|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 22, No 4 (February 15), 2004: pp. 754-756 © 2004 American Society of Clinical Oncology. DOI: 10.1200/JCO.2004.99.305
In Reply:
Department of Palliative Care and Rehabilitation Medicine, The University of Texas M.D. Anderson Cancer Center, Houston, TX Erwin et al raised several issues—from the perspective of psychiatry—related to our article [1] in the May 15th issue of the Journal of Clinical Oncology. The comments focused on the methodologic limitations of our study and our interpretation of the findings and their implications. At the root of these issues are differences in perspective in both conceptual and practical terms. First, Erwin and colleagues chose to hold this trial to the standard of research on drug trials for major depression. As such, they were critical that the instrument used for screening was not validated and that a measure of depression was not the primary end point. It is crucial to understand that this study was directed at identifying patients who were receiving oncology care in an outpatient setting and expressed at least mild impairment in their quality of life as marked by depressed mood or anhedonia. We could identify no validated and appropriate screening tool for this specific purpose; therefore, we chose a two-question screening instrument because of its face validity and ease of use. We purposefully screened out people who had been diagnosed with major depression or who expressed suicidal ideation. With respect to our choice of the brief version of the Zung Self-Rating Depression Scale as a measure of depressive symptoms, the literature shows nearly a dozen brief instruments concerning primary care patients who are evaluated for major depression, and no significant differences between instruments have been identified [2]. The references related to our choice of instruments were outlined in our article. None of the available instruments have been formally validated in patients with advanced cancer, and our choice was reasonable and appropriate. The goal in caring for these patients was to improve their quality of life, and thus the primary end point was assessed using a validated instrument that measures overall quality of life in cancer patients. It is conceivable that treatment with an antidepressant could reduce depressive symptoms but result in a lower overall quality of life. As such, the study was designed to determine whether the modulation of serotonin using fluoxetine would produce enough favorable effects to outweigh possible adverse effects or unfavorable drug interactions to produce a net benefit in the patients' overall quality of life. The authors also criticized our failure to conform to CONSORT (Consolidated Standards for Reporting Trials) standards [3] and cited a possible selection bias in our study. We agree that further research will be needed to determine how well these data generalize to other settings. This issue was acknowledged and addressed in our article. In particular, we were not able to provide the top of a typical CONSORT study flow diagram that showed how many patients were assessed for eligibility. As readers of the Journal of Clinical Oncology are well aware, this is a difficult feature of the CONSORT criteria to produce in cooperative group cancer research. This journal would be far thinner if manuscripts were excluded for their failure to collect information on all patients who might have been eligible for a cancer study but were not enrolled. Indeed, fewer than 1% of studies of serotonin reuptake inhibitors as first-line treatment for depression were believed to be devoid of significant limitations in reporting when strict standards were applied [4]. In this regard, the nature of the patients, the health-care setting, and the extent of funding for the research can influence how close to perfection a given trial can be. Erwin et al suggested that proper management of major depression was not provided. The serotonin pathway is implicated in mood regulation in general and affects significant depression and anxiety disorders [5]. Consistent with this, depressed mood and/or anhedonia were conceptualized as symptoms that might be part of a complex of symptoms for which fluoxetine, as an intervention that targets the serotonin pathway, might be helpful. The standards regarding patient education, dose adjustment, and follow-up for use of fluoxetine for this purpose in this setting have not been established. Our study involved a detailed explanation about the study drug (fluoxetine or placebo), a fixed dose of the drug throughout the study, and follow-up every 3 to 6 weeks (depending on the existing cadence of outpatient visits). Erwin et al asserted that our method of monitoring the intervention was limited to assessment of vomiting episodes. In complex patients, it is often difficult to attribute symptoms to a particular drug (versus the cancer itself) or to comorbidities. Our reporting regarding nausea outcomes was meant to summarize one of the more common side effects of fluoxetine, and it did not imply that toxicity assessment was limited to this one parameter. Again using the standard of research on drug trials for major depression, these authors were critical that our protocol stopped short of establishing whether each patient did or did not meet the diagnostic criteria for major depression. Our assessment of depressive symptoms, including a direct inquiry about suicidal ideation at every visit, far exceeded the usual care for this patient population. The study also did not attempt to formally rule out major depression by applying a diagnostic interview using specially trained personnel. Doing so has proven to be exceedingly difficult to accomplish on a timely basis in a typical outpatient oncology setting, and it does not represent the current standard of care for outpatients who express some degree of depressed mood or anhedonia. Our study protocol was peer-reviewed on clinical and scientific grounds, and was reviewed by multiple institutional review boards familiar with the standards of care in outpatient oncology. In this regard, cancer patients face barriers to ideal mental health care exceeding those that already exist in primary care medicine. We are perplexed by these authors' suggestion that our study itself "raises yet another barrier" to cancer patients receiving appropriate care. We designed a practical clinical trial exploring a novel approach to managing cancer patients in the real world. Most patients like those enrolled on our study are managed within the skill set of their medical oncologists, who do not typically have ready access to more elaborate mental health screening and treatment techniques. We did not claim that the findings of our study provided a basis for a new standard of care. Although we used a placebo-controlled design typical of a phase III study, our study had implications more consistent with those of a typical phase II study. That is, we provided a better understanding of the efficacy and feasibility of one easily reproducible approach to identifying and treating depression in patients with advanced cancer. We documented the significant, albeit modest, value of a widely used serotonin reuptake inhibitor on patients' quality of life. This finding was consistent with the recognized role of serotonin in general mood regulation. Considerable variations in practice patterns exist in the management of symptomatic patients with cancer. There is an emerging field of symptom science arising from patient need and the efforts of researchers from multiple disciplines, including (but not limited to) academic general medicine, psychiatry, psychology, palliative care, and medical oncology. Dr Gary Morrow, speaking at the 20th anniversary celebration of the National Cancer Institute's Community Clinical Oncology Program, described cancer control research as "evolutionary, not revolutionary." Taking this cue, our efforts should not be viewed as a revolution against the conventional paradigms in psychiatry. Rather, this research applies a symptom management approach to a common clinical scenario in outpatient oncology. As usual, further research will be needed to refine and evolve our assessment and management of these patients such that outcomes improve and the variation in practice patterns is reduced. Authors' Disclosures of Potential Conflicts of Interest The authors indicated no potential conflicts of interest. REFERENCES
1. Fisch MJ, Loehrer PJ, Kristeller J, et al: Fluoxetine versus placebo in advanced cancer outpatients: A double-blinded trial of the Hoosier Oncology Group. J Clin Oncol 21:1937-1943, 2003
2. Williams JW Jr, Noel PH, Cordes JA, et al: Is this patient clinically depressed? JAMA 287:1160-1170, 2002
3. Altman DG, Schultz KF, Moher D, et al: The revised CONSORT statement for reporting randomized trials: Explanation and elaboration. Ann Intern Med 134:663-694, 2001
4. Moher D, Schultz KF, Altman DG: The CONSORT statement: Revised recommendations for improving the quality of reports of parallel-group randomized trials. JAMA 285:1987-1991, 2001 5. Stahl SM: Essential psychopharmacology: Neuroscientific basis and practical application. Cambridge, Cambridge University Press, 2000, pp 236-239
Related Article
Related Correspondence
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2004 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
