Antidepressants are often ineffective. Can a blood test change that?

Researchers claim that a blood test can identify which medication is most likely to effectively treat depression, according to a report published in Psychoneuroendocrinology. (But see our analysis below.)

The blood test, developed by the University of Texas’s Brain Institute, measures levels of a patient’s C-reactive protein through a finger-prick sample.

Doctors currently rely on patient questionnaires and their own personal preferences when choosing a treatment for depression. A US study found that up to a third of depressed patients don’t improve on the first antidepressant they are prescribed, and about 40 per cent of people who start taking antidepressants stop taking them within three months.

The researchers say that clinical guidance could improve these figures. They were able to demonstrate that for patients whose CRP levels were less than one milligram per litre, escitalopram (known in the UK as Cipralex) alone was the most effective drug, with a 57 per cent remission rate compared to less than 30 per cent on the other medication.

For patients with higher CRP levels, it was found that escitalopram plus bupropion (which is only licensed in the UK for smoking cessation) was more likely to work, with a 51 per cent remission rate compared to 33 per cent on escitalopram alone.

Dr Madhukar Trivedi, the study’s lead author, said: ‘Currently, our selection of depression medications is not any more superior than flipping a coin, and yet that is what we do. Now we have a biological explanation to guide treatment of depression.’

Instant analysis
The study randomly assigned patients to either a single antidepressant (escitalopram) plus a placebo or two antidepressants together (escitalopram plus bupropion). Escitalopram works by selectively blocking serotonin re-uptake in the brain, where as buproprion selectively blocks norepinephrine and dopamine re-uptake. Serotonin, norepinephrine and dopamine are all neurotransmitters. By blocking re-uptake into cells, this increases the extracellular concentration bathing the cells, which is believed to improve mood.

The key thing to note is that there was no difference between the groups in terms of depressive symptom or side effect outcomes, so more medications didn’t improve the patients’ mood. However, lots of biomarkers such as CRP, serum amyloid P component, and alpha-2-macroglobulin were measured in these patients too, often called acute phase reactants, as they are realised in the acute (early) phase of illness. CRP stands for C-reactive protein and is a marker of inflammation released by the liver and amyloid P is a distant relative of CRP, as is alpha-2-macroglobulin. As far as I am aware depression doesn’t cause inflammation, but it may well be a consequence and most of the patients did indeed have elevated levels of CRP reflecting this.

The ‘highlight’ of the study is that ‘pre-treatment C-reactive protein (CRP) levels predict differential response to currently available antidepressant treatments’. However, this was a short (12-week) study on a small (100) number of patients, so there would never be any definitive answers, only hypotheses generated. And, bizarrely, the link between higher baseline CRP and lower depression severity was found in only one group, not all patients. But if depression is an inflammatory condition then surely higher CRP would mean more severe depression, not the reverse?

It appears in the study that if the CRP was higher, then two drugs were better than one. Furthermore, a correlation between CRP levels and which drug regimen improved symptoms is just that – an association rather than cause and effect. It can work both ways, with CRP influencing symptoms, symptoms influencing CRP or a third factor altogether.

The biomarker finding is interesting but this study is not sufficient to reach any definitive conclusions. It needs to be repeated in a much larger and more diverse population, with longer follow-up, and prospectively assigned by the basis of patients’ CRP at baseline into four groups, eg, high CRP group given two drugs, high CRP group receiving one drug, low CRP group given two drugs, and low CRP group receiving one drug.

The press release claims the study ‘unlocks a new frontier in treating depression’. Well, not so much.
Research score: 1/5