Diagnosis, Classication, and Treatment of Retinoblastoma

Szilard Kiss, MD Yannek I. Leiderman, MD, PhD Shizuo Mukai, MD

Introduction

Retinoblastoma represents approximately 4% of all pediatric malignancies and is the most common primary intraocular cancer in children.1,2 Worldwide, it is estimated that there are 5000 to 8000 newly diagnosed cases of retinoblastoma yearly.2,3 In the United States, approximately 250 to 350 cases are diagnosed each year.2,4 The rst description of a child successfully treated with radiation for retinoblastoma was reported in 1903.5 Since that time, improvement in survival of retinoblastoma has been faster than any other cancer in children or adults. In 2003, retinoblastoma was the pediatric cancer with the highest survival.2 In the Western world, 99% of children will survive this cancer and over 90% will retain normal vision in at least 1 eye.2,3,6 Unfortunately, primarily to late detection of advanced retinoblastoma, in medically underdeveloped nations survival drops to approximately 50%.4 A genetic basis for the development of retinoblastoma was recognized over 70 years ago.2 In 1971, after noting differences in tumor development in patients with unilateral versus bilateral retinoblastoma, Knudson7 proposed his now famous two-hit model of tumorgenesis. The retinoblastoma gene (RB1) was the rst human cancer gene cloned and was the rst of a whole new class of human tumor-suppressor genes.2,3 The protein responsible for intraocular retinoblastoma is vital for terminal differentiation of many cell lineages, including differentiation of retinal progenitor cells.8 It is now recognized that the loss of normal functioning RB1 is an important step in the development of many adult nonocular cancers.2,3,9
INTERNATIONAL OPHTHALMOLOGY CLINICS Volume 48, Number 2, 135147 r 2008, Lippincott Williams & Wilkins

135

136

Kiss et al

Epidemiology

The cumulative lifetime incidence of retinoblastoma is 1 in 18,000 to 30,000 live births worldwide.10,11 In the United States, the incidence is estimated at 3.6 cases for each million children under the age of 15 years.6 The incidence of heritable retinoblastoma is remarkably constant among the various populations of the world, with no sex or race predilection and no signicant environmental or socio-economic predisposing factors.12 In contrast to heritable retinoblastoma, there are marked geographic differences in the incidence of nonheritable, unilateral retinoblastoma. Sporadic, unilateral retinoblastoma seems to occur more commonly in poorer, tropical, and subtropical regions of the world. Preliminary evidence suggests a possible viral etiology (human papilloma virus) or perhaps a diet decient in fruits and vegetables.13,14 Advanced paternal age has also been associated with new sporadic retinoblastoma gene mutations and sporadic retinoblastoma.1520 New germline RB1 mutations apparently arise more frequently during spermatogenesis.20,21 Dryja and colleagues16 observed that the proportion of new germline RB1 gene mutations arising from a paternally derived allele was approximately 85%. However, in cases with somatic mutations, the initial mutation occurs on paternal or maternal chromosomes with roughly equal frequency.16,20

Clinical Features

Approximately two-thirds of all retinoblastoma cases are unilateral and one-third of cases bilateral.4 Bilateral and familial cases of retinoblastoma have a germline mutation and are heritable. Patients with germline mutations are also at signicant risk for developing second and third nonocular tumors, a risk that increases throughout the lifetime.22 Retinoblastoma is a childhood disease with most cases being recognized by 3 years of age and 90% diagnosed by age the age of 5 years.23 The average age at presentation for germline retinoblastoma is 3 to 18 months. For sporadic, nongermline retinoblastoma, the average age is 18 to 24 months. Patients with known family history of retinoblastoma typically undergo early screening examinations before developing symptoms, with a funduscopic examination at birth and then every 2 to 3 months for the rst 3 years.6 Patients are then examined at increasing time intervals until it is safe to assume that no retinoblastoma will develop. The oldest age at which a child with a family history of retinoblastoma was reported to develop a tumor in a previously documented tumor-free eye is 48 months.2,24 In the 5-10% of patients diagnosed with retinoblastoma after the age of 5 years, the presentation

Retinoblastoma

137

is almost universally unilateral, sporadic retinoblastoma.25 Misdiagnosis in these older patients is common and the prospects for globe salvage poor.23,25 Leukocoria is the most common presenting sign in patients with retinoblastoma.26 In the United States, nearly 60% of all cases of retinoblastoma are diagnosed after observation of leukocoria in the affected eye, most often by a parent or family member. In a retrospective review by Abramson et al,26 the observation of leukocoria correlated with the presence of advanced disease. Because a retinoblastoma in the posterior pole has to be sufciently large to generate a leukocoria reex, leukocoria is a late presenting sign with poorer prognosis for globe salvage, but nonetheless has a good survival rate (approximately 88% at 5 y).3,27 Strabismus is the second major presenting sign of retinoblastoma, detected in approximately 1 in 5 retinoblastoma patients in the United States.26 A small tumor in the fovea or foveola or tumor-associated subretinal uid in the macula can signicantly reduce visual acuity and result in strabismus. In contrast to leukocoria, tumors presenting with strabismus as the initial sign are associated with a higher survival rate and a higher chance of globe salvage.3,27 The remaining 20% of the cases of retinoblastoma present with atypical signs and symptoms, including red, painful eye with glaucoma, cloudy cornea, poor vision, vitreous hemorrhage, or signs of orbital inammation mimicking orbital cellulitis.26 These uncommon presentations are usually late signs associated with more advanced disease and poorer survival and globe salvage prognosis.3,28 The clinical presentation of retinoblastoma is dependent on tumor growth pattern, duration of growth, degree of tumor vascularity, and the presence or absence of calcications, vitreous seeding, and retinal detachment. Endophytic retinoblastomas, accounting for nearly 60% of cases, exhibit growth toward the vitreous cavity as a result of cell division within the inner retinal layers.29 It presents as one or multiple foci, isolated or coalesced, variably sized, round or oval-shaped, and calcied creamy-white or vascularized pink retinal mass. These tumors have a tendency for vitreous seeding. The location of endophytic retinoblastoma is age-dependent, with the earliest tumors developing closer to the posterior pole and more anterior tumors arising at a later age.30 In a retrospective review by Palazzi et al,29 a disproportionately higher percentage of patients with endophytic retinoblastoma had a positive family history of retinoblastoma. Approximately 39% of retinoblastomas demonstrate an exophytic growth pattern with tumor development in the outer retinal layers and expansion beneath the retina into the subretinal space.29 Patients with exophytic retinoblastoma present with retinal detachment surrounding the tumor. A disproportionately high percentage of patients with

138

Kiss et al

exophytic retinoblastoma develop glaucoma. Choroidal invasion also occurs signicantly more often in patients with exophytic retinoblastoma than in those with endophytic retinoblastoma.29 Both exophytic and endophytic patterns can be seen with mixed growth retinoblastoma. The remaining 1% to 2% of cases of retinoblastoma exhibit a diffuse inltrating, plaquelike growth pattern. Tumor growth is diffusely within the retina without apparent signs of a mass.31 An irregular, grayish plaque is sometimes visible on the retinal surface. Diffuse inltrating retinoblastoma typically presents with signs suggestive of inammation, and, as a result, patients are often misdiagnosed as having uveitis.23 The average age at presentation of this atypical form is 6 years.31 Despite an often delayed diagnosis, the prognosis for survival is >95% after enucleation of the affected eye.31 Any child with unexplained, atraumatic hyphema or vitreous hemorrhage should be evaluated for retinoblastoma.4 Vitrectomy or retinal detachment repair should be avoided until the diagnosis of retinoblastoma is reliably ruled out.32 If vitrectomy is performed in an eye with unsuspected retinoblastoma, enucleation combined with adjuvant chemotherapy, radiotherapy, or both without delay is advised to prevent systemic tumor dissemination.32

Diseases Simulating Retinoblastoma

Shields and colleagues33 retrospectively examined 500 consecutive patients referred to the Ocular Oncology Service at Wills Eye Hospital with the diagnosis of possible retinoblastoma. In all, 58% of patients were found, on clinical evaluation, to have retinoblastoma and 42% had lesions that simulated retinoblastoma. A total of 23 different conditions accounted for these pseudoretinoblastomas. The 3 most common lesions imitating retinoblastomas were persistent hyperplastic primary vitreous (28%), Coats disease (16%), and presumed ocular toxocariasis (16%).33 Other less common conditions that may mimic retinoblastoma include congenital cataract, retinopathy of prematurity, familial exudative vitreoretinopathy, Norrie disease, and incontinentia pigmenti. Family history, clinical examination including indirect ophthalmoscopic examination (possibly under anesthesia), uorescein angiogram, and B-scan ultrasonography (oftentimes demonstrating the classic shadowing of intralesional calcication of retinoblastoma) can facilitate the diagnosis and differentiation of retinoblastoma from pseudoretinoblastoma.

Classication

There are 4 clinical stages of retinoblastoma: intraocular, regional, central nervous system, and hematogenous.23 Several classication systems have been developed specically for intraocular retinoblastoma.

Retinoblastoma

139

The Reese-Ellsworth Classication, developed in 1963, is based on intraocular tumor staging and globe salvage prediction after external beam radiation.34 This system divides eyes according to location, focality, and size of the tumors as determined by ophthalmoscopy into 5 groups (I to V) and 10 subgroups (a and b for each group). Group I consists of eyes with the lowest risk of enucleation and group V with the highest risk. At the time of the Reese-Ellsworth Classication development, external beam radiotherapy (EBRT) was the most popular nonenucleation treatment for retinoblastoma. Peripheral retinoblastomas, multifocal tumors, and larger tumors were more difcult to treat with EBRT than smaller, single tumors. The smaller macular tumors were presumed to be more aggressive and earned a higher ranking in the classication.34 Over the past 10 years, the treatment approaches to retinoblastoma have changed dramatically.2 In an effort to avoid the use of EBRT, which can result in considerable disgurement and increase the risk of developing additional cancers in patients with germline mutations, most retinoblastoma referral centers are using systemic chemotherapy (chemoreduction) and focal treatments, such as laser photocoagulation and cryotherapy, as the primary treatment modality. The limiting variables for successful retinoblastoma therapy with chemoreduction are considerably different than with EBRT. Tumor location, multifocality, and size were of lesser concern than vitreous and subretinal seeding.35 The Reese-Ellsworth Classication system does not address subretinal seeding nor does it differentiate between focal and diffuse seeding. As such, this classication system has been found to be a poor predictor of success (dened as salvage of the globe) with chemotherapy.4,36 In 2003, a new classication system for intraocular retinoblastoma was nalized (Table 1). The International Classication of Retinoblastoma (ICRB) is based mainly on the extent of tumor seeding in the vitreous and subretinal space, with consideration for tumor size and location.37 The ICRB is based both on the natural history of retinoblastoma (early disease, group A to late disease, group E) and on the likelihood of salvaging the eye when systemic chemotherapy and adjuvant focal therapy is used as the primary treatment paradigm. Group A eyes have lesions that are small and are away from critical visual structures (Table 1). In both groups A and B, the tumor remains conned to the retina. In groups C and D, the tumor has spread into the subretinal space and vitreous cavity. The tumor in group E eyes is extensive and has destroyed the eye. Group A eyes have the lowest risk of treatment failure and enucleation, whereas group E eyes are rarely salvageable and require enucleation as the primary treatment.36,37 The chances of retaining useful vision decreases from groups A to E. Like the Reese-Ellsworth Classication system, the ICRB addresses only the risk of loss of the eye and is not a staging system for the disease in the child. ICRB is not intended to predict life prognosis or visual outcome; rather,

140

Kiss et al

Table 1. The International Classication of Retinoblastoma* Group A B Subgroup A B Quick Reference Small tumor Larger tumor Macula Juxtapapillary Subretinal uid C C1 C2 C3 D D1 D2 D3 E E Extensive retinoblastoma Diffuse seeds Focal seeds Specic Features Retinoblastoma r3 mm in size Retinoblastoma >3 mm in size or Macular retinoblastoma location (r3 mm to foveola) Juxtapapillary retinoblastoma location (r1.5 mm to disc) Clear subretinal uid r3 mm from margin Retinoblastoma with Subretinal seeds r3 mm from retinoblastoma Vitreous seeds r3 mm from retinoblastoma Both subretinal and vitreous seeds r3 mm from retinoblastoma Retinoblastoma with Subretinal seeds >3 mm from retinoblastoma Vitreous seeds >3 mm from retinoblastoma Both subretinal and vitreous seeds >3 mm from retinoblastoma Extensive retinoblastoma occupying >50% globe or Neovascular glaucoma Opaque media from hemorrhage in anterior chamber, vitreous, or subretinal space Invasion of postlaminar optic nerve, choroid (>2 mm), sclera, orbit, anterior chamber

*Modied from Murphree37 and Shields and Shields.4

ICRB predicts globe outcome and the probability of avoidance of enucleation or EBRT after primary chemotherapy.36

Metastasis

There seem to be no inherent differences in the risk of metastatic disease between patients with germline mutations and those with sporadic retinoblastoma.2 Delay in diagnosis is the major risk factor in affecting survival. Kopelman and colleagues38 reported that the chances of metastasis and death were 2.5 times greater in patients in whom the clinical diagnosis of retinoblastoma was delayed. Metastasis from

Retinoblastoma

141

retinoblastoma typically develops within 1 year after diagnosis of the intraocular tumor.4 The poorly cohesive, high-mitotic-index retinoblastoma cells have a natural propensity to invade any readily accessible space, including the vitreous cavity, subretinal uid, the anterior segment, and the optic nerve.3 Retinoblastoma can spread anteriorly via the vitreous cavity or subretinal space toward the anterior segment, invading the ciliary body and anterior chamber and resulting in a pseudohypopyon. The tumor can extend posteriorly toward the optic nerve with intracranial invasion. Retinoblastoma may invade the orbit directly through the choroid and sclera or by following the scleral emissary canals of the ciliary arteries and nerves. Metastatic dissemination may also occur via lymphatic or vascular pathways.4 The greatest risks for metastatic disease include histopathologic features of retinoblastoma invasion beyond the lamina cribrosa in the optic nerve or orbital invasion with extension outside the sclera.24,6 The mortality rate is 50% to 85% if the tumor cells have reached the surgical resection line of the optic nerve, 13% to 69% if the tumor is posterior to the lamina cribrosa but not at the resection line, and insignicant if the cells are anterior to the lamina cribrosa.3,39 The association of optic nerve invasion and massive choroidal invasion carries the greatest risk of metastasis and requires preventative or adjuvant chemotherapy.4,40

Secondary Nonocular Tumors

Patients with hereditary retinoblastoma are at risk for developing an intracranial neuroblastic tumor, most often a pinealoblastoma or other parasellar tumor, in the rst 5 years of life.41 This trilateral retinoblastoma is found in approximately 3% of all children with retinoblastoma and 10% of those with bilateral or familial retinoblastoma.4 The median time from diagnosis of retinoblastoma to the diagnosis of the neuroblastic tumor is 21 months.41 Unfortunately, pinealoblastoma is nearly universally fatal with only a few survivors.41 A recent retrospective analysis by Shields and colleagues42 indicates that chemoreduction for retinoblastoma may prevent the development of intracranial neuroblastic tumors. None of the 142 patients treated with chemoreduction developed pinealoblastoma. It was estimated that 5 to 15 of these patients should have manifested the trilateral retinoblastoma.42 Carriers of mutations in the retinoblastoma gene have an increased incidence of developing second tumors.43 Patients with bilateral or heritable retinoblastoma have the RB1 mutation present in all cells predisposing them to develop nonocular tumors. Second tumors develop in 4.4% of the patients during the rst 10 years of follow-up, in 18.3% after 20 years, and in 26.1% after 30 years.44 Second cancers

142

Kiss et al

most often include osteogenic sarcoma, spindle cell sarcoma, chondrosarcoma, rhabdomyosarcoma, neuroblastoma, glioma, leukemia, sebaceous cell carcinoma, squamous cell carcinoma, and malignant melanoma.2,4,43 Second nonocular tumors are the greatest cause of death in these patients, rather than the primary retinoblastoma itself.22,43 Survivors of retinoblastoma in whom second malignant neoplasms develop are at an even higher risk for the development of additional tumors (third, fourth, and even fth nonocular tumors) than they were for the development of a second tumor.45 The locations and expected ages at which additional tumors develop are consistent with the patterns seen in second tumors. The incidence rate for the development of nonocular tumors is further increased in hereditary retinoblastoma patients who have been treated with EBRT.44 In a retrospective analysis by Abramson and colleagues,45 96% of patients in whom third tumors developed had received radiation therapy for their intraocular retinoblastoma.

Treatment

Management of a child with retinoblastoma is individualized and is based on the threat of metastatic disease, risks for secondary cancers, systemic status, laterality of disease, size and location of tumors, anticipated visual prognosis, and response to treatment.4 The most important objective of a treatment plan is the survival of the child, followed by salvage of the globe, and nally the preservation of the best possible visual function. Systemic chemotherapy (most commonly carboplatin, etoposide, and vincristine with or without cyclosporin) has replaced EBRT as the most commonly used primary treatment for intraocular retinoblastoma.2,6,35,46,47 Chemotherapy is used to shrink the tumor until it is accessible to adjuvant focal treatment such as photocoagulation, thermotherapy, cryotherapy, or brachytherapy.46,48 The number and frequency of chemotherapy cycles vary from institution to institution and are dependent on the stage of the tumor.46,47 Chemotherapy reduces the size of the tumor but does not usually cure the retinoblastoma; focal therapy is necessary to consolidate the chemotherapy response. Chemoreduction, followed by focal consolidation, has reduced the frequency of enucleation and spared the complications associated with primary EBRT (such as the development of secondary nonocular tumors).3 In general, eyes classied as ICRB group A are treated with cryotherapy or laser photocoagulation. Eyes classied as group B or C usually receive chemoreduction, followed by focal consolidation. Eyes classied as group D are treated with either chemoreduction or enucleation depending on the laterality of the disease. Group E eyes are treated by primarily enucleation (Table 2). Shields and colleagues36

Retinoblastoma

143

recently demonstrated that the ICRB is predictive of globe outcome after chemoreduction. Treatment success was achieved in 100% of group A, 93% of group B, 90% of group C, and 47% of group D eyes.36 Over half the patients in group D required enucleation or EBRT. All group E eyes in this series were treated with enucleation. Conservative treatment is possible in approximately 25% of all unilateral retinoblastoma; enucleation is necessary in the remaining 75%.27,49 This high rate of enucleation is a result of the delay in diagnosis of unilateral sporadic retinoblastoma, which is most often detected by the parents or family members as leukocoria or strabismus.27 For those with less advanced unilateral tumor, chemoreduction with focal consolidation is benecial (Table 2).40 Unilateral group A tumors are treated with cryotherapy or laser photocoagulation, chemoreduction or plaque or proton radiotherapy is used for group B or C, and unilateral group D or E retinoblastoma most often requires enulceation.4 The treatment of bilateral disease is based on the stage of the most advanced eye (Table 2). Bilateral retinoblastoma require enucleation of at least 1 eye in 60% of cases for dangerously advanced tumor.40 Bilateral enucleation is necessary in only about 1% of cases.49 On the
Table 2. Treatment Strategy on the Basis of Laterality and Retinoblastoma Grouping* Bilateral (on the Basis of Most Advanced Eye) Cryotherapy or laser Photocoagulation Vincristine, carboplatin plus thermotherapy/cryotherapy Vincristine, carboplatin, and etoposide plus thermotherapy/cryotherapy Vincristine, carboplatin, etoposide plus thermotherapy/cryotherapy, and subconjunctival carboplatin Enucleation, but if both eyes equally advanced then vincristine, carboplatin, etoposide plus thermotherapy/cryotherapy and subconjunctival carboplatin, and proton radiotherapy

International Classication of Retinoblastoma A B

Unilateral Cryotherapy or laser Photocoagulation Vincristine, carboplatin plus thermotherapy/cryotherapy, or plaque radiotherapy Vincristine, carboplatin, etoposide plus thermotherapy/cryotherapy, or plaque or proton radiotherapy Vincristine, carboplatin, etoposide plus thermotherapy/cryotherapy and subconjunctival carboplatin, or enucleation Enucleation

*Modied from Shields and Shields.4

144

Kiss et al

basis of the worst eye, eyes with only group A disease are treated with laser or cryotherapy; group B or C eyes receive chemoreduction followed by focal consolidation; group D eyes receive chemoreduction, focal consolidation and subconjunctival or subTenons carboplatin or proton radiotherapy, or enucleation and; group E eyes are treated with enucleation (Table 2). If both eyes are equally advanced (both group E), the patient is treated with chemoreduction, subconjunctival or subTenons carboplatin, and proton radiotherapy.3,4 The principal complication of chemoreduction is the recurrence of subretinal or vitreous seeds, oftentimes remote from the original tumor.35,40,47 Tumor recurrence rate is highest for retinoblastoma located in the macula and for retinoblastoma with greater thickness.50 In some cases, recurrent seeds can be treated with cryotherapy or plaque radiotherapy.4 Patients with hereditary retinoblastoma can also develop new tumors during or after chemotherapy.51 The mean interval to new tumor development is roughly 5 months after the initiation of chemoreduction. New tumors can be detected in 24% of patients by 5 years of follow-up.51 Children with retinoblastoma treated with chemoreduction and focal consolidation should be followed at regular intervals for both recurrence and new tumor development. Most complications of the chemotherapeutic agents currently used to treat intraocular retinoblastoma are dose related and are of short duration. Unlike EBRT, these chemotherapeutic agents do not seem to predispose children with heritable retinoblastoma to the same tumors for which they are genetically at risk.24,6 However, long-term follow-up of patients treated with primary chemoreduction is still necessary for better understanding of primary tumor control and potential effects on secondary tumor development.

Conclusion

Retinoblastoma is the most common primary intraocular tumor in children. The causative gene was the rst in the class of tumorsuppressor genes identied. Remarkable advances in the treatment of this disease have been made over the past century. Chemoreduction, followed by adjuvant focal consolidation, in combination with early detection, has lead to a dramatic increase in overall survival and a substantial decrease in morbidity. Patients eyes are now able to be preserved, and oftentimes, useful visual function is also maintained. Future directions include preimplantation diagnosis in families with a genetic predisposition for developing retinoblastoma and targeted molecular therapy for retinoblastoma, taking advantage of the specic genetic defects responsible for the tumor.3

Retinoblastoma

145

Financial Support: None. Conict of Interest: No conicting relationship exists.

References

1. Mahoney MC, Burnett WS, Majerovics A, et al. The epidemiology of ophthalmic malignancies in New York State. Ophthalmology. 1990;97:11431147. 2. Abramson DH. Retinoblastoma in the 20th century: past success and future challenges the Weisenfeld lecture. Invest Ophthalmol Vis Sci. 2005;46:26832691. 3. Balmer A, Zografos L, Munier F. Diagnosis and current management of retinoblastoma. Oncogene. 2006;25:53415349. 4. Shields CL, Shields JA. Basic understanding of current classication and management of retinoblastoma. Curr Opin Ophthalmol. 2006;17:228234. 5. Hilgartner H. Report of a case of double glioma treated by x-ray. Texas Med J. 1903;18:322323. 6. Abramson DH, Scheer AC. Update on retinoblastoma. Retina. 2004;24:828848. 7. Knudson AG Jr. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci USA. 1971;68:820823. 8. Zhang J, Gray J, Wu L, et al. Rb regulates proliferation and rod photoreceptor development in the mouse retina. Nat Genet. 2004;36:351360. 9. Zheng L, Lee WH. The retinoblastoma gene: a prototypic and multifunctional tumor suppressor. Exp Cell Res. 2001;264:218. 10. Seregard S, Lundell G, Svedberg H, et al. Incidence of retinoblastoma from 1958 to 1998 in Northern Europe: advantages of birth cohort analysis. Ophthalmology. 2004;111:12281232. 11. Abramson DH. Retinoblastoma incidence in the United States. Arch Ophthalmol. 1990;108:1514. 12. Buckley JD. The aetiology of cancer in the very young. Br J Cancer Suppl. 1992;18: S8S12. 13. Orjuela M, Castaneda VP, Ridaura C, et al. Presence of human papilloma virus in tumor tissue from children with retinoblastoma: an alternative mechanism for tumor development. Clin Cancer Res. 2000;6:40104016. 14. Orjuela MA, Titievsky L, Liu X, et al. Fruit and vegetable intake during pregnancy and risk for development of sporadic retinoblastoma. Cancer Epidemiol Biomarkers Prev. 2005;14:14331440. 15. Yip BH, Pawitan Y, Czene K. Parental age and risk of childhood cancers: a population-based cohort study from Sweden. Int J Epidemiol. 2006;35:14951503. 16. Dryja TP, Morrow JF, Rapaport JM. Quantication of the paternal allele bias for new germline mutations in the retinoblastoma gene. Hum Genet. 1997;100:446449. 17. Greger V, Debus N, Lohmann D, et al. Frequency and parental origin of hypermethylated RB1 alleles in retinoblastoma. Hum Genet. 1994;94:491496. 18. Moll AC, Imhof SM, Kuik DJ, et al. High parental age is associated with sporadic hereditary retinoblastoma: the Dutch retinoblastoma register 1862-1994. Hum Genet. 1996;98:109112. 19. Kato MV, Ishizaki K, Shimizu T, et al. Parental origin of germ-line and somatic mutations in the retinoblastoma gene. Hum Genet. 1994;94:3138. 20. Zhu XP, Dunn JM, Phillips RA, et al. Preferential germline mutation of the paternal allele in retinoblastoma. Nature. 1989;340:312313. 21. Munier F, Spence MA, Pescia G, et al. Paternal selection favoring mutant alleles of the retinoblastoma susceptibility gene. Hum Genet. 1992;89:508512. 22. Abramson DH. Second nonocular cancers in retinoblastoma: a unied hypothesis. The Franceschetti Lecture. Ophthalmic Genet. 1999;20:193204.

146

Kiss et al

23. Walton DS, Mukai S, Grabowski EF, et al. Case records of the Massachusetts General Hospital. Case 5-2006. An 11-year-old girl with loss of vision in the right eye. N Engl J Med. 2006;354:741748. 24. Moll AC, Imhof SM, Meeteren AY, et al. At what age could screening for familial retinoblastoma be stopped? A register based study 1945-98. Br J Ophthalmol. 2000; 84:11701172. 25. Shields CL, Shields JA, Shah P. Retinoblastoma in older children. Ophthalmology. 1991;98:395399. 26. Abramson DH, Frank CM, Susman M, et al. Presenting signs of retinoblastoma. J Pediatr. 1998;132(3 Pt 1):505508. 27. Shields CL, Gorry T, Shields JA. Outcome of eyes with unilateral sporadic retinoblastoma based on the initial external ndings by the family and the pediatrician. J Pediatr Ophthalmol Strabismus. 2004;41:143149; quiz 7273. 28. Butros LJ, Abramson DH, Dunkel IJ. Delayed diagnosis of retinoblastoma: analysis of degree, cause, and potential consequences. Pediatrics. 2002;109:E45. 29. Palazzi M, Abramson DH, Ellsworth RM. Endophytic vs exophytic unilateral retinoblastoma: is there any real difference? J Pediatr Ophthalmol Strabismus. 1990;27:255258. 30. Abramson DH, Gombos DS. The topography of bilateral retinoblastoma lesions. Retina. 1996;16:232239. 31. Bhatnagar R, Vine AK. Diffuse inltrating retinoblastoma. Ophthalmology. 1991; 98:16571661. 32. Shields CL, Honavar S, Shields JA, et al. Vitrectomy in eyes with unsuspected retinoblastoma. Ophthalmology. 2000;107:22502255. 33. Shields JA, Parsons HM, Shields CL, et al. Lesions simulating retinoblastoma. J Pediatr Ophthalmol Strabismus. 1991;28:338340. 34. Reese AB, Ellsworth RM. The evaluation and current concept of retinoblastoma therapy. Trans Am Acad Ophthalmol Otolaryngol. 1963;67:164172. 35. Shields JA, Shields CL, Meadows AT. Chemoreduction in the management of retinoblastoma. Am J Ophthalmol. 2005;140:505506. 36. Shields CL, Mashayekhi A, Au AK, et al. The international classication of retinoblastoma predicts chemoreduction success. Ophthalmology. 2006;113: 22762280. 37. Linn Murphree A. Intraocular retinoblastoma: the case for a new group classication. Ophthalmol Clin North Am. 2005;18:4153, viii. 38. Kopelman JE, McLean IW, Rosenberg SH. Multivariate analysis of risk factors for metastasis in retinoblastoma treated by enucleation. Ophthalmology. 1987;94: 371377. 39. Uusitalo MS, Van Quill KR, Scott IU, et al. Evaluation of chemoprophylaxis in patients with unilateral retinoblastoma with high-risk features on histopathologic examination. Arch Ophthalmol. 2001;119:4148. 40. Shields CL, Meadows AT, Leahey AM, et al. Continuing challenges in the management of retinoblastoma with chemotherapy. Retina. 2004;24:849862. 41. Kivela T. Trilateral retinoblastoma: a meta-analysis of hereditary retinoblastoma associated with primary ectopic intracranial retinoblastoma. J Clin Oncol. 1999; 17:18291837. 42. Shields CL, Meadows AT, Shields JA, et al. Chemoreduction for retinoblastoma may prevent intracranial neuroblastic malignancy (trilateral retinoblastoma). Arch Ophthalmol. 2001;119:12691272. 43. Fletcher O, Easton D, Anderson K, et al. Lifetime risks of common cancers among retinoblastoma survivors. J Natl Cancer Inst. 2004;96:357363. 44. Roarty JD, McLean IW, Zimmerman LE. Incidence of second neoplasms in patients with bilateral retinoblastoma. Ophthalmology. 1988;95:15831587.

Retinoblastoma

147

45. Abramson DH, Melson MR, Dunkel IJ, et al. Third (fourth and fth) nonocular tumors in survivors of retinoblastoma. Ophthalmology. 2001;108:18681876. 46. Chan HS, Gallie BL, Munier FL, et al. Chemotherapy for retinoblastoma. Ophthalmol Clin North Am. 2005;18:5563, viii. 47. Shields CL, Honavar SG, Meadows AT, et al. Chemoreduction plus focal therapy for retinoblastoma: factors predictive of need for treatment with external beam radiotherapy or enucleation. Am J Ophthalmol. 2002;133:657664. 48. Shields CL, Shields JA, Needle M, et al. Combined chemoreduction and adjuvant treatment for intraocular retinoblastoma. Ophthalmology. 1997;104:21012111. 49. Epstein JA, Shields CL, Shields JA. Trends in the management of retinoblastoma: evaluation of 1196 consecutive eyes during 1974 to 2001. J Pediatr Ophthalmol Strabismus. 2003;40:196203; quiz 1718. 50. Shields CL, Mashayekhi A, Cater J, et al. Chemoreduction for retinoblastoma. Analysis of tumor control and risks for recurrence in 457 tumors. Am J Ophthalmol. 2004;138:329337. 51. Shields CL, Shelil A, Cater J, et al. Development of new retinoblastomas after 6 cycles of chemoreduction for retinoblastoma in 162 eyes of 106 consecutive patients. Arch Ophthalmol. 2003;121:15711576.