*2.5. Case 80002: Core Biopsy at Resection of Brain Metastasis*

Patient 80002 presented with a solitary brain metastasis that was surgically resected. His PSA was elevated and morphology of the tumour specimen was consistent with an adenocarcinoma of prostatic origin; immunohistochemistry (IHC) markers for neuroendocrine differentiation were negative.

The relevant genomic features of this *TMPRSS2-ERG* fusion positive case are summarized in Figure 7 and include: *TP53* mutation (rs1057519999, c.716A > C; p.Asn239Thr) and SCNAs in *CDK12, RAD51C, RNF43, TP53,* and *BRAF*. This tumour presented with a high rate of SVs, including a large deletion overlapping *TP53,* a partial deletion of *LRP1B,* and an interchromosomal translocation involving *CTNNA1*, the downregulation of which is associated with a worse prognosis in PCa [91]. Using our WGM approach, we identified additional large heterozygous deletions. Two overlap TSGs including *TP53* and *KCTD11,* and another overlaps with *TBX3* [92] and *NRF2*. NRF2 has been shown to suppress PCa cell mitosis and migration [93,94]. Another large deletion on chromosome 2 overlapped *HOXD10* and *HOXD3*. Decreased HOXD10 expression promotes an aggressive phenotype in PCa in knockdown mice, as well on retrospective review of clinical outcomes [95] and *HOXD3* methylation predicts earlier BCR [96].

**Figure 7.** Summary of relevant genomic alterations for case 80002 with a brain metastasis at the time of sampling (80002) and Circos plots as per Figure 3.

Although COSMIC Mutational Signatures 1, 5, 8 and 9 are present, Signatures 17 and 18 contribute >5% each. Signature 18 may be associated with failure of base excision repair [97] and enriched in metastatic PCa [8]. Signature 17, predominantly found in gastric and oesophageal cancers, has been shown to co-occur with Signature 18 in mouse models of these cancers and this signature may be a by-product of oxidative damage [98,99].

Brain metastases are uncommon in prostate adenocarcinoma and tend to occur in cases with neuroendocrine differentiation [100]. However, gains in *FOXA1,* as seen in this case, are thought to protect against neuroendocrine trans-differentiation [101] and the *TMPRSS2-ERG* fusion supports the prostatic origin. This patient has a number of targets impacting androgen signalling, DDR and MAPK pathways. His clinical presentation would already support aggressive therapy with combination therapy and his genomic data include several poor prognostic features. The partial *LRP1B* deletion may produce sensitivity to pembrolizumab but the evidence for this is limited so this should only be considered as part of a clinical trial potentially upfront with docetaxel or later in his clinical course at development of CRPC. *KCTD11* is a negative regulator of hedgehog pathway signalling [102] and therefore its loss, identified using WGM, may increase signalling and imply this tumour would be sensitive to pathway inhibitors.
