One size does not fit all…or does it?: Heterogeneity within depression

Depressive disorders are “the world’s most important cause of disability”, as emphasized in the session title. Depression is a prevalent, chronic, relapsing condition, and – unlike many chronic physical disorders – it has an early onset. Hence there are many years of life during which its damaging effects are experienced.1 Professor Brenda Penninx, Amsterdam University Medical Center, the Netherlands, also pointed to under- or mis-treatment and lack of maintenance strategies as contributors to disease burden.

Professor Penninx presented a strong case for subtypes within depression – more specifically “typical” and “atypical” types – which, although of similar severity, can be distinguished by symptomatology, demography, prior life experiences, the course of disease, underlying pathophysiology, and genetics. If validated, this distinction has clear implications for therapy.

While typical depression often presents with decreased appetite and weight and insomnia, the opposite symptoms characterize atypical depression, Professor Penninx reported.2

Smoking, and negative life events are associated with typical presentation, while atypical patients are more likely to be female and have an earlier disease onset then their typical counterparts. Persistent anxiety is found in typical depression while atypical patients show a persistently poor metabolic profile.

The need is to find out what works for which patients

In terms of pathophysiology, hyperactivity of the HPA axis (evident, for example, in raised cortisol) is found in patients with typical depression. Atypical patients show immune-inflammatory features, metabolic dysregulation and leptin resistance.

Typical patients are more likely to have gene markers which overlap with those seen in schizophrenia, while the genetics of atypical patients suggests metabolic dysregulation.

For typical patients, antidepressant and/or cognitive therapy are typical entries for treatment, while atypical depression patients may require other interventions including lifestyle changes.

Professor Penninx emphasized that we are only at the start of the large-scale multidisciplinary research needed to understand subtypes; and the presence or absence of immunometabolic features is only one of many possible means of distinguishing between groups of patients.

Reducing the disability burden

The scale of the task is illustrated by the Netherlands Study of Depression and Anxiety (NESDA), which is the source of the findings presented above. More than 200 researchers have been involved in following a naturalistic cohort of more than 3000 people (patients and controls) for 13 years.

We treat episodes but should also have a long-term strategy to prevent relapse

That said, such efforts are clearly worthwhile, since there is an emerging consensus that the heterogeneity within depression is a major reason that our efforts at therapy are less successful than we would like to be.

The theme of heterogeneity was reinforced by Sidney Kennedy, University of Toronto, Canada, who confirmed that there are 227 different combinations of symptoms, each of which could lead to a diagnosis of major depression.

Heterogeneity within depression is a major reason that our efforts at therapy are less successful than we would like

We do better with measurement-based care

There is a disconnect between our diagnostic systems and our choice of treatments, he argued. The future is one in which biological and imaging markers will play a major role. But, even with current understanding, we can and should still able to do better.

Care that involves repeated measurement using an instrument such as the Patient Health Questionnaire (PHQ-9) leads to fewer people continuing to receive suboptimal doses. We can also make better use of algorithms that select treatment based on the side effects most relevant to our patients, he said.


  1. Vos T et al. Lancet 2012
  2. Penninx B et al. Neurosci Biobehav Rev 2016