2†Pathways to detection of non-infectious childhood uveitis in the UK: findings from the UNICORN cohort study.

INTRODUCTION: Prompt detection of childhood uveitis is key to minimising negative impact. From an internationally unique inception cohort, we report pathways to disease detection.UNICORNS is a national childhood non-infectious uveitis study with longitudinal collection of a standardised clinical dataset and patient reported outcomes. Descriptive analysis of baseline characteristics are reported.Amongst 150 recruited children (51% female, 31% non-white ethnicity) age at detection ranged from 2-18yrs (median 10). In 69%, uveitis was diagnosed following onset of symptoms: time from first symptoms to uveitis detection ranged from 0-739days (median 7days), with longer time to detection for those presenting initially to their general practitioner. Non symptomatic children were detected through JIA/other disease surveillance (16%), routine optometry review (5%) or child visual health screening (1%). Commonest underlying diagnoses at uveitis detection were JIA (17%), TINU (9%, higher than pre-pandemic reported UK disease frequency) and sarcoid (1%). 60% had no known systemic disease at uveitis detection. At disease detection, in at least one eye: 34% had structural complications (associated with greater time to detection - 17 days versus 4 days for uncomplicated presentation).The larger relative proportions of children with non-JIA uveitis reported here increase the importance of improving awareness of childhood uveitis amongst the wider clinical communities. There is scope for improvement of pathways to detection. Forthcoming analysis on the full cohort (251 recruited to date across 33 hospitals and 4 nations) will provide nationally representative data on management and the determinants of visual and broader developmental/well-being outcomes.

Germline whole exome sequencing and large-scale replication identifies FANCM as a likely high grade serous ovarian cancer susceptibility gene.

We analyzed whole exome sequencing data in germline DNA from 412 high grade serous ovarian cancer (HGSOC) cases from The Cancer Genome Atlas Project and identified 5,517 genes harboring a predicted deleterious germline coding mutation in at least one HGSOC case. Gene-set enrichment analysis showed enrichment for genes involved in DNA repair (p = 1.8×10-3). Twelve DNA repair genes - APEX1, APLF, ATX, EME1, FANCL, FANCM, MAD2L2, PARP2, PARP3, POLN, RAD54L and SMUG1 - were prioritized for targeted sequencing in up to 3,107 HGSOC cases, 1,491 cases of other epithelial ovarian cancer (EOC) subtypes and 3,368 unaffected controls of European origin. We estimated mutation prevalence for each gene and tested for associations with disease risk. Mutations were identified in both cases and controls in all genes except MAD2L2, where we found no evidence of mutations in controls. In FANCM we observed a higher mutation frequency in HGSOC cases compared to controls (29/3,107 cases, 0.96 percent; 13/3,368 controls, 0.38 percent; P=0.008) with little evidence for association with other subtypes (6/1,491, 0.40 percent; P=0.82). The relative risk of HGSOC associated with deleterious FANCM mutations was estimated to be 2.5 (95% CI 1.3 - 5.0; P=0.006). In summary, whole exome sequencing of EOC cases with large-scale replication in case-control studies has identified FANCM as a likely novel susceptibility gene for HGSOC, with mutations associated with a moderate increase in risk. These data may have clinical implications for risk prediction and prevention approaches for high-grade serous ovarian cancer in the future and a significant impact on reducing disease mortality.