Neurology -- Correspondence for Wolkowitz et al., 60 (7) 1071-1076

Correspondence published:

DHEA treatment of Alzheimer’s disease: A randomized, double-blind, placebo-controlled study: Friedrich Leblhuber, Herbert Haller, Kostja Steiner, Dietmar Fuchs (20 August 2003)

Reply to Leblhuber: Owen M. Wolkowitz, Joel H. Kramer, Psy.D.,Victor I. Reus, MD,Kristine Yaffe, MD (20 August 2003)

DHEA treatment of Alzheimer’s disease: A randomized, double-blind, placebo-controlled study 20 August 2003

Friedrich Leblhuber,
MD
Wagner Jaureggweg 15, 4020 Linz, Austria,
Herbert Haller, Kostja Steiner, Dietmar Fuchs
Send Correspondence to journal:
Re: DHEA treatment of Alzheimer’s disease: A randomized, double-blind, placebo-controlled study

friedrich.leblhuber{at}gespag.at Friedrich Leblhuber, et al.

We read with interest the paper by Wolkowitz et al [1], who compared the efficacy and tolerability of DHEA versus placebo in Alzheimer’s disease (AD). They could not find significant improvement of cognitive performance in their DHEA treated group of patients. Cortisol plasma measurements were not performed in this study. Unfortunately, the loss of 43% of the participants by the end of the study makes any interpretation of their data difficult. The questionable clinical evidence of a trial designed like this has already been discussed. [2] DHEA was seen to have antiglucocorticoid effects and higher cortisol concentrations in patients with AD. [3] In our series [3], a decreased DHEA per cortisol ratio was found in AD patients (aged 75,4 +/- 4,6 years) compared to age matched controls, with the female cohort significantly contributing to these results. There was no difference in DHEA serum concentrations between patients and age-matched controls. Similar results were seen in Huntington’s chorea in a small number of male patients (39,4 +/- 13,8 years) [4], indicating that sex and age differences have to be considered especially in hormone studies like these and that stress and glucocorticoids may play an important role in memory and cognitive performance in neuropsychiatric conditions like other forms of dementia. These may include depression, posttraumatic stress disorder as well as Cushing’s disease.
In 1998, Lupien et al. demonstrated that aged humans with prolonged cortisol elevations showed reduced hippocampal volume and deficits in hippocampus-dependent memory tasks compared to controls with normal cortisol concentrations and that the degree of hippocampal atrophy correlated strongly with the degree of cortisol elevation over time and current basal cortisol concentrations. [5] The latter authors concluded that basal cortisol elevation leads to hippocampal damage. [5] Consequently, DHEA medication could protect hippocampus against elevated basal cortisol in patients with AD mediated by its antiglucocorticoid mechanisms, but this has to be proven in future studies in which these hormones are measured consecutively before and during a placebo controlled trial with a sufficiently powered number of male and female patients. In addition, these hormones could be combined with other agents and antidementia medications like anticholinesterase inhibitors. [2]
References
1. Wolkowitz OM, Kramer JH, Reus VI, et al. DHEA treatment of Alzheimer’s disease: a randomized, double-blind, placebo-controlled trial. Neurology 2003; 6: 1071-1076.
2. Knopman D. and Henderson VW. DHEA for Alzheimer’s disease. A modest showing by a superhormone. Neurology 2003; 6: 1060-1061.
3. Leblhuber F, Neubauer C, Peichl M, et al. Age and sex differences of dehydroepiandrosterone sulphate (DHEAS) and cortisol (CRT) plasma levels in normal controls and Alzheimer’s disease (AD). Psychopharmacology 1993; 111: 23-26.
4. Leblhuber F, Peichl M, Neubauer C, et al. Serum dehydroepiandrosterone and cortisol measurements in Huntington’s chorea. J Neurol Sci 1995; 132:76-79.
5. Lupien SJ, de Leon M, de Santi S, et al. Cortisol levels during human aging predict hippocampal atrophy and memory deficits. Nat Neurosci 1998; 1 :69-73.

Reply to Leblhuber 20 August 2003

Owen M. Wolkowitz,
Dept. Psychiatry UCSF School Med
401 Parnassus Ave., Box F-0984, San Francisco, CA 94143,
Joel H. Kramer, Psy.D.,Victor I. Reus, MD,Kristine Yaffe, MD
Send Correspondence to journal:
Re: Reply to Leblhuber

owenw{at}itsa.ucsf.edu Owen M. Wolkowitz, et al.

We are grateful for Dr. Leblhuber et al's comments on our paper [1] and for the opportunity to present additional data, which space did not permit in the original report. Dr. Leblhuber et al reviewed their own data showing a trend towards low serum DHEA-sulfate (DHEAS) /cortisol ratios in AD patients. Similar findings have been occasionally, but not universally reported, and decreases in this ratio are not specific for AD [2, 3]. According to Dr. Leblhuber and colleagues’ hypothesis, DHEA treatment should be especially helpful in patients with hypercortisolemia or with low DHEA/cortisol or DHEAS/cortisol ratios. We expected this as well. However, our data indicated that baseline DHEA and DHEAS levels and the DHEAS/cortisol ratio did not predict changes in ADAS-Cog ratings (p > 0.20 in all cases). Baseline cortisol levels and the baseline DHEA/cortisol ratio were significantly correlated with changes in ADAS-Cog ratings (r= 0.61, p= 0.008; and r= -0.49, p< 0.05, respectively), but in the direction seemingly counter to Dr. Leblhuber and colleagues’ hypothesis. Specifically, subjects with relatively higher baseline cortisol levels and lower DHEA/cortisol ratios responded less well to DHEA treatment. Further, treatment-associated changes in DHEA, DHEAS and cortisol levels and in DHEA/cortisol and DHEAS/cortisol ratios were not significantly correlated with changes in ADAS-COG ratings.
Dr. Leblhuber et al also commented on our study’s high drop- out rate. The impact of this on the interpretation of our data is somewhat mitigated by similar findings in our Last Observation Carried Forward and Completers analyses. Our modest sample size and relatively high drop-out rate speak to the difficulty recruiting and retaining AD subjects in trials of solitary experimental agents when effective and safe alternatives (e.g., donepezil) are available. From methodological as well as scientific standpoints, future DHEA trials in AD should test it as an augmentation of standard therapies. Despite our high drop-out rate, our pilot study succeeded in two of its major goals: determining effect sizes of DHEA’s efficacy in AD and determining DHEA’s tolerability in this population.
We agree with Dr. Leblhuber et al that there are ample preclinical data (e.g., antiglucocorticoid effects) to justify enthusiasm regarding DHEA as a novel treatment in AD. However, the evolving clinical literature on DHEA treatment is replete with examples of preclinical data not translating into clinical reality (3-5). Whether DHEA proves to have efficacy as an adjunctive treatment in AD must await future large scale clinical trials.
References
1. Wolkowitz OM, Kramer JH, Reus VI, Costa MM, Yaffe K, Walton P, et al. DHEA treatment of Alzheimer’s disease: A randomized, double-blind, placebo-controlled study. Neurology 2003;60:1071-1076.
2. Wolkowitz OM, Reus VI, Roberts E. Role of DHEA and DHEA-S in Alzheimer's disease: Reply. Am J Psychiatry 1993;150(9):1433.
3. Wolkowitz OM, Reus VI. Neuropsychiatric Effects of Dehydroepiandrosterone (DHEA). In: Kalimi M, Regelson W, editors. Dehydroepiandrosterone (DHEA): Biochemical, Physiological and Clinical Aspects. Berlin: Walter De Gruyter; 2000. p. 271-298.
4. Huppert FA, Van Niekerk JK. Dehydroepiandrosterone (DHEA) supplementation for cognitive function. Cochrane Database Syst Rev 2001;2:CD000304.
5. Wolf OT, Kirschbaum C. Wishing a dream came true: DHEA as a rejuvenating treatment? J Endocrinol Invest 1998;21(2):133-135.