In the Netherlands, an estimated 80 000 patients suffer from treatment-resistant depression (TRD), who do not respond to all conventional treatments (psychotherapy, different classes of antidepressants and electroconvulsive therapy). TRD patients have a very high suicide risk, with more than 30% of patients attempting suicide at least once during lifetime. Besides the extremely high level of suffering, societal costs of these patients are twice as high as those with non-resistant depression, because of the high health care utilization and loss of work.
Deep brain stimulation (DBS) of the ventral anterior limb of the internal capsule (vALIC) is a promising treatment for TRD patients. DBS is an existing treatment for neurologic and psychiatric patients, which modulates activity in specific brain areas with electric pulses given off by implanted electrodes connected with a neurostimulator. Efficacy can be optimized and side effects minimized by changing electrical parameters (eg. voltage). In a previous trial 40% of the patients lost at least half their symptoms after one year of vALIC DBS and active DBS proved to be far more effective than sham DBS.
Although this is promising in a group of severely resistant patients, in the other patients 27% improved only partially and 33% did not improve at all. In addition, it could take up to 6 months to find a therapeutic effect in some patients. Furthermore, relatively energy inefficient parameters were needed (eg. high voltages), shortening the battery life of the neurostimulator. This necessitates a surgical replacement of the battery every 1.5 years on average.
A probable way of improving these outcomes is personalizing DBS. The current practice ignores the considerable heterogeneity of the brain anatomy of specific patients. New insights show several white matter bundles run through the vALIC, one of which is the medial forebrain bundle (MFB). The MFB is a central pathway in the reward circuitry of the brain, showing aberrant functioning in patients with depression. Preliminary analyses show proximity of the electrode to the MFB is associated with a better outcome and the first, small, open-label pilot studies targeting the MFB show response rates of more than 75%.
We propose to treat 24 TRD patients with personalized DBS. With diffusion weighted imaging (DWI) we will visualize white matter bundles in each patient individually and use these images to implant electrodes in the MFB. After surgery, DBS parameters will be optimized for a maximum of 6 months, after which patients will be randomized to active followed by sham DBS, or vice versa to investigate a possible placebo effect. One year after a long term follow-up will establish how stable the effect is. Outcome measures include symptom and quality of life scales, adverse events, neuropsychological tests, and (functional) magnetic resonance imaging (MRI) scans using reward anticipation tasks.
We hypothesize personalized DBS targeted to the MFB is more effective than placebo DBS, will result in a stable decline of depressive symptoms and will be tolerated well. In comparison to conventional DBS, we expect a higher full response rate and a shorter time to response. Personalized DBS will reduce the suffering of patients and caregivers, and decrease health care utilization and costs. In addition, this project could give essential insight into the neurobiology of TRD, which could lead to further optimization of DBS and the development of new treatments.