[1] 12w Journal Pre-proof Hyper-reflective foci in the retina of active relapse-onset multiple sclerosis

[1] 94w This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

[1] 203w Ninety-one eyes of 46 RMS patients and 74 eyes of 37 HC (Healthy Controls) were evaluated (see Supplementary Materials 2 for clinical and demographic data). One eye of a RMS patient was excluded due to the presence of a choroidal nevus at the posterior pole. Mean disease duration at SD-OCT acquisition was 2.5 months (range 0-15). Notably, 43 (93.5%) patients had disease duration <12 months. Mean HRF number was significantly higher in RMS compared to HC both in the whole retina (28.8±9.3 vs 23.9±7.1, p=0.0007) and in the Inner Retina (IR) (24.5±8.4 vs 19.4±6.5; p<0.0001) (Figure 2). No difference in HRF number was observed in the Outer Retina (OR) (4.3±2.0 in RMS vs 4.4±1.2 in HC, p=0.94). The intergrader agreement was at least substantial for all measurements (intraclass correlation coefficient: 0.82). RMS and HC did not differ in the macular volume throughout all the retinal single layers and in the retinal thickness in the macular ETDRS map any analyzed layers (Supplementary Materials 2). No correlation was observed between HRF number and other SD-OCT parameters. With regard to the retinal thickness in the macular ETDRS only the internal nasal sector of ONL was slightly lower in RMS compared to HC (p=0.02) (Supplementary Materials 3).

[1] 222w Activated and proliferating microglia, a peculiar histological feature of Multiple Sclerosis (MS) pathology, is observed not only in inflammatory lesions but also in normal appearing white and grey matter. 1,2 Several lines of evidence suggest that microglia may play a pivotal role in MS pathology, promoting local inflammation and demyelination, and possibly contributing to the neurodegenerative process that characterizes disease progression. 3 Retinal microglia, that share morphological features and functional properties of brain microglia, 4 is also thought to play a role in local inflammation and neurodegeneration, as also suggested by findings in experimental autoimmune encephalomyelitis (EAE) and experimental optic neuritis (EON). 5 Spectral domain optical coherence tomography (SD-OCT) allows the identification of unique features inside the retinal layers in pathological conditions, such as the socalled "hyper-reflective intra-retinal foci" (HRF), 6 indicated as aggregates of activated proliferating microglial cells and possible in vivo biomarker of local inflammation. [7][8][9] We investigated the presence of HRF in the retina of relapse-onset MS patients (RMS) at the time of diagnosis. Inasmuch as HRF reflect a highly dynamic process, we selected patients with active disease and not yet treated with anti-inflammatory or immunosuppressive drugs (See supplementary Materials 1 for inclusion criteria, OCT protocols and statistic analysis). Since retinal microglia can likely be triggered by acute optic nerve inflammation, we excluded patients with history of optic neuritis.

[2] 58w In this explorative study, we analyzed the retina morphological features in the eyes of a highly homogeneous cohort of RMS (clinically and/or radiologically active, with very short disease duration, no history of optic neuritis and never treated with immunosuppressive/immunomodulatory drugs) and found that these patients had a significantly higher number of HRF in the IR compared to HC.

[3] 476w HRF are small, intra-retinal, hyper-reflective lesions, visible on structural linear SD-OCT scan, in either normal subjects and in patients with both retinal and choroidal diseases. 7,10 Although no consensus about HRF origin currently exists, 11 the most accredited hypothesis suggests that HRF, having specific features (i.e., small in size with a maximum diameter of 30 μm, with a moderate reflectivity (similar to RNFL) and without back shadowing) represent activated and aggregated clusters of microglial cells physiologically resident in the IR. 8 In the eyes of diabetic patients, HRF are initially observed in the IR where microglia is usually resident, but progressively increase in number 32 and spread towards middle and outer retinal layers, 7 mimicking microglia activation and migration from IR to OR described in histopathological studies. Moreover, increasing evidence points out a role of microglial cells inflammatory processes taking place in the retina. 11 Finally, an increase of HRF were recently identified after phacoemulsification, further pointing out their inflammatory origin, and suggesting their use as SD-OCT biomarkers of intra-retinal microglia activation. 8 In MS brain, activated and proliferating microglial cells were described in focal lesions, areas of neurodegeneration and normal appearing WM and GM, and "microglia nodules" were associated to degenerating axons, stressed oligodendrocytes, or deposits of activated products of the complement pathway. 12 Moreover, microglia nodules were observed in absence of lymphocyte infiltration and demyelination and interpreted as "preactive lesions", i.e., preceding the formation of active demyelinating lesions. 2 This is the first report of an increased HRF number in the IR of RMS patients at the time of the diagnosis. This finding suggests that microglia activation in the retina likely parallels microglia activation in the normal-appearing brain tissue and merits to be investigated as a possible prognostic marker in MS. Interestingly, in the RMS patients included in this study, both single retinal layer thickness and volume did not differ from HC. Moreover, no differences in thickness were found in any sector of pRNFL. It is well known that in more advanced MS phases both macular thickness and pRNFL, mainly in the temporal sector, are reduced even in eyes with no history of ON. [13][14][15] This discrepancy can be explained by the different demographic and clinical features of the RMS population enrolled in the present study. The presence of an increased HRF number in the IR in absence of retinal layer thinning or other signs of local pathology might indicate that in early RMS phases microglial activation precedes any neurodegenerative process in the retina. The patients included in this study are currently enrolled in a longitudinal prospective follow-up that may give us the possibility of 1) analyzing quantitative (number) and qualitative (location) microglia changes in the retina, 2) evaluating whether HRF have a prognostic clinical values, and 3) confirming whether the appearance of HRF precedes RFNL thinning in RMS with no history of ON.

[4] 82w Although our preliminary observations must be confirmed, we would like to stress the strengths of our study: the number of patients/eyes analyzed and the unique features of the RMS population, namely, very short disease duration, lack of immunomodulatory/ immunosuppressive therapies, active disease, no history of clinically evident optic neuritis, no evidence of subclinical optic neuritis. Bars show mean±SD and range of hyper-reflective foci (HRF) number in the inner (IR) and outer (OR) retina in relapse-onset multiple sclerosis (RMS) and healthy controls (HC).