Presentation Authors: Alberto Martini, Joanna Wang, Nicholas Brown, Shivaram Cumarasamy*, Eithan Wajswol, John P Sfakianos, Ardeshir R Rastinehad, Kenneth G Haines III, N Peter Wiklund, Sujit S Nair, Ashutosh K Tewari, New York, NY
Introduction: The presence of tertiary pattern 5 (TP5) on radical prostatectomy specimens has been associated with worse long-term outcomes. However, the molecular underpinnings of TP5 is poorly understood. We hypothesized that a comparative analysis of transcript profiles between Gleason matched tumors with and without TP5 would provide us a unique opportunity to identify genes that contribute to TP5 and investigated if genomic features of TP5 could explain the poor clinical outcomes associated with this disease subtype.
Methods: Data from 159 men who underwent radical prostatectomy and had Gleason Grade Group (GGG) 3 or 4 without primary or secondary pattern 5 on final pathology were considered. All patients had Decipher diagnostic test (assesses 5-year risk of metastasis post-RP). Additionally, transcript profiles and SCAN-normalized expression of coding genes were available for all 159 patients.The relationship between Decipher score and the presence of TP5 was investigated by means of linear and binary logistic regression. A differential transcriptomic analysis between patients with and without TP5 was performed in order to identify the genes associated with TP5 as a proxy of early dedifferentiation. The prognostic role of these genes in identifying patients with worse progression-free survival (PFS) and overall survival (OS) was then evaluated by utilizing The Cancer Genome Atlas provisional (TCGAp) accessed through the cBioPortal for Cancer Genomics.
Results: Overall, 52/159 (33%) patients had GGG 3-4 disease with TP5 while 107/159 (67%) harbored GGG 3-4 disease alone. The presence of TP5 was associated with a higher Decipher score (B: 0.07, 95% CI: 0.02, 0.13, p=0.04) and a higher likelihood of falling within the intermediate- or high-risk categories (OR: 3.34, 95% CI: 1.34, 8.35, p=0.01) rather than the low-risk category. Analysis of microarray data revealed an 18 gene signature that was differentially expressed in patients with TP5. Genes CDKN2B, CHIT1, PDSS1, CDK4, SRD5A1, CXCR6, DCK, PLK1, CDC20, FGF14, FLRT2, RARA, and CXXC5 were over-expressed in the TP5 cohort.A systematic analysis of genes in the TCGAp revealed that overexpression of CDKN2B, PLK1, and CDC20 was associated with worse PFS. When combined, the group harboring overexpression of at least one of these genes had a 5-year PFS rate of 50% versus 74% in the group without overexpression, p < 0.001.
Conclusions: Our study has elucidated unique genomic features of TP5 while confirming previous findings that patients harboring TP5 tend to have worse prognosis. Analysis of transcript profiles revealed that the expression levels of 18 genes were significantly altered in TP5 positive versus negative specimens. Furthermore, there is evidence that alterations in the expression levels of cell cycle genes CDKN2B, PLK1, and CDC20 are critical drivers of TP5. Analysis of independent prostate cancer datasets demonstrated that the overexpression of all three genes was associated with worse PFS. To the best of our knowledge, this is the first RNA-based study to investigate the molecular diversity of TP5 and the first one correlating CKDN2B to poorer prognosis in patients with prostate cancer.