Histopathologic examination of the enucleated globe of an untreated 3-month-old transgenic retinoblastoma mouse (control). Note the large retinal tumor.
Histopathologic examination of enucleated globes of transgenic retinoblastoma mice after 6 serial subconjunctival carboplatin injections beginning at age 5 weeks. A, A 50-µg injection. B, A 125-µg injection. C, A 200-µg injection. A moderately sized tumor is present at 50 µg of carboplatin. Note the complete absence of retinal tumor at carboplatin doses of 125 and 200 µg.
Histopathologic examination of enucleated globes of transgenic retinoblastoma mice after 6 serial subconjunctival carboplatin injections beginning at age 10 weeks. A, A 50-µg injection. B, A 125-µg injection. C, A 200-µg injection. Large tumors are present at carboplatin doses of 50 and 125 µg. There is no evidence of tumor at 200 µg of carboplatin.
Histopathologic examination of enucleated globes of transgenic retinoblastoma mice after 12 serial subconjunctival carboplatin injections beginning at age 10 weeks. A, A 50-µg injection. B, A 125-µg injection. C, A 200-µg injection. A moderately sized tumor is present at 50 µg of carboplatin. Note the complete absence of retinal tumor at carboplatin doses of 125 and 200 µg.
Carboplatin dose-response curves demonstrating the proportion of tumor controlled by subconjunctival carboplatin treatment initiated at ages 5 and 10 weeks. Increased tumor burden is associated with decreased tumor control. Tumor control is reestablished in animals with a higher tumor burden when the number of injections is increased from 6 to 12 serial injections.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Hayden BH, Murray TG, Scott IU, et al. Subconjunctival Carboplatin in Retinoblastoma: Impact of Tumor Burden and Dose Schedule. Arch Ophthalmol. 2000;118(11):1549–1554. doi:10.1001/archopht.118.11.1549
To evaluate the impact of tumor burden and chemotherapy dose scheduling on the response to subconjunctival carboplatin treatment in a murine transgenic retinoblastoma model.
Eighty simian virus 40 T antigen–positive mice were treated at age 5 or 10 weeks. Six control animals received placebo treatment. Twenty-four 5-week-old mice received 6 subconjunctival carboplatin injections at doses of 30 to 300 µg delivered at 72-hour intervals. Fifty 10-week-old mice received either 6 or 12 subconjunctival carboplatin injections at doses of 30 to 300 µg delivered at 72-hour intervals. All eyes were obtained at age 16 weeks for histopathologic examination. Eyes were graded as positive if any tumor was present.
All simian virus 40 T antigen–positive control eyes contained large tumor foci throughout the retina. Subconjunctival carboplatin injections controlled tumors in a dose-dependent manner. Tumor control was observed in 50% of treated eyes at 138.3 µg for the 10-week-old 6-injection group, 94.3 µg for the 5-week-old 6-injection group, and 85.9 µg for the 10-week-old 12-injection group.
Increased tumor burden requires an increase in subconjunctival carboplatin dose scheduling to maintain local tumor control in this murine model of retinoblastoma.
This study documents the efficacy of subconjunctival carboplatin in the treatment of an animal model of retinoblastoma. These data establish a framework for further human clinical trials.
RETINOBLASTOMA is the most common primary intraocular malignancy of childhood and the third most common pediatric cancer.1,2 In the past century, earlier diagnosis and treatment of retinoblastoma have improved survival rates in developed countries dramatically.3-5 Historically, treatment of retinoblastoma was limited to enucleation.6,7 More recent treatment options in the management of retinoblastoma (including external beam radiotherapy, episcleral plaque radiotherapy, cryotherapy, laser photoablation, and systemic chemotherapy) offer the potential of sight preservation.8-10 External beam radiotherapy has played a prominent role in the treatment of retinoblastoma, especially in children with a germline Rb mutation predisposing them to bilateral eye involvement.11-13 Although the efficacy of ionizing radiation treatment has been demonstrated, this therapeutic modality is associated with such complications as facial deformities, cataract, and radiation retinopathy and an increased incidence of second tumors.14-20 To avoid these radiation-related complications, systemic chemotherapy has become standard in the management of retinoblastoma.8,21-24
Cyclophosphamide and vincristine sulfate are among the more effective chemotherapeutic agents, but their use is associated with substantial toxic effects, including bone marrow suppression, nephrotoxicity, and myelotoxicity.25,26 Platinum compounds, such as cisplatin and carboplatin, demonstrate relatively low toxicity compared with other chemotherapeutic agents.27,28 Clinical trials29,30 have documented the efficacy of systemic carboplatin therapy in the management of multiple pediatric malignancies (neuroblastoma, Ewing sarcoma, and Wills tumor) and adult neoplasms (ovarian, testicular, and lung malignancies). For these reasons, carboplatin is included in the chemotherapeutic regimen for most patients with pediatric extraocular retinoblastoma.31,32
Although carboplatin has a relatively low toxic effect profile compared with other chemotherapeutic agents, concerns regarding second malignancies and other complications still exist.27,29,33,34 Systemic chemotherapy might increase the risk of second cancers in childhood survivors.16,21 One approach to the concerns related to systemic carboplatin as a treatment strategy is the introduction of the drug directly into periorbital tissues. Focal carboplatin chemotherapy, delivered intravitreally and subconjunctivally, has been demonstrated35-38 to be efficacious in the treatment of a transgenic murine model of retinoblastoma. To our knowledge, the impact of tumor burden at the time of treatment and the impact of carboplatin dose scheduling has not been investigated.
The present study compared the efficacy of subconjunctival carboplatin administered in 6 vs 12 cycles for the treatment of murine transgenic retinoblastoma. The treatment was applied to animals at age 5 or 10 weeks to evaluate the effect of increased tumor burden at the time of treatment onset on therapeutic efficacy.
The study protocol was approved by the University of Miami School of Medicine Animal Care and Use Review Board, Miami, Fla. All experiments in this study were conducted in accordance with the Association for Research in Vision and Ophthalmology guidelines for the use of animals in ophthalmologic and vision research.
Eighty large simian virus 40 T antigen transgene–bearing mice were treated at age 5 or 10 weeks. The transgenic mouse model used in this study has been characterized previously.39-43 Briefly, a highly expressed murine oncogenic transgene drives bilateral retinal tumor development by using a retinal-specific promoter sequence to direct the expression of simian virus 40 T antigen, resulting in ocular tumor growth. Transgenic animals were identified through polymerase chain reaction analysis of tail DNA. Samples positive and negative for the transgene were detected by visualizing ethidium bromide–stained agarose gels. Transgenic animals develop bilateral, heritable retinoblastoma that resembles human retinoblastoma. Pathological evidence of tumor is noted by age 4 weeks, small intraocular tumor (corresponding to Reese-Ellsworth group 1) is present at 5 weeks, moderate to large intraocular tumor (corresponding to Reese-Ellsworth group 3-4) is noted at 10 weeks, and tumor fills the globe by 16 weeks. Tumors in this animal model are typically small at 5 weeks, appearing with intraretinal involvement only (occupying <1% of retinal area and correspondingly less ocular volume), and at 10 weeks the tumor is moderate in size (occupying approximately 20%-25% of the retinal area and 10%-25% of the ocular volume).
Twenty-four 5-week-old mice received subconjunctival carboplatin (Paraplatin; Bristol-Myers Squibb, Princeton, NJ) injections in the right eye at doses of 30, 50, 62.5, 85, 125, 150, 200, or 300 µg given twice per week for a total of 6 injections. Fifty 10-week-old mice received 6 (25 mice) or 12 (25 mice) serial subconjunctival carboplatin injections in the right eye at doses of 30, 50, 62.5, 85, 125, 200, or 300 µg administered twice per week for a total of 6 injections. Injections were delivered with a 33-gauge needle inserted into the nasal and superior subconjunctival spaces. A microvolume delivery pump was used to ensure accurate and reproducible delivery of a 25-µL volume. The left eyes of these animals served as untreated controls. Six transgenic litter-matched animals received 25-µL subconjunctival injections of balanced salt solution to provide a positive placebo control. Following each injection, all of the animals underwent serial ophthalmologic examination.
At age 16 weeks, all animals were killed with an overdose of ketamine hydrochloride and xylazine hydrochloride. Both eyes were enucleated and immediately immersion fixed in 10% formalin. The eyes were sectioned serially and stained with hematoxylin-eosin. Light microscopic examination was performed on all histopathologic sections in a masked fashion. Eyes were graded positive for tumor development if any histopathologic evidence of tumor was present. Eyes were also evaluated for evidence of corneal, lenticular, retinal, or scleral toxic effects.
Outcomes were analyzed and logit statistical modeling was used to calculate dose-response curves. Doses at which 50% of the eyes had complete tumor control (TCD50) were calculated using the probit model, and logistic regression analysis was used to compare the 3 dose-response curves generated. Carboplatin dose, dose schedule, and tumor burden were analyzed as independent variables.
Histopathologic examination at age 16 weeks revealed that all untreated left eyes (n = 74) and placebo-treated eyes (n = 6) exhibited multiple large intraocular tumors (Figure 1). None of the mice treated with 30 µg of carboplatin exhibited tumor control. All 6 mice treated with 300 µg of carboplatin exhibited complete tumor control; 4 eyes at this dose exhibited mild corneal toxic effects and neovascularization. No histopathologic evidence of ocular toxic effects was noted for any of the other treated eyes.
Increased tumor burden at the time of subconjunctival carboplatin injection was significantly associated with decreased tumor control. Reestablishment of tumor control was accomplished by doubling the treatment duration of subconjunctival carboplatin injection from 6 to 12 total injections (Figure 2, Figure 3, and Figure 4). The TCD50 was 138.3 µg for 10-week-old mice receiving 6 injections, 94.3 µg for 5-week-old mice receiving 6 injections, and 85.9 µg for 10-week-old mice receiving 12 injections.
Probit regression analysis was used for estimating the relation between dose and tumor control, and logistic regression analysis demonstrated that the dose-response curves were significantly different among the 3 treatment groups (P = .03) (Figure 5).
Although effective in controlling intraocular retinoblastoma, external beam radiotherapy has been shown to be associated with such complications as radiation retinopathy, optic neuropathy, and midfacial hypoplasia and an increased incidence of secondary malignancies.1 In an effort to avoid radiation-associated morbidity, systemic chemotherapy, which had previously been reserved for patients with orbital extension or metastatic retinoblastoma,24,44 has been used with increasing frequency in the treatment of intraocular retinoblastoma.
Clinical studies45,46 have demonstrated that intravenous carboplatin is an effective treatment option in children with retinoblastoma. There is much interest in carboplatin as a primary agent in the treatment of retinoblastoma because it is relatively nontoxic compared with other chemotherapeutic agents available. The proposed mechanism of carboplatin action is that it covalently binds the diamino platinum radical to DNA, which inhibits DNA replication.27
Despite the potential advantages of systemic chemotherapy, including reduced tumor size, enhanced efficacy of other treatment modalities, and the possible prevention of micrometastasis, serious complications may occur.14,15,27 Systemic chemotherapy has been associated with myelosuppression, nephrotoxicity, ototoxicity, and sepsis and may increase the risk of secondary malignancies later in life.24 Acute nonlymphatic leukemia has been reported after treatment of retinoblastoma with combined external beam radiotherapy and systemic chemotherapy.28 Subconjunctival delivery of carboplatin chemotherapy directly into the periorbital tissues may be associated with increased drug delivery and decreased toxic effects.
Systemic administration of chemotherapy as a single-modality treatment strategy is not currently used in patients with retinoblastoma. Intravenous administration of carboplatin is associated with relatively low penetrance into the vitreous. However, a recent investigation33 of intraocular carboplatin concentrations after peribulbar administration in non–tumor-bearing primates has shown dramatically higher concentrations of carboplatin in the vitreous and aqueous humor than in control animals receiving intravenous administration of the drug. The latter data, in combination with data from the present study, suggest that subconjunctival delivery of carboplatin might be effective in the control of intraocular retinoblastoma, with the optimal chemotherapeutic dose schedule dependent on the tumor burden at the time of treatment onset.
Previous studies35-38 have demonstrated the efficacy of carboplatin administered intravitreally and subconjunctivally in the treatment of murine retinoblastoma. Direct injection into the vitreous cavity, however, creates the possibility of tumor dissemination from the injection tract and, therefore, may increase the risk of extraocular metastasis. For this reason, use of subconjunctival injections alone is preferable.
Results of clinical studies47,48 indicate that systemic chemotherapy may be most valuable as a treatment strategy for retinoblastoma when it is combined with local treatment strategies to enhance efficacy. Systemic chemoreduction has been shown to decrease the size of intraocular retinoblastoma, allowing for more conservative, globe-preserving treatment options.25,32,47 Furthermore, increased cycle number has resulted in enhanced chemoreduction of moderately sized tumors when systemic chemotherapy is coupled with adjuvant local therapy.47 Systemic carboplatin administered with cyclosporine after a session of cryotherapy and focal ablation has been shown to be highly effective in treating retinoblastoma in children.49,50 In addition, results of laboratory experiments51 indicate that subconjunctival delivery of carboplatin enhances the efficacy of other treatment modalities in the treatment of murine retinoblastoma. The most effective and least toxic combination of modalities for the treatment of human pediatric retinoblastoma may be subconjunctivally administered carboplatin in addition to other local treatments such as cryotherapy and laser hyperthermia photoablation.
One limitation of the present study is that a direct extrapolation of the murine model of retinoblastoma to human retinoblastoma is not possible. Intraocular volumes and vascular supply differ between the 2 species, which complicates data interpretation. A recent phase I/II clinical study52 of subconjunctival carboplatin administration for intraocular retinoblastoma has documented significant efficacy in control of vitreous seeding and primary retinal tumor. Tumor control of pediatric retinoblastoma using subconjunctival carboplatin may prove to be effective and safe when administered at optimum carboplatin concentrations and dose schedules.
Experimental data from the present study suggest that clinically relevant tumor burdens respond to serial focal carboplatin therapy and that increased tumor burden at the time of treatment will require an increase in dose scheduling to maintain local tumor control. These data may be important in determining the optimal cycle number for the management of human intraocular retinoblastoma with local or systemic chemotherapy alone or in combination with other local treatment modalities.
Accepted for publication April 15, 2000.
This study was supported by Fight for Sight and Research to Prevent Blindness Inc, New York, NY; the American Cancer Society Florida Division, Tampa; and the Knights Templar Eye Foundation, Chicago, Ill. Carboplatin was generously donated by Bristol-Myers Squibb, Princeton, NJ.
Reprints: Timothy G. Murray, MD, Bascom Palmer Eye Institute, PO Box 016880, Miami, FL 33101 (e-mail: firstname.lastname@example.org).