**3. Materials and Methods**

Included cases had adenocarcinoma of prostatic origin and were selected based on availability of tissue and matched blood specimens and micro- or overt metastatic disease either at the time of sampling or subsequent to radical prostatectomy. Patients sampled at the time of radical prostatectomy (primary tissue) had either pathologically confirmed lymph node metastases (19011, 19145, 19260, 19651) at diagnosis or subsequent metastatic relapse post-surgery (5545, 5684, 12543, 13179). Patients recruited at presentation of distant metastases had bone (1135, 147, A153, PCSD13) or brain (80002) tissue sampled.

All samples were obtained with written informed consent, as per the study approval granted from the St. Vincent's Human Research Ethics Committee (HREC), SVH/12/231 and HREC/12/SVH/323, Melbourne Health Human Research Ethics Committee HREC/12/MH/272 and Epworth Health 55512, or University of California Institute Review Board (IRB) approval 090401. Samples were shipped to the Garvan Institute of Medical Research in accordance with institutional Material Transfer Agreements (MTAs), and genomic screening and analysis were performed in accordance with approval granted by St. Vincent's Hospital HREC SVH/15/227 and governance review authorisation granted for human research at the Garvan Institute of Medical Research GHRP1522.

Primary tumour samples were collected at the time of radical prostatectomy and two core biopsies were taken from the prostate regions with cancer on preoperative biopsy. Lymph node tissue was collected at the time of radical prostatectomy from nodal masses with palpable tumour. Metastatic samples were obtained by image guided biopsy or at surgical resection (80002, PCSD13). All tissue samples were snap frozen. The presence of prostate cancer and its location within the samples was confirmed by a pathologist prior to dissection for DNA extraction. DNA was extracted from tissue and buffy coat or whole blood using one of two commercially available kits: the DNeasy blood and tissue kit protocol (Qiagen, Maryland), or for high molecular weight (HMW) DNA, the Bionano Prep Frozen Human Blood and Animal Tissue DNA isolation protocols (Bionano Genomics, San Diego document #30246 and #30077).

Demographic, clinical and pathological data were collected for each patient and are summarised in Figure 1 and Table 1. The median patient age at the time of PCa diagnosis was 65 years (range 51–77). The median time to biochemical recurrence (BCR) for those that underwent definitive first-line treatment and subsequently relapsed (*n* = 8) was 43.5 months (range 6–93); six of these patients relapsed with metastatic disease detectable on standard imaging at a median time of 86.5 months from initial diagnosis (range 24–120).
