Somatic Variants Research Articles

Abstract 4620: Assessing the impact of germline and somatic variants in breast cancer treatment

Abstract Germline genetic variants are hereditary DNA alterations, whereas cancer somatic variants originate in the tumor. For patients with breast cancer, treatment approvals focus on either germline or somatic data with no consideration of co-occurrence. As such, most clinical datasets lack either germline or somatic variant designation, and most molecular oncology datasets contain limited treatment information, focusing primarily on survival. Datasets that contain germline and somatic sequencing and treatment information are typically from single institutions, limiting their statistical power. This project seeks to aggregate datasets inclusive of germline and somatic sequencing and treatment information with the goal of starting to assess the effect of co-occurring variants on treatment response in breast cancer. We reviewed breast cancer datasets in CBioPortal and papers published in the past 15 years for germline and somatic sequencing with reported variants, treatment information, and associated responses. We identified 3 datasets: The Cancer Genome Atlas, The Metastatic Breast Cancer Project, and Fudan University Shanghai Cancer Center (BC cohort). We evaluated a total of 7 co-occurring gene pairs obtained from literature review: BRCA1 and BRCA2; BRCA2 and RB1, BRCA1 or 2 and TP53, PTEN, PIK3CA, MYC; and PALB2 and NOTCH2. We used the NIH Integrated Data Analysis Platform to harmonize variant calls across these datasets. For early-stage disease, we assessed disease-free survival (DFS) in response to the classes of chemotherapy or hormone therapy. For metastatic disease, we assessed duration of treatment response for specific treatments. We were able to assess one gene pair using our dataset – BRCA1/2 and TP53. Other gene pairs had 0-2 patients with co-occurring variants, limiting statistical analysis. We observed statistically significant differences in DFS among early-stage patients who received chemotherapy (Kruskal-Wallis test: <0.05), but observed patients with BRCA1/2 alone having shorter DFS relative to those with TP53 regardless of co-occurrence:No variants (n=212): Mean=33.32 months (SD=26.7 months)BRCA1/2 alone (n=141): Mean=18.32 months (SD=13.91 months)Co-occurring (n=10): Mean=49.17 months (SD=38.14 months)TP53 alone (n=95): Mean=41.45 months (SD=31.88 months) In the context of treatment, significantly more publicly available, better annotated data will be required to understand the influence of co-occurring variants, germline origin, and somatic origin on treatment response. In this preliminary analysis, we observed differential patterns of changes in treatment response based on the presence or absence of BRCA1/2 and TP53. These genes commonly co-occur – where co-occurrence is less common, we lacked power for our analysis. This question is under-addressed and may potentially influence treatment response and decision-making. Citation Format: Jonathan V. Wooten, Santiago Avila, Becca Pollard, Sheila Rajagopal. Assessing the impact of germline and somatic variants in breast cancer treatment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 4620.

Abstract 1907: A novel DNA library prep method enables high sensitivity variant detection from matched fresh frozen and FFPE patient tissues

Abstract In cancer genomics, a common source of DNA is formalin-fixed, paraffin-embedded (FFPE) tissue from patient surgical samples, where in most cases high quality fresh or frozen tissue samples are not available. FFPE DNA poses many notable challenges for preparing NGS libraries, including low input amounts and highly variable damage from fixation, storage, and extraction methods. It is difficult to obtain libraries with sufficient coverage and the sequencing artifacts arising from damaged DNA bases confound somatic variant detection. Additionally, many laboratories process FFPE tumor samples alongside matched, high quality, normal DNA and many library prep workflows are not readily compatible with both sample types. We developed a novel NGS library prep method compatible with both high quality and very low quality FFPE DNA samples, employing a novel enzymatic DNA repair mix, enzymatic fragmentation mix, and PCR master mix. To validate this workflow on real patient sample sets, we obtained DNA extracted from matched tumor and normal tissue of various tissue types preserved by both fresh frozen and FFPE methods. The FFPE DNA samples ranged in quality from DNA integrity number (DIN) 1.5 to 6.8 and fresh frozen DNA was DIN 8 to 9. We prepared libraries using this method compared against other library prep workflows and sequenced them by WGS and target capture. We assessed the performance of these workflows by library yield, library quality metrics, depth and evenness of coverage, and somatic variant calling with fresh frozen-extracted DNA providing the gold standard for library quality and mutation content. This new enzymatic fragmentation-based library prep workflow not only reduced the false positive rate in somatic variant detection by repairing damage-derived mutations in FFPE DNA samples but also improved the library yield, library quality metrics, library complexity, coverage depth and uniformity, and hybrid capture library quality metrics. Comparing the variant calls from matched FFPE and frozen tissues revealed an improved sensitivity and accuracy of variant calling using this library prep method compared to mechanical shearing and other enzymatic fragmentation library prep approaches. Furthermore, library quality and variant detection sensitivity were equivalent when using this FFPE workflow compared to gold standard mechanically sheared libraries. This new suite of enzyme mixes allows even highly damaged FFPE samples to achieve high quality libraries with a greater sensitivity for somatic variant identification. The workflow is robust and flexible, compatible with both FFPE DNA and matched high quality DNA samples as well as automation-friendly for convenience in sample processing. Citation Format: Chen Song, Margaret R. Heider, Jian Sun, Adrian Reich, Brittany S. Sexton, Bradley W. Langhorst, Laura Blum, Keerthana Krishnan. A novel DNA library prep method enables high sensitivity variant detection from matched fresh frozen and FFPE patient tissues [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1907.

Abstract 6299: Strainy: Multi-allelic phasing and assembly of bacterial strains and tumor clones using long reads

Abstract INTRODUCTION: Metagenomic sequencing of human microbiome and other complex microbial communities reveals extensive heterogeneity on sub-species levels. Detecting these heterogeneities (and strains) in the human microbiome is particularly important because different strains of the same bacterial species often display distinct phenotypes, such as pathogenicity in humans and antimicrobial resistance. For the context of cancer, studies show that the host microbiome may slow or contribute to the development and progression of cancer as well as the patient response to certain cancer therapies. However, reconstruction of heterogeneous bacterial species (represented by multiple strains) from a single metagenomic sample is challenging. Here we present an algorithm for phasing and assembly of closely related strains from long reads, which could also be applied to phased somatic variants in cancer. METHODS: The goal of the Strainy phasing procedure is to cluster long reads based on their strain (or haplotype) of origin. Given a set of all reads aligned to a unitig, Strainy builds a connection graph, where nodes correspond to aligned reads, and edges connect reads that share the same SNP genotypes. Then, Strainy iteratively uses a community detection algorithm to partition the connection graph into densely connected components that are assembled into strain unitigs. Finally, Strainy uses the overlap graph approach to extend strain unitigs and integrate them back to the original de novo assembly graph. RESULTS: We benchmark Strainy against Hifiasm-meta, Strainberry, metaFlye and metaMDBG using several simulated, mock and real datasets with ONT and PacBio HiFi reads.On simulated and mock metagenomic datasets, Strainy phased and reconstructed a substantially higher portion of unique strain sequence compared to the other tools, while having fewer errors. We then applied Strainy to untangle strains in metagenomic sequencing of activated sludge from an anaerobic digester bacterial community previously sequenced with Nanopre R9, R10 and PacBio HiFi. The number of recovered strains per bacterial species varied from 1 to 4. On average, 8.6 heterozygous structural variants per bacterial species were recovered. Non-synonymous to synonymous substitutions rates (dN/dS) revealed a few hotspots with evidence of selection that were specific to different species. Strainy represents a first practical algorithmic framework for multi-allelic phasing with long reads. Tumor genomes often contain multiple clones defined by characteristic somatic variants. Phasing of somatic variants into clonal haplotypes is therefore a promising approach to better characterize clonal heterogeneity in cancer. We tested Strainy for multi-allelic phasing on a mix of multiple human genomes and successfully phased distinct haplotypes. Motivated by the promising results, we are working on extending our approach to cancer clone phasing. Citation Format: Ataberk Donmez, Ekaterina Kazantseva, Maria Frolova, Mihai Pop, Mikhail Kolmogorov. Strainy: Multi-allelic phasing and assembly of bacterial strains and tumor clones using long reads [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 6299.

Abstract 5925: Circulating cell-free methylated tumor DNA measurements correlate with plasma VAF-based tumor fraction estimates

Abstract Introduction: Cell-free DNA (cfDNA) consists of circulating tumor DNA (ctDNA) and DNA shed by non-tumor cells such as immune and endothelial cells. Tumor fraction (TF) is the proportion of ctDNA present in cfDNA. TF is often estimated by summary statistics on somatic variants such as the mean or maximum variant allele fraction (VAF). The Northstar Response (NSR) assay quantifies ctDNA by accurately and precisely counting methylated cfDNA molecules across over 500 loci uniquely methylated in tumor tissue but not in any non-cancerous cells. We hypothesize that the NSR quantification of plasma methylated DNA, via the Tumor Methylation Score (TMS), correlates with tumor fraction as estimated by VAF. Methods: A pan-cancer cohort of 271 longitudinal blood samples across 74 patients was gathered and run on the Northstar Select (NSS), an 84-gene liquid biopsy panel for somatic variant detection, and NSR assays. Samples were required to have positive detection in both assays to ensure precise TF measurements (max(VAF)>0.01%, TMS>25 methylated molecules). Tumor fractions were estimated by removing germline variants and evaluating the max(VAF) and sum(VAF), with the latter measurement paralleling the summation of methylated molecules used by the TMS. Due to the high variability of tumor mutational burden (TMB) and tumor methylation levels between tumors, correlations between VAF and TMS were calculated on a per-patient basis for patients with at least 3 longitudinal samples. Results: TMS ranged from 25 to 253k molecules with a mean of 16k; sum(VAF) ranged from 0.06 to 782 with a mean of 67; max(VAF) ranged from 0.06% to 100% with a mean of 34%. TMS correlated most strongly with sum(VAF) with mean(R2)=0.77, median(R2)=0.96, whereas for max(VAF), mean(R2)=0.67, median(R2)=0.91. A linear regression model between max(VAF) and TMS demonstrated a mean increase of 1.3k methylated molecules per unit percent increase in max(VAF). Conclusions: Previous studies have demonstrated that VAF decreases are prognostic to targeted therapy response including immune checkpoint blockade and tyrosine kinase inhibitors. However, VAF-based tumor burden measurements are limited by the paucity of somatic variants. In contrast, aberrant DNA methylation is a well-known hallmark of cancer that is an abundant source of ctDNA. The strong, linear correlation between TMS and VAF shown here suggests that TMS is an effective measure of ctDNA burden and can similarly be used to monitor response to therapy. Furthermore, the large amount of methylated DNA in patient samples (mean=16k molecules) suggests that epigenetic approaches can be highly sensitive even at very low tumor fraction (<1%). Citation Format: Lee Davis McDaniel, David Tsao. Circulating cell-free methylated tumor DNA measurements correlate with plasma VAF-based tumor fraction estimates [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5925.

Abstract 4905: Mosaic chromosomal alterations overlapping hotspot and coldspot sites of somatic structural variation are associated with increased odds of hematological malignancy

Abstract BACKGROUND: Clonal hematopoiesis (CH) occurs when hematopoietic cells acquire somatic mutations and proliferate to generate clones in blood. CH can be driven by mosaic chromosomal alterations (mCAs), which are large somatic structural variants, and is associated with increased risk of blood malignancy. We conducted a genome-wide search for sites significantly more or less impacted by mCA events, termed hotspots and coldspots. We tested whether the odds of blood malignancy in participants harboring mCAs overlapping hotspots or coldspots differed from the overall association between mCAs and blood malignancy. METHODS: This study utilized two cohorts of the Canadian Partnership for Tomorrow’s Health, including the Ontario Health Study (OHS; n=7070) and CARTaGENE (n=28,639). Participants were genotyped with the UK Biobank Axiom or Infinium Global Screening Array and completed a baseline cancer questionnaire. All mCA calling was performed using MoChA software. We identified hotspot and coldspot sites of autosomal mCA accumulation using binomial tests scaled by chromosome and array type. We determined whether the number of overlapping mCAs at each query site was greater or less than would be expected under the null expectation. Then, we calculated the prevalence of cancer in participants with or without mCAs overlapping hotspots/coldspots and conducted Fisher’s exact tests to generate odds ratios. RESULTS: In OHS, participants with an mCA had a significantly greater odds of having a hematological malignancy at baseline (OR=7.29, 95% CI=3.72-13.38, p=7.84e-08). They were also more likely to have a cancer diagnosis of any type (OR=1.63, 95% CI=1.14-2.30, p=5.59e-03). Participants with an mCA overlapping a hotspot site had a further increased odds of harboring hematological malignancy (OR=9.79, 95% CI=3.94-21.29, p=5.05-e06). For participants with an mCA overlapping a coldspot site, the OR reached 20.52 (95% CI=3.68-76.85, p=7.36e-04). In CARTaGENE, participants with an mCA were also at increased odds of hematological malignancy (OR=11.31, 95% CI=6.78-18.17, p=9.77e-16), and any cancer diagnosis (OR=1.93, 95% CI=1.53-2.41, p=5.95e-08). Those carrying an mCA overlapping a hotspot site were 14.6 times more likely to have a hematological malignancy (95% CI=7.94-25.29, p=3.30e-13), and those with an mCA overlapping a coldspot site were 14.2 times more likely (95% CI=5.0-32.9, p=7.63e-06). CONCLUSIONS: These results suggest that there may be genomic locations at which somatic structural variation has a larger impact on the development of blood malignancy, relative to other regions. Further work is needed to characterize the functional consequences of somatic mutation at these key regions and explore whether these associations of hotspots, coldspots, and cancer incidence are conserved across a range of tissues. Citation Format: Jasmine Ryu Won Kang, Vanessa Bruat, Kimberly Skead, Mawusse Agbessi, June Kim, Elias Gbeha, Marie-Julie Fave, Philip Awadalla. Mosaic chromosomal alterations overlapping hotspot and coldspot sites of somatic structural variation are associated with increased odds of hematological malignancy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 4905.

Abstract 5050: Comparative analysis of somatic copy number variations and allelic imbalance in whole-genome bisulfite sequencing: challenges and insights

Goal: To assess the feasibility, reliability, and limitations of using Whole Genome Bisulfite Sequencing (WGBS) for detecting somatic copy number variation (SCNV) and tumor-specific allelic imbalance (AI). Background: In cancer genomics, the use of WGBS for SCNV and AI detection enables simultaneous analysis of genetic variations and methylation landscapes, potentially providing insights into the interplay between somatic alterations and DNA methylation in tumorigenesis. However, WGBS data is inherently noisier compared to whole-genome sequencing (WGS) or microarray data due to artifacts introduced during bisulfite conversion. These challenges, including difficulties in read alignment, genotype calling, and phasing, the lack of strand complementarity, and the need to distinguish true C>T SNPs from bisulfite-induced changes, collectively complicate SCNV and AI detection. Method: We analyzed WGBS data from The Cancer Genome Atlas (TCGA), including 39 tumor samples and 8 adjacent normal tissues across 9 tumor types. SCNV detection was performed using GATK-CNV, while AI callings were conducted with BisSNP, BISCUIT, Revelio (with Unified Genotyper), and BSsnp. Confusion matrices were used to compare GATK-CNV (WGBS) and hapLOH (microarray) at the base, gene, and arm levels. F1 score, specificity, and sensitivity were calculated to evaluate their performance across different genomic scales. Additionally, AI identified from WGBS data using BisSNP, BISCUIT, Revelio (with Unified Genotyper), and BSsnp were compared to those detected from array and WGS data using hapLOH/seq. To determine detection thresholds and refine results, sequencing data were downsampled, with filters applied to exclude low-quality data and retain high-confidence SNPs and genotypes. Result: GATK-CNV exhibited variable performance across samples, with most samples showing high true positives and low false positives, indicating good sensitivity for detecting SCNVs and capturing a genome burden similar to hapLOHseq data. However, for samples with a high number of events, normalization disrupts event callings, resulting in inconsistencies in SCNV identification compared to array data. For AI detection, Revelio (with Unified Genotyper) showed the best overall performance among the tools evaluated. Downsampling analysis showed that reducing sequencing coverage impaired event detection, becoming less effective at lower coverage levels. BSsnp, BISCUIT, and BisSNP produced widespread and inconsistent variant calls, and initial filtering for high-quality SNPs showed limited improvement in precision. This study highlights the potential and challenges of WGBS as a tool for analyzing SCNV/AI alongside DNA methylation with phased data, offering insights into tumorigenesis and advancements in cancer diagnostics and treatment. Citation Format: Yiru Gu, Jian Gu, Paul Scheet, Justin Wong. Comparative analysis of somatic copy number variations and allelic imbalance in whole-genome bisulfite sequencing: challenges and insights [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5050.

Abstract 6639: Optimized selection of tumor-informed MRD panels enhances sensitivity of ctDNA detection

Abstract Circulating tumor DNA (ctDNA) is a highly sensitive and specific biomarker for molecular residual disease (MRD) and can be by identified by the presence of tumor-specific somatic variants. We have developed a second-generation hybrid-capture-based tumor-informed MRD assay to detect the presence and quantity of ctDNA in plasma-derived cell-free DNA (cfDNA). Our assay surveys the whole tumor genome to identify the superset of high-confidence somatic variant targets that can be used to distinguish ctDNA from non-tumor cfDNA. Each panel consists of up to 1000 targets that are selected based on target-specific and tumor-specific features that impact the representation of tumor variants or the effective molecular depth of a panel, such as somatic call confidence, zygosity, copy number, background error rate and expected molecular depth following capture and sequencing. cfDNA captured using a patient-matched panel undergoes targeted sequencing and that sequence data is fit to a statistical model that evaluates the presence and fraction of ctDNA in the sample. Here, we evaluated the performance of our target-ranking algorithm using contrived mixtures of matched tumor and normal cell lines, with tumor DNA concentrations ranging from 5 to 10, 000 parts per million (ppm) and samples processed across multiple days, reagent lots, operators and instruments. Identical sets of contrived samples were analyzed using subsets of panel targets, including top-ranking and bottom-ranking target sets. We found the Limit of Detection with 95% sensitivity (LoD95) for top-ranking targets was 40% of the LoD95 for low-ranking targets, demonstrating the effectiveness of our ranking algorithm in stratifying high and low sensitivity targets. In conclusion, the constituents of tumor-informed MRD panels have a measurable impact on the sensitivity of ctDNA detection and, with optimization, can enhance the resolution of residual disease detection. Citation Format: Ashley Acevedo, Ravi Patel, Kyle Trettin, Nafei Xu, Genevieve Gould, Dale Muzzey. Optimized selection of tumor-informed MRD panels enhances sensitivity of ctDNA detection [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 6639.

Abstract 1977: Implementation of a comprehensive clinical cfDNA analysis platform for aqueous humor liquid biopsy in retinoblastoma

Introduction: The development of liquid biopsy testing based on sequencing analysis of cell-free DNA (cfDNA) has the potential to transform cancer care in pediatric tumors, which are characterized by a wide array of genomic alterations including copy number alterations (CNAs), sequence variants, epigenetic alterations, and structural rearrangements in tumor-associated genes. In patients with retinoblastoma where direct tumor biopsy is not possible due to the risk of tumor spread, aqueous humor (AH) is a rich source of cfDNA. Comprehensive cfDNA testing, combining a low pass whole genome sequencing (LP-WGS) assay to detect CNAs and a custom targeted sequencing panel (TSP) to detect mutations, can provide a highly effective platform to inform diagnosis, risk stratification, response to therapy, and surveillance in patients with retinoblastoma. LP-WGS Results: We validated a liquid biopsy LP-WGS assay, LBSeq4Kids, to detect CNAs in cfDNA from plasma, cerebrospinal fluid, or AH. Clinical testing has been performed for the past two years for diagnostic evaluation and monitoring response to therapy using AH from retinoblastoma patients. Thirty-six children with unilateral or bilateral retinoblastoma (44 eyes) underwent LBSeq4Kids LP-WGS analysis. CNAs were detected in 30/32 (94%) samples at diagnosis or recurrence. Serial sampling of 22/44 (50%) eyes was performed using a range of 1-10 samples per patient. Seventeen of those 22 (77%) eyes demonstrated clearance of CNAs with treatment. Five eyes showed persistence of abnormal profiles, necessitating enucleation of 2 eyes. Treatment and evaluation are ongoing for the remaining three eyes. Targeted Sequencing Panel Results: We are validating a custom cfDNA TSP to detect single nucleotide variants and small indels in the coding sequence of 136 genes and fusions involving EWSR1, FOXO1, and BRAF for patients with ocular disease, solid tumors and brain tumors. To date, 17 children (21 eyes) diagnosed with retinoblastoma have had TSP testing using AH. All eleven (100%) patients with germline variants demonstrated the same RB1 alterations in the AH using the panel. Six of these patients (55%) had a second somatic alteration in RB1 detected with the panel. Two of 11 (18%) with known germline RB1 mutations also demonstrated mutations in BCOR, ARID1A, or MSH6. Somatic variants in RB1, ARID1A and BCOR were detected in the remaining six patients with sporadic retinoblastoma. Targeted sequencing of the AH surveillance samples from these patients is in progress. Conclusion: Here we demonstrate clinical utility for aqueous humor liquid biopsy molecular testing as a complementary means of diagnosis and monitoring treatment response in patients with retinoblastoma. Continued surveillance of these patients is ongoing to monitor disease status and rule out recurrence, as well as identify novel prognostic biomarkers. Citation Format: Laura A. Kagami, Eirini Christodoulou, Venkata Yellapantula, Dong Xu, Cindy Fong, Dejerianne Ostrow, Liya Xu, Jesse Berry, Jaclyn A. Biegel. Implementation of a comprehensive clinical cfDNA analysis platform for aqueous humor liquid biopsy in retinoblastoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1977.

Abstract 5802: Somatic variant predictors of cutaneous immune-related adverse events in cancer patients treated with immunotherapy

Cutaneous immune-related adverse events (cirAEs) affect nearly one-third of patients treated with immune checkpoint inhibitors (ICIs). Identifying patients at high risk for cirAE can help guide monitoring and therapeutic management. Whether tumor somatic mutations are associated with the risk of cirAE development remains unknown. We analyzed tumor somatic single nucleotide variants (SNVs) from a discovery cohort of 733 patients with stage III/IV melanoma receiving ICIs at Dana Farber Cancer Institute and Massachusetts General Hospital. CirAE development was determined by chart review. SNVs were detected by OncoPanel v1-3. Nonsynonymous SNVs were used to calculate tumor mutational burden (TMB) and determine gene mutation status. Logistic regression models were used to test the association between gene mutation status and the binary outcome cirAE, while Cox regression models were used to examine the association between TMB or gene mutation status with time to cirAE development. Models were adjusted for demographic covariates (age at ICI, sex, self-reported race/ethnicity) and clinical covariates (melanoma subtype and ICI type). To validate significant associations identified in the discovery cohort, we used an independent cohort of 792 non-melanoma cancer patients treated with ICIs. Consistent with prior studies, TMB-high (>10 mutations/Mb) was significantly associated with improved overall survival (hazard ratio (HR) [95% CI] = 0.72 [0.53, 0.98]) in melanoma. However, TMB-high was not associated with cirAE development (HR = 0.92 [0.69, 1.22]). Binary logistic regression revealed that somatic mutations in PRKCI (OR = 2.76 [1.27, 6.33]), STAG2 (OR = 2.25 [1.09, 4.80]), and CXCR4 (OR = 2.56 [1.02, 6.80]) were associated with increased odds of cirAE. Furthermore, time-to-event Cox regression identified that somatic mutations in PRKCI, STAG2, and CXCR4 as well as five additional genes (MEN1, AKT3, FANCG, DAXX, H3F3B) were significantly associated with increased risk of cirAE development (nominal p-values<.05). Mutations in CXCR4 replicated in the validation cohort as significantly associated with time to cirAE (HR = 4.02 [1.22, 13.21]) adjusting for cancer type, ICI type, and demographics. Of the CXCR4 mutations identified in patients with cirAE, 5/15 (33%) localized to the C-terminus, were predicted deleterious by SIFT and PolyPhen2, and altered key residues known to be involved in CXCR4 degradation, suggesting a mechanistic link between CXCR4 gain-of-function mutations and increased risk of cirAE. We identified somatic mutations in several genes as significant predictors of cirAE development in melanoma. The association between mutations in CXCR4 and cirAE was replicated in a pan-cancer validation cohort, providing mechanistic insight into cirAE development across cancers. Citation Format: Ninghui Hao, Jenny Lai, Crystal Chang, Ahmad Rajeh, Chuck Lin, Cameron Moseley, Arjun Mahajan, Matthew Tran, Coleen Valery, Christopher J. Thang, Guihong Wan, Alexander Gusev, Yevgeniy R. Semenov. Somatic variant predictors of cutaneous immune-related adverse events in cancer patients treated with immunotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5802.

Abstract 5895: Development and analytical validation of a highly sensitive tumor fraction estimation method using blood samples without matched normal controls

Abstract Background: Tumor fraction (TF) in cell-free DNA (cfDNA) is a quantitative indicator of circulating tumor DNA (ctDNA) and has become an important biomarker for disease monitoring and tracking therapeutic responses. We developed a longitudinal TF estimation method that does not require matched normal control samples and analytically validated its ability to accurately estimate TF when longitudinal blood samples are available. Methods: Both reference materials and clinical blood samples were used in the analytical validation. We first determined the limit of detection (LoD) of TF for our PredicineCARE cfDNA assay at targeted sequencing depth of 20, 000x using serially diluted reference materials with 12 gradient values from 0 to 4% TF. We further used a sequentially diluted clinical sample (with known TF) with healthy donor cfDNA to a gradient of expected TF values. We evaluated the variant calling and TF calculation algorithm by treating each dilution separately or treating the immediately nearby samples as longitudinal samples. We further evaluated our longitudinal-aware analysis method on 20 clinical subjects tested with PredicineCARE cfDNA assay with ≥2 time points to evaluate the improvement in TF estimation compared to individual estimation. Results: With the titrated reference materials and manufacturer-labelled somatic variants, we achieved an LoD of 0.01% TF for 30ng cfDNA input and a sequencing depth at 20, 000x using PredicineCARE cfDNA assay. When approaching the LoD, the number of reads for the somatic variants becomes more stochastic, which resulted in big variances in the calculated TF values. When independently estimating the TF without referring to matched normal or longitudinal samples, the LoD was 0.5% TF. When the longitudinal sample information was included, the LoD went down to 0.01% TF, contingent on the number of trackable variants. Our simulated results showed that when there are 4 or more tracking variants between longitudinal samples, the LoD can go down to 0.01% TF most of the time. We further applied this longitudinal sample aware dynamic TF estimation method to 20 clinical subjects tested with PredicineCARE cfDNA assay and with ≥2 time points, most showed improved TF estimation with ≥50% of the subjects having questionable somatic/germline calls if called separately. Conclusions: We developed a dynamic TF estimation methodology that leverages longitudinal samples to improve somatic/germline variant calling and exclude CHIP mutations. Our results demonstrated that with longitudinal samples, the accuracy in TF estimation can approach the ideal situation where all somatic/germline status of the variants are known. Notably, longer interval between time points and a higher number of tracking somatic mutations (≥2) in longitudinal samples generally enhance the effectiveness of this dynamic estimation methodology. Citation Format: Junmei Wang, Wanshu He, Lisha Zhu, Yong Huang, Xiaohong Wang, Shidong Jia, Pan Du. Development and analytical validation of a highly sensitive tumor fraction estimation method using blood samples without matched normal controls [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5895.

Abstract 3305: Notch mutations in subjects with recurrent or metastatic adenoid cystic carcinoma treated with rivoceranib from study RM-202

Abstract Background: Activating NOTCH mutations are present in approximately 20% of metastatic adenoid cystic carcinomas (ACC) leading to an aggressive tumor phenotype and poor clinical prognosis, yet less common pathogenic NOTCH mutations are poorly characterized. RM-202 was an international, single-arm, open-label phase 2 trial of oral VEGFR-2 tyrosine kinase inhibitor, rivoceranib, in patients with histologically confirmed recurrent or metastatic (R/M) ACC (Hanna et al. 2023 Clin Cancer Res 29(22): 4555). Herein, we report the mutational analysis NOTCH genes 1-4 (NOTCH1-4) in a subset of ACC samples from RM-202. Methods: Patients (n=72) enrolled in RM-202 received oral rivoceranib 700 mg once daily for a 28-day cycle. The co-primary endpoints were objective response rate (ORR) assessed by both investigator and a blinded independent review committee (BIRC). RNA seq was performed on ACC tumor samples from 27 patients. Complete coding sequences were analyzed with NOTCH1-4 variants identified and cross-referenced with Tempus and public databases to categorize mutations according to pathogenicity. Results: The investigator and BIRC assessed objective response rates were 15.3% (95%CI, 7.9-25.7) and 9.7% (95% CI, 4.0-19.0), respectively. The median duration of response was 14.9 months (95% CI, 4.9-17.3) and median progression free survival was 9.0 months (95% CI, 7.3-11.5). Grade 3 treatment related adverse events occurred in 70% of patients, with hypertension (42.5%) and stomatitis (7.5%) being the most common. A total of 477 variants were identified in NOTCH1-4 from 27 RM-202 ACC tumor samples, of which 463 (97%) mutants had been previously reported by Tempus DNA sequencing of germline and somatic variants across multiple tumor types. Three pathogenic NOTCH1 mutations (two nonsense mutations and 1 in-frame deletion) were identified across 2 tumor samples. Two of these mutations had not been previously reported in ACC. Forty-five additional non-synonymous variants were identified in NOTCH1-4. Of those, NOTCH4 L16 mutations occurred in 17/27 (63%) of tumors, with 12/27 (44%) tumors harboring L16 deletions, 4/27 (15%) of tumors with L16 duplications, and 1/27 (4%) that had both an L16 deletion and an L16 duplication. The frequency of L16 mutant sequences in tumors ranged from 8%-67% suggesting subclonal variation and potential pathogenicity. Conclusions: Previously unreported pathogenic NOTCH1-4 mutations were identified in ACC tumor samples from the RM-202 phase II trial. High rates of NOTCH4 L16 mutations suggest a novel mechanism of Notch pathway activation in ACC and represents a potential therapeutic target in this rare disease. Citation Format: Ari Rosenberg, James Bugni, Lissa Nazal, Chris Galloway, Kristin Ryan. Notch mutations in subjects with recurrent or metastatic adenoid cystic carcinoma treated with rivoceranib from study RM-202 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3305.

Abstract 7171: Comparison of clinical next generation sequencing and whole genome sequencing for acute myeloid leukemia characterization and classification

Abstract Acute myeloid leukemia (AML) is a hematopoietic neoplasm characterized by uncontrolled proliferation of myeloid blast cells. AML is the most common acute leukemia, with over 20, 000 new U.S. cases diagnosed annually. The 5-year survival rate varies greatly by age, from 63% in younger patients to 11% in older patients. AML is a genetically heterogeneous disease, with over 25 genes implicated in classification. Clinical next-generation sequencing (NGS) is crucial for AML management, aiding in relapse risk assessment, therapy guidance, and understanding of the genetic landscape. In AML, this typically involves sequencing a panel of 50-500 genes. In practice, only diagnostic samples are tested, without germline controls. In research settings, whole genome sequencing (WGS) is performed on paired diagnosis and remission samples, with remission serving as a germline control. This study assessed the accuracy and utility of AML clinical NGS reports by comparing with matched WGS data. We explored the benefits of germline controls and how discrepancies impact risk classification. 65 AML patients treated in the Northwell Health System and enrolled in an IRB-approved biospecimen study were analyzed. Clinical NGS testing was performed using FoundationOne Heme (n=51) or GenPath OnkoSight (n=14). We analyzed mutation data from clinical NGS reports (median 12 variants per report; range 1-26), identifying DNMT3A, FLT3, TP53, and NPM1 as the most frequently mutated genes, consistent with prior studies. Matched WGS data from 15 patients classified clinically reported variants as somatic (27%; n=57), germline (65%; n=135), or not observed (8%; n=17), using variant allele frequencies (VAFs). There was a strong but imperfect correlation between somatic/germline labels and known/unknown status, provided in the clinical NGS reports: 89% of germline variants were of unknown significance (n=120), while 83% of somatic variants were of known significance (n=43). Clinical NGS reports lacked data such as population variant frequency, pathogenicity scores, and VAFs. Germline and unknown significance variants had higher population frequencies and lower pathogenicity scores than somatic and known significance variants (p<0.0001). European LeukemiaNet (ELN) risk groups assigned using clinical NGS data (Adverse: 51%, n=33; Favorable: 29%, n=19) were affected by the inclusion of germline or unobserved variants; excluding those reclassified two patients from Adverse to Intermediate risk. This study finds general agreement between AML clinical NGS and WGS in the overlapping genes. The clinical NGS reports studied here lack annotation of germline and subclonal variant status. We highlight the limitations of clinical NGS without germline controls and its implications for AML classification, clinical decision making, and broader scientific understanding. Citation Format: Annette A. Prah, Derek Van Delden, Joan Alexander, Asya Stepansky, Erin Boyle, Nicholas Chiorazzi, Jonathan Kolitz, Stephanie Boisclair, Steven L. Allen, David Chitty, Andrea B. Moffitt. Comparison of clinical next generation sequencing and whole genome sequencing for acute myeloid leukemia characterization and classification [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 7171.

Abstract 7471: Mutational profiles of pancreatic cancer patients from Eastern Asia using somatic exome sequencing

Abstract Introduction: Pancreatic adenocarcinoma (PC) is one of the most lethal diseases. The incidence of PC is rising in the Asia-Pacific population, with some of the highest rates in Japan. As such, the utilization of molecular testing has increased, especially for advanced disease, to identify molecular drivers and therapeutic targets. Methods: 146 patients from Japan (ages 45-81) who were diagnosed with PC underwent exome-based, tumor molecular profiling (TMP) (Labcorp Genetics Inc). 96/146 (65.8%) patients had unresectable tumors. Results were correlated with clinical information. Variants were interpreted according to the AMP/ASCO/CAP guidelines. Signaling pathway analysis was conducted using the Reactome database, with significance determined by binomial test with corrected p-values (FDR; Benjamini-Hochberg procedure). Fisher’s exact test was utilized for statistical assessment of clinical parameters. Results: TMP detected 404 Tier 1/2 somatic variants in 135 (92.5%) patients. Genes known to be frequently mutated in PC were identified in the cohort: KRAS (122/146; 83.6%), TP53 (108/146; 74.0%), SMAD4 (35/146; 24.0%) and CDKN2A (25/146; 17.1%). A potential enrichment of hotspot U2AF1 (5/146; 3.4%) codon 34 mutations was identified, compared to published frequencies in PC in cancer databases (1%) that trended toward statistical significance (p=0.07). Pathway analysis identified a statistically significant enrichment of mutations in several kinase signaling pathways including PI3K/AKT (p-value: 2.1e-11; FDR: 9.8e-09), ERBB2 (p-value: 5.3e-07; FDR: 1.2e-04), and RAS/MAPK (p-value: 4.2e-06; FDR: 5.5e-04). Mutations associated with homologous recombination or mismatch repair pathways were statistically associated with early age of onset (<50 years of age; p= 0.036). Hotspot KRAS mutations were associated with unresectability (p=0.002), late tumor stage (stages 3-4, p=0.026), and metastasis (p=0.010). Conclusion: Alterations enriched in targetable pathways were identified in the patient cohort in addition to statistically significant findings associated with more aggressive PC. TMP testing can aid in the identification of alterations that are pertinent to the identification of relevant therapies and clinical trials as well as identify PC patients with potentially more aggressive disease. Citation Format: Jacquelyn Reuther, Taro Shibuki, Megan Hawley, Hideaki Bando, Keir Violaceae, Makoto Ueno, Kenji Ikezawa, Takuji Okusaka, Teiichi Sugiura, Keiko Kamei, Yoshito Komatsu, Michiaki Unno, Hiroki Yukami, Takuji Iwashita, Emily Westheimer, Kumiko Umemoto, Robert Daber, Takayuki Yoshino, Jianhua Zhao, Masafumi Ikeda. Mutational profiles of pancreatic cancer patients from Eastern Asia using somatic exome sequencing [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 7471.

Abstract 3901: Charting copy number gain histories from mutational divergence among multi-samples

Abstract In cancer, a somatic copy number gain (SCNG) on a tumor chromosomal region is often assumed to result from a series of minimum number of copy number events, but this parsimonious-trajectory assumption may be violated in reality. For example, an SCNG could develop through a complex trajectory involving a combination of gain and loss events, rather than exclusively through gains. While prior efforts have been made to infer the SCNG trajectories from patient data, the presence of complex trajectories remains to be validated. Furthermore, these methods are limited by either only utilizing SCNG profiles, or inferring trajectories from only a single sample. Here, we hypothesize that SCNG trajectories can be reflected by somatic single nucleotide variants (SSNVs) shared between samples from the same patient. By analyzing multi-regional whole genome sequencing data of tumor specimens, we found evidence where the multiplicities of shared SSNVs between samples pinpoint complex SCNG trajectories. Here, the SSNV multiplicity refers to the number of alleles harboring the corresponding SSNV in the dominating clone of a tumor sample. We also define the joint multiplicity distribution as the proportion of shared SSNVs that occur at different allelic states (multiplicities) across distinct tumor samples. We developed a probabilistic computational pipeline to automate the identification of non-parsimonious multi-sample SCNG trajectories utilizing shared SSNV multiplicities. Extending an existing method GRITIC from single sample to multi-sample setting, this pipeline enumerates all trajectories leading to SCNGs states in multiple samples, and characterizes the probability of observing the joint multiplicity distribution in the real patient data given each possible trajectory. We applied the method on prostate cancer and breast cancer datasets, and identified conclusive cases of complex trajectories on genomic regions affected by SCNGs. On regions showing different copy number states in multiple samples, we identified cases experiencing both gains and losses. On regions showing the same copy number states in multiple samples, we also identified a few of them experiencing independent acquisition of SCNGs in different samples. Our findings call for a stochastic modeling of SCNGs dynamics during tumor evolution. Citation Format: Yunong Xia, Ruping Sun. Charting copy number gain histories from mutational divergence among multi-samples [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3901.

Abstract 3892: Multi-sample whole genome sequencing unveils complex mutational dynamics and clonal evolutionary patterns in young onset breast cancer from Nigeria

Abstract Background: Breast cancer (BC) is a heterogenous disease, and the rising global epidemic of premenopausal BC remains poorly understood. Mutational landscape and evolutionary dynamics of BC from diverse geography and populations are largely unknown, preventing the global acceleration of progress towards precision prevention. Methods: To examine etiology and heterogeneity of BC in indigenous Africans, 43 multi-region tumor samples and blood samples from 18 Nigerian women with BC (mean age 53 +/-12.7) were analyzed. Whole Genome Sequencing (WGS) was performed to identify somatic single nucleotide variants, insertions and deletions (ID), copy number alterations (CN), and structural variants (SV). Five mutational signature types (single base substitutions (SBS), double base substitutions (DBS), ID, CN, and SV), were analyzed and compared to COSMIC human cancer signatures. Multi-DPclust was used to classify mutations to clonal and subclonal cluster and create mutational phylogenetic trees. Results: Driver gene analysis highlighted key driver genes, such as TP53, GATA3, and PIK3CA, corroborating prior findings. The most common signature was clock-like (CL) SBS5, followed by APOBEC-related SBS2 and SBS13. Other notable signatures included hypoxia-linked SBS18 and HRD-related DBS13. Signature profiles showed moderate heterogeneity across samples, with distinct patterns observed across mutational clusters. Intra-cluster correlation coefficients for signatures within samples from same patient range from 0.37 to 0.99 (median 0.84). TP53 and GATA3 mutations are common in clonal (early) clusters, with GATA3 often appearing in consecutive subclones. The CL SBS5 signature was prevalent early, while SBS18, linked to hypoxia, became prominent late, highlighting evolving mutational processes across disease progression. Distinct BC subtypes also displayed unique mutational profiles. HR+/HER2+ tumors exhibited higher levels of signature SBS91, while HR-/HER2- tumors exhibited higher levels of the HRD signature SBS3 compared to the rest of the tumors. HR-/HER2- tumors demonstrated lower levels of the mismatch repair-related signature ID1 but higher levels of the HRD signature ID6 and TOP2A signature ID8. Significant differences were observed in SV signatures, with higher SV9 signature activities in HR+/HER2+ tumors and elevated BRCA-related SV3 signature activities in HR-/HER2- tumors. Heterogeneity was also evident in the DBS and CN mutational signatures, further highlighting tumor complexity. Conclusion: This analysis sheds light on the diverse mutational dynamics within and across molecular subtypes, providing insights into the mutational evolution of BC in a non-screen detected young onset population. Ongoing work integrating WGS and transcriptome data will be presented at the conference. Citation Format: Avraam Tapinos, Toshio Yoshimatsu, Ilona Siljander, Mustapha A. Ajan, Ayodele Sanni, Atara Ntekim, Abayomi Odetunde, Elisabeth Sveen, Jeffrey Mueller, Galina Khramtsova, Sulin Wu, Dorothy Nyamai, Mihai Giurcanu, Dezheng Huo, Yonglan Zheng, David C. Wedge, Olufunmilayo I. Olopade. Multi-sample whole genome sequencing unveils complex mutational dynamics and clonal evolutionary patterns in young onset breast cancer from Nigeria [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3892.

Abstract 2780: Topoisomerase IIb binding underlies frequently mutated elements in cancer genomes

Abstract Type-II topoisomerases resolve topological stress in DNA through controlled double-strand breaks. While TOP2A is a chemotherapy target in proliferating cells, the ubiquitously expressed TOP2B is a potential off-target. Here we explore roles of TOP2B in mutagenesis by generating DNA-binding maps of TOP2B, CTCF, and RAD21 in human cancer samples and analyzing these maps for driver mutations and mutational processes in 6500 whole cancer genomes. TOP2B-CTCF-RAD21 and TOP2B-RAD21 sites are enriched in somatic mutations and structural variants (SVs), especially at evolutionary conserved sites displaying high transcription and long-range chromatin interactions. TOP2B binding underlies SVs and hotspot mutations in cancer-driving genes such as TP53, MYC, FOXA1, and VHL, and many cis-regulatory elements. We show that the TOP2B-bound mutational hotspot at RMRP drives tumor initiation and growth in vivo. These data highlight TOP2B as a protector of the genome from topological challenges whose aberrant activity promotes driver and passenger mutations in cancer genomes. Citation Format: Jüri Reimand, 1 Christian A. Lee, 1 Robin H. Oh, 2 Zoe P. Klein, 1 Nina Adler, 1 Sana Akhtar Alvi, 3 Ellen Langille, 2 Elisa Pasini, 4 Kevin C. Cheng, 1 Diala Abd-Rabbo, 1 Huayun Hou, 3 Ricky Tsai, 2 Mamatha Bhat, 1 Daniel Schramek, 2 Michael D. Wilson, 3 Liis Uusküla-Reimand3. Topoisomerase IIb binding underlies frequently mutated elements in cancer genomes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2780.

Abstract 3899: Genetic insights into clonal evolution from normal prostate epithelium to cancer via HGPIN

Backgrounds: Prostate cancer (PC) is one of the most prevalent cancers in the developed world, where its incidence is still increasing. Most PCs are thought to arise from high-grade prostatic intraepithelial neoplasia (HGPIN), a well-recognized precursor lesion. However, the genetic landscape of HGPIN and its relationship with normal prostate epithelium (NPE) and PC remains to be fully explored. Methods: To address these issues, we collected prostate tissue samples from five patients who underwent radical prostatectomy. Prostate tissue sections (3 mm thick) were fixed with an alcohol-based solution and prepared for dual immunostaining using AMACR/p63, in addition to hematoxylin-eosin staining, and subjected to laser-capture microdissection (LMD). Then, PC (n = 35), HGPIN (n = 71), and normal epithelium (n =244) samples were dissected and subjected to whole-exome sequencing (WES). Variant pathogenicity was assessed based on COSMIC and ClinVar databases. Furthermore, to investigate the mutation acquisition rate in the prostate epithelium, single cell-derived prostatic organoids established from three patients were analyzed using WES. Somatic single nucleotide variants (SNVs), indels, and copy number variations (CNVs) were identified. Results: A total of 10,047 somatic SNVs and indels were detected. The mutation rate was estimated to be 0.28 mutations/year/exon. Shared mutations were observed between NPE and HGPIN as well as between HGPIN and PC in some cases, suggesting a common clonal origin. Pathogenic FOXA1 mutations were frequently identified in both HGPIN and PC. While pathogenic FOXA1 mutations were also observed in NPE, none of these mutations were shared with HGPIN or PC, indicating that they likely represent independent mutational events. In one case, spatially separated multifocal cancers exhibited distinct FOXA1 mutations, each of which was shared with the adjacent HGPIN. In other cases, mutations were also shared between PCa, HGPIN, and NPE samples, although no known driver mutations were detected therein. These observations suggest that these cancers were derived from their precancer lesions with apparently normal histology, which progressed through HGPIN and PC. Phylogenetic analysis of these samples revealed that HGPIN and PCa diverged recently from a common ancestor. Conclusions: We successfully characterized somatic mutations and CNVs in apparently normal prostate epithelium, which were shared between HGPIN and adjacent PC in some cases. These findings provide new insights into the early genetic events driving PC development. Further studies are needed to fully elucidate the clonal evolution from NPE to PC, which may offer novel targets for early detection and intervention. Citation Format: Kohsuke Hishiki, Nobuyuki Kakiuchi, Yuki Teramoto, Koichi Watanabe, Kosuke Ieiri, Hirona Maeda, Tomonori Hirano, Yuki Kita, Takashi Kobayashi, Seishi Ogawa. Genetic insights into clonal evolution from normal prostate epithelium to cancer via HGPIN [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3899.

Abstract 2103: Pharmacogenomic biomarkers of oncology drugs approved by the US FDA classified on the basis of BEST (Biomarkers, EndpointS, and other Tools) resource: systematic review

Abstract Objective: To characterize pharmacogenomic biomarkers of oncology indications approved by the US FDA classified on the basis of BEST (Biomarkers, EndpointS, and other Tools). Resource Design: Systematic review. Data sources: FDA oncology/hematologic malignancies approvals and databases, BEST (Biomarkers, EndpointS, and other Tools) Resource, and CDER / CBER’s DDT Qualification Project Search database: Stage 1: Letter of Intent (LOI), Stage 2: Qualification Plan (QP), and Stage 3: Full Qualification Package (FQP). Study Inclusion: All Biomarkers include germline or somatic gene variants (polymorphisms, mutations), functional deficiencies with a genetic etiology, gene expression differences, chromosomal abnormalities; and selected protein biomarkers in all approved labeling sections published by the FDA. Results: BEST Resource defines seven biomarker categories: susceptibility/risk, diagnostic, monitoring, prognostic, predictive, pharmacodynamic/response, and safety. The biomarkers are more linked to three cancers: breast, lung and colorectal cancers. Citation Format: Tariq H. Alqurayshah, SFDA Oncology and Hematology Group. Pharmacogenomic biomarkers of oncology drugs approved by the US FDA classified on the basis of BEST (Biomarkers, EndpointS, and other Tools) resource: systematic review [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2103.

Abstract 5890: Personalized detection of circulating cell-free tumor DNA in gynecologic cancer

Abstract Endometrial and ovarian cancers are among the most common gynecologic malignancies in the U.S., with 5-year overall survival rates of 81% and 50% respectively. Ovarian cancer often presents late due to asymptomatic early stages and nonspecific symptoms in advanced stages. Few markers exist for early detection, and they lack the sensitivity and specificity needed for high-risk screening. Given the absence of informative protein blood biomarkers, cell-free DNA analysis has become a promising tool for early detection, treatment evaluation, and monitoring. However, previous studies of circulating tumor DNA (ctDNA) in gynecological cancers had limited sensitivity. We adapted MASQ (Multiplex Accurate Sensitive Quantitation), originally developed to measure residual cells in leukemia, to the measurement of ctDNA in solid cancers. MASQ detects patient-specific mutations simultaneously in up to 50 loci with exceptional sensitivity, below one part per million. This study uses MASQ to measure tumor-specific variants in blood (cell-free DNA) from endometrial and ovarian cancer patients, aiming to determine the range of ctDNA signals at presentation and explore correlations with clinicopathological features and patient outcomes. We analyzed samples from 46 gynecologic cancer patients (12 ovarian, 34 endometrial) treated at Northwell Health. Samples were collected in collaboration with gynecologic oncology research coordinators and the Northwell Health Biorepository. Peripheral blood was collected at diagnosis (n=46) and post-surgery (n=14), along with surgical tumor specimens. Whole genome sequencing of tumor and blood identified thousands of somatic tumor variants per patient. For each patient, 30 representative variants were selected for MASQ sequencing. MASQ libraries were generated from cell-free DNA isolated from pre- and post-surgery plasma. Ultrasensitive variant counts across 30 loci were aggregated and analyzed for clinical correlations. ctDNA signal was detected in 35 of 46 cases at presentation, with variant allele frequencies ranging from below 10^-4 to 10^-1. ctDNA signal was higher in ovarian cancer (13/13 detected) than endometrial cancer, and higher in aggressive Type II endometrial cancer (15/16) vs. Type I (5/13). ctDNA levels correlated with tumor size and copy number imbalance, but not with total cell-free DNA. Detection of ctDNA predicted worse overall survival (p=0.023) and was the strongest predictor of survival in multivariate analysis. Post-surgery follow-up in 10 of 14 patients showed significant ctDNA reduction following treatment, highlighting MASQ’s sensitivity in tracking treatment response. This study highlights the potential of sensitive ctDNA detection for early diagnosis and monitoring in gynecologic cancers. However, limited cell-free DNA abundance necessitates even more sensitive methods to enhance clinical utility. Citation Format: Andrea B. Moffitt, Jude Kendall, Joan Alexander, Asya Stepansky, Arisa Kapedani, Zihua Wang, Dan Levy, Kenny Ye, Abba M. Krieger, Marina Frimer, Gary L. Goldberg, Michael Wigler. Personalized detection of circulating cell-free tumor DNA in gynecologic cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5890.

Abstract 1283: Targeted genetic profiling of a de novo neuroendocrine patient-derived xenograft (PDX) and organoid model of prostate cancer from patient with African Ancestry

Abstract Patient-derived xenograft (PDX) and organoid (PDO) models provide platforms to study prostate cancer (PCa) progression and therapy, especially in underrepresented populations. Here we report the first de novo neuroendocrine prostate cancer (NEPC) from a consented patient of African Ancestry with localized disease and no prior treatment before radical prostatectomy. African American patient had a pTN2 tumor, Gleason7 (4+3) and PSA or 3.6. Acquired tissue was collected for PDX and PDO establishment as well as for downstream molecular characterization. To uncover genetic drivers of aggressive phenotypes, we performed whole exome sequencing on DNA from the primary tumor and the established PDX using the Illumina Exome Kit and Twist Biosciences Exome Panel 2.0, followed by bioinformatic analysis. Computational workflows using Nextflow, including nf-core/sarek, identified somatic single nucleotide polymorphisms (SNPs), mutations (Mutect2, Strelka), and copy number variations (CNVs) (CNVkit, Control-FREEC). Variant annotation with Variant Effect Predictor and SnpEff linked variants to cancer progression. Analysis of the tumor samples revealed CNAs and mutations in genes similar to those seen in previously published NEPC. Protein expression by western blot and IHC identified altered protein expression peculiar to the NEPC phenotype further validating our primary tissue and its derived PDX models to have the NEPC phenotype. Furthermore, novel markers identified provide insight into molecular drivers of this prostate cancer phenotype and may inform targeted therapies for high-risk localized PCa. Citation Format: Diana Rodriguez, Surendra Gulla, Tej Sharma, John Bodkin, Roberto Pili, John E. Tomaszewski, Remi M. Adelaiye-Ogala. Targeted genetic profiling of a de novo neuroendocrine patient-derived xenograft (PDX) and organoid model of prostate cancer from patient with African Ancestry [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1283.