Antiviral treatment perspective against Borna disease virus 1 infection in major depression: a double-blind placebo-controlled randomized clinical trial

Abstract

Background: Whether Borna disease virus (BDV-1) is a human pathogen remained controversial until recent encephalitis cases showed BDV-1 infection could even be deadly. This called to mind previous evidence for an infectious contribution of BDV-1 to mental disorders. Pilot open trials suggested that BDV-1 infected depressed patients benefitted from antiviral therapy with a licensed drug (amantadine) which also tested sensitive in vitro. Here, we designed a double-blind placebo-controlled randomized clinical trial (RCT) which cross-linked depression and BDV-1 infection, addressing both the antidepressant and antiviral efficacy of amantadine.

Methods: The interventional phase II RCT (two 7-weeks-treatment periods and a 12-months follow-up) at the Hannover Medical School (MHH), Germany, assigned currently depressed BDV-1 infected patients with either major depression (MD; N = 23) or bipolar disorder (BD; N = 13) to amantadine sulphate (PK-Merz®; twice 100 mg orally daily) or placebo treatment, and contrariwise, respectively. Clinical changes were assessed every 2-3 weeks by the 21-item Hamilton rating scale for depression (HAMD) (total, single, and combined scores). BDV-1 activity was determined accordingly in blood plasma by enzyme immune assays for antigens (PAG), antibodies (AB) and circulating immune complexes (CIC).

Results: Primary outcomes (≥25% HAMD reduction, week 7) were 81.3% amantadine vs. 35.3% placebo responder (p = 0.003), a large clinical effect size (ES; Cohen's d) of 1.046, and excellent drug tolerance. Amantadine was safe reducing suicidal behaviour in the first 2 weeks. Pre-treatment maximum infection levels were predictive of clinical improvement (AB, p = 0.001; PAG, p = 0.026; HAMD week 7). Respective PAG and CIC levels correlated with AB reduction (p = 0,001 and p = 0.034, respectively). Follow-up benefits (12 months) correlated with dropped cumulative infection measures over time (p < 0.001). In vitro, amantadine concentrations as low as 2.4-10 ng/mL (50% infection-inhibitory dose) prevented infection with human BDV Hu-H1, while closely related memantine failed up to 100,000-fold higher concentration (200 μg/mL).

Conclusions: Our findings indicate profound antidepressant efficacy of safe oral amantadine treatment, paralleling antiviral effects at various infection levels. This not only supports the paradigm of a link of BDV-1 infection and depression. It provides a novel possibly practice-changing low cost mental health care perspective for depressed BDV-1-infected patients addressing global needs.

Trial registration: The trial was retrospectively registered in the German Clinical Trials Registry on 04th of March 2015. The trial ID is DRKS00007649; https://www.drks.de/drks_web/setLocale_EN.do.

Keywords: Borna disease virus 1 (BDV-1); Major depression; amantadine; antiviral treatment; bipolar disorder; double-blind placebo-controlled randomized clinical trial (RCT).

Fig. 1

Study design. The graphic illustrates the timeline of the two treatment periods and the follow-up, as well as the cross-over design of intervention by either amantadine or placebo and vice versa of the study, and indicates the number of patients who finished each period (N)

Fig. 2

CONSORT Flow Diagram. The CONSORT flow chart illustrates the enrolment of the patients, their allocation to intervention groups in treatment period I, period II, and optional treatment in the follow-up period, as well as the finally analysed patients

Fig. 3

Course of treatment 1 (HAMD-21 and suicidal behaviour, item 3). At the top: treatment effect of amantadine compared to placebo measured by the difference in total HAMD-score, week 7 marked by an arrow. At the bottom: treatment effect by the difference in HAMD item 3 “suicidal behaviour”. For p-values and effect size indicators at different time points, see Table 3 and Table 4

Fig. 4

Course of treatment 2 (HAMD items 1, 2, 8, and 10). Treatment effects of amantadine and placebo, respectively, measured by the difference in indicated items 1 (depressed mood), 2 (feeling of guilt), 8 (retardation), and 10 (anxiety) of the HAMD-score. For p-values and effect size indicators at different time points see Table 3 and Table 4

Fig. 5

Course of treatment 3 (BfS and HAMD items 7, 18, 14). Treatment effects of amantadine and placebo, respectively, measured by the difference in indicated items 7 (work and activities), 14 (sexual function), and 18 (day-variations) of the HAMD-score, as well as by the self-rated depression-related well-being score (BfS). For p-values and effect size indicators at different time points see Table 3 and Table 4

Fig. 6

Course of treatment 4 (combined HAMD cluster). Treatment effects of amantadine and placebo, respectively, measured by the difference in combined items of the HAMD-score, clustering for either melancholic (DENDOG) or retardation (DRETARD) features. For p-values and effect size indicators at different time points see Table 3 and Table 4

Fig. 7

a-d Pre-treatment infection variables and primary clinical outcome. Pearson’s correlation coefficient of treatment outcome of amantadine and placebo on week 7, as based on differences of the Hamilton score of depression (rDHAMD7) and the self-rated well-being scale (DBfS7) after period I, with pre-treatment values of BDV-1 antibodies (maxab) and antigen (maxpag, meanpag), as indicated in parts (a-d)

Fig. 8

In vitro prevention of infection comparing amantadine and memantine. Prevention-of-infection test by indicated doses of amantadine and memantine in vitro, using young rabbit spleen cells (YRS p108) and human BDV strain Hu-H1 (p55); p = passage