Author + information
- Published online March 25, 2019.
- aDuke Clinical Research Institute, Durham, North Carolina
- bDivision of Cardiology, Duke University School of Medicine, Durham, North Carolina
- cDuke Forge, Duke University School of Medicine, Durham, North Carolina
- dDepartment of Medicine, Stanford University, Stanford, California
- eVerily Life Sciences (Alphabet), South San Francisco, California
- ↵∗Address for correspondence:
Dr. Stephen J. Greene, Duke Clinical Research Institute, 200 Morris Street, Durham, North Carolina 27701.
Phase III clinical trials for heart failure (HF) have rapidly moved toward globalization in recent years, a movement that has fostered inclusive and collaborative research consistent with HF as a global health problem. However, this globalization has largely arisen from mixed motives. The expansion of clinical trials into additional populations has been laudable, but sponsors may seek such environments for expediency to escape poor enrollment and high costs in North America and Western Europe, their primary markets (1). This tradeoff between the positive motive of globalization and the questionable motive of “offshoring” may also apply within regions and individual countries; the populations included in clinical trials are less reflective of representative samples, and more reflective of speeds of enrollment. Therefore, it is not surprising that despite identical trial selection criteria for all enrolling sites, participants enrolled in these global programs often vary markedly across geographic regions, including in socioeconomic status and perhaps other “social determinants of health.”
These observations are supported by several secondary analyses of HF trials demonstrating varying degrees and patterns of global heterogeneity in patient profiles and outcomes (2,3). Although some degree of global variation may be expected and beneficial for worldwide generalizability, the magnitude of difference in some prior HF programs and the absence of a clear explanation is cause for concern and skepticism. One striking example, the ASTRONAUT (Aliskiren Trial on Acute Heart Failure Outcomes) trial, found rates of 12-month cardiovascular mortality ranging from 6.5% in North America to 24.8% in Asia/Pacific (3). In this and other analyses, marked regional differences in mortality and hospitalization endpoints persist despite adjustment for traditional biomedical prognostic factors (2,3). Some recent trials have also observed differential treatment effects of the study intervention across regions (4,5). Thus, globalization presents an added challenge for regulators, guideline writers, payers, and clinicians who must judge whether results from global trials are generalizable to their respective patients.
Although prior analyses almost uniformly describe patients by geography or continent, it is obvious that simple location on a map is not mediating differences. Geographic location may be a crude surrogate for underlying driving factors that may vary across multiple important domains relevant to clinical outcomes (Table 1) (6). At every level, region-, country-, and site-specific factors blend together to produce enrollment of patients with different mixes of clinical and social risk, as well as different event rates, operational performance, and data quality.
In this issue of JACC: Heart Failure, Dewan et al. (7) examine global heterogeneity within HF trials through the lens of country-level income inequality. The authors pooled data from the PARADIGM-HF (Prospective Comparison of ARNI [Angiotensin Receptor Neprilysin Inhibitor] with ACEI [Angiotensin Converting Enzyme Inhibitor] to Determine Impact on Global Mortality and Morbidity in Heart Failure) and the ATMOSPHERE (Aliskiren Trial to Minimize Outcomes in Patients with Heart Failure) trials, 2 global programs that began enrollment in 2009 with near identical selection criteria (8,9). The combined dataset included 15,126 participants from 54 counties. The investigators then applied the Gini coefficient, a measure of economic inequality, to categorize participants by level of income inequality of the enrolling country. Patient and country characteristics and clinical outcomes differed markedly by Gini coefficient tertile. These differences in survival persisted after adjustment for patient-level and country-level clinical factors and in both tertile and continuous-variable analyses. By contrast, there was no significant continuous relationship between income inequality and risk of HF hospitalization. In fact, although adjusted HF hospitalization risk was highest in the tertile of greatest inequality, adjusted risk was lower in the intermediate tertile than the group with least inequality.
We congratulate the authors for a novel analysis that provides a new perspective on global heterogeneity within HF trials (7). Nonetheless, some limitations should be noted. First, great caution is needed when generalizing these trial-specific findings to real-world health outcomes in the respective countries. Participants in the PARADIGM-HF and ATMOSPHERE trials represent a population fully treated with angiotensin-converting enzyme/angiotensin II receptor blocker therapy at screening who successfully completed a multiweek run-in period during which they received high doses of study drugs. Likewise, patients who agreed to trial participation and received care at trial centers are unlikely to fully represent the country’s overall HF population. It is also unclear whether barriers or incentives to trial participation (and hence country-level representativeness of trial participants) varied with country income inequality. Thus, although these important data may generate hypotheses for public health policies, the purest application concerns future clinical trial design and the countries included in global trials. Second, concomitant medical and device therapies were not included in multivariable models. Notably, it is unclear whether accounting for very low use of implantable cardioverter-defibrillators (4.4%) and cardiac resynchronization therapy (2.7%) within countries with greatest income inequality would have attenuated the excess mortality risk for those patients. Third, although associations with income inequality are clear, these data cannot identify key features of income equality that influence trial outcomes. In this context, it is striking that patients from countries with greatest inequality tended to have markedly better baseline quality of life, functional class, and clinical congestion (as well as other favorable prognostic signs), yet paradoxically poor survival with event curves separating before 6 months. It is unclear to what degree these findings reflect inconsistencies with HF diagnosis/interpretation of selection criteria, differences in follow-up care, social and behavioral variation, or other factors.
To our knowledge, the only prior study examining economic characteristics of enrolling countries in a HF trial comes from the ASCEND-HF (Acute Study of Clinical Effectiveness of Nesiritide in Decompensated Heart Failure) trial (10). After categorizing patients by per-capita gross national product and health development index of the enrolling country, investigators found no associations with 180-day mortality after adjustment for geographic region and patient factors. However, an independent association between higher country income and excess risk of hospitalization and trial protocol discontinuation was observed. Reconciling the exact nature of the socioeconomic relationships seen in the ASCEND-HF trial with the current analysis is difficult and potentially affected by differences in trial population (i.e., acute versus chronic HF), inclusion criteria, and countries included. Nevertheless, both analyses show that geographic region is an imperfect surrogate for key economic characteristics. Although Dewan et al. (7) show some overlap between geographic location and degree of income inequality (e.g., North America entirely within intermediate group), Western European, Central European, and Asia/Pacific countries extended across all tertiles. Likewise, in the ASCEND-HF analysis, all country-level income groupings included countries from multiple geographic regions (10). We would expect similar variation within any geographic unit (region, country, city, or county), pointing to policies that produce income inequality as targets for change, and to examination of social factors as covariates relevant to trial recruitment and enrollment.
In summary, the findings of Dewan et al. (7) introduce country-level income inequality as an important dimension of variation within global HF trials. As we work toward understanding the mechanisms of global variation and properly accounting and preparing for it within HF trial design, similar investigations studying the role of site- and country-level factors on protocol adherence, enrollment rate, and study endpoints are encouraged. Regardless, the continued execution of large HF trials with the degree of global variation in enrollment, study conduct, and patient outcomes seen in prior experiences requires ongoing assessment. Globalization of clinical research can be a powerful social good, but we must understand and mitigate unintended consequences if we are to harness its full potential.
↵∗ Editorials published in JACC: Heart Failure reflect the views of the authors and do not necessarily represent the views of JACC: Heart Failure or the American College of Cardiology.
Dr. Greene is supported by the National Heart, Lung, and Blood Institute T32 post-doctoral training grant T32HL069749-14 and a Heart Failure Society of America/Emergency Medicine Foundation Acute Heart Failure Young Investigator Award funded by Novartis; and has received research support from Amgen and Novartis. Dr. Califf was the Commissioner of Food and Drugs for the U.S. Food and Drug Administration (FDA) from February 2016 to January 2017 and Deputy Commissioner for Medical Products and Tobacco for the U.S. FDA from February 2015 to January 2016; has served on the corporate board for Cytokinetics; has been the board chair for the People-Centered Research Foundation; has received consulting fees from Merck, Biogen, Genentech, Eli Lilly, and Boehringer Ingelheim; and is employed as a scientific advisor by Verily Life Sciences (Alphabet).
- 2019 American College of Cardiology Foundation
- Pfeffer M.A.,
- Claggett B.,
- Assmann S.F.,
- et al.
- Packer M.,
- O'Connor C.,
- McMurray J.J.V.,
- et al.
- Dewan P.,
- Rørth R.,
- Jhund P.S.,
- et al.
- Greene S.J.,
- Hernandez A.F.,
- Sun J.L.,
- et al.