The VAST-D study: Optimizing treatment for major depression

Less than one-third of patients with major depressive disorder (MDD) achieve remission with their first antidepressant treatment. The Veterans Affairs (VA) Augmentation and Switching Treatments for Improving Depression Outcomes (VAST-D) study randomized 1522 military veterans in the US with a diagnosis of MDD who were unresponsive to at least one antidepressant course (meeting minimal standards for treatment dose and duration) to one of three commonly used next-step treatment options to determine their relative effectiveness and safety. A small beneficial effect was demonstrated at 12 weeks for those augmenting their index antidepressant treatment with an atypical antipsychotic (a partial D2 agonist) medication.

The primary data were published in JAMA in 2017.1 An editorial in JAMA noted some important questions that remained, however, including whether MDD comorbid with post-traumatic stress disorder (PTSD) or anxiety may have potentially favored one of the treatment options of VAST-D.2 The VAST-D investigators presented further analyses, addressing the effects of PTSD comorbidity, anxiety, mixed features and medical comorbidity on treatment outcome coupled with longer-term results from the VAST-D study at APA 2018. 

VAST-D was a 35-site Veteran Affairs (VA) Cooperative Study funded by the VA Office of Research and Development and carried out between December 2012 and August 2015.

Augmentation resulted in a modest statistically significantly increased likelihood of remission compared with switch

1522 veterans, 85% men, with severe non-psychotic MDD who had not achieved optimal response after an adequate trial with their clinician’s choice of antidepressant monotherapy were randomized to one of three commonly used next-step treatment options:

  • switch to an aminoketone antidepressant (SWI)
  • combination of the index antidepressant with an aminoketone antidepressant (COM)
  • augmentation of the index antidepressant with an atypical antipsychotic medication (AUG)

The results showed that AUG resulted in a modest statistically significantly increased likelihood of remission during 12 weeks of treatment compared with SWI.1

Anxiety was more frequent in the SWI (24.3%) and COM (22.5%) groups, and akathisia (14.9%), somnolence (14.5%), and weight gain (5.7%) were more frequent in the AUG group.1

Remission does not necessarily occur within 12 weeks

Sidney Zisook, Professor of Psychiatry, University of California, San Diego, CA, emphasized that most antidepressant treatment trials only last 8 to 12 weeks, but depression is a long-term illness and the goal is to maintain relapse prevention.

Among those who had not remitted at 12 weeks, 60% remitted during the continuation phase

He highlighted the following important findings from analyses of the 725 patients who continued VAST-D beyond 12 weeks:

  • relapse rates do not depend on the treatment and was similar (38–40% at 6 months)
  • remission in the acute phase was associated with a lower risk of relapse in the continuation phase
  • among those who had not remitted at 12 weeks, 60% remitted during the continuation phase
  • similar improvements in symptoms of depression, anxiety, suicidal ideation and quality of life by 36 weeks were seen in all three treatment groups

PTSD comorbidity did not affect the results of VAST-D

PTSD is a common comorbidity in MDD, said Somaia Mohamed, Associate Director of the Northeast Program Evaluation Center and Associate Clinical Professor in the Yale Department of Psychiatry.

In VAST-D, AUG increased the likelihood of remission, but 47% of these patients also had PTSD. The question is whether this high rate of comorbid PTSD favored AUG over the other treatment options, noted Dr Mohamed.

In order to assess this question, Dr Mohamed carried out detailed analyses of the data to identify differences in the baseline characteristics and outcomes of patients with MDD and those with MDD and PTSD, and between the treatment options.

The results, that have been submitted for publication, lead to the conclusion that PTSD comorbidity did not affect the results of VAST-D and should not limit the generalizability of the findings of VAST-D.

Effect of anxiety, mixed features and medical comorbidity on treatment outcome

Higher anxiety was associated with lower rates of remission

Lori Lynne Davis, Clinical Professor of Psychiatry and Behavioral Neurobiology, University of Alabama, AL, presented her analyses of the VAST-D data to determine the effects of anxiety, mixed features and medical comorbidity on overall treatment outcome and/or response to specific interventions. They demonstrate that:

  • comorbid anxiety was prevalent, and higher anxiety was associated with lower rates of remission
  • the presence of mixed features at baseline was a potentially meaningful moderator of treatment
  • baseline severity of medical comorbidity was not associated with rate of remission except in the subgroup with severe medical comorbidity, for whom remission was more likely for those in the AUG group

Use of antidepressant response trajectories to predict remission

Paul B. Hicks, Professor of Psychiatry and Behavioral Science, Texas A&M College of Medicine, TX, explained how charting the 16-item Quick Inventory of Depressive Symptomatology-Clinician Rated (QIDS-C16) score against duration of treatment for each patient results in between six and nine response trajectories, which can be used to optimize treatment.

The augmentation group had higher rates of both response and remission

Professor Hicks’ analysis of the trajectories for different treatment groups showed:

  • AUG treatment was more likely to be associated with the remitter group (details below)
  • COM and SWI treatment was more likely to be associated with the non-responders outcome group

Remission rates at 12 weeks were 22.3% (SWI), 26.9% (COM) and 28.9% (AUG). The AUG group exceeded the SWI group in remission rates (relative risk 1.3; p=0.02), however, other remission comparisons were not significant. The AUG group had higher rates of response (74.3%) than either the SWI (62.4%) or COM (65.6%) treatment groups.

Treatment group allocation did not affect the trajectory group assignment.

Early improvement was more likely to be associated with remission, and Professor Hicks considered whether the data could be used to predict remission. He explained that the chance of remission for patients on the worst trajectory was less than 8%.

Professor Hicks showed a significant effect of treatment allocation on the response trajectory and that AUG patients were more likely to remit.

Wealth of VAST-D data enables person-centered base analysis

John Rush, Adjunct Professor of Psychiatry and Behavioral Sciences, Duke University School of Medicine, Durham, NC closed the session by highlighting the heterogeneity of response to antidepressant treatment and the importance of developing tools to predict outcome for patients.

He further highlighted the richness of the data provided by VAST-D and the opportunities they provide for developing person-centered analyses.

References

  1. Mohamed S, et al. JAMA 2017;318(2):132-45.
  2. Fava M. JAMA 2017,318(2):126-8.