The purpose of this report is to assess the efficacy and adverse event profile of photon-based stereotactic radiosurgery (SRS) compared with fractionated stereotactic radiotherapy (fSRT) for the treatment of uveal melanoma. Primary outcomes include incidence proportions of local control, enucleation, metastatic progression, disease-specific, and overall mortality. Treatment-related toxicities such as incidence proportions of radiation retinopathy, neovascular glaucoma, optic neuropathy, and cataract formation were examined as secondary outcomes. Five-year survival and 5-year local control rates were also assessed.

PubMed, Embase, Web of Science, Scopus, and 2 Cochrane databases were searched up to December 31, 2018. Random effects models were used to calculate pooled incidence proportions of outcome measures. Meta-regression was carried out to explore the potential impact of dose per fraction on local control.

Twenty-four articles with a total of 1745 patients were included in the meta-analysis. There were no statistically significant differences between photon-based fSRT and SRS for all primary, secondary and 5-year outcome measures, including local control (P=0.28), enucleation (P=0.51), and neovascular glaucoma (P=0.40). The 5-year local control rate was 90% (95% confidence interval: 76%, 96%) for fSRT and 89% (70%, 97%) for SRS.

Our meta-analysis showed no difference in tumor control, survival and toxicities, as defined in this paper, between SRS and fSRT for uveal melanoma. Confounding biases remain an expected limitation in this study of novel treatment modalities deployed in rare tumors. Further investigation is needed to validate outcomes and compare stereotactic treatment techniques.