Pretutorial
    Assessment

Tutorial
Introduction
Prognostic Factors
Conclusion
References

Slides

Prognostic Factors in Retinoblastoma

Arun D. Singh, MD · Carol L. Shields, MD · Jerry A. Shields, MD

Reprinted with permission from the Journal of Pediatric Ophthalmology and Strabismus. 2000;37(3):134-141.

Introduction

Retinoblastoma is the most common primary intraocular malignant tumor in children with an incidence of 1 in 15,000 live births.¹ The average annual incidence of retinoblastoma in the United States has been estimated at 10.9 per million for children younger than 5 years, and the incidence has remained stable from 1974 to 1985.² Most children present with a white pupillary reflex or leukocoria.³ Ancillary studies such as ultrasonography, computed tomography, and magnetic resonance imaging can assist in confirming the diagnosis.⁴

In recent years, there has been a trend away from enucleation⁵ and an increase in the use of alternative globe-conserving methods of treatment including external beam radiotherapy,⁶ plaque radiotherapy,⁷ laser photocoagulation,⁸ cryotherapy,⁹ and transpupillary thermotherapy.^l0 Currently, chemotherapy is being used increasingly as chemoreduction with other adjuvant therapy to avoid external beam radiotherapy and enucleation.^11,12 Recent advances in the treatment of retinoblastoma have led to a reduced risk of metastasis and improved survival rates in developed countries. Five-year survival rates of 91%, 93%, and 88% have been reported from the United States,² Japan,¹³ and the United Kingdom,⁵ respectively. However, in underdeveloped countries, retinoblastoma is still associated with high mortality. Of 44 Nigerian patients with retinoblastoma, 57% (25 patients) had a fatal outcome.¹⁴

Retinoblastoma is a familial disorder with an autosomal dominant inheritance. Approximately 60% of patients are considered sporadic and 40% are heritable. Of the heritable type of retinoblastoma, 5% to 15% have an existing family history and the remaining 25% are due to new germline mutations.¹⁵ The heritable variant of retinoblastoma tends to be bilateral and presents earlier with a mean age of 12 months. The sporadic type of retinoblastoma tends to be unilateral and presents later with a mean age of 24 months. All bilateral cases are considered heritable and harbor germline mutation. In addition, approximately 15% of all unilateral cases are also heritable. Patients with the heritable variant of retinoblastoma are particularly susceptible to the development of pineoblastoma and other second malignant neoplasms, particularly osteosarcoma.

Metastatic disease at the time of retinoblastoma diagnosis is rare. Therefore, staging procedures such as bone scans, lumbar puncture, and bone marrow aspirations at initial presentation are not recommended.^16,17 The metastasis in retinoblastoma usually occurs within 1 year of diagnosis of retinoblastoma. If there is no metastatic disease within 5 years of retinoblastoma diagnosis, the child is considered cured. ¹⁸ Involvement of the central nervous system and hematogenous spread are common. Survival with metastatic retinoblastoma is limited, with death occurring generally within 6 months.^18,19

This review provides an update on the prognostic factors that may influence decisions regarding the use of adjuvant therapy in high-risk cases of retinoblastoma. The prognosis can be related to clinical features such as delay in diagnosis, prior intraocular surgery, and the genetic subtype of retinoblastoma (Table 1). Treatment factors that can affect prognosis include use of external beam radiotherapy and chemoreduction. Tumor histopathologic factors include tumor invasion of the choroid, optic nerve, and orbit.

Prognostic Factors

Clinical Prognostic Factors
Delay in Diagnosis. The most common presentations of retinoblastoma include leukocoria and strabismus.²⁰ Children presenting with such symptoms or signs should be examined by an ophthalmologist to exclude retinoblastoma. In a study of 100 patients with retinoblastoma in the United Kingdom, a delay in diagnosis did not increase the likelihood of treatment by enucleation, but the risk of local tumor invasion was increased.²¹ The impact of such a delay on survival is unknown.

In another study of 153 children with retinoblastoma from Brazil, 80 (52%) patients received treatment within 6 months of symptom onset (early diagnosis) and 73 (48%) received treatment more than 6 months after symptom onset (late diagnosis). The 3-year survival rate was 82% in the early diagnosis group compared with 44% in the late diagnosis group. The worse prognosis in the late diagnosis group correlated with the tendency of patients in this group to exhibit extraocular extension of the retinoblastoma.^22,23

Prior Intraocular Surgery. Extraocular spread of retinoblastoma is a poor prognostic factor. Intraocular surgical procedures such as pars plana vitrectomy, inadvertently performed on patients with retinoblastoma, predisposes to orbital extension or metastasis of retinoblastoma. Stevenson and colleagues²⁴ reported three such patients, all of whom developed orbital retinoblastoma following intraocular surgery. Shields and associates²⁵ reported a series of 11 patients after vitrectomy. Each child was treated by prompt enucleation and most received adjuvant prophylactic orbital radiotherapy and chemotherapy. Metastasis developed in one case with a mean follow-up of 83 months.

Genetic Subtype. The heritable form of retinoblastoma has significant long-term prognostic implications. Second malignant neoplasms, which occur predominantly in the heritable form of retinoblastoma, have now become significant contributors to overall mortality.²⁶ In a large study from the United Kingdom, the overall 3-year survival rate was 88% in 431 patients with retinoblastoma who underwent follow-up for up to 17 years.²⁶ Patients with bilateral tumors had worse long-term survival rates because of later deaths from second primary neoplasms.⁵

Similar results were observed in a study from Japan.¹³ In 1147 cases of retinoblastoma from the National Registry of Retinoblastoma of Japan, the 5-year survival rate of 93% for 757 unilateral cases was similar to that of 92% for 390 bilateral cases. However, the 10-year survival rates were 92.3% and 86.7% for the unilateral and bilateral cases, respectively. The occurrence of a second neoplasm was the main cause of death in the hereditary cases 10 years after diagnosis.¹³

Second Malignant Neoplasms
It is now recognized that more children with retinoblastoma die from the second malignant neoplasm than from the retinoblastoma itself.²⁷ The incidence of second malignant neoplasm in the heritable form of retinoblastoma has been reported to be between 5% and 90% in various studies due to differences in the methods of ascertainment of the cases and analysis of the data (Table 2).²⁷ In a study from the United Kingdom, Draper and colleagues²⁸ reported a cumulative

SLIDE 1

There are many types of second malignant neoplasms, and at least 35 distinct types have been reported in association with retinoblastoma.³³ The majority of the neoplasms are osteosarcoma (37%) and soft tissue sarcoma (7%).³⁴ Cutaneous melanoma represents approximately 7% of all second malignant neoplasms.³⁴ The risk of developing bone tumors and connective and soft tissue tumors in heritable retinoblastoma is increased approximately 400 times and 100 times, respectively, compared with the normal population.³² The role of radiotherapy in the pathogenesis of second malignant neoplasm is discussed below.

Trilateral Retinoblastoma
The association between heritable retinoblastoma and a primary intracranial malignancy especially of the pineal gland is termed trilateral retinoblastoma.³⁵ In patients with bilateral retinoblastoma, the presence of an intracranial malignancy is now recognized to be the most frequent cause of death in the first decade of life, accounting for 50% of all deaths.³⁶

Trilateral retinoblastoma occurs in approximately 8% of all heritable retinoblastoma.³⁷ The nature of primary intracranial malignant tumor can vary in its location and histopathologic features.⁵ The majority of tumors are located in the pineal region but can also occur in the suprasellar and parasellar regions.^38,39 The overall prognosis in the presence of trilateral retinoblastoma is poor.⁴⁰ Kivela⁴¹ reported findings of a meta-analysis based on all published reports of trilateral retinoblastoma between 1966 and 1998. Data on 106 children were analyzed. The median time from retinoblastoma to the diagnosis of intracranial tumor was 21 months with 75% of intracranial tumor cases diagnosed within 1 year of retinoblastoma diagnosis. Screening by neuroimaging was able to detect intracranial tumor at an earlier age but did not prolong survival. The median survival time after diagnosis of trilateral retinoblastoma was 9 months.⁴¹ It is hoped that earlier neuroimaging studies may detect the intracranial tumor and allow earlier treatment.

External Beam Radiotherapy. The relationship between radiation and second malignant neoplasms is complex.²⁷ At least two effects of external beam radiotherapy in the pathogenesis of second malignant neoplasms in patients with retinoblastoma have been identified. These include enhancement of the baseline risk and a localizing effect.³¹ The effects of radiotherapy are influenced by the age of the recipient and are believed to be dose dependent.^31,32

SLIDE 2

The influence of radiotherapy in causing second malignant neoplasm appears to be dependent on the age of the child, when radiotherapy is administered, and the dose of radiation applied.^31,32 If radiation is given to children who are younger than 1 year of age, the cumulative incidence of a second malignant neoplasm is 9% at age 10 years and 34% at 30 years compared to 5% and 22%, respectively, for children who are older than 1 year when treated.³¹ In addition, the effect of radiotherapy may be dose dependent with elevated risk associated with doses higher than 50 Gy.³² Currently, with megavoltage techniques, an average dose of 44 Gy is used to treat retinoblastoma, and the high-energy photon beams deliver a uniform dose to the bone and soft tissues.⁶ With present techniques of external beam radiotherapy for retinoblastoma, the enhancing influence of external beam radiotherapy in the pathogenesis of second malignant neoplasm may be lower than the published data.

Use of Chemoreduction. Chemoreduction involves the administration of chemotherapy for advanced intraocular retinoblastoma to achieve tumor reduction so that focal ophthalmic therapy can be applied to partially regressed tumors.^11,12 In a series of 160 children with intraocular retinoblastoma treated with chemoreduction, none of the children developed trilateral retinoblastoma.⁴² In the same cohort, ^5-16 cases were expected to develop trilateral retinoblastoma. It is suspected that chemoreduction may play a protective role against the development of intracranial tumor.⁴² However, chemoreduction may affect the development of other second malignant neoplasms positively or negatively, and longer follow-up is necessary.

Histopathologic Prognostic Factors
Many studies have evaluated the histopathologic prognostic factors including tumor size, tumor growth pattern, tumor differentiation, extension into the choroid, optic nerve, anterior chamber, and extrascleral tissues.^18,23,43-47 Multivariate analysis, in which correction for confounding effects of multiple risk factors is made, is particularly helpful in identifying the influence of individual prognostic factors (Table 3).^18,23,43-45

Tumor parameters such as size, growth pattern (endophytic or exophytic),⁴⁷ and differentiation⁴³ do not significantly influence the systemic prognosis. The Reese-Ellsworth classification traditionally has been used to grade the retinoblastoma from stage I to stage V.⁴⁸ The classification system is based on the number, size, and location of the tumors. The Reese-Ellsworth stage of retinoblastoma indicates the prognosis for globe conservation rather than systemic prognosis due to metastatic disease. The pattern of tumor growth (endophytic or exophytic) has no significant influence on the prognosis. In a retrospective review of 297 cases of enucleated and histologically proven unilateral retinoblastoma, endophytic retinoblastoma (181 cases) was found more often than exophytic retinoblastoma (116 cases). With a 5-year minimum follow-up, the clinical appearance of either endophytic or exophytic tumor did not correlate with the optic nerve invasion, orbital recurrence, or survival of the patient.⁴⁷ A similar lack of significant influence of tumor differentiation on systemic prognosis has been observed.⁴³

Choroidal Invasion. There are conflicting results as to whether extension of retinoblastoma into the choroid is a poor prognostic factor.^18,49 A histopathologic review of 74 patients with retinoblastoma revealed choroidal invasion correlated with 100% survival provided the sclera, iris, and optic nerve were not involved.⁵⁰ In a multivariate analysis of risk factors for metastasis in retinoblastoma treated by enucleation, statistical analysis of 361 cases indicated histologic detection of choroidal invasion was not significantly associated with a fatal outcome.¹⁸

In a detailed study of metastatic potential of choroidal invasion of retinoblastoma, clinical and histopathologic data were evaluated on 289 eyes enucleated for retinoblastoma.⁴⁶ Sixty-seven (23%) eyes had histopathologic evidence of choroidal invasion. Patients with choroidal invasion were more likely to develop metastasis than those without choroidal invasion (P = 0.0001). When patients with optic nerve invasion were excluded, there was no significant risk, but there was a trend toward the development of metastasis (P = 0.10). The choroidal invasion could be predicted by iris neovascularization. The authors concluded choroidal invasion of retinoblastoma is a risk for metastasis, especially if it is associated with any degree of optic nerve invasion.⁴⁶ Studies using methods of multivariate analysis have supported the view that choroidal invasion by retinoblastoma increases the risk of metastatic disease.^23,44

Optic Nerve Invasion. Optic nerve invasion of retinoblastoma is a poor prognostic factor for patients with retinoblastoma.⁵¹ In fact, multivariate statistical analysis has indicated histologic detection of optic nerve invasion by retinoblastoma is one of the most highly predictive factors for death from retinoblastoma.¹⁸ Since 1960, the frequency and severity of optic nerve invasion has been less extensive due to the earlier diagnosis of retinoblastoma.⁵¹

Optic nerve invasion is present in approximately one third of enucleated globes for retinoblastoma and correlates with the presence of choroidal or scleral extension.^45,50,52 Optic nerve invasion is categorized as prelaminar, laminar, postlaminar, and up to the line of transection.⁴⁵ Prelaminar optic nerve involvement is most common, representing about half of all cases of optic nerve involvement.⁵² The laminar and the retrolaminar involvement in equal portions accounts for the remaining half of the cases. In our experience, optic nerve involvement up to the line of transection is rare and was seen in two (1%) of 289 cases.⁵²

There is increasing mortality with increasing extent of optic nerve involvement.⁴⁵ However, it is generally agreed that prelaminar involvement of the optic nerve does not increase the risk of metastasis.^44,45,52,53 The impact of laminar involvement on metastasis is debated.^45,52 Retrolaminar involvement is a poor prognostic factor, and optic nerve invasion by retinoblastoma up to the line of transection predicts the worst prognosis.^23,44,52 In a study of 361 cases, the multivariate odds ratio for metastasis was 9 when the optic nerve was involved up to the resection line compared to 4 when the retinoblastoma extended retrolaminar but did not extend up to the resection line.¹⁸

In another study of 172 patients with retinoblastoma, the 3-year disease-free survival rate was 97% when the optic nerve was not involved compared to 55% when the optic nerve invasion was up to the line of transection. The prognosis was intermediate with retrolaminar involvement.⁴⁴ Further analysis of 149 patients, after excluding those with extrascleral extension and optic nerve involvement up to the resection line, revealed retrolaminar involvement of the optic nerve was a significant prognostic factor for metastasis.⁴⁴ Similar prognostic implications of retrolaminar invasion of the optic nerve by retinoblastoma have been supported by other studies.²³

Optic nerve invasion can be predicted by the presence of a large exophytic retinoblastoma with secondary glaucoma and vitreous hemorrhage. These findings should alert the clinician to suspect optic nerve invasion. When enucleation is performed for retinoblastoma, an attempt should be made to salvage a long stump of optic nerve (10 mm to 15 mm) so as to transect the optic nerve beyond the extent of involvement by retinoblastoma.⁵⁴

Orbital Invasion. Extrascleral or orbital extension of retinoblastoma is possibly the worst prognostic factor for death from retinoblastoma.¹⁸ The multivariate odds ratio of 27 for metastasis is highest when there is invasion into the orbit indicating that orbital invasion is the worst prognostic indicator in retinoblastoma with up to 90% mortality in 2 years.^18,55 The presence of orbital invasion of retinoblastoma is associated with a 10 times greater risk of metastasis compared to cases with no orbital invasion.⁴⁴

Prophylactic chemotherapy with adjuvant orbital radiotherapy has been advocated in retinoblastoma cases with poor prognostic indicators such as prior intraocular surgery, orbital extension of retinoblastoma, and anterior chamber seeding, as well as in cases with extension of retinoblastoma up to the line of optic nerve transection.^44,53,54 The role of such prophylactic treatment in instances with isolated choroidal invasion and retrolaminar optic nerve invasion remains controversial.⁵⁶

In a recent retrospective study, outcome was analyzed in 80 unilateral cases treated by primary enucleation with histopathologic risk factors for metastasis. The presence or absence of metastasis at a minimum of 12 months follow-up was the main outcome measure. Forty-six patients were treated by prophylactic chemotherapy (treated group) for at least 6 months and 34 patients were untreated. A significant difference (P = 0.01) was found in the incidence of metastasis between the treated (two [4%] of 46) and the untreated (eight [24%] of 34) groups.⁵⁷ Our data suggest beneficial effects of prophylactic therapy in reducing the risk of metastatic disease in a select group of high-risk retinoblastoma.

Conclusion

Various clinical and histopathologic prognostic factors have been identified in retinoblastoma. Early recognition of symptoms with a high index of suspicion and prompt referral to a specialized center will further improve survival in patients with retinoblastoma. Patients with germline mutations of retinoblastoma gene are at risk for development of pineoblastoma and other second malignant neoplasms. The use of external beam radiotherapy enhances the risk for development of second malignant neoplasms. Uncontrolled clinical studies suggest the use of prophylactic chemotherapy in high-risk retinoblastoma patients is effective in reducing the risk of metastasis. Due to the limited number of cases treated at a given center, there is a need for a collaborative prospective randomized clinical trial evaluating the efficacy of prophylactic chemotherapy in retinoblastoma patients with poor prognostic factors.

References

1. Bishop JO, Madsen EC. Retinoblastoma: review of current status. Surv Ophthalmol. 1975;19:342-366.
2. Tamboli A, Podgor MJ, Horm JW. The incidence of retinoblastoma in the United States: 1974 through 1985. Arch Ophthalmol. 1990;108:128-132.
3. Shields JA, Shields CL. Intraocular Tumors: A Text and Atlas. Philadelphia, Pa: WB Saunders Co; 1992:305-319.
4. Shields JA, Shields CL. Intraocular Tumors: A Text and Atlas. Philadelphia, Pa: WB Saunders Co; 1992:363-376.
5. Shields CL, Shields JA. Recent developments in the management of retinoblastoma. J Pediatr Ophthalmol Strabismus. 1999;36:8-18.
6. Hernandez JC, Brady LW, Shields JA, et al. External beam radiation for retinoblastoma: results, patterns of failure and proposal for treatment guidelines. Int J Radiat Oncol Biol Phys. 1996;35:125-132.
7. Shields JA, Shields CL, De Potter P, Hernandez JC, Brady LW. Plaque radiotherapy for residual or recurrent retinoblastoma in 91 cases. J Pediatr Ophthalmol Strabismus. 1994;31:242-245.
8. Shields CL, Shields JA, Kiratli H, De Potter PV. Treatment of retinoblastoma with indirect ophthalmoscope laser photocoagulation. J Pediatr Ophthalmol Strabismus. 1995;32:317-322.
9. Shields JA, Parsons H, Shields CL, Giblin ME. The role of cryotherapy in the management of retinoblastoma. Am J Ophthalmol. 1989;108:260-264.
10. Shields CL, Santos MC, Diniz W, et al. Thermotherapy for retinoblastoma. Arch Ophthalmol. 1999;117:885-893.
11. Ferris FL, Chew KY. A new era for the treatment of retinoblastoma. Arch Ophthalmol. 1996;114:1412.
12. Shields CL, Shields JA, Needle M, et al. Combined chemoreduction and adjuvant treatment for intraocular retinoblastoma. Ophthalmology. 1997;104:2101-2111.
13. Anonymous. Survival rate and risk factors for patients with retinoblastoma in Japan. The Committee for the National Registry of Retinoblastoma. Jpn J Ophthalmol. 1992;36:121-131.
14. Ajaiyeoba IA, Akang EE, Campbell OB, Olurin IO, Aghadiuno PU. Retinoblastomas in Ibadan: treatment and prognosis. West Afr J Med. 1993;12:223-227.
15. Vogel F. Genetics of retinoblastoma. Hum Genet. 1979;52:1-54.
16. Pratt CB, Meyer D, Chenaille P, Crom DB. The use of bone marrow aspirations and lumbar punctures at the time of diagnosis of retinoblastoma. J Clin Oncol. 1989;7:140-143.
17. Mohney BG, Robertson DM. Ancillary testing for metastasis in patients with newly diagnosed retinoblastoma. Am J Ophthalmol. 1994;118:707-711.
18. Kopelman JE, McLean IW, Rosenberg SH. Multivariate analysis of risk factors for metastasis in retinoblastoma treated by enucleation. Ophthalmology. 1987;94:371-377.
19. McCay CJ, Abramson DH, Ellsworth RM. Metastatic patterns of retinoblastoma. Arch Ophthalmol. 1984;102:391-396.
20. Abramson DH, Frank CM, Susman L, et al. Presenting signs of retinoblastoma. J Pediatr. 1998;25:210-211.
21. Goddard AG, Kingston JE, Hungerford JL. Delay in diagnosis of retinoblastoma: risk factors and treatment outcomes. Br J Ophthalmol. 1999;83:1320-1323.
22. Erwenne CM, Franco EL. Age and lateness of referral as determinants of extra-ocular retinoblastoma. Ophthalmic Genet. 1989;10:179-184.
23. Messmer EP, Heinrich T, Hopping W, et al. Risk factors for metastases in patients with retinoblastoma. Ophthalmology. 1991;98:136-141.
24. Stevenson KE, Hungerford JL, Garner A. Local extraocular extension of retinoblastoma following intraocular surgery. Br J Ophthalmol. 1989;73:739-742.
25. Shields CL, Honavar SG, Shields JA, et al. Vitrectomy in eyes with unsuspected retinoblastoma. In press.
26. Sanders BM, Draper GJ, Kingston JE. Retinoblastoma in Great Britain 1969-80: incidence, treatment, and survival. Br J Ophthalmol. 1983;72:576-583.
27. Abramson DH. Second nonocular cancers in retinoblastoma: a unified hypothesis–The Franceschetti Lecture. Ophthalmic Genet. 1999;20:193-204.
28. Draper GJ, Sanders BM, Kingston JE. Second primary neoplasms in patients with retinoblastoma. Br J Cancer. 1986;53:661-671.
29. Roarty JD, McLean IW, Zimmerman LE. Incidence of second neoplasms in patients with bilateral retinoblastoma. Ophthalmology. 1988;95:1583-1587.
30. Mohney BG, Robertson DM, Schomberg PJ, Hodge DO. Second nonocular tumors in survivors of heritable retinoblastoma and prior radiation therapy. Am J Ophthalmol. 1998;126:230-237.
31. Abramson DH, Frank CM. Second nonocular tumors in survivors of bilateral retinoblastoma: a possible age effect on radiation-related risk. Ophthalmology. 1998;105:573-579.
32. Wong FL, Boice ID, Abramson DH, et al. Cancer incidence after retinoblastoma. Radiation dose and sarcoma risk. JAMA. 1997;278:1262-1267.
33. Dunkel IJ, Gerald WL, Rosenfield NS, et al. Outcome of patients with a history of bilateral retinoblastoma treated for a second malignancy: the Memorial Sloan-Kettering experience. Med Pediatr Oncol. 1998;30:59-62.
34. Moll AC, Imhof SM, Bouter LM, Tan KEWP. Second primary tumors in patients with retinoblastoma. Ophthalmic Genet. 1997;18:27-34.
35. Singh AD, Shields CL, Shields JA. New insights into trilateral retinoblastoma. Cancer. 1999;86:3-5.
36. Blach LE, McCormick B, Abramson DH, Ellsworth RM. Trilateral retinoblastoma incidence and outcome: a decade of experience. Int J Radiat Oncol Biol Phys. 1994;29:729-733.
37. DePotter PV, Shields CL, Shields JA. Clinical variations of trilateral retinoblastoma: a report of 13 cases. J Pediatr Ophthalmol Strabismus. 1994;31:26-31.
38. Paulino AC. Trilateral retinoblastoma: is the location of the intracranial tumor important? Cancer. 1999;86:135-141.
39. Becker LE, Hinton D. Primitive neuroectodermal tumors of the central nervous system. Hum Pathol. 1983;14:538-550.
40. Marcus DM, Brooks SE, Leff G, et al. Trilateral retinoblastoma: insights into histogenesis and management. Surv Ophthalmol. 1998;43:59-70.
41. Kivela T. Trilateral retinoblastoma. A meta-analysis of hereditary retinoblastoma associated with primary ectopic intracranial retinoblastoma. J Clin Oncol. 1999;17:1829-1837.
42. Shields CL, Shields JA, Meadows AT. Chemoreduction for retinoblastoma may prevent trilateral retinoblastoma. J Clin Oncol. 2000;18:236-237. Letter.
43. Tosi P, Cintorino M, Toti P, et al. Histopathological evaluation for the prognosis of retinoblastoma. Ophthalmic Genet. 1989;10:173-177.
44. Khelfaoui F, Validire P, Auperin A, et al. Histopathologic risk factors in retinoblastoma: a retrospective study of 172 patients treated in a single institution. Cancer. 1996;77:1206-1213.
45. Magramm I, Abramson DH, Ellsworth RM. Optic nerve involvement in retinoblastoma. Ophthalmology. 1989;96:217-222.
46. Shields CL, Shields JA, Baez KA, Cater J, De Potter PV. Choroidal invasion of retinoblastoma: metastatic potential and clinical risk factors. Br J Ophthalmol. 1993;77:544-548.
47. Palazzi M, Abramson DH, Ellsworth RM. Endophytic vs exophytic unilateral retinoblastoma: is there any real difference? J Pediatr Ophthalmol Strabismus. 1990;27:255-258.
48. Ellsworth RM. Retinoblastoma: an overview. In: Blodi FC, ed. Contemporary Issues in Ophthalmology: Retinoblastoma. New York, NY: Churchill Livingston; 1985:1-10.
49. Wolter JR. Retinoblastoma extension into the choroid. Pathological study of the neoplastic process and thoughts about its prognostic significance. Ophthalmic Genet. 1987;8:151-157.
50. Stannard C, Lipper S, Sealy R, Sevel D. Retinoblastoma: correlation of invasion of the optic nerve and choroid with prognosis and metastasis. Br J Ophthalmol. 1979;63:560-570.
51. Rootman J, Hofbauer J, Ellsworth RM, Kitchen D. Invasion of the optic nerve by retinoblastoma: a clinicopathological study. Can J Ophthalmol. 1976;11:106-114.
52. Shields CL, Shields JA, Baez K, Cater JR, De Potter P. Optic nerve invasion of retinoblastoma. Metastatic potential and clinical risk factors. Cancer. 1994;73:692-698.
53. Chantada GL, de Davila MT, Fandino A, et al. Retinoblastoma with low risk for extraocular relapse. Ophthalmic Genet. 1999;20:133-140.
54. Shields JA, Shields CL, De Potter P. Enucleation technique for children with retinoblastoma. J Pediatr Ophthalmol Strabismus. 1992;29:213-215.
55. Rootman J, Ellsworth RM, Hofbauer J, Kitchen D. Orbital extension of retinoblastoma: a clinicopathological study. Can J Ophthalmol. 1978;13:72-80.
56. White L. The role of chemotherapy in the treatment of retinoblastoma. Retina. 1983;3:194-199.
57. Honavar SO, Singh AD, Shields CL, et al. Does post-enucleation prophylactic chemotherapy in high-risk retinoblastoma prevent metastasis? In press.
58. Wong FL, Boice JD, Abramson DH, et al. Cancer incidence after retinoblastoma. Radiation dose and sarcoma risk. JAMA. 1997;278:1262-1267.