|
|
||
 |
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 22, No 1 (January 1), 2004: pp. 206-208 © 2004 American Society of Clinical Oncology. DOI: 10.1200/JCO.2004.99.248
In Reply:Department of Palliative Care and Rehabilitation Medicine, University of Texas M.D. Anderson Cancer Center, Houston, TX
Department of Psychology, Indiana State University, Terre Haute, IN
Department of Palliative Care, University of Kentucky, Lexington, KY
Division of Hematology/Oncology, Indiana University, Indianapolis, IN We appreciate the thoughtful comments sent by Drs Alliot and Coyne et al related to our article in the May 15 issue of the Journal of Clinical Oncology [1]. Dr Coyne et al are particularly concerned that this study did not meet established guidelines for the diagnosis and treatment of depression in medical settings. We feel that Dr Coyne et al have mistakenly evaluated this study as an explanatory clinical trial targeting depression in cancer, whereas the study was intended as a practical clinical trial directed at improving the quality of life of outpatients with advanced cancer. Practical clinical trials are trials that address pragmatic questions about the risks, benefits, and costs of an intervention as they would occur in practice [2]. These trials include a diverse population of study participants, and the participants are recruited from heterogenous practice settings [3]. Such trials are scarce in outpatient oncology, and there is tremendous variability in practice patterns as a result. This is certainly the case related to the prescribing of antidepressants, where variability in practice patterns is quite obvious [4]. As Dr Coyne et al point out, there are formidable barriers to mental health assessment and management for patients with advanced cancer. The actual diagnosis of depressive disorders is notoriously difficult in this setting, reimbursement for specialized mental health care is often lacking, and other problems including patient and family attitudes about mental health care and scheduling issues also limit patients from receiving what a mental health professional might consider ideal assessment, treatment, and follow-up. Given the reality that many patients are essentially trapped within the skill set of oncologists practicing in busy outpatient settings with limited resources, we identified patients as at higher risk for quality of life impairment by virtue of their self-report of some degree of depressed mood and/or anhedonia. Also informing our choice of design and patient inclusion criteria is the theoretical conceptualization of fluoxetine as an intervention that targets the serotonin pathway, because this pathway is implicated in mood regulation in general, as well as influencing significant depression and anxiety disorders [5]. Therefore, we wished to include as broad a range of patients as possible who might potentially benefit from the therapy, excluding those reporting virtually no distress but including all others. We believe that this is one of the strengths of the study. By including a broad range of patients, we provided provocative evidence for the potential benefit of a serotonin-reuptake inhibitor on quality of life and mild depression. In addition, our subset analysis suggested that the greatest benefit from fluoxetine appeared to be in those patients with a score of 4 or higher on the two-question screening survey instrument. This two-question screening approach has been validated, albeit not in a pure oncology population [6]. We believe that interest in exploring practical techniques to alleviate suffering in cancer patients does not equate with dismissing the need to prevent cancer, cure existing cancer, or advocate for more resources that might enable more impeccable treatment strategies. We often use "we wish it were different" phrases [7] with our patients when acknowledging the limitations of the status quo. Thereafter, we take aim at the realistic ideal given the current circumstances. Dr Alliot and Dr Coyne et al commented on the generalizability of our findings. As noted in a recent editorial about the globalization of oncology, standards of care do differ, and patterns of care can affect the interpretation of studies [8]. The globalization of oncology can be understood not only in geographic terms, but also in terms of the growing trend toward interdisciplinary care of this patient population. Our data should not be assumed to generalize well to international settings, outside of the discipline of medical oncology, or to outpatient oncology settings rich in resources related to mental health care. Dr Alliot pointed out that our study population does not entirely reflect the global epidemiology of cancer. Although this is true, no study could uphold such a standard. Our study population does expand to men and women and includes a broad mix of common solid tumors. As we acknowledged in the discussion of the manuscript, possible selection bias is an important limitation of our study. Clearly, these data cannot be applied to those patients who are reluctant to take antidepressant medications. However, even if it were somehow possible to include such patients in a practical clinical trial, it would only have diminished the generalizability of the results. Additional research is needed to determine whether our findings will generalize to other similar academic and community settings. Dr Alliot questioned the success of the study blinding procedure because of the expected side effects of fluoxetine. We doubt that patients or clinicians could easily discern the treatment assignment on the basis of nausea and vomiting. Chronic nausea is a common symptom, and advanced cancer patients often have multiple risk factors for nausea. Dizziness and diarrhea may occur more often in patients treated with fluoxetine, but the rate of those side effects is low. Concern was also expressed about the imbalance of the treatment groups with regard to a family history of depression and the presence or absence of psychological counseling. As noted above, the inception cohort for this study was not patients diagnosed with major depression. Because the primary end point was overall quality of life, the patients were stratified by performance status. Family history of depression and presence of psychological counseling are not established to have strong influence in overall quality of life. Although there was some imbalance in those factors on the basis of the randomization, the impact of that minor imbalance is difficult to ascertain. The appropriateness of treating the study population with fluoxetine specifically was questioned. It also was suggested that there should be no further use of placebo controls for this type of research. We agree that clinical practice should involve tailored prescribing of antidepressants when the decision to prescribe an antidepressant has been made. Most antidepressants are similar in efficacy for their known indications, and the major differences are related to their side effects and pharmacologic differences. Although fluoxetine need not be assumed to be the best or most appropriate treatment for all cancer patients with depressed mood and/or anhedonia, as a practical matter, it is difficult to incorporate tailored prescribing in the design of this kind of trial. Finally, we do not agree that comparison of antidepressant therapy to placebo is no longer justified for the purpose of improving the quality of life of patients with advanced cancer. We again emphasize that this was not a study of major depression in cancer patients. Our study had a phase III design, but it was intended as a practical clinical trial providing phase II information about the feasibility and efficacy of a particular strategy of identifying and treating advanced cancer patients with depressed mood and/or anhedonia. Confirmatory trials are needed, and the placebo-control design remains reasonable and appropriate for patients with quality of life impairment associated with at least minor depressive symptoms. Authors' Disclosures of Potential Conflicts of Interest The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Received more than $2,000 a year from a company for either of the last 2 years: Lawrence H. Einhorn, Eli Lilly; Steven Passik, Eli Lilly. REFERENCES
Related Article
Related Correspondence
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2004 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
