Genomic Analysis of Circulating Tumor DNA in 3,334 Patients with Advanced Prostate Cancer Identifies Targetable BRCA Alterations and AR Resistance Mechanisms

Abstract

Purpose: Comprehensive genomic profiling (CGP) is of increasing value for patients with metastatic castration-resistant prostate cancer (mCRPC). mCRPC tends to metastasize to bone, making tissue biopsies challenging to obtain. We hypothesized CGP of cell-free circulating tumor DNA (ctDNA) could offer a minimally invasive alternative to detect targetable genomic alterations (GA) that inform clinical care.

Experimental design: Using plasma from 3,334 patients with mCRPC (including 1,674 screening samples from TRITON2/3), we evaluated the landscape of GAs detected in ctDNA and assessed concordance with tissue-based CGP.

Results: A total of 3,129 patients (94%) had detectable ctDNA with a median ctDNA fraction of 7.5%; BRCA1/2 was mutated in 295 (8.8%). In concordance analysis, 72 of 837 patients had BRCA1/2 mutations detected in tissue, 67 (93%) of which were also identified using ctDNA, including 100% of predicted germline variants. ctDNA harbored some BRCA1/2 alterations not identified by tissue testing, and ctDNA was enriched in therapy resistance alterations, as well as possible clonal hematopoiesis mutations (e.g., in ATM and CHEK2). Potential androgen receptor resistance alterations were detected in 940 of 2,213 patients (42%), including amplifications, polyclonal and compound mutations, rearrangements, and novel deletions in exon 8.

Conclusions: Genomic analysis of ctDNA from patients with mCRPC recapitulates the genomic landscape detected in tissue biopsies, with a high level of agreement in detection of BRCA1/2 mutations, but more acquired resistance alterations detected in ctDNA. CGP of ctDNA is a compelling clinical complement to tissue CGP, with reflex to tissue CGP if negative for actionable variants.See related commentary by Hawkey and Armstrong, p. 2961.

Figure 1.. Genomic landscape of prostate cancer in liquid biopsies.

A) Distribution of estimated tumor fraction within each liquid biopsy dataset B) Frequency and cooccurrence of alterations in genes associated with mCRPC across 2,213 liquid biopsy samples assayed with 70 gene panel. Variant type indicated by color legend at the top of the oncoprint. MSI-H status indicated in last row. Estimated tumor fraction indicated by bar below the oncoprint. Copy number deletions not reported by the liquid biopsy assay versions used in this study. C) Frequency of short variants detected in metastatic tissue samples versus frequency of short variants detected in liquid samples. Genes with short variants with significantly different frequencies in tissue and liquid are color coded to reflect p-value. *TP53 off-scale (45.4% in liquid vs 40.8% in tissue, p = 0.0096).

Figure 2.. BRCA1/2 alterations in liquid biopsy.

A) Prevalence of BRCA1/2 alterations in the three cohorts B) Types of detected BRCA1/2 alterations. 174 frameshifts among 157 patients for BRCA2, 25 frameshifts among 25 patients for BRCA1, 58 BRCA1/2 nonsense point mutations among 57 cases, 51 patients with rearrangements, 28 missense mutations among 22 patients, and 16 splice site alterations among 14 patients, 49 non-frameshift deletions among 7 patients (these indels were all reversion mutations). C) Numbers of BRCA1/2 alterations per patient D) Variant allele frequencies of BRCA1/2 short variants in ten patients with detected reversion mutations. Variants with unknown functional status are splice site mutations. See table S2 for details. E) Variant allele frequencies of short variants in BRCA1 and BRCA2 compared to the ctDNA fraction of the liquid biopsy. Germline variants were predicted using heuristic scoring of observed instances across all FMI datasets.

Figure 3.. Concordance of BRCA1/2 detection in liquid and tissue biopsy.

A) 92 tissue/liquid pairs where BRCA1/2 variants were detected among the tissue sample alone (n=5), the liquid sample alone (n=20), or both (n=67). Samples in each group arranged in ascending ctDNA fraction (gray bar). Variant allele frequency is indicated for each short variant. Rearrangements, for which VAF was not reported, indicated at the top of the chart. Tissue only variants, with no associated liquid VAF, also indicated at the top of the chart. For simplicity and clarity, all analyses presented in this figure omit 9 BRCA reversion mutations detected in one sample. All alterations presented are predicted deleterious to BRCA1/2 function. B) Patient-level BRCA1/2 mutant status was assigned in the presence of at least one deleterious alteration in BRCA1 or BRCA2 in a sample. No patient in this study had multiple discordant BRCA1/2 variants assigned in tissue and liquid tests. Positive percentage agreement (PPA): the number of patients assigned BRCA1/2-mutant status by both liquid and tissue biopsies divided by the total number of BRCA1/2-mutant patients identified by tissue biopsy. Negative percent agreement (NPA), was also calculated with tissue biopsy taken as standard. OPA calculated as patients assigned similarly by both tests divided by total patients in paired comparison. C) SGZ algorithm predictions of variants germline/somatic status using the tissue biopsy, and the proportions of these variants also detected in the matching liquid biopsy. D) Comparison of VAF of short variants in liquid versus matched tissue biopsy. Variants were classified as detected in liquid only, or detected in tissue.

Figure 4.. AR alterations in liquid biopsy.

A) Oncoprints of 940 AR-altered samples, divided as separate cohorts and the aggregate B) Polyclonality of AR activating mutations: numbers of AR short variants and rearrangements per sample. SV= short variant, RE= rearrangement C) Distribution of oncogenic missense mutations in AR. Letters indicate amino acid change when there is >1 missense mutation at the position D) Rare AR alterations identified near the C-terminus of the ligand binding domain: Compound missense mutations in cis (gold) and in-frame deletions (grey) spanning important androgen-binding residues H875, F877, and T878, all resulting in S885 moving into the 878 position (red). One sample contained an isoleucine insertion (light blue). F877L/T878A double mutants are predicted to have enhanced resistance to enzalutamide (bold). 16 compound mutations were found in 12 patients, and 11 in-frame deletions among as many patients. All patients were confirmed to have progressed on at least one of abiraterone, enzalutamide or apalutamide, except 2 patients with compound mutations for whom treatment information was not available. E) Map of AR rearrangements that describe breakpoints for translocations and deleted, duplicated, or inverted regions (22 translocations, 60 deletions, 53 duplications, and 25 inversions). X-axis is a schematic representations of the 8 exons in the AR gene (not to scale). Among the 160 patients with AR rearrangements, 138 were confirmed to have progressed on at least one of abiraterone, enzalutamide or apalutamide, and 22 had no available treatment information. F) Patient-matched sample pairs collected within 30 days of each other with ≥1 AR short variant detected, in ascending order of ctDNA fraction. Bar represents estimated ctDNA fraction of the liquid biopsy. Tumor fraction of tissue biopsy listed on left. Table to the right lists short variants identified exclusively in tissue (orange), in both tissue and liquid (blue), or exclusively in liquid (green). Ratio of VAF/ctDNA fraction listed in parentheses after each variant detected in liquid biopsy. * = VAF can exceed ctDNA fraction if mutation is an amplified copy of AR. G) Tissue-liquid pairs in which an AR amplification was detected in tissue. Correlation of copy number, ctDNA fraction and detection in the matched liquid biopsy.