Prognostic value of 8q gain in relation to BAP1 and SF3B1 mutated uveal melanoma

Uveal melanoma (UM) is the most common primary intraocular malignancy in adults [1]. Recurrent mutations in secondary driver genes BAP1, SF3B1, and EIF1AX, as well as characteristic copy number variations (CNV) and gene expression profiles (GEP) are used in prognostication. Previously, gain of chromosome 8q has been associated with poor prognosis, with increased numbers of 8q correlating with shorter survival [2]. Whereas in BAP1-mutated (BAP1^MUT) UM, gain of 8q is the result of whole chromosome 8 gain or the formation of isochromosome 8q, in case of SF3B1-mutated (SF3B1 ^MUT) UM structural, often partial gain of 8q is predominant [3]. Some of these SF3B1 ^MUT UM patients develop early-onset metastatic disease [4], prompting to investigate the relationship between survival, gain of 8q and SF3B1 ^MUT UM.

Patients with SF3B1 ^MUT tumors (n=59) from the Rotterdam Ocular Melanoma Study group diagnosed between 1994 and 2022 were included in this study. CNVs were assessed using single nucleotide polymorphism (SNP) array (n=55) or in the past with karyotyping (n=17) and or fluorescence in situ hybridization (FISH, n=28) and when material was available used for transcriptome profiling (n=19) [4]. The summary plot showing the chromosomal patterns of the SNP arrays can be found in the supplementary (available at www.aaojournals.org). Disease-free survival (DFS) was determined using a cut-off of 60 months to identify early-onset metastatic disease in UM patients [4].

Gain of chromosome 8q was present in 48 tumors (81%) (Figure 1a). Cox proportional hazard analysis could not confirm reported independent [2] prognostic value of gain of 8q in general (>2 copies) in SF3B1 ^MUT UM (HR: 1.042 (95%;CI:0.3105-4.733)), nor the number of additional 8q copies (3 copies of 8q, HR: 1.213 (95%;CI:0.3363-5.736); ≥4 copies of 8q, HR: 0.6913 (95%;CI:0.08939-4.264)) in SF3B1 ^MUT tumors. Kaplan-Meier survival analysis also indicated no difference in survival in patients with or without 8q gain (p=0.9854; Figure 1b) nor is there a difference between 2, 3 or ≥4 copies (p=0.6927; Figure 1d). However, since 8q gain is a characteristic of SF3B1 ^MUT as well as BAP1 ^MUT tumors, the prognostic value of 8q gain could also be only attributed to BAP1 tumors. Therefore, we assessed the survival of 211 UM patients with immunohistochemically BAP1-negative tumors or tumors with BAP1 mutations. Gain of chromosome 8q was present in 181 tumors (86%) and this was correlated with a worse survival in BAP1 ^MUT UM (p=0.0134; Figure 1c). Three copies of 8q was not associated with decreased survival (HR: 1.667 (95%;CI:0.9755-3.042) but more copies (≥ 4 copies) had predictive value (HR: 1.907 (95%;CI:1.103-3.507, Figure 1e). Since (partial) gain of 8q can also be accompanied with changes in 8p copy number, 8p status in SF3B1 ^MUT tumors was also assessed in 8q gain tumors (1 copy 8p (2.1%); 2 copies 8p (77.1%); ≥3 copies 8p (20.8%)) (Figure 1a). Next, the survival plot was stratified on 8p status. Patients with a SF3B1 tumor and loss or gain of 8p (1 copy 8p, n=1; 3 copies 8q, n=4) in combination with 3 copies of 8q had the shortest survival (p=0.0352). Since BAP1 ^MUT tumors have recurrent isochromosome 8q, we explored the combination of 8q and 8p status, where 8q gain was accompanied with 1 copy (40%), 2 copies (35%) and ≥3 copies of 8p (25%). However, no difference was found in 8p status between SF3B1 ^MUT and BAP1 ^MUT tumors (p=0.4149). Nonetheless, many aberrations can be overlooked with single probe detection. No difference was found in copies of 8p (p=0.2274) and 8q (p=0.8237) between DFS<60 months (n=8) and DFS≥60 months (n=51) in SF3B1 ^MUT tumors (Figure 1a). Distal gains (8q23-8q24.3) were similar in patients with a DFS<60 months and DFS≥60 months (71% vs 78%; p=0.7122), more proximal (8q1-8q23) gains were more prominent in DFS<60 patients, though this difference was not significant (71% vs 41%; p=0.1461). Fifteen patients have metastatic disease (DFS<60 months, n=8 (88% 8q gain); DFS≥60 months, n=7 (71.4% 8q gain)) [4]. When grouping patients based on DFS, the number of 8q copies was not associated with survival (DFS<60, p=0.5319; DFS≥60, p=0.2328) (Figure 1f-g). Interestingly, ≥4 copies of 8q corresponded to the best survival (100%) in patients with DFS≥60 months. Of the patients with DFS≥60 months and ≥4 copies of 8q, 50% had ≥3 copies of 8p, whereas 25% of the patients with DFS<60 months and ≥4 copies of 8q had ≥3 copies of 8p. Transcriptome profiling and evaluating the genes from the GEP test on SF3B1 ^MUT (n=12) and BAP1 ^MUT (n=7) tumors showed no discriminating factor with 8q status (p=0.4149). One of the genes distinguishing DFS<60 and DFS≥60, TOP1MT [4] is located at 8q24.3 and did not cluster to any group (Figure 1h).

• Download figure
• Open in new tab

Figure 1.

Frequency of 8p and 8q copies in (a) all SF3B1 ^MUT tumors, 8q gain SF3B1 ^MUT tumors, and early (< 60 months) vs late (≥ 60 months) DFS; chromosome 8q status in (b) SF3B1 ^MUT tumors and (c) BAP1 ^MUT tumors; chromosome 8q copy numbers in (d) SF3B1^MUT tumors and (e) BAP1^MUT tumors; chromosome 8q copy numbers in (f) early and (g) late DFS SF3B1^MUT tumors; (h) heatmap using Z-scores of the GEP test of UM including TOP1MT on SF3B1^MUT and BAP1^MUT tumors [4]. Samples were clustered based on mutational and 8q status. No BAP1^MUT tumors had normal copies of 8q.

SF3B1 ^MUT and BAP1 ^MUT tumors often have (partial) 8q gain [3, 5]. Gain of 8q is correlated to prognosis in UM, with increased metastatic risk correlates with increased 8q copy numbers, often a result of isochromosome formation [2, 5]. Here, we show that 8q gain has no discriminating association with survival in early versus late metastasizing SF3B1 ^MUT tumors. Nevertheless, gain of 8q could still play a role in metastases in SF3B1 ^MUT tumors, independent of DFS. BAP1 ^MUT tumors contribute to >50% of all UM and are characterized by gains and losses of entire chromosomes or chromosome arms. Gain of 8q is often accompanied with monosomy 3, and its combination correlates with a worse prognosis [6, 7]. In contrast, SF3B1 ^MUT UM karyotypes are more complex and CNV often have recurrent distal breakpoints on chromosome 6 and 8. This difference in CNV patterns could indicate a separate tumorigenesis mechanism in both groups. Since 8q gain was correlated with survival in only BAP1 ^MUT tumors and not SF3B1 ^MUT tumors, the reported decreased survival in 8q gain is predominantly due to BAP1 ^MUT tumors.

To conclude, we have evaluated gain of chromosome 8q and its role on DFS in SF3B1 ^MUT and BAP1 ^MUT UM. There is no correlation between 8q gain and early-onset metastasis in SF3B1 ^MUT tumors. Gain of 8q has no additional predictive value in SF3B1 ^MUT tumors. In contrast, 8q gain is predictive for a worse prognosis in BAP1 ^MUT tumors. Thus, 8q gain has additional predictive value for BAP1 ^MUT tumors, but not for SF3B1 ^MUT tumors.