Retinocytoma: Clinical and Genetic Characteristics in 16 Pediatric and Adult Cases

Abstract

Introduction

Retinocytomas are rare benign intraocular tumors that may mimic spontaneously regressed retinoblastoma (Rb).

Materials and Methods

This was a retrospective monocentric study of patients with retinocytoma in a French tertiary ocular oncology center, with an inclusion period from January 1999 to January 2024.

Results

Sixteen patients with retinocytoma were identified, and 1,351 Rbs were diagnosed during the same 25-year period. Age at diagnosis ranged from 11 months to 75 years (mean 28.1 years). Thirteen cases were asymptomatic, while three presented with floaters, decreased visual acuity, or strabismus. Clinical presentation was a whitish or grayish retinal tumor with a translucent (87%) and/or fragmented appearance (76%) at diagnosis. A proportion of lesions were surrounded by atrophy (52%) and/or pigmentation (30%). Eleven patients had unilateral retinocytoma (one of which was multifocal), and five had bilateral lesions. Eight patients had a known family history of Rb. Patients were followed regularly and underwent genetic counseling. The eight patients with a family history had a germline pathogenic variation of the RB1 gene. None showed malignant transformation during follow-up (mean 79.5 months, median 35.5 months).

Conclusion

The presentation of retinocytoma is most often asymptomatic. Thus, diagnosis may be delayed to far later ages than expected with Rb. Retinocytoma may be associated with germline pathogenic variants of the RB1 gene, and follow-up is recommended due to rare but possible malignant transformation.

Introduction

Retinocytoma is a rare benign retinal tumor, reported in 1977 by Gallie et al. [1, 2], who proposed coining this term for retinal tumors sharing features with retinoblastoma (Rb) but showing signs of spontaneous regression. Subsequently, several case series have provided a more precise clinical description of these lesions and their typical or atypical presentations [3–14].

In 1983, Margo et al. [13] identified in 5 enucleated eyes retinocytoma areas within Rbs and showed that they were composed of photoreceptor differentiation foci (rosettes), monomorphic round nuclei, an abundant eosinophilic fibrillar stroma, and an absence of mitotic activity. It has also been demonstrated that inactivation of both alleles of the RB1 gene is present in retinocytomas, as in Rb [15].

Several cases of transformation of retinocytoma into Rb have also been described in recent years [16–19], but the causes of these malignant transformations remain unclear [20, 21]. To our knowledge, only a few French cases of retinocytomas have been published [22, 23]. Here, we reviewed the Rb French reference center database over 25 years to identify patients with retinocytomas and describe their clinical characteristics, management, outcome, and genetic implications.

Methods

The clinical records of all patients with the diagnosis of retinocytoma (spontaneously regressed/arrested Rb) managed by the ocular oncology department at Institut Curie, Paris, between January 1, 1999, and January 1, 2024, were reviewed. Diagnosis was based on ophthalmological examination with indirect ophthalmoscopy and fundus photography, conducted by senior ophthalmologists specialized in pediatric ocular oncology.

All Rb patients seen in our institution are included in a prospective database. Their parents or legal guardians agreed at diagnosis that their clinical data could be registered in a prospective database and used for publication. Patients or parents/legal guardians for minors have given their written informed non-objection for ulterior analysis of their data and publication. We used CONSORE [24], a computerized search tool of the patient database using the keywords retinocytoma and retinoma.

All the charts identified with the CONSORE search were reviewed, and all patients with a clinically inactive retinal tumor with features of an inactive Rb, such as translucent retinal mass, calcifications, retinal pigment epithelial alterations, surrounding chorioretinal atrophy, and possible vitreous seeds, were included for analysis.

The collected data included patient’s features, such as sex, age at diagnosis, presenting complaints, presence of family history of Rb; clinical features of the tumor including laterality, number of tumors, location (posterior pole, superior vascular arch, inferior vascular arch, superior, inferior, nasal, temporal), ophthalmoscopy features. Follow-up and evolution were analyzed. Genetic testing and the presence of germinal pathogenic variation of RB1 were also collected.

This manuscript was prepared in accordance with the CARE Checklist, which is provided as supplementary material (for all online suppl. material, see https://doi.org/10.1159/000548389).

Case Series

Over the inclusion period, 1.351 consecutive patients with Rb were diagnosed and managed in our institution. The CONSORE search identified 189 records with the keywords retinocytoma or retinoma, which, after manual chart analysis, resulted in 16 cases (1.2%) with retinocytoma, who had 24 tumors in 20 eyes.

Nine patients (56%) were male and 7 were female (44%). The mean age at diagnosis was 28.1 years (median 27 years, range 11 months–75 years). The patient demographics are shown in Table 1, and their detailed clinical characteristics are shown in Table 2.

Table 1.

Demographic and clinical data in patients with retinocytoma

​	Total, n = 16
Age, mean (median; range), years	28.1 (27; 1–75)
Sex, n (%)	​
Male	9 (56)
Female	7 (44)
Presenting complaint, n (%)	​
None	4 (25)
Blurred vision	1 (6)
Myodesopsia	1 (6)
Strabismus	1 (6)
Screening	9 (56)
Family history of Rb, n (%)	​
Present	8 (50)
Absent	8 (50)

Age values are mean (median, range), and results otherwise reported are numbers (%).

Table 2.

Detailed patients characteristics

Patient	Sex	Presenting complaint	Family history of Rb	Age at diagnosis, years	Follow-up, months	Affected eye	Focality	Location	Number of tumor(s)	RB1 mutation
1	M	Family screening	Yes	39	100	Bilateral	Multifocal	Temporal RE/posterior pole LE	1 RE (+1 coloboma) + 1 LE	Present
2	F	None	No	14	0	Left	Unifocal	Temporal superior	1	Absent
3	M	None	No	75	0	Left	Unifocal	Temporal superior	1	No genetic testing
4	F	Family screening	Yes	8	99	Bilateral	Multifocal	Posterior pole RE/temporal LE	1 RE + 1 LE	Present
5	F	Family screening	Yes	30	90	Bilateral	Multifocal	Temporal + temporal inferior RE/nasal superior LE	2 RE + 1 LE	Present
6	F	Family screening	Yes	34	0	Right	Unifocal	Inferior vascular arch	1	Present
7	M	Gorlin syndrome in sister	No	8	59	Right	Unifocal	Superior	1	Absent
8	M	None	No	65	0	Left	Unifocal	Nasal	1	Absent
9	F	Strabismus/nystagmus	No	1	253	Right	Multifocal	Temporal inferior + nasal superior	2 LE	Absent
10	M	Family screening	Yes	45	12	Bilateral	Multifocal	Posterior pole RE/2 temporal LE	1 RE + 2 LE	Present
11	M	Family screening	Yes	41	0	Right	Unifocal	Inferior vascular arch	1	Present
12	F	Blurred vision	No	5.5	252	Right	Unifocal	Posterior pole	1	Absent
13	M	Family screening	Yes	31	108	Bilateral	Multifocal	Posterior pole RE/inferior vascular arch LE	1 RE + 1 LE	Present
14	M	None	No	15	0	Right	Unifocal	Superior	1	Absent
15	M	Family screening	Yes	24	0	Right	Unifocal	Temporal inferior	1	Present
16	F	Myodesopsia	No	14	299	Left	Unifocal	Inferior	1	Absent

RE, right eye; LE, left eye; RPE, retinal pigment epithelium.

Thirteen patients (81%) were asymptomatic, with a diagnosis made after screening due to a family history of Rb in 8 cases (50%), 1 screening for Gorlin syndrome in a sister, and incidentally in 4 cases (25%). Other symptoms included decreased visual acuity, myodesopsia, or strabismus (shown in Tables 1, 2).

Eleven patients had unilateral retinocytoma (of the right eye in 7 patients [44%] and the left eye in 4 patients [25%]), and bilateral involvement was observed in 5 patients (31%). The equatorial retina was the most frequently affected area (67%) with variable distribution across different quadrants (Table 2). Twenty-three tumors in 15 patients underwent fundus imaging, enabling a precise clinical description. The tumors presented with a combination of 2–3 clinical characteristics. A translucent mass was observed in 87% (n = 20/23) of the tumors, 74% (n = 17/23) were calcified, and 26% (n = 6/23) were non-calcified. Pigment epithelial changes were found in 30% (n = 7/23) of the lesions, chorioretinal atrophy in 52% (n = 12/23), and vitreous lesions in 1 case (4%). Tumor features are described in Table 3 and illustrated in Figure 1.

Table 3.

Clinical features of the tumors

​	Parameters, n (%)
Focalitya, n (%)	​
Unifocal	10 (63)
Multifocal	6 (37)
Tumor location quadrantb, n (%)	​
Posterior pole	5 (21)
Superior vascular arcade	0 (0)
Inferior vascular arcade	3 (12.5)
Temporal	10 (42)
Nasal	3 (12.5)
Superior	1 (4)
Inferior	2 (8)
Clinical featuresc, n (%)	​
Translucent mass	20 (87)
Calcification	17 (74)
No calcification	6 (26)
RPE alteration	7 (30)
Chorioretinal atrophy	12 (52)
Vitreous lesion	1 (4)

Total t = 24 tumors in 20 eyes, clinical features analysis realized on 15 patients/23 tumors with quality fundus imaging available.

RPE, retinal pigment epithelium.

^aBased on 16 patients.

^bBased on 24 tumors.

^cBased on 23 tumors.

Fig. 1.

Clinical spectrum of retinocytomas. Tumors demonstrating intrinsic calcification and mild to extensive surrounding retinal pigment epithelial (RPE) and chorioretinal alterations (a–b). c Retinocytoma with intrinsic calcification and chorioretinal atrophy without RPE alterations.

Genetic counseling and analysis were conducted for 15 patients, in whom germline pathogenic variation of the RB1 gene was identified in 8 patients (53%), corresponding to those with a known family history. The 5 patients with bilateral retinocytomas were among the 8 patients with confirmed germline pathogenic variation. The mean follow-up period was 79.5 months [0–299], median 35.5 months. We did not record any malignant transformation among the 16 patients. One patient was treated with focal laser therapy for a neovascular process associated with the lesion.

Discussion

Most reports of retinocytomas have been conducted in North America, Asia, and a limited number in Europe, but only two case reports have been published on French patients (one of which is included in the present study). In our cohort, we found an estimated percentage of retinocytomas of 1.2% compared to all Rb cases evaluated over a 25-year study period. This proportion varies from 0.04% to 1% in Saudi Arabia, 3% in the USA, 5.3% in Taiwan, and 9% in China [3, 16, 25–27]. Discrepancies between these reported proportions of retinocytoma/Rb may be explained by differences in exhaustivity of local/national databases, registration of cases, screening strategies for families of Rb patients, study period, epidemiological variation, proportion of germline Rb variants in the population, or by the recruitment of tertiary centers, among others. The study by Xu et al. [27], with the highest reported proportion of retinocytomas compared to Rb, in a Chinese population, had a study period limited to 6 years, whereas all other reports analyzed a period of 24 years or more. Referral bias can also be cited, as a diagnosis of retinocytoma is likely to be made only in symptomatic cases or in individuals with a family history of Rb, and more rarely as an incidental finding, as described in our cohort, where only 4 of 16 cases were diagnosed incidentally.

In our study, the mean age at diagnosis was 28.1 years (median 27 years, range 1–75 years). It is higher than reported in other studies on retinocytomas: 13 years in a series by Shields et al. [16] (62 patients, range 0.4–88 years), 16 years in the series by Marback et al. [4] (5 patients, 2.5–27.5 years), and a median of 15 years in the series by Singh et al. [12] (17 patients, range 4–45 years). In our reference center, we ask parents to undergo an ophthalmologic exam after an Rb diagnosis is made on one of their children. Most retinocytoma diagnoses are made in adults because one of their children was diagnosed with Rb.

The clinical features observed in our series are consistent with previous studies. We described localized calcified vitreous seeds, which is an atypical presentation for retinocytoma but has already been described in the literature [8–10].

Fifteen of our 16 patients had genetic testing, and half of them had a germline RB1 pathogenic variation. In the single published series where the genetic status is explored, only 18 of the 62 patients had genetic testing, among whom 44% had a germline RB1 pathogenic variation [16]. The genetic changes underlying Rb and retinocytoma have been recently investigated [20, 21]. Presumably, both diseases require a biallelic pathogenic variation of the RB1 gene. It is suspected that retinocytomas remain quiescent until further pathogenic variations, probably involving other driver oncogenes [20] or the acquisition of instability or gene rearrangements [21]. The complete pathophysiology of retinocytoma transformation into Rb is not yet fully elucidated. Identification of a germline RB1 pathogenic variation in patients with retinocytoma has the same clinical implications regarding the risks of Rb in relatives, so specialized genetic counseling is essential to adapt their follow-up. Two Rb subtypes have been recently described [28], but little is known about retinocytoma biology. The analysis of aqueous humor cell-free DNA is clearly useful to study Rb biology [29], but, to our knowledge, there is not yet such experience in the case of retinocytoma: such analysis should be attempted, but it is uncertain whether it would contribute to retinocytoma diagnosis and subtyping, given that the high levels of cell-free DNA in aqueous or plasma of Rb patients result from the rapid multiplication and necrosis of Rb cells.

None of the patients in our study showed retinocytoma growth over a mean follow-up > 6 years. This risk has been described in several cohorts. Singh et al. [12] demonstrated transformation into Rb in one case (4%) with a median follow-up of 4 years. Abouzeid et al. [30] showed tumor growth or reactivation in 8 retinocytomas (16%) with a mean follow-up of 3.7 years. Shields et al. [16] followed 78 retinocytomas over 20 years and found 6 tumors with malignant transformation, with a mean period between diagnosis and transformation of 9.8 years.

As malignant transformation is very rare, preventive treatment of these lesions could be discussed on a case-by-case basis. While chemotherapy is typically not indicated for retinocytoma unless there is malignant transformation, laser has not yet been evaluated to potentially decrease the risk of transformation. It is currently known as a treatment for exudation or vascular complications. However, in most cases, careful monitoring is preferred. In cases without germline pathogenic variation of the RB1 gene, the risk of progression and malignant transformation is low, and a reasonable monitoring cycle would include clinical and imaging follow-up at regular intervals, depending on patient age and lesion characteristics.

Our study has limitations, including its retrospective nature and the fact that retinocytoma diagnosis is clinical, based on tumor characteristics and absence of progression over time. Absolute proof of diagnosis was impossible due to the prohibition on performing a tumor biopsy. Finally, the duration of follow-up is variable, mainly because a proportion of patients were followed by their local ophthalmologist, due to the long distance from our reference tertiary center.

Yet, the strengths of this study include the review of a large, exhaustive national database in an Rb reference center. All patients were examined by senior ocular oncologists, and the follow-up was based on iterative color fundus photographs, allowing for precise clinical descriptions. Finally, all but 1 patient underwent genetic investigations within our tertiary center or in affiliated academic hospitals.

In summary, retinocytoma is an inactive form of Rb, often diagnosed incidentally. Subjects diagnosed with retinocytoma should be offered genetic analyses and counseling similar to Rb patients. Their rare but possible risk of malignant transformation justifies regular lifelong follow-up. Collaborative international guidelines would be useful to recommend the frequency of ophthalmological follow-up and the modality of genetic explorations of retinocytoma patients.

Statement of Ethics

This study was performed in accordance with the Declaration of Helsinki. Written informed consent was obtained from the legal guardian of the individuals for publication of the details of their medical case and any accompanying images. Written informed consent was obtained from the individuals for publication of the details of their medical case and any accompanying images. This retrospective review of patient data did not require ethical approval in accordance with national guidelines.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

This study was not supported by any sponsor or funder.

Author Contributions

C.W. collected, analyzed, and organized the patient data, drafted the initial manuscript based on Karger guidelines and the CARE checklist, and contributed to its revision with the support of the other authors. A.M. and L.L.-Le Rouic supervised this study and the writing of this article. A.M., N.C., L.L.L.R., D.M., I.A., and F.D. took care of these patients. M.G.V. did the genetic counseling.

Funding Statement

This study was not supported by any sponsor or funder.

Data Availability Statement

The data that support the findings of this study are not publicly available due to patient confidentiality but are available from the corresponding author upon reasonable request.