Phenotypic and Genotypic Characterization of RP1L1-Associated Retinopathy

Abstract

Purpose: Pathogenic variants in RP1L1 are associated with autosomal dominant occult macular dystrophy (OMD) and autosomal recessive retinitis pigmentosa (RP). In this study, we investigated the phenotypic and genotypic landscape of RP1L1-associated retinopathy in an ethnically heterogeneous cohort.

Methods: This multicenter cohort study retrospectively collected the following data: best-corrected visual acuity (BCVA), color fundus photograph (CFP), optical coherence tomography (OCT), short-wavelength fundus autofluorescence (SW-AF), and full-field electroretinography (ffERG). Patients were classified based on their clinical phenotype in OMD or RP. Atypical cases were analyzed separately and reappraised according to their clinical and genetic findings.

Results: This study included 20 patients (40 eyes) from 19 families: 12 (60%) with OMD, 4 (20%) with RP, and 4 (20%) atypical cases (3 “non-occult” macular dystrophy, 1 rod-cone dystrophy with vitelliform maculopathy). Autosomal dominant OMD was the most common phenotype, with one autosomal recessive OMD case identified. Autosomal recessive RP had the latest onset, best visual acuity, and highest refractive error. OMD BCVA declined by ∼0.5 lines/year over a median follow-up of 3.2 years.

Conclusions: Mutations in RP1L1 cause a spectrum of diseases, including autosomal dominant OMD, autosomal recessive OMD, and autosomal recessive rod-cone dystrophies, occasionally presenting with pseudovitelliform maculopathy.

The retinitis pigmentosa-1-like-1 (RP1L1) gene encodes a component of the photoreceptor axoneme that is assumed to participate in the morphogenesis and maintenance of photoreceptor outer segments.^1,2 The RP1L1 gene spans approximately 105 kilobases on chromosome 8p23.1, comprising 4 exons, 3 of which contain the coding sequence.¹ Its protein product, comprising 2400 amino acids, features several structural domains: an N-terminal sequence, with homology to RP1, including doublecortin (DCX) domains crucial for microtubule stabilization; a signature RP1 domain, likely essential for photoreceptor function; and a C-terminal repeat region, which is believed to be involved in protein interaction of the photoreceptor axoneme.^1,3 

Despite RP1L1 being expressed both in rod and cone photoreceptors, according to studies on monkey retinas, RP1L1-associated retinopathies generally result in either a macular or a rod-cone disease.^4,5 The most common phenotype linked to pathogenic variants in this gene is occult macular dystrophy (OMD; MIM #613587), first reported by Miyake et al. in 1989 and later attributed to pathogenic variants in RP1L1 in 2010.^4,6 OMD is an autosomal dominant macular dystrophy with normal fundus appearance and reduced macular function, as documented by multifocal electroretinogram. Interestingly, full-field electroretinogram (ffERG) responses are generally within normal limits, although OMD cases associated with biallelic missense variants in RP1L1 may result in a mild cone dystrophy.⁷ The first and most common variant associated with OMD is the c.133C>T p.(Arg45Trp), affecting the first doublecortin domain. However, most of the other variants resulting in OMD are found in a hotspot region between amino acid residues 1196 and 1201, a region of unknown protein domain,⁸ highlighting the significant gaps in our current understanding of RP1L1.⁵ 

The “classic” forms of OMD are particularly prevalent in East Asia, where their imaging characteristics and natural history are the subject of a multicenter ethnicity-based study (EAOMD).^9–13 Interestingly, whereas OMD has generally been associated with dominant pedigrees among Asians, sporadic cases seem to be more frequent in Europe.^14,15 This might be due to undiagnosed cases, where individuals maintain good visual acuity with minimal changes on optical coherence tomography (OCT),¹⁶ or due to incomplete penetrance of the disease. 

Alternatively, variants in RP1L1 can cause recessive rod-cone dystrophy (retinitis pigmentosa [RP] 88, MIM #618826), typically resulting from the association of two null variants, although exceptions exist.^14,15 Last, in a recent work, Noel and McDonald documented a series of non-OMD, non-RP phenotypes, including pseudovitelliform maculopathies with or without associated generalized rod-cone dysfunction on ffERG. A few additional case reports corroborated the hypothesis of variants in RP1L1 as a cause of macular dystrophies with vitelliform lesions.^17–19 Still, scant data are available about the retinal phenotypes associated with RP1L1, particularly in European or multiethnic cohorts, as well as their genotypic correlations and natural history. 

Methods

Subjects

This was a retrospective cohort study including patients carrying at least one pathogenic variant or a variant of uncertain significance (VUS) in the RP1L1 gene, reflecting its dual inheritance pattern (autosomal recessive and autosomal dominant), which allows for the possibility of multiple variants contributing to the phenotype.²⁰ VUSs were included if their potential relevance to the phenotype was supported by available clinical and genetic data, and the phenotype was not explained by other gene variants. All participants were followed at the reference centers for rare diseases of the Quinze-Vingts Hospital (REFERET, Paris, France) and Centre Hospitalier Universitaire of Rennes, France, underwent genetic analysis and signed a written informed consent. This study adhered to the principles outlined in the Declaration of Helsinki and received approval from a national ethics committee (CPP Ile de France V, Project number 06693, N°EUDRACT 2006-A00347-44, December 11, 2006). 

Medical records of included patients were reviewed to gather the following data: best-corrected visual acuity (BCVA) measured with a standard Early Treatment of Diabetic Retinopathy Study (ETDRS) chart and converted to the logarithm of the minimum angle resolution (LogMAR); color fundus photograph (CFP); OCT acquired with a Spectralis (HRA + OCT) device (Heidelberg Engineering, Heidelberg, Germany) and short-wavelength fundus autofluorescence (SW-AF) as well as near-infrared fundus autofluorescence (NIR-AF) acquired with HRA2 (Heidelberg Engineering, Heidelberg, Germany); and ffERG recorded according to the standards of the International Society for Clinical Electrophysiology of Vision (ISCEV).²¹ The standard ETDRS grid was superimposed onto fovea-centered OCT raster scans using the Heidelberg Eye Explorer software to generate and annotate data on central subfield thickness (CST). 

The “baseline” visit, used for reporting cross-sectional data, was determined as the earliest examination that included BCVA measurements and at least one OCT and SW-AF acquisition. On the other hand, BCVA records from earlier examinations without imaging data were also included in the longitudinal analyses. 

Phenotype Grading

Each study subject was categorized based on his/her clinical phenotype, determined by integrating information collected from all available imaging. These phenotypes included: (i) OMD, or Miyake disease, characterized by typical mild ellipsoid zone (EZ) and interdigitation zone (IZ) alterations on OCT, without obvious fundus or SW-AF abnormalities¹⁶; (ii) RP, if presenting a consistent clinical phenotype¹⁵; (iii) atypical RP1L1-retinopathy, when the phenotype did not clearly align with the two aforementioned entities or matched rare alternative presentations described in the literature.⁵ Patients exhibiting the OMD phenotype were further classified in accordance with the OCT staging system proposed by Nakamura and colleagues: stage I, with no visual symptoms and minimal structural changes; stage II, with undetectable IZ with blurred and dome shaped EZ; and stage III, flat EZ.¹⁶ A functional classification was pursued separately by reviewing the electrophysiological records of each patient. 

Genetic Analysis

Genetic testing was conducted using a Next-Generation Sequencing (NGS) approach, with identified variants subsequently confirmed through direct Sanger sequencing, as described elsewhere.²² Homozygous variants were confirmed by checking NGS coverage of the specific variant regions, which excluded the possibility of a heterozygous deletion on one allele. 

DNA samples were extracted from peripheral blood leukocytes from all patients for all exons and flanking exonic regions of the RP1L1 (RefSeq NM_178857.6) gene. In one patient (CIC13807), Whole-Genome Sequencing (WGS) was used with a trio-based approach to investigate structural variants, such as the large intragenic deletion identified in this study, and to document the phase of the variant(s). Detailed protocols will be delivered on request. A de-identification number (e.g. CICXXXXX, CIC standing for Clinical Investigation Center) was assigned to all patients who underwent a blood draw. Novel variants identified in this study were interpreted based on the American College of Human Genetics (ACMG) classification.²³ 

Statistical Analysis

Continuous variables were presented as mean ± standard deviation (SD) or median and interquartile range (IQR), whereas categorical variables were expressed as frequency and percentages. The Shapiro-Wilk test was used for normality testing of continuous variables. Based on the necessity to account for inter-eye correlations, comparisons between groups were made through either the ANOVA test or linear mixed models (LMMs). 

Baseline BCVA was analyzed using an LMM where the phenotype was considered as the fixed factor and data were nested between patients. Changes in BCVA over time were estimated using separate models for the different phenotypes. Disease duration was included as the fixed factor, and the intercept was allowed to vary randomly across individual patients and each eye within patients. For statistical purposes, BCVA measurements recorded as “counting fingers” were calculated as 1.98 LogMAR in our models. The significance threshold was set at alpha < 0.05 for all analyses. All tests were performed with R studio (version 2023.03.0+386). 

Results

Twenty patients (9 women, 45%) from 19 families were included in the study, for a total of 40 eyes. Our cohort was ethnically diverse, comprising nine (45%) patients of European descent and five (25%) subjects of African descent, based on their country of origin. Ethnic background information was unavailable for six patients. Based on their clinical phenotype, 12 (60%) patients were classified as OMD and 4 (20%) subjects as RP. The remaining four (20%) patients exhibited atypical phenotypes that did not conform to the standard OMD or RP classifications. Individual-level and pooled patients’ demographics, clinical, imaging, and ffERG findings are reported in Tables 1 and 2, respectively. Both patients with OMD and patients with atypical presentations exhibited variable inheritance patterns, including autosomal dominant, autosomal dominant with incomplete penetrance, and autosomal recessive pedigrees. In contrast, all RP cases consistently displayed an autosomal recessive mode of inheritance (Supplementary Fig. S1). 

Table 1.

 

Patients’ Demographics, Clinical, Genetic, and Functional Findings

 

Table 2.

 

Pooled Patients’ Demographics and Clinical Findings

 

The mean baseline age was 42.5 ± 19.4 years, and was not statistically different between the groups (P = 0.35). Common presenting symptoms were visual acuity loss, photophobia, and dyschromatopsia in the OMD group, nyctalopia and peripheral visual field constriction in the RP group, and a variable combination of these symptoms in patients with atypical presentations (see Table 1). Except for one patient who was unable to recall a specific age of onset, the mean age of initial symptoms was 30.8 ± 17.2 years, without a statistically significant difference among the 3 groups (P = 0.927). This resulted in a mean disease duration of 31.4 ± 13.1 years (range = 0.6–46.8 years) at the last examination visit. 

The median baseline BCVA was 0.6 logMAR (IQR = 0.4–0.8), roughly corresponding to 20/80 Snellen, with significantly better BCVA observed in patients with RP compared with patients with OMD (P = 0.0174), and similar BCVA in atypical cases (P = 0.505; Supplementary Table S1). The rate of BCVA change over time was assessed independently within each group, using eye-level data from 18 eyes of 9 patients with OMD (62 observations over a median follow-up of 3.2 years, IQR = 0.6–5.8), 4 eyes of 2 patients with RP (34 observations over follow-ups of 24.1 and 23.0 years), and 6 eyes of 3 patients with atypical presentations (18 observations over follow-ups of 0.8, ∼10, and ∼28 years, respectively; Supplementary Fig. S2). 

Our estimates indicate that both patients with OMD and those with atypical presentations experienced a loss of approximately 1 ETDRS line every 2 years (both P < 0.01), whereas BCVA remained stable over more than 20 years in the RP group (P = 0.99; Supplementary Table S2). Conversely, we observed a statistically significant but not clinically meaningful thinning of the CST over time in OMD and patients with RP; no such changes were detected in those with atypical phenotypes (Supplementary Table S3). Moreover, no notable qualitative OCT and SW-AF changes were observed across all patients over a median follow-up of 2.5 years (IQR = 1.5–3.4 years). 

For 17 patients, an ffERG was available. All examinations were performed within 7 months of the baseline visit, with the exception of 1 case conducted a year later and 2 cases within 3 years. Overall, there were eight subjects with ffERG responses within normal limits, four with cone-system dysfunction (3 of them being mild), two with a dysfunction of both the cone and rod photoreceptors, and three with undetectable responses. 

The next section will provide a detailed report of patients’ genetic, clinical, imaging, and electroretinographic findings stratified by phenotype. 

Occult Macular Dystrophy

Among the 3 phenotypes, OMD was the most common, affecting 12 patients out of 20 (60%) from 11 families. The c.133C>T p.(Arg45Trp) was the most prevalent variant in the OMD group (9/12 patients, 75%), whereas the remaining patients carried known variants located in the second mutational hotspot.^8,9 All patients were heterozygous, except for CIC08179, who harbored the c.133C>T p.(Arg45Trp) variant homozygously. This patient was born to consanguineous parents, although neither of them nor his uninvestigated siblings complained of any symptoms. Intriguingly, CIC08179 experienced the earliest disease onset in the OMD group at 9 years of age, in contrast with the group's overall mean onset at 31.5 ± 19.3 years. Mild-to-moderate myopia was the most common refractive error in OMD, with a median spheric equivalent of −1.25 diopter (D; IQR:[−3.50] − [+0.25]). BCVA was reduced in all patients, the median being 0.6 LogMAR (IQR = 0.4–0.8). On SW-AF, subtle and symmetric macular hyperautofluorescent changes could be detected in six(50%) patients, likely due to the window defect related to loss or disruption of the outer retinal photoreceptors’ bands. All OMD eyes were classified according to Nakamura et al. based on their OCT findings.¹⁶ In detail, just 1 eye was classified as stage I, 19 eyes as stage II, and 4 eyes as stage III. Last, ffERG was available in 9 of 12 patients. As expected, most patients exhibited normal waveforms with both amplitudes and implicit times within normal limits. However, 2 patients (patients CIC03268 and CIC09080) exhibited mild cone dysfunction with delayed 30 hertz (Hz) flicker period. In addition, patient CIC09080, who harbored the p.(Ser1198Phe), also showed reduced photopic amplitudes. Last, patient CIC08179, homozygous for the p.(Arg45Trp) variant, showed severely reduced cone-driven responses, with a milder involvement of dark-adapted responses (Fig. 1). 

Figure 1.

 

Imaging and electroretinography findings in occult macular dystrophy (OMD). (A) For reference, short-wavelength autofluorescence (SW-AF), optical coherence tomography (OCT), and full-field electroretinography (ffERG) of a healthy control are displayed. (B) Patient CIC10507, who carries the “classic” p.(Arg45Trp) mutation at a heterozygous state, shows a subtle increase in parafoveal autofluorescence on SW-AF, blurred interdigitation zone, and disrupted ellipsoid zone (stage III) on OCT, and normal ffERG responses. (C) Patient CIC08179, carrying the same mutation but in a homozygous state, displays similar imaging findings, but with severely altered photopic waveforms and reduced scotopic amplitudes on ffERG. (D) Patient CIC09080, carrying the p.(Ser1198Phe) mutation, exhibits milder OCT changes (stage II) and mild cone dysfunction on ffERG. Peak-to-peak times for the 30 Hz flicker were A 32.9 ms, B 33 ms, C 33 ms, and D 32.6 ms, respectively.

Retinitis Pigmentosa

Four patients in this cohort showed a typical RP phenotype with autosomal recessive inheritance (Fig. 2). Both patients CIC12592 and CIC13415 carried two null variants (p.(Trp374*); p.(Glu567Glyfs*13) and p.(Arg658*); p.(Gln838*), respectively), three of which, to our knowledge, had never been published. Patient CIC12882 was compound heterozygous for the c.329C>T (Pro110Leu) missense variant and c.3955_3956ins86 p.(Ala1319Glyfs*33), a long frameshift insertion. Last, patient CIC10834 was homozygous for the same c.329C>T p.(Pro110Leu) variant. This patient had high myopia, with a bilateral spherical equivalent of −10 D, and was the oldest of 10 siblings, all asymptomatic except for 1 sister who also had high myopia, as far as she could recall. Conversely, among the remaining three patients with RP, one had moderate myopia, one was hyperopic, and for one patient with pseudophakia we could not retrieve the natural refraction. Patient CIC10834 also presented the earliest symptoms at the age of 6 years, in contrast with the other subjects, in which RP manifested at a mean age of 44.7 ± 10.1 years. As previously stated, the RP group had the best BCVA, with a median of 0.20 LogMAR (IQR = 0.03–0.38). In contrast, they had the worst ffERGs, with undetectable responses under all stimulus conditions. 

Figure 2.

 

Multimodal imaging of RP1L1-associated retinitis pigmentosa (RP). Patient CIC10834 displays typical imaging findings, with a central hyperautofluorescent ring visible on both short-wavelength autofluorescence (SWAF) and near-infrared autofluorescence (NIR-AF) imaging, peripheral loss of autofluorescence, and measurable ellipsoid zone (EZ) width on optical coherence tomography (OCT). In contrast, patient CIC12592 lacks a clear hyperautofluorescent ring, showing only diffuse loss of autofluorescence in the periphery. OCT reveals an epiretinal membrane and drusenoid lesions below the retinal pigment epithelium temporal to the macula. Patient CIC12882 exhibits an irregularly shaped hyperautofluorescent ring, better seen with NIR-AF. Finally, although patient CIC13415 has a regular hyperautofluorescent ring, OCT shows an asymmetrically constricted EZ with irregular thickening, most notably in the fovea.

SW-AF showed characteristic findings of concentric hypoautofluorescence due to the loss of retinal pigment epithelium (RPE). A hyperautofluorescent ring was present in three patients, one of which had an irregular shape and was better observed on NIR-AF than SW-AF (see Fig. 2). On OCT scans, all patients showed a measurable EZ width, whereas a remarkable thickening of this layer was present subfoveally in patient CIC13415. Last, patient CIC12592 also displayed drusenoid changes in both eyes, although his older age made it difficult to attribute them to his underlying RP1L1-related condition. 

Atypical Cases of RP1L1-Retinopathy

Four unrelated patients presented with atypical phenotypes that diverged from standard OMD and RP (Fig. 3). Two of them were heterozygous for the common c.133C>T p.(Arg45Trp) variant. Specifically, patient CIC10408 exhibited a well-demarcated foveal cavitation on OCT,²⁴ with a noticeable optic gap and irregular thickening of the surrounding EZ, precluding the use of the OMD staging system for classification. Patient CIC09380, although meeting the criteria for stage III OMD on OCT, exhibited a distinctly atypical phenotype characterized by sharply defined foveal hyperautofluorescence on SW-AF, a feature that contrasts with the subtler autofluorescence changes typically observed in patients with OMD. 

Figure 3.

 

Atypical presentations of RP1L1-retinopathy. Patient CIC09380 exhibits markedly increased central autofluorescence on short-wavelength autofluorescence (SW-AF), with an extinguished interdigitation zone and blurred ellipsoid zone (EZ) on optical coherence tomography (OCT). Patient CIC11582 displays abnormal central hyperautofluorescence with parafoveal hyperautofluorescent lesions, corresponding to severe foveal thinning and paracentral retinal pigment epithelium migration. Patient CIC10408 demonstrates a loss of the normal foveal hypoautofluorescence on SW-AF and presents an “optic gap” appearance surrounded by a thickened EZ on OCT. Last, SW-AF for patient CIC13807 reveals abnormal central hyperautofluorescence that corresponds to a subfoveal vitelliform lesion on OCT. In this case, the macular EZ is also markedly thickened.

A third male patient (patient CIC11582) carried the c.3598G>T p.(Gly1200Cys) variant in heterozygosity. Although this specific nucleotide change has not been previously published, its corresponding amino acid change has already been associated with OMD.⁸ Alongside severe foveal thinning, this patient exhibited speckled pigmentary abnormalities in the macular region that were hyperautofluorescent on SW-AF and hyper-reflective on OCT. 

Last, patient CIC13807 carried two likely pathogenic variants: a missense variant resulting in the formation of a stop codon (c.1130C>G p.(Ser377*); inherited from the mother) and a large intragenic deletion of exon 3 (c.609+346_751+791del; inherited from the father). On fundus examination, this patient had bilateral vitelliform maculopathy (Fig. 4), and detailed inspection of the retinal periphery revealed bilaterally hypopigmented retina with minimal bone-spicule-like pigment migration in the left eye (Supplementary Fig. S3). OCT demonstrated subfoveal hyper-reflective material below the RPE alongside with a diffuse EZ thickening. The vitelliform material was inhomogenously hyperautofluorescent on SW-AF and hypoautofluorescent on NIR-AF. Notably, patients in this group were either emmetropic or hyperopic, except for patient CIC09380 with high myopia on both eyes. The median BCVA was 0.8 LogMAR (IQR = 0.70–1.98), with a median age of onset of 29.5 years (range = 20–37 years). In this cohort, only patient CIC09380 had normal ffERG responses, whereas the patients with foveal cavitation and pigmentary abnormalities exhibited mild cone dysfunction, and the patient with vitelliform maculopathy showed greater rod than cone dysfunction. Electrooculography performed for patient CIC13807 revealed reduced Arden ratios (1.13 in the right eye and 1.22 in the left). These results have to be interpreted considering the abnormal ffERG responses, photoreceptor dysfunction explaining the abnormal electro-oculogram (EOG) light rise although additional RPE dysfunction cannot be ruled out. The distribution of RP1L1 variants identified in this study, along with those associated with RP and atypical presentations reported in the literature, is summarized in Figure 5. 

Figure 4.

 

Multimodal imaging and full-field electroretinography (ffERG) of CIC13807. Color fundus photograph reveals bilateral yellowish lesions in the fovea, which appear hyperautofluorescent on short-wavelength autofluorescence (SW-AF) – second row – and hypoautofluorescent on near-infrared autofluorescence (NIR-AF) – third row. On optical coherence tomography (OCT), the sub-RPE material is as reflective as the ganglion cell layer, exhibiting a flat, domed aspect in the right eye and more irregular contours in the left eye. On full-field electroretinography (ffERG) testing, this patient shows greater rod than cone dysfunction, with markedly reduced scotopic amplitudes and slightly reduced and delayed photopic responses. Due to the patient’s severe photophobia, numerous blinking artifacts are apparent on the light-adapted 30 Hz flicker recordings.

Figure 5.

 

Schematic representation of the variants in the RP1L1 gene observed in our cohort (upper section). Variants associated with typical and “atypical” occult macular dystrophy (OMD) are in red, retinitis pigmentosa (RP) are in italics, and pseudovitelliform maculopathy is in bold. A literature review (PubMed: accessed August 30, 2024) was conducted to identify all published variants in the RP1L1 gene associated with autosomal recessive RP or pseudovitelliform maculopathies and related phenotypes (lower section). With the exception of p.(Pro110Leu), additional variants associated with OMD in literature are not included in this figure. †Homozygous or compound heterozygous with a missense variant. #Homozygous or compound heterozygous with a null variant.

Discussion

This study was designed to explore the natural history of RP1L1-related retinopathy and uncover novel genotype-phenotype correlations. We classified our patients into three groups to compare their clinical courses, imaging characteristics, and genetic findings. To ensure accurate interpretation of our results, patients with atypical presentations will be reappraised based on their findings before proceeding to the discussion. 

In our cohort, autosomal dominant OMD was caused by missense variants leading to amino acid substitutions located in the first doublecortin domain (p.(Arg45Trp)) or in the second mutational hotspot (p.(Gly1200Val) and p.(Ser1198Phe)). The same variants were also found in three out of four patients with an atypical presentation. The consistent genotype, their similar rate of BCVA loss, and the fact that pathological changes were confined to the macula justified their categorization as “atypical OMD.” Although the small sample size for this group limits generalizability, the estimated BCVA loss aligns with reliable findings from the OMD cohort and further supports their inclusion in this category. In contrast, patient CIC13807 demonstrated a generalized photoreceptor dysfunction alongside vitelliform deposits. This patient matched an existing phenotype in the literature and was thus designated as “rod-cone dystrophy with pseudovitelliform maculopathy.”⁵ 

As a result of this categorization, OMD emerged as the most common phenotype, accounting for 75% of cases (60% “classic” and 15% “atypical”), followed by 25% rod-cone dystrophies (4 patients with typical RP and 1 patient with pseudovitelliform maculopathy). Patients with OMD were designated as atypical for the following reasons: (1) non-occult maculopathy in patient CIC11582, who presented with complete foveal outer retinal atrophy and pigmentary changes, (2) atypical OCT findings in patient CIC10408, and (3) evident changes on SW-AF due to severe parafoveal EZ disruption in patient CIC09380. Nonetheless, both genotype and clinical course in these patients were consistent with OMD, suggesting that in severe and advanced cases, the current OCT staging may be inadequate, and the “occult” state of the disease may be lost over time.^5,16,25 Considering typical and atypical autosomal dominant OMD cases together, we estimated disease penetrance as the ratio of affected individuals to carriers. In sporadic pedigrees, parents were excluded as carriers unless molecular testing confirmed their status, as OMD-causing variants frequently arise in mutational hotspots and are presumed de novo. In pedigrees with multiple affected offspring and healthy parents, one parent was assumed to be an obligate carrier. Using this method, we estimated a penetrance of 85%, aligning with previous reports from German and Asian cohorts but intriguingly differing significantly from the English cohort reported by Davidson and colleagues.^4,14,15,26 Nevertheless, this estimation may be overestimated in large families with fewer affected offspring than anticipated or underestimated due to age-related penetrance.⁴ 

This aligns with the clinical observation that patients typically reported their first symptoms in their 30s, with OMD showing the widest age range at onset—from the first to the seventh decade—and atypical RP1L1-retinopathies tending to manifest earlier. Both typical and atypical OMD presented with a marked reduction in visual acuity (VA), followed by a gradual deterioration of approximately 1 line every 2 years, whereas individuals with RP had nearly normal BCVA, which remained stable over the follow-up. Myopia was the most common refractive error, being more pronounced in patients with RP than patients with OMD. 

Focusing on the genetic findings, both classic and atypical OMD resulted from amino acid substitutions located either in the first doublecortin domain or in the second mutational hotspot. These mutations typically caused autosomal dominant OMD with normal retinal function or mild cone dysfunction, with the notable exception of patient CIC08179. This patient, affected by an uncommon autosomal recessive form of OMD, harbored the classic p.(Arg45Trp) variant in a homozygous state and exhibited generalized cone-rod dysfunction on ffERG. Notably, his parents were asymptomatic. Zobor et al. described a similar case of autosomal recessive OMD with the same homozygous p.(Arg45Trp) variant and healthy parents, although with normal ffERG values despite being roughly the same age of our patient.¹⁴ These findings suggest that, whereas the possibility of a later onset should always be considered, part of the variable expressivity of OMD may be linked to reduced penetrance. This is further exemplified in family FQV01, where the proband's healthy father also carried the p.(Gly1200Val) in heterozygosity. 

In a compelling contrast, another patient (patient CIC10834) carried the p.(Pro110Leu) missense variant in a homozygous state, previously associated with OMD,¹⁵ but developed RP instead, setting her apart from the rest of her asymptomatic family. Interestingly, this variant was also present in trans with a long frameshift insertion between nucleotides 3955 and 3956 in patient CIC12882, who also suffered from RP. Homozygous frameshift insertions at this position have been previously associated with RP,²⁷ further reinforcing the p.(Pro110Leu) variant's unique capacity to cause both OMD and RP. A similar behavior has been reported also for the p.(Pro101Thr) BEST1 variant, which can result in autosomal dominant macular dystrophies with incomplete penetrance as well as autosomal recessive bestrophinopathy.²⁸ 

Patients with RP1L1-related RP generally displayed a “classic” appearance, except for macular pigmentary abnormalities in patient CIC12592 and a thickened EZ in patient CIC13415. Similarly, patient CIC13807, who carried a stop variant in exon 4 in trans with a large intragenic deletion of exon 3, exhibited bilateral vitelliform maculopathy with thickened EZ associated with rod-cone dysfunction on ffERG. Recent studies, including cases reported by Noel and McDonald, have described similar manifestations with variants in RP1L1. They highlighted a mother and daughter with a heterozygous missense variant in exon 4, affected with adult onset foveomacular vitelliform dystrophy (AOFVD).⁵ These patients showed bilateral foveal “mottling” corresponding to a small serous retinal detachment and photoreceptor outer segment elongation on OCT. Another patient, homozygous for the p.(Tyr405*) stop variant in exon 4, displayed similar alterations along with severe rod-cone dysfunction on ffERG.⁵ Manayath et al. reported a 39-year-old male patient with a homozygous base pair deletion (p.Lys1138Serfs*24) in exon 4 of RP1L1, presenting with AOFVD and mild cone dysfunction on ffERG.¹⁷ Furthermore, Takahashi et al. described a case of dominant bilateral subfoveal serous retinal detachment associated with the p.(Ser1199Pro) variant.²⁹ In our cohort, patient CIC11582, an atypical patient with OMD who was heterozygous for the p.(Gly1200Cys) variant, manifested bilateral pigmentary abnormalities with foveal atrophy, which could be suggestive of a longstanding case of AOFVD, although age-related or pachychoroid-related changes cannot be definitely excluded. In summary, our findings indicate that amino acid substitutions in the second mutational hotspot of RP1L1 can occasionally result in pseudovitelliform maculopathies and related phenotypes in addition to autosomal dominant OMD. Conversely, null variants in exons 3 and 4 result in a similar autosomal recessive phenotype with generalized photoreceptor dysfunction. 

In addition to the retinal phenotype, we classified patients also based on their ffERG findings. Five out of 12 patients affected by autosomal dominant OMD and 2 out of 3 patients with atypical OMD demonstrated mild cone dysfunction. By contrast, recessive cases were associated with worse retinal function. This was particularly evident when patients developed typical RP, but it also held true for patient CIC08179 with autosomal recessive OMD and patient CIC13807 with vitelliform material plus generalized photoreceptor dysfunction. Thus, the term “maculopathy” does not fully capture the condition of some patients, who would be more accurately described as having “RP1L1-retinopathy.” 

This study is limited by its retrospective design, which is responsible for uneven follow-up lengths and data collection inconsistencies. Moreover, the limited sample size and the disproportion among patient groups hindered our statistical analyses and the robustness of our conclusions. Although the OMD group provided reliable estimates over a median follow-up of 3.2 years, and patients with RP demonstrated stable BCVA over extended follow-up periods, a larger cohort might reveal nonlinear trends, including potential floor effects in advanced disease. On the other hand, our approach allowed for a careful reappraisal of patients presenting with atypical phenotypes and highlighted the complex spectrum of RP1L1-associated retinopathy, encompassing maculopathies (OMD and pseudovitelliform phenotypes), cone dystrophy, cone-rod dystrophy, and rod-cone dystrophy with or without associated pseudovitelliform maculopathy. Moreover, we emphasize the importance of ffERG in accurately phenotyping these patients, as it may reveal more severe retinal dysfunction than expected. Further multicenter studies gathering more “atypical” forms of RP1L1-associated retinopathy will elucidate their natural history and genetic determinants influencing phenotypic manifestations of the disease. 

Acknowledgments

The authors are thankful to the patients and family members who participated in this study. DNA samples included in this study originate from NeuroSensCol DNA bank, dedicated for research in neurosensory disorders (PI: I Audo, partner with CHNO des Quinze-Vingts, Inserm and CNRS). 

Supported by IHU FOReSIGHT (ANR‐18‐IAHU‐0001) French state funds managed by the Agence Nationale de la Recherche within the Investissements d'Avenir program. 

Disclosure: A. Antropoli, None; L. Bianco, None; X. Zanlonghi, None; A. Benadji, None; C. Condroyer, None; A. Antonio, None; J. Navarro, None; C.-M. Dhaenens, None; J.-A. Sahel, None; C. Zeitz, None; I. Audo, Novartis Pharma (Rueil Malmaison, France) (C), and Janssen Pharmaceuticals (Headquarters, Leiden, Netherlands) (C)