SCOTTSDALE, ARIZ. – Failing to see depression as a chronic condition that needs to be managed has hampered the ability of psychiatry to help patients with the mood disorder, according to the principal investigator of a recently published prospective study on the most refractory of depression cases.

“The majority of research [in our field] has been on how we manage acute episodes of depression. But, these are chronic, often lifelong conditions. We need to pay attention to that and come up with solutions,” said Scott T. Aaronson, MD, program chair of this year’s annual meeting of the American College of Psychiatrists, in an interview.

For Dr. Aaronson, who also directs clinical research at the Sheppard Pratt Health System in Baltimore, such solutions would include basing depression staging on the oncology model to help determine treatment algorithms. He said that he also would like to see the expanded use of electrical current therapies and trial designs that are more inclusive, with clinical endpoints that have lower thresholds than are typically used now.

“There are people who have these terrible depressions, that, even if you could get them 25% better over the course of years instead of 75% better for just a few months, they’d probably have a much better prognosis and a better quality of life. We need to pay attention to that and think of a longer horizon than we currently do,” Dr. Aaronson said.

During a scientific session at the meeting, he presented data from a 5-year, observational registry study, conducted in nearly 800 people with severe treatment-resistant depression – a population for whom no current evidence-based treatments exists – showing that adjunctive vagus nerve stimulation (VNS) had superior outcomes and mortality, compared with treatment as usual (Am J Psychiatry. 2017 Mar 31. doi: 10.1176/appi.ajp.2017.16010034).

Between January 2006 and May 2015, the multicenter study enrolled adults with unremitting unipolar or bipolar depression lasting at least 2 years. It also enrolled adults who had experienced three or more depressive episodes and had failed four or more depression treatments, including electroconvulsive therapy (ECT). People with a history of psychosis or rapid-cycling bipolar disorder were excluded. One cohort came from a patient registry designed as a postmarketing surveillance study stipulated by the Food and Drug Administration for the approval of the refractory depression indication for VNS. Another cohort came from a study that compared patients with refractory depression who received VNS therapy at various doses. People in the registry cohort were seen at 61 U.S. sites in different settings. Patients were assigned to treatment as usual or treatment as usual with adjunctive VNS based on their preference of a treatment arm. Often, patients’ treatment arm depended on whether implantation was available at their site or what their insurance would cover.

In all, 494 patients were in the VNS study arm, and 301 were in the treatment-as-usual arm. The FDA approved use of the pooled data. People in the dose-finding cohort all had VNS implants when they entered the study, and, unless lost to follow-up, all were observed for 60 months, regardless of the point at which they entered the study. About two-thirds of the dose-finding patients remained in the study for all 5 years, as did about half of the registry study cohort. Of those involved, 22 patients exercised their option of switching treatment arms, but their data were censored from the efficacy analysis. At baseline, the mean Montgomery-Asberg Depression Rating Scale (MADRS) score was 29.3 for the treatment-as-usual group and 33.1 for those in the VNS adjunct group. Responders were those who had a 50% or greater reduction in MADRS scores at any point post baseline.

The 5-year cumulative response to treatment rate in the adjunctive VNS group was significantly higher at 67.6%, compared with 40% in the treatment-as-usual arm (P less than .001). The cumulative percentage of first-time responders in the VNS adjunctive arm was nearly double that of the treatment-as-usual group at all follow-up points in the study, and they tended to respond by 1 year, compared with 2 years in the treatment-as-usual group (P less than .001).

A secondary efficacy endpoint was changes in the Clinical Global Impression–Improvement (CGI-I) scores. These also favored the VNS adjunctive group, which had a 75.9% cumulative CGI-I response rate, compared with a 48.6% rate in the treatment-as-usual arm (P less than .001). Scores on the Quick Inventory of Depressive Symptomatology–Self-Report (QIDS-SR) were consistent with these results: The cumulative response rate in the VNS group was 64.7%, compared with 41.7% in the treatment-as-usual arm (P less than .001).

There were 15 deaths during the study, 7 in the VNS arm and 8 in the treatment-as-usual arm. There were two suicides in each arm, meaning the VNS arm, which was larger, experienced a lesser rate of suicides as the treatment-as-usual group. “This is a fairly key point,” Dr. Aaronson said in the interview.

The remission rate also was significantly higher in the adjunctive group at 43.3%, compared with 25.7% in the treatment-as-usual group (P less than .001). Differences in CGI-I and QIDS-SR scores also were statistically significant and were both higher in the VNS adjunct group, with cumulative response rates from baseline CGI-I scores of 49.7% vs. 21.4%, and changes in QIDS-SR scores of 40.4% vs. 25.0% (P less than .001).

Yet, these statistics do not describe the full potential affect of adjunctive VNS on refractory depression, according to Dr. Aaronson. “There were people in this study who did not meet the endpoint but who were no longer suicidal. Some [reported that they] could now just enjoy riding their bicycle,” he said. “If you asked them, ‘Did they think participating was worthwhile?’ they would tell you there was no question about it. Small differences can be incredibly meaningful for these folks. We need to rethink what success means when we treat chronic depression.”

In addition to higher mean depression rating scores at baseline, those in the VNS group also had higher rates of psychiatric hospitalizations and suicide attempts, suggesting more severe illness in this group. However, it is the fact that this group, with its higher response rate overall, also had higher baseline rates of exposure to ECT that excited Dr. Aaronson the most about the study.

A subanalysis showed that 58.7% of the adjunct VNS group and 36.2% of the treatment-as-usual arm had all had at least seven right lateral treatments of ECT, typically an exclusionary criterion in depression treatment trials. For patients in the VNS arm who previously had responded to ECT, the cumulative response rate at 5 years, based on MADRS scores, was 71.3%, compared with 56.9% of those who had responded to ECT in the treatment-as-usual group, a statistically significant difference (P less than .006). Further, a significant difference in response was recorded at 9 months and then sustained throughout the study.

For the ECT nonresponders in the VNS arm, the response rate in this study was 59.6%, compared with 34.1% for the ECT nonresponders who were receiving treatment as usual (P less than .001). Statistical separation of the two arms began after about 2 years and continued throughout the study.

“This is my personal, favorite part,” Dr. Aaronson said in the interview. “We don’t consider ECT very much, but it is, without question, one of the single most effective acute treatments we have in all of psychiatry for depression. The problem is that, for the majority who respond to it, they are sick again within 6 months. The point here is that, if you’ve ever responded to anything, including ECT, we now have a marker for who will respond to VNS.”

In part because VNS is a chronic, and comparatively less expensive, treatment, the study also has implications for patients on maintenance ECT, Dr. Aaronson said. “Wouldn’t it be terrific if I could offer them VNS rather than continuous ECT, which I worry in the long run can be hard on brains and which is expensive and inconvenient?”

More than one-third of people diagnosed with depression have the treatment-resistant type, the standard definition of which is that a person previously has failed two or more treatments. In this patient population, between 10% and 15% will go on to fail at least four treatments, Dr. Aaronson said.

Although previous failure at least 4 previous depression treatment regimens was one of the inclusion criteria in this study, the VNS population had failed an average of 8.2 previous treatments, compared with 7.3 in the treatment-as-usual arm. These data, together with the subanalysis data on ECT responders, make a compelling case for staging depression, Dr. Aaronson said.

He said, he believes that, with more study and differently structured trials, it can be demonstrated that there also should be a clinical diagnosis of “severe” treatment-resistant depression.

Dr. Aaronson and his colleagues are currently seeking funding to conduct a national study that is randomly controlled using VNS or a sham treatment. Once efficacy data are sufficient, making the case for staging depression will be easier, Dr. Aaronson said.

“I am a firm believer that we should look at psychiatric illnesses the same way we do cancers – using levels of severity,” he said. “The neat thing about [these data on] VNS is that it gives me the bully pulpit to start preaching that gospel.”

Dr. Aaronson’s relevant disclosures include Genomind, LivaNova, Neuronetics, Otsuka, Sunovion, and Takeda.

wmcknight@frontlinemedcom.com

On Twitter @whitneymcknight

Clinical staging of depression endorsed