Author: Ana Marta (Portugal)
Co-authors: Sara Sara Geada Batista, Ana Luísa Carvalho, Joaquim Murta, Jorge Saraiva, Rufino Silva, João Pedro Marques
Retinitis pigmentosa (RP) is the most common inherited retinal dystrophy. The changes are usually bilateral with a high degree of inter-eye symmetry. However, the disease exhibits an immense phenotypic variability and several atypical RP phenotypes have been described, including unilateral or asymmetric cases. One contributing factor is the genetic heterogeneity that characterizes RP. Additionally, a substantial number (30-50%) of RP cases remain genetically unsolved despite state-of-the-art genetic testing. Possible explanations include undetected/unknown genotypes (e.g. inappropriate genetic test selection; hypomorphic variants; deep intronic mutations; variants within non-coding regions or variants in genes that have not yet been associated with RP) or an incorrect clinical diagnosis (i.e. disease entities that mimic RP such as paraneoplastic retinopathy, inflammation, infection or autoimmune disease). Clinical/demographic and multimodal imaging patient characteristics may also impact clinical pre-selection for efficient genetic testing and patient counselling. With the development of new therapeutic options and a growing number of gene therapy clinical trials, the importance of deep phenotyping and genetic testing cannot be overemphasized. The purpose was to compare clinical/demographic, functional testing and multimodal imaging features between genetically solved and genetically unsolved non-syndromic retinitis pigmentosa (nsRP) patients.
Department of Ophthalmology, Centro Hospitalar e Universitário de Coimbra (CHUC); Clinical Academic Center of Coimbra (CACC); University Clinic of Ophthalmology, Faculty of Medicine, University of Coimbra (FMUC); Department of Ophthalmology, Centro Hospitalar e Universitário do Porto (CHUP); Instituto Ciências Biomédicas Abel Salazar (ICBAS); Department of Medical Genetics, Centro Hospitalar e Universitário de Coimbra (CHUC); University Clinic of Medical Genetics, Faculty of Medicine, University of Coimbra (FMUC); University Clinic of Pediatrics, Faculty of Medicine, University of Coimbra (FMUC)
Cross-sectional study conducted at the Retinal Dystrophies Clinic and Medical Genetics Unit of Centro Hospitalar e Universitário de Coimbra (CHUC), an IRD reference center and only Portuguese healthcare provider represented in the ERN-EYE. The IRD-PT registry was used to identify consecutive patients with non-syndromic RP and available genetic testing results. All patients with clinically suspected and/or genetically confirmed syndromic RP were excluded. Clinical/demographic, functional and anatomical features were compared between genetically unsolved (Group 1) and genetically solved (Group 2) patients. Genetic testing was clinically-oriented in all probands and coordinated by a medical geneticist from the Medical Genetics Unit of CHUC. Variants were classified in accordance with the American College of Medical Genetics and Genomics (ACMG). All variants classified as pathogenic (class V) or likely pathogenic (class IV) were further confirmed by Sanger sequencing. The diagnostic yield was calculated from the number of variants classified as pathogenic or likely pathogenic. Genetic counselling provided by a medical geneticist was granted to all subjects.
The study included 175 patients (350 eyes) of 146 families: 68 patients (59 families; 136 eyes) in group 1 and 107 patients (87 families; 214 eyes) in group 2. The average age at diagnosis was significantly higher in group 1 (p<0.001). Age of first symptoms <25 years (p<0.001), consanguinity (p=0.019), evidence for a particular inheritance pattern (p=0.044) and absence of indicators for phenocopies (p=0.007) were significantly more prevalent in G2. The onset of symptoms in childhood (OR=1.650; IC 95%=1.021-2.665), history of consanguinity in family (OR=1.821; IC95%=1.040-3.191) and absence of potential phenocopies (OR=2.689; IC95%=1.215-5.951) were predictors of a solved case in our nsRP cohort (x2(1)=16.101;p=0.001, R2Negekerke=0.064). No significant differences were observed on best-corrected visual acuity (p=0.099). The visual field index (VFI) was significantly higher (p=0.020) in group 1 than in group 2. VFI (AUC=0.688, p=0.037, 95%CI=0.518-0.857) was able to identify solved and unsolved cases with <53.50% as the optimal cut-off (75% sensitivity and 63% specificity). The frequency of atypical features on multimodal imaging did not differ between groups. On optical coherence tomography, the mean retinal layer thickness was higher (p=0.032) in group 1 than in group 2.
We found that significant clinical/demographic, functional, and multimodal imaging differences exist between genetically solved and unsolved nsRP cases. Overall, our study provides evidence that an older age of symptom onset, absence of consanguinity, absence for a particular inheritance pattern, presence of indicators for phenocopies, larger fields, a VFI>53.5% and a higher RLT negatively impact the diagnostic yield in patients with nsRP. Careful medical history taking and deep phenotyping have shown to impact the genetic diagnostic yield and prognosis in nsRP. Individual clinical/demographic, functional testing and multimodal imaging features should be considered when counselling patients about the probability of identifying disease-causing variants. Given the genotypic and phenotypic heterogeneity that characterizes nsRP, it is critical to obtain predictive models for different populations in order to increase the efficiency of genetic testing and provide accurate genetic counselling.