|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2008.19.7343 on October 27 2008 © 2008 American Society of Clinical Oncology.
Quality of Reporting Primary Outcomes in Phase II Cancer Trials
Departments of Internal Medicine, Federal University of Sao Paulo, Sao Paulo, Brazil
Departments of Medicine, University of Toronto, and Medical Oncology & Hematology, Princess Margaret Hospital, Toronto, Canada To the Editor: There has been a dramatic rise in the number of early phase trials in oncology.1 In the cancer drug development process, the main goals of phase II trials are to evaluate drug efficacy and safety, and to select promising therapies to compare against standard of care in randomized phase III trials.2 In order for the human and financial investments made in these trials to be returned, it is essential that study design and reporting be rigorous and without bias. Studies on the quality of reporting of phase II trials demonstrated that both statistical design and study results have been poorly reported.3,4 Even though these studies reported on the type of end points selected by phase II trials, neither of them looked specifically at the quality of reporting of primary end points (PEP), and this is the focus of the this correspondence.
We analyzed all phase II trials as part of a cross-sectional survey of phase II and III trials of systemic anticancer drugs and supportive care agents published in five major clinical oncology journals in 2005, with the purpose to evaluate the relationship(s) between study sponsorship and/or authors’ self-disclosed financial conflicts of interest and study results and interpretation.5 From all single-arm phase II trials, we (R.P.R., V.D.) independently abstracted information on whether the PEP was clearly stated (explicitly stated in the article's abstract or methods as being primary and/or if the end point was used to calculate the sample size), what the end point was, whether it was standard and the actual result. End points broadly defined as feasibility, efficacy, or safety were considered as unclearly stated; PEP considered standard for phase II trials were response rate, symptom control, and time to progression, whereas overall survival and prespecified response rate lower than 20% were considered nonstandard.2 The results of PEP were classified as either positive, negative, or unclear. For clearly stated PEP, their results were considered positive if it was better than a pre-especified value described in the text; otherwise a response rate From our sample, 125 phase II trials were published in the selected five journals in 2005 and their characteristics have been described previously.5 Characteristics of the reporting of PEP are presented in Table 1.
This report shows that the quality of reporting PEP in phase II cancer clinical trials needs improvement, with almost one third not explicitly describing the study primary goal. In addition, although the great majority of trials used response rate as their primary outcome, there were studies that used nonstandard end points as their PEP. Mariani et al3 found that 96% of 295 phase II trials published in 1995 and 1997 used response rate as the primary end point. In this analysis, we found that 80% of phase II with clear definition of PEP and only 58% of all phase II used response rate as their main outcome. Such difference is likely related to the fact that almost one third of trials did not explicitly describe the PEP. We are unaware of studies that have evaluated the quality of reporting in phase II cancer trials. A survey of the quality of abstracts of randomized phase IIII trials on lung, breast, and colorectal cancer published in six major oncology journals from 1975 to 2004 showed that only 38% of them reported the study PEP.7 With respect to phase II trials, a study of phase II trials published a decade ago demonstrated that the quality of statistical design reporting and study results was poor, with only 19% of articles reporting statistical components.3 Although a subsequent study showed that the quality of statistical design reporting in articles published in 2000 was better, there was room for further improvement.4 These last two studies did not analize the reporting of study PEP. Despite the originality of this report and the use of standard definitions of study end point and results,2,8 our analysis has a number of limitations. First, we selected high impact clinical cancer journals, with more strict peer-reviewing process, what may have underestimate the proportion of phase II cancer trials with poor reporting. We included only single-arm phase II trials, did not assess the quality of statistical methods, which has already been covered by other studies.3,4,9 Lastly, the reasons why the reporting of PEP in phase II trials was poor were not examined but less strict peer-review process and insufficient methodologic knowledge on the part of authors may have influenced our findings. The quality of reporting of end points in cancer phase II trials directly influences the interpretation of study results, and thus affect the design and development of subsequent phase III trials. We suggest journals improve the peer-review process and be more rigorous about the quality of methodology reporting. Improving the quality of study reporting will likely help readers better interpret clinical cancer research and put studies into context. AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a "U" are those for which no compensation was received; those relationships marked with a "C" were compensated. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors. Employment or Leadership Position: Rachel P. Riechelmann, Novartis Pharmaceuticals (C) Consultant or Advisory Role: None Stock Ownership: None Honoraria: None Research Funding: None Expert Testimony: None Other Remuneration: None
NOTES published online ahead of print at www.jco.org on October 27, 2008 REFERENCES
1. Hillner BE: Trends in clinical trials reports in common cancers between 1989 and 2000. J Clin Oncol 21:1850-1858, 2003 2. Eisenhauer EA: Phase I and II trials of novel anti-cancer agents: Endpoints, efficacy and existentialism. Ann Oncol 9:1047-1052, 1998 3. Mariani L, Marubini E: Content and quality of currently published phase II cancer trials. J Clin Oncol 18:429-436, 2000 4. Thezenas S, Duffour J, Culine S, et al: Five-year change in statistical designs of phase II trials published in leading cancer journals. Eur J Cancer 40:1244-1249, 2004[CrossRef][Medline] 5. Riechelmann RP, Dounaevskaia V, Taback N, et al: Source of funding, conflict of interest (COI), and the interpretation of cancer clinical trials. J Clin Oncol 25:329s, 2007 (suppl; abstr 6530) 6. Fleming TR: One-sample multiple testing procedure for phase II clinical trials. Biometrics 38:143-151, 1982[CrossRef][Medline] 7. Cescon DW, Booth CM, Wang L, et al: Trends in the quality of abstracts of articles describing randomized controlled cancer trials (RCTs) from 1975-2004. J Clin Oncol 25:332s, 2007 (suppl; abstr 6541) 8. Wittes RE, Marsoni S, Simon R, et al: The phase II trial. Cancer Treat Rep 69:1235-1239, 1985[Medline] 9. Perrone F, Di Maio M, De Maio E, et al: Statistical design in phase II clinical trials and its application in breast cancer. Lancet Oncol 4:305-311, 2003[CrossRef][Medline]
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2008 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
20% was considered positive.
