Unraveling PTSD and comorbid personality disorders: insights from neuroimaging and clinical effect studies

Approximately 70% of individuals experience at least one traumatic event in their lifetime, and around 4% develop post-traumatic stress disorder (PTSD). PTSD is characterized by four main symptom clusters: re-experiencing/intrusions, avoidance, negative alterations in cognition and mood, and changes in arousal and reactivity. Dissociative symptoms are also common. About 35% of those diagnosed with PTSD have a comorbid personality disorder, with paranoid, borderline, and Cluster C (avoidant, obsessive-compulsive, and dependent) personality disorders being the most prevalent. Psychotherapy is the primary treatment for both PTSD and borderline personality disorder (BPD). However, no established treatment guidelines exist for Cluster C personality disorders (CPD). Treatment for individuals with comorbid PTSD and personality disorders is often less effective. It remains unclear whether PTSD symptoms should be treated first or concurrently with personality disorders. To address these challenges, the PROSPER (PRediction and Outcome Study in PTSD and PERsonality disorders) study was initiated. This study comprises two randomized clinical trials. Participants were assigned to either trauma treatment (12-18 sessions) alone or trauma treatment plus a year-long group therapy focusing on the personality disorder. A total of 254 participants were involved, with 89 undergoing MRI scans. Extensive research has examined the neurobiological correlates of PTSD, identifying two subtypes: re-experiencing/hyperarousal and dissociative, with newer models including several networks that are disrupted in PTSD. Regarding personality disorders, a few neurobiological studies have yielded inconsistent results. Factors such as dissociative symptoms, medication use, and the choice of scanning tasks may contribute to these discrepancies. Only a few small studies have explored PTSD with comorbid personality disorders. In this thesis, we present the results of neurobiological studies conducted within the PROSPER study on individuals with PTSD and comorbid PDs. Chapter 2 describes the design of a neuroimaging study on a subsample of the PROSPER trials, including participants with PTSD and comorbid BPD and/or CPD, as well as control subjects. We collected resting-state and task-based functional and structural imaging data. Chapter 3 examines whether the type of comorbid personality disorder affects brain response to an emotional face task. Traditional analyses found no consistent differences in brain activation between personality disorder comorbidities. However, Bayesian analyses indicated higher activation in several brain regions in participants with PTSD, BPD, and CPD compared to those with only PTSD and BPD or PTSD and CPD. We also found associations between dissociation severity, emotion regulation problems, and activation in specific brain regions, supporting a dimensional approach to personality disorders. Chapter 4 presents a meta-analysis of 12 studies on psychotherapy-induced brain activation changes during negative emotional processing in PTSD. The analysis revealed a decrease in activation in fear and cognitive control network areas after therapy, though these findings were not significant after correcting for multiple comparisons. Chapter 5 explores changes in brain activity pre- and post-treatment in the PROSPER samples. No significant changes in brain activation were found despite clinical symptom reduction, suggesting that the emotional face task may not have been sensitive enough to detect changes. In Chapter 6, we studied predictors of treatment attendance, finding that 40% of participants attended fewer sessions than recommended. Predictors of higher dropout included lower educational status and inadequate social support. The discussion in Chapter 7 synthesizes these results, highlighting the importance of a dimensional approach to personality disorders and suggesting that future research focus on treatment responders and the investigation of "biotypes." In conclusion, this dissertation aims to fill a gap in the understanding of PTSD and comorbid personality disorders, particularly Cluster C disorders. While no evidence was found for changes in brain activation following psychotherapy, our findings support a dimensional model of personality pathology and the use of Bayesian statistics.