Signature Response to Light Therapy in Unipolar and Bipolar Major Depressive Episode (MDE)

Major Depressive Episodes (MDE) are severe and common conditions that affect up to 20% of the general population. MDE cause a decrease in psychosocial functioning, quality of life, and is associated with a high rate of suicides. They will be the leading cause of disability by 2030 according to the World Health Organization. The international effort carried out to identify biomarkers of MDE has been hampered by the heterogenous nature of MDE (unipolar, bipolar, seasonal, non-seasonal) and their heterogeneous response to treatment. Response rate to antidepressant drugs is only 40 to 50%, leading to the use of drug combinations and development of alternative therapeutics such as light therapy (LT). A recent meta-analysis demonstrated that LT, as a first line treatment of MDE with and without seasonal pattern (± SP), has comparable efficacy to antidepressants. LT has the advantage of being also effective in improving both sleep, alertness and circadian rhythms, which may be altered in depression, contrary to antidepressant drugs that target mainly mood. Further research is warranted to determine the most efficient lighting parameters to use depending on depression characteristics, as well as to identify signature biomarkers of response.

Light exerts powerful biological effects on mood regulation. A growing body of research supports the efficacy of light therapy (LT) for MDE ± SP. Light information is conveyed to the brain through intrinsically photosensitive retinal ganglion cells (ipRGCs) expressing the melanopsin photopigment. However, the respective contribution of the different photoreceptive systems (rods, cones, melanopsin) and the precise brain circuits that mediate the impact of light on depressive behaviours are not well understood. Recently, we summarised the two distinct retinabrain pathways of light effects on mood in humans and animals (mostly demonstrated in animal research): i) a suprachiasmatic nucleus (SCN)-dependent pathway (a.k.a the circadian pathway) with mood actions via synchronisation of circadian rhythms and other direct effects, and ii) a SCN-independent pathway with mood effects through modulation of the homeostatic process of sleep, alertness, and emotion regulation. It is now well-known that light plays a main role in mood regulation mainly via melanopsin-containing ipRGCs and their direct projections on brain structures. Interestingly, the amplitude of the prolonged pupil constriction after exposure to blue light, reflects the sensitivity of the melanopsin pathway, found to be reduced in MDE with SP and MDE without SP. The examination of retinal sensitivity using electroretinography (ERG) reported winter rod photoreceptors hyposensitivity to light in MDE with SP. Very few studies have shown an impact of LT on networks involved in MDE: emotion regulation networks (prefrontal-limbic anatomy and connectivity) and the glutamatergic system in the ACC. Most studies used LT in combination with sleep deprivation and/or SSRIs. The retina-brain pathways involved are little known, probably because of their small anatomical size, requiring very high spatial resolution, but preclinical models strongly suggest that habenula and thalami nuclei (SCN independent pathway) are crucial for the antidepressive effects of light therapy. Interestingly, thanks to 7T high resolution MRI, a recent study in unmedicated MDE patients found an increased habenula volume associated with disease severity, and preliminary results indicate that functional connectivity of the thalamic subnuclei is associated with mood variation in MDE. Moreover, immune disruption is evident in MDD, and light may influence mood through its effects on the immune system, which is regulated by the circadian clock. The number of immune cells circulating in the blood oscillates throughout the day, with a peak during the resting phase, and recruitment to specific tissues occurring preferentially during the active phase. This migration is so important that injecting pro-inflammatory agents into mice during the resting phase results in 70% mortality, compared with 30% during the active phase, when immune cells are unavailable and mainly found in tissues. This circadian migration pattern also influences the vaccine response, with a better response during the resting phase when lymphocytes are circulating.

In fact, there is a bidirectional relationship where inflammatory mediators regulate sleep by acting directly in specific brain systems and conversely, sleep deprivation leads to alterations in circulating immune cells.

It is now recognized that inflammation plays a role in the pathophysiology of depression and that several migration markers are affected. Given this potential alteration in immune cell migration and the fact that analyses are performed during the day, current studies therefore fail to capture the complexity of immune disturbances associated with MDD.

Beyond immune-related mechanisms, other biological factors may also play a role in the complex relationship between mood and disturbances of sleep and circadian rhythms. One such factor is Toxoplasma gondii (Tg) infection, a common parasitic condition affecting around 40% of the general population in France. Growing epidemiological evidence suggest a potential link between Tg infection and various neuropsychiatric disorders, including schizophrenia, mood disorders, and other behavioral and neurological conditions . Notably, Tg infection has been identified as a potential risk factor for the development of Major Depressive Disorder (MDD) as well as for insomnia in men over the age of 50. However, the heterogeneity in study designs and reported outcomes limits the generalizability and consistency of current findings.

More recently, Professor Wallon's team demonstrated that Tg-infected mice exhibit altered sleep-wake cycles. More specifically, Tg-infected mice had reduced total sleep time and increased diurnal activity as compared to uninfected mice. These findings suggest that Tg infection may influence sleep architecture and host behavior, potentially facilitating parasite dissemination by increasing the host's exposure to predators.

Despite its high prevalence and suspected relevance to psychiatric and neurological health, the relationship between Tg infection, mood regulation, and sleep remains poorly understood. Therefore, it is important to assess the prevalence of both recent and past Tg infections in MDD patients to better characterize this population and explore possible associations with sleep, biological rhythms and mood disturbances.

MDE with winter SP (defined by DSM-5 TR criteria or with the SPAQ we validated in French) may occur annually in the autumn and winter due to the reduction of natural daylight, and affect approximately 1 to 10% of the general population depending on the latitude (5% in France). MDE with winter SP, also named seasonal affective disorder and including both unipolar and bipolar disorders, are associated with specific abnormalities of mood-photic regulation, in both SCN-dependent and SCN-independent pathways. For instance, the serotonin reuptake transporter (5-HTT), considered as an important central biomarker in SP pathophysiology, has been found significantly reduced in the anterior cingulate (ACC) and the prefrontal cortex (PFC) after 2 weeks of LT, in association with decreased symptomatology. In addition, recent studies in SP reported several sleep objective abnormalities using polysomnography (PSG), but these studies did not compare MDE ± SP. In addition, frequent sleep comorbidities, such as sleep apnea, have been reported, with deleterious impact on response to MDE treatment. PSG gave access to macro- and micro-architecture of sleep and to sleep homeostasis via slow wave density analysis, but also to interhemispheric connection (reported to be altered in MDE), to ultradian sleep regulation and to the interplay between sleep and sympathovagal balance. Our recent meta-analysis reported that patients with SP showed increased REM sleep and decreased REM latency during acute phase and also during remitted phases, without alteration of the slow wave sleep. Our work rather confirmed the chronobiological hypotheses of SP with a phase advance of circadian rhythms as a trait marker. Finally, some studies observed specific biological clock alterations like a longer duration of melatonin secretion during winter than in summer for patients with SP. These changes of the secretion pattern of melatonin may desynchronize endogenous rhythms and alter sleep architecture in these patients.

To date, no studies have directly evaluated both subjective and objective biomarkers of sleep, wake, biological rhythms, and light signalling pathways and activation in patients with unipolar and bipolar MDE ± SP. In this global context, this timely project aims to identify a signature of a good response to LT in order to implement personalised medicine in psychiatry and increase probability of antidepressant success, using an innovative multimodal approach that allows the identification of new composite signatures and new combinations of biomarkers. The main objective of the research will be to identify the signature of response to LT examining the correlation between the measures of biological and clinical parameters before LT and their evolution at the end of the procedure, and the therapeutic response. This is a prospective multi-centers clinical exploratory one-arm one-intervention.

Eligible MDD patients presenting to the emergency room, or to a psychiatric consultation, or hospitalised for a depressive episode with or without SP (balanced exposure 1:1) will be recruited. SP will be defined according to DSM-5 TR criteria and further investigated with the seasonal pattern assessment questionnaire (SPAQ) that examine subdimensions such as mood, sleep, social contacts, food intakes and weight, and validated in French in patients with MDD. Patients will be evaluated clinically by a psychiatrist and all the inclusion and exclusion criteria will be checked.

Patients will be included at V1 and will be treated for 4 weeks with LT. We chose 4 weeks of treatment as an acceptable and recommended delay to change the antidepressant strategy in case of no response. The primary endpoint of the study will be the therapeutic response to LT measured by the difference of MADRS score between Visit 1 and Visit 4 (end of the therapeutic protocol).