Single-Masked Randomized Phase 2 Study Assessing 2 Forms of Hypofractionated Proton Therapy in Patients With Large Choroidal Melanomas.

PURPOSE: Patients with large uveal melanomas are at major risk of liver metastases. Some patients are reluctant to undergo the standard treatment (ie, immediate enucleation). Proton therapy yields 5-year local control rates and eyeball retention of >85% and ≈20% in large uveal melanomas. Patients with T3/T4 uveal melanomas refusing enucleation were randomized between standard 4 to 13 Gy-fraction or moderately hypofractionated 8 to 6.5 Gy-fraction proton therapy. The main endpoint was the 2-year local recurrence-free survival without enucleation. METHODS AND MATERIALS: A single-masked 1:2 randomized phase 2 trial was conducted between 2015 and 2017 with planned endoresection and distance to the posterior pole as strata. Local events were defined as local relapse, or enucleation due to complications or relapse. RESULTS: The 32 patients, with a mean age of 64 years, had T3/4 (N = 17/15), M1 (N = 2) uveal melanomas, of mean tumor diameter and thickness of 16.5 mm and 9.1 mm, and of posterior location in 56.5%. Median follow-up was 56.7 months. The 2-year local recurrence-free survival rate without enucleation was 79% (95% confidence interval, 65%-96%), similar in both arms. There were 9 enucleations, 3 at relapse and 6 for toxicities. Twelve patients had distant metastases. The 2-year-overall survival was 72% (95% confidence interval, 58%-89%). At baseline, visual acuity by average logarithm value of the minimum angle of resolution was 0.68 and 0.70 in the standard and experimental arms, and at last follow-up 2 and 1.7, with mean differences of 1.44 and 1.01, respectively (P = .39). CONCLUSION: An 8-times 6.5 Gy scheme is feasible without deteriorating local control and with similar toxicity rates in patients with large uveal melanomas. Larger studies incorporating adjuvant treatments are warranted.

Management of Invasive Squamous Cell Carcinomas of the Conjunctiva.

PURPOSE: Ocular surface squamous neoplasia includes a spectrum of diseases from dysplasia to invasive squamous cell carcinoma (SCC) of the conjunctiva. Whether the degree of invasion influences outcomes is debated. We evaluated the outcomes and management of conjunctival carcinomas defined as ≤0.2 mm invasion of the chorion (microinvasive; miSCC) or over (SCC). DESIGN: Retrospective case series. METHODS: Clinical, tumor, and therapeutic characteristics and outcomes were collected for consecutive patients with histology-proven invasive conjunctival miSCC/SCC treated between 2002 and 2017. RESULTS: Patients were 70% men, ≥70 years old (56%), with carcinomas of the bulbar conjunctiva (83.0%). Limbal, corneal, and/or scleral involvement were present in 70.4%, 42.6%, and 27.8%, respectively. Patient characteristics, tumor characteristics, and no-touch surgery rates were similar between the 39 SCC and 15 miSCC. However, mitomycin was performed in 93.3% and 20.5% of miSCC and SCC, respectively (P < .001). Proton therapy was performed in 0% and 92.0% of miSCC and SCC, respectively (P < .001). SCC received mitomycin in case of tumoral resection margins, respectively (P = .018). The 24-month incidence of local relapse was 14.8%, including 20% and 12% for miSCC and SCC, respectively (P = .079). Irradiation was the only prognostic factor associated with a lower risk for local relapse (hazard ratio [0.25]; P = .045). There were 2 cancer-related deaths (2%). Mild/moderate anterior segment complications occurred in one third of the patients. CONCLUSIONS: miSCC had slightly worse relapse rates compared with SCC. Postoperative proton therapy, performed in SCC only, was associated with a lower risk for relapse.

Dry Eye Syndrome After Proton Therapy of Ocular Melanomas.

PURPOSE: To investigate whether proton therapy (PT) performs safely in superotemporal melanomas, in terms of risk of dry-eye syndrome (DES). METHODS AND MATERIALS: Tumor location, DES grade, and dose to ocular structures were analyzed in patients undergoing PT (2005-2015) with 52 Gy (prescribed dose, not accounting for biologic effectiveness correction of 1.1). Prognostic factors of DES and severe DES (sDES, grades 2-3) were determined with Cox proportional hazard models. Visual acuity deterioration and enucleation rates were compared by sDES and tumor locations. RESULTS: Median follow-up was 44 months (interquartile range, 18-60 months). Of 853 patients (mean age, 64 years), 30.5% had temporal and 11.4% superotemporal tumors. Five-year incidence of DES and sDES was 23.0% (95% confidence interval [CI] 19.0%-27.7%) and 10.9% (95% CI 8.2%-14.4%), respectively. Multivariable analysis showed a higher risk for sDES in superotemporal (hazard ratio [HR] 5.82, 95% CI 2.72-12.45) and temporal tumors (HR 2.63, 95% CI 1.28-5.42), age ≥70 years (HR 1.90, 95% CI 1.09-3.32), distance to optic disk ≥5 mm (HR 2.71, 95% CI 1.52-4.84), ≥35% of retina receiving 12 Gy (HR 2.98, 95% CI 1.54-5.77), and eyelid rim irradiation (HR 2.68, 95% CI 1.49-4.80). The same risk factors were found for DES. Visual acuity deteriorated more in patients with sDES (0.86 ± 1.10 vs 0.64 ± 0.98 logMAR, P=.034) but not between superotemporal/temporal and other locations (P=.890). Enucleation rates were independent of sDES (P=.707) and tumor locations (P=.729). CONCLUSIONS: Severe DES was more frequent in superotemporal/temporal melanomas. Incidence of vision deterioration and enucleation was no higher in patients with superotemporal melanoma than in patients with tumors in other locations. Tumor location should not contraindicate PT.