Article Tools
ORIGINAL REPORTS
Cost Effectiveness
Article Tools
Cost-Effectiveness of the International Late Effects of Childhood Cancer Guideline Harmonization Group Screening Guidelines to Prevent Heart Failure in Survivors of Childhood Cancer
Abstract
Survivors of childhood cancer treated with anthracyclines and/or chest-directed radiation are at increased risk for heart failure (HF). The International Late Effects of Childhood Cancer Guideline Harmonization Group (IGHG) recommends risk-based screening echocardiograms, but evidence supporting its frequency and cost-effectiveness is limited.
Using the Childhood Cancer Survivor Study and St Jude Lifetime Cohort, we developed a microsimulation model of the clinical course of HF. We estimated long-term health outcomes and economic impact of screening according to IGHG-defined risk groups (low [doxorubicin-equivalent anthracycline dose of 1-99 mg/m2 and/or radiotherapy < 15 Gy], moderate [100 to < 250 mg/m2 or 15 to < 35 Gy], or high [≥ 250 mg/m2 or ≥ 35 Gy or both ≥ 100 mg/m2 and ≥ 15 Gy]). We compared 1-, 2-, 5-, and 10-year interval-based screening with no screening. Screening performance and treatment effectiveness were estimated based on published studies. Costs and quality-of-life weights were based on national averages and published reports. Outcomes included lifetime HF risk, quality-adjusted life-years (QALYs), lifetime costs, and incremental cost-effectiveness ratios (ICERs). Strategies with ICERs < $100,000 per QALY gained were considered cost-effective.
Among the IGHG risk groups, cumulative lifetime risks of HF without screening were 36.7% (high risk), 24.7% (moderate risk), and 16.9% (low risk). Routine screening reduced this risk by 4% to 11%, depending on frequency. Screening every 2, 5, and 10 years was cost-effective for high-risk survivors, and every 5 and 10 years for moderate-risk survivors. In contrast, ICERs were > $175,000 per QALY gained for all strategies for low-risk survivors, representing approximately 40% of those for whom screening is currently recommended.
Survivors of childhood cancer treated with anthracycline chemotherapy or chest-directed radiotherapy (RT) are at an increased risk of developing cardiomyopathy,1-3 which is well recognized to have a period of asymptomatic left ventricular dysfunction (ALVD) that precedes overt heart failure (HF).4 Numerous guidelines have been developed to facilitate early detection and treatment of ALVD,5 yet they differ regarding definitions of risk groups and recommended surveillance modalities and frequency, hindering effective implementation. To address this, the International Late Effects of Childhood Cancer Guideline Harmonization Group (IGHG) reviewed and graded existing evidence and formulated recommendations for cardiomyopathy risk stratification, screening, and follow-up.6 Classification of survivors as having low, moderate, or high risk of cardiomyopathy was based on cumulative anthracycline and chest RT exposure and was supported by moderate- to high-quality evidence. Limited data were available to inform screening frequency or duration; therefore, harmonized strategies were largely consensus based.
Key Objectives
The International Late Effects of Childhood Cancer Guideline Harmonization Group (IGHG) recommends routine echocardiography screening for survivors of childhood cancer at elevated risk for cardiomyopathy, yet the clinical benefits and cost-effectiveness of these recommendations are uncertain.
Knowledge Generated
Using data from the Childhood Cancer Survivor Study and St Jude Lifetime Cohort, we developed a microsimulation model of the clinical course of heart failure and demonstrated that although currently recommended screening is cost-effective in moderate- and high-risk survivors given commonly cited cost-effectiveness thresholds, more frequent screening may be indicated for those in the high-risk group. Conversely, screening does not appear to be cost-effective in low-risk survivors.
Relevance
Our findings suggest that although IGHG-defined high-risk survivors may benefit from more frequent screening echocardiograms, discontinuation of screening should be considered for low-risk survivors.
Given the disease latency, large numbers needed to detect a significant effect, and the lack of clinical equipoise, trials evaluating effectiveness of different cardiomyopathy screening strategies are infeasible. Simulation models, however, can inform clinical surveillance strategies in such instances. Building on prior modeling work7,8 and leveraging newly available Childhood Cancer Survivor Study (CCSS) and St Jude Lifetime Cohort (SJLIFE) data, we used a model-based approach to estimate the clinical benefits, costs, and cost-effectiveness of various screening strategies for IGHG-defined cardiomyopathy risk groups.
We developed a microsimulation model of the clinical course of HF among survivors of childhood cancer to estimate long-term health and economic outcomes associated with recommended IGHG screening. We used data from CCSS9 (n = 24,297) and SJLIFE10 (n = 3,010), 2 large, institutional review board–approved, and previously described survivorship cohorts that include participant demographics, cancer treatment–related exposures, and graded cardiac events, as previously described,9-11 to establish risk estimates for ALVD and HF (Data Supplement). The CCSS is a retrospectively assessed, prospectively followed cohort of 5-year survivors of childhood cancer assessed periodically for self-reported events,9 representative of US survivors.12 SJLIFE is unique in its ability to prospectively assess and validate the health status of long-term survivors of childhood cancer, facilitating diagnosis of asymptomatic conditions. The model was calibrated to SJLIFE data on asymptomatic and symptomatic disease to ensure modeled outcomes were consistent with observed data. Model parameters are listed in Table 1. For each IGHG risk group, no screening (baseline) and 1-, 2-, 3-, 5-, and 10-year interval screening were evaluated. Screening was initiated at entry into study cohorts (5 years from cancer diagnosis) and continued through age 74 years. Outcomes included number of screens, HF risk at age 40 years, lifetime HF risk, life expectancy, quality-adjusted life expectancy, and lifetime costs. To illustrate the trade-offs between incremental harms and benefits of screening, we calculated the number of screens per HF case averted.
|
To evaluate the relative performance of each strategy, we calculated incremental cost-effectiveness ratios (ICERs), defined as the additional cost of a strategy divided by additional clinical benefit, compared with the next least expensive strategy, and expressed as cost per quality-adjusted life-year (QALY) gained. We discounted future costs and clinical consequences at 3% annually following established recommendations.13,14 We assumed that interventions with ICERs < $100,000 per QALY gained provided good value for resources invested and were considered cost-effective.15,16 The model was programmed in Java (v1.8.0; Oracle, Redwood City, CA), and statistical analyses were performed in R (v3.3.1).
The lifetime risk of HF was based on data from the CCSS9 and the Framingham Heart Study.17,18 To reflect treatment-related risk, we fit a statistical model using CCSS data, which included age at cancer diagnosis, cumulative doxorubicin-equivalent anthracycline dose,19 and mean RT heart dose.20 Mean heart doses were estimated by reconstructing an individual’s treatment fields on age-scaled computational phantoms.20 Survivors faced additional age-related risk of developing HF (46% systolic) based on US general population estimates.21,22 Lifetime risk for HF beyond current CCSS follow-up was estimated by assuming that survivors continued to face an elevated treatment-related risk of HF for the remainder of their lifetime based on the relative risk of observed rates among survivors to the general population at age 45 years. To project HF risks for each IGHG risk group, we bootstrapped individual-level data on CCSS participants, which allowed us to maintain covariance between patient characteristics (eg, age at diagnosis) and treatment exposures (eg, cumulative anthracycline dose). To account for potential nonparticipation bias, we used inverse probability weights when calculating all modeled outcomes.23
The model was constructed to simulate the lifetime clinical course of HF (up to age 100 years) after childhood cancer (Fig 1). Survivors entered the model facing a monthly risk of ALVD. Survivors with ALVD faced risk for progression to HF, and those who developed HF were at risk for HF-related mortality, including additional monthly risks of dying from excess background and noncardiac late effect–related mortality. To infer risk of progression from untreated ALVD to HF, we used model calibration to ensure simulated outcomes were consistent with observed SJLIFE data for asymptomatic (grade 2) and symptomatic (grade 3 or 4) cardiomyopathy (Data Supplement).11,24
FIG 1. Health states and treatment status for the simulated heart failure (HF) model are provided. Data informing risks of transition between health states (shown in solid blue lines), including modifying factors, are represented by numerical indicators as follows. (1) Risk of developing asymptomatic left ventricular dysfunction (ALVD) and subsequent progression to HF was based on data from the Childhood Cancer Survivor Study (CCSS) and St Jude Lifetime Cohort (see Patients and Methods and Data Supplement for details). (2) Mortality rates were based on Centers for Disease Control and Prevention WONDER and CCSS mortality data. (3) Attenuation in the progression from ALVD to HF and mortality from non-HF cardiac causes associated with angiotensin-converting enzyme inhibitor and β-blocker treatment were estimated from the SOLVD trial data (shown in dashed red lines).
We made the following simplifying assumptions: risk of progression from ALVD to HF is constant across ages25; HF-specific mortality risk varies by age (< 30 v ≥ 30 years at HF diagnosis) and increases with time since HF diagnosis26; late mortality risks (including late cancer recurrence and non-HF late effects) for survivors diagnosed in 1980-1999 were equal to or lower than rates observed for those diagnosed in 1970-197927; and age-, sex-, and calendar-specific background mortality rates remained constant at 2014 rates in subsequent years.28
For each IGHG risk group, we evaluated a strategy of no screening (baseline) versus 1-, 2-, 3-, 5-, and 10-year interval screening. We assumed general practitioners would direct routine screening, with subsequent referral for positive echocardiograms to cardiology for a confirmatory echocardiogram. We based 2-dimensional echocardiogram sensitivity and specificity (and interrater agreement between sequential echocardiograms) on published reports (Data Supplement).29 Patients with a positive confirmatory echocardiogram received treatment with an angiotensin-converting enzyme inhibitor (ACE-I) and β-blocker per guideline-based care and remained under a cardiologist’s care.30
Approximate Bayesian computation was used to estimate a distribution of parameters for treatment effectiveness. Because no published studies inform treatment effectiveness in long-term survivors of childhood cancer, our estimates were based on the Studies of Left Ventricular Dysfunction (SOLVD) prevention trial.31 Specifically, for a cohort representative of the SOLVD data (1,594 people in the baseline arm and 513 in the intervention arm with ALVD without a history of childhood cancer; mean age, 58 years), we projected HF risk and all-cause mortality after 4 years by sampling probabilities of progression from ALVD to HF from the simulation model, incorporating age-specific background mortality rates for the study year. We assumed that treatment reduced ALVD progression to HF and mortality from cardiac causes among individuals with ALVD, and we varied model parameters accordingly until simulated outcomes were consistent with the SOLVD results (4-year baseline mortality rate, 0.22; all-cause mortality associated with treatment relative risk, 0.64; Data Supplement).31
Medical costs associated with routine cardiac assessment and follow-up care were based on 2017 Medicare reimbursement rates.32 Survivors with ALVD received guideline-based treatment (ACE-I [lisinopril 20 mg daily] and β-blocker [carvedilol 25 mg twice daily] and an annual physician visit and echocardiogram). Survivors with HF received guideline-based care (same pharmacotherapy as ALVD, a physician visit every 3 months, and echocardiography every 6 months). Drug prices were based on mean wholesale manufacturer acquisition costs.33 We assumed that a proportion of survivors with HF would require hospitalization34 and reflected this in the per-patient costs and that, once ALVD or HF was diagnosed, abnormal echocardiography results did not lead to subsequent diagnostic tests, procedures, or diagnoses that would incur additional costs. Indirect patient costs, such as lost work time, were based on the 2017 median hourly wage (Data Supplement).35
To estimate QALYs, we incorporated age-, sex-, and condition-specific36 utility weights to reflect survivors’ quality of life.37 As a result of its asymptomatic nature, no additional decrement was applied for ALVD. We accounted for decrements in quality of life (disutility) associated with taking daily ALVD medication based on published estimates (Data Supplement).38
For all analyses, to account for first-order (stochastic) and second-order (parameter) uncertainty, we conducted 1,000 Monte Carlo simulations, bootstrapping the modeled cohort and sampling a parameter set (from the subset of good-fitting parameter sets on late mortality risks and accepted parameter sets on treatment effectiveness). To reflect the uncertainty in modeled outcomes, we report the mean and 95% uncertainty interval (UI; calculated as the 2.5 and 97.5 percentiles among simulated results) and include cost-effectiveness acceptability curves identifying the optimal strategy for a range of willingness-to-pay thresholds up to $150,000 per QALY gained. We conducted 1- and 2-way sensitivity analyses to understand how results varied across plausible ranges of key model parameters.
Among a cohort representative of CCSS participants, the cumulative risks of HF at age 40 years were 9.9% (95% UI, 8.7% to 11.1%), 4.5% (95% UI, 2.5% to 6.6%), and 2.2% (95% UI, 0.8% to 3.8%) for the high-, moderate-, and low-risk subgroups, respectively. Lifetime cumulative risks were 36.7% (95% UI, 28.7% to 43.9%), 24.7% (95% UI, 17.3% to 33.5%), and 16.9% (95% UI, 11.2% to 23.8%) in the high-, moderate-, and low-risk groups, respectively (Table 2). Lifetime risk of HF decreased with more frequent screening across all groups, ranging from 4% for 10-year intervals to 11% for 1-year interval screening. Screening delayed the average age of HF onset by as much as 1.6-2.0 years with screening at 1-year intervals.
|
The average number of lifetime per-person echocardiograms varied by risk group and surveillance strategy (Table 2). Screening high-risk survivors at 1-, 2-, 5-, and 10-year intervals averted 1 case of HF every 1,012, 667, 445, and 383 screening echocardiograms, respectively. Screening moderate-risk survivors at the same frequencies prevented 1 case per 1,660, 1,103, 745, and 654 screening echocardiograms, respectively. No strategy resulted in < 1,000 echocardiograms required to avert 1 case of HF in low-risk survivors (Fig 2).
FIG 2. Number of screens by screening strategy (1-, 2-, 5-, and 10-year intervals, reflected by colors as noted in key) per heart failure case averted in high-, moderate-, and low-risk survivors. Shown are estimates for number of screens per heart failure case averted for each risk group, defined as the mean number of screens divided by the mean number of heart failure cases averted, among 50 best-fitting parameter sets.
Table 2 lists the cost-effectiveness results. For high-risk survivors, the ICERs for 10-year ($34,604 per QALY gained), 5-year ($37,703 per QALY gained), and 2-year ($77,877 per QALY gained) interval strategies were all < $100,000 per QALY gained, with the 2-year strategy yielding the greatest gain in QALYs. Among the 1,000 simulations, screening every 2 years had the highest probability of being the preferred strategy (0.52) given a $100,000 per QALY threshold. In moderate-risk survivors, ICERs were < $100,000 for screening every 10 years ($79,312 per QALY gained) and 5 years ($94,575 per QALY gained), although the probability that each strategy was preferred was only 0.14 and 0.35, respectively. Conversely, for low-risk survivors, ICERs were > $175,000 per QALY gained for all strategies, and the probability that no screening was the preferred strategy was 0.70. Cost-effectiveness acceptability curves are show in the Data Supplement.
One-way sensitivity analyses identified preferred screening strategies across a range of model parameters (Fig 3). In high-risk survivors, results were largely unchanged, and 2-year screening remained preferred across most plausible ranges for costs, treatment disutility, lifetime HF-related mortality, and echocardiogram screening parameters. Similarly, no screening remained preferred for the low-risk subgroup across all reasonable values for the same parameters. The optimal strategy for moderate-risk survivors was uncertain, however, because results varied greatly depending on parameter estimates. For example, no screening was favored at higher costs or treatment disutility (Figs 3D and 3E; Data Supplement) and at lower treatment effectiveness and echocardiogram sensitivities and specificities (Figs 3A-3C; Data Supplement).
FIG 3. Sensitivity analyses on key model parameters, including (A) echocardiogram sensitivity, (B) echocardiogram specificity, (C) reduction in lifetime heart failure (HF)–related mortality from angiotensin-converting enzyme inhibitor or β-blocker treatment, (D) screening and congestive HF treatment costs (varied ± 50%), and (E) disutility associated with taking daily medication. Colored regions indicate the range of values for the corresponding variable over which a specific interval strategy would be preferred. Vertical solid gray bars indicate discrete parameter intervals simulated. Vertical dashed line represents the base case assumptions supported by published data. For example, in panel A, when screening sensitivity was varied between 20% and 70%, in high-risk survivors, a 2-year interval screening strategy was favored, whereas no screening was favored in low-risk survivors.
Using a model-based approach that leverages data from the CCSS and SJLIFE, we estimated that lifetime HF risk varied between 16.9% and 36.7% among IGHG risk-stratified survivors of childhood cancer exposed to anthracyclines or chest-directed RT. Screening high-risk survivors at 2-year intervals could potentially avert 8.4% of HF cases. Given the $100,000 per QALY gained benchmark for good value, this strategy appears to be cost-effective. In contrast, for low-risk survivors, routine screening may reduce HF risk, but because of their lower absolute HF risk, screening even every 10 years does not appear to be cost-effective, suggesting a need to reconsider current practices. These data provide much needed evidence to strengthen current IGHG consensus surveillance recommendations and evidence-based screening intervals (Table 3).
|
To date, existing survivor guidelines have had limited ability to recommend evidence-based screening strategies for conditions in which disease latency precludes randomized trials necessary to inform decision making, consequently reflecting a hybrid of evidence-based risk stratification merged with expert opinion and consensus-based surveillance recommendations.6 Other guideline organizations, such as the US Preventive Services Task Force, have used simulation models to strengthen recommendations for adult-onset cancers.39,40 A key strength of simulation models is their ability to test the impact of various baseline model assumptions on screening effectiveness. In our study, we demonstrate that surveillance echocardiograms are not cost-effective in survivors exposed to low-risk doses of anthracyclines and chest-directed RT despite testing a wide range of model parameters. Given both the gap in existing literature regarding the efficacy of pharmacologic intervention for ALVD in survivors of childhood cancer and previous data suggesting that ALVD treatment efficacy is the key determinant of the cost-effectiveness of screening survivors for cardiomyopathy,8 our ability to demonstrate that screening low-risk survivors is not cost-effective even at substantially higher than anticipated ALVD treatment efficacies reinforces the need to reconsider current strategies for this group.
Recent challenges observed in the cost and delivery of health care in the United States and abroad have raised awareness of the need to reform health care approaches and spending.41,42 By using newly available data from the CCSS including survivors diagnosed between 1970 and 1999 to evaluate IGHG-endorsed risk groups, our work builds upon previous studies that also suggest that less frequent screening than currently recommended may be cost-effective.7,8 Our study addresses 2 key aspects of the existing IGHG cardiomyopathy guidelines that may enhance alignment of future recommendations with these objectives. First, our data support ongoing surveillance, including effective intervals, in high-risk survivors, whereas previous evidence for various aspects of these practices were only weak to moderate (Table 3). Specifically, we strengthen prior consideration to screen high-risk individuals more frequently than every 5 years, demonstrating that 2-year screening can prevent 1 in every 12 cases of HF and is cost-effective. Second, we provide compelling data to suggest that although low-risk individuals may benefit from screening, screening may not be cost-effective given their lower absolute HF risk. Importantly, we found that the preferred strategy for the moderate-risk group is uncertain, and additional data on screening performance and treatment effectiveness are needed to inform refinement of existing IGHG guidelines for these survivors.
Our study has a number of strengths. First, we bootstrapped individual-level CCSS data, allowing us to maintain covariance between patient characteristics and treatment exposures and to thus project more informative and accurate risk estimates. We also used the prospectively assessed and validated SJLIFE cohort to estimate ALVD risk and for model calibration. These factors facilitated robust model input regarding ALVD, HF, and mortality risk, rather than depending only upon published estimates. Second, we aligned tested groups with the IGHG risk stratification, which accounted for high-quality evidence to put forth practical, widely accepted guidelines. Thus, our results inform a broad-reaching community of medical providers and affect a global population of survivors.
Our results should be interpreted in the context of the following limitations. First, a number of assumptions informed our models. Among the most notable assumptions was the efficacy of treatment based on the SOLVD prevention trial, which studied the effect of treatment in individuals with HF unrelated to anthracyclines. To account for such assumptions, we incorporated a wide range of values into our sensitivity analyses, drawn from the strongest existing data regarding these parameters, and we are reassured by the stability of our high- and low-risk findings and our ability to conclude which strategies are most favorable. Similarly, we assumed a constant rate of progression from ALVD to HF as survivors aged; however, by using multiple calibrated parameter sets in our models, we believe we have accounted for this uncertainty and its impact on our results. In addition, we did not have access to an external cohort for model validation; however, we built our models on estimations from 2 large, long-standing survivor cohorts. In addition, overlap between these cohorts and similarities in grading strategies strengthened our ability to use both studies to estimate progression of ALVD to HF. We also assumed 100% compliance with screening recommendations because we focused on evaluating guideline recommendations; additional resources may be needed to increase adherence to and awareness of recommendations, which have been shown to be low among survivors and clinicians.43,44 Notably, IGHG risk groups do not account for age, sex, or modifiable health conditions that may increase the risk for ALVD or HF in an individual survivor. Other factors, such as risk factors for HF and individual preferences, will be important for survivors and providers to consider for individual screening decisions. Conversely, variability in model output across a range of model parameters precludes our ability to strongly endorse one surveillance strategy over another in moderate-risk survivors. It should also be noted that there are no benchmarks regarding the ideal number of screens per HF case averted in our population to provide context on whether the trade-offs in potential harms and benefits are reasonable. Finally, our results used health care cost estimates based on Medicare reimbursement rates, which may underestimate costs for younger survivors.
In summary, our model-based findings suggest screening survivors of childhood cancer at high risk for anthracycline-associated cardiomyopathy with echocardiograms at 2-year intervals is cost-effective. However, our data do not support the cost-effectiveness of screening to detect ALVD with echocardiography in the low-risk survivors, which in our study population would eliminate screening for approximately 40% of currently screened individuals. Future efforts to maximize guideline awareness and access to screening are necessary to implement cost-effective surveillance strategies.
Presented, in part, at the 51st Congress of the International Society of Paediatric Oncology, Lyon, France, October 23-26, 2019; the 55th Annual Meeting of the American Society of Clinical Oncology, Chicago, IL, May 31-June 4, 2019; and the 2019 North American Symposium on Late Complications after Childhood Cancer, Atlanta, GA, June 20-22, 2019.
Supported by the National Institutes of Health Grants No. P30CA21765 (Roberts), U24CA55727 (G.T.A.), and U01CA195547 (M.M.H. and L.L.R.); American Cancer Society Grant No. RSG-16018-01-CPHPS (J.M.Y.); and the American Lebanese Syrian Associated Charities. The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
See accompanying editorial on page 3827
Conception and design: Matthew J. Ehrhardt, Zachary J. Ward, Joy M. Fulbright, Kevin C. Oeffinger, Wendy M. Leisenring, Gregory T. Armstrong, Melissa M. Hudson, Lisa Diller, Saro H. Armenian, Jennifer M. Yeh
Financial support: Leslie L. Robison, Gregory T. Armstrong, Jennifer M. Yeh
Administrative support: Gregory T. Armstrong
Provision of study materials or patients: Leslie L. Robison, Gregory T. Armstrong, Melissa M. Hudson
Collection and assembly of data: Matthew J. Ehrhardt, Daniel A. Mulrooney, Wendy M. Leisenring, Todd M. Gibson, Rebecca M. Howell, Leslie L. Robison, Gregory T. Armstrong, Jennifer M. Yeh
Data analysis and interpretation: Matthew J. Ehrhardt, Zachary J. Ward, Qi Liu, Aeysha Chaudhry, Anju Nohria, William Border, Daniel A. Mulrooney, Kevin C. Oeffinger, Paul C. Nathan, Louis S. Constine, Eric J. Chow, Gregory T. Armstrong, Lisa Diller, Yutaka Yasui, Saro H. Armenian, Jennifer M. Yeh
Manuscript writing: All authors
Final approval of manuscript: All authors
Accountable for all aspects of the work: All authors
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.
Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).
Consulting or Advisory Role: Takeda Oncology
Research Funding: Amgen (Inst)
Uncompensated Relationships: Triple Gene
Honoraria: UpToDate, Springer, Lippincott
Travel, Accommodations, Expenses: Particle Therapy Cooperative Group of North America
Research Funding: The University of Texas MD Anderson Cancer Center
Consulting or Advisory Role: Oncology Research Information Exchange Network, Princess Máxima Center
Stock and Other Ownership Interests: Novartis (I), Amgen (I), Roche (I), Crispr Therapeutics (I), Baxter (I), Spark Therapeutics (I), Regenxbio (I), LabCorp (I), Portola Pharmaceuticals (I)
No other potential conflicts of interest were reported.
| 1. | Mulrooney DM, Hyun G, Ness KK, et al: Major cardiac events for adult survivors of childhood cancer treated in more recent eras: A report from the Childhood Cancer Survivor Study Cohort. BMJ 368:I6794, 2020 Crossref, Google Scholar |
| 2. | Kremer LC, van Dalen EC, Offringa M, et al: Anthracycline-induced clinical heart failure in a cohort of 607 children: Long-term follow-up study. J Clin Oncol 19:191-196, 2001 Link, Google Scholar |
| 3. | van der Pal HJ, van Dalen EC, van Delden E, et al: High risk of symptomatic cardiac events in childhood cancer survivors. J Clin Oncol 30:1429-1437, 2012 Link, Google Scholar |
| 4. | Lipshultz SE, Adams MJ, Colan SD, et al: 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 128:1927-1995, 2013 Crossref, Medline, Google Scholar |
| 5. | Kremer LC, Mulder RL, Oeffinger KC, et al: A worldwide collaboration to harmonize guidelines for the long-term follow-up of childhood and young adult cancer survivors: A report from the International Late Effects of Childhood Cancer Guideline Harmonization Group. Pediatr Blood Cancer 60:543-549, 2013 Crossref, Medline, Google Scholar |
| 6. | Armenian SH, Hudson MM, Mulder RL, 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 16:e123-e136, 2015 Crossref, Medline, Google Scholar |
| 7. | Yeh JM, Nohria A, Diller L: Routine echocardiography screening for asymptomatic left ventricular dysfunction in childhood cancer survivors: A model-based estimation of the clinical and economic effects. Ann Intern Med 160:661-671, 2014 Crossref, Medline, Google Scholar |
| 8. | Wong FL, Bhatia S, Landier W, et al: Cost-effectiveness of the Children’s Oncology Group long-term follow-up screening guidelines for childhood cancer survivors at risk for treatment-related heart failure. Ann Intern Med 160:672-683, 2014 Crossref, Medline, Google Scholar |
| 9. | Robison LL, Armstrong GT, Boice JD, et al: The Childhood Cancer Survivor Study: A National Cancer Institute-supported resource for outcome and intervention research. J Clin Oncol 27:2308-2318, 2009 Link, Google Scholar |
| 10. | Hudson MM, Ness KK, Nolan VG, et al: Prospective medical assessment of adults surviving childhood cancer: Study design, cohort characteristics, and feasibility of the St. Jude Lifetime Cohort study. Pediatr Blood Cancer 56:825-836, 2011 Crossref, Medline, Google Scholar |
| 11. | Hudson MM, Ehrhardt MJ, Bhakta N, et al: Approach for classification and severity grading of long-term and late-onset health events among childhood cancer survivors in the St. Jude Lifetime Cohort. Cancer Epidemiol Biomarkers Prev 26:666-674, 2017 Crossref, Medline, Google Scholar |
| 12. | Phillips SM, Padgett LS, Leisenring WM, et al: Survivors of childhood cancer in the United States: Prevalence and burden of morbidity. Cancer Epidemiol Biomarkers Prev 24:653-663, 2015 Crossref, Medline, Google Scholar |
| 13. | Caro JJ, Briggs AH, Siebert U, et al: Modeling good research practices—Overview: A report of the ISPOR-SMDM Modeling Good Research Practices Task Force-1. Value Health 15:796-803, 2012 Crossref, Medline, Google Scholar |
| 14. | Neumann PJ, Sanders GD, Russell LB, et al: Cost-Effectiveness in Health and Medicine (ed 2). New York, NY, Oxford University Press, 2017 Google Scholar |
| 15. | Eichler HG, Kong SX, Gerth WC, et al: Use of cost-effectiveness analysis in health-care resource allocation decision-making: How are cost-effectiveness thresholds expected to emerge? Value Health 7:518-528, 2004 Crossref, Medline, Google Scholar |
| 16. | Neumann PJ, Sandberg EA, Bell CM, et al: Are pharmaceuticals cost-effective? A review of the evidence. Health Aff (Millwood) 19:92-109, 2000 Crossref, Medline, Google Scholar |
| 17. | Levy D, Kenchaiah S, Larson MG, et al: Long-term trends in the incidence of and survival with heart failure. N Engl J Med 347:1397-1402, 2002 Crossref, Medline, Google Scholar |
| 18. | Mahmood SS, Levy D, Vasan RS, et al: The Framingham Heart Study and the epidemiology of cardiovascular disease: A historical perspective. Lancet 383:999-1008, 2014 Crossref, Medline, Google Scholar |
| 19. | Children’s Oncology Group: Long-Term Follow-Up Guidelines for Survivors of Childhood, Adolescent, and Young Adult Cancers, Version 5.0. Monrovia, CA, Children’s Oncology Group, 2018 Google Scholar |
| 20. | Howell RM, Smith SA, Weathers RE, et al: Adaptations to a generalized radiation dose reconstruction methodology for use in epidemiologic studies: An update from the MD Anderson Late Effect Group. Radiat Res 192:169-188, 2019 Crossref, Medline, Google Scholar |
| 21. | National Heart, Lung, and Blood Institute: Incidence and Prevalence: 2006 Chart Book on Cardiovascular and Lung Diseases. Bethesda, MD, National Institutes of Health, 2006 Google Scholar |
| 22. | Owan TE, Redfield MM: Epidemiology of diastolic heart failure. Prog Cardiovasc Dis 47:320-332, 2005 Crossref, Medline, Google Scholar |
| 23. | Halloran M, Berry D (eds): Marginal structural models versus structural nested models as tools for causal inference, in Statistical Models in Epidemiology: The Environment and Clinical Trials. New York, NY, Springer-Verlag, 1999, pp 95-133 Google Scholar |
| 24. | National Cancer Institute: Common Terminology Criteria for Adverse Events. https://ctep.cancer.gov/protocolDevelopment/electronic_applications/ctc.htm#ctc_40 Google Scholar |
| 25. | Wang TJ, Evans JC, Benjamin EJ, et al: Natural history of asymptomatic left ventricular systolic dysfunction in the community. Circulation 108:977-982, 2003 Crossref, Medline, Google Scholar |
| 26. | Jones NR, Roalfe AK, Adoki I, et al: Survival of patients with chronic heart failure in the community: A systematic review and meta-analysis. Eur J Heart Fail 21:1306-1325, 2019 Crossref, Medline, Google Scholar |
| 27. | Armstrong GT, Chen Y, Yasui Y, et al: Reduction in late mortality among 5-year survivors of childhood cancer. N Engl J Med 374:833-842, 2016 Crossref, Medline, Google Scholar |
| 28. | Arias E, Heron M, Xu J: United States life tables, 2014. Natl Vital Stat Rep 66:1-64, 2017 Google Scholar |
| 29. | Greupner J, Zimmermann E, Grohmann A, et al: Head-to-head comparison of left ventricular function assessment with 64-row computed tomography, biplane left cineventriculography, and both 2- and 3-dimensional transthoracic echocardiography: Comparison with magnetic resonance imaging as the reference standard. J Am Coll Cardiol 59:1897-1907, 2012 Crossref, Medline, Google Scholar |
| 30. | Yancy CW, Jessup M, Bozkurt B, et al: 2013 ACCF/AHA guideline for the management of heart failure: A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 62:e147-e239, 2013 Crossref, Medline, Google Scholar |
| 31. | Exner DV, Dries DL, Waclawiw MA, et al: Beta-adrenergic blocking agent use and mortality in patients with asymptomatic and symptomatic left ventricular systolic dysfunction: A post hoc analysis of the Studies of Left Ventricular Dysfunction. J Am Coll Cardiol 33:916-923, 1999 Crossref, Medline, Google Scholar |
| 32. | Centers for Medicare and Medicaid Services: Research, statistics, data and systems. https://www.cms.gov/Research-Statistics-Data-and-Systems/Research-Statistics-Data-and-Systems Google Scholar |
| 33. | Truven Health Analytics: IBM Micromedex Red Book. https://www.ibm.com/products/micromedex-red-book Google Scholar |
| 34. | McMurray JJ, Ostergren J, Swedberg K, et al: Effects of candesartan in patients with chronic heart failure and reduced left-ventricular systolic function taking angiotensin-converting-enzyme inhibitors: The CHARM-Added trial. Lancet 362:767-771, 2003 Crossref, Medline, Google Scholar |
| 35. | Bureau of Labor Statistics: U.S. Department of Labor: Occupational Employment and Wages. Washington, DC, US Government Printing Office, 2017 Google Scholar |
| 36. | Göhler A, Geisler BP, Manne JM, et al: Utility estimates for decision-analytic modeling in chronic heart failure: Health states based on New York Heart Association classes and number of rehospitalizations. Value Health 12:185-187, 2009 Crossref, Medline, Google Scholar |
| 37. | Yeh JM, Hanmer J, Ward ZJ, et al: Chronic conditions and utility-based health-related quality of life in adult childhood cancer survivors. J Natl Cancer Inst 108:djw046, 2016 Crossref, Medline, Google Scholar |
| 38. | Hutchins R, Viera AJ, Sheridan SL, et al: Quantifying the utility of taking pills for cardiovascular prevention. Circ Cardiovasc Qual Outcomes 8:155-163, 2015 Crossref, Medline, Google Scholar |
| 39. | Knudsen AB, Zauber AG, Rutter CM, et al: Estimation of benefits, burden, and harms of colorectal cancer screening strategies: Modeling study for the US Preventive Services Task Force. JAMA 315:2595-2609, 2016 Crossref, Medline, Google Scholar |
| 40. | Mandelblatt JS, Stout NK, Schechter CB, et al: Collaborative modeling of the benefits and harms associated with different U.S. breast cancer screening strategies. Ann Intern Med 164:215-225, 2016 Crossref, Medline, Google Scholar |
| 41. | Cassel CK, Guest JA: Choosing wisely: Helping physicians and patients make smart decisions about their care. JAMA 307:1801-1802, 2012 Crossref, Medline, Google Scholar |
| 42. | Norton WE, Chambers DA, Kramer BS: Conceptualizing de-implementation in cancer care delivery. J Clin Oncol 37:93-96, 2019 Link, Google Scholar |
| 43. | Yan AP, Chen Y, Henderson TO, et al: Adherence to surveillance for second malignant neoplasms and cardiac dysfunction in childhood cancer survivors: A Childhood Cancer Survivor Study. J Clin Oncol 38:1711-1722, 2020 Link, Google Scholar |
| 44. | Suh E, Daugherty CK, Wroblewski K, et al: General internists’ preferences and knowledge about the care of adult survivors of childhood cancer: A cross-sectional survey. Ann Intern Med 160:11-17, 2014 Crossref, Medline, Google Scholar |
| 45. | Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics: Compressed mortality file: Years 1968-1978 with ICD-8 codes, 1979-1998 with ICD-9 codes and 1999-2016 with ICD-10 codes. https://wonder.cdc.gov/wonder/help/cmf.html Google Scholar |
| 46. | Yeh JM, Ward ZJ, Chaudhry A, et al: Life expectancy of adult survivors of childhood cancer over 3 decades. JAMA Oncol 6:350, 2020 Crossref, Medline, Google Scholar |
