OPTIC NERVE INFILTRATION BY RETINOBLASTOMA: Predictive Clinical Features and Outcome.

PURPOSE: To identify the clinical features predictive of any optic nerve infiltration and postlaminar optic nerve infiltration by retinoblastoma on histopathology and to report the outcome (metastasis and death) in these patients. METHODS: Retrospective study. RESULTS: Of the 403 patients who underwent primary enucleation for retinoblastoma, 196 patients had optic nerve tumor infiltration (Group 1) and 207 patients had no evidence of optic nerve tumor infiltration (Group 2). Group 1 included patients with prelaminar (n = 47; 24%), laminar (n = 74; 38%), and postlaminar tumor infiltration with or without involving optic nerve transection (n = 74; 38%). Comparing Group 1 and Group 2, the patients in Group 1 had prolonged duration of symptoms (>6 months) (16% vs. 8%; P = 0.02) and were associated with no vision at presentation (23% vs. 10%; P = 0.01), higher rates of secondary glaucoma (42% vs. 12%; P < 0.0001), iris neovascularization (39% vs. 23%; P < 0.001), and larger tumors (mean tumor thickness, 12.8 mm vs. 12 mm; P = 0.0001). There was a higher prevalence of metastasis in Group 1 than in Group 2 (4% vs. 0%; P = 0.006). On multivariate analysis, clinical features predictive of any optic nerve tumor infiltration secondary glaucoma (hazard ratio = 5.38; P < 0.001) and those predictive of postlaminar optic nerve tumor infiltration included iris neovascularization (hazard ratio = 2.66; P = 0.001) and secondary glaucoma (hazard ratio = 3.13; P < 0.001). CONCLUSION: In this study, clinical features predictive of any optic nerve tumor infiltration included secondary glaucoma and those predictive of postlaminar optic nerve tumor infiltration included iris neovascularization and secondary glaucoma. Despite adjuvant treatment in those with postlaminar optic nerve tumor infiltration, metastasis occurred in 8% of patients.

Clinical features predictive of high-risk retinoblastoma in 403 Asian Indian patients: a case-control study.

PURPOSE: To identify the clinical features predictive of high-risk retinoblastoma on histopathology. DESIGN: Case-control study. PARTICIPANTS: A total of 145 cases with histopathologic high-risk features of retinoblastoma and 258 controls without high-risk features. METHODS: Enucleation and adjuvant chemotherapy. MAIN OUTCOME MEASURES: High-risk features on histopathology were defined as the presence of anterior chamber seeds, iris infiltration, ciliary body infiltration, massive (≥ 3 mm) choroidal invasion, postlaminar optic nerve invasion, invasion of optic nerve transection, combined nonmassive choroidal and prelaminar/laminar optic nerve invasion, or scleral/extrascleral infiltration. RESULTS: Of 403 patients who underwent primary enucleation for the treatment of retinoblastoma, 145 (36%) had high-risk features on histopathology (cases) and 258 (64%) had no high-risk features (controls). High-risk retinoblastoma occurred in 16% of (8/50) group D eyes and 39% of (137/353) group E eyes. The histopathologic high-risk features in these 145 patients included anterior chamber seeds (n = 25, 17%), iris infiltration (n = 12, 8%), ciliary body infiltration (n = 17, 12%), massive (≥3 mm) choroidal invasion (n = 69, 48%), postlaminar optic nerve invasion (n = 71, 49%), invasion of optic nerve transection (n = 3, 2%), combined choroidal and optic nerve invasion (n = 17, 12%), scleral infiltration (n = 20, 14%), and extrascleral involvement (n = 8, 6%). The mean number of high-risk features was 2 (median, 2; range, 1-7). The significant clinical features in cases versus controls included prolonged duration of symptoms of >6 months (21% vs. 7%; P < 0.001), poor visual acuity at presentation (74% vs. 64%; P = 0.05), buphthalmos (16% vs. 7%; P = 0.005), secondary glaucoma (47% vs. 15%; P < 0.001), iris neovascularization (46% vs. 22%; P < 0.001), ectropion uveae (39% vs. 14%; P < 0.001), and orbital cellulitis (3% vs. <1%; P = 0.05). On the basis of International Classification of Intraocular Retinoblastoma, group E tumor had a statistically significant higher incidence of high-risk retinoblastoma compared with controls (39% vs. 16%; P = 0.01). Multivariate analysis of clinical features at presentation that predicted high-risk features on histopathology included prolonged duration of symptoms of >6 months (P = 0.008) and secondary glaucoma (P = 0.021). CONCLUSIONS: In this study, the clinical features at presentation predictive of high-risk features on histopathology included prolonged duration of symptoms of >6 months and secondary glaucoma. Globe-preserving methods of treatment should be used with caution in patients with these features.