Abstract
Purpose::
To characterize and compare a xenograft model of vitreous seeding of retinoblastoma in rodents and rabbits. To compare and contrast tumor growth and implantation after utilizing different routes of tumor cell delivery.
Methods::
White New Zealand rabbits were immunosupressed with Cyclosporin A at doses of 10 - 15 mg/Kg starting at day 1. At day 3, animals received an injection of retinoblastoma cells (1 to 1.5 x 106 cells) distributed as follows: (n = 4) intravitreal injection of CHLA 203; (n = 4) suprachoroidal injection of Y79; and (n = 4) suprachoroidal injection of CHLA 203. To develop the vitreous seeding model of retinoblastoma in rodents CHLA 203 cells were injected into the vitreous cavity of 16 nude rats. The animals were examined every 7-10 days for a period of 60 days, and euthanasia was carried out at different time points. Clinical characterization was based on tumor implantation rate, progression and clinical-histopathologic correlation. Histology was carried out in enucleated eyes.
Results::
In rabbits, the intravitreal tumor seeds were evident as soon as two weeks after injection in all injected animals. Tumors developed in the vitreous cavity as multiple seeds progressing as a confluent lesion towards the optic nerve and posterior retina. After suprachoroidal injection, tumor growth was slow and the first sign of tumor implantation, in one animal, occurred at day 22 as an exudative retinal detachment. In nude rats, approximately 79% of the animals developed tumor implantation. Among them, 10 presented vitreous seeding as the initial manifestation of the disease. Of the previous 10, 60% of them exhibited seeding up to 20 days after injection. Indirect ophthalmoscopy and fundus photography allowed diagnosis and follow-up of the disease. These findings correlated with the histologic diagnosis in human eyes with vitreous disease.
Conclusions::
The animal model of retinoblastoma presenting as vitreous seeding, was demonstrated to be viable both in rodents and rabbits. The use of rabbits to host the vitreous seeding model will facilitate testing of new therapies for retinoblastoma, particularly for treatments based on local drug delivery. We plan further studies with a panel of newly developed human retinoblastoma cell lines to evaluate variability in treatment response.
Keywords: retinoblastoma • tumors • oncology