Introduction:

Primary prostate cancers (PCa) harbor multiple spatially distinct tumors with significant inter- and intra-tumoral molecular heterogeneity. This genomic diversity gives rise to many competing subclones that may drive the biological trajectory of localized PCa. Previous large scale sequencing efforts have focused on the evolutionary process of metastatic PCa, revealing a potential clonal progression to castration resistance. However, the clonal origin of lymph node (LN) metastases in primary disease is still unknown. Here, we performed multiregional targeted DNA/RNA next generation sequencing (NGS) of primary PCa with synchronous LN metastases to better define the cancer foci or region capable of metastasis.

Methods:

Patients who underwent radical prostatectomy and LN dissection that revealed node positive disease were identified for this IRB-approved study. All pathology slides were re-reviewed by a dedicated genitourinary pathologist. Punch biopsies were performed from pre-identified multiple regions of cancer on formalin fixed paraffin embedded (FFPE) prostate and LN specimens from which DNA/RNA samples were co-isolated. High depth, targeted, multiplexed, polymerase chain reaction (PCR)-based DNA NGS was performed to characterize the genomic profile of each tumor region using two targeted DNA NGS panels: a custom PCa NGS panel (135 genes, 3,127 amplicons) and the comprehensive cancer panel (CCP; 409 genes and 15,992 amplicons). Targeted RNA NGS sequencing was performed to evaluate gene fusion status of each sample. We determined and compared somatic DNA mutations, copy number alterations (CNA), and gene fusion status between primary and LN disease. Phylogenetic analysis was performed to determine the likely clonal source of LN metastasis.

Results:

We analyzed 88 primary tumor (1^°) and 23 LN-metastases samples from 14 patients. After quality control, 11 patients (69 and 18 primary and LN tumor samples, respectively) had sufficient quality data for analyses. Seven had evidence of extraprostatic extension (EPE), with phylogenetic analysis supporting this as the source of LN-metastasis in 4 cases. In patient #1 (Figure 1), while all 1^° regions showed concordant TP53 and TPR non-synonymous mutations and broad CNAs with two LN-metastasis foci, only two of the 1^° regions showed high-level CNAs with both LN-metastasis foci. In two patients with pT2N1 disease, sub-clonal seeding and clonal evolution was observed with metastases arising from GG5 area. Seven had cribriform pattern in both LN and dominant 1^° foci. One patient showed CDK12 mutation within the 1^° and LN foci. FOXA1 was mutated in two patients within 1^° and LN foci. 

Conclusion:

Using targeted genomic and transcriptomic NGS to assess primary PCa and synchronous LN metastases, we demonstrate that LN metastases is idiosyncratic, likely related to a combination of histopathologic and genomic factors, including presence of high-grade disease, EPE, cell morphology such as cribriform pattern, and shared driver mutations such as CDK12 and FOXA1. These findings highlight the need to develop robust prognostic biomarkers for identification of novel therapeutic targets for LN metastasis.

Funding: Prostate Cancer Foundation Young Investigator Award, University of Michigan Prostate SPORE