Prediction of visual function from automatically quantified optical coherence tomography biomarkers in patients with geographic atrophy using machine learning.

Geographic atrophy (GA) is a vision-threatening manifestation of age-related macular degeneration (AMD), one of the leading causes of blindness globally. Objective, rapid, reliable, and scalable quantification of GA from optical coherence tomography (OCT) retinal scans is necessary for disease monitoring, prognostic research, and clinical endpoints for therapy development. Such automatically quantified biomarkers on OCT are likely to further elucidate structure-function correlation in GA and thus the pathophysiological mechanisms of disease development and progression. In this work, we aimed to predict visual function with machine-learning applied to automatically acquired quantitative imaging biomarkers in GA. A post-hoc analysis of data from a clinical trial and routine clinical care was conducted. A deep-learning automated segmentation model was applied on OCT scans from 476 eyes (325 patients) with GA. A separate machine learning prediction model (Random Forest) used the resultant quantitative OCT (qOCT) biomarkers to predict cross-sectional visual acuity under standard (VA) and low luminance (LLVA). The primary outcome was regression coefficient (r2) and mean absolute error (MAE) for cross-sectional VA and LLVA in Early Treatment Diabetic Retinopathy Study (ETDRS) letters. OCT parameters were predictive of VA (r2 0.40 MAE 11.7 ETDRS letters) and LLVA (r2 0.25 MAE 12.1). Normalised random forest feature importance, as a measure of the predictive value of the three constituent features of GA; retinal pigment epithelium (RPE)-loss, photoreceptor degeneration (PDR), hypertransmission and their locations, was reported both on voxel-level heatmaps and ETDRS-grid subfields. The foveal region (46.5%) and RPE-loss (31.1%) had greatest predictive importance for VA. For LLVA, however, non-foveal regions (74.5%) and PDR (38.9%) were most important. In conclusion, automated qOCT biomarkers demonstrate predictive significance for VA and LLVA in GA. LLVA is itself predictive of GA progression, implying that the predictive qOCT biomarkers provided by our model are also prognostic.

Fractionated Stereotactic Radiation Therapy for Pituitary Adenomas: An alternative escalating protocol of hypofractionated stereotactic radiotherapy delivering 35Gy in 5 fractions.

PURPOSE: Evaluate efficacy and toxicity of hypofractionated stereotactic radiotherapy (HSRT) for patients treated for pituitary adenoma (PA) with an alternative HSRT escalating protocol delivering 35Gy in 5 fractions. MATERIAL AND METHODS: From June 2007 to March 2017, 29 patients with pituitary adenoma were treated in Antoine Lacassagne Cancer Centre with an alternative HSRT protocol. Prescribed dose was 35Gy in 5 fractions of 7Gy. Radiographic responses were assessed by annual MRI. Hormone blood samples were evaluated each year after HSRT. RESULTS: A total of 29 patients aged between 23 and 86 years (median 54 years) were included. Twelve patients received HSRT for recurrent cases and 12 received postoperative adjuvant HSRT, 5 patients did not have surgery. After a median follow-up period of 47 months local control rate was 96%. One patient presented an out-field tumor regrowth 73 months after HSRT. The majority of PA were endocrine-active (18 patients, 62%). After HSRT, 8 patients (44%) presented complete response on initial secretion, 4 patients (23%) presented partial response on initial secretion. Four patients (14%) presented grade 2 or more acute radiation toxicities. One grade 4 visual disorder was observed for one patient. CONCLUSIONS: HSRT delivering 35Gy in 5 fractions represents a feasible treatment and shows promising results to reduce hormonal overproduction and to improve local control in PA.

Rare coding variant analysis in a large cohort of Ashkenazi Jewish families with inflammatory bowel disease.

Rare variants are thought to contribute to the genetics of inflammatory bowel disease (IBD), which is more common amongst the Ashkenazi Jewish (AJ) population. A family-based approach using exome sequencing of AJ individuals with IBD was employed with a view to identify novel rare genetic variants for this disease. Exome sequencing was performed on 960 Jewish individuals including 513 from 199 multiplex families with up to eight cases. Rare, damaging variants in loci prioritized by linkage analysis and those shared by multiple affected individuals within the same family were identified. Independent evidence of association of each variant with disease was assessed. A number of candidate variants were identified, including in genes involved in the immune system. The ability to achieve statistical significance in independent case/control replication data was limited by power and was only achieved for variants in the well-established Crohn's disease gene, NOD2. This work demonstrates the challenges of identifying disease-associated rare damaging variants from exome data, even amongst a favorable cohort of familial cases from a genetic isolate. Further research of the prioritized rare candidate variants is required to confirm their association with the disease.

A novel missense mutation in HSF4 causes autosomal-dominant congenital lamellar cataract in a British family.

PurposeInherited cataract, opacification of the lens, is the most common worldwide cause of blindness in children. We aimed to identify the genetic cause of isolated autosomal-dominant lamellar cataract in a five-generation British family.MethodsWhole exome sequencing (WES) was performed on two affected individuals of the family and further validated by direct sequencing in family members.ResultsA novel missense mutation NM_001040667.2:c.190A>G;p.K64E was identified in the DNA-binding-domain of heat-shock transcription factor 4 (HSF4) and found to co-segregate with disease.ConclusionWe have identified a novel mutation in HSF4 in a large British pedigree causing dominant congenital lamellar cataract. This is the second mutation in this gene found in the British population. This mutation is likely to be dominant negative and affect the DNA-binding affinity of HSF4.