Abstract 41: Multi-focal genomic dissection of synchronous primary and metastatic tissue from de novo metastatic prostate cancer

Abstract

Background: 10% of newly diagnosed prostate cancer presents with metastases. Known as de novo metastatic castrate-sensitive prostate cancer (mCSPC), it is disproportionally responsible for >50% of prostate cancer deaths. Cancer genotyping can identify vulnerabilities exploitable by targeted therapies, and promises to help prognosticate. However, tissue from de novo mCSPC is scarce; neither prostatectomy nor metastatic biopsy is standard, and it is unknown if diagnostic biopsies are representative of synchronous metastases. The potential for plasma circulating tumor DNA (ctDNA) to inform on tumor genotype is also unknown. Methods: We performed comprehensive pathological and genomic assessment of all spatially or phenotypically-distinct tumor foci (n=523) in 43 patients with de novo mCSPC who underwent prostatectomy, pelvic lymph node dissection, and plasma collection. Results: 91% (478/523) of tissue foci had evidence of prostate cancer by targeted DNA sequencing, with a median tumor fraction of 48%. When modeling random selection of a single primary foci (mirroring biopsy tissue availability in clinic), tumor fraction was <20% in 19% of patients. Only 46% of plasma cell-free DNA samples prior to systemic therapy had a ctDNA fraction above 0.3% (limit of detection); median tumor fraction of 5% in samples with confirmed ctDNA. We observed recurrent alterations in major driver genes, including TP53, FOXA1, PTEN, and RB1, and the genomic landscape was very similar to published cohorts of castration-resistant prostate cancer (excluding AR). Primary site genomic heterogeneity was pervasive, including secondary (clonally distinct) prostate cancer populations in 14% of patients. Polyclonal seeding of metastases was detected in 26% of patients. Biallelic inactivation of TP53, PTEN, and/or RB1 was observed in 63% of tumors, and was frequently found in synchronous metastases and ctDNA. The two patients with compound disruption of TP53, PTEN, and RB1 experienced rapid progression to castration-resistance and death within two years of diagnosis, despite initial low-risk clinical features. Across the cohort, biallelic disruption of TP53 together with high-risk clinical features at diagnosis was associated with rapid progression (HR 4.64 (95% CI: 1.70-12.69); P = 0.003). Conclusions: One fifth of patients with de novo mCSPC have pervasive low tumor fraction in their primary tumor and blood plasma. Many tumors exhibit spatial heterogeneity within the primary site, with evidence of multiple clones seeding metastases. This data raises concerns about accurate tumor genotyping in routine clinical practice where needle biopsy specimens are the only available tissue for profiling. Nevertheless, some de novo mCSPC are marked by aggressive genomics and experience rapid progression to lethal disease, suggesting that tailored multi-focal genomic profiling can further segment the disease. Citation Format: Evan Warner, Kim Van der Eecken, Andrew J. Murtha, Edmond M. Kwan, Sarah W. Ng, Xinyi E. Chen, Cecily Q. Bernales, Grainne Donnellan, Elena Schonlau, Sofie Verbeke, Nicolaas Lumen, Jo Van Dorpe, Gillian Vandekerkhove, Elie Ritch, Matti Annala, Bram De Laere, Piet Ost, Alexander W. Wyatt. Multi-focal genomic dissection of synchronous primary and metastatic tissue from de novo metastatic prostate cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 41.

Optimal systemic therapy in men with low-volume prostate cancer

“…Local consolidation refers to managing local disease in addition to systemic therapy. Recently, Warner et al [32] suggested that the prostate could potentially serve as a reservoir for distinct primary tumor clones, enabling multiple metastatic seeding events in patients with synchronous prostate cancer. The implication may provide a compelling biological rationale for local consolidated therapy in this setting.…”

“…Further risk stratification is needed in this setting, and advanced imaging in conjunction with genomics and liquid biomarkers may suggest who may benefit from MDT. Warner et al [32] recently postulated that among patients with SLV prostate cancer, particularly those with tumors lacking TP53 inactivation, may derive the greatest benefit from MDT. Recently, it was also demonstrated that PSMApositive extracellular vesicles may be used to select patients for Stereotactic Body Radiation Therapy (SBRT) with the potential of improving radiographic and PSA progression free survival without the need for systemic treatment [12 && ].…”

Current evidence on local therapy in oligometastatic prostate cancer

“…Thus, treatment of the primary location of cancer may improve clinical outcomes simply by tumor volume reduction [16]. Additionally, specifically regarding de-novo OMPC, as recently demonstrated by Warner et al [17], genomic assessment of the primary tumor post-radical prostatectomy (RP) can offer valuable insights for the implementation of genomics-guided patient management. In their study, Warner et al [17] observed a significant discordance in genotypes between certain areas of the primary tumor and synchronous metastases.…”

“…Additionally, specifically regarding de-novo OMPC, as recently demonstrated by Warner et al [17], genomic assessment of the primary tumor post-radical prostatectomy (RP) can offer valuable insights for the implementation of genomics-guided patient management. In their study, Warner et al [17] observed a significant discordance in genotypes between certain areas of the primary tumor and synchronous metastases. Another concept with the widespread adoption of PET imaging is that when conventional imaging is normal even in high-risk disease, yet PET demonstrates OMPC, robust data do suggest a benefit of RP versus nonlocal therapy.…”

Towards clinical implementation of circulating tumor DNA in metastatic prostate cancer: Opportunities for integration and pitfalls to interpretation

“…Low pathological tumor content and poor DNA yield were frequently responsible for unsuccessful sequencing, an issue that pervasively impacts prostate core needle biopsies obtained at initial diagnosis. Furthermore, marked genomic spatial heterogeneity within the prostate can undermine the ability of a single prostate biopsy specimen to correctly identify the dominant genotype responsible for metastatic disease development ( 8 , 9 ). In patients with metachronous metastatic disease, substantial time can separate initial diagnosis and subsequent development of clinical metastases ( 10 ).…”