Fig. 1: Discovery of PTN as a potential serum-based biomarker for clinically significant prostate cancer through high-multiplex immunoassays.

a Schematic representation of Olink targeted proteomics method. Immuno-Oncology and Oncology II panels utilise proximity extension assay (PEA) to measure 92 human proteins associated with immuno-oncology and 92 human protein biomarkers related to oncology. The detection of each protein is achieved by paired oligonucleotide-labelled antibodies. Binding of the antibodies to the protein target leads to DNA hybridisation upon proximity, resulting in the formation of a DNA reporter sequence. The DNA reporter sequence is quantified by qPCR. A list of the proteins in each panel is shown in Supplementary Table S2. The illustration was generated using BioRender (https://biorender.com). b Eighty serum samples from four different patient groups: (1) cancer-free group, (2) the Cambridge Prognostic Group 1 (CPG1), disease with 97% 10-year survival, (3) CPG5, disease with 50% 10-year survival and (4) men with metastatic disease since diagnosis were analysed using Immuno-Oncology and Oncology II panels. N = 20 per group. Heatmap of nine proteins found to be elevated in at least one group in the proteomics. Protein level (log₂, z-score). c Normalised protein expression (NPX) for pleiotrophin (PTN) is plotted. d PTN levels in the 80 serum samples used for proteomics analysis were analysed by Sandwich ELISA. Error bars represent standard deviation (SD). For all, *P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001 and n.s. not significant. Significance was determined by Student’s t test (two-tailed). e IHC staining of PTN on Stanford University tissue microarrays (TMAs). PTN staining intensity was scored from 0 to 3 (0 is negative, 1 is low, 2 is medium, 3 is high). Representative images for negative, low, medium and high PTN staining are shown in the right panel. Scale bars represent 50 μm (upper panel) and 25 μm (lower panel), respectively. PTN IHC on Stanford University TMAs with associated clinical data is shown in the left panel. High PTN expression correlates with lower 10-year recurrence-free survival. P = 0.049. f Comparative PTN IHC scores between those with CPG1 and CPG5 disease. g The distribution of PTN intensity scores as a percentage of patient samples. PTN score 0: PTN score of four different cores = 0; PTN score ≤ 1.5: average PTN score of four different cores ≤ 1.5; PTN score > 1.5: average PTN score of four different cores > 1.5. Strong staining for PTN correlates with prostate cancer recurrence after prostatectomy (recurrent patients: n = 56; non-recurrent patients: n = 171) (left panel). The statistical significance of the differences between non-recurrent and recurrent proportions was calculated by z-score normal distribution N (0,1). P = 0.03. High PTN levels correlate with prostate cancer metastasis (right panel). Samples from patients with no metastasis (n = 212) and patients with metastasis (n = 15, including nine patients with lymph node metastasis, three patients with bone metastasis, and three patients with both lymph node and bone metastasis) were analysed. The statistical significance of the differences between no metastasis and metastasis proportions was calculated by the normal distribution N (0,1) of z-scores. P = 0.04. h Multivariate analysis of high levels of PTN as an independent predictor of biochemical recurrence in patients with pathological Gleason grades 3 + 3 and 3 + 4. P < 0.05.