Objective: To describe platinum-based chemotherapy combined with local treatment modalities as an alternative to external beam radiotherapy for intraocular retinoblastoma.
Design: Platinum levels were measured by atomic absorption analysis in the tumors of 2 patients with retinoblastoma given carboplatin 5 or 2.5 hours before enucleation. Platinum levels in heated vs nonheated Greene melanoma tumors in rabbits were compared. A retrospective review of 172 affected eyes in 136 consecutive patients treated for retinoblastoma between January 1990 and December 1995 was performed. From 1990 to 1992, all treatable eyes initially received systemic carboplatin, 560 mg/m2, followed by 15 to 30 minutes of continuous diode laser hyperthermia (thermochemotherapy). Since 1992, larger tumors were treated initially with 3 monthly cycles of carboplatin, etoposide, and vincristine sulfate to reduce tumor volume (chemoreduction) followed by sequential aggressive local therapy (SALT) during examinations under anesthesia every 2 to 3 weeks.
Outcome measure: Treatment success was defined as eradication of tumor without enucleation or external beam radiotherapy.
Results: Significant therapeutic platinum levels were measured in the human tumors 2.5 and 5 hours after carboplatin administration. Increasing the temperature by 9 degrees C for 15 minutes doubled platinum levels in the rabbit model. Of the 38 eyes with Reese-Ellsworth group 1 through 5b tumors that were treated primarily with thermochemotherapy, all 24 eyes with group 1 and 2 tumors were treated successfully and two of the 4 eyes with group 3 tumors and all 10 eyes with group 5b tumors were treated unsuccessfully. Chemoreduction plus SALT was the primary treatment in 35 eyes and was successful in all 10 eyes with group 1 through 4 tumors and unsuccessful in all 7 eyes with extensive subretinal seeding and all 18 eyes with group 5b tumors with vitreous seeding. Seventy patients received carboplatin or carboplatin, vincristine, and etoposide, with myelosuppression, occasionally associated with bacteremia, being the main side effect. Transfusions were required in 15% of patients. Radiation retinopathy occurred in all 6 eyes treated with iodine 125 plaques.
Conclusions: Thermochemotherapy is successful primary treatment for Reese-Ellsworth group 1 and 2 retinoblastomas. For larger tumors in the absence of vitreous or extensive subretinal seeding, 3 cycles of chemoreduction followed by SALT eradicates residual viable tumor. Chemoreduction plus SALT was not successful in eyes with diffuse vitreous or extensive subretinal seeding. Prior chemotherapy increases the risk for radiation retinopathy following 125I plaque therapy. External beam radiotherapy can safely be avoided in the primary treatment of Reese-Ellsworth groups 1 through 4 nondispersed retinoblastoma.