|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2008.21.2951 on March 2 2009 © 2009 American Society of Clinical Oncology.
Lapatinib Plus Paclitaxel As First-Line Therapy for Patients With Human Epidermal Growth Factor Receptor 2–Positive Metastatic Breast Cancer: Inappropriate Conclusions From a Company-Sponsored Study?Division of Medical Oncology and Hematology, Princess Margaret Hospital, Toronto, Ontario, Canada To the Editor: DiLeo et al1 report the results of a well-designed phase III trial comparing lapatinib plus paclitaxel with placebo plus paclitaxel as first-line therapy for women with metastatic breast cancer with human epidermal growth factor receptor 2 (HER-2) expression either negative or untested. The study addressed the important question as to whether lapatinib, a dual kinase inhibitor, might have beneficial effects in this population. The results of the study, as stated appropriately and succinctly in the first concluding sentence of the abstract, showed that such patients did not benefit from the addition of lapatinib to paclitaxel. The authors also undertook a preplanned subgroup analysis of outcome in the patients who were found to be HER-2 positive on retrospective central testing. This analysis suggested that there might be benefit of lapatinib in this subgroup. The final paragraph of their article stated: "These retrospective results in a limited number of patients should be considered as hypothesis generating."1 Unfortunately the authors do not heed that caution and instead conclude in both the abstract and in the main body of the article that therapy with lapatinib plus paclitaxel significantly improved clinical outcomes in a subgroup of HER-2–positive patients. We caution readers against this interpretation of the data and suggest that it reflects over-enthusiasm in the reporting of a company-sponsored trial. In large randomized controlled trials, it is appropriate to conduct a limited number of preplanned subgroup analyses to generate hypotheses that might be evaluated in further studies. Such subgroup analyses do not compare randomized groups of patients, and it is not appropriate to make robust conclusions from them. Despite this, the second concluding sentence of the abstract of the article by DiLeo et al1 states that "first-line therapy with paclitaxel-lapatinib significantly improved clinical outcomes in HER-2–positive patients," while the first sentence of the discussion declares that this "study demonstrated that the primary activity of lapatinib in breast cancer patients is mediated through HER-2 inhibition." These statements are misleading, and fail to acknowledge the limitations of a posthoc subgroup analysis. Another concern with the report of DiLeo et al1 is failure to emphasize the increased mortality and morbidity in the lapatinib-paclitaxel arm. The results section of the article indicates that mortality was four times higher in the lapatinib than the placebo arm (eight v two deaths), while serious adverse events were reported in 102 and 63 patients, respectively. In contrast to the authors' enthusiasm for including a statistical analysis for (nonrandomly assigned) patients in the HER-2–positive subgroup, an appropriate statistical analysis of these toxicity data for the entire population of randomized patients is missing. Our own analysis shows that morbidity (as represented by serious adverse events) was significantly greater in the lapatinib-paclitaxel arm (Mann-Whitney U, P < .0001) while differences in mortality approached significance (Mann-Whitney U, P = .058). While the low rate of cardiac events was described in the abstract, this crucial information regarding the overall increase in morbidity and mortality was not stated explicitly. Abstracts are the most widely read summaries of research findings and are an important source of information for healthcare professional and policy makers. A recent review of abstracts of than 350 randomized controlled trials published in high-impact medical journals demonstrated that they frequently underreport harm, even when harm is reported in the main body of the article2; unfortunately the article by DiLeo et al conforms to this trend. In industry-sponsored research, benefit from therapy has a propensity to be reported more often than harm.3,4 The association between authors' opinions about the safety of developmental drugs and their financial relationships with pharmaceutical manufacturers has been demonstrated.5 According to the disclosures, six authors of the article by DiLeo et al1 are employees of the pharmaceutical company that markets lapatinib, and three authors have been financially supported by it. We recommend that widely-read high-impact journals such as Journal of Clinical Oncology apply greater scrutiny to the reporting of research results, especially of those published with significant input from industry. The study reported by DiLeo et al1 answers an important question about the possible role of lapatinib for women with breast cancer whose tumors are HER-2 negative or untested. The appropriate conclusions to this study are that when administered with paclitaxel, lapatinib did not have a significant effect on time to progression, but was associated with a significantly increased risk of morbidity and a trend toward increased mortality. AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST The author(s) indicated no potential conflicts of interest. REFERENCES
1. DiLeo A, Gomez HL, Aziz Z, et al: Phase III, double-blind, randomized study comparing lapatinib plus paclitaxel with placebo plus paclitaxel as first-line treatment for metastatic breast cancer. J Clin Oncol 26:5544–5552, 2008. 2. Bernal-Delgado E, Fisher ES: Abstracts in high profile journals often fail to report harm. BMC Med Res Methodol 8:14; 2008.[CrossRef][Medline] 3. Beckelman JE, Li Y, Gross CP: Scope and impact of financial conflicts of interest in biomedical research. JAMA 289:454–465, 2003. 4. Friedman LS, Richter ED: Relationship between conflicts of interest and research results. J Gen Intern Med 19:51–56, 2004.[CrossRef][Medline] 5. Stelfox HT, Chua G, O'Rourke K, et al: Conflict of interest in the debate over calcium-channel antagonists. N Engl J Med 338:101–106, 1998.
Related Articles
Related Reply
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2009 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
