Optical coherence tomography assessment of disease activity in cryopyrin‐associated periodic syndrome

Abstract Background and purpose Cryopyrin‐associated periodic syndrome is a rare autoinflammatory disease caused by gain‐of‐function mutations or variants in the NLRP3 gene. Clinically, patients suffer from a broad spectrum of both systemic and neurological symptoms. The aim of this study was to determine whether systemic inflammation demonstrated by serum amyloid A (SAA) elevation is associated with neuroinflammation assessed by optical coherence tomography (OCT). Methods Thirty eyes of 15 patients with NLRP3 low penetrance mutations (PwNLRP3) and 20 eyes of 10 age‐ and sex‐matched healthy controls were examined by spectral‐domain OCT as part of routine clinical care. All retinal layers and clinical features were evaluated. Results At baseline no significant retinal neuroaxonal inflammation or degeneration was observed in all measured retinal layers amongst PwNLRP3 compared with healthy controls. In a pooled analysis of all individual OCT time points a significant difference regarding the macular retinal nerve fibre layer was detected. Increased levels of SAA showed a positive association with averaged combined outer plexiform layer and outer nuclear layer volumes (ρ < 0.0001, r 2 = 0.35). Conclusion In cryopyrin‐associated periodic syndrome increased combined outer plexiform layer and outer nuclear layer volumes are mirrored by SAA increase, an acute phase reactant indicating systemic inflammation. Our findings identify OCT as a candidate biomarker to monitor subclinical neuroinflammation and to assess disease activity in PwNLRP3.


evaluated.
Results: At baseline no significant retinal neuroaxonal inflammation or degeneration was observed in all measured retinal layers amongst PwNLRP3 compared with healthy controls.In a pooled analysis of all individual OCT time points a significant difference regarding the macular retinal nerve fibre layer was detected.Increased levels of SAA showed a positive association with averaged combined outer plexiform layer and outer nuclear layer volumes (ρ < 0.0001, r 2 = 0.35).

Conclusion:
In cryopyrin-associated periodic syndrome increased combined outer plexiform layer and outer nuclear layer volumes are mirrored by SAA increase, an acute phase reactant indicating systemic inflammation.Our findings identify OCT as a candidate biomarker to monitor subclinical neuroinflammation and to assess disease activity in PwNLRP3.

K E Y W O R D S
autoinflammation, NLRP3 variants, OCT deafness to vision loss and are often irreversible and detrimental [3].
It was recently shown that two-thirds of patients with NLRP3 low penetrance variants (PwNLRP3) are at high risk to develop severe neuroinflammation; however, neurologists are still lacking biomarkers to assess disease activity [4,5].Along that line, this study aimed to investigate the relationship between systemic inflammation and inflammation of the central nervous system (CNS) in PwNLRP3 by retinal optical coherence tomography (OCT) usage.

Optical coherence tomography
Optical coherence tomography examination was performed using a SD-OCT (Spectralis, Heidelberg Engineering, Heidelberg, Germany) with automatic real time (ART) function for image averaging.One experienced rater checked all scans for sufficient quality and segmentation errors and corrected, if required.The volumes of the peripapillary retinal nerve fibre layer (pRNFL), macular retinal nerve fibre layer (mRNFL), combined ganglion cell and inner plexiform layer GCIPL, the inner nuclear layer (INL), outer plexiform layer and outer nuclear layer (OPONL) and the total macular volume (TMV) were analysed (Figure 1f).Calculation of macular layers is given for a 3 mm diameter cylinder around the fovea from a macular volume scan (20° × 20°, 25 vertical B-scans, ART ≤49).The pRNFL was measured with an activated eye tracker using 3.4 mm ring scans around the optic nerve (12°, 1536 Ascans, ART ≤100).Segmentation of all layers was performed semiautomatically using software provided by the OCT manufacturer (Eye Explorer 1.9.10.0 with viewing module 6.3.4.0,Heidelberg Engineering).In addition, the presence of microcystic macular oedema (MME) was examined in the INL and OPONL.MME was defined as the presence of cystic microlesions on at least two adjacent B-scans.OCT data in this study are reported according to the APOSTEL and OSCAR-IB recommendations [6][7][8].

Statistical analysis
Group differences were assessed with the non-parametric Mann-Whitney U test.Frequencies and percentages were used as descriptive statistics for categorical variables.Correlations were calculated using the Spearman's rank correlation coefficient r.Additionally, cross-sectional differences of OCT values between groups were analysed pairwise by generalized estimating equation (GEE) models to account for inter-eye within-patient correlations of monocular measurements.Analyses of all OCT time points were performed by a linear mixed effects model.Mixed effects models and GEE analysis were both adjusted for relevant confounders including age and sex.
Data of the GEE and linear mixed effect models are not shown but are made available upon request.All tests and graphical representations were performed with R V.3.3.1 (http:// www.R-proje ct.org).

Statistical significance was established at p < 0.05, and all results
were interpreted in the context of an exploratory analysis without adjusting for multiple comparisons.

Clinical features
The median age at symptom onset was 33 ± 14 years.Most patients showed a recurrent (n = 9; 60%) disease course with flares.

Optical coherence tomography changes during follow-up
Next, group differences during follow-up visits between HCs and PwNLRP3 were analysed.Average follow-up time was 2.9 years F I G U R E 1 Cerebral magnetic resonance imaging and OCT images of a PwNLRP3 was diagnosed with relapsing inflammatory optic neuropathy due to recurrent episodes of ON in both eyes at the age of 43.Not long after her first neurological consultation she started manifesting pain and oedema in her left Achilles tendon.Clinical progression over the following years, typically accompanied by signs of inflammation in both blood (elevated SAA and CRP levels) and CSF (pleocytosis and CSF-specific oligoclonal bands) led to further diagnostics.The systemic workup was negative for immunological markers (aquaporin-4 antibodies, MOG-IgG, ANA, c-ANCA and p-ANCA, s-IL-2 receptor) or infectious aetiologies (HSV, VZV, CMV, measles, rubella and lues IgG, borrelia and toxoplasma IgG/IgM).Genetic analyses revealed a homozygous low penetrance variant in the NLRP3 gene compatible with CAPS.The prevalence of MME in a CAPS cohort is unknown; in our cohort of PwNLRP3 MME could be detected in 6.7% of the eyes and only in the INL.Overall, the findings are consistent with a pronounced neuroaxonal retinal degeneration.In this patient, there was only one follow-up OCT examination at an interval of 6 months.In this short course, no further retinal neuroaxonal degeneration was observed.(f) Retinal layers adapted from 'Structure of the retina' by BioRe nder.com (2020).Retrieved from https:// app.biore nder.com/ biore nder-templ ates.(g) Correlation of OPONL and mean SAA levels (assessed at the OCT examination).SAA levels of >5.0 mg/L were considered abnormal.Spearman's rank correlation coefficient r and exact p values are reported.PwNLRP3 with ON 0.49 ± 0.20, p = 0.03).

Correlation of retinal layers and serum amyloid A levels
Interestingly, mean SAA levels showed a positive association with averaged OPONL volumes (ρ < 0.0001, r 2 = 0.35) (Figure 1g).This effect was not dependent on the presence of MME in the OPONL.The examination of the OPONL revealed no MME in any of the 30 eyes of PwNLRP3.However, microcysts within the INL were observed in 2 out of 30 eyes (n=6.7%).Also associations with other laboratory markers such as CRP and leucocyte counts were not observed.
Further analysis of SAA and other retinal layers including INL demonstrated a lack of correlation.

DISCUSS ION
The results of this study provide support that an increase in OPONL volume mirrors concomitant chronic, systemic inflammation in CAPS patients, thereby suggesting OCT as an emerging, candidate biomarker tool for disease monitoring in patients with autoinflammatory syndromes.
It was recently demonstrated that patients with NLRP3 variants are at particular risk of developing severe CNS inflammation and cranial nerve affection [3][4][5].Of those, ON was the most detrimental manifestation leading to substantial neuroaxonal damage and blindness after multiple and recurrent clinical attacks [10].Along that line our data indicate a clear retinal neuroaxonal degeneration (global pRNFL and GCIPL) in patients with a history of ON versus patients without ON.These data are consistent with ON manifestation in other chronic inflammatory CNS diseases such as multiple sclerosis [11].The vulnerability in PwNLRP3 of cranial nerves and especially the proneness of the optic nerve to chronic, sterile inflammation can be readily explained by an increased retinal expression of the NLRP3 inflammasome [12].Previous animal studies suggested that optic nerve injuries activate the NLRP3 inflammasome in retinal microglial cells, typically distributed in GCIPL, OPONL and nerve fibre layers, which then release pro-inflammatory cytokines, thus further perpetuating neuroinflammation [13,14].Vice versa, NLRP3 knockout studies demonstrated a delayed loss and degeneration of retinal ganglion cells following the optic nerve injury [14].
Our study has several limitations including the lack of patients

For
this analysis clinical, laboratory (C-reactive protein [CRP], serum amyloid A [SAA], leucocyte counts) and OCT data collected between 2014 and 2019 were retrospectively included.Clinical details of an index patient are shown in Figure 1a-e.SAA values >5.0 mg/L were considered abnormal and were routinely determined at the Institute of Laboratory Medicine, LMU University Hospital, Munich, during the identical time point of OCT assessment.Inclusion criteria encompassed established diagnosis of PwNLRP3, the genetic verification of mutation or variants in the NLRP3 gene and the presence of follow-up OCT imaging data.Healthy volunteers from the Institute of Clinical Neuroimmunology were recruited to serve as for this study.The study was approved by the local ethics committee of the LMU Munich (project no.600-15 and 427-14).Written informed consent was obtained from each participant according to the Declaration of Helsinki.
(a) Axial fluid attenuated inversion recovery sequence showing massive swelling of the chiasma-pituitary region and both nervi optici.(b) Gadolinium enhanced T1-weighted images demonstrating contrast enhancement involving both nervi optici.(c) OCT images showing the ring scan and (d) the macula scan of the index patient.Global subtotal atrophy of the pRNFL is shown in (c) as well as morphological evidence of bilateral volume reduction (TMV) (d).(e) OCT images of the inner nuclear layer indicating MME (red arrows).
acteristics and disease-modifying therapies and the small sample size, potentially leading to an underestimation of retinal inflammatory effects.Larger and continuous multicentre studies are desired.Taken together, our study reveals that OPONL and chronic, systemic inflammation in CAPS closely parallel each other over time, suggesting a potential role for OCT as candidate biomarker, not TA B L E 1