Tumor, node, metastasis classification of malignant ciliary body and choroidal melanoma evaluation of the 6th edition and future directions.

PURPOSE: The tumor, node, metastasis classification of malignant uveal melanoma has been revised. We evaluated how the 6th edition (TNM6) improves on the previous one (TNM5).
DESIGN: Population-based, retrospective, cross-sectional study. PARTICIPANTS: Two hundred eighty-nine consecutive patients who had a ciliary body and choroidal melanoma treated in the district of the Helsinki University Central Hospital, Finland, between 1962 and 1981.
METHODS: Tumor dimensions, ciliary body involvement, and extraocular extension were evaluated from histopathologic sections and pathology reports. Tumors were assigned into categories and stages according to TNM6, TNM5, and 2 previously proposed size classifications. MAIN OUTCOME MEASURES: Proportion of tumors classified in each category and melanoma-specific survival by category and stage.
RESULTS: Of the 289 melanomas, 5% were classified as pT1, 63% as pT2, 22% as pT3, and 7% as pT4 according to TNM6. The corresponding percentages based on TNM5 were 8%, 17%, 63%, and 10%. Of pT2 tumors in TNM6, 4% came from pT1, 65% from pT3, and 4% from pT4 category of TNM5. Of 28 melanomas with extraocular growth, 29% were classified as pT2 in TNM6 rather than pT4 in TNM5. The 10-year survival estimate was 2 percentage points lower for pT1, 7 percentage points higher for pT2, 17 percentage points lower for pT3, and 13 percentage points lower for pT4 by TNM6 compared with TNM5; TNM6 (P<0.0001) and the modified alternative size classifications (P = 0.0022 and P = 0.0026) divided tumors according to prognosis better than TNM5. The 10-year survival for stage I, II, and III tumors was 2 percentage points lower, 7 points higher, and 23 points lower by TNM6, which was not better than TNM5 in separating patients according to prognosis (P = 0.47). The alternative size classifications provided more equal categories and fitted the data set better than TNM5 regarding prognosis.
CONCLUSIONS: TNM6 is an improvement over TNM5 in some, but not all, respects. Areas for development include taking into account ciliary body involvement and extraocular extension in more detail and combining into each stage tumor categories with similar rather than different prognosis. An evidence-based, multicenter approach would be beneficial.