135 Trilateral Retinoblastoma Is the Location of the Intracranial Tumor Important? Arnold C. Paulino, M.D. Department of Radiotherapy, Cardinal Bernardin Cancer Center and Ronald McDonald Children’s Hospital, Loyola University Medical Center, Maywood, Illinois. Address for reprints: Arnold C. Paulino, M.D., Division of Radiation Oncology, The University of Iowa Hospitals and Clinics, 200 Hawkins Drive, Room W189-Z, Iowa City, IA 52242. Received June 15, 1998; revision received October 19, 1998; accepted December 10, 1998. © 1999 American Cancer Society BACKGROUND. Trilateral retinoblastoma refers to bilateral retinoblastoma associated with an intracranial primitive neuroectodermal tumor in the pineal or suprasellar region. The purpose of this study was to review patient and tumor characteristics and treatment outcome in patients with trilateral retinoblastoma and to determine whether there is a difference in presentation or outcome according to the location of the intracranial tumor. METHODS. A MEDLINE search of all English language articles pertaining to trilateral retinoblastoma published between 1977–1997 was performed. A total of 94 different cases were identified and analyzed. RESULTS. The gender was male in 39 patients (41.5%), female in 50 patients (53.2%), and unknown in 5 patients (5.3%). Family history for retinoblastoma was positive in 44 patients (46.8%), negative in 39 patients (41.5%), and unknown in 11 patients (11.7%). The median age at the time of diagnosis of retinoblastoma was 6 months. The median time interval to the development of an intracranial tumor from the time of diagnosis of retinoblastoma was 21 months. In 78 patients (83.0%) the intracranial tumor was in the pineal region and in 16 patients (17.0%) it was in the suprasellar region. The median time interval from the time of diagnosis of retinoblastoma to the development of a pineal region tumor was 24 months whereas the median time interval for the development of a suprasellar region tumor was 1 month. At 6 months after the diagnosis of intraocular tumors, 6 of 61 children with pineal region tumors and 10 of 14 patients with suprasellar region tumors had developed intracranial disease (P 5 0.005). Unilateral intraocular retinoblastoma associated with intracranial tumor was more likely to occur in patients with suprasellar region tumors than pineal region tumors (P , 0.015). The median survival after the diagnosis of an intracranial tumor was 6 months regardless of the location of the intracranial tumor. For patients who received no treatment for the intracranial tumor the median survival was 1 month whereas it was 8 months for those who received treatment. Children who were asymptomatic at the time of diagnosis of the intracranial tumor had a better overall survival than those who were symptomatic (P 5 0.002). CONCLUSIONS. The prognosis of children who develop trilateral retinoblastoma is dismal with current treatment strategies. Tumors of the suprasellar region present earlier than tumors of the pineal region after the diagnosis of intraocular tumors. Because patients who were asymptomatic at the time of diagnosis of intracranial disease had a better overall survival than those who were symptomatic, screening for intracranial tumors may be a valuable strategy in the management of patients with bilateral and/or hereditary retinoblastoma. [See editorial on pages 3–5, this issue.] Cancer 1999;86:135– 41. © 1999 American Cancer Society. KEYWORDS: trilateral retinoblastoma, intracranial tumor, childhood cancer, radiotherapy, chemotherapy, screening. 136 CANCER July 1, 1999 / Volume 86 / Number 1 T he trilateral retinoblastoma syndrome is a rare condition that traditionally refers to bilateral retinoblastoma in association with an ectopic intracranial tumor indistinguishable histologically from the intraocular tumors. The association first was described in detail by Jakobiec et al. in 19771 and the term “trilateral retinoblastoma” was popularized in 1980 by Bader et al.2 When Bader et al. described their initial ten cases with trilateral retinoblastoma, the intracranial tumors identified were located in the pineal gland region.2 Since that time, multiple reports have identified a location in the suprasellar region associated with bilateral retinoblastoma. The purpose of this study was to determine whether there is a difference in presentation or outcome in relation to the location of the intracranial tumor in children with trilateral retinoblastoma. An additional aim of the study was to provide patient and tumor characteristics, pattern of failure analysis, and treatment outcome in patients with this unusual entity. METHODS A MEDLINE search of all English language articles between 1970 –1997 pertaining to trilateral retinoblastoma was performed and identified a total of 125 cases.1,3-42 Table 1 lists the authors, institutions, and the number of cases described in each study. Overlap of case reports occurred in some patients2– 4,12–13, 16 –17,19 –23,25–26,28,30 –31,33,35–37,39 – 40,42 and after correction for repetition of cases a total of 94 different children with trilateral retinoblastoma was identified. Special attention was paid to characteristics of the host and tumor, treatment, patterns of failure, and outcome; family history of retinoblastoma also was examined. An attempt was made to contact the first authors of the articles when important information regarding the patient and tumor characteristics was lacking. The Fisher exact test was employed to determine whether there was a difference in the time interval from the time of the diagnosis of retinoblastoma to the occurrence of an intracranial tumor between children who developed tumors of the pineal and suprasellar regions. Likewise the same test was used to determine whether unilateral intraocular retinoblastomas with germline mutations were more likely to be associated with the suprasellar region than pineal tumors. Survival estimates of children who were symptomatic and asymptomatic were calculated using the Kaplan– Meier method43; the log rank test was employed to compare survival curves of patients who were symptomatic and asymptomatic at the time of diagnosis for intracranial disease. RESULTS Patient and Tumor Characteristics Gender was male in 39 patients (41.5%), female in 50 patients (53.2%), and unknown in 5 patients (5.3%). A family history of retinoblastoma was found to be positive in 44 patients (46.8%), negative in 39 patients (41.5%), and unknown in 11 (11.7%). The median age at the time of diagnosis of an intraocular tumor was 6 months. Eighty-five patients had bilateral retinoblastoma associated with an intracranial tumor. Nine of the 85 children with bilateral retinoblastoma who subsequently developed an intracranial primitive neuroectodermal tumor (PNET) had asynchronous intraocular tumors. Laterality of initial ocular involvement was known in eight of nine asynchronous patients; four had initial right eye involvement and four had initial left eye involvement. Seven children with trilateral retinoblastoma who were reported in the literature had a unilateral retinoblastoma with germline mutation associated with an intracranial tumor. In two patients, the unilateral intraocular tumor was diagnosed at the same time as the intracranial tumor; for the remaining five patients, the time interval from the diagnosis of the eye tumor to the development of the brain tumor ranged from 4 –30 months (median, 12 months). Of the seven patients with unilateral retinoblastoma and an associated intracranial malignancy, four had right intraocular tumors and three had left intraocular tumors; four children had tumors of the suprasellar region and three children had tumors of the pineal region. A unilateral intraocular tumor associated with an intracranial PNET was more likely to be found in the suprasellar region than in the pineal region (4 of 16 patients vs. 3 of 78 patients; P 5 0.015). Two patients had intracranial tumors (one in the suprasellar region and the other in the pineal region) in the absence of intraocular disease. Both patients had a sibling with bilateral retinoblastoma; one sibling had a tumor in the pineal region in addition to bilateral intraocular tumors. Intracranial Tumor Intracranial tumor location was known in all patients. Seventy-eight children (83.0%) had tumor of the pineal region and 16 (17.0%) had a PNET of the suprasellar region. The median time interval from the diagnosis of an intraocular tumor to the development of an intracranial neoplasm for 75 evaluable children was 21 months. Figure 1 shows the various time intervals for all evaluable patients according to the location of the intracranial tumor. Trilateral Retinoblastoma/Paulino 137 TABLE 1 Studies Reporting Cases of Trilateral Retinoblastoma Authors Year of publication Institution No. of cases Jakobiec et al.1 Bullitt and Crain3 Judisch and Shivanand4 Bader et al.5 Zimmerman et al.6 Dudgeon and Lee7 Ehlers et al.8 Brownstein et al.9 Michaud et al.10 Tarkkanen et al.11 Johnson et al.12 Kingston et al.13 Stannard et al.14 Whittle et al.15 Lueder et al.16 Malik et al.17 Meadows18 Towbin and Bisset19 Helveston et al.20 Bomanji et al.21 Pesin and Shields22 Shields et al.23 Schwartz et al.24 Holladay et al.25 Lueder et al.26 Mukherjee et al.27 Nelson et al.28 Stolovitch et al.29 Marks et al.30 Scott and Richard31 Blach et al.32 DePotter et al.33 Finelli et al.34 McCormack et al.35 Nunna et al.36 Provenzale et al.37 Amoaku et al.38 Bagley et al.39 Bejjani et al.40 Moll et al.41 Skulski et al.42 1977 1981 1981 1982 1982 1983 1983 1984 1984 1984 1985 1985 1985 1985 1986 1986 1986 1986 1987 1989 1989 1989 1990 1991 1991 1991 1992 1992 1993 1993 1994 1994 1995 1995 1995 1995 1996 1996 1996 1996 1997 George Washington University, Washington, DC Duke University, Durham, NC University of Iowa, Iowa City, IA National Cancer Institute, Bethesda, MD University of Missouri, Columbia, MO University of Glasgow, Glasgow, United Kingdom University of Aarhus, Aarhus, Denmark McGill University, Montreal, Canada University of Montreal, Montreal, Canada University of Helsinki, Helsinki, Finland Children’s Hospital National Medical Center, Washington, DC St. Bartholomew’s Hospital, London, United Kingdom University of Cape Town, Cape Town, South Africa Royal Alexandria Hospital for Children, Sydney, Australia University of Iowa, Iowa City, IA Duke University, Durham, NC Children’s Hospital of Philadelphia, Philadelphia, PA University of Cincinnati, Cincinnati, OH Indiana University, Indianapolis, IN St. Bartholomew’s Hospital, London, United Kingdom Wills Eye Hospital, Philadelphia, PA Wills Eye Hospital, Philadelphia, PA Medical College of Virginia, Richmond, VA Indiana University, Indianapolis, IN and University of Cincinnati, Cincinnati, OH University of Iowa, Iowa City, IA Diwan Aggarwal Research Center, New Delhi, India Duke University, Durham, NC Ichilov Hospital Tel Aviv, Israel Duke University, Durham, NC University of Oklahoma, Oklahoma City, OK Memorial Sloan-Kettering Cancer Center, New York, NY Wills Eye Hospital, Philadelphia, PA Case Western University, Cleveland, OH University of Oklahoma, Oklahoma City, OK All India Institute, New Delhi, India Duke University, Durham, NC West Midlands Regional Children’s Tumour, Research Group, Birmingham, United Kingdom Wills Eye Hospital, Philadelphia, PA George Washington University, Washington, DC Dutch Retinoblastoma Register, Amsterdam, The Netherlands University of Cincinnati, Cincinnati, OH 2 1 1 14 1 2 1 2 1 1 3 12 1 1 3 2 1 1 1 1 7 1 1 3 4 1 3 1 1 3 6 13 2 1 1 1 5 10 1 7 1 For the 16 patients with suprasellar tumors, 9 patients (56.3%) were female, 6 patients (37.5%) were male, and the gender was unknown in 1 patient (6.3%). The median time interval from the diagnosis of an ocular tumor to the development of an intracranial tumor was 1 month. In two children, the tumor in the suprasellar region preceded bilateral intraocular tumors. Six children (37.5%) with tumors of the suprasellar region had a positive family history of retinoblastoma. For the 78 patients with pineal region tumors, the gender was female in 41 patients (52.6%), male in 33 patients (42.3%), and unknown in 4 patients (5.1%). The median time interval from the diagnosis of an intraocular tumor to the development of a pineal region tumor was 24 months. Thirty-eight children (48.7%) with tumors of the pineal region had a positive family history of retinoblastoma. At 6 months after the diagnosis of intraocular tumors, 6 of 61 evaluable children with tumors of the pineal region and 10 of 14 children with tumors of the suprasellar region had developed intracranial disease (P 5 0.005). 138 CANCER July 1, 1999 / Volume 86 / Number 1 Treatment Outcome and Pattern of Failure The median survival after a diagnosis of intracranial tumor for all patients was 6 months. The median survivals for children with tumors of the pineal region (range, 0 –96 months) and PNETs of the suprasellar region (range, 0 –23 months) both were 6 months. The median survival for children who received no treatment for the intracranial tumor was 1 month whereas it was 8 months for children who were treated for the intracranial tumor. Table 2 lists the 6 patients who survived at least 2 years after the diagnosis of an intracranial tumor. All six children had a pineal region tumor. Treatment included chemotherapy in all six children and craniospinal irradiation (CSRT) was given to four patients. The type of treatment given for the intracranial tumor in all patients is shown in Table 3. The most common type of treatment was chemotherapy in combination with radiation therapy (RT) followed by RT alone. Eighteen patients received CSRT; 15 of these patients developed disseminated neuraxis disease despite CSRT. Forty-one of the 75 evaluable patients (54.7%) died with disseminated neuraxis disease. Survival rates for children who were symptomatic and those who were asymptomatic at the time of diagnosis of intracranial disease were compared and showed a statistically significant survival advantage for patients with asymptomatic intracranial disease (Fig. 2) (P 5 0.002). DISCUSSION In this review of trilateral retinoblastoma from 1977– 1997, four groups of children were identified with regard to the presentation of intraocular tumors in relation to the intracranial tumor. The most common presentation found in 76 of 94 children (80.1%) was the “traditional” definition of trilateral retinoblastoma: synchronous bilateral intraocular tumors observed in association with a PNET of the pineal or suprasellar region. Bilateral retinoblastoma may be asynchronous and may be associated with an intracranial tumor. Such was the case in 9 of 94 children (9.6%). The third group of children included 7 patients (7.4%) with a unilateral retinoblastoma with a germline mutation and intracranial PNET; this first was described by Whittle et al.,15 who coined the term “forme fruste” of trilateral retinoblastoma for this presentation. Our study showed that tumors of the suprasellar region were more likely to be associated with unilateral intraocular tumors when compared with tumors of the pineal region. Finally, two children with germline mutations did not have an intraocular tumor but had an intracranial PNET; one patient had a sib- ling with bilateral retinoblastoma and the other had an identical twin with trilateral retinoblastoma. The pineal region presentation is more frequent than the suprasellar region presentation. There appeared to be no difference based on gender or family history between patients with tumors in the suprasellar region and those with tumors in the pineal region. Regardless of intracranial location, trilateral retinoblastoma was detected more frequently in girls than boys. The median time interval from the diagnosis of an intraocular tumor to the development of an intracranial tumor was 21 months. Suprasellar region tumors presented earlier than pineal region PNETs. It appears that 89% and 98.7% of all patients with trilateral retinoblastoma developed intracranial disease at 4 years and 7 years, respectively, from the time of the initial diagnosis of an intraocular tumor. This information is important for physicians who treat children with bilateral retinoblastoma because screening for intracranial tumors would be most helpful during the first 4 years after the diagnosis of an intraocular tumor. A small but significant proportion of children with bilateral or hereditary retinoblastoma develop intracranial disease. At the Memorial Sloan-Kettering Cancer Center, 6% of patients with bilateral retinoblastoma and 10% of patients with a family history of retinoblastoma developed intracranial PNET.32 Another report from the Wills Eye Hospital showed that trilateral retinoblastoma occurred in 13 of 440 consecutive children with retinoblastoma (3%).33 Some may argue that screening for trilateral retinoblastoma may not be very helpful given the median survival of only 6 months and the small number of patients with bilateral or hereditary retinoblastoma who develop intracranial disease. However, in 1 study trilateral retinoblastoma accounted for 6 of 12 deaths (50%) during the first 5 years after the diagnosis of retinoblastoma.32 Trilateral retinoblastoma is an underappreciated cause of mortality in the first 5 years after the diagnosis of bilateral retinoblastoma. Screening for intracranial disease in children with bilateral and hereditary retinoblastoma may be beneficial because our study showed that patients who were asymptomatic at the time of diagnosis of intracranial disease had a better survival compared with those who were symptomatic. Screening may be performed by the use of magnetic resonance imaging of the brain. It is not clear how often brain imaging should be performed except that it should be performed for at least 4 years after the diagnosis of intraocular tumors. In a report from Duke University, the 2 longest survivors (at 33 months and 96 months, respectively, after the diagnosis of an intracranial tumor) were diagnosed using routine imaging of the Trilateral Retinoblastoma/Paulino 139 TABLE 3 Type of Treatment of Intracranial Tumors in the 94 Children with Trilateral Retinoblastomas CSRT 1 chemotherapy CSRT alone No treatment Unknown Chemotherapy alone CSRT 1 surgery Chemotherapy 1 surgery CSRT 1 chemotherapy 1 surgery Surgery alone FIGURE 1. Time intervals between the diagnosis of an intraocular tumor and 25 20 16 12 10 6 2 2 1 RT: radiation therapy. the development of an intracranial neoplasm according to the location of the intracranial tumor in 75 evaluable children with trilateral retinoblastoma. TABLE 2 Treatment Characteristics of Patients Surviving at Least 2 Years after the Diagnosis of an Intracranial Tumor Case no. Survival (mos) Treatment 4 40 41 45 69 78 31 96 33 26 30 28 CSRT 1 CCNU, VCR CPM, MTX, ARA-C, HYDRO CSRT 1 CPM, ARA-C, HYDRO CSRT 1 CPM, MTX, VCR, DOX Unspecified chemotherapy CSRT 1 unspecified chemotherapy CSRT: craniospinal irradiation; CCNU: lomustine; VCR: vincristine; CPM: cyclophosphamide; MTX: methotrexate; ARA-C: cytosine arabinoside; HYDRO: hydrocortisone; DOX: doxorubicin. brain prior to the development of symptoms.28 Data from the Wills Eye Hospital also argue for the screening of children with bilateral retinoblastoma. In a series of 13 patients, children who were asymptomatic had a significantly longer survival compared with those who were symptomatic (P 5 0.05).33 DePotter et al. recommend screening for intracranial tumors in children with bilateral and/or familial retinoblastoma until age 4 years.33 This recommendation can be modified further using the results of our analysis, which showed that 89% of patients will have developed an intracranial tumor 4 years after the initial diagnosis of intraocular retinoblastoma. Treatment for trilateral retinoblastoma resulted in a median survival of 8 months. There was no difference in survival outcome regardless of whether the tumor was in the pineal versus the suprasellar region. Of the 6 children who survived . 2 years after a diagnosis of intracranial PNET, all received chemotherapy and 4 received CSRT. For the 75 evaluable children in whom the pattern of failure was reported, 41 (54.7%) had disseminated neuraxis disease. Only 3 FIGURE 2. Comparison of overall survival rates after a diagnosis of intracranial tumor according to symptomatology. Children who were asymptomatic for intracranial disease at the time of diagnosis had a better overall survival than children who were symptomatic at the time of diagnosis (P 5 0.002). of 18 children who received CSRT achieved control of tumor dissemination along the neuraxis; 2 of these children received chemotherapy. Therefore, efforts in achieving a cure for this tumor should be directed toward the prevention of tumor dissemination along the neuraxis. The use of chemotherapy and CSRT appears to be reasonable because this form of treatment was used in four of six “long-term” survivors. Furthermore, the use of chemotherapy and CSRT resulted in 3-year overall and progression free survival rates of 57% and 45%, respectively, in children with primary supratentorial PNET treated on a Children’s Cancer Group protocol.44 In a retrospective study of 36 children at the Hospital for Sick Children in Toronto, all six survivors with primary PNET of the brain received CSRT and four received chemotherapy.45 Certainly, there are limitations in this report. First, this is a 20-year review of patients treated using a variety of therapeutic approaches. The data gathered 140 CANCER July 1, 1999 / Volume 86 / Number 1 from each case were not all complete because some articles did not evaluate all the patient, tumor, and treatment characteristics. Another potential problem with this study is that we do not know whether the early detection of intracranial disease in asymptomatic patients produces a lag time bias similar to that observed in the detection of women with breast carcinoma using mammography. The median survival of children with trilateral retinoblastoma is dismal regardless of the location of the intracranial tumor. Pineal region tumors occur more frequently than suprasellar PNET. Children with tumors of the suprasellar region develop intracranial disease earlier and are more likely to have intracranial disease associated with unilateral intraocular retinoblastoma with germline mutations when compared with children with tumors of the pineal region. In children with bilateral or hereditary retinoblastoma, screening for intracranial disease for at least 4 years may be warranted because patients who are asymptomatic at the time of diagnosis of an intracranial tumor have been found to have a better overall survival than patients who were symptomatic. Efforts to control this disease should be directed toward the prevention of tumor dissemination along the neuraxis, which is a pattern of failure observed in . 50% of patients. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. REFERENCES 1. Jackobiec FA, Tso MOM, Zimmerman LE, Davis P. Retinoblastoma and intracranial malignancy. Cancer 1977;39: 2048 –58. 2. Bader JL, Miller RW, Meadows AT, Zimmerman LE, Champion LAA, Voute PA. Trilateral retinoblastoma. Lancet 1980; 2:582–3. 3. Bullitt E, Crain BJ. Retinoblastomas as a possible primary intracranial tumor. Neurosurgery 1981;9:706 –9. 4. Judisch GF, Shivanand RP. Concurrent hereditable retinoblastoma, pinealoma and trisomy X. Arch Ophthalmol 1981; 99:1767–9. 5. Bader JL, Meadows AT, Zimmerman LE, Rorke LB, Voute PA, Champion LAA, et al. Bilateral retinoblastoma with ectopic intracranial retinoblastoma: trilateral retinoblastoma. Cancer Genet Cytogenet 1982;5:203–13. 6. Zimmerman LE, Burns RP, Wankum G, Tully R, Esterly JA. Trilateral retinoblastoma: ectopic intracranial retinoblastoma associated with bilateral retinoblastoma. J Pediatr Ophthalmol Strabismus 1982;19:320 –5. 7. Dudgeon J, Lee WR. The trilateral retinoblastoma syndrome. Trans Ophthalmol Soc UK 1983;103:523–9. 8. Ehlers N, Kaal S, Rasmussen K, Ratjen E. Hereditary bilateral retinoblastoma, pinealoma and normal chromosomes: a case report. Acta Opthalmol (Copenh) 1983;61:838 – 43. 9. Brownstein S, de Chadarevian JP, Little JM. Trilateral retinoblastoma: report of two cases. Arch Ophthalmol 1984;102: 257– 62. 10. Michaud J, Jacob JL, Demers J, Dumas J. Trilateral retino- 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. blastoma: bilateral retinoblastoma with pinealoblastoma. Can J Ophthalmol 1984;19:36 –9. Tarkkanen A, Haltia M, Karjalainin K. Trilateral retinoblastoma. Pineoblastoma (ectopic intracranial retinoblastoma) associated with bilateral retinoblastoma. Ophthalmic Paediatr Genet 1984;4:1– 6. Johnson DL, Chandra R, Fisher WS, Hammock MK, McKeown CA. Trilateral retinoblastoma: ocular and pineal retinoblastomas. J Neurosurg 1985;63:367–70. Kingston JE, Plowman PN, Hungerford JL. Ectopic intracranial retinoblastoma in childhood. Br J Ophthalmal 1985;69: 742– 8. Stannard C, Knight BK, Sealy R. Pineal malignant neoplasm in association with hereditary retinoblastoma. Br J Ophthalmol 1985;69:749 –53. Whittle IR, McClellan K, Martin FJ, Johnson IHl. Concurrent pineoblastoma and unilateral retinoblastoma: a forme fruste of trilateral retinoblastoma? Neurosurgery 1985;17: 500 –5. Lueder GT, Judisch GF, O’Gorman TW. Second nonocular tumors in survivors of heritable retinoblastoma. Arch Ophthalmol 1986;104:372–3. Malik RK, Friedman HS, Djang WT, Falletta JM, Buckley E, Kurtzberg J, et al. Treatment of trilateral retinoblastoma with vincristine and cyclophosphamide. Am J Ophthalmol 1986;102:650 – 6. Meadows A. Trilateral retinoblastoma. Med Pediatr Oncol 1986;14:323– 6. Towbin RB, Bisset GS. Pediatric case of the day. Radiographics 1986;6:1100 – 4. Helveston EM, Knuth KR, Ellis FD. Retinoblastoma. J Pediatr Ophthalmol Strabismus 1987;24:296 –300. Bomanji J, Kingston JE, Hungerford JL, Britton KE. 123 IMetaiodobenzylguanidine scintigraphy of ectopic intracranial retinoblastoma. Med Pediatr Oncol 1989;17:66 – 8. Pesin SR, Shields JA. Seven cases of trilateral retinoblastoma. Am J Ophthalmol 1989;107:121– 6. Shields JA, Pesin SR, Shields CL. Trilateral retinoblastoma. J Clin Neuro Ophthalmol 1989;9:222–3. Schwartz AM, Ghatak NR, Laine FJ. Intrasellar primitive neuroectodermal tumor (PNET) in familial retinoblastoma: a variant of “trilateral retinoblastoma”. Clin Neuropathol 1990;9:55–9. Holladay DA, Holladay A, Montebello JF, Redmond KP. Clinical presentation, treatment, and outcome of trilateral retinoblastoma. Cancer 1991;67:710 –5. Lueder GT, Judisch GF, Wen BC. Heritable retinoblastoma and pinealoma. Arch Ophthalmol 1991;109:1707–9. Mukherjee M, Gothi R, Doda SS, Aggarwal SK. Trilateral retinoblastoma. AJR Am J Roentgenol 1991;157:198 –9. Nelson SC, Friedman HS, Oakes WJ, Halperin EC, Tien R, Fuller GN, et al. Successful therapy for trilateral retinoblastoma. Am J Ophthalmol 1992;114:23–9. Stolovitch C, Loewenstein A, Varssano D, Lazar M. Trilateral retinoblastoma. Metab Pediatr Syst Ophthalmol 1992;15: 57–9. Marks LB, Bentel G, Sherouse GW, Spencer DP, Light K. Craniospinal irradiation for trilateral retinoblastoma following ocular irradiation. Med Dosim 1993;18:125– 8. Scott MH, Richard JM. Retinoblastoma in the state of Oklahoma: a clinicopathologic review. J Okla State Med Assoc 1993;86:111– 8. Trilateral Retinoblastoma/Paulino 32. 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 –33. 33. DePotter P, Sheilds CL, Shields JA. Clinical variations of trilateral retinoblastoma: a report of 13 cases. J Pediatr Ophthalmol Strabismus 1994;31:26 –31. 34. Finelli DA, Shurin SB, Bardenstein DS. Trilateral retinoblastoma: two variations. AJNR Am J Neuroradiol 1995;16:166–70. 35. McCormack S, Raftery R, Palmgren D, Leonard JC. Pineoblastoma imaged during bone scintigraphy. Clin Nucl Med 1995;20:1023– 4. 36. Nunna SVV, Sharma R, Vashisht S, Berry M. Trilateral retinoblastoma: CT evaluation. Indian J Ophthalmol 1995;43:27–9. 37. Provenzale JM, Weber AL, Klintworth GK, McLendon RE. Radiographic-pathologic correlation. Bilateral retinoblastoma with coexistent pineoblastoma (trilateral retinoblastoma). AJNR Am J Neuroradiol 1995;16:157– 65. 38. Amoaku WMK, Willshaw HE, Parkes SE. Trilateral retinoblastoma. A report of five patients. Cancer 1996;78:858 – 63. 39. Bagley LJ, Hurst RW, Zimmerman RA, Shields JA, Shields CL, DePotter P. Imaging in trilateral retinoblastoma syndrome. Neuroradiology 1996;38:166 –70. 141 40. Bejjani GK, Donahue DJ, Selby D, Cogen PH, Packer R. Association of a suprasellar mass and intraocular retinoblastoma: a variant of pineal trilateral retinoblastoma? Ped Neurosurg 1996;25:269 –75. 41. Moll AC, Imhaf SM, Bouter LM, Kuik DJ, Den Otter W, Bezember PD, et al. Second primary tumors in patients with hereditary retinoblastoma: a register based follow-up study, 1945-1994. Int J Cancer 1996;67:515–9. 42. Skulski M, Egelhoff JC, Kollias SS, Mazeniski C, Ball WS. Trilateral retinoblastoma with suprasellar involvement. Neuroradiology 1997;39:41–3. 43. Kaplan EL, Meier P. Nonparametric estimation from the incomplete observations. J Am Stat Assoc 1958;54:457– 81. 44. Cohen BH, Zeltzer PM, Boyett JM, Geyer JR, Allen JC, Finlay JL, et al. Prognostic factors and treatment results for supratentorial primitive neuroectodermal tumors in children, using radiation and chemotherapy: a Children’s Cancer Group randomized trial. J Clin Oncol 1995;13:1687–96. 45. Dirks PB, Harris L, Hoffman HJ, Humphreys RP, Drake JM, Rutka JT. Supratentorial primitive neuroectodermal tumors in children. J Neurooncol 1996;29:75– 84.