- Email: [email protected]
Multimodal Therapy for Stage III Retinoblastoma (International Retinoblastoma Staging System)
Multimodal Therapy for Stage III Retinoblastoma (International Retinoblastoma Staging System) A Prospective Comparative Study Bhavna Chawla, MS,1,2 Fahmi Hasan, PhD,1,2 Rachna Seth, DNB,3 Sushmita Pathy, MD,4 Rajesh Pattebahadur, MD,1,2 Sanjay Sharma, MD,5 Ashish Upadhyaya, MA,6 Rajvardhan Azad, MD1,2 Purpose: To compare the efficacy of 2 chemotherapeutic drug combinations as part of multimodal therapy for orbital retinoblastoma. Design: Prospective, comparative, study. Participants: Patients with stage III retinoblastoma (International Retinoblastoma Staging System). Methods: Demographic and clinical features were recorded at presentation. Treatment consisted of a multimodal protocol with neoadjuvant chemotherapy, enucleation, orbital external-beam radiotherapy, and adjuvant chemotherapy. For chemotherapy, patients were randomized into 2 groups: group A patients were treated with vincristine, etoposide, and carboplatin (VEC) and group B patients were treated with carboplatin and etoposide, alternating with cyclophosphamide, idarubicin, and vincristine. Treatment outcomes and adverse effects were recorded. Efficacy parameters were compared between the groups. Main Outcome Measures: Survival probability, cause of death, and chemotherapy-related toxicity. Results: A total of 54 children were recruited (27 in each group). The mean SD follow-up was 21.311.34 months. The overall Kaplan-Meier survival probability was 80% (95% confidence interval [CI], 0.67e0.89) and 42% (95% CI, 0.24e0.59) at 1 year and 4 years, respectively. There were 9 deaths in group A and 15 deaths in group B. The Kaplan-Meier survival probability at 1 year was similar between the groups: 81% (95% CI, 0.60e0.91) and 79% (95% CI, 0.58e0.9) for groups A and B, respectively. At 4 years, the survival probability for group A was higher (63% [95% CI, 0.41e0.79] vs. 25% [95% CI, 0.08e0.46] for groups A and B, respectively), with a strong trend of better survival in group A over time (P ¼ 0.05). The major cause of death was central nervous system relapse (8 patients in group A and 7 patients in group B). Two patients in group B died of sepsis after febrile neutropenia. Grade 3 and grade 4 hematologic toxicities were more common in group B, with a significant difference in grade 4 neutropenia (P ¼ 0.002). Conclusions: This study compared the outcomes of VEC chemotherapy with a 5-drug combination of vincristine and carboplatin, alternating with cyclophosphamide, idarubicin, and vincristine, for stage III retinoblastoma. The VEC combination was found to be more effective and may be recommended as neoadjuvant and adjuvant chemotherapy. Ophthalmology 2016;-:1e7 ª 2016 by the American Academy of Ophthalmology.
Retinoblastoma is a curable cancer if detected in the early stages.1 However in untreated patients, extraocular spread can occur, either through the optic nerve or through the sclera.2 Although extraocular disease is rare in developed countries, it is not an unusual feature in the developing world.1e3 In low-income countries of Asia and Africa, extraocular disease constitutes 20% to 50% of all retinoblastoma cases.4,5 The management of orbital retinoblastoma remains a challenge: orbital involvement is associated with a 10- to 27-times higher risk of metastasis when compared with patients without orbital extension.6,7 Although there is no definitive treatment for orbital retinoblastoma, the preferred management involves a multimodal approach that includes neoadjuvant chemotherapy (NAC), enucleation surgery, external-beam 2016 by the American Academy of Ophthalmology Published by Elsevier Inc.
radiotherapy (EBRT), and adjuvant chemotherapy.8e10 Initially, a combination of high-dose chemotherapeutic agents is used for inducing tumor regression. After enucleation, orbital EBRT and adjuvant chemotherapy are given to eradicate microscopic residual disease and to prevent metastasis.3,9,11 The literature on orbital retinoblastoma is sparse and treatment strategies are still evolving. Most published studies are limited by inclusion of heterogeneous populations in various stages of orbital disease, a small number of cases, and a retrospective design. Furthermore, different staging systems (Children’s Co-operative Group staging system, Grabowski-Abramson staging, and the International Retinoblastoma Staging System [IRSS]) have been used, http://dx.doi.org/10.1016/j.ophtha.2016.05.034 ISSN 0161-6420/16
1
Ophthalmology Volume -, Number -, Month 2016 Table 1. Chemotherapeutic Regimen for Groups A and B Group A
B
Drug Vincristine Etoposide Carboplatin Vincristine Etoposide Carboplatin Cyclophosphamide Mesna Idarubicin
Dose (per Day)
Cycles
0.025 mg/kg 12 mg/kg 28 mg/kg 1.5 mg/m2 0.05 mg/kg for children <3 yrs 100 mg/m2 3.3 mg/kg for children <3 yrs 560 mg/m2 18.6 mg/kg for children <3 yrs 65 mg/kg 60mg/kg 10 mg/m2
3 weekly, 12 cycles 3 weekly, 12 cycles 3 weekly, 12 cycles weeks 0, 6, 12, and 18
1 1 and 2 1 1
Day(s)
weeks 3, 9, 15, and 21
1, 2, and 3
weeks 3, 9, 15, and 21
1 and 2
weeks 0, 6, 12, and 18
1
weeks 0, 6, 12, and 18
1
Mesna ¼ 2-mercaptoethane sulfonate Na.
making comparison and interpretation difficult.8e12 There is also a lack of consensus regarding the choice of chemotherapeutic drug combinations to be used as NAC and adjuvant chemotherapy. Chantada et al9,11 introduced the combination of carboplatin plus etoposide in their chemotherapeutic protocol for orbital retinoblastoma and used a 5-drug combination of cyclophosphamide, vincristine, etoposide, idarubicin, and carboplatin. Although the results were encouraging, they agreed that their patient population was too small to make a definitive recommendation.9 In 2005, Honavar and Singh3 reported their experience with high-dose triple-drug chemotherapy followed by appropriate surgery, orbital radiotherapy, and adjuvant chemotherapy. The chemotherapy combination in their series consisted of 12 cycles of vincristine, etoposide, and carboplatin (VEC).3 They found encouraging results, but agreed that their protocol needs validation and further studies.3 The current management of orbital retinoblastoma involves a lengthy and intensive treatment protocol, and it is desirable to treat these children with a standard chemotherapy combination that is most effective and least toxic. Therefore, a pilot study was undertaken to investigate whether there was any difference in the efficacy of the VEC combination and the 5-drug chemotherapy combination when used as a part of multimodal treatment for IRSS stage III retinoblastoma.
Methods The study was a prospective, randomized, comparative study. Prior approval was obtained from the ethics committee of our institution. Patients who sought treatment at the retinoblastoma clinic between 2011 and 2013 were recruited after obtaining informed consent from the guardians. Children with stage IIIa retinoblastoma (regional extension with overt orbital disease) according to the IRSS classification were enrolled. The criteria for exclusion were patients with metastatic disease at initial diagnosis (IRSS stage IV), history of prior treatment, or parents who were not willing to participate in the study. In addition, patients with stage IIIb disease (regional lymph node extension) also were excluded to maintain a homogeneous population for comparison between the groups. Patients were considered to have metastatic disease if they had positive bone marrow aspiration or biopsy results, positive
2
cerebrospinal fluid results, or evidence of central nervous system disease on magnetic resonance imaging (MRI) scans. A baseline history and demographic details such as the child’s age, gender, and laterality were noted. Thorough ophthalmic, general physical, and systemic examinations were performed. Diagnostic investigations included B-scan ultrasonography, contrast-enhanced MRI of the orbits and the brain, and incision biopsy of the orbital mass, when required. The extent of orbital soft tissue invasion by the tumor was assessed on MRI scans. Systemic investigations consisted of hemogram, liver and kidney function tests, ultrasonography of the abdomen, chest x-ray, echocardiography, cerebrospinal fluid examination, and bone marrow aspiration and biopsy. Children who fulfilled the inclusion criteria were randomized into 2 groups according to computer-generated random numbers. Group A children were treated with triple-drug chemotherapy consisting of VEC. Etoposide and carboplatin were administered in higher doses (12 mg/kg daily for 2 days and 28 mg/kg, respectively) as compared with the standard dose (5 mg/kg daily for 2 days and 18.6 mg/kg, respectively) used for patients of group E intraocular retinoblastoma with high-risk histopathologic features after enucleation. The dose of vincristine was lower (0.025 mg/kg) compared with the standard dose of 0.05 mg/kg. Group B children were treated with a 5-drug combination of carboplatin and etoposide, alternating with cyclophosphamide, idarubicin, and vincristine. The dosage and schedule of chemotherapy in both groups is shown in Table 1. The response to treatment was assessed on clinical examination as well as on MRI of the brain and orbits that was performed after 3 cycles of chemotherapy (Fig 1). The response was graded on MRI scans as follows: complete response, phthisical globe and no residual optic nerve disease; partial response, partial shrinkage of tumor mass or residual optic nerve disease present; no response, no change in tumor mass or extent of optic nerve invasion; and progressive disease, increase in tumor mass or extent of optic nerve invasion. If the eye was not considered amenable to enucleation after 3 cycles, another 3 cycles of chemotherapy were administered and MRI examination was repeated. Enucleation was carried out under general anaesthesia and a primary polymethyl methacrylate implant was placed. A detailed histopathologic examination was carried out on the enucleated eyeballs for the presence of viable tumor cells and for any evidence of scleral or extrascleral invasion, optic nerve invasion, choroidal invasion, and iris or ciliary body involvement. Adjuvant EBRT was administered to the affected orbit within 4 to 6 weeks of enucleation. Radiotherapy was administered in doses
Chawla et al
Multimodal Therapy for Orbital Retinoblastoma
Figure 1. Orbital retinoblastoma in the left eye of a 3-year-old girl. Clinical photographs at (A) presentation and (B) after 3 cycles of neoadjuvant chemotherapy showing a marked reduction in proptosis after chemotherapy. Axial magnetic resonance imaging scans obtained (C) at presentation to show a large orbital mass with thickening of the retrobulbar portion of the optic nerve and (D) after 3 cycles of chemotherapy. Note the phthisical globe and a reduction in thickening and enhancement of the optic nerve.
of 40 Gy in 20 fractions over 4 weeks as a 2-Gy fraction daily for 5 days per week. The technique used was computerized tomographybased, three-dimensional, conformal radiotherapy. Adjuvant chemotherapy was continued after surgery according to the treatment protocol (Table 1). During treatment, patients were monitored closely for adverse effects of chemotherapy, if any. Granulocyte colony-stimulating factor was administered in specific instances, either as primary prophylaxis in patients who were at high risk of febrile neutropenia or as secondary prophylaxis in patients who experienced a neutrophilic complication from a prior cycle of chemotherapy. The National Cancer Institute Common Terminology Criteria for Adverse Events Scale, version 4.03, was used to grade toxicities.13 The number of episodes of hematologic toxicities (neutropenia, thrombocytopenia, anaemia) were recorded for each child. Children were followed up at regular intervals and investigations were performed, as appropriate. Disease progression was assessed on MRI scans and by bone marrow or cerebrospinal fluid analysis for the presence of tumor cells. On completion of treatment, MRI scans of the orbits and brain were repeated to rule out any orbital recurrence or intracranial spread. The main outcome measure of our study was comparison of efficacy of therapy between the groups. For efficacy, the parameters were defined as (1) effectiveness of NAC in inducing local tumor control as assessed on MRI scans, (2) histopathologic evidence of viable tumor cells in the enucleated eye and invasion of ocular tissues by tumor cells, and (3) survival probability of patients. Adverse effects of treatment, such as deaths resulting from chemotherapy-related toxicity, if any, and the frequency of hematologic toxicities, also were studied. Data were analyzed by Stata software version 11.1 (StataCorp., College Station, TX). Data were represented as mean standard deviation (SD) or median (range; minimumemaximum) and frequency (percent). Categorical variables were analyzed by the Fisher exact test. Continuous variables following a normal distribution were compared by the independent t test and
Wilcoxon rank-sum test (skewed data). Kaplan-Meier survival analysis using the log-rank test was used to study the survival probability. A P value of 0.05 or less was considered statistically significant.
Results A total of 71 patients with orbital retinoblastoma sought treatment at our center during the study period. Of these, 17 patients were excluded because they had IRSS stage IV or IIIb disease or a history of prior therapy, or they declined to participate in the study. Therefore, 54 patients who fulfilled the inclusion criteria (n ¼ 27 in each group) were recruited. The median age at presentation was 34.5 months (range, 4e132 months). Thirty-two patients were males (59%) and 22 patients were females (41%). The disease was unilateral in 34 patients (63%) and bilateral in 20 patients (37%). None of the patients had bilateral extraocular disease. In group A, the right eye showed orbital retinoblastoma in 18 patients (66.7%) and the left eye showed orbital retinoblastoma in 9 patients (33.3%). In group B, the right eye showed orbital retinoblastoma in 17 patients (63%) and the left eye showed orbital retinoblastoma in 10 patients (37%). Demographic features at presentation were comparable between the groups. No significant difference was noted between groups A and B with regard to age at presentation (P ¼ 0.24), gender (P ¼ 0.21), or laterality (P ¼ 0.22). The varying grades of tumor response on MRI after NAC treatment are depicted in Figure 2. On comparative evaluation, a complete response was achieved in more patients in group A as compared with group B (P ¼ 0.008). In addition, more patients in group B showed evidence of progressive disease as compared with group A (P ¼ 0.04), suggesting better local tumor control by chemotherapy in group A. During the course of NAC, 15 children died (7 in group A and 8 in group B) and another 4 children (2 in each group) abandoned therapy. The remaining 35 patients underwent enucleation surgery
3
Ophthalmology Volume -, Number -, Month 2016
Figure 2. The grades of tumor response on magnetic resonance imaging (MRI) after neoadjuvant chemotherapy.
(18 and 17 patients in groups A and B, respectively). Histopathologic examination showed that 17 of the 35 eyes (49%) that were enucleated had evidence of viable tumor cells. There was presence of tumor infiltration into the optic nerve head in 34.2% of eyes, into the retrolaminar area in 25.7% of eyes, in the resected margin of the optic nerve in 11.4% of eyes, in the iris and ciliary body in 25.7% of eyes, in the choroid in 28.6% of eyes, and in the sclera in 17.1% of eyes. Other microscopic features included the presence of necrotic tissue, foamy macrophages, calcification, and lipofuscin pigments. A comparison of histopathologic findings between the groups showed that more patients in group B had tumor infiltration into the optic nerve head (16.7% in group A vs. 53% in group B; P ¼ 0.03) and the retrolaminar area (11.1% in group A vs. 41.2% in group B; P ¼ 0.04). However, the presence of viable tumor cells was comparable between the groups (44.4% vs. 53%; P ¼ 0.61), as were other features, such as the presence of scleral invasion (11.1% vs. 23.5%; P ¼ 0.40), choroidal invasion (22.2% vs. 35.3%; P ¼ 0.47), and iris or ciliary body infiltration (27.8% vs. 23.5%; P ¼ 0.77). None of the patients in our study were treated with orbital exenteration. Of the 35 children who underwent enucleation surgery, adjuvant EBRT was given to 31 children (18 patients in group A and 13 patients in group B). The remaining 4 patients were in group B and died of progressive disease before receiving radiotherapy.
The mean SD follow-up was 21.311.34 months, with a range of 5 to 52 months for group A and 4 to 45 months for group B. Of 54 patients, 24 children died (44%). The mean SD interval between diagnosis and death was 13.36.88 months. The overall Kaplan-Meier survival probability was 80% at 1 year (95% confidence interval [CI], 0.67e0.89), 49% at 2 years (95% CI, 0.33e0.63), and 42% at 4 years (95% CI, 0.24e0.59), as shown in Figure 3A. Group-wise analysis showed that there were 9 deaths in group A and 15 deaths in group B. The Kaplan-Meier survival probability was similar in the groups at 1 year: 81% (95% CI, 0.60e0.91) and 79% (95% CI, 0.58e0.91) for groups A and B, respectively. However, the survival probability at 4 years was higher in group A as compared with group B: 63% (95% CI, 0.41e0.79) versus 25% (95% CI, 0.08e0.46), respectively. As is evident from Figure 3B, a strong trend toward better survival with time was observed in group A as compared with group B (P ¼ 0.05). The mean SD interval from diagnosis to death was 11.85.39 months for group A and 14.17.7 months for group B, with no significant difference between the 2 groups (P ¼ 0.51). The major cause of death in both groups was central nervous system relapse (15/24 [62.5%]; 8 children in group A and 7 children in group B). In group B, 4 children died of hematogenous metastasis (16.6%). In 3 children (1 in group A and 2 in group B), the exact cause of death could not be ascertained because it occurred in a remote area. Chemotherapy-related toxicity was more significant among group B children, and 2 children died of sepsis after drug toxicity. The percentage of patients who had 1 or more episodes of grade 3 or 4 chemotoxicity (all hematologic toxicities) in each group was determined. Figure 4 depicts the occurrence of neutropenia, thrombocytopenia, and anemia in the groups. As is evident, grade 4 neutropenia was more significant in group B as compared with group A (55% vs. 12%; P ¼ 0.002). Grade 4 thrombocytopenia and anemia also were more frequent in group B, although the difference was not statistically significant. Similarly, grade 3 hematologic toxicities were more common in group B compared with group A. No case of secondary malignancy was noted in our study.
Discussion The management of orbital retinoblastoma is challenging.3 Although survival rates reported from developed nations range from 88% to 93%,14e16 mortality rates in developing countries have been as high as 50% to 90%.5,17,18 In the past, orbital retinoblastoma was treated with orbital exenteration, which is a mutilating and disfiguring surgery.19 Several years
Figure 3. Kaplan-Meier survival probability curve showing (A) overall survival and (B) group-wise survival.
4
Chawla et al
Multimodal Therapy for Orbital Retinoblastoma
Figure 4. Grade 3 and 4 hematologic toxicity in the study groups. Grade 3 neutropenia was an absolute neutrophil count (ANC) between 500/mm3 and 1000/mm3, grade 4 neutropenia was an ANC of less than 500/mm3, grade 3 thrombocytopenia was a platelet count between 25 000/mm3 and 50 000/mm3, grade 4 thrombocytopenia was a platelet count of less than 25 000/mm3, grade 3 anemia was a decrease in hemoglobin to 6 to 8 g/dl, and grade 4 anemia was a hemoglobin of less than 6 g/dl.
ago, White20 used a combination of cyclophosphamide, teniposide, cisplatin, vincristine, and doxorubicin in conjunction with EBRT. Later, encouraging results were shown with the addition of etoposide to other agents.21 Chantada et al formulated management protocols in which patients were treated initially with NAC to achieve substantial tumor reduction, making limited surgery possible, which was followed by orbital radiotherapy.9 In the present era, exenteration no longer is used as a first-line treatment, and there is increasing evidence that a multimodal approach with NAC, enucleation, orbital radiotherapy, and adjuvant chemotherapy improves patient survival.3,9e11 Although there is universal agreement on the use of multimodal therapy, confusion persists about the chemotherapy protocols to be used, and different investigators have used various combinations of drugs and doses, making therapeutic recommendations difficult. Our study showed that NAC was effective in inducing tumor regression, thereby obviating the need for exenteration surgery. This finding is consistent with previous studies that have reported effective local tumor control with NAC.9,10 Despite the marked response on clinical examination and MRI, 49% of enucleated eyes in our study had viable tumor cells after NAC. Histopathologic evidence of viable tumor cells after NAC also has been reported by previous investigators and supports the use of a multimodal approach.3,8,10 In our patients, the cut end of the optic nerve showed positive results in 11.4% of patients, which is comparable with a previous study that reported a positive transection end in 13.6% of patients.10 On comparative evaluation, the VEC regimen was found to be more effective as NAC in achieving local tumor control as
assessed on MRI scans than the 5-drug regimen. An interesting observation was that the number of eyes showing tumor infiltration into the optic nerve head and the retrolaminar area on histopathologic examination was significantly lower in the VEC group. Previous studies have reported a 3- to 5-year survival ranging from 40% to 84% for orbital retinoblastoma.9,10 In our study, the overall survival probability was 80% and 42% at 1 year and 4 years, respectively. Although the survival probability at 1 year was similar in the 2 groups, there was a trend toward better survival over time in the VEC-treated patients as compared with those treated with the 5-drug combination. In a previous study, Chantada et al11 reported a survival of 70% in patients that were treated with a similar 5-drug combination. The better outcome noted in their study could be explained by the fact that most of their patients had stage II disease (n ¼ 21), and very few patients with stage III disease were included (n ¼ 5). In contrast, our study comprised only patients with stage III disease.11 In our VEC-treated patients, the survival probability was 81% and 63% at 1 and 4 years, respectively, which is higher than the previously reported survival of 40.4% in another study of patients with stage III disease treated with the VEC combination.10 A possible explanation for this could be the inclusion of more advanced cases of orbital retinoblastoma with regional lymph node extension (stage IIIb) in the previous study that were excluded in the present study.10 Central nervous system metastasis was the most common cause of relapse and death in our study. Hematogenous metastasis also resulted in 4 deaths in group B. Progressive disease has been reported to be the cause of death in previous
5
Ophthalmology Volume -, Number -, Month 2016 studies also, with the central nervous system being the most common site for relapse.8e11 The mean interval between diagnosis and death was 13.3 months, highlighting that most deaths are the result of progressive disease and occur early because of the aggressive nature of the disease. Deaths resulting from secondary malignancies such as osteogenic sarcomas and leukemia have been reported by other investigators.8,9,11 We did not find any case of secondary malignancy, possibly because of a relatively shorter follow-up. Toxicity is a major concern in these patients because the chemotherapy course is intensive and prolonged. Sepsis resulting from febrile neutropenia resulted in 2 deaths in group B. Chantada et al9,11 also reported deaths resulting from febrile neutropeniad1 patient in each of 2 different studiesdusing the same protocol as was used for group B patients. However, no death resulting from chemotherapy toxicity was observed in group A. To our knowledge, there are no previous reports of death resulting from chemotherapy toxicity with the VEC protocol. Hematologic toxicity was evaluated in detail in our study by examining the occurrence of neutropenia, thrombocytopenia, and anemia. The occurrence of grade 4 neutropenia was significantly lower in the VEC-treated group (55% vs. 12%; P ¼ 0.002). Also, the frequencies of grade 4 thrombocytopenia and anemia and of grade 3 hematologic toxicities were lower in the VEC-treated group, reflecting the more toxic nature of group B drugs. A literature search revealed only 1 study to have evaluated the frequency of grade 3 or 4 hematologic toxicity using VEC chemotherapy for stage III retinoblastoma.10 In another study where a combination of vincristine, etoposide, carboplatin, and cyclophosphamide was used, neutropenic fever was seen in 41.3% of the children.22 However, the frequency of thrombocytopenia and anemia was not mentioned in that study.22 To summarize, our study found triple-drug chemotherapy with the VEC combination to be more effective compared with etoposide and carboplatin, alternating with cyclophosphamide, idarubicin, and vincristine. Hence, VEC may be recommended as NAC and adjuvant chemotherapy for orbital retinoblastoma. Further research to determine whether a shorter course of VEC can be as effective as the current protocol of 12 cycles and to investigate the longterm effects of toxicity is recommended.
References 1. Broaddus E, Topham A, Singh AD. Survival with retinoblastoma in the USA: 1975e2004. Br J Ophthalmol 2009;93:24–7. 2. Ali MJ, Honavar SG, Reddy VAP. Orbital retinoblastoma: present status and future challenges. A review. Saudi J Ophthalmol 2011;25:159–67. 3. Honavar SG, Singh AD. Management of advanced retinoblastoma. Ophthalmol Clin N Am 2005;18:65–73.
6
4. Chawla B, Hasan F, Azad R, et al. Clinical presentation and survival of retinoblastoma in Indian children. Br J Ophthalmol 2016;100:172–8. 5. Badhu B, Sah SP, Thakur SK, et al. Clinical presentation of retinoblastoma in Eastern Nepal. Clin Experiment Ophthalmol 2005;33:386–9. 6. Gündüz K, Müftüoglu O, Günalp I, et al. Metastatic retinoblastoma clinical features, treatment, and prognosis. Ophthalmology 2006;113:1558–66. 7. Messmer EP, Heinrich T, Hopping W, et al. Risk factors for metastasis in patients with retinoblastoma. Ophthalmology 1991;98:136–41. 8. Antoneli CB, Steinhorst F, de Cassia Braga RK, et al. Extraocular retinoblastoma: a 13-year experience. Cancer 2003;98: 1292–8. 9. Chantada GL, Fandino AC, Casak S, et al. Treatment of overt extraocular retinoblastoma. Med Pediatr Oncol 2003;40: 158–61. 10. Radhakrishnan V, Kashyap S, Pushker N, et al. Outcome, pathologic findings, and compliance in orbital retinoblastoma (International Retinoblastoma Staging System stage III) treated with neo-adjuvant chemotherapy: a prospective study. Ophthalmology 2012;119:1470–7. 11. Chantada GL, Guitter MR, Fandino AC, et al. Treatment results in patients with retinoblastoma and invasion to the cut end of the optic nerve. Pediatr Blood Cancer 2009;52:218–22. 12. Chantada GL, Doz F, Antoneli CBG, et al. A proposal for an international retinoblastoma staging system. Pediatr Blood Cancer 2006;47:801–5. 13. National Cancer Institute, Program CTE. Common terminology criteria for adverse events, version 4.03. Available at: http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE4.032010-06-14Quick Reference8.5x11.pdf. Accessed December 22, 2010. 14. Sanders BM, Draper GJ, Kingston JE. Retinoblastoma in Great Britain 1969-1980: incidence, treatment and survival. Br J Ophthalmol 1988;72:576–83. 15. 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: 121–31. 16. Abramson DH, Niksarli K, Ellsworth RM, et al. Changing trends in the management of retinoblastoma 1951e1965 versus 1966e1980. Strabismus 1994;31:32–7. 17. Chantada GL, Fadino AC, Manzitti J, et al. Late diagnosis of retinoblastoma in developing countries. Arch Dis Child 1999;80:171–4. 18. Schvartzman E, Chantada GL, Fandino AC, et al. Results of a stage-based approach for the treatment of retinoblastoma. J Clin Oncol 1996;14:1532–6. 19. Reese AB. Tumors of the eye. New York: Harper and Row; 1963:155–6. 20. White L. Chemotherapy in retinoblastoma: current status and future direction. Am J Pediatr Hematol Oncol 1991;13: 189–201. 21. Menon BS, Reddy SC, Wan Maziah WN, et al. Extraocular retinoblastoma. Med Pediatr Oncol 2000;35:75–6. 22. Künkele A, Wilm J, Holdt M, et al. Neoadjuvant/adjuvant treatment of high-risk retinoblastoma: a report from the German Retinoblastoma Referral Centre. Br J Ophthalmol 2015;99:949–53.
Chawla et al
Multimodal Therapy for Orbital Retinoblastoma
Footnotes and Financial Disclosures Originally received: April 3, 2016. Final revision: May 21, 2016. Accepted: May 23, 2016. Available online: ---.
Supported by the Department of Science and Technology, Science and Engineering Research Board, Government of India, New Delhi, India. Author Contributions: Manuscript no. 2016-670.
1
Ocular Oncology Service, All India Institute of Medical Sciences, New Delhi, India.
2
Rajendra Prasad Centre for Ophthalmic Sciences, All India Institute of Medical Sciences, New Delhi, India.
3
Pediatric Oncology Division, Department of Pediatrics, All India Institute of Medical Sciences, New Delhi, India.
4
Department of Radiotherapy, All India Institute of Medical Sciences, New Delhi, India.
5
Department of Radiodiagnosis, All India Institute of Medical Sciences, New Delhi, India.
6
Department of Biostatistics, All India Institute of Medical Sciences, New Delhi, India. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article.
Conception and design: Chawla, Seth Analysis and interpretation: Chawla, Hasan, Seth, Sharma, Upadhyaya Data collection: Chawla, Hasan, Pathy, Pattebahadur, Azad Obtained funding: Chawla Overall responsibility: Chawla, Hasan, Seth, Pathy, Pattebahadur, Sharma, Upadhyaya, Azad Abbreviations and Acronyms: CI ¼ confidence interval; EBRT ¼ external-beam radiotherapy; IRSS ¼ International Retinoblastoma Staging System; Mesna ¼ 2-mercaptoethane sulfonate Na; MRI ¼ magnetic resonance imaging; NAC ¼ neoadjuvant chemotherapy; SD ¼ standard deviation; VEC ¼ vincristine, etoposide, and carboplatin. Correspondence: Bhavna Chawla, MS, Rajendra Prasad Centre for Ophthalmic Sciences, All India Institute of Medical Sciences, New Delhi, India. E-mail: [email protected].
7
Retinoblastoma is a curable cancer if detected in the early stages.1 However in untreated patients, extraocular spread can occur, either through the optic nerve or through the sclera.2 Although extraocular disease is rare in developed countries, it is not an unusual feature in the developing world.1e3 In low-income countries of Asia and Africa, extraocular disease constitutes 20% to 50% of all retinoblastoma cases.4,5 The management of orbital retinoblastoma remains a challenge: orbital involvement is associated with a 10- to 27-times higher risk of metastasis when compared with patients without orbital extension.6,7 Although there is no definitive treatment for orbital retinoblastoma, the preferred management involves a multimodal approach that includes neoadjuvant chemotherapy (NAC), enucleation surgery, external-beam 2016 by the American Academy of Ophthalmology Published by Elsevier Inc.
radiotherapy (EBRT), and adjuvant chemotherapy.8e10 Initially, a combination of high-dose chemotherapeutic agents is used for inducing tumor regression. After enucleation, orbital EBRT and adjuvant chemotherapy are given to eradicate microscopic residual disease and to prevent metastasis.3,9,11 The literature on orbital retinoblastoma is sparse and treatment strategies are still evolving. Most published studies are limited by inclusion of heterogeneous populations in various stages of orbital disease, a small number of cases, and a retrospective design. Furthermore, different staging systems (Children’s Co-operative Group staging system, Grabowski-Abramson staging, and the International Retinoblastoma Staging System [IRSS]) have been used, http://dx.doi.org/10.1016/j.ophtha.2016.05.034 ISSN 0161-6420/16
1
Ophthalmology Volume -, Number -, Month 2016 Table 1. Chemotherapeutic Regimen for Groups A and B Group A
B
Drug Vincristine Etoposide Carboplatin Vincristine Etoposide Carboplatin Cyclophosphamide Mesna Idarubicin
Dose (per Day)
Cycles
0.025 mg/kg 12 mg/kg 28 mg/kg 1.5 mg/m2 0.05 mg/kg for children <3 yrs 100 mg/m2 3.3 mg/kg for children <3 yrs 560 mg/m2 18.6 mg/kg for children <3 yrs 65 mg/kg 60mg/kg 10 mg/m2
3 weekly, 12 cycles 3 weekly, 12 cycles 3 weekly, 12 cycles weeks 0, 6, 12, and 18
1 1 and 2 1 1
Day(s)
weeks 3, 9, 15, and 21
1, 2, and 3
weeks 3, 9, 15, and 21
1 and 2
weeks 0, 6, 12, and 18
1
weeks 0, 6, 12, and 18
1
Mesna ¼ 2-mercaptoethane sulfonate Na.
making comparison and interpretation difficult.8e12 There is also a lack of consensus regarding the choice of chemotherapeutic drug combinations to be used as NAC and adjuvant chemotherapy. Chantada et al9,11 introduced the combination of carboplatin plus etoposide in their chemotherapeutic protocol for orbital retinoblastoma and used a 5-drug combination of cyclophosphamide, vincristine, etoposide, idarubicin, and carboplatin. Although the results were encouraging, they agreed that their patient population was too small to make a definitive recommendation.9 In 2005, Honavar and Singh3 reported their experience with high-dose triple-drug chemotherapy followed by appropriate surgery, orbital radiotherapy, and adjuvant chemotherapy. The chemotherapy combination in their series consisted of 12 cycles of vincristine, etoposide, and carboplatin (VEC).3 They found encouraging results, but agreed that their protocol needs validation and further studies.3 The current management of orbital retinoblastoma involves a lengthy and intensive treatment protocol, and it is desirable to treat these children with a standard chemotherapy combination that is most effective and least toxic. Therefore, a pilot study was undertaken to investigate whether there was any difference in the efficacy of the VEC combination and the 5-drug chemotherapy combination when used as a part of multimodal treatment for IRSS stage III retinoblastoma.
Methods The study was a prospective, randomized, comparative study. Prior approval was obtained from the ethics committee of our institution. Patients who sought treatment at the retinoblastoma clinic between 2011 and 2013 were recruited after obtaining informed consent from the guardians. Children with stage IIIa retinoblastoma (regional extension with overt orbital disease) according to the IRSS classification were enrolled. The criteria for exclusion were patients with metastatic disease at initial diagnosis (IRSS stage IV), history of prior treatment, or parents who were not willing to participate in the study. In addition, patients with stage IIIb disease (regional lymph node extension) also were excluded to maintain a homogeneous population for comparison between the groups. Patients were considered to have metastatic disease if they had positive bone marrow aspiration or biopsy results, positive
2
cerebrospinal fluid results, or evidence of central nervous system disease on magnetic resonance imaging (MRI) scans. A baseline history and demographic details such as the child’s age, gender, and laterality were noted. Thorough ophthalmic, general physical, and systemic examinations were performed. Diagnostic investigations included B-scan ultrasonography, contrast-enhanced MRI of the orbits and the brain, and incision biopsy of the orbital mass, when required. The extent of orbital soft tissue invasion by the tumor was assessed on MRI scans. Systemic investigations consisted of hemogram, liver and kidney function tests, ultrasonography of the abdomen, chest x-ray, echocardiography, cerebrospinal fluid examination, and bone marrow aspiration and biopsy. Children who fulfilled the inclusion criteria were randomized into 2 groups according to computer-generated random numbers. Group A children were treated with triple-drug chemotherapy consisting of VEC. Etoposide and carboplatin were administered in higher doses (12 mg/kg daily for 2 days and 28 mg/kg, respectively) as compared with the standard dose (5 mg/kg daily for 2 days and 18.6 mg/kg, respectively) used for patients of group E intraocular retinoblastoma with high-risk histopathologic features after enucleation. The dose of vincristine was lower (0.025 mg/kg) compared with the standard dose of 0.05 mg/kg. Group B children were treated with a 5-drug combination of carboplatin and etoposide, alternating with cyclophosphamide, idarubicin, and vincristine. The dosage and schedule of chemotherapy in both groups is shown in Table 1. The response to treatment was assessed on clinical examination as well as on MRI of the brain and orbits that was performed after 3 cycles of chemotherapy (Fig 1). The response was graded on MRI scans as follows: complete response, phthisical globe and no residual optic nerve disease; partial response, partial shrinkage of tumor mass or residual optic nerve disease present; no response, no change in tumor mass or extent of optic nerve invasion; and progressive disease, increase in tumor mass or extent of optic nerve invasion. If the eye was not considered amenable to enucleation after 3 cycles, another 3 cycles of chemotherapy were administered and MRI examination was repeated. Enucleation was carried out under general anaesthesia and a primary polymethyl methacrylate implant was placed. A detailed histopathologic examination was carried out on the enucleated eyeballs for the presence of viable tumor cells and for any evidence of scleral or extrascleral invasion, optic nerve invasion, choroidal invasion, and iris or ciliary body involvement. Adjuvant EBRT was administered to the affected orbit within 4 to 6 weeks of enucleation. Radiotherapy was administered in doses
Chawla et al
Multimodal Therapy for Orbital Retinoblastoma
Figure 1. Orbital retinoblastoma in the left eye of a 3-year-old girl. Clinical photographs at (A) presentation and (B) after 3 cycles of neoadjuvant chemotherapy showing a marked reduction in proptosis after chemotherapy. Axial magnetic resonance imaging scans obtained (C) at presentation to show a large orbital mass with thickening of the retrobulbar portion of the optic nerve and (D) after 3 cycles of chemotherapy. Note the phthisical globe and a reduction in thickening and enhancement of the optic nerve.
of 40 Gy in 20 fractions over 4 weeks as a 2-Gy fraction daily for 5 days per week. The technique used was computerized tomographybased, three-dimensional, conformal radiotherapy. Adjuvant chemotherapy was continued after surgery according to the treatment protocol (Table 1). During treatment, patients were monitored closely for adverse effects of chemotherapy, if any. Granulocyte colony-stimulating factor was administered in specific instances, either as primary prophylaxis in patients who were at high risk of febrile neutropenia or as secondary prophylaxis in patients who experienced a neutrophilic complication from a prior cycle of chemotherapy. The National Cancer Institute Common Terminology Criteria for Adverse Events Scale, version 4.03, was used to grade toxicities.13 The number of episodes of hematologic toxicities (neutropenia, thrombocytopenia, anaemia) were recorded for each child. Children were followed up at regular intervals and investigations were performed, as appropriate. Disease progression was assessed on MRI scans and by bone marrow or cerebrospinal fluid analysis for the presence of tumor cells. On completion of treatment, MRI scans of the orbits and brain were repeated to rule out any orbital recurrence or intracranial spread. The main outcome measure of our study was comparison of efficacy of therapy between the groups. For efficacy, the parameters were defined as (1) effectiveness of NAC in inducing local tumor control as assessed on MRI scans, (2) histopathologic evidence of viable tumor cells in the enucleated eye and invasion of ocular tissues by tumor cells, and (3) survival probability of patients. Adverse effects of treatment, such as deaths resulting from chemotherapy-related toxicity, if any, and the frequency of hematologic toxicities, also were studied. Data were analyzed by Stata software version 11.1 (StataCorp., College Station, TX). Data were represented as mean standard deviation (SD) or median (range; minimumemaximum) and frequency (percent). Categorical variables were analyzed by the Fisher exact test. Continuous variables following a normal distribution were compared by the independent t test and
Wilcoxon rank-sum test (skewed data). Kaplan-Meier survival analysis using the log-rank test was used to study the survival probability. A P value of 0.05 or less was considered statistically significant.
Results A total of 71 patients with orbital retinoblastoma sought treatment at our center during the study period. Of these, 17 patients were excluded because they had IRSS stage IV or IIIb disease or a history of prior therapy, or they declined to participate in the study. Therefore, 54 patients who fulfilled the inclusion criteria (n ¼ 27 in each group) were recruited. The median age at presentation was 34.5 months (range, 4e132 months). Thirty-two patients were males (59%) and 22 patients were females (41%). The disease was unilateral in 34 patients (63%) and bilateral in 20 patients (37%). None of the patients had bilateral extraocular disease. In group A, the right eye showed orbital retinoblastoma in 18 patients (66.7%) and the left eye showed orbital retinoblastoma in 9 patients (33.3%). In group B, the right eye showed orbital retinoblastoma in 17 patients (63%) and the left eye showed orbital retinoblastoma in 10 patients (37%). Demographic features at presentation were comparable between the groups. No significant difference was noted between groups A and B with regard to age at presentation (P ¼ 0.24), gender (P ¼ 0.21), or laterality (P ¼ 0.22). The varying grades of tumor response on MRI after NAC treatment are depicted in Figure 2. On comparative evaluation, a complete response was achieved in more patients in group A as compared with group B (P ¼ 0.008). In addition, more patients in group B showed evidence of progressive disease as compared with group A (P ¼ 0.04), suggesting better local tumor control by chemotherapy in group A. During the course of NAC, 15 children died (7 in group A and 8 in group B) and another 4 children (2 in each group) abandoned therapy. The remaining 35 patients underwent enucleation surgery
3
Ophthalmology Volume -, Number -, Month 2016
Figure 2. The grades of tumor response on magnetic resonance imaging (MRI) after neoadjuvant chemotherapy.
(18 and 17 patients in groups A and B, respectively). Histopathologic examination showed that 17 of the 35 eyes (49%) that were enucleated had evidence of viable tumor cells. There was presence of tumor infiltration into the optic nerve head in 34.2% of eyes, into the retrolaminar area in 25.7% of eyes, in the resected margin of the optic nerve in 11.4% of eyes, in the iris and ciliary body in 25.7% of eyes, in the choroid in 28.6% of eyes, and in the sclera in 17.1% of eyes. Other microscopic features included the presence of necrotic tissue, foamy macrophages, calcification, and lipofuscin pigments. A comparison of histopathologic findings between the groups showed that more patients in group B had tumor infiltration into the optic nerve head (16.7% in group A vs. 53% in group B; P ¼ 0.03) and the retrolaminar area (11.1% in group A vs. 41.2% in group B; P ¼ 0.04). However, the presence of viable tumor cells was comparable between the groups (44.4% vs. 53%; P ¼ 0.61), as were other features, such as the presence of scleral invasion (11.1% vs. 23.5%; P ¼ 0.40), choroidal invasion (22.2% vs. 35.3%; P ¼ 0.47), and iris or ciliary body infiltration (27.8% vs. 23.5%; P ¼ 0.77). None of the patients in our study were treated with orbital exenteration. Of the 35 children who underwent enucleation surgery, adjuvant EBRT was given to 31 children (18 patients in group A and 13 patients in group B). The remaining 4 patients were in group B and died of progressive disease before receiving radiotherapy.
The mean SD follow-up was 21.311.34 months, with a range of 5 to 52 months for group A and 4 to 45 months for group B. Of 54 patients, 24 children died (44%). The mean SD interval between diagnosis and death was 13.36.88 months. The overall Kaplan-Meier survival probability was 80% at 1 year (95% confidence interval [CI], 0.67e0.89), 49% at 2 years (95% CI, 0.33e0.63), and 42% at 4 years (95% CI, 0.24e0.59), as shown in Figure 3A. Group-wise analysis showed that there were 9 deaths in group A and 15 deaths in group B. The Kaplan-Meier survival probability was similar in the groups at 1 year: 81% (95% CI, 0.60e0.91) and 79% (95% CI, 0.58e0.91) for groups A and B, respectively. However, the survival probability at 4 years was higher in group A as compared with group B: 63% (95% CI, 0.41e0.79) versus 25% (95% CI, 0.08e0.46), respectively. As is evident from Figure 3B, a strong trend toward better survival with time was observed in group A as compared with group B (P ¼ 0.05). The mean SD interval from diagnosis to death was 11.85.39 months for group A and 14.17.7 months for group B, with no significant difference between the 2 groups (P ¼ 0.51). The major cause of death in both groups was central nervous system relapse (15/24 [62.5%]; 8 children in group A and 7 children in group B). In group B, 4 children died of hematogenous metastasis (16.6%). In 3 children (1 in group A and 2 in group B), the exact cause of death could not be ascertained because it occurred in a remote area. Chemotherapy-related toxicity was more significant among group B children, and 2 children died of sepsis after drug toxicity. The percentage of patients who had 1 or more episodes of grade 3 or 4 chemotoxicity (all hematologic toxicities) in each group was determined. Figure 4 depicts the occurrence of neutropenia, thrombocytopenia, and anemia in the groups. As is evident, grade 4 neutropenia was more significant in group B as compared with group A (55% vs. 12%; P ¼ 0.002). Grade 4 thrombocytopenia and anemia also were more frequent in group B, although the difference was not statistically significant. Similarly, grade 3 hematologic toxicities were more common in group B compared with group A. No case of secondary malignancy was noted in our study.
Discussion The management of orbital retinoblastoma is challenging.3 Although survival rates reported from developed nations range from 88% to 93%,14e16 mortality rates in developing countries have been as high as 50% to 90%.5,17,18 In the past, orbital retinoblastoma was treated with orbital exenteration, which is a mutilating and disfiguring surgery.19 Several years
Figure 3. Kaplan-Meier survival probability curve showing (A) overall survival and (B) group-wise survival.
4
Chawla et al
Multimodal Therapy for Orbital Retinoblastoma
Figure 4. Grade 3 and 4 hematologic toxicity in the study groups. Grade 3 neutropenia was an absolute neutrophil count (ANC) between 500/mm3 and 1000/mm3, grade 4 neutropenia was an ANC of less than 500/mm3, grade 3 thrombocytopenia was a platelet count between 25 000/mm3 and 50 000/mm3, grade 4 thrombocytopenia was a platelet count of less than 25 000/mm3, grade 3 anemia was a decrease in hemoglobin to 6 to 8 g/dl, and grade 4 anemia was a hemoglobin of less than 6 g/dl.
ago, White20 used a combination of cyclophosphamide, teniposide, cisplatin, vincristine, and doxorubicin in conjunction with EBRT. Later, encouraging results were shown with the addition of etoposide to other agents.21 Chantada et al formulated management protocols in which patients were treated initially with NAC to achieve substantial tumor reduction, making limited surgery possible, which was followed by orbital radiotherapy.9 In the present era, exenteration no longer is used as a first-line treatment, and there is increasing evidence that a multimodal approach with NAC, enucleation, orbital radiotherapy, and adjuvant chemotherapy improves patient survival.3,9e11 Although there is universal agreement on the use of multimodal therapy, confusion persists about the chemotherapy protocols to be used, and different investigators have used various combinations of drugs and doses, making therapeutic recommendations difficult. Our study showed that NAC was effective in inducing tumor regression, thereby obviating the need for exenteration surgery. This finding is consistent with previous studies that have reported effective local tumor control with NAC.9,10 Despite the marked response on clinical examination and MRI, 49% of enucleated eyes in our study had viable tumor cells after NAC. Histopathologic evidence of viable tumor cells after NAC also has been reported by previous investigators and supports the use of a multimodal approach.3,8,10 In our patients, the cut end of the optic nerve showed positive results in 11.4% of patients, which is comparable with a previous study that reported a positive transection end in 13.6% of patients.10 On comparative evaluation, the VEC regimen was found to be more effective as NAC in achieving local tumor control as
assessed on MRI scans than the 5-drug regimen. An interesting observation was that the number of eyes showing tumor infiltration into the optic nerve head and the retrolaminar area on histopathologic examination was significantly lower in the VEC group. Previous studies have reported a 3- to 5-year survival ranging from 40% to 84% for orbital retinoblastoma.9,10 In our study, the overall survival probability was 80% and 42% at 1 year and 4 years, respectively. Although the survival probability at 1 year was similar in the 2 groups, there was a trend toward better survival over time in the VEC-treated patients as compared with those treated with the 5-drug combination. In a previous study, Chantada et al11 reported a survival of 70% in patients that were treated with a similar 5-drug combination. The better outcome noted in their study could be explained by the fact that most of their patients had stage II disease (n ¼ 21), and very few patients with stage III disease were included (n ¼ 5). In contrast, our study comprised only patients with stage III disease.11 In our VEC-treated patients, the survival probability was 81% and 63% at 1 and 4 years, respectively, which is higher than the previously reported survival of 40.4% in another study of patients with stage III disease treated with the VEC combination.10 A possible explanation for this could be the inclusion of more advanced cases of orbital retinoblastoma with regional lymph node extension (stage IIIb) in the previous study that were excluded in the present study.10 Central nervous system metastasis was the most common cause of relapse and death in our study. Hematogenous metastasis also resulted in 4 deaths in group B. Progressive disease has been reported to be the cause of death in previous
5
Ophthalmology Volume -, Number -, Month 2016 studies also, with the central nervous system being the most common site for relapse.8e11 The mean interval between diagnosis and death was 13.3 months, highlighting that most deaths are the result of progressive disease and occur early because of the aggressive nature of the disease. Deaths resulting from secondary malignancies such as osteogenic sarcomas and leukemia have been reported by other investigators.8,9,11 We did not find any case of secondary malignancy, possibly because of a relatively shorter follow-up. Toxicity is a major concern in these patients because the chemotherapy course is intensive and prolonged. Sepsis resulting from febrile neutropenia resulted in 2 deaths in group B. Chantada et al9,11 also reported deaths resulting from febrile neutropeniad1 patient in each of 2 different studiesdusing the same protocol as was used for group B patients. However, no death resulting from chemotherapy toxicity was observed in group A. To our knowledge, there are no previous reports of death resulting from chemotherapy toxicity with the VEC protocol. Hematologic toxicity was evaluated in detail in our study by examining the occurrence of neutropenia, thrombocytopenia, and anemia. The occurrence of grade 4 neutropenia was significantly lower in the VEC-treated group (55% vs. 12%; P ¼ 0.002). Also, the frequencies of grade 4 thrombocytopenia and anemia and of grade 3 hematologic toxicities were lower in the VEC-treated group, reflecting the more toxic nature of group B drugs. A literature search revealed only 1 study to have evaluated the frequency of grade 3 or 4 hematologic toxicity using VEC chemotherapy for stage III retinoblastoma.10 In another study where a combination of vincristine, etoposide, carboplatin, and cyclophosphamide was used, neutropenic fever was seen in 41.3% of the children.22 However, the frequency of thrombocytopenia and anemia was not mentioned in that study.22 To summarize, our study found triple-drug chemotherapy with the VEC combination to be more effective compared with etoposide and carboplatin, alternating with cyclophosphamide, idarubicin, and vincristine. Hence, VEC may be recommended as NAC and adjuvant chemotherapy for orbital retinoblastoma. Further research to determine whether a shorter course of VEC can be as effective as the current protocol of 12 cycles and to investigate the longterm effects of toxicity is recommended.
References 1. Broaddus E, Topham A, Singh AD. Survival with retinoblastoma in the USA: 1975e2004. Br J Ophthalmol 2009;93:24–7. 2. Ali MJ, Honavar SG, Reddy VAP. Orbital retinoblastoma: present status and future challenges. A review. Saudi J Ophthalmol 2011;25:159–67. 3. Honavar SG, Singh AD. Management of advanced retinoblastoma. Ophthalmol Clin N Am 2005;18:65–73.
6
4. Chawla B, Hasan F, Azad R, et al. Clinical presentation and survival of retinoblastoma in Indian children. Br J Ophthalmol 2016;100:172–8. 5. Badhu B, Sah SP, Thakur SK, et al. Clinical presentation of retinoblastoma in Eastern Nepal. Clin Experiment Ophthalmol 2005;33:386–9. 6. Gündüz K, Müftüoglu O, Günalp I, et al. Metastatic retinoblastoma clinical features, treatment, and prognosis. Ophthalmology 2006;113:1558–66. 7. Messmer EP, Heinrich T, Hopping W, et al. Risk factors for metastasis in patients with retinoblastoma. Ophthalmology 1991;98:136–41. 8. Antoneli CB, Steinhorst F, de Cassia Braga RK, et al. Extraocular retinoblastoma: a 13-year experience. Cancer 2003;98: 1292–8. 9. Chantada GL, Fandino AC, Casak S, et al. Treatment of overt extraocular retinoblastoma. Med Pediatr Oncol 2003;40: 158–61. 10. Radhakrishnan V, Kashyap S, Pushker N, et al. Outcome, pathologic findings, and compliance in orbital retinoblastoma (International Retinoblastoma Staging System stage III) treated with neo-adjuvant chemotherapy: a prospective study. Ophthalmology 2012;119:1470–7. 11. Chantada GL, Guitter MR, Fandino AC, et al. Treatment results in patients with retinoblastoma and invasion to the cut end of the optic nerve. Pediatr Blood Cancer 2009;52:218–22. 12. Chantada GL, Doz F, Antoneli CBG, et al. A proposal for an international retinoblastoma staging system. Pediatr Blood Cancer 2006;47:801–5. 13. National Cancer Institute, Program CTE. Common terminology criteria for adverse events, version 4.03. Available at: http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE4.032010-06-14Quick Reference8.5x11.pdf. Accessed December 22, 2010. 14. Sanders BM, Draper GJ, Kingston JE. Retinoblastoma in Great Britain 1969-1980: incidence, treatment and survival. Br J Ophthalmol 1988;72:576–83. 15. 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: 121–31. 16. Abramson DH, Niksarli K, Ellsworth RM, et al. Changing trends in the management of retinoblastoma 1951e1965 versus 1966e1980. Strabismus 1994;31:32–7. 17. Chantada GL, Fadino AC, Manzitti J, et al. Late diagnosis of retinoblastoma in developing countries. Arch Dis Child 1999;80:171–4. 18. Schvartzman E, Chantada GL, Fandino AC, et al. Results of a stage-based approach for the treatment of retinoblastoma. J Clin Oncol 1996;14:1532–6. 19. Reese AB. Tumors of the eye. New York: Harper and Row; 1963:155–6. 20. White L. Chemotherapy in retinoblastoma: current status and future direction. Am J Pediatr Hematol Oncol 1991;13: 189–201. 21. Menon BS, Reddy SC, Wan Maziah WN, et al. Extraocular retinoblastoma. Med Pediatr Oncol 2000;35:75–6. 22. Künkele A, Wilm J, Holdt M, et al. Neoadjuvant/adjuvant treatment of high-risk retinoblastoma: a report from the German Retinoblastoma Referral Centre. Br J Ophthalmol 2015;99:949–53.
Chawla et al
Multimodal Therapy for Orbital Retinoblastoma
Footnotes and Financial Disclosures Originally received: April 3, 2016. Final revision: May 21, 2016. Accepted: May 23, 2016. Available online: ---.
Supported by the Department of Science and Technology, Science and Engineering Research Board, Government of India, New Delhi, India. Author Contributions: Manuscript no. 2016-670.
1
Ocular Oncology Service, All India Institute of Medical Sciences, New Delhi, India.
2
Rajendra Prasad Centre for Ophthalmic Sciences, All India Institute of Medical Sciences, New Delhi, India.
3
Pediatric Oncology Division, Department of Pediatrics, All India Institute of Medical Sciences, New Delhi, India.
4
Department of Radiotherapy, All India Institute of Medical Sciences, New Delhi, India.
5
Department of Radiodiagnosis, All India Institute of Medical Sciences, New Delhi, India.
6
Department of Biostatistics, All India Institute of Medical Sciences, New Delhi, India. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article.
Conception and design: Chawla, Seth Analysis and interpretation: Chawla, Hasan, Seth, Sharma, Upadhyaya Data collection: Chawla, Hasan, Pathy, Pattebahadur, Azad Obtained funding: Chawla Overall responsibility: Chawla, Hasan, Seth, Pathy, Pattebahadur, Sharma, Upadhyaya, Azad Abbreviations and Acronyms: CI ¼ confidence interval; EBRT ¼ external-beam radiotherapy; IRSS ¼ International Retinoblastoma Staging System; Mesna ¼ 2-mercaptoethane sulfonate Na; MRI ¼ magnetic resonance imaging; NAC ¼ neoadjuvant chemotherapy; SD ¼ standard deviation; VEC ¼ vincristine, etoposide, and carboplatin. Correspondence: Bhavna Chawla, MS, Rajendra Prasad Centre for Ophthalmic Sciences, All India Institute of Medical Sciences, New Delhi, India. E-mail: [email protected].
7