PARP inhibitors (PARPi) are effective in tumors with homologous recombination repair (HRR) deficiency (HRD), typically identified by germline/tumor mutations. However, genetic testing may miss intrinsic PARPi sensitivity and resistance. We evaluated strategies to improve detection and longitudinal monitoring of HRD across >500 tumor samples and >20 paired liquid biopsies, integrating genetic, genomic, and functional readouts. HRD was more frequent in high-grade ovarian cancer (HGOC; 52%) than in metastatic breast (mBC; 12%) or prostate cancer (mPC; 20%). Assay concordance was low-to-moderate, underscoring complementarity. RAD51 testing and the genomic instability score identified HRD in tumors lacking pathogenic HRR mutations (6% and 30% in mBC, 38% and 46% in HGOC, 14% and 27% in mPC, respectively). Longitudinal ctDNA sequencing revealed BRCA1/BRCA2 reversion mutations in >30% of post-PARPi mBC samples, which were associated with poor response to subsequent platinum therapy. These findings support the use of complementary HRD biomarkers in tissue and liquid biopsy to guide PARPi use and monitor response.

Analytical validation of the VHIO-YWBC protocol

Complete information of the Case cohort, with clinical features and results from the somatic mutation analysis

Analytical validation of the VHIO-YWBC protocol for NGS

Breast milk charactherization as liquid biopsy specim and comparison to plasma

Diagnostic images of the PPBC from case #15 right breast

AURORA NGS panel gene list.

VHIO-YWBC NGS panel gene list

List of oligos and probes for each ddPCR assay

Table 1. Samples analyzed by NGS Table 2. VHIO-300 results from solid tumor Table 3. VHIO-YWBC results from BM positive for ctDNA

Sensitivity of the BM ctDNA detection assays.

Women aged ≤45 years face a 10-30% increased risk of developing breast cancer (BC) within the first decade postpartum compared to nulliparous women, with the risk rising with the age of pregnancy. Furthermore, the prognosis for BC diagnosed up to 5-10 years postpartum (PPBC) is worse even after adjusting for clinicopathological factors. Previous studies point to short lactation and subsequent abrupt breast involution as major tumorigenic factors. However, the lack of systematically collected breastfeeding and parity data from YWBC obscures the validation of this theory. FFPE primary tumors from 35 cases of HR+ YWBC were categorized as nulliparous, pregnant, or postpartum (PPBC ≤5 years) based on the time of diagnosis. Reproductive history and breastfeeding duration data were collected. GeoMx Digital Spatial Protein platform was employed to analyze an average of 3 regions of interest per tumor, focusing on 41 markers for immune cell and cell death profiling. Tumor and stromal compartments were marked by PanCK and CD45, respectively. In YWBC, Luminal B tumors exhibited higher immune cell presence in both tumor and stroma compared to Luminal A. HR+ lymph node-positive disease displayed elevated stromal infiltrating lymphocytes. In the PPBC subgroup, the CD68 macrophage marker showed significant up regulation in both compartments, compared to other samples (pval 0,001 FDR<0,05). Longitudinal analysis showed no substantial changes in the immune milieu across the three trimesters of pregnancy. However, markers of M2 macrophages, endothelial progenitor cells, GZMB, among others, significantly increased in the stromal and tumor compartments in correlation with breastfeeding duration (p-value <0.01). Our results, with the resolution provided by GeoMX, suggest that immune cell markers remain stable during pregnancy, possibly due to the mammary gland focusing on epithelial proliferation at that stage. After birth, breastfeeding duration significantly impacts the immune milieu, correlating with an increase in both immunosuppressive and tumor-killing markers, even before formal involution.

Liquid biopsy has proven valuable in identifying individual genetic alterations; however, the ability of plasma ctDNA to capture complex tumor phenotypes with clinical value is unknown. To address this question, we have performed 0.5X shallow whole-genome sequencing in plasma from 459 patients with metastatic breast cancer, including 245 patients treated with endocrine therapy and a CDK4/6 inhibitor (ET + CDK4/6i) from 2 independent cohorts. We demonstrate that machine learning multi-gene signatures, obtained from ctDNA, identify complex biological features, including measures of tumor proliferation and estrogen receptor signaling, similar to what is accomplished using direct tumor tissue DNA or RNA profiling. More importantly, 4 DNA-based subtypes, and a ctDNA-based genomic signature tracking retinoblastoma loss-of-heterozygosity, are significantly associated with poor response and survival outcome following ET + CDK4/6i, independently of plasma tumor fraction. Our approach opens opportunities for the discovery of additional multi-feature genomic predictors coming from ctDNA in breast cancer and other cancer-types.

Abstract Breast cancer occurring during pregnancy (PrBC) and postpartum (PPBC) is usually diagnosed at more advanced stages compared with other breast cancer, worsening its prognosis. PPBC is particularly aggressive, with increased metastatic risk and mortality. Thus, effective screening methods to detect early PrBC and PPBC are needed. We report for the first time that cell-free tumor DNA (ctDNA) is present in breast milk (BM) collected from patients with breast cancer. Analysis of ctDNA from BM detects tumor variants in 87% of the cases by droplet digital PCR, while variants remain undetected in 92% of matched plasma samples. Retrospective next-generation sequencing analysis in BM ctDNA recapitulates tumor variants, with an overall clinical sensitivity of 71.4% and specificity of 100%. In two cases, ctDNA was detectable in BM collected 18 and 6 months prior to standard diagnosis. Our results open up the potential use of BM as a new source for liquid biopsy for PPBC detection. Significance: For the first time, we show that BM obtained from patients with breast cancer carries ctDNA, surpassing plasma-based liquid biopsy for detection and molecular profiling of early-stage breast cancer, even prior to diagnosis by image. See related commentary by Cunningham and Turner, p. 2125. This article is featured in Selected Articles from This Issue, p. 2109

Fanconi anemia (FA) patients display an exacerbated risk of oral squamous cell carcinoma (OSCC) and oral potentially malignant lesions (OPMLs) at early ages. As patients have defects in their DNA repair mechanisms, standard-of-care treatments for OSCC such as radiotherapy and chemotherapy, give rise to severe toxicities. New methods for early diagnosis are urgently needed to allow for treatment in early disease stages and achieve better clinical outcomes. We conducted a prospective, longitudinal study wherein liquid biopsies from sixteen patients with no clinical diagnoses of OPML and/or OSCC were analyzed for the presence of mutations in cancer genes. The DNA from saliva and plasma were sequentially collected and deep-sequenced, and the clinical evaluation followed over a median time of approximately 2 years. In 9/16 FA patients, we detected mutations in cancer genes (mainly TP53) with minor allele frequencies (MAF) of down to 0.07%. Importantly, all patients that had mutations and clinical follow-up data after mutation detection (n = 6) developed oral precursor lesions or OSCC. The lead-time between mutation detection and tumor diagnosis ranged from 23 to 630 days. Strikingly, FA patients without mutations displayed a significantly lower risk of developing precursor lesions or OSCCs. Therefore, our diagnostic approach could help to stratify FA patients into risk groups, which would allow for closer surveillance for OSCCs or precursor lesions.

There is a great need for non-invasive tools that inform of an early molecular response to cancer therapeutic treatment. Here, we tested the hypothesis that proteolytically resistant proteins could be candidate circulating tumor biomarkers for cancer therapy. Proteins resistant to proteolysis are drastically under-sampled by current proteomic workflows. These proteins could be reliable sensors for the response to therapy since they are likely to stay longer in circulation. We selected manganese superoxide dismutase (SOD2), a mitochondrial redox enzyme, from a screening of proteolytic resistant proteins in breast cancer (BC). First, we confirmed the robustness of SOD2 and determined that its proteolytic resistance is mediated by its quaternary protein structure. We also proved that the release of SOD2 upon chemotherapy treatment correlates with cell death in BC cells. Then, after confirming that SOD2 is very stable in human serum, we sought to measure its circulating levels in a cohort of BC patients undergoing neoadjuvant therapy. The results showed that circulating levels of SOD2 increased when patients responded to the treatment according to the tumor shrinkage during neoadjuvant chemotherapy. Therefore, the measurement of SOD2 levels in plasma could improve the non-invasive monitoring of the therapeutic treatment in breast cancer patients. The identification of circulating biomarkers linked to the tumor cell death induced by treatment could be useful for monitoring the action of the large number of cancer drugs currently used in clinics. We envision that our approach could help uncover candidate tumor biomarkers to measure a tumor’s response to cancer therapy in real time by sampling the tumor throughout the course of treatment.

Circulating tumor DNA (ctDNA) is increasingly being used as a biomarker in early breast cancer (EBC). We performed a systematic review and meta-analysis to investigate the prognostic value of ctDNA in patients with EBC treated with neoadjuvant therapy (NAT). We searched Medline, Web of Science and Embase for observational or interventional studies that included patients with EBC undergoing NAT, reported outcomes related to the predefined endpoints, and had full text articles available. Study selection followed the PRISMA guidelines and quality assessment the REMARK tool for biomarker studies. Primary endpoint was impact of ctDNA detection in different time points (baseline, on-treatment, and after NAT) on relapse-free survival (RFS) and overall survival (OS). Secondary endpoints included the association of ctDNA detection with pathologic complete response (pCR), and the positive and negative predictive value of ctDNA detection in predicting residual disease after NAT. From the 2908 studies initially identified, 11 met the eligibility criteria and were included in the meta-analysis. Detection of ctDNA, both at baseline and after completion of NAT, significantly associated to worse RFS (HR 4.22, 95% CI: 1.29-13.82 and HR 5.67, 95% CI: 2.73-11.75, respectively) and worse OS (HR 19.1, 95% CI: 6.9-53.04 and HR 4.00, 95% CI: 1.90-8.42, respectively). In contrast, detection of ctDNA did not associate with the probability of achieving a pCR. Our results suggest that ctDNA assessment during NAT for EBC merits further evaluation as a stratification risk factor in prospective trials, in order to better individualize patient's treatment.