|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2003 American Society for Clinical Oncology
In Reply:
1 University of Rochester Medical Center, Golisano Children’s Hospital at Strong, and University of Rochester School of Medicine and Dentistry, Rochester, NY Dr van Dalen et al correctly state that our study1 found transient improvement with enalapril therapy in this population but incorrectly imply that asymptomatic patients and symptomatic patients with heart failure were only presented together. We also presented patients with asymptomatic left ventricular dysfunction separately from patients with congestive heart failure, and we reported outcomes during follow-up for both groups separately.1 Dr van Dalen et al point out methodological limitations associated with case series, such as selection bias. Our article represented a complete consecutive series with extremely long follow-up.1 No patient who met eligibility criteria was excluded. We made no recommendation for lifelong therapy with angiotensin-converting enzyme inhibitors.1 We have agreed that a randomized, controlled trial is needed to answer the question posed by van Dalen et al regarding the benefit of lifetime enalapril therapy in this population.1 In this reply, we will review necessary components of such a clinical trial, which would at least be able to address the clinical efficacy during 4 years of therapy, and review some of the barriers to conducting such a clinical trial resulting from the unique aspects of late effects in long-term survivors of childhood cancer. In the United States, most National Institutes of Health–funded clinical trials are limited to 4 years, leading us to consider a 4-year enalapril clinical trial. We agree that it is important to study anthracycline-treated childhood cancer survivors with asymptomatic cardiac dysfunction, and we will base this design on that population alone. However, our data suggest that it becomes at least as important to understand whether this intervention, or others, offers benefit for those with heart failure since our paper indicates that the development of late congestive heart failure portends a bleak future.1 There are currently many biomarkers that could be measured during an enalapril clinical trial in this population, including echocardiographic, electrocardiographic, exercise-associated, functional status, and serum-associated biomarkers. Unfortunately, no biomarker has yet been validated as a surrogate end point for clinically significant end points such as heart failure, heart transplantation, cardiac, and all-cause mortality in this population.2 Therefore, the use of these clinically-significant primary end points becomes necessary for such a clinical trial to determine the clinical efficacy of this therapy. Randomized clinical trials have certain advantages over case series (in particular, a greater chance to control bias), and they should be encouraged.2 A randomized design by itself, however, does not ensure that the most accurate answers will be generated. In this case, for example, our data showed that the use of enalapril in long-term survivors of childhood cancer resulted in a delay in progressive late doxorubicin cardiotoxicity, but not a cure.1 Based on our work, early results from randomized clinical trials are likely to be misleading; long-term follow-up is essential.1 A randomized trial using measurement techniques that are unable to accurately detect outcomes of interest is also problematic.3 In our study, we not only used centrally remeasured serial echocardiographic parameters to improve reliability, but we also assessed for primary clinical end points, such as heart failure, cardiac transplantation, and death.1 We endorse properly conducted and analyzed randomized clinical trials. Case series like ours, however, provide valuable data to improve the quality of such randomized trials and for care for patients today.2 To properly perform such a clinical trial, an appropriate sample size needs to be accrued. An estimate of the enalapril effect size is given from the Studies of Left Ventricular Dysfunction prevention trial, in which 4,228 asymptomatic adults with left ventricular dysfunction not related to chemotherapy were enrolled onto a randomized trial of enalapril versus placebo.4 This study noted a 20% 4-year cardiovascular mortality in the placebo group.4 The treated group had a 12% reduction in mortality from cardiac causes, which was not a statistically significant improvement. During the same 4-year interval, there was a 25% incidence of hospitalization for new or worsening heart failure in the placebo group and 20% in the treatment group, which was statistically significant (P < .001).4 We therefore estimate the enalapril effect size as a 20% reduction in the risk for new or worsening heart failure.
The time to development of congestive heart failure in survivors of childhood cancer was measured by Kremer et al.5 They reported an incidence of heart failure of 2.8% over 6.3 years of follow-up, which is 0.44% per year, or a 1.8% cumulative incidence throughout 4 years. Using a However, the situation is actually much worse than that, given that 13 of 17 of the incident cases in the series by Kremer et al actually occurred within the first year of therapy.5 The group with early heart failure is a troublesome group of cases but is not the cohort with which we are primarily concerned in our paper1 and in this proposed clinical trial. If one considers patients with asymptomatic left ventricular dysfunction who are more than one year after therapy, Kremer et al 5 had 4 of 594 cases over a 6.2-year interval, which is an incidence of 0.1% per year. The sample size required to detect a reduction from 0.4% to 0.32% over 4 years would be 176,000. Similarly, Kremer et al 5 determined an incidence of death of 4 of 607 over 6.2 years, which is again 0.1% per year. The ability to detect a reduction in deaths from 0.4% down to 0.35% over 4 years would require a considerably larger sample size than noted above for heart failure—and, in fact, some of these deaths may well have been early after therapy. (This is not specified in the paper by Kremer et al.)5 Unfortunately, given the sample size required to reach a clinically meaningful end point, our conclusion is that the appropriate clinical trial of enalapril in this population will never happen. REFERENCES
1. Lipshultz SE, Lipsitz SR, Sallan SE, et al: Long-term enalapril therapy for left ventricular dysfunction in doxorubicin-treated survivors of childhood cancer. J Clin Oncol 20:4517–4522, 2002 2. Lipshultz SE: Ventricular dysfunction clinical research in infants, children and adolescents. Prog Pediatr Cardiol 12:1–28, 2000[CrossRef][Medline]
3. Lipshultz SE, Easley K, Orav EJ, et al: The reliability of multicenter pediatric echocardiographic measurements of left ventricular structure and function. Circulation 104:310–316, 2001 4. The SOLVD Investigators: Effect of enalapril on mortality and the development of heart failure in asymptomatic patients with reduced left ventricular ejection fractions. N Engl J Med 327:685–691, 1992[Abstract]
5. Kremer LCM, van Dalen EC, Offringa M, et al: Anthracycline-induced clinical heart failure in a cohort of 607 children: A long-term follow-up study. J Clin Oncol 19:191–196, 2001
Related Article
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2003 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
2 test for equal proportions with a 5% Type I error rate and 80% power, the sample size required to detect a 20% reduction in heart failure incidence (1.8% down to 1.55% in the treated group) for a 4-year trial is 83,000. 
