4.2.1. Psychotherapy and HPA Axis

PTSD is considered to be a memory-related fear disorder, which is characterised by an over-consolidation of TE-related fear, or, on the other hand, a failure to extinguish fear memories [54]. GCs have potentiating e ffects on emotional memories consolidation and impair memory retrieval. Several theories exist as to which mechanisms are involved in PTSD memory-related pathophysiology [54], including over-consolidation due to excessive noradrenergic signalling as a consequence of low levels of GCs [192] or due to GC-enhanced (as a result of GR hypersensitivity) fear memory consolidation [42]. After consolidation, memories can be retrieved. Important consequences after retrieval are the reconsolidation and extinction of memories, which seem to share the same role of GCs in both the amygdala and hippocampus as that of consolidation [54,193]. This is an important factor, as some PTSD psychotherapeutic interventions rely on fear learning paradigms which can be enhanced by GCs, depending on the knowledge of the adequate moment, dose, and choice of the GC to be administered. On the other hand, a GR antagonist might be more adequate if our aim is the GR hypersensitivity pathophysiological hypothesis of PTSD [54]. One example of this is the case of the most studied psychotherapy indicated for treatment of PTSD, that of PE, which relies on fear learning paradigms, particularly extinction learning, which, in turn, depend largely on the actions of GCs [54]. Extinction includes repeated presentations of the feared stimulus without reinforcement. Interindividual variability in extinction learning can be mediated in part by SNPs in *FKBP5* [180] and constitute one of the reasons for di fferent treatment responses in patients with PTSD.

The first study to address the association of HPA axis-related risk factors with psychotherapeutic outcomes in PTSD patients showed that after brief eclectic psychotherapy, significant changes occurred in levels of cortisol and dehydroepiandrosterone in PTSD patients with civilian-related trauma. Those who responded to therapy showed an increase in cortisol and dehydroepiandrosterone levels, while for those who did not respond, both hormone levels decreased [32].

Accordingly, a higher bedtime salivary cortisol and 24-h urinary cortisol excretion predicted positive treatment outcomes in PTSD veterans after PE therapy or a weekly minimal attention intervention for 12 consecutive weeks [35]. Another study of combat-exposed male PTSD patients found that a high cortisol awakening response (CARi) before trauma-focussed psychotherapy predicted

symptom reduction after 6–8 months. In this study, no di fferences were found between the PTSD and the non-PTSD group who had received no treatment with regard to a decrease in CARi at the end of the treatment period [194]. Furthermore, a study of veterans with PTSD showed that an increased salivary cortisol response to personal trauma script-driven imagery task prior to PTSD therapy was significantly and uniquely related to reductions in the core symptoms of PTSD after PE [195]. Interestingly, an 8-week psychosocial intervention with war-a ffected adolescents regulated hair cortisol levels. While this intervention decreased hair cortisol levels for young patients with hypersecretion and medium secretion, it increased hair cortisol levels for young patients with hyposecretion, in relation to the control group [196]. Steudte-Schmiedgen et al. [197] sugges<sup>t</sup> that hair cortisol levels can be valuable to complement research into long-term HPA axis predictors and correlates of clinical outcomes in response to psychotherapy for PTSD patients.

A literature review revealed preliminary evidence that the lower levels of pre-treatment GC are related to poorer PTSD psychotherapeutic treatment gains [12]. The authors hypothesised that HPA axis dysregulation increases both behavioural and emotional avoidance, which interferes with the efficacy of psychotherapy, especially exposure-based interventions. This could constitute a strong argumen<sup>t</sup> for GC-enhanced psychotherapy. Another argumen<sup>t</sup> could be related to the fact that GCs can potentiate extinction learning [180] or decrease the retrieval of TEs memories on exposure to stimuli [198], either spontaneously [199], or during the application of PE therapy [179].

Imaging studies have also been carried out to understand whether structural modifications occur after psychotherapy. A metanalysis revealed that psychotherapy e ffects in PTSD increased prefrontal and decreased limbic activity, particularly by normalising hippocampal function and morphology [200]. Later studies continued to confirm these findings. Butler et al. [201] found hippocampal grey matter increases after six weeks of multimodal psychological treatment for combat-related PTSD, when compared with a waiting-list control group. Another study also found that clinical improvement during cognitive behavioural therapy in PTSD was predicted by increased *FKBP5* expression and increased hippocampal size [138]. Increased hippocampal volume and elevated *FKBP5* expression were significantly correlated. The hippocampus has been shown to be highly sensitive to the e ffects of GCs [202]. These studies highlight the key role that hippocampal neuroplasticity can play in improving resilience and recovery from traumatic stress, especially bearing in mind its involvement in the HPA axis regulation of memory extinction [54] and also that it is one of the major structures of neurogenesis in the adult human brain [203].

### 4.2.2. Psychotherapy Interaction with Genetic Makeup in PTSD

Since G × E interactions are determined by a certain confluence of environmental factors on SNPs and epigenomic mechanisms, and as these interactions exert an influence on the HPA axis-related pathophysiology of PTSD development and maintenance, we can accordingly presume that G × E interactions could be beneficial for the treatment of these and other disorders—if we could determine the terms of the interaction. In other words, what environmental change would be required for which genetic makeup? This environmental change could be a psychotherapeutic intervention.

Psychotherapeutic e ffects result from learning and memory-related neuronal plasticity [204] which ultimately depend on intervention-associated gene expression [205]). As SNPs and epigenetic changes have direct e ffects on gene expression, several lines of investigation have tried to better understand these relationships.

HPA axis-related genetic makeup information can be useful in psychotherapy practice through several of the following ways. Firstly, the SNPs of genes regulating the HPA axis function, such as *NR3C1*, *FKBP5*, and *CRHR1*, can interact with psychotherapeutic interventions in patients with specific PTSD characteristics and result in better outcomes. This knowledge can enable informed choices of the best intervention. Likewise, knowledge of the epigenome that is associated with better outcomes can also inform the decision regarding the most appropriate psychotherapy. This information could be constituted by biomarkers for both the indication and outcome of the psychotherapies for specific

patients, as, for example, appears to happen in the case of a high level of pre-treatment cortisol as a predictor of better treatment outcomes [35,194,195]. In addition, psychotherapy addresses the trauma which seems to be the main culprit in HPA axis dysregulation interacting with genes.

Furthermore, the usefulness of genetic makeup information for the prescription of psychotherapy and the respective outcomes is promising, based on the results of several studies which have researched the interaction between psychotherapy and both SNPs and the epigenome.

The B*cl*I SNP G carrier states predicted treatment gains after 12 consecutive weeks with PE therapy, or a weekly minimal attention intervention for patients with PTSD [35]. Homozygotes of the same B*cl*I SNP risk allele evidenced more traumatic memories of intensive care unit treatment at six months after cardiac surgery [143]. Taken together, this information could help to identify those subjects who are at higher risk of PTSD symptoms and provide them with a psychotherapeutic intervention before, or shortly after the TE exposure, depending on its predictability. Interestingly, a randomised controlled trial study showed that a pre-operation minimal cognitive behavioural intervention which targeted homozygous carriers of this SNP's G high-risk allele reduced traumatic memories and posttraumatic stress disorder symptoms after heart surgery [206]. Interestingly, GG carriers of this SNP showed increased emotional memory performance when compared to GC and CC carriers [82].

Traumatised Ugandan carriers of the *FKBP5* rs1360780 risk allele were at increased risk of PTSD symptoms relapse after 10 months of Narrative Exposure Therapy (a form of exposure-based short-term therapy), whereas non-carriers continued to show a reduction in symptoms [36]. This is an interesting result, as the genetic makeup which correlates with functional distinct roles impacts on di fferent treatment outcomes [14,152].

Gene expression studies have also shown that increased *FKBP5* expression after treatment is associated with successful response to trauma-focussed psychotherapy [33,138]. Levy-Gigi et al. [138] found significant increases in *FKBP5* expression and in hippocampal size, which were correlated, in PTSD patients after 12 weekly 1.5-h sessions of trauma-focussed cognitive behavioural therapy. Improvement in PTSD symptoms was predicted by increased *FKBP5* expression and increased hippocampal volume, although the primary predictor was *FKBP5* expression [138].

A recent review of studies addressing the associations between various psychiatric disorders and several genes methylation before and after psychotherapeutic interventions concluded that DNA methylation change can be considered a marker of treatment outcome [100]. However, the author advises adopting caution in relation to the interpretation of results, owing to di fferent methodologies across studies and also the need for the biological functional characterisation of the findings.

Higher *NR3C1* exon 1F promoter methylation levels predict a better treatment response after 12 weeks of PE psychotherapy, whereas *FKBP5* exon 1 promoter region methylation levels decreased in those subjects who responded with remission to the same psychotherapy. Correspondingly, those subjects who responded to treatment had higher *FKPP5* gene expression than those who did not respond [33]. The authors argue that GR gene methylation did not change in response to treatment, due to association with early environmental changes which could be more enduring. The methylation changes could constitute biomarkers of PTSD prognosis and symptom severity, respectively, as the environment part of G × E interactions could be attributed to epigenetic mechanisms [14].

Other studies addressing epigenomic mechanism in other genes associated with PTSD psychotherapy are currently being carried out. A longitudinal study of genome-wide DNA methylation levels found that the successful treatment of PTSD patients with trauma-focused psychotherapy (cognitive behavioural therapy, with or without EMDR) was accompanied by changes in DNA methylation at 12 di fferentially methylated genomic regions and particularly of the *ZFP57* gene with regards to prospective evidence [34,207]. *ZFP57* codes for a transcriptional regulator of genomic imprinting which has been associated with hippocampus-related stress vulnerability [208].

The previous research reviewed shows that the HPA axis is an important mediator of treatment approaches to PTSD, based on the fear learning paradigm of PTSD pathophysiology. Although promising, pharmacotherapy targeting the HPA axis still has a long way to go. In addition, GC-enhanced

psychotherapy also represents a possible approach which needs to be investigated better. Indeed, psychotherapy research for treating PTSD needs to be pursued, as outcomes are far from good. It seems that certain endophenotypes can be useful in predicting treatment outcomes and psychotherapy has shown to normalise HPA axis-related functions in several ways, including levels or cortisol, gene expression, and CNS function and structure. The way genetic makeup (namely SNPs and DNA methylation) interacts with psychotherapy can also be useful to inform us of the best treatment options. However, further studies of these relationships are needed before we can apply these interesting discoveries to clinical practice.
