|
Introduction: Activation of the hypothalamic-pituitary-adrenal (HPA) axis by cytokine therapies such as interleukin-2 (IL-2) appears to be related to the development of depression in cancer patients. We sought to characterize the effects of IL-2 treatment on the HPA axis and the relationship between psychiatric symptoms and treatment adherence in patients with malignant melanoma.
Methods: Patients with stage three or four malignant melanoma were recruited for an IL-2 treatment parent study. Prior to, during, and after treatment we examined HPA axis activation by measuring plasma concentrations of adrenocorticotropic hormone (ACTH) and cortisol (CORT), depressive symptoms and cognitive function, and IL-2 treatment adherence. The Friedman test, Wilcoxon signed-rank test, Spearman’s correlation, and multiple regression were used in the analysis.
Results: Neither ACTH nor cortisol plasma concentrations post IL-2 administration significantly increased from baseline during Cycle 1. During Cycles 2 and 3 these levels peaked significantly following the initial dose of IL-2 but returned to baseline following the end of each cycle. A positive correlation between baseline ACTH levels and the first significant elevation in ACTH became insignificant when controlling for number of IL-2 doses received. Neither the severity of depressive symptoms nor number of IL-2 doses significantly predicted the initial increase in ACTH or cortisol.
Conclusions: Activation of the HPA axis by IL-2 treatment initially occurs with a second treatment cycle and is greatest after the initial dose of each cycle. ACTH production remains activated between the second and third cycles. It seems the number of IL-2 doses received per treatment cycle affects HPA axis activation; patient recruitment continues to improve the statistical power of analysis.
One of the most reproducible findings in biological psychiatry is the hyperactivity of the HPA axis in patients with major depression, evidenced in part by elevated concentrations of cortisol in blood.1
IL-2 administration is associated with HPA axis activation as manifested by increased ACTH and CORT levels and has been found to lead to persistent elevations in cortisol.2, 3
IL-2 activation of HPA axis pathways may play an important role in the pathophysiology of IL-2 induced behavioral changes which often cause patients to terminate treatment.
Patients were recruited and screened for a parent study examining the ability of escitalopram to ameliorate the adverse neuropsychiatric effects of IL-2.
Eligible subjects (n=14) were studied before and during a 12-week treatment with intravenous IL-2 [720,000 units/kg every 8 hours for 5 days (1 cycle), every 3 weeks for 4 cycles]. Blood draws were performed at regular intervals just prior to and during each IL-2 cycle.
Plasma ACTH and cortisol levels were measured via radioimmunoassay according to the manufacturer’s protocol (Nichols Institute and ICN Biomedicals, respectively). IL-2 treatment adherence was evaluated according to the patient’s medical records.
PSYCHIATRIC ASSESSMENTS
The HAM-D is a 24-question, clinician-administered interview that was used to determine the presence of depressive symptoms.4
The Zung Self-Rating Scale is a 20-item patient-reported scale that was used to quantify the symptoms of depression.5
BIOSTATISTICAL METHODS
Changes in plasma ACTH and cortisol levels were analyzed using descriptive statistics. Differences between time points were assessed using the Friedman test and the Wilcoxon signed-rank test, both non-parametric statistical methods.
Relationships between HPA axis activation, assessments of neuropsychiatric functioning, and number of IL-2 doses tolerated were determined via Spearman’s correlation and multiple regression analyses.
Spearman’s correlation showed ACTH levels at screening and at Cycle 2 Day 1 9:00 to be positively correlated, r(9)=.783, p<0.05. When controlling for number of doses, this correlation is not significant, r(9)=.170, p=0.687.
Initial HPA activation levels measured by ACTH at Cycle 2 Day 1 9:00 were not significantly predicted by HAM-D scores, β =.208, t(7)=.565, p=0.590, or number of IL-2 doses received to that point β =.102, t(7)=.276, p=0.791.
Spearman’s correlation showed Zung score and CORT levels at Cycle 4
Day 1 9:00 to be positively correlated, r(6)=0.899, p<0.05.
Initial HPA activation levels measured by CORT at Cycle 2 Day 1 9:00 were
not significantly predicted by number of IL-2 doses received to that point
β=0.368, t(7)=1.081, p=0.315 or by HAM-D scores, β=0.218, t(7)=0.641,
p=0.542.
Initial activation of the HPA axis (ACTH and CORT) by IL-2 administration occurs upon re-exposure to the cytokine during Cycle 2 of treatment while the greatest activation occurs during Cycle 3.
ACTH production is chronically elevated by IL-2 treatment continued through Cycle 2.
The number of IL-2 doses received appears to affect the level of HPA axis activation and should be further examined with a greater number of patients.
An effective examination of the relationship between neuropsychiatric symptoms and level of HPA activation requires a larger study population
Continued recruitment of patients will improve the statistical power of analysis, particularly with regard to multiple regression.
Completion of parent study will allow for analysis of depressed vs. non-depressed patients as well as those receiving escitalopram vs. placebo.
Future studies should examine the clinical relevance of HPA activation due to IL-2 treatment in patients with malignant melanoma.
Research sponsored by: Howard Hughes Medical Institute Grant #52005873,
National Institutes of Mental Health Grant R01MH71580 to Dominique L. Musselman,
and an Atlanta Clinical and Translational Science Institute award.
1. Musselman DL, et al. Biology of Mood Disorders, in Textbook of Psychopharmacology, 2nd Edition.
Washington, D.C., American Psychiatric Press, 1998.
2. Raab C, et al. Clinical Endocrinology (Oxf) 1999; 50:37-44.
3. Hanisch UK, et al. Endocrinology 1994; 135:2465-2472
4. Miller IW, et al. Psychiatric Research 14:131-142, 1984.
5. Zung W. Arch Gen Psychiatry, 1965; 12(1):63-70
|
