Laboratory Tests to Aid in Psychiatric Diagnosis: Title and subTitle BreakAre We Making Progress?

IN THE ARTICLE by Berman et al,¹ in this issue of the ARCHIVES, the authors report that administration of α-methylparatyrosine to subjects with a history of depressive episodes evoked a brief "relapse" of depressive symptoms. In related articles, the authors have reported that healthy subjects without a history of depressive episodes do not display this sensitivity to α-methylparatyrosine.² The authors' interpretation of these findings is that the behavioral sensitivity of individuals to α-methylparatyrosine may be a "phenotypic trait marker for depression," and useful for identifying individuals who are prone to develop depressive episodes even though they have no current symptoms. If true, such a marker might also be useful for identifying individuals with latent mood disorders during a preclinical phase of their illness. The development of a laboratory test to aid in the diagnosis of individuals with mood disorders would be an important advance in our capacity to treat depression; however, many issues need to be resolved to provide a stronger foundation for the authors' interpretation of their findings.

The authors found that α-methylparatyrosine administration to 14 previously depressed, but currently untreated, subjects over a 2-day period produced greater increases in Hamilton Depression Rating Scale (HDRS) scores than did diphenhydramine (ie, active placebo) administration to the same subjects during another period. Changes in the subjects' mood were measured by comparing HDRS ratings just before the initial administration of drug with 1 of 2 time points on the next day when the greatest severity of symptoms was observed. In the subjects who were administered α-methylparatyrosine and experienced depressive symptoms, HDRS scores reverted to nondepressed levels on the day after the final day of α-methylparatyrosine administration. Final HDRS ratings were not different between the group that had been administered α-methylparatyrosine and the group that had been administered diphenhydramine. On the basis of these observations, the authors concluded that α-methylparatyrosine administration produced a "clinically significant depressive reaction" in the patients. However, caution is warranted in concluding that the observed effect was clinically significant. By definition, brief increases in depressive symptoms do not constitute recurrence of a depressive episode. Clinical experience suggests that patients with remitted depression do experience spontaneous fluctuations in depressive symptoms, and we are not aware of research establishing the clinical significance of fluctuations similar to those observed by the authors. In further regard to this issue, the authors measured depressive symptoms at 6 time points, but did not present HDRS scores for all of these time points. Without being able to evaluate the overall variability in depression ratings during the protocol, it is more difficult to gauge the significance of the transient effects that were observed.

The authors also interpreted their findings as supportive of "catecholaminergic dysfunction in depression." While this may be true, the neurochemical specificity of α-methylparatyrosine in this study should be reconsidered. α-Methylparatyrosine has specific in vitro effects on tyrosine hydroxylase within catecholamine terminals. However, research has shown that the catecholamine neurotransmitters, norepinephrine and dopamine, have functional linkages with other neurotransmitter systems, such as serotonin. Concentrations of norepinephrine, serotonin, and dopamine metabolites in cerebrospinal fluid (CSF) are significantly intercorrelated, via mechanisms other than those related to their elimination from the CSF space.³ Also, while the direct effects of serotonin selective reuptake inhibitors are specific for the serotonin transporter, these drugs increase the CSF concentrations of norepinephrine metabolites, as well as serotonin metabolites, in CSF.⁴ Thus, α-methylparatyrosine might have had indirect effects on neurotransmitter systems other than the catecholamines in these subjects, in particular, serotonin. Likewise, the phenotypic marker potentially identified in these subjects may have been serotonergic and not catecholaminergic in nature. It would have been interesting to assess an index of serotonergic function in these subjects or to determine the sensitivity of these same subjects to indoleamine depletion.

The authors observed that baseline plasma cortisol levels predicted the appearance of depressive symptoms during α-methylparatyrosine administration, and they suggest that the subjects' increased sensitivity to α-methylparatyrosine may also involve an abnormality in the hypothalamic-pituitary-adrenal (HPA) axis. While the catecholamine hypothesis of depression was the stated rationale for this study, it is fair to say that HPA axis abnormalities in depression have been found more consistently than abnormalities of noradrenergic or dopaminergic function.⁵ There is also abundant evidence for important interactions between the HPA axis and the catecholamine neurotransmitter systems. Noradrenergic inputs are major regulators of hypothalamic secretion of corticotropin-releasing factor, and dopaminergic systems also exert some regulatory effect on corticotropin-releasing factor release.⁶ Corticotropin-releasing factor may function as an integrative mediator of behavioral, endocrine, autonomic, and hemodynamic components of the stress response, and has major effects on the activity of noradrenergic and dopaminergic systems in brain.⁷ Receptors for the glucocorticoid hormone, cortisol, are also widely distributed in the brain, and cortisol has been shown to exert effects on noradrenergic and serotonergic systems.⁸ Thus, the relationship between cortisol levels and catecholamine depletion observed in this study is consistent with the view that the pathophysiology of depression involves dysregulation of a neural network including the monoaminergic neurotransmitter systems and HPA-axis modulators. There are numerous possibilities as to the particular neurochemical marker or markers responsible for these subjects' sensitivity to α-methylparatyrosine that go far beyond the catecholaminergic synapse. Nevertheless, a phenotypic marker for mood disorders would have great clinical value as an aid to diagnosis, regardless of the nature of the neurochemical systems involved.

Finally, one must give greater consideration to the possibility that the authors have identified a long-lasting consequence of prior depressive episodes and not a trait marker for vulnerability to future depressive episodes in these subjects. The authors acknowledge this alternative explanation for their findings, but do not consider possible pathophysiological mechanisms. Sheline et al⁴ reported that hippocampal volume reductions occur in elderly depressed individuals as compared with nondepressed controls, and that the degree of volume loss was correlated with the cumulative length of prior depressive episodes.⁹ This hippocampal volume loss has been proposed to be both a consequence and a cause of hypercortisolemia in subjects with depression. Thus, prior depressive episodes may have long-lasting effects on the hippocampus and on the HPA axis, which could then bring about an abnormal sensitivity of monoaminergic neurotransmitter systems. It would be interesting to determine whether the length or severity of prior depressive episodes was correlated with the magnitude of the subject's sensitivity to α-methylparatyrosine in the study reported by Berman et al.¹

Despite limitations in the interpretability of the data presented in this study, Berman et al¹ have made an important advance in our search for tools to aid in the diagnosis of mood disorders. Psychiatry is unfortunately unique among medical specialties in that we have no useful laboratory tests that can aid the clinician in resolving the differential diagnosis when clinical symptoms are equivocal or confusing. Markers for the preclinical diagnosis of neuropsychiatric disorders are sorely needed. After all, no cardiologist would want to make therapeutic decisions about the treatment of new onset chest pain without first referring to an electrocardiogram. In the future, our patients deserve the availability of such tests so that they can better avoid the potentially disabling complications of depressive episodes.