Access the full text.
Sign up today, get DeepDyve free for 14 days.
A prevalent ideology in the oncology community holds that clinical trials represent the best treatment option for patients with cancer. This belief, which likely developed in response to concerns over the low proportions of adult patients with cancer who enroll in trials,1 the historical reluctance of third-party payers to cover trials' routine-care costs,2 and the psychological need to reconcile the simultaneous pursuit of scientific and therapeutic aims,3 finds expression in the National Comprehensive Cancer Network's assertion, reiterated in each of its guidelines for cancer treatment, that “the best management for any cancer patient is in a clinical trial.”4 That this viewpoint persists in authoritative, otherwise evidence-based guidance is surprising in light of accumulating data suggesting that patients treated within trials have outcomes that are on average neither better nor worse than those of comparable patients treated outside trials.5-8 At least 3 overlapping literatures rely on data from studies, almost always observational in nature, that compare outcomes among patients treated within vs outside trials. The first of these asks whether a “trial effect,” or outcome benefit due to trial participation, exists.5,7,9 Authors who pose this question typically seek empirical support for arguments, directed at both patients and at clinicians, favoring greater trial enrollment. The second, which reflects a desire to challenge regulatory and other barriers to trial completion, asks whether there is any measurable disadvantage, at least with respect to major clinical outcomes, associated with trial enrollment.6 The third literature, which addresses questions about the generalizability of data from clinical trials and consequently about the quality of the evidence base for medical decision making, asks whether trial participants differ in prognostically important ways from nonparticipants.10 In this issue of the Archives, Koschmann et al11 shed important new light on these clinical and policy questions. They identify a retrospective cohort of children with newly diagnosed acute lymphoblastic leukemia (ALL) treated with curative intent at Seattle Children's Hospital during 1997 to 2005. They then compare the sociodemographic and clinical characteristics, as well as the event-free survival, of the 157 children who enrolled in one of several cooperative-group clinical trials with those of the 165 children treated outside a trial. Two important findings emerge from these comparisons. First, there were no statistically significant sociodemographic or prognostic differences between the 2 groups, arguing against any important selection bias in trial enrollment in the context of pediatric ALL. Second, the event-free survival curves of the 2 groups were indistinguishable from one another, both overall and within the standard-risk and high-risk subgroups of children. This observation suggests that, at least when considering pediatric ALL at a tertiary care center during the current era, treatment outside a trial does not confer any meaningful outcome disadvantage as compared with trial enrollment. Although these findings have limitations—they reflect a single center, may not extend to diseases for which baseline outcomes are less good or for which routine treatment is less regimented, and may not hold during periods of rapid improvement in outcomes due to the introduction of effective new therapeutic approaches—they are undoubtedly valid within the context from which they were derived. Should we lament or celebrate the findings of Koschmann et al? Those who seek evidence with which to encourage clinicians to recommend trials, and to persuade patients and parents to enroll, will no doubt be disappointed. One may advocate for trial enrollment on the grounds that trial enrollment is likely to be directly beneficial to participants, or on the grounds that participation contributes to the public good. Data such as those of Koschmann et al undercut the former argument, leaving the latter as the primary reason for considering a trial. Other readers, however, will find much to celebrate. For example, those whose primary concern is to support informed decision making about trial participation by patients and families will find these data helpful in validating either trial enrollment or receipt of standard care outside a trial as legitimate treatment options for children with ALL, as well as in combating therapeutic misconceptions about the nature and purpose of trials.12 Second, the excellent outcomes observed outside the trial setting will provide comfort to clinicians and families when, because of ineligibility or the absence of an open study, children do not have the option to enroll in a trial. Third, those who wish to present institutional review boards and other oversight bodies with evidence that trial participation does not systematically disadvantage children with ALL will welcome these data. Finally, in light of the comparability of the trial and nontrial populations, those who worry about whether the results of pediatric oncology clinical trials are generalizeable to childhood cancer treatment more broadly will be reassured. Should the absence of evidence that clinical trial enrollment per se improves outcomes reduce public, professional, or political enthusiasm for a robust, vigorously funded clinical trials program? Considerable evidence indicates that clinical trials are associated with substantial benefits as measured by both public health and economic gains. An analysis of the costs and benefits of all phase 3 randomized trials supported by the National Institute of Neurological Disorders and Stroke prior to 2000 estimated the 10-year public health and economic returns on the institute's $335 million investment in the 28 trials at 470 000 quality-adjusted life-years and $15.5 billion, respectively.13 Second, and more subtly, some evidence suggests that the conduct of clinical trials raises standards of care and outcomes for all patients treated at participating centers, not just for those who actually take part in the trials. Karjalainen and Palva14 used registry data to conduct a population-based study of outcomes among patients younger than 71 years with multiple myeloma in Finland before and after the introduction of a clinical trials program into 1 of 2 regions of the country. Among all patients in the trial region, including those who did not enroll in the trials, 5-year survival improved by 14% after the introduction of the trial program; in contrast, among patients in the nontrial region, there was no improvement in survival between the 2 periods. Taken together, these studies suggest that, even in the absence of a trial effect, there are multiple mechanisms by which clinical trials programs contribute to improved outcomes for patients. In the modern era, 5-year survival rates for children with ALL are 80% to 90%.15 It is likely that these remarkable outcomes owe much not only to the legacy of treatment insights derived from past trials, but also to the happy side effect of improved quality attributable to the existence of a cooperative clinical trials program and infrastructure in the settings in which most children with cancer receive their care. Correspondence: Dr Joffe, Dana-Farber Cancer Institute, 44 Binney St, Boston, MA 02115 (firstname.lastname@example.org). Author Affiliation: Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts. Financial Disclosure: Dr Joffe is a paid member of a Data Monitoring Committee for the Genzyme Corporation. References 1. Murthy VHKrumholz HMGross CP Participation in cancer clinical trials: race-, sex-, and age-based disparities. JAMA 2004;291 (22) 2720- 2726PubMedGoogle ScholarCrossref 2. Rettig RAJacobson PAFarquhar CMAubry WM False Hope: Bone Marrow Transplantation for Breast Cancer. Oxford, England Oxford University Press2007; 3. Miller FGRosenstein DLDeRenzo EG Professional integrity in clinical research. JAMA 1998;280 (16) 1449- 1454PubMedGoogle ScholarCrossref 4. NCCN Clinical Practice Guidelines in Oncology. National Comprehensive Cancer Network Web site. http://www.nccn.org/professionals/physician_gls/f_guidelines.asp. Accessed November 24, 2009Google Scholar 5. Peppercorn JMWeeks JCCook EFJoffe S Comparison of outcomes in cancer patients treated within and outside clinical trials: conceptual framework and structured review. Lancet 2004;363 (9405) 263- 270PubMedGoogle ScholarCrossref 6. Gross CPKrumholz HMVan Wye GEmanuel EJWendler D Does random treatment assignment cause harm to research participants? PLoS Med 2006;3 (6) e188PubMedGoogle ScholarCrossref 7. Braunholtz DAEdwards SJLilford RJ Are randomized clinical trials good for us (in the short term)? evidence for a “trial effect”. J Clin Epidemiol 2001;54 (3) 217- 224PubMedGoogle ScholarCrossref 8. Vist GEBryant DSomerville LBirminghem TOxman AD Outcomes of patients who participate in randomized controlled trials compared to similar patients receiving similar interventions who do not participate. Cochrane Database Syst Rev 2008; (3) MR000009PubMedGoogle Scholar 9. Robinson WRRitter JRogers ASTedjarati SLieberenz C Clinical trial participation is associated with improved outcome in women with ovarian cancer. Int J Gynecol Cancer 2009;19 (1) 124- 128PubMedGoogle ScholarCrossref 10. Stevens JMMacdougall FJenner MOakervee HCavenagh JLister AT Patterns of recruitment into acute myeloid leukaemia (AML) 15 and outcome for young patients with AML at a single referral centre. Br J Haematol 2009;145 (1) 40- 44PubMedGoogle ScholarCrossref 11. Koschmann CThomson BHawkins DS No evidence of a trial effect in newly diagnosed pediatric acute lymphoblastic leukemia. Arch Pediatr Adolesc Med 2010;164 (3) 214- 217Google ScholarCrossref 12. Appelbaum PSRoth LHLidz CWBenson PWinslade W False hopes and best data: consent to research and the therapeutic misconception. Hastings Cent Rep 1987;17 (2) 20- 24PubMedGoogle ScholarCrossref 13. Johnston SCRootenberg JDKatrak SSmith WSElkins JS Effect of a US National Institutes of Health programme of clinical trials on public health and costs. Lancet 2006;367 (9519) 1319- 1327PubMedGoogle ScholarCrossref 14. Karjalainen SPalva I Do treatment protocols improve end results? a study of survival of patients with multiple myeloma in Finland. BMJ 1989;299 (6707) 1069- 1072PubMedGoogle ScholarCrossref 15. 5-year relative survival rates by year dx, by age at diagnosis/death, acute lymphocytic leukemia, all races, both sexes, 1975-2001. Surveillance Epidemiology and End Results Web site. http://seer.cancer.gov/faststats/selections.php?run&output=1&data=4&statistic=6&cancer=90&year=200905&race=1&sex=1&subSite=92&series=age&age=15. Accessed November 25, 2009 Google Scholar
Archives of Pediatrics & Adolescent Medicine – American Medical Association
Published: Mar 1, 2010
Access the full text.
Sign up today, get DeepDyve free for 14 days.