most significant variable. No tests for interactions were performed in logistic regression due to the limited number of events.
      Logistic regression
      All variables in the linear models were tested for interactions with being a CCS by stepwise addition starting with the most significant variable. No tests for interactions were performed in logistic regression due to the limited number of events.
      Multivariable β (95%CI)pMultivariable OR (95%CI)p
      LV ejection fraction
       CCS (vs no)−5.76(−11.0 to −0.53)0.0313.07(1.0–11.6)0.065
       Female sex (vs male)1.67(0.34–3.0)0.0141.07(0.37–3.2)0.896
       Age at echo (per year)0.04(−0.06–0.16)0.4260.94(0.88–1.0)0.075
       BMI >25 kg/m2 (vs ≤25)−0.70(−2.0–0.63)0.300
      Univariable p > 0.2.
       Diastolic BP (per 10 mmHg)
      Multicollinear with systolic blood pressure and equally related to cardiovascular risk in the general population under age 50 years [33]. Diastolic blood pressure had a slightly better fit in most of the models and was therefore chosen. Too few participants were diagnosed with hypertension.
      −0.70(−1.4 to −0.05)0.0321.42(0.85–2.4)0.182
       Ever smoked >1 year (vs no)−2.21(−3.6 to −0.77)0.003
      Univariable p > 0.2.
      Interaction term:
       Age at echo*CCS0.15(0.01–0.30)0.041
      R-squared
      Adjusted R-squared for linear models, Nagelkerke pseudo-R-squared for logistic models.
      0.1580.082
      Global longitudinal strain
      Global longitudinal strain is a negative number. Negative correlations indicate ‘better’ values and vice versa.
       CCS (vs no)0.69(−0.03–1.4)0.0603.55(1.37–10.1)0.012
       Female sex (vs male)−0.90(−1.6 to −0.18)0.0151.76(0.70–4.62)0.237
       Age at echo (per year)−0.03(−0.07–0.02)0.2121.02(0.97–1.08)0.490
       BMI >25 kg/m2 (vs ≤25)0.80(0.08–1.5)0.0311.89(0.76–4.94)0.178
       Diastolic BP (per 10 mmHg)
      Multicollinear with systolic blood pressure and equally related to cardiovascular risk in the general population under age 50 years [33]. Diastolic blood pressure had a slightly better fit in most of the models and was therefore chosen. Too few participants were diagnosed with hypertension.
      0.65(0.31–0.99)< 0.0011.39(0.92–2.18)0.130
       Ever smoked >1 year (vs no)0.09(−0.68–0.86)0.815
      Univariable p > 0.2.
      R-squared
      Adjusted R-squared for linear models, Nagelkerke pseudo-R-squared for logistic models.
      0.1950.131
      Childhood cancer survivors and controls were combined in the models.
      BMI = body mass index; BP = blood pressure; CCS = childhood cancer survivors; LV = left ventricle.
      Bold values indicate statistical significance (p < 0.05).
      a All variables in the linear models were tested for interactions with being a CCS by stepwise addition starting with the most significant variable. No tests for interactions were performed in logistic regression due to the limited number of events.
      b Univariable p > 0.2.
      c Multicollinear with systolic blood pressure and equally related to cardiovascular risk in the general population under age 50 years [
      • Williams B.
      • Mancia G.
      • Spiering W.
      • Agabiti Rosei E.
      • Azizi M.
      • Burnier M.
      • et al.
      2018 ESC/ESH guidelines for the management of arterial hypertension.
      ]. Diastolic blood pressure had a slightly better fit in most of the models and was therefore chosen. Too few participants were diagnosed with hypertension.
      d Adjusted R-squared for linear models, Nagelkerke pseudo-R-squared for logistic models.
      e Global longitudinal strain is a negative number. Negative correlations indicate ‘better’ values and vice versa.
      Abnormal LVEF was not significantly associated with being a CCS versus control (OR 3.07) but had wide 95% confidence intervals (1.0 to 11.6). The model for LVEF as a continuous outcome showed a significant interaction between CCS status and age at echo, which should be interpreted as younger CCS having a significantly lower LVEF than controls. The difference between CCS and controls for abnormal GLS, but not GLS as a continuous outcome, did reach significance (OR 3.55, 95%CI 1.37 to 10.1 and β 0.69, 95%CI -0.03 to 1.4, respectively).
      Factors that significantly contributed to variation in LVEF were sex, age at echocardiography (in CCS), diastolic blood pressure and smoking history, whereas sex, obesity and diastolic blood pressure contributed to variation in GLS. Hypertension was too rare (n = 4) to include in the models. Instead, continuous blood pressure values were included. Since only few CCS used antidiabetics or lipid lowering medication, these variables were not included in the models, but sensitivity analyses showed similar results when these subjects were excluded from the models for GLS. In the linear regression model for GLS, no significant interactions were found between the included covariates and being a CCS versus control.
      Within the CCS group, cumulative vincristine dose and the number of vincristine administrations did not satisfy our univariable selection criterion to be included in our multivariable models for either LVEF or GLS (Supporting Table 3 and Supporting Fig. 1). No significant non-linearity was observed. After adjusting for age and sex, cumulative vincristine dose and the number of vincristine administrations were not associated with abnormal LVEF (OR 0.88 per 50 mg/m2, 95%CI 0.24 to 3.4; OR 0.58 per 10 administrations, 95%CI 0.23 to 1.3) or abnormal GLS (OR 1.14 per 50 mg/m2, 95%CI 0.40 to 3.4; OR 1.04 per 10 administrations, 95%CI 0.48 to 2.6). Including diastolic blood pressure in the models did not alter these findings. Results in multivariable linear regression analysis were similar.

      4. Discussion

      Our systematic approach addressed mild echocardiographic abnormalities in long-term CCS treated with vincristine, but without other established or potentially cardiotoxic therapies. We showed an increased prevalence of abnormal GLS in these CCS compared to untreated controls, independent of other cardiovascular risk factors. Only mild LVEF abnormalities were found and important diastolic dysfunction was not encountered in these CCS. Although vincristine treatment is the common denominator in this sub-population of CCS, we did not find a dose-response relationship of measures of vincristine exposure with any of the systolic dysfunction measurements.
      It is encouraging that none of the CCS in the current study reported a diagnosis of clinical heart failure, although this small cross-sectional cohort study prevents drawing conclusions on heart failure incidence. The prevalence of an abnormal LVEF (14%) in vincristine-treated CCS may seem higher than the 4.3% in anthracycline-treated CCS [
      • Merkx R.
      • Leerink J.M.
      • de Baat E.C.
      • Feijen E.A.M.
      • Kok W.E.M.
      • Mavinkurve-Groothuis A.M.C.
      • et al.
      Asymptomatic systolic dysfunction on contemporary echocardiography in anthracycline-treated long-term childhood cancer survivors: a systematic review.
      ], but we emphasize that we defined abnormal LVEF with a higher cut-off value than in previous reports, following the latest chamber quantification guidelines [
      • Lang R.M.
      • Badano L.P.
      • Mor-Avi V.
      • Afilalo J.
      • Armstrong A.
      • Ernande L.
      • et al.
      Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
      ]. Only two CCS showed an LVEF just below 50% in our cohort.
      The prevalence of abnormal GLS (24%), however, is comparable to that found in anthracycline-treated CCS [
      • Merkx R.
      • Leerink J.M.
      • de Baat E.C.
      • Feijen E.A.M.
      • Kok W.E.M.
      • Mavinkurve-Groothuis A.M.C.
      • et al.
      Asymptomatic systolic dysfunction on contemporary echocardiography in anthracycline-treated long-term childhood cancer survivors: a systematic review.
      ]. GLS is considered an earlier indicator of systolic dysfunction compared to LVEF, with superior predictive value for future heart failure and related events in various populations with (risk of) cardiovascular disease and adult cardio-oncology [
      • Kalam K.
      • Otahal P.
      • Marwick T.H.
      Prognostic implications of global LV dysfunction: a systematic review and meta-analysis of global longitudinal strain and ejection fraction.
      ,
      • Oikonomou E.K.
      • Kokkinidis D.G.
      • Kampaktsis P.N.
      • Amir E.A.
      • Marwick T.H.
      • Gupta D.
      • et al.
      Assessment of prognostic value of left ventricular global longitudinal strain for early prediction of chemotherapy-induced cardiotoxicity: a systematic review and Meta-analysis.
      ]. The occurrence of abnormal GLS among young survivors warrants investigation of its natural course and the clinical consequences later in life. Clinicians should be aware that also these CCS may have an elevated risk of future cardiovascular events.
      Whether vincristine causes direct cardiac damage has been investigated in preclinical studies. In rats, interstitial cardiac endothelial cells went into arrest or apoptosis hours after high dose vincristine or vinblastine administrations, but no myocardial necrosis was found [
      • Mikaelian I.
      • Buness A.
      • de Vera-Mudry M.C.
      • Kanwal C.
      • Coluccio D.
      • Rasmussen E.
      • et al.
      Primary endothelial damage is the mechanism of cardiotoxicity of tubulin-binding drugs.
      ]. Endothelial cell damage was confirmed in porcine aortic cells [
      • Yamada T.
      • Egashira N.
      • Imuta M.
      • Yano T.
      • Yamauchi Y.
      • Watanabe H.
      • et al.
      Role of oxidative stress in vinorelbine-induced vascular endothelial cell injury.
      ]. Tochinai et al., by contrast, did show cardiomyocyte necrosis in rats and indeed suggested endothelial damage as causative mechanism [
      • Tochinai R.
      • Ando M.
      • Suzuki T.
      • Suzuki K.
      • Nagata Y.
      • Hata C.
      • et al.
      Histopathological studies of microtubule disassembling agent-induced myocardial lesions in rats.
      ]. Clinical studies may support the hypothesis of vascular toxicity, but mainly in peripheral vasculature and vinca-alkaloids were only part of a regimen [
      • Bokemeyer C.
      • Berger C.C.
      • Kuczyk M.A.
      • Schmoll H.J.
      Evaluation of long-term toxicity after chemotherapy for testicular cancer.
      ,
      • Stoter G.
      • Koopman A.
      • Vendrik C.P.
      • Struyvenberg A.
      • Sleyfer D.T.
      • Willemse P.H.
      • et al.
      Ten-year survival and late sequelae in testicular cancer patients treated with cisplatin, vinblastine, and bleomycin.
      ].
      We generated working hypotheses on the associations and pathophysiology of the observed systolic dysfunction. An absent dose-response relation between vincristine (cumulative dose or number of administrations) and systolic function measurements, does not exclude a ‘single-hit’ or threshold association, individual variation in pharmacokinetics or vincristine-susceptibility, or indirect toxicity. GLS was more often abnormal in vincristine-treated CCS than in controls, independent of sex, age at echocardiography and diastolic blood pressure; such residual confounding likely results from unmeasured risk factors. Investigating GLS in CCS who received only surgical treatment, may provide useful insights in whether CCS are at higher risk of cardiac dysfunction, irrespective of received chemotherapy.
      We found GLS and LVEF to be associated with diastolic blood pressure. Hypertension is highly prevalent in young CCS [
      • Armstrong G.T.
      • Oeffinger K.C.
      • Chen Y.
      • Kawashima T.
      • Yasui Y.
      • Leisenring W.
      • et al.
      Modifiable risk factors and major cardiac events among adult survivors of childhood cancer.
      ], but since only few CCS in our cohort used antihypertensive medication, analysis of blood pressure as a continuous variable, rather than a binary indicator of hypertension, was more feasible. Both LVEF and GLS are known afterload-dependent measurements, and blood pressure can influence GLS measurement [
      • Sugimoto T.
      • Dulgheru R.
      • Bernard A.
      • Ilardi F.
      • Contu L.
      • Addetia K.
      • et al.
      Echocardiographic reference ranges for normal left ventricular 2D strain: results from the EACVI NORRE study.
      ,
      • D’Elia N.
      • Caselli S.
      • Kosmala W.
      • Lancellotti P.
      • Morris D.
      • Muraru D.
      • et al.
      Normal global longitudinal strain: an individual patient meta-analysis.
      ,
      • Yingchoncharoen T.
      • Agarwal S.
      • Popovic Z.B.
      • Marwick T.H.
      Normal ranges of left ventricular strain: a meta-analysis.
      ,
      • Blomstrand P.
      • Sjöblom P.
      • Nilsson M.
      • Wijkman M.
      • Engvall M.
      • Länne T.
      • et al.
      Overweight and obesity impair left ventricular systolic function as measured by left ventricular ejection fraction and global longitudinal strain.
      ], even independent of arterial hypertension [
      • Stylidis M.
      • Leon D.A.
      • Rӧsner A.
      • Schirmer H.
      Global myocardial longitudinal strain in a general population—associations with blood pressure and subclinical heart failure: the Tromsø study.
      ]. The hypothesized vascular toxicity of vincristine may as well result in an increased afterload. However, the cross-sectional nature of our study precludes any conclusion on the possible contributions of afterload dependency of the outcome measurements and of cardiac damage due to longitudinal exposure to higher blood pressures. The association of being overweight with decreased GLS has also been previously shown in CCS and the general population [
      • Ng A.C.T.
      • Prevedello F.
      • Dolci G.
      • Roos C.J.
      • Djaberi R.
      • Bertini M.
      • et al.
      Impact of diabetes and increasing body mass index category on left ventricular systolic and diastolic function.
      ,
      • Blomstrand P.
      • Sjöblom P.
      • Nilsson M.
      • Wijkman M.
      • Engvall M.
      • Länne T.
      • et al.
      Overweight and obesity impair left ventricular systolic function as measured by left ventricular ejection fraction and global longitudinal strain.
      ].
      In CCS treated with anthracyclines and/or radiotherapy to the heart region, Armstrong et al. showed less increasing risk ratios for abnormal GLS than for abnormal LVEF with increasing cardiotoxic doses, and a stronger correlation of abnormal GLS than abnormal LVEF with traditional cardiovascular risk factors [
      • Armstrong G.T.
      • Joshi V.M.
      • Ness K.K.
      • Marwick T.H.
      • Zhang N.
      • Srivastava D.
      • et al.
      Comprehensive echocardiographic detection of treatment-related cardiac dysfunction in adult survivors of childhood Cancer: results from the St. Jude lifetime cohort study.
      ]. As such, GLS may be representative of the traditional cardiovascular risk factors present in a long-term CCS, rather than being a sole indicator of treatment-related cardiotoxicity. Modification of common cardiovascular risk factors may thus be one key strategy to prevent future cardiovascular events for young patients whose chances to survive their respective childhood cancer depend on treatment modalities that may be cardiotoxic.

      4.1 Limitations

      In this heterogeneous historical cohort, many unknown potential confounders that may explain the difference in abnormal GLS between CCS and controls remained unmeasured. These confounders may relate to the cancer diagnosis and treatment period as well as the long period until the current study, and may include other somatic late effects, lifestyle and environmental factors. A dose-response relation is one of the few signs suggestive of causality in observational studies. We assessed alternative dose-response hypothesis with the number of vincristine administrations, but data were limited. Logistic regression analysis of abnormal LVEF may have been underpowered due to few events. Three-dimensional LVEF measurements were not systematically obtained in all participating centres and carried a high risk of selection bias [
      • Merkx R.
      • Leerink J.M.
      • Feijen E.A.M.
      • Kremer L.C.M.
      • de Baat E.C.
      • Bellersen L.
      • et al.
      Echocardiography protocol for early detection of cardiac dysfunction in childhood cancer survivors in the multicenter DCCSS LATER 2 CARD study: design, feasibility, and reproducibility.
      ]. Questionnaires were currently the best data source for medical history and medication use, since new diagnoses of heart failure, diabetes or hypertension at the current visit required referral, follow-up visits or additional investigations which were not recorded.

      5. Conclusions

      Long-term CCS treated with vincristine, but without anthracyclines, radiotherapy involving the heart region, cyclophosphamide or ifosfamide, showed an increased prevalence of abnormal GLS compared to controls, independent of age, sex, obesity and blood pressure. This finding remained partially unexplained since there was no dose-response relation between vincristine exposure and LVEF or GLS. Traditional cardiovascular risk factors contributed to abnormal GLS and LVEF. Whether vincristine-treated CCS with abnormal GLS are at risk of cardiovascular events later in life and would benefit from early cardiovascular risk factor modification, should be evaluated.

      Availability of data

      De-identified data may be shared with investigators who would like to collaborate after the main analyses of the LATER CARD study are finished. Applications of intent can be sent to the LATER study group via [email protected] .

      Grant support

      This work was supported by grant 2015-021 of the Dutch Heart Foundation and grant 171 ‘DCOG LATER program’ of KiKa/ODAS.

      Protocol registration

      Netherlands Trial Registry number 7481.

      Author statement

      Remy Merkx: Investigation, Data curation, Formal analysis, Visualization, Writing draft. Lieke Feijen: Conceptualization, Project administration, Funding acquisition, Supervision, Writing draft. Jan Leerink: Data curation, Interpretation, Writing review. Esmée de Baat: Interpretation, Writing review. Louise Bellersen: Resources, Writing review. Elvira van Dalen: Conceptualization, Writing review. Eline van Dulmen-den Broeder: Conceptualization, Writing review. Margriet van der Heiden-van der Loo: Data curation, Project administration, Writing review. Marry van den Heuvel-Eibrink: Conceptualization, Writing review. Chris de Korte: Resources, Interpretation, Writing review. Jacqueline Loonen: Conceptualization, Funding acquisition, Writing review. Marloes Louwerens: Conceptualization, Writing review. Cécile Ronckers: Conceptualization, Writing review. Arco Teske: Methodology, Writing review. Wim Tissing: Conceptualization, Writing review. Andrica de Vries: Conceptualization, Writing review. Annelies Mavinkurve-Groothuis: Methodology, Funding acquisition, Writing review. Helena van der Pal: Conceptualization, Project administration, Writing review. Gert Weijers: Software, Writing review. Wouter Kok: Methodology, Funding acquisition, Writing review. Leontien Kremer: Conceptualization, Funding acquisition, Writing review. Livia Kapusta: Methodology, Funding acquisition, Writing draft, Supervision.

      Declaration of Competing Interest

      None declared.

      Acknowledgements

      We thank the other members of the DCOG LATER consortium (Birgitta Versluys, Martha Grootenhuis, Flora van Leeuwen, Sebastian Neggers, Lideke van der Steeg, Geert Janssens, Hanneke van Santen, Margreet Veening, Jaap den Hartogh, Saskia Pluijm, Lilian Batenburg, Hanneke de Ridder, Nynke Hollema, Lennart Teunissen, Anke Schellekens) and all physicians, research nurses, sonographers, data managers and participating patients, parents and siblings for their contribution.

      Appendix A. Supplementary data

      References

        • Leerink J.M.
        • de Baat E.C.
        • Feijen E.A.M.
        • Bellersen L.
        • Dalen E.C.
        • Grotenhuis H.B.
        • et al.
        Cardiac disease in childhood cancer survivors.
        JACC: CardioOncol. 2020; 2: 363-378https://doi.org/10.1016/j.jaccao.2020.08.006
        • Fidler M.M.
        • Reulen R.C.
        • Henson K.
        • Kelly J.
        • Cutter D.
        • Levitt G.A.
        • et al.
        Population-based long-term cardiac-specific mortality among 34 489 five-year survivors of childhood Cancer in Great Britain.
        Circulation. 2017; 135: 951-963https://doi.org/10.1161/CIRCULATIONAHA.116.024811
        • Armenian S.H.
        • Hudson M.M.
        • Mulder R.L.
        • Chen M.H.
        • Constine L.S.
        • Dwyer M.
        • et al.
        Recommendations for cardiomyopathy surveillance for survivors of childhood cancer: a report from the international late effects of childhood Cancer guideline harmonization group.
        Lancet Oncol. 2015; 16 (e123-e36)https://doi.org/10.1016/s1470-2045(14)70409-7
        • Madsen M.L.
        • Due H.
        • Ejskjær N.
        • Jensen P.
        • Madsen J.
        • Dybkær K.
        Aspects of vincristine-induced neuropathy in hematologic malignancies: a systematic review.
        Cancer Chemother. Pharmacol. 2019; 84: 471-485https://doi.org/10.1007/s00280-019-03884-5
        • Gros R.
        • Hugon V.
        • Thouret J.M.
        • Peigne V.
        Coronary spasm after an injection of vincristine.
        Chemotherapy. 2017; 62: 169-171https://doi.org/10.1159/000455224
        • Taniguchi T.
        • Nakamura T.
        • Sawada T.
        Arterial stiffness, endothelial dysfunction and recurrent angina post-chemotherapy.
        QJM: An Int. J. Med. 2014; 108: 653-655https://doi.org/10.1093/qjmed/hcu184
        • Stoter G.
        • Koopman A.
        • Vendrik C.P.
        • Struyvenberg A.
        • Sleyfer D.T.
        • Willemse P.H.
        • et al.
        Ten-year survival and late sequelae in testicular cancer patients treated with cisplatin, vinblastine, and bleomycin.
        J. Clin. Oncol. 1989; 7: 1099-1104https://doi.org/10.1200/JCO.1989.7.8.1099
        • Roy A.
        • Khanna N.
        • Senguttuvan N.B.
        Rituximab-vincristine chemotherapy-induced acute anterior wall myocardial infarction with cardiogenic shock.
        Tex. Heart Inst. J. 2014; 41: 80-82https://doi.org/10.14503/THIJ-12-2853
        • Goli A.K.
        • Osman M.N.
        • Koduri M.
        • Byrd R.P.
        • Roy T.M.
        A case report of vinorelbine monotherapy-related acute bronchospasm and non-ST elevation acute coronary syndrome.
        Tenn Med. 2011; 104: 47-48
        • Mulrooney D.A.
        • Yeazel M.W.
        • Kawashima T.
        • Mertens A.C.
        • 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; 339b4606https://doi.org/10.1136/bmj.b4606
        • Feijen E.
        • Font-Gonzalez A.
        • Van der Pal H.J.H.
        • Kok W.E.M.
        • Geskus R.B.
        • Ronckers C.M.
        • et al.
        Risk and temporal changes of heart failure among 5-year childhood Cancer survivors: a DCOG-LATER study.
        J. Am. Heart Assoc. 2019; 8e009122https://doi.org/10.1161/JAHA.118.009122
        • van der Pal H.J.H.
        • van Dalen E.C.
        • Hauptmann M.
        • Kok W.E.M.
        • Caron H.N.
        • van den Bos C.
        • et al.
        Cardiac function in 5-year survivors of childhood cancer: a long-term follow-up study.
        Arch. Intern. Med. 2010; 170: 1247-1255https://doi.org/10.1001/archinternmed.2010.233
        • Leerink J.M.
        • Feijen E.A.M.
        • van der Pal H.J.H.
        • Kok W.E.M.
        • Mavinkurve-Groothuis A.M.C.
        • Kapusta L.
        • et al.
        Diagnostic tools for early detection of cardiac dysfunction in childhood cancer survivors: methodological aspects of the Dutch late effects after childhood cancer (LATER) cardiology study.
        Am. Heart J. 2019; 219: 89-98https://doi.org/10.1016/j.ahj.2019.10.010
        • Merkx R.
        • Leerink J.M.
        • Feijen E.A.M.
        • Kremer L.C.M.
        • de Baat E.C.
        • Bellersen L.
        • et al.
        Echocardiography protocol for early detection of cardiac dysfunction in childhood cancer survivors in the multicenter DCCSS LATER 2 CARD study: design, feasibility, and reproducibility.
        Echocardiography. 2021; https://doi.org/10.1111/echo.15081
        • Lang R.M.
        • Badano L.P.
        • Mor-Avi V.
        • Afilalo J.
        • Armstrong A.
        • Ernande L.
        • et al.
        Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
        Eur. Heart J. Cardiovasc. Imaging. 2015; 16: 233-271https://doi.org/10.1093/ehjci/jev014
        • Nagueh S.F.
        • Smiseth O.A.
        • Appleton C.P.
        • Byrd 3rd, B.F.
        • Dokainish H.
        • Edvardsen T.
        • et al.
        Recommendations for the evaluation of left ventricular diastolic function by echocardiography: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
        J. Am. Soc. Echocardiogr. 2016; 29: 277-314https://doi.org/10.1016/j.echo.2016.01.011
        • Sugimoto T.
        • Dulgheru R.
        • Bernard A.
        • Ilardi F.
        • Contu L.
        • Addetia K.
        • et al.
        Echocardiographic reference ranges for normal left ventricular 2D strain: results from the EACVI NORRE study.
        Eur. Heart J. Cardiovasc. Imaging. 2017; 18: 833-840https://doi.org/10.1093/ehjci/jex140
        • Voigt J.U.
        • Pedrizzetti G.
        • Lysyansky P.
        • Marwick T.H.
        • Houle H.
        • Baumann R.
        • et al.
        Definitions for a common standard for 2D speckle tracking echocardiography: consensus document of the EACVI/ASE/industry task force to standardize deformation imaging.
        Eur. Heart J. Cardiovasc. Imaging. 2015; 16: 1-11https://doi.org/10.1093/ehjci/jeu184
        • Merkx R.
        • Leerink J.M.
        • de Baat E.C.
        • Feijen E.A.M.
        • Kok W.E.M.
        • Mavinkurve-Groothuis A.M.C.
        • et al.
        Asymptomatic systolic dysfunction on contemporary echocardiography in anthracycline-treated long-term childhood cancer survivors: a systematic review.
        J. Cancer Surviv. 2021; https://doi.org/10.1007/s11764-021-01028-4
        • Kalam K.
        • Otahal P.
        • Marwick T.H.
        Prognostic implications of global LV dysfunction: a systematic review and meta-analysis of global longitudinal strain and ejection fraction.
        Heart. 2014; 100: 1673-1680https://doi.org/10.1136/heartjnl-2014-305538
        • Oikonomou E.K.
        • Kokkinidis D.G.
        • Kampaktsis P.N.
        • Amir E.A.
        • Marwick T.H.
        • Gupta D.
        • et al.
        Assessment of prognostic value of left ventricular global longitudinal strain for early prediction of chemotherapy-induced cardiotoxicity: a systematic review and Meta-analysis.
        JAMA Cardiol. 2019; 4: 1007-1018https://doi.org/10.1001/jamacardio.2019.2952
        • Mikaelian I.
        • Buness A.
        • de Vera-Mudry M.C.
        • Kanwal C.
        • Coluccio D.
        • Rasmussen E.
        • et al.
        Primary endothelial damage is the mechanism of cardiotoxicity of tubulin-binding drugs.
        Toxicol. Sci. 2010; 117: 144-151https://doi.org/10.1093/toxsci/kfq189
        • Yamada T.
        • Egashira N.
        • Imuta M.
        • Yano T.
        • Yamauchi Y.
        • Watanabe H.
        • et al.
        Role of oxidative stress in vinorelbine-induced vascular endothelial cell injury.
        Free Radic. Biol. Med. 2010; 48: 120-127https://doi.org/10.1016/j.freeradbiomed.2009.10.032
        • Tochinai R.
        • Ando M.
        • Suzuki T.
        • Suzuki K.
        • Nagata Y.
        • Hata C.
        • et al.
        Histopathological studies of microtubule disassembling agent-induced myocardial lesions in rats.
        Exp. Toxicol. Pathol. 2013; 65: 737-743https://doi.org/10.1016/j.etp.2012.09.008
        • Bokemeyer C.
        • Berger C.C.
        • Kuczyk M.A.
        • Schmoll H.J.
        Evaluation of long-term toxicity after chemotherapy for testicular cancer.
        J. Clin. Oncol. 1996; 14: 2923-2932https://doi.org/10.1200/JCO.1996.14.11.2923
        • Armstrong G.T.
        • Oeffinger K.C.
        • Chen Y.
        • Kawashima T.
        • Yasui Y.
        • Leisenring W.
        • et al.
        Modifiable risk factors and major cardiac events among adult survivors of childhood cancer.
        J. Clin. Oncol. 2013; 31: 3673-3680https://doi.org/10.1200/JCO.2013.49.3205
        • D’Elia N.
        • Caselli S.
        • Kosmala W.
        • Lancellotti P.
        • Morris D.
        • Muraru D.
        • et al.
        Normal global longitudinal strain: an individual patient meta-analysis.
        JACC Cardiovasc. Imaging. 2020; 13: 167-169https://doi.org/10.1016/j.jcmg.2019.07.020
        • Yingchoncharoen T.
        • Agarwal S.
        • Popovic Z.B.
        • Marwick T.H.
        Normal ranges of left ventricular strain: a meta-analysis.
        J. Am. Soc. Echocardiogr. 2013; 26: 185-191https://doi.org/10.1016/j.echo.2012.10.008
        • Blomstrand P.
        • Sjöblom P.
        • Nilsson M.
        • Wijkman M.
        • Engvall M.
        • Länne T.
        • et al.
        Overweight and obesity impair left ventricular systolic function as measured by left ventricular ejection fraction and global longitudinal strain.
        Cardiovasc. Diabetol. 2018; 17: 113https://doi.org/10.1186/s12933-018-0756-2
        • Stylidis M.
        • Leon D.A.
        • Rӧsner A.
        • Schirmer H.
        Global myocardial longitudinal strain in a general population—associations with blood pressure and subclinical heart failure: the Tromsø study.
        Int. J. Card. Imaging. 2020; 36: 459-470https://doi.org/10.1007/s10554-019-01741-3
        • Ng A.C.T.
        • Prevedello F.
        • Dolci G.
        • Roos C.J.
        • Djaberi R.
        • Bertini M.
        • et al.
        Impact of diabetes and increasing body mass index category on left ventricular systolic and diastolic function.
        J. Am. Soc. Echocardiogr. 2018; 31: 916-925https://doi.org/10.1016/j.echo.2018.02.012
        • Armstrong G.T.
        • Joshi V.M.
        • Ness K.K.
        • Marwick T.H.
        • Zhang N.
        • Srivastava D.
        • et al.
        Comprehensive echocardiographic detection of treatment-related cardiac dysfunction in adult survivors of childhood Cancer: results from the St. Jude lifetime cohort study.
        J. Am. Coll. Cardiol. 2015; 65: 2511-2522https://doi.org/10.1016/j.jacc.2015.04.013
        • Williams B.
        • Mancia G.
        • Spiering W.
        • Agabiti Rosei E.
        • Azizi M.
        • Burnier M.
        • et al.
        2018 ESC/ESH guidelines for the management of arterial hypertension.
        Eur. Heart J. 2018; 39: 3021-3104https://doi.org/10.1093/eurheartj/ehy339