Extraocular retinoblastoma: A 13-year experience

1292

Extraocular Retinoblastoma: A 13-Year Experience Ce´lia Beatriz Gianotti Antoneli, M.D., Ph.D.1 Fla´vio Steinhorst, M.D.1 Karina de Ca´ssia Braga Ribeiro, D.D.S., Ph.D.2 Paulo Eduardo R. S. Novaes, M.D., Ph.D.3 Martha M. M. Chojniak, M.D.4 Victor Arias, M.D.5 Beatriz de Camargo, M.D., Ph.D.1 1

Department of Pediatric Oncology, Hospital do Caˆncer, A. C. Camargo, Centro de Tratamento e Pesquisa, Sa˜o Paulo, Brazil.

2

Hospital-Based Pediatric Cancer Registry, Hospital do Caˆncer, A. C. Camargo, Centro de Tratamento e Pesquisa, Sa˜o Paulo, Brazil.

3

Department of Radiation therapy, Hospital do Caˆncer, A. C. Camargo, Centro de Tratamento e Pesquisa, Sa˜o Paulo, Brazil.

4

Department of Ophthalmology, Hospital do Caˆncer, A. C. Camargo, Centro de Tratamento e Pesquisa, Sa˜o Paulo, Brazil.

5

Department of Pathology, Hospital do Caˆncer, A. C. Camargo, Centro de Tratamento e Pesquisa, Sa˜o Paulo, Brazil

BACKGROUND. The current study was performed to evaluate two regimens of treatment and to describe clinical and epidemiologic characteristics in patients with extraocular retinoblastoma. METHODS. Eighty-three patients with extraocular retinoblastoma according to Childrens Cancer Group (CCG) classification were admitted to the Pediatric Department of the A. C. Camargo between 1987–2000. The age, gender, race, lag time, first clinical presentation, staging, laterality, and treatment regimen were analyzed. Treatment was comprised of cisplatin, teniposide, vincristine, doxorubicin, and cyclophosphamide during the first treatment period (1987–1991) or cisplatin and teniposide with alternating courses of ifosfamide and etoposide during the second treatment period (1992–2000).

RESULTS. The mean age of the patients was 32.9 months (range, 2–145 months). The mean lag time was 10.5 months. Forty-three patients were treated in the first period and 40 patients were treated in the second period. Locally advanced tumors (Class I–III) were present in 83.1% of the patients. There was a positive correlation between lag time and age for unilateral tumors (correlation coefficient [r] ⫽ 0.35; P ⫽ 0.006), whereas the correlation was negative for bilateral tumors (r ⫽ -0.12; P ⫽ 0.63). The 5-year overall survival was 55.1% in the first treatment period and 59.4% in the second treatment period (P ⫽ 0.69). No significant differences with regard to survival rates were noted for unilateral tumors between the two treatment periods (44.6 noted for unilateral tumors vs. 59.1 noted for unilateral tumors). CONCLUSIONS. In the current study, the addition of ifosfamide and etoposide to a treatment regimen comprised of cisplatin, teniposide, vincristine, doxorubicin, and cyclophosphamide did not appear to improve the survival of patients with extraocular retinoblastoma. Patients with dissemination to the central nervous system or metastatic disease remain incurable and die of progressive disease, despite the aggressive treatment. A multicenter trial should be considered to evaluate the best strategy for these situations. Cancer 2003;98:1292– 8. © 2003 American Cancer Society.

KEYWORDS: extraocular retinoblastoma, classification, treatment regimens, combined chemotherapy, prognosis.

R

The authors thank Victor Santana, M.D. and Marcio H. Malogolowkin, M.D. for their helpful comments. Address for reprints: Ce´lia Beatriz Gianotti Antoneli, M.D., Ph.D., Department of Pediatric Oncology, Hospital do Caˆncer, A. C. Camargo, Centro de Tratamento e Pesquisa, Rua Vergueiro 3645, apto511 Cep 04101-300, Sa˜o Paulo, Brazil; Fax: (011) 55 11 3272 5088; E-mail: cantoneli@ terra.com.br Received December 31, 2002; revision received June 12, 2003; accepted June 20, 2003. © 2003 American Cancer Society DOI 10.1002/cncr.11647

etinoblastoma is the most common intraocular malignancy diagnosed in children, accounting for 3% of all childhood malignant tumors in developed countries.1 Advanced and metastatic tumors occur frequently in developing countries,2 and late referral might account for the delayed diagnosis.3 The survival of patients with retinoblastoma has gradually improved over the years,4 in part because of the introduction of a multimodality approach. When tumor extends beyond the globe into the orbit, a combination of radiation therapy and chemotherapy has been used.5 Multidrug treatment strategies warrant further investigation to improve outcome as well as minimize toxicity.6 Prognosis remains relatively poor for patients whose disease disseminates into the central nervous system (CNS) and those with distant metastatic disease.7

Extraocular Retinoblastoma/Gianotti Antoneli et al. TABLE 1 CCG Classification for Extraocular Retinoblastoma Class I Class II Class III Class IV Class V

Microscopic involvement of the scleral emissaries Microscopic involvement of the cut end of the optic nerve Orbital disease in the biopsy CNS disease with brain mass or CSF with positive tumor cells Blood-born metastases to bone marrow, bone, or lymphatic metastases to lymph nodes

CCG: Childrens Cancer Group, CNS: central nervous system; CSF: cerebrospinal fluid.

The high response rate observed with the combination of ifosfamide and etoposide in other embryonal neuroectodermal tumors occurring in children8,9 encouraged us to test a combination of this pair of drugs in the second treatment period in patients with extraocular retinoblastoma. According to Pratt et al.,8 cyclophosphamide and ifosfamide produce measurable antitumor effects in patients with metastatic retinoblastoma. The same authors reported the activity of single drugs and drug combinations in 11 cases of extraocular retinoblastoma. Responses to vincristine, cyclophosphamide, doxorubicin, cisplatin, teniposide, and ifosfamide as single agents or in combination were observed.8 The combination of etoposide and carboplatin in the treatment of patients with extraocular retinoblastoma was associated with a very high response rate.7 We decided to incorporate an upfront “window” with ifosfamide and etoposide to a regimen that included cisplatin and teniposide to improve the overall survival in children with extraocular disease. The current study presents the clinical and epidemiologic characteristics of patients with extraocular retinoblastoma who presented to a single institution in Brazil and the results of two treatment programs with chemotherapy, surgery, and radiation therapy.

MATERIALS AND METHODS Eighty-three children with newly diagnosed extraocular retinoblastoma who were classified according Childrens Cancer Group (CCG) classification criteria10 (Table 1) were admitted to the Pediatrics and Ophthalmology Departments at the Centro de Tratamento e Pesquisa Hospital do Caˆncer A. C. Camargo in Sa˜o Paulo, Brazil between January 1987 and December 2000, and accounted for 20% of all new retinoblastoma patients (n ⫽ 417) in the same period. All patients were evaluated at the time of diagnosis with a complete blood count, serum blood urea nitrogen (BUN) and creatinine levels, liver function tests, cerebrospinal fluid (CSF) examination, bone marrow aspiration, bone scan, and orbital computed tomography (CT) scan. A single pathologist reviewed all pathologic material. Staging in patients with bilateral disease

1293

was based on the most compromised eye. None of the patients had bilateral extraocular disease. The contralateral eye was treated according to the Reese–Ellsworth staging system. For the purpose of the current study, patients were divided into 2 groups based on the treatment period: first period (1987–1991) (43 patients) and second period (1992–2000) (40 patients). During the first period, the treatment approach was as follows: for patients with Class I tumors, treatment was comprised of enucleation followed by cyclophosphamide at a dose of 30 mg/kg/day or 900 mg/m2/day ⫻ 1 and vincristine at a dose of 0.05 mg/kg/day ⫻ 1 or 1.5 mg/m2/day every 21 days for 10 cycles. None of patients with Class I tumors received orbital external radiation therapy. For patients with Class II-V tumors, treatment was comprised of enucleation at the time of diagnosis for Class II tumors only followed by cisplatin, teniposide, doxorubicin, cyclophosphamide, and vincristine (Table 2) and external radiation therapy to the orbit (approximately 40 –50 grays [Gy]) concomitant to chemotherapy. Intrathecal methotrexate (MTX) at a dose of 0.4 mg/kg/dose (maximum dose, 10 mg) was given every 6 weeks. Patients with Class III–V tumors underwent enucleation after three cycles of induction chemotherapy. During the second treatment period, in patients with Class I disease, we used enucleation at the time of diagnosis and the same chemotherapy schedule as used for patients with Class II-V tumors. For Class II-V tumors, patients were treated with ifosfamide at a dose of 1.8 g/m2/day ⫻ 5 and etoposide at a dose of 100 mg/m2/day ⫻ 5 (3 cycles), alternating with cisplatin at a dose of 90 mg/m2/day ⫻ 1 and tenoposide at a dose of 100 mg/m2/day ⫻ 1 every 21 days until the completion of 34 weeks (Table 3). Enucleation was performed at the time of diagnosis in patients with Class II tumors, whereas patients with Class III–V tumors underwent enucleation after three cycles of induction chemotherapy. External radiation therapy to the orbit (at a dose of 40 –50 Gy) concomitant to chemotherapy also was prescribed. Intrathecal chemotherapy comprised of MTX and cytarabine was administered according to age (age ⬍ 2 years, MTX at a dose of 6.0 mg plus cytarabine at a dose of 12.0 mg; age ⬎ 2–3 years, MTX at a dose of 9.0 mg plus cytarabine at a dose of 18.0 mg; age ⬎ 3–9 years, MTX at a dose of 12.0 mg plus cytarabine at a dose of 24.0 mg; and age ⬎ 9 years, MTX at a dose of 12.0 mg plus cytarabine at a dose of 30.0 mg, plus dexamethasone given at a dose of 2.0 mg/m2 [maximum dose, 2.0 mg]) all during the treatment, every 6 weeks. Children were evaluated using a complete blood cell count, hepatic enzymes and serum electrolytes,

1294

CANCER September 15, 2003 / Volume 98 / Number 6

TABLE 2 Therapeutic Schedule in the First Treatment Period for Patients with Class II–V Disease Phase

Drug

Dose

Starting “day”

Frequency

Induction

CDDP Teniposide CDDP (Regimen A) Teniposide (Regimen A) Cyclophosphamide (Regimen B) Vincristine (Regimen B) Doxorubicin (Regimen B) MTX Ara-C Dexamethasone

90 mg/m2/day ⫻ 1 100 mg/m2/day ⫻ 1 90 mg/m2/day ⫻ 1 100 mg/m2/day ⫻ 1 30 mg/kg/day ⫻ 1 0.05 mg/kg/day ⫻ 1 2 mg/kg/day ⫻ 1 According to age

1 3 1 3 1 1 1 1

Every 21 days ⫻ 3 cycles A and B “alternated” Every 21 days

Maintenancea

Intratecal therapyb

Every 6 weeks

CDDP: cisplatin; MTX: methotrexate; Ara-C: cytarabine. a Patients without disease progression or recurrence after 3 cycles received 60 weeks of treatment with alternating Regimens A and B. b Intrathecal therapy was administered only to patient-with Class II-V disease.

TABLE 3 Therapeutic Schedule in the Second Treatment Period for Patients with Class I–V Disease Phase

Drug

Dose

Starting “day”

Frequency

Induction

IFO VP-16 IFO (Regimen A) VP-16 (Regimen A) CDDP (Regimen B) Teniposide (Regimen B) MTX Ara-C Dexamethasone

1.8 g/m2/day ⫻ 5 100 mg/m2/day ⫻ 5 1.8 mg/m2/day ⫻ 5 100 mg/m2/day ⫻ 5 90 mg/m2/day ⫻ 1 100 mg/m2/day ⫻ 1 According to age

1–5 1–5 1–5 1–5 1 3 1

Every 21 days ⫻ 3 cycles A and B “alternated” Every 21 days

Maintenancea

Intratecal “therapy”b

Every 6 weeks

IFO: ifosfamide; VP-16: etoposide; CDDP: cisplatin; MTX: methotrexate; Ara-C: cytarabine. a Patients without disease progression or recurrence after 3 cycles received 34 weeks of treatment with alternating Regimens A and B. b Intrathecal therapy was given only to patients with Class II–V disease.

BUN, and creatinine determinations before each cycle.Visual outcome were not assessed. The radiation therapy guidelines were the same for both treatment periods. Patients with Class I disease did not receive orbital radiation therapy. Patients with Class II and III disease were treated with megavoltage photons from a linear accelerator of 4 megaelectron volts, using 2 angled fields to encompass the orbital cavity and the optical nerve through the chiasm. A median total dose of 46 Gy was delivered in 23 fractions of 2 Gy each. All patients were immobilized with individual masks and in the younger patients (those age ⬍ 2 years), anesthesia was used. Patients with CNS disease received radiation therapy to the whole brain and neuroaxis. The total dose was 36 Gy in 20 fractions. A boost of 10 –15 Gy was added with localized fields to any residual mass. Patients with bone metastasis received approximately 30 – 40 Gy to the lesion. Data regarding age, gender, race, lag time, first sign, staging, laterality, and treatment regimen were collected

using a form designed for this purpose. Lag time was defined as the duration of symptoms before diagnosis.11 In all patients, follow-up assessments included clinical evaluation, CSF examinations, and brain CT imaging. Eight children were lost to follow-up (9.6%). Patients were defined as lost to follow-up after two consecutive medical consultations were missed. The 5-year overall survival was calculated using the Kaplan–Meier method. Statistical methods also included the Student t test, Pearson correlation, and the chi-square test. For all tests, the ␣ error was 5%.

RESULTS There was a slight predominance of males (53%) and the majority of the patients were white (62.7%). The age of the patients ranged from 2–145 months with a mean age of 32.9 months. Patients with unilateral disease presented with a median age of 33.2 months and those with bilateral disease had a median age of 23.7 months. The mean lag time was 10.5 months. Leukocoria was the first sign/symptom in 57 children

Extraocular Retinoblastoma/Gianotti Antoneli et al.

1295

TABLE 4 Patients Distribution According to Gender, Class, and Laterality in the Two Treatment Periodsa Laterality Variable Gender Male Female Class I/III IV/V Mean age at diagnosis (mos) Mean lag time (mos) Treatment period 1 2 a

Unilateral

Bilateral

30 33

14 6

52 11 35.9 9.0

17 3 23.6 6.0

32 31

11 9

P value

Period 1

Period 2

21 22

23 17

36 7 29.7 8 —

33 7 30.8 7 —

0.081

0.429

0.798

0.023 0.743

P value

0.882

0.480 — —

Period 1 included 1987–1991 and Period 2 included 1992–2000.

(68.7%). Table 4 summarizes patients characteristics according to gender, disease staging, and laterality in the two treatment periods. Locally advanced tumors (Classes I–III) were found to be present in 83.1% of the patients (Class I: 4 unilateral tumors and 4 bilateral tumors; Class II: 21 unilateral tumors and 8 bilateral tumors; and Class III: 27 unilateral tumors and 5 bilateral tumors). Seven patients (all with unilateral tumors) were classified as having Class IV disease and seven patients were classified as having Class V disease (four unilateral and three bilateral tumors). Sixtythree patients presented with unilateral tumors (75.9%) and the remaining 20 patients were found to have bilateral disease (24.1%). Bilateral tumors were diagnosed at an earlier age (mean age, 23.6 months) compared with unilateral tumors (mean age, 35.9 months) (P ⫽ 0.023). Bilateral and unilateral tumors did not appear to differ with regard to staging distribution (P ⫽ 0.798) (Table 4). There was a positive correlation between lag time and age for unilateral tumors (correlation coefficient [r] ⫽ 0.35; P ⫽ 0.006), whereas the correlation was negative for bilateral tumors (r ⫽ ⫺0.12; P ⫽ 0.63). The 3-year and 5-year disease-free survival estimates were equivalent for both treatment periods, corresponding to 59.6% and 69.5%, respectively, for the first and second treatment periods (P ⫽ 0.351). The 5-year overall survival rate was 55.1% in the first treatment period and 59.4% in the second treatment period (P ⫽ 0.690) (Fig. 1) In the first treatment period, patients with Class I-III disease presented with a 5-year overall survival rate of 65.3% whereas the survival rate for patients with Class IV-V disease was 0% (P ⫽ 0.003). In the second treatment period, the 5-year overall survival rate for patients with Class I-III disease was reported to have improved (75.5%) whereas pa-

FIGURE 1. Five-year overall survival according the period of treatment (first versus second).

tients with Class IV-V tumors continued to have a poor survival rate (20%) (P ⬍ 0.001) (Figs. 2 and 3). During the first treatment period we treated four patients classified with Class I tumors (microscopic involvement of the scleral emissaries) with a combination of cyclophosphamide plus vincristine. At the time of last follow-up, 1 patient was alive without disease, 1 patients was lost of follow-up 29 months after diagnosis, and 2 patients had died with progressive disease. During the second treatment period, four patients with Class I disease received intensive chemotherapy and were reported to be alive without disease at the time of last follow-up (P ⫽ 0.187). With regard to those patients classified as having Class II disease, there was a total of 29 patients when the 2 treatment periods were compared. In the first

1296

CANCER September 15, 2003 / Volume 98 / Number 6

FIGURE 2. Five-year overall survival for unilateral tumors according the period of treatment (first versus second).

mor in the surgical specimen, whereas in the second treatment period seven patients achieved a complete clinical response, with three patients having no viable tumor in the surgical specimen. Patients with CNS/ CSF disease were not reported as achieving any response in either treatment period. One patient with bone marrow disease achieved a complete disease remission and was alive without evidence of disease at the time of last follow-up. Fourteen patients presented with metastatic disease at the time of diagnosis: 7 patients presented with CNS disease at the time of diagnosis and 7 patients had bone or bone marrow disease at the time of diagnosis. None of the patients with CNS disease at the time of diagnosis (Class IV) were alive at the time of last follow-up (five patients were admitted during the first treatment period and two patients were admitted during the second treatment period). Despite therapy, all these patients died of disease progression, with a median survival time of 13 months. Five patients with metastatic disease (Class V disease) who were treated during the second period had a 5-year overall survival rate of 20% (P ⫽ 0.708). Local control with enucleation was obtained for all patients with Class I and Class II disease. One patient with Class III disease presented with orbital recurrence. Sites of distant recurrence included the CSF plus bone in 1 patient, the CSF in 19 patients, and the CNS in 4 patients. In the first treatment period, 2 patients presented with osteogenic sarcoma (1 in the femur and the other in the tibia, 7 years and 9 years, respectively, after the diagnosis of retinoblastoma) and in the second treatment period, 1 patient presented with a nonlymphocytic leukemia 3 years after the initiation of treatment for retinoblastoma.

FIGURE 3. Five-year overall survival in the second treatment period, accord-

DISCUSSION

ing to tumor class.

Extraocular retinoblastoma remains a challenging disease to treat. The management of extraocular disease is complex, and in many cases, various combinations of treatment may be needed to achieve a satisfactory result. Among the patients in the current study, we observed a discrete predominance of males (53%). In the literature, no gender predilection has been reported to our knowledge.12,13 The age of the patients at the time of presentation ranged from 2–145 months (mean, 32.9 months). Menon et al.14 and Doz et al.7 reported an average age of 30 months and 38 months, respectively, for patients with extraocular disease. The most common sign/symptom was leukokoria, which was observed in 68% of the cases in the current study and also is described by Menon et al. (72%).14 In that series, the authors also reported that the most com-

period, the overall survival was 76.5% (14 of 18 patients were alive without disease at the time of last follow-up) versus 69.3% (8 of 11 patients were alive without disease at the time of last follow-up) in the second period (P ⫽ 0.830). Among 32 patients classified as having Class III disease, 53% (8 of 14 patients) were alive during the first treatment period and 63.6% (12 of 18 patients) were alive during the second treatment period (P ⫽ 0.431) Patients with Class I and II disease were not evaluated for response because they underwent enucleation at the time of diagnosis. Eleven patients with Class III disease underwent an assessment of response after 3 cycles of chemotherapy. In the first treatment period, three patients achieved a complete clinical response with viable tu-

Extraocular Retinoblastoma/Gianotti Antoneli et al.

mon presenting sign was proptosis (83%).14 The rate of proptosis as the most common presenting sign was 75.9% versus 24.1%, respectively, in patients with unilateral and bilateral disease . In the literature,1–9,13,15–24 the reported estimated frequency of bilaterally ranges from 20 –30%. The prognosis of patients with extraocular disease remains relatively poor. The use of chemotherapy (including cyclophosphamide, teniposide, cisplatin, vincristine, and doxorubicin) associated with external radiation therapy has been reported previously.25 Doz et al.7 and Menon et al.14 have used schedules with etoposide to treat patients with extraocular retinoblastoma and their good results have encouraged the use of this drug in combination with other agents in many retinoblastoma protocols. Acquaviva et al.26 used chemotherapy in a series of 51 patients with retinoblastoma. Chemotherapy was given to all patients (including 32 with unilateral tumors and 19 with bilateral tumors). Patients with retinoblastoma extending to the optic nerve head, choroid, and emissaries received cyclophosphamide and vincristine in addition to local radiation therapy. To our knowledge, the literature provides little information regarding treatment for patients with Class I disease. We did not use radiation therapy in these patients and do not recommend it. For patients with Class II (optic nerve involvement) or Class III (orbital involvement) retinoblastoma, the addition of doxorubicin resulted in an overall survival rate of 90.6% in patients with unilateral disease and 84.2% in patients with bilateral disease. The current study results did not demonstrate improvement in the survival of patients with Class II and III disease who received ifosfamide plus etoposide and external beam radiation therapy (at a dose of 45 Gy). These data suggest that the aggressive chemotherapy used in the current study did not improve survival significantly in patients with extraocular retinoblastoma presenting with orbital disease. The surgery that was performed in patients with Class II and III disease was enucleation in the current study, but Doz et al. and Chantada et al. did not use surgery that was more aggressive than exenteration in these patients, even those presenting with orbital disease at the time of diagnosis.7,27 The use of a carboplatin and etoposide regimen could potentially be effective in these cases. In patients with CNS involvement, intrathecal chemotherapy with MTX, cytarabine, and dexamethasone and cranial-spinal radiation therapy was added.26 In the current study results, no patient with CNS disease at the time of diagnosis (Class IV) was alive at the time of last follow-up. All patients died with progressive disease, with a median survival time of 13

1297

months. These data confirm the poor prognosis of those patients with CNS dissemination. The combination of high-dose carboplatin, etoposide, and cyclophosphamide is reportedly effective in patients with high-risk retinoblastoma, especially those with chemosensitive distant metastatic disease, but without involvement of the CNS. The 3-year disease-free survival rate was 67.1% with the CARBOPEC regimen (carboplatin, etoposide, and cyclophosphamide followed by hematopoietic stem cell rescue), which is in agreement with our results.28 The treatment strategy using high-dose chemotherapy with carboplatin and thiotepa with or without etoposide plus autologous stem cell rescue was found to be effective for patients with metastatic retinoblastoma that does not involve the CNS. However, to our knowledge there are too few patients to conclude that the drugs are effective for patients with metastatic retinoblastoma.29 Chantada et al. reported that among 15 patients with orbital or preauricular lymph node disease the 5-year event-free survival was 84%,27 whereas the 5-year overall survival in the current study during the second treatment period for this group of patients was 63.6%. Clinical trials8,9 have documented the efficacy of systemic carboplatin therapy in the management of multiple pediatric malignancies and for these reasons, carboplatin is included in the chemotherapeutic regimen for the majority of patients with pediatric extraocular retinoblastoma . Although the majority of authors15–24,30 conclude that the combination of etoposide plus ifosfamide when added to cyclophosphamide, vincristine, doxorubicin, cisplatin, and teniposide appears to be highly effective in the treatment of extraocular retinoblastoma, we continue to observe a 5-year overall survival rate of 60% using this combination. Patients with CNS disease remain incurable. In the last 10 years, proposals regarding the optimal duration of therapy were changed. The majority of solid pediatric tumors recur during the first year of treatment and the majority of centers treat these patients with 1-year therapy. Bone marrow transplantation is used in patients with metastatic retinoblastoma and appears to be effective, but to our knowledge the number of reported successful cases still is small enough to recommend this procedure as the gold standard of treatment in these patients.

Conclusions Chemotherapy and external beam radiation therapy continue to be a mainstay in the treatment of patients with advanced retinoblastoma. The current study data suggest that despite the use of aggressive chemother-

1298

CANCER September 15, 2003 / Volume 98 / Number 6

apy, including ifosfamide and etoposide, we were unable to improve survival for patients with extraocular retinoblastoma with any statistical significance. Radiation therapy is important to achieve local control in patients with optical nerve disease and orbital involvement. Patients with CNS dissemination or metastatic disease continue to die with progressive disease and remain incurable, despite the aggressive treatment. A multicenter trial should be considered to evaluate the best treatment strategy for these cases. Cyclophosphamide, carboplatin, and etoposide are known to be mutagenic and may be responsible for the development of second tumors, especially in patients with a constitutional abnormality of the retinoblastoma gene. Careful cumulative dosing and follow-up evaluation must be the rule in these patients. Awareness through education and outreach to the community can help in early referral, so that vision can be preserved and survival may be improved.

12. 13. 14.

15.

16.

17. 18.

19.

REFERENCES ´ , Chevis-Barrios P, HurHurwitz RL, Shields CL, Shields JA witz MY, Chintagumpala MM. In: Pizzo PA, Poplack DG, editors. Principles and practice of pediatric oncology, 4th edition. Philadelphia: Lippincott- Raven, 2001:28:828 – 846. 2. Chantada GL, Fandin ˜ o A, Manzitti J, Urrutia L, Schwartzman E. Late diagnosis of retinoblastoma in a developing country. Arch Dis Child. 1999;80:171–174. 3. Erwenne CM, Franco EL. Age and lateness of referral as determinants of extraocular retinoblastoma. Ophthalmic Paediatr Genet. 1989;10:179 –184. 4. Shields CL, Shields AS, Baez K, Cater JR. Optic nerve invasion of retinoblastoma. Metastatic potential and clinical risk factors. Cancer. 1994;73:692– 698. 5. Ellsworth RM. The management of retinoblastoma. Jpn J Ophthalmol. 1978;22:424 – 430. 6. Schouten-van Meeteren AYN, Moll AC, Imhof SM, Veerman AJP. Overview chemotherapy for retinoblastoma: an expanding area of clinical research. Med Pediatr Oncol. 2002; 38:428 – 438. 7. Doz F, Neuenschwander S, Plantaz D, et al. Etoposide and carboplatin in extraocular retinoblastoma: a study by the Societe´ Franc¸aise d’Oncologie Pe´diatrique. J Clin Oncol. 1995;13:902–909. 8. Pratt CB, Crom DB, Howarth C. The use of chemotherapy for extraocular retinoblastoma. Med Pediatr Oncol. 1985;13: 330 –333. 9. Frappaz J, Michon O, Hartmann E, et al. Etoposide and carboplatin in neuroblastoma: a French Society of Pediatric Oncology phase II study. J Clin Oncol. 1992;10:1592–1601. 10. Wolff JA, Boesel C, Ellsworth R, et al. Extraocular retinoblastoma. Children Cancer Study Group- Protocol CCSG 962. New York: Children Cancer Group, 1978. 11. Saha V, Love S, Eden T, Micallef-Eynaud P, MacKin G. 1.

20.

21. 22.

23. 24.

25.

26.

27.

28.

29. 30.

Determinants of symptom interval in chidhood cancer. Arch Dis Child. 1993;68:771–774. Pawlak BR. Retinoblastoma: an epidemiological study (survey and review). J Surg Oncol. 1975;7:45–55. Abramson DH. Retinoblastoma: diagnosis and management. CA Cancer J Clin. 1982:32:130 –140. Menon BS, Reddy SC, Wan Maziah WN, Ham A, Rosline H. Extraocular retinoblastoma. Med Pediatr Oncol. 2000:35:75– 76. Schwartzman E, Chantada G, Fandin ˜ o A, De Da´vila MT, Raslawski E, Mazitti J. Results of a stage-based protocol for the treatment of retinoblastoma. J Clin Oncol. 1996;14:1532– 1536. Goble RR, McKenzie J, Kingston JE, Plowman PN, Hungerford JL. Orbital recurrence of retinoblastoma successfully treated by combined therapy. Br J Ophthalmol. 1990;2:97– 98. Hopping W, Schimitt G, Haver W. The treatment of retinoblastoma. Jpn J Ophthalmol. 1978;22:420 – 423. Pratt CB, Fontanesi J, Chenaille P, et al. Chemotherapy for extraocular retinoblastoma. Pediatr Hematol Oncol. 1994;11: 301–309. Saariam UM, Sariola H, Hovi L. Recurrent disseminated retinoblastoma treated by high-dose chemotherapy, total body irradiation and autologous bone marrow-rescue. Am J Pediatr Hematol Oncol. 1991;13:315–319. ´ , Boesel CP, Dyment PG, et al. Treatment of retinoWolff JA blastoma: a preliminary report. Int Congr Ser. 1981;570:364 – 368. ´ , Pratt CB, Sitarz AL. Chemotherapy of metastatic Wolff JA retinoblastoma. Cancer Chemother Rep. 1962;16:435– 437. Londsale D, Berry DH, Holcomb TM, et al. Chemotherapeutic trials in patients with metastatic retinoblastoma. Cancer Chemother Rep. 1968;52:631– 634. Merriam GR Jr. Retinoblastoma: some considerations. Int J Radiat Oncol Biol Phys. 1978;4:523–524. Doz F, Khelfaoui F, Mosseri V, et al. The role of chemotherapy in orbital involvement of retinoblastoma: the experience at a single institution with 33 patients. Cancer. 1994; 74:722–732. White L. Chemotherapy in retinoblastoma: current status and future directions. Am J Pediatr Hematol Oncol. 1991;13: 189 –201. Acquaviva A, Barberi L, Bernardini C, et al. Medical therapy of retinoblastoma in children. J Neurosurg Sci. 1982;26:49 – 52. Chantada G, Fandin ˜ o A, Casak S, Manzitti J, Raslawski E, Schvartzman E. Treatment of overt extraocular retinoblastoma. Med Pediatr Oncol. 2003;40:158 –161. Namoumi F, Doz F, Tanguy ML, et al. High-dose chemotherapy with carboplatin, etoposide and cyclophosphamide followed by a haematopoietic stem cell rescue in patients with high-risk retinoblastoma: a SFOP and SFGM study. Eur J Cancer. 1997:33(14):2368 –2375. Dunkel IJ, Aledo A, Kernan NA, et al. Successful treatment of metastatic retinoblastoma. Cancer. 2000;89(10):2117–2121. Zelter M, Gonzales G, Schwartz L, et al. Treatment of retinoblastoma: results obtained from a prospective study of 51 patients. Cancer. 1991;68:1685–1690.