Blood will tell:
a new diagnostic
approach for PTSD

People with PTSD experience strong, persistent distress when reminded of a traumatic event. It’s estimated to affect around three in 100 people and can result in feelings of isolation, guilt and anger. But Marmar explains that contrary to popular belief, PTSD is not a modern affliction at all. The first mention of the condition actually emerged in the literature more than 3,000 years ago, in a report from the physician to the king of Assyria.

“A warfighter presented to his doctor because he was having disturbed sleep. The patient would awaken in the middle of the night, having had a dream of being in combat. When he awakened from the dream, he would feel he was on the battlefield. He would try to scream but he was frozen.”

These are classic symptoms of PTSD, Marmar argues, even if the illness hadn’t been given a name back then. So why is diagnosis so challenging if we’ve known about the symptoms for literally thousands of years? The problem, as is so often the case in psychiatry, is shame. There’s so much stigma surrounding trauma, leading to many patients suffering in silence, and perhaps even lying in psychiatrist assessments to avoid the diagnosis. This is why we need new tests that don’t rely on self-reporting, but are instead based on physical measures.

“People with PTSD have been complaining about nightmares and flashbacks and trouble falling asleep for the past 3,400 years. So what do we need blood markers for?” asks Marmar. “The answer is, especially if you're dealing with police officers and military personnel, some people can be very ashamed of having these symptoms. If you screen them and give them a questionnaire, they'll say they’re fine. But if you ask their wife, she's getting kicked in the middle of the night because they're having disturbing dreams.”

To tackle this problem, a high-throughput, efficient screening mechanism for the illness is sorely needed, says Marmar. He wants to do for PTSD what mammography and PSA tests have done for breast and prostate cancers.

Working with several other PTSD leaders and institutions, including Dr Frank Doyle, Dean of Harvard John A Paulson School of Engineering and Applied Sciences, as well as Dr Marti Jett from the US Army Medical Research and Development Command, the NYU team set out to develop a blood-based biomarker test to diagnose PTSD in veterans.

They recruited 83 men with confirmed PTSD who had served in Iraq or Afghanistan. An equal number of healthy controls, who had also been deployed to war zones but did not have the condition, were chosen too.

Determining the biomarkers that might be relevant for PTSD would require an awful lot of blood — 21 tubes from each volunteer. Marmar explains that the participants needed to be screened for anaemia first.

“We studied a very deep range of markers that are detectable in blood, including genes and gene expression products, metabolites and proteins and even changes in the genome which are caused by stress called epigenetics,” he says.

The scientists tested nearly one million biomarkers. Using a combination of techniques, including machine-learning and genomic and molecular testing, they were able to narrow them down — first to 343, then to 77, and finally to 28 markers, which would form the basis of a ‘PTSD blood test’.

“The final 28 markers performed well. They were able to accurately classify who went to Iraq and Afghanistan and came back with PTSD as validated by four hour, super-expert clinical assessments,” reveals Marmar.

Next on the agenda was replicating the work, applying the blood test to an independent group of veterans who had been assessed using the standard clinical questionnaire for the condition. Luckily, the replication study was a success and the team found the blood test gave an accuracy rate of 77%.

The biomarker test is an incredible start. With further validation, the US Department of Defense is even considering using it as a screening tool, to identify service members before and after deployment with features of unresolved PTSD. But far more research will be required before the test is used in standard care.

We don’t yet know why these biomarkers are so important, for instance. Or why PTSD results in these changes. Additionally, the tool has only been tested in male veterans so far. It could be completely different for women, or men with PTSD that does not relate to the battleground, Marmar is keen to point out.

“In a perfect world, we'll find some universal features that are associated with PTSD, or not having PTSD, in both men and women, and in veterans and civilians. And maybe eventually in children and adolescents But first, there's a lot of work to do.”