PureCN: copy number calling and SNV classification using targeted short read sequencing

Neoantigen-targeted dendritic cell vaccination in lung cancer patients induces long-lived T cells exhibiting the full differentiation spectrum

Non-small cell lung cancer (NSCLC) is known for high relapse rates despite resection in early stages. Here, we present the results of a phase I clinical trial in which a dendritic cell (DC) vaccine targeting patient-individual neoantigens is evaluated in patients with resected NSCLC. Vaccine manufacturing is feasible in six of 10 enrolled patients. Toxicity is limited to grade 1-2 adverse events. Systemic T cell responses are observed in five out of six vaccinated patients, with T cell responses remaining detectable up to 19 months post vaccination. Single-cell analysis indicates that the responsive T cell population is polyclonal and exhibits the near-entire spectrum of T cell differentiation states, including a naive-like state, but excluding exhausted cell states. Three of six vaccinated patients experience disease recurrence during the follow-up period of 2 years. Collectively, these data support the feasibility, safety, and immunogenicity of this treatment in resected NSCLC.

Longitudinal molecular profiling elucidates immunometabolism dynamics in breast cancer

Although metabolic reprogramming within tumor cells and tumor microenvironment (TME) is well described in breast cancer, little is known about how the interplay of immune state and cancer metabolism evolves during treatment. Here, we characterize the immunometabolic profiles of tumor tissue samples longitudinally collected from individuals with breast cancer before, during and after neoadjuvant chemotherapy (NAC) using proteomics, genomics and histopathology. We show that the pre-, on-treatment and dynamic changes of the immune state, tumor metabolic proteins and tumor cell gene expression profiling-based metabolic phenotype are associated with treatment response. Single-cell/nucleus RNA sequencing revealed distinct tumor and immune cell states in metabolism between cold and hot tumors. Potential drivers of NAC based on above analyses were validated in vitro. In summary, the study shows that the interaction of tumor-intrinsic metabolic states and TME is associated with treatment outcome, supporting the concept of targeting tumor metabolism for immunoregulation.

Classification of multiple primary lung cancer in patients with multifocal lung cancer: assessment of a machine learning approach using multidimensional genomic data

Background

Multiple primary lung cancer (MPLC) is an increasingly well-known clinical phenomenon. However, its molecular characterizations are poorly understood, and still lacks of effective method to distinguish it from intrapulmonary metastasis (IM). Herein, we propose an identification model based on molecular multidimensional analysis in order to accurately optimize treatment.

Methods

A total of 112 Chinese lung cancers harboring at least two tumors (n = 270) were enrolled. We retrospectively selected 74 patients with 121 tumor pairs and randomly divided the tumor pairs into a training cohort and a test cohort in a 7:3 ratio. A novel model was established in training cohort, optimized for MPLC identification using comprehensive genomic profiling analyzed by a broad panel with 808 cancer-related genes, and evaluated in the test cohort and a prospective validation cohort of 38 patients with 112 tumors.

Results

We found differences in molecular characterizations between the two diseases and rigorously selected the characterizations to build an identification model. We evaluated the performance of the classifier using the test cohort data and observed an 89.5% percent agreement (PA) for MPLC and a 100.0% percent agreement for IM. The model showed an excellent area under the curve (AUC) of 0.947 and a 91.3% overall accuracy. Similarly, the assay achieved a considerable performance in the independent validation set with an AUC of 0.938 and an MPLC predictive value of 100%. More importantly, the MPLC predictive value of the classification achieved 100% in both the test set and validation cohort. Compared to our previous mutation-based method, the classifier showed better κ consistencies with clinical classification among all 112 patients (0.84 vs. 0.65, p <.01).

Conclusion

These data provide novel evidence of MPLC-specific genomic characteristics and demonstrate that our one-step molecular classifier can accurately classify multifocal lung tumors as MPLC or IM, which suggested that broad panel NGS may be a useful tool for assisting with differential diagnoses.

Beyond basics: Key mutation selection features for successful tumor-informed ctDNA detection

Tumor-informed mutation-based approaches are frequently used for detection of circulating tumor DNA (ctDNA). Not all mutations make equally effective ctDNA markers. The objective was to explore if prioritizing mutations using mutational features-such as cancer cell fraction (CCF), multiplicity, and error rate-would improve the success rate of tumor-informed ctDNA analysis. Additionally, we aimed to develop a practical and easily implementable analysis pipeline for identifying and prioritizing candidate mutations from whole-exome sequencing (WES) data. We analyzed WES and ctDNA data from three tumor-informed ctDNA studies, one on bladder cancer (Cohort A) and two on colorectal cancer (Cohorts I and N). The studies included 390 patients. For each patient, a unique set of mutations (median mutations/patient: 6, interquartile 13, range: 1-46, total n = 4023) were used as markers of ctDNA. The tool PureCN was used to assess the CCF and multiplicity of each mutation. High-CCF mutations were detected more frequently than low-CCF mutations (Cohort A: odds ratio [OR] 20.6, 95% confidence interval [CI] 5.72-173, p = 1.73e-12; Cohort I: OR 2.24, 95% CI 1.44-3.52, p = 1.66e-04; and Cohort N: OR 1.78, 95% CI 1.14-2.79, p = 7.86e-03). The detection-likelihood was additionally improved by selecting mutations with multiplicity of two or above (Cohort A: OR 1.55, 95% CI 1. 14-2.11, p = 3.85e-03; Cohort I: OR 1.78, 95% CI 1.23-2.56, p = 1.34e-03; and Cohort N: OR 1.94, 95% CI 1.63-2.31, p = 2.83e-14). Furthermore, selecting the mutations for which the ctDNA detection method had the lowest error rates, additionally improved the detection-likelihood, particularly evident when plasma cell-free DNA tumor fractions were below 0.1% (p = 2.1e-07). Selecting mutational markers with high CCF, high multiplicity, and low error rate significantly improve ctDNA detection likelihood. We provide free access to the analysis pipeline enabling others to perform qualified prioritization of mutations for tumor-informed ctDNA analysis.

Somatic Tumor Testing in Prostate Cancer: Experience of a Tertiary-Care Center Including Pathologist-Driven Reflex Testing of Localized Tumors at Diagnosis

Somatic tumor testing in prostate cancer (PCa) can guide treatment options by identifying clinically actionable variants in DNA damage repair (DDR) genes, including acquired variants not detected by germline testing alone. Guidelines currently recommend performing somatic tumor testing in metastatic PCa, while there is no consensus on the role of testing in regional disease and the optimal testing strategy is involving. This study evaluates the frequency, distribution, and pathologic correlates of somatic DDR mutations in metastatic and localized PCa following the implementation of pathologist-driven reflex testing at diagnosis. A cohort of 516 PCa were sequenced using a custom NGS panel including homologous recombination repair (HRR) and mismatch repair (MMR) genes. Variants were classified based on the AMP/ASCO/CAP guidelines. In total 183 (35.5%) patients had at least one variant: 72/516 (13.9%) had at least one Tier I or Tier II variant while 111/516 (21.5%) had a Tier III variant. Tier I/II variant(s) were identified in 27% (12/44) of metastatic biopsies and 13% (61/472) of primary samples. Overall, 12% (62/516) of patients had at least one Tier I/II variant in a HRR gene while 2.9% (10/516) had at least one Tier I/II variant in a MMR gene. The presence of a Tier I/II variant was not significantly associated with the grade group (GG) or presence of intraductal/cribriform carcinoma in the primary tumor. Among 309 reflex-tested hormone-naïve primary tumors, Tier I/II variants were identified in 10% (31/309) of cases: 9.2% (9/98) GG2; 9% (9/100) GG3; 9.1% (4/44) GG4 and 13.4% (9/67) GG5 cases. Our findings confirm the utility of somatic tumor testing to detect variants of clinical significance in PCa and provide insights that can inform the design of testing strategies. Pathologist-initiated reflex testing streamlines the availability of the results for clinical decision-making, however pathologic parameters such as grade group and intraductal/cribriform carcinoma may not be reliable to guide patient selection.

A deep-learning-based genomic status estimating framework for homologous recombination deficiency detection from low-pass whole genome sequencing

Genome-wide sequencing allows for prediction of clinical treatment responses and outcomes by estimating genomic status. Here, we developed Genomic Status scan (GSscan), a long short-term memory (LSTM)-based deep-learning framework, which utilizes low-pass whole genome sequencing (WGS) data to capture genomic instability-related features. In this study, GSscan directly surveys homologous recombination deficiency (HRD) status independent of other existing biomarkers. In breast cancer, GSscan achieved an AUC of 0.980 in simulated low-pass WGS data, and obtained a higher HRD risk score in clinical BRCA-deficient breast cancer samples (p = 1.3 × 10-4, compared with BRCA-intact samples). In ovarian cancer, GSscan obtained higher HRD risk scores in BRCA-deficient samples in both simulated data and clinical samples (p = 2.3 × 10-5 and p = 0.039, respectively, compared with BRCA-intact samples). Moreover, HRD-positive patients predicted by GSscan showed longer progression-free intervals in TCGA datasets (p = 0.0011) treated with platinum-based adjuvant chemotherapy, outperforming existing low-pass WGS-based methods. Furthermore, GSscan can accurately predict HRD status using only 1 ng of input DNA and a minimum sequencing coverage of 0.02 × , providing a reliable, accessible, and cost-effective approach. In summary, GSscan effectively and accurately detected HRD status, and provide a broadly applicable framework for disease diagnosis and selecting appropriate disease treatment.

Unraveling the potential clinical utility of circulating tumor DNA detection in colorectal cancer-evaluation in a nationwide Danish cohort

BACKGROUND

Increasingly, circulating tumor DNA (ctDNA) is proposed as a tool for minimal residual disease (MRD) assessment. Digital PCR (dPCR) offers low analysis costs and turnaround times of less than a day, making it ripe for clinical implementation. Here, we used tumor-informed dPCR for ctDNA detection in a large colorectal cancer (CRC) cohort to evaluate the potential for post-operative risk assessment and serial monitoring, and how the metastatic site may impact ctDNA detection. Additionally, we assessed how altering the ctDNA-calling algorithm could customize performance for different clinical settings.

PATIENTS AND METHODS

Stage II-III CRC patients (N = 851) treated with a curative intent were recruited. Based on whole-exome sequencing on matched tumor and germline DNA, a mutational target was selected for dPCR analysis. Plasma samples (8 ml) were collected within 60 days after operation and-for a patient subset (n = 246)-every 3-4 months for up to 36 months. Single-target dPCR was used for ctDNA detection.

RESULTS

Both post-operative and serial ctDNA detection were prognostic of recurrence [hazard ratio (HR) = 11.3, 95% confidence interval (CI) 7.8-16.4, P < 0.001; HR = 30.7, 95% CI 20.2-46.7, P < 0.001], with a cumulative ctDNA detection rate of 87% at the end of sample collection in recurrence patients. The ctDNA growth rate was prognostic of survival (HR = 2.6, 95% CI 1.5-4.4, P = 0.001). In recurrence patients, post-operative ctDNA detection was challenging for lung metastases (4/21 detected) and peritoneal metastases (2/10 detected). By modifying the cut-off for calling a sample ctDNA positive, we were able to adjust the sensitivity and specificity of our test for different clinical contexts.

CONCLUSIONS

The presented results from 851 stage II-III CRC patients demonstrate that our personalized dPCR approach effectively detects MRD after operation and shows promise for serial ctDNA detection for recurrence surveillance. The ability to adjust sensitivity and specificity shows exciting potential to customize the ctDNA caller for specific clinical settings.

Clinicopathological and genomic features of superficial esophageal squamous cell carcinomas in nondrinker, nonsmoker females

BACKGROUND

Esophageal squamous cell carcinoma (ESCC) is sometimes detected in non-drinker and non-smoker females who are considered to have very low risk of ESCC development in daily practice. This study examined the clinicopathological and genomic characteristics of ESCCs in females with no history of drinking and smoking.

METHODS

The sample comprised 118 ESCC lesions occurring in 95 female patients who underwent endoscopic submucosal dissection at our department between January 2008 and December 2019. The patients were categorized into two groups: 51 lesions in 49 patients with no history of drinking and smoking (nondrinker/nonsmoker [NDNS] group) and 69 lesions in 45 patients with a history of drinking or smoking (drinker/smoker [DS] group). We analyzed the differences in clinicopathological and cancerous genomic characteristics between the groups. Significant genomic alterations were validated using immunohistochemistry.

RESULTS

Multiple logistic regression revealed that older age, fewer multiple Lugol-voiding lesions (LVLs), and reflux esophagitis (RE) were independently associated with the occurrence of ESCCs in the NDNS group. ESCC lesions in the NDNS group were predominantly located in the mid-thoracic esophagus, posterior wall side, with 0-IIa, the aspect ratio of the lesion >2 (vertical/horizontal), and endoscopic keratinization. Genetic analysis showed that CDKN2A driver alterations were significantly more frequent and KMT2D alterations were significantly less frequent in the NDNS group than in the DS group. KMT2D alterations were strongly correlated with immunostaining.

CONCLUSION

Older nondrinker, nonsmoker females with RE and fewer multiple LVLs may develop longitudinal 0-IIa ESCC with keratinization of the posterior wall of the mid-thoracic esophagus. ESCCs in nondrinker, nonsmoker females had fewer KMT2D alterations and more CDKN2A alterations, which may be a biomarker for treatment.

Circulating tumor DNA-based copy-number profiles enable monitoring treatment effects during therapy in high-grade serous carcinoma

Circulating tumor DNA (ctDNA) analysis has emerged as a promising tool for detecting and profiling longitudinal genomics changes in cancer. While copy-number alterations (CNAs) play a major role in cancers, treatment effect monitoring using copy-number profiles has received limited attention as compared to mutations. A major reason for this is the insensitivity of CNA analysis for the real-life tumor-fraction ctDNA samples. We performed copy-number analysis on 152 plasma samples obtained from 29 patients with high-grade serous ovarian cancer (HGSC) using a sequencing panel targeting over 500 genes. Twenty-one patients had temporally matched tissue and plasma sample pairs, which enabled assessing concordance with tissues sequenced with the same panel or whole-genome sequencing and to evaluate sensitivity. Our approach could detect concordant CNA profiles in most plasma samples with as low as 5% tumor content and highly amplified regions in samples with ∼1% of tumor content. Longitudinal profiles showed changes in the CNA profiles in seven out of 11 patients with high tumor-content plasma samples at relapse. These changes included focal acquired or lost copy-numbers, even though most of the genome remained stable. Two patients displayed major copy-number profile changes during therapy. Our analysis revealed ctDNA-detectable subclonal selection resulting from both surgical operations and chemotherapy. Overall, longitudinal ctDNA data showed acquired and diminished CNAs at relapse when compared to pre-treatment samples. These results highlight the importance of genomic profiling during treatment as well as underline the usability of ctDNA.

Genomic profiling of subcutaneous patient-derived xenografts reveals immune constraints on tumor evolution in childhood solid cancer

Subcutaneous patient-derived xenografts (PDXs) are an important tool for childhood cancer research. Here, we describe a resource of 68 early passage PDXs established from 65 pediatric solid tumor patients. Through genomic profiling of paired PDXs and patient tumors (PTs), we observe low mutational similarity in about 30% of the PT/PDX pairs. Clonal analysis in these pairs show an aggressive PT minor subclone seeds the major clone in the PDX. We show evidence that this subclone is more immunogenic and is likely suppressed by immune responses in the PT. These results suggest interplay between intratumoral heterogeneity and antitumor immunity may underlie the genetic disparity between PTs and PDXs. We further show that PDXs generally recapitulate PTs in copy number and transcriptomic profiles. Finally, we report a gene fusion LRPAP1-PDGFRA. In summary, we report a childhood cancer PDX resource and our study highlights the role of immune constraints on tumor evolution.

Genomic profiling and pre-clinical modelling of breast cancer leptomeningeal metastasis reveals acquisition of a lobular-like phenotype

Breast cancer leptomeningeal metastasis (BCLM), where tumour cells grow along the lining of the brain and spinal cord, is a devastating development for patients. Investigating this metastatic site is hampered by difficulty in accessing tumour material. Here, we utilise cerebrospinal fluid (CSF) cell-free DNA (cfDNA) and CSF disseminated tumour cells (DTCs) to explore the clonal evolution of BCLM and heterogeneity between leptomeningeal and extracranial metastatic sites. Somatic alterations with potential therapeutic actionability were detected in 81% (17/21) of BCLM cases, with 19% detectable in CSF cfDNA only. BCLM was enriched in genomic aberrations in adherens junction and cytoskeletal genes, revealing a lobular-like breast cancer phenotype. CSF DTCs were cultured in 3D to establish BCLM patient-derived organoids, and used for the successful generation of BCLM in vivo models. These data reveal that BCLM possess a unique genomic aberration profile and highlight potential cellular dependencies in this hard-to-treat form of metastatic disease.

Pooled ctDNA analysis of MONALEESA phase III advanced breast cancer trials

BACKGROUND

The phase III MONALEESA trials tested the efficacy and safety of the cyclin-dependent kinase (CDK)4/6 inhibitor ribociclib with different endocrine therapy partners as first- or second-line treatment of hormone receptor-positive/human epidermal growth factor receptor 2-negative advanced breast cancer (ABC). Using the largest pooled biomarker dataset of the CDK4/6 inhibitor ribociclib in ABC to date, we identified potential biomarkers of response to ribociclib.

PATIENTS AND METHODS

Baseline circulating tumour DNA from patients in the MONALEESA trials was assessed using next-generation sequencing. An analysis of correlation between gene alteration status and progression-free survival (PFS) was carried out to identify potential biomarkers of response to ribociclib.

RESULTS

Multiple frequently altered genes were identified. Alterations in ERBB2, FAT3, FRS2, MDM2, SFRP1, and ZNF217 were associated with a greater PFS benefit with ribociclib versus placebo. Patients with high tumour mutational burden (TMB) and with ANO1, CDKN2A/2B/2C, and RB1 alterations exhibited decreased sensitivity to ribociclib versus placebo.

CONCLUSIONS

Although exploratory, these results provide insight into alterations associated with the improved response to ribociclib treatment and may inform treatment sequencing in patients with actionable alterations following progression on CDK4/6 inhibitors. Validation of potential biomarkers identified here and development of prospective trials testing their clinical utility are warranted.

CLINICALTRIALS

GOV IDENTIFIERS

NCT01958021, NCT02422615, NCT02278120.

Expressed Barcoding Enables High-Resolution Tracking of the Evolution of Drug Tolerance

For a majority of patients with non-small cell lung cancer with EGFR mutations, treatment with EGFR inhibitors (EGFRi) induces a clinical response. Despite this initial reduction in tumor size, residual disease persists that leads to disease relapse. Elucidating the preexisting biological differences between sensitive cells and surviving drug-tolerant persister cells and deciphering how drug-tolerant cells evolve in response to treatment could help identify strategies to improve the efficacy of EGFRi. In this study, we tracked the origins and clonal evolution of drug-tolerant cells at a high resolution by using an expressed barcoding system coupled with single-cell RNA sequencing. This platform enabled longitudinal profiling of gene expression and drug sensitivity in response to EGFRi across a large number of clones. Drug-tolerant cells had higher expression of key survival pathways such as YAP and EMT at baseline and could also differentially adapt their gene expression following EGFRi treatment compared with sensitive cells. In addition, drug combinations targeting common downstream components (MAPK) or orthogonal factors (chemotherapy) showed greater efficacy than EGFRi alone, which is attributable to broader targeting of the heterogeneous EGFRi-tolerance mechanisms present in tumors. Overall, this approach facilitates thorough examination of clonal evolution in response to therapy that could inform the development of improved diagnostic approaches and treatment strategies for targeting drug-tolerant cells.

SIGNIFICANCE

The evolution and heterogeneity of EGFR inhibitor tolerance are identified in a large number of clones at enhanced cellular and temporal resolution using an expressed barcode technology coupled with single-cell RNA sequencing.

Mismatch repair deficiency is not sufficient to elicit tumor immunogenicity

DNA mismatch repair deficiency (MMRd) is associated with a high tumor mutational burden (TMB) and sensitivity to immune checkpoint blockade (ICB) therapy. Nevertheless, most MMRd tumors do not durably respond to ICB and critical questions remain about immunosurveillance and TMB in these tumors. In the present study, we developed autochthonous mouse models of MMRd lung and colon cancer. Surprisingly, these models did not display increased T cell infiltration or ICB response, which we showed to be the result of substantial intratumor heterogeneity of mutations. Furthermore, we found that immunosurveillance shapes the clonal architecture but not the overall burden of neoantigens, and T cell responses against subclonal neoantigens are blunted. Finally, we showed that clonal, but not subclonal, neoantigen burden predicts ICB response in clinical trials of MMRd gastric and colorectal cancer. These results provide important context for understanding immune evasion in cancers with a high TMB and have major implications for therapies aimed at increasing TMB.

Integrated molecular and multiparametric MRI mapping of high-grade glioma identifies regional biologic signatures

Sampling restrictions have hindered the comprehensive study of invasive non-enhancing (NE) high-grade glioma (HGG) cell populations driving tumor progression. Here, we present an integrated multi-omic analysis of spatially matched molecular and multi-parametric magnetic resonance imaging (MRI) profiling across 313 multi-regional tumor biopsies, including 111 from the NE, across 68 HGG patients. Whole exome and RNA sequencing uncover unique genomic alterations to unresectable invasive NE tumor, including subclonal events, which inform genomic models predictive of geographic evolution. Infiltrative NE tumor is alternatively enriched with tumor cells exhibiting neuronal or glycolytic/plurimetabolic cellular states, two principal transcriptomic pathway-based glioma subtypes, which respectively demonstrate abundant private mutations or enrichment in immune cell signatures. These NE phenotypes are non-invasively identified through normalized K2 imaging signatures, which discern cell size heterogeneity on dynamic susceptibility contrast (DSC)-MRI. NE tumor populations predicted to display increased cellular proliferation by mean diffusivity (MD) MRI metrics are uniquely associated with EGFR amplification and CDKN2A homozygous deletion. The biophysical mapping of infiltrative HGG potentially enables the clinical recognition of tumor subpopulations with aggressive molecular signatures driving tumor progression, thereby informing precision medicine targeting.

Potential value of ctDNA monitoring in metastatic HR + /HER2 - breast cancer: longitudinal ctDNA analysis in the phase Ib MONALEESASIA trial

BACKGROUND

There is increasing interest in the use of liquid biopsies, but data on longitudinal analyses of circulating tumor DNA (ctDNA) remain relatively limited. Here, we report a longitudinal ctDNA analysis of MONALEESASIA, a phase Ib trial evaluating the efficacy and safety of ribociclib plus endocrine therapy (ET) in Asian patients with hormone receptor-positive, human epidermal growth factor receptor-2-negative advanced breast cancer.

METHODS

MONALEESASIA enrolled premenopausal and postmenopausal Japanese and postmenopausal non-Japanese Asian patients. All patients received ribociclib with ET (letrozole, fulvestrant, or tamoxifen with goserelin). ctDNA was analyzed using a targeted next-generation sequencing panel of 572 cancer-related genes and correlated by best overall response (BOR).

RESULTS

Five hundred seventy-four cell-free DNA samples from 87 patients were tested. The most frequently altered genes at baseline included PIK3CA (29%) and TP53 (22%). Treatment with ribociclib plus ET decreased ctDNA in most patients at the first on-treatment time point, regardless of dose or ET partner. Patients with partial response and stable disease had lower ctDNA at baseline that remained low until data cutoff if no progressive disease occurred. Most patients with progressive disease as the best response had higher ctDNA at baseline that remained high at the end of treatment. For patients with partial response and stable disease with subsequent progression, ctDNA increased towards the end of treatment in most patients, with a median lead time of 83 days (14-309 days). In some patients with BOR of partial response who experienced disease progression later, specific gene alterations and total ctDNA fraction increased; this was sometimes observed concurrently with the development of new lesions without a change in target lesion size. Patients with alterations in PIK3CA and TP53 at baseline had shorter median progression-free survival compared with patients with wild-type PIK3CA and TP53, 12.7 and 7.3 months vs 19.2 and 19.4 months, respectively (P = .016 and P = .0001, respectively).

CONCLUSIONS

Higher ctDNA levels and PIK3CA and TP53 alterations detected at baseline were associated with inferior outcomes. On-treatment ctDNA levels were associated with different patterns based on BOR. Longitudinal tracking of ctDNA may be useful for monitoring tumor status and detection of alterations with treatment implications.

TRIAL REGISTRATION

ClinicalTrials.gov NCT02333370 . Registered on January 7, 2015.

Proteogenomic analysis of chemo-refractory high-grade serous ovarian cancer

To improve the understanding of chemo-refractory high-grade serous ovarian cancers (HGSOCs), we characterized the proteogenomic landscape of 242 (refractory and sensitive) HGSOCs, representing one discovery and two validation cohorts across two biospecimen types (formalin-fixed paraffin-embedded and frozen). We identified a 64-protein signature that predicts with high specificity a subset of HGSOCs refractory to initial platinum-based therapy and is validated in two independent patient cohorts. We detected significant association between lack of Ch17 loss of heterozygosity (LOH) and chemo-refractoriness. Based on pathway protein expression, we identified 5 clusters of HGSOC, which validated across two independent patient cohorts and patient-derived xenograft (PDX) models. These clusters may represent different mechanisms of refractoriness and implicate putative therapeutic vulnerabilities.

Development of the NOGGO GIS v1 Assay, a Comprehensive Hybrid-Capture-Based NGS Assay for Therapeutic Stratification of Homologous Repair Deficiency Driven Tumors and Clinical Validation

The worldwide approval of the combination maintenance therapy of olaparib and bevacizumab in advanced high-grade serous ovarian cancer requires complex molecular diagnostic assays that are sufficiently robust for the routine detection of driver mutations in homologous recombination repair (HRR) genes and genomic instability (GI), employing formalin-fixed (FFPE) paraffin-embedded tumor samples without matched normal tissue. We therefore established a DNA-based hybrid capture NGS assay and an associated bioinformatic pipeline that fulfils our institution's specific needs. The assay´s target regions cover the full exonic territory of relevant cancer-related genes and HRR genes and more than 20,000 evenly distributed single nucleotide polymorphism (SNP) loci to allow for the detection of genome-wide allele specific copy number alterations (CNA). To determine GI status, we implemented an %CNA score that is robust across a broad range of tumor cell content (25-85%) often found in routine FFPE samples. The assay was established using high-grade serous ovarian cancer samples for which BRCA1 and BRCA2 mutation status as well as Myriad MyChoice homologous repair deficiency (HRD) status was known. The NOGGO (Northeastern German Society for Gynecologic Oncology) GIS (GI-Score) v1 assay was clinically validated on more than 400 samples of the ENGOT PAOLA-1 clinical trial as part of the European Network for Gynaecological Oncological Trial groups (ENGOT) HRD European Initiative. The "NOGGO GIS v1 assay" performed using highly robust hazard ratios for progression-free survival (PFS) and overall survival (OS), as well a significantly lower dropout rate than the Myriad MyChoice clinical trial assay supporting the clinical utility of the assay. We also provide proof of a modular and scalable routine diagnostic method, that can be flexibly adapted and adjusted to meet future clinical needs, emerging biomarkers, and further tumor entities.

Salivary gland cancer organoids are valid for preclinical genotype-oriented medical precision trials

Salivary gland cancers (SGCs) are heterogeneous tumors, and precision oncology represents a promising therapeutic approach; however, its impact on SGCs remains obscure. This study aimed to establish a translational model for testing molecular-targeted therapies by combining patient-derived organoids and genomic analyses of SGCs. We enrolled 29 patients, including 24 with SGCs and 5 with benign tumors. Resected tumors were subjected to organoid and monolayer cultures, as well as whole-exome sequencing. Organoid and monolayer cultures of SGCs were successfully established in 70.8% and 62.5% of cases, respectively. Organoids retained most histopathological and genetic profiles of their original tumors. In contrast, 40% of the monolayer-cultured cells did not harbor somatic mutations of their original tumors. The efficacy of molecular-targeted drugs tested on organoids depended on their oncogenic features. Organoids recapitulated the primary tumors and were useful for testing genotype-oriented molecular targeted therapy, which is valuable for precision medicine in patients with SGCs.

Genomic and transcriptomic profiling reveal molecular characteristics of parathyroid carcinoma

Genomic and transcriptomic profiling has enhanced the diagnostic and treatment options for many cancers. However, the molecular characteristics of parathyroid cancer remain largely unexplored, thereby limiting the development of new therapeutic interventions. Herein, we conducted genomic and transcriptomic sequencing of 50 parathyroid tissues (12 carcinomas, 28 adenomas, and 10 normal tissues) to investigate the intrinsic and comparative molecular features of parathyroid carcinoma. We confirmed multiple two-hit mutation patterns in cell division cycle 73 (CDC73) that converged to biallelic inactivation, calling into question the presence of a second hit in other genes. In addition, allele-specific repression of CDC73 in copies with germline-truncating variants suggested selective pressure prior to tumorigenesis. Transcriptomic analysis identified upregulation of the expression of E2F targets, KRAS and TNF-alpha signaling, and epithelial-mesenchymal transition pathways in carcinomas compared to adenomas and normal tissues. A molecular classification model based on carcinoma-specific genes clearly separated carcinomas from adenomas and normal tissues, the clinical utility of which was demonstrated in two patients with uncertain malignant potential. A deeper analysis of gene expression and functional prediction suggested that Wilms tumor 1 (WT1) is a potential biomarker for CDC73-mutant parathyroid carcinoma, which was further validated through immunohistochemistry. Overall, our study revealed the genomic and transcriptomic profiles of parathyroid carcinoma and may help direct future precision diagnostic and therapeutic improvements.

A phase I trial of LXS196, a protein kinase C (PKC) inhibitor, for metastatic uveal melanoma

BACKGROUND

Up to 50% of patients with uveal melanoma develop metastases (MUM) with a poor prognosis and median overall survival of approximately 1 year.

METHODS

This phase I study evaluated the safety, tolerability, pharmacokinetics, pharmacodynamics and efficacy of the oral protein kinase C inhibitor LXS196 in 68 patients with MUM (NCT02601378). Patients received LXS196 doses ranging from 100-1000 mg once daily (QD; n = 38) and 200-400 mg twice daily (BID; n = 30).

RESULTS

First cycle dose-limiting toxicities (DLTs) were observed in 7/38 (18.4%) QD and 2/17 (11.8%) BID patients. Hypotension was the most common DLT, occurring at doses ≥500 mg/day, and manageable with LXS196 interruption and dose reduction. Median duration of exposure to LXS196 was 3.71 months (range: 1.81-15.28) for QD and 4.6 months (range: 0.33-58.32) for BID dosing. Clinical activity was observed in 6/66 (9.1%) evaluable patients achieving response (CR/PR), with a median duration of response of 10.15 months (range: 2.99-41.95); 45/66 had stable disease (SD) per RECIST v1.1. At 300 mg BID, the recommended dose for expansion, 2/18 (11.1%) evaluable patients achieved PR and 12/18 (66.7%) had SD.

CONCLUSION

These results suggest manageable toxicity and encouraging clinical activity of single-agent LXS196 in patients with MUM.

Impact of Whole Genome Doubling on Detection of Circulating Tumor DNA in Colorectal Cancer

OBJECTIVE

Circulating tumor DNA (ctDNA) is a candidate biomarker of cancer with practice-changing potential in the detection of both early and residual disease. Disease stage and tumor size affect the probability of ctDNA detection, whereas little is known about the influence of other tumor characteristics on ctDNA detection. This study investigates the impact of tumor cell whole-genome doubling (WGD) on the detection of ctDNA in plasma collected preoperatively from newly diagnosed colorectal cancer (CRC) patients.

METHODS

WGD was estimated from copy numbers derived from whole-exome sequencing (WES) data of matched tumor and normal DNA from 833 Danish CRC patients. To explore if tumor WGD status impacts ctDNA detection, we applied tumor-informed ctDNA analysis to preoperative plasma samples from all patients.

RESULTS

Patients with WGD+ tumors had 53% increased odds of being ctDNA positive (OR = 1.53, 95%CI: 1.12-2.09). After stratification for UICC stage, the association persisted for Stage I (OR = 2.44, 95%CI: 1.22-5.03) and Stage II (OR = 1.76, 95%CI: 1.11-2.81) but not for Stage III (OR = 0.83, 95%CI: 0.44-1.53) patients.

CONCLUSION

The presence of WGD significantly increases the probability of detecting ctDNA, particularly for early-stage disease. In patients with more advanced disease, the benefit of WGD on ctDNA detection is less pronounced, consistent with increased DNA shedding from these tumors, making ctDNA detection less dependent on the amount of ctDNA released per tumor cell.

Persistent mutation burden drives sustained anti-tumor immune responses

Tumor mutation burden is an imperfect proxy of tumor foreignness and has therefore failed to consistently demonstrate clinical utility in predicting responses in the context of immunotherapy. We evaluated mutations in regions of the genome that are unlikely to undergo loss in a pan-cancer analysis across 31 tumor types (n = 9,242) and eight immunotherapy-treated cohorts of patients with non-small-cell lung cancer, melanoma, mesothelioma, and head and neck cancer (n = 524). We discovered that mutations in single-copy regions and those present in multiple copies per cell constitute a persistent tumor mutation burden (pTMB) which is linked with therapeutic response to immune checkpoint blockade. Persistent mutations were retained in the context of tumor evolution under selective pressure of immunotherapy and tumors with a high pTMB content were characterized by a more inflamed tumor microenvironment. pTMB imposes an evolutionary bottleneck that cancer cells cannot overcome and may thus drive sustained immunologic tumor control in the context of immunotherapy.

Fast, accurate, and racially unbiased pan-cancer tumor-only variant calling with tabular machine learning

Accurately identifying somatic mutations is essential for precision oncology and crucial for calculating tumor-mutational burden (TMB), an important predictor of response to immunotherapy. For tumor-only variant calling (i.e., when the cancer biopsy but not the patient's normal tissue sample is sequenced), accurately distinguishing somatic mutations from germline variants is a challenging problem that, when unaddressed, results in unreliable, biased, and inflated TMB estimates. Here, we apply machine learning to the task of somatic vs germline classification in tumor-only solid tumor samples using TabNet, XGBoost, and LightGBM, three machine-learning models for tabular data. We constructed a training set for supervised classification using features derived exclusively from tumor-only variant calling and drawing somatic and germline truth labels from an independent pipeline using the patient-matched normal samples. All three trained models achieved state-of-the-art performance on two holdout test datasets: a TCGA dataset including sarcoma, breast adenocarcinoma, and endometrial carcinoma samples (AUC > 94%), and a metastatic melanoma dataset (AUC > 85%). Concordance between matched-normal and tumor-only TMB improves from R2 = 0.006 to 0.71-0.76 with the addition of a machine-learning classifier, with LightGBM performing best. Notably, these machine-learning models generalize across cancer subtypes and capture kits with a call rate of 100%. We reproduce the recent finding that tumor-only TMB estimates for Black patients are extremely inflated relative to that of white patients due to the racial biases of germline databases. We show that our approach with XGBoost and LightGBM eliminates this significant racial bias in tumor-only variant calling.

ETV6 Deficiency Unlocks ERG-Dependent Microsatellite Enhancers to Drive Aberrant Gene Activation in B-Lymphoblastic Leukemia

Distal enhancers play critical roles in sustaining oncogenic gene-expression programs. We identify aberrant enhancer-like activation of GGAA tandem repeats as a characteristic feature of B-cell acute lymphoblastic leukemia (B-ALL) with genetic defects of the ETV6 transcriptional repressor, including ETV6-RUNX1+ and ETV6-null B-ALL. We show that GGAA repeat enhancers are direct activators of previously identified ETV6-RUNX1+/- like B-ALL "signature" genes, including the likely leukemogenic driver EPOR. When restored to ETV6-deficient B-ALL cells, ETV6 directly binds to GGAA repeat enhancers, represses their acetylation, downregulates adjacent genes, and inhibits B-ALL growth. In ETV6-deficient B-ALL cells, we find that the ETS transcription factor ERG directly binds to GGAA microsatellite enhancers and is required for sustained activation of repeat enhancer-activated genes. Together, our findings reveal an epigenetic gatekeeper function of the ETV6 tumor suppressor gene and establish microsatellite enhancers as a key mechanism underlying the unique gene-expression program of ETV6-RUNX1+/- like B-ALL.

SIGNIFICANCE

We find a unifying mechanism underlying a leukemia subtype-defining gene-expression signature that relies on repetitive elements with poor conservation between humans and rodents. The ability of ETV6 to antagonize promiscuous, nonphysiologic ERG activity may shed light on other roles of these key regulators in hematolymphoid development and human disease. See related commentary by Mercher, p. 2. This article is highlighted in the In This Issue feature, p. 1.

Genetic Ancestry Inference from Cancer-Derived Molecular Data across Genomic and Transcriptomic Platforms

Genetic ancestry-oriented cancer research requires the ability to perform accurate and robust genetic ancestry inference from existing cancer-derived data, including whole-exome sequencing, transcriptome sequencing, and targeted gene panels, very often in the absence of matching cancer-free genomic data. Here we examined the feasibility and accuracy of computational inference of genetic ancestry relying exclusively on cancer-derived data. A data synthesis framework was developed to optimize and assess the performance of the ancestry inference for any given input cancer-derived molecular profile. In its core procedure, the ancestral background of the profiled patient is replaced with one of any number of individuals with known ancestry. The data synthesis framework is applicable to multiple profiling platforms, making it possible to assess the performance of inference specifically for a given molecular profile and separately for each continental-level ancestry; this ability extends to all ancestries, including those without statistically sufficient representation in the existing cancer data. The inference procedure was demonstrated to be accurate and robust in a wide range of sequencing depths. Testing of the approach in four representative cancer types and across three molecular profiling modalities showed that continental-level ancestry of patients can be inferred with high accuracy, as quantified by its agreement with the gold standard of deriving ancestry from matching cancer-free molecular data. This study demonstrates that vast amounts of existing cancer-derived molecular data are potentially amenable to ancestry-oriented studies of the disease without requiring matching cancer-free genomes or patient self-reported ancestry.

SIGNIFICANCE

The development of a computational approach that enables accurate and robust ancestry inference from cancer-derived molecular profiles without matching cancer-free data provides a valuable methodology for genetic ancestry-oriented cancer research.

Differences in genome, transcriptome, miRNAome, and methylome in synchronous and metachronous liver metastasis of colorectal cancer

Despite distant metastases being the critical factor affecting patients' survival, they remain poorly understood. Our study thus aimed to molecularly characterize colorectal cancer liver metastases (CRCLMs) and explore whether molecular profiles differ between Synchronous (SmCRC) and Metachronous (MmCRC) colorectal cancer. This characterization was performed by whole exome sequencing, whole transcriptome, whole methylome, and miRNAome. The most frequent somatic mutations were in APC, SYNE1, TP53, and TTN genes. Among the differently methylated and expressed genes were those involved in cell adhesion, extracellular matrix organization and degradation, neuroactive ligand-receptor interaction. The top up-regulated microRNAs were hsa-miR-135b-3p and -5p, and the hsa-miR-200-family while the hsa-miR-548-family belonged to the top down-regulated. MmCRC patients evinced higher tumor mutational burden, a wider median of duplications and deletions, and a heterogeneous mutational signature than SmCRC. Regarding chronicity, a significant down-regulation of SMOC2 and PPP1R9A genes in SmCRC compared to MmCRC was observed. Two miRNAs were deregulated between SmCRC and MmCRC, hsa-miR-625-3p and has-miR-1269-3p. The combined data identified the IPO5 gene. Regardless of miRNA expression levels, the combined analysis resulted in 107 deregulated genes related to relaxin, estrogen, PI3K-Akt, WNT signaling pathways, and intracellular second messenger signaling. The intersection between our and validation sets confirmed the validity of our results. We have identified genes and pathways that may be considered as actionable targets in CRCLMs. Our data also provide a valuable resource for understanding molecular distinctions between SmCRC and MmCRC. They have the potential to enhance the diagnosis, prognostication, and management of CRCLMs by a molecularly targeted approach.

Multiregional Sequencing Analysis Reveals Extensive Genetic Heterogeneity in Gastric Tumors from Latinos

Gastric cancer is a leading cause of cancer mortality and health disparities in Latinos. We evaluated gastric intratumoral heterogeneity using multiregional sequencing of >700 cancer genes in 115 tumor biopsies from 32 patients, 29 who were Latinos. Analyses focused on comparisons with The Cancer Genome Atlas (TCGA) and on mutation clonality, druggability, and signatures. We found that only approximately 30% of all mutations were clonal and that only 61% of the known TCGA gastric cancer drivers harbored clonal mutations. Multiple clonal mutations were found in new candidate gastric cancer drivers such as EYS, FAT4, PCDHA1, RAD50, EXO1, RECQL4, and FSIP2. The genomically stable (GS) molecular subtype, which has the worse prognosis, was identified in 48% of our Latino patients, a fraction that was >2.3-fold higher than in TCGA Asian and White patients. Only a third of all tumors harbored clonal pathogenic mutations in druggable genes, with most (93%) GS tumors lacking actionable clonal mutations. Mutation signature analyses revealed that, in microsatellite-stable (MSS) tumors, DNA repair mutations were common for both tumor initiation and progression, while tobacco, POLE, and inflammation signatures likely initiate carcinogenesis. MSS tumor progression was likely driven by aging- and aflatoxin-associated mutations, as these latter changes were usually nonclonal. In microsatellite-unstable tumors, nonclonal tobacco-associated mutations were common. Our study, therefore, contributed to advancing gastric cancer molecular diagnostics and suggests clonal status is important to understanding gastric tumorigenesis. Our findings of a higher frequency of a poor prognosis associated molecular subtype in Latinos and a possible new aflatoxin gastric cancer etiology also advance cancer disparities research.

SIGNIFICANCE

Our study contributes to advancing our knowledge of gastric carcinogenesis, diagnostics, and cancer health disparities.

Deriving tumor purity from cancer next generation sequencing data: applications for quantitative ERBB2 (HER2) copy number analysis and germline inference of BRCA1 and BRCA2 mutations

Tumor purity, or the relative contribution of tumor cells out of all cells in a pathological specimen, influences mutation identification and clinical interpretation of cancer panel next generation sequencing results. Here, we describe a method of calculating tumor purity using pathologist-guided copy number analysis from sequencing data. Molecular calculation of tumor purity showed strong linear correlation with purity derived from driver KRAS or BRAF variant allele fractions in colorectal cancers (R² = 0.79) compared to histological estimation in the same set of colorectal cancers (R² = 0.01) and in a broader dataset of cancers with various diagnoses (R² = 0.35). We used calculated tumor purity to quantitate ERBB2 copy number in breast carcinomas with equivocal immunohistochemical staining and demonstrated strong correlation with fluorescence in situ hybridization (R² = 0.88). Finally, we used calculated tumor purity to infer the germline status of variants in breast and ovarian carcinomas with concurrent germline testing. Tumor-only next generation sequencing correctly predicted the somatic versus germline nature of 26 of 26 (100%) pathogenic TP53, BRCA1 and BRCA2 variants. In this article, we describe a framework for calculating tumor purity from cancer next generation sequencing data. Accurate tumor purity assessment can be assimilated into interpretation pipelines to derive clinically useful information from cancer genomic panels.

TOSCA: an automated Tumor Only Somatic CAlling workflow for somatic mutation detection without matched normal samples

Motivation

Accurate classification of somatic variants in a tumor sample is often accomplished by utilizing a paired normal tissue sample from the same patient to enable the separation of private germline mutations from somatic variants. However, a paired normal sample is not always available, making a reliable somatic variant calling more challenging. In silico screening of variants against public or private databases and other filtering approaches are often used in absence of a paired normal sample. Nevertheless, difficulties in performing a tumor-only calling with sufficient accuracy and lack of open-source software have limited their applications in clinical research.

Results

To address these limitations, we developed TOSCA, the first automated tumor-only somatic calling workflow in whole-exome sequencing and targeted panel sequencing data which performs an end-to-end analysis from raw read files, via quality checks, alignment and variant calling to functional annotation, databases filtering, tumor purity and ploidy estimation and variant classification. Application of our workflow to tumor-only data provides estimates of somatic and germline variants that are consistent with results from paired analyses.

Availability and implementation

TOSCA is a Snakemake-based workflow and freely available at https://github.com/mdelcorvo/TOSCA.

Supplementary information

Supplementary data are available at Bioinformatics Advances online.

Unsupervised detection of fragment length signatures of circulating tumor DNA using non-negative matrix factorization

Sequencing of cell-free DNA (cfDNA) is currently being used to detect cancer by searching both for mutational and non-mutational alterations. Recent work has shown that the length distribution of cfDNA fragments from a cancer patient can inform tumor load and type. Here, we propose non-negative matrix factorization (NMF) of fragment length distributions as a novel and completely unsupervised method for studying fragment length patterns in cfDNA. Using shallow whole-genome sequencing (sWGS) of cfDNA from a cohort of patients with metastatic castration-resistant prostate cancer (mCRPC), we demonstrate how NMF accurately infers the true tumor fragment length distribution as an NMF component - and that the sample weights of this component correlate with ctDNA levels (r=0.75). We further demonstrate how using several NMF components enables accurate cancer detection on data from various early stage cancers (AUC=0.96). Finally, we show that NMF, when applied across genomic regions, can be used to discover fragment length signatures associated with open chromatin.

Responses to the Tepotinib in Gastric Cancers with MET Amplification or MET Exon 14 Skipping Mutations and High Expression of Both PD-L1 and CD44

Both MET exon 14 skipping mutation (METex14SM) and high copy-number variation (CNV) lead to enhanced carcinogenesis; additionally, programmed-death ligand 1 (PD-L1) is often upregulated in cancers. In this study, we characterized the expression of MET (including METex14SM), PD-L1, and CD44 in human gastric cancer (GC) cells as well as the differential susceptibility of these cells to tepotinib. Tepotinib treatments inhibited the growth of five GC cells in a dose-dependent manner with a concomitant induction of cell death. Tepotinib treatments also significantly reduced the expression of phospho-MET, total MET, c-Myc, VEGFR2, and Snail protein in SNU620, MKN45, and Hs746T cells. Notably, tepotinib significantly reduced the expression of CD44 and PD-L1 in METex14SM Hs746T cells. By contrast, tepotinib was only slightly active against SNU638 and KATO III cells. Migration was reduced to a greater extent in the tepotinib-treated group than in the control group. Tepotinib may have therapeutic effects on c-MET-amplified GC, a high expression of both PD-L1 and CD44, and METex14SM. Clinical studies are needed to confirm these therapeutic effects.

Baseline and post-treatment biomarkers of resistance to anti-PD-1 therapy in acral and mucosal melanoma: an observational study

Background

Immunotherapies targeting programmed cell death-1 (PD-1) and its ligands have improved clinical outcomes for advanced melanoma. However, many tumors exhibit primary resistance or acquire secondary resistance after an initial positive response. The mechanisms of resistance are not well understood, and no validated predictive biomarkers are available. This exploratory study aimed to characterize baseline differences and molecular changes arising during treatment in acral and mucosal melanomas that exhibited primary or secondary resistance to anti-PD-1 monotherapy.

Methods

This was an observational retrospective study of 124 patients who had been treated for metastatic acral or mucosal melanoma with anti-PD-1 monotherapy. Tumor samples were collected at baseline (all patients) and post-treatment (resistant tumors only) and were assayed by immunohistochemistry, whole-exome sequencing, and RNA sequencing.

Results

At baseline, more non-progressor than resistant tumors exhibited expression of PD-L1, immune cell infiltration, and high tumor mutational burden (TMB); baseline PD-L1 expression was also more common in secondary-resistant than in primary-resistant tumors as well as in late versus early secondary-resistant tumors. Non-progressor tumors also had higher median baseline expression of an 18-gene T cell-inflamed gene expression profile (Tcell_infGEP). Among resistant tumors, the proportion of PD-L1-positive melanomas and the expression of the Tcell_infGEP mRNA signature increased during treatment, while the expression of mRNA signatures related to WNT and INFA1 signaling decreased. There was evidence for greater changes from baseline in secondary-resistant versus primary-resistant tumors for some markers, including expression of RAS-related and WNT-related mRNA signatures and density of CD11c⁺ and FOXP3⁺ T cells. Greater changes in CD11c⁺ cell density were observed in early compared with late secondary-resistant tumors.

Conclusions

Our findings suggest that Tcell_infGEP and PD-L1 expression, TMB, immune cell infiltration, and RAS and WNT signaling warrant further investigation as potential mechanisms and/or biomarkers of anti-PD-1 therapy resistance in acral and mucosal melanomas. Confirmation of these findings in larger populations is needed.

Molecular characterization of low-grade serous ovarian carcinoma identifies genomic aberrations according to hormone receptor expression

Hormone receptor expression is a characteristic of low-grade serous ovarian carcinoma (LGSOC). Studies investigating estrogen receptor (ER) and progesterone receptor (PR) expression levels suggest its prognostic and predictive significance, although their associations with key molecular aberrations are not well understood. As such, we sought to describe the specific genomic profiles associated with different ER/PR expression patterns and survival outcomes in a cohort of patients with advanced disease. The study comprised fifty-five advanced-staged (III/IV) LGSOCs from the Canadian Ovarian Experimental Unified Resource (COEUR) for which targeted mutation sequencing, copy-number aberration, clinical and follow-up data were available. ER, PR, and p16 expression were assessed by immunohistochemistry. Tumors were divided into low and high ER/PR expression groups based on Allred scoring. Copy number analysis revealed that PR-low tumors (Allred score <2) had a higher fraction of the genome altered by copy number changes compared to PR-high tumors (p = 0.001), with cancer genes affected within specific loci linked to altered peptidyl-tyrosine kinase, MAP-kinase, and PI3-kinase signaling. Cox regression analysis showed that ER-high (p = 0.02), PR-high (p = 0.03), stage III disease (p = 0.02), low residual disease burden (p = 0.01) and normal p16 expression (p<0.001) were all significantly associated with improved overall survival. This study provides evidence that genomic aberrations are linked to ER/PR expression in primary LGSOC.

Allele-informed copy number evaluation of plasma DNA samples from metastatic prostate cancer patients: the PCF_SELECT consortium assay

Sequencing of cell-free DNA (cfDNA) in cancer patients' plasma offers a minimally-invasive solution to detect tumor cell genomic alterations to aid real-time clinical decision-making. The reliability of copy number detection decreases at lower cfDNA tumor fractions, limiting utility at earlier stages of the disease. To test a novel strategy for detection of allelic imbalance, we developed a prostate cancer bespoke assay, PCF_SELECT, that includes an innovative sequencing panel covering ∼25 000 high minor allele frequency SNPs and tailored analytical solutions to enable allele-informed evaluation. First, we assessed it on plasma samples from 50 advanced prostate cancer patients. We then confirmed improved detection of genomic alterations in samples with <10% tumor fractions when compared against an independent assay. Finally, we applied PCF_SELECT to serial plasma samples intensively collected from three patients previously characterized as harboring alterations involving DNA repair genes and consequently offered PARP inhibition. We identified more extensive pan-genome allelic imbalance than previously recognized in prostate cancer. We confirmed high sensitivity detection of BRCA2 allelic imbalance with decreasing tumor fractions resultant from treatment and identified complex ATM genomic states that may be incongruent with protein losses. Overall, we present a framework for sensitive detection of allele-specific copy number changes in cfDNA.

Ductal and acinar components of mixed prostatic adenocarcinoma frequently have a common clonal origin

BACKGROUND

Ductal adenocarcinoma (DA) is an aggressive subtype of prostate cancer. It is most commonly seen in mixed tumors together with conventional acinar adenocarcinoma (AA). The genetic profile of DA and its clonal origin is not fully characterized.

OBJECTIVE

To investigate whether DA represents a distinct genetic subtype and to investigate the somatic relationship between the ductal and acinar components of mixed cancers.

DESIGN, SETTING, AND PARTICIPANTS

In 17 radical prostatectomy specimens ductal and acinar tumor components from the same tumor foci were dissected. DNA was extracted and genomic sequencing performed. After exclusion of two cases with low cell yield, 15 paired samples remained for analysis.

RESULTS

In 12 of 15 cases a common somatic denominator was identified, while three cases had clonally separate components. In DA, TMPRSS2-ERG gene fusions were detected in 47% (7/15), clonal FOXA1 alterations in 33% (5/15) and SPOP alterations in 27% (4/15) of cases. In one case KIAA1549-BRAF fusion was identified. Genome doubling events, resulting in an increased ploidy, were identified in the DA in 53% (8/15) of cases, but not seen in any AA. PTEN and CTNNB1 alterations were enriched in DA (6/15) but not seen in any AA. No cancers showed microsatellite instability or high tumor mutation burden.

CONCLUSIONS

Ductal and acinar prostate adenocarcinoma components of mixed tumors most often share the same origin and are clonally related. DA components in mixed tumor often exhibit genome doubling events resulting in aneuploidy, consistent with the aggressive nature of high grade prostate cancer.

VHL mosaicism: the added value of multi-tissue analysis

Von Hippel-Lindau disease (VHL) is an autosomal dominant, inherited syndrome with variants in the VHL gene causing predisposition to multi-organ benign and malignant neoplasms. A germline VHL variant is identified in 95-100% of individuals with a clinical diagnosis of VHL. Here, we present the case of an individual with a clinical diagnosis of VHL disease where peripheral blood DNA analysis did not detect a VHL variant. Sequencing of four tumor tissues (ccRCC, pheochromocytoma, lung via sputum, liver) revealed a VHL c.593 T > C (p.Leu198Pro) variant at varying allele fractions (range: 10-55%) in all tissues. Re-examination of the peripheral blood sequencing data identified this variant at 6% allele fraction. Tumor analysis revealed characteristic cytomorphological, immunohistochemical reactivity for alpha-inhibin, and CAIX, and reduced pVHL reactivity supported VHL-related pseudohypoxia. This report of a rare case of VHL mosaicism highlights the value of tissue testing in VHL variant negative cases.

CNViz: An R/Shiny Application for Interactive Copy Number Variant Visualization in Cancer

Copy number variants (CNVs) comprise a class of mutation which includes deletion, duplication, or amplification events that range in size from smaller than a single-gene or exon, to the size of a full chromosome. These changes can affect gene expression levels and are thus implicated in disease, including cancer. Although a variety of tools and methodologies exist to detect CNVs using data from massively parallel sequencing (also referred to as next-generation sequencing), it can be difficult to appreciate the copy number profile in a list format or as a static image. CNViz is a freely accessible R/Bioconductor package that launches an interactive R/Shiny visualization tool to facilitate review of copy number data. As inputs, it requires genomic locations and corresponding copy number ratios for probe, gene, and/or segment-level data. If supplied, loss of heterozygosity (LOH), focal variant data [single nucleotide variants (SNVs) and small insertions and deletions (indels)], and metadata (e.g., specimen purity and ploidy) can also be incorporated into the visualization. The CNViz R/Bioconductor package is an easy-to-use tool built with the intent of encouraging visualization and exploration of copy number variation. CNViz can be used in a clinical setting as well as for research to study patterns in human cancers more broadly. The intuitive interface allows users to visualize the copy number profile of a specimen, dynamically change resolution to explore gene and probe-level copy number changes, and simultaneously integrate LOH, SNV, and indel findings. CNViz is available for download as an R package via Bioconductor. An example of the application is available at rebeccagreenblatt.shinyapps.io/cnviz_example.

Co-occurrence CDK4/6 amplification serves as biomarkers of de novo EGFR TKI resistance in sensitizing EGFR mutation non-small cell lung cancer

Despite the development of predictive biomarkers to shape treatment paradigms and outcomes, de novo EGFR TKI resistance advanced non-small cell lung cancer (NSCLC) remains an issue of concern. We explored clinical factors in 332 advanced NSCLC who received EGFR TKI and molecular characteristics through 65 whole exome sequencing of various EGFR TKI responses including; de novo (progression within 3 months), intermediate response (IRs) and long-term response (LTRs) (durability > 2 years). Uncommon EGFR mutation subtypes were significantly variable enriched in de novo resistance. The remaining sensitizing EGFR mutation subtypes (exon 19 del and L858R) accounted for 75% of de novo resistance. Genomic landscape analysis was conducted, focusing in 10 frequent oncogenic signaling pathways with functional contributions; cell cycle, Hippo, Myc, Notch, Nrf2, PI-3-Kinase/Akt, RTK-RAS, TGF-β, p53 and β-catenin/Wnt signaling. Cell cycle pathway was the only significant alteration pathway among groups with the FDR p-value of 6 × 10^-4. We found only significant q-values of < 0.05 in 7 gene alterations; CDK6, CCNE1, CDK4, CCND3, MET, FGFR4 and HRAS which enrich in de novo resistance [range 36-73%] compared to IRs/LTRs [range 4-22%]. Amplification of CDK4/6 was significant in de novo resistance, contrary to IRs and LTRs (91%, 27.9% and 0%, respectively). The presence of co-occurrence CDK4/6 amplification correlated with poor disease outcome with HR of progression-free survival of 3.63 [95% CI 1.80-7.31, p-value < 0.001]. The presence of CDK4/6 amplification in pretreatment specimen serves as a predictive biomarker for de novo resistance in sensitizing EGFR mutation.

GSA: an independent development algorithm for calling copy number and detecting homologous recombination deficiency (HRD) from target capture sequencing

BACKGROUND

The gain or loss of large chromosomal regions or even whole chromosomes is termed as genomic scarring and can be observed as copy number variations resulting from the failure of DNA damage repair.

RESULTS

In this study, a new algorithm called genomic scar analysis (GSA) has developed and validated to calculate homologous recombination deficiency (HRD) score. The two critical submodules were tree recursion (TR) segmentation and filtering, and the estimation and correction of the tumor purity and ploidy. Then, this study evaluated the rationality of segmentation and genotype identification by the GSA algorithm and compared with other two algorithms, PureCN and ASCAT, found that the segmentation result of GSA algorithm was more logical. In addition, the results indicated that the GSA algorithm had an excellent predictive effect on tumor purity and ploidy, if the tumor purity was more than 20%. Furtherly, this study evaluated the HRD scores and BRCA1/2 deficiency status of 195 clinical samples, and the results indicated that the accuracy was 0.98 (comparing with Affymetrix OncoScan™ assay) and the sensitivity was 95.2% (comparing with BRCA1/2 deficiency status), both were well-behaved. Finally, HRD scores and 16 genes mutations (TP53 and 15 HRR pathway genes) were analyzed in 17 cell lines, the results showed that there was higher frequency in HRR pathway genes in high HRD score samples.

CONCLUSIONS

This new algorithm, named as GSA, could effectively and accurately calculate the purity and ploidy of tumor samples through NGS data, and then reflect the degree of genomic instability and large-scale copy number variations of tumor samples.

Genomic Profiling of Premenopausal HR+ and HER2- Metastatic Breast Cancer by Circulating Tumor DNA and Association of Genetic Alterations With Therapeutic Response to Endocrine Therapy and Ribociclib

PURPOSE

This analysis evaluated the genomic landscape of premenopausal patients with hormone receptor–positive and human epidermal growth factor receptor 2–negative advanced breast cancer and the association of genetic alterations with response to ribociclib in the phase III MONALEESA-7 trial.

METHODS

Premenopausal patients were randomly assigned 1:1 to receive endocrine therapy plus ribociclib or placebo. Plasma collected at baseline was sequenced using targeted next-generation sequencing for approximately 600 relevant cancer genes. The association of circulating tumor DNA alterations with progression-free survival (PFS) was evaluated to identify biomarkers of response and resistance to ribociclib.

RESULTS

Baseline circulating tumor DNA was sequenced in 565 patients; 489 had evidence of ≥ 1 alteration. The most frequent alterations included PIK3CA (28%), TP53 (19%), CCND1 (10%), MYC (8%), GATA3 (8%), receptor tyrosine kinases (17%), and the Chr8p11.23 locus (12%). A treatment benefit of ribociclib was seen with wild-type (hazard ratio [HR] 0.45 [95% CI, 0.33 to 0.62]) and altered (HR 0.57 [95% CI, 0.36 to 0.9]) PIK3CA. Overall, patients with altered CCND1 had shorter PFS regardless of treatment, suggesting CCND1 as a potential prognostic biomarker. Benefit with ribociclib was seen in patients with altered (HR 0.21 [95% CI, 0.08 to 0.54]) or wild-type (HR 0.52 [95% CI, 0.39 to 0.68]) CCND1, but greater benefit was observed with altered, suggesting predictive potential of CCND1. Alterations in TP53, MYC, Chr8p11.23 locus, and receptor tyrosine kinases were associated with worse PFS, but ribociclib benefit was independent of alteration status.

CONCLUSION

In this study—to our knowledge, the first large study of premenopausal patients with hormone receptor–positive and human epidermal growth factor receptor 2–negative advanced breast cancer—multiple genomic alterations were associated with poor outcome. A PFS benefit of ribociclib was observed regardless of gene alteration status, although in this exploratory analysis, a magnitude of benefits varied by alteration.

Comparison of the multigene panel test and OncoScan™ for the determination of HER2 amplification in breast cancer

The diagnostic accuracy of the multigene panel test (MPT) and OncoScan™ in the determination of HER2 amplification in breast tumors remains controversial. In the present study, HER2 copy number was analyzed using both MPT and OncoScan™ in 45 breast tumors and was compared with that in fluorescent in situ hybridization (FISH) analysis. Tumors with low cellularity were examined using tumor cell enrichment and fluorescence‑activated cell sorting. Both MPT and OncoScan™ exhibited significant correlations with FISH with respect to the determination of HER2 amplification in breast tumors. However, the correlation coefficient was significantly higher for the comparison of MPT and FISH (r=0.770) compared with that between OncoScan™ and FISH (r=0.564). The accuracy of MPT (93.3%) was slightly higher compared with that in OncoScan™ (84.4%) in determining the HER2 status, which was mostly explained by the higher sensitivity of MPT in tumors with low cellularity (83.3 vs. 33.3%), but not in those with high cellularity (81.8 vs. 72.7%). The specificity was 100% for both tests. The MPT exhibited higher sensitivity in the determination of the amplification of other genes, including MYC, fibroblast growth factor receptor 1 and GATA binding protein 3 in tumors with low cellularity compared with that in tumors with high cellularity. OncoScan™ exhibited low sensitivity without tumor cell enrichment. The results suggested that MPT could be a promising method to determine HER2 status in breast tumors and that it could exhibit improved accuracy compared with that in OncoScan™ in tumors with low cellularity.

A BAYESIAN NONPARAMETRIC MODEL FOR INFERRING SUBCLONAL POPULATIONS FROM STRUCTURED DNA SEQUENCING DATA

There are distinguishing features or "hallmarks" of cancer that are found across tumors, individuals, and types of cancer, and these hallmarks can be driven by specific genetic mutations. Yet, within a single tumor there is often extensive genetic heterogeneity as evidenced by single-cell and bulk DNA sequencing data. The goal of this work is to jointly infer the underlying genotypes of tumor subpopulations and the distribution of those subpopulations in individual tumors by integrating single-cell and bulk sequencing data. Understanding the genetic composition of the tumor at the time of treatment is important in the personalized design of targeted therapeutic combinations and monitoring for possible recurrence after treatment. We propose a hierarchical Dirichlet process mixture model that incorporates the correlation structure induced by a structured sampling arrangement and we show that this model improves the quality of inference. We develop a representation of the hierarchical Dirichlet process prior as a Gamma-Poisson hierarchy and we use this representation to derive a fast Gibbs sampling inference algorithm using the augment-and-marginalize method. Experiments with simulation data show that our model outperforms standard numerical and statistical methods for decomposing admixed count data. Analyses of real acute lymphoblastic leukemia cancer sequencing dataset shows that our model improves upon state-of-the-art bioinformatic methods. An interpretation of the results of our model on this real dataset reveals co-mutated loci across samples.

Technical and biological constraints on ctDNA-based genotyping

Circulating tumor DNA (ctDNA) enables real-time genomic profiling of cancer without the need for tissue biopsy. ctDNA-based technology is seeing rapid uptake in clinical practice due to the potential to inform patient management from diagnosis to advanced disease. In metastatic disease, ctDNA can identify somatic mutations, copy-number variants (CNVs), and structural rearrangements that are predictive of therapy response. However, the ctDNA fraction (ctDNA%) is unpredictable and confounds variant detection strategies, undermining confidence in liquid biopsy results. Assay design also influences which types of genomic alterations are identifiable. Here, we describe the relationships between ctDNA%, methodology, and sensitivity-specificity for major classes of genomic alterations in prostate cancer. We provide recommendations to navigate the technical complexities that constrain the detection of clinically relevant genomic alterations in ctDNA.

BRN2 is a non-canonical melanoma tumor-suppressor

While the major drivers of melanoma initiation, including activation of NRAS/BRAF and loss of PTEN or CDKN2A, have been identified, the role of key transcription factors that impose altered transcriptional states in response to deregulated signaling is not well understood. The POU domain transcription factor BRN2 is a key regulator of melanoma invasion, yet its role in melanoma initiation remains unknown. Here, in a BrafV600E PtenF/+ context, we show that BRN2 haplo-insufficiency promotes melanoma initiation and metastasis. However, metastatic colonization is less efficient in the absence of Brn2. Mechanistically, BRN2 directly induces PTEN expression and in consequence represses PI3K signaling. Moreover, MITF, a BRN2 target, represses PTEN transcription. Collectively, our results suggest that on a PTEN heterozygous background somatic deletion of one BRN2 allele and temporal regulation of the other allele elicits melanoma initiation and progression.

reconCNV: interactive visualization of copy number data from high-throughput sequencing

SUMMARY

Copy number variation (CNV) is an important category of unbalanced structural rearrangement. While methods for detecting CNV in high-throughput targeted sequencing have become increasingly sophisticated, dedicated tools for interactive and dynamic visualization of CNV from these data are still lacking. We describe reconCNV, a tool that produces an interactive and annotated web-based dashboard for viewing and summarizing CNVs detected in next-generation sequencing (NGS) data. reconCNV is designed to work with delimited result files from most NGS CNV callers with minor adjustments to the configuration file. The reconCNV output is an HTML file that is viewable on any modern web browser, requires no backend server, and can be readily appended to existing analysis pipelines. In addition to a standard CNV track for visualizing relative fold change and absolute copy number, the tool includes an auxiliary variant allele fraction track for visualizing underlying allelic imbalance and loss of heterozygosity. A feature to mask assay-specific technical artifacts and a direct HTML link out to the UCSC Genome Browser are also included to augment the reviewer experience. By providing a light-weight plugin for interactive visualization to existing NGS CNV pipelines, reconCNV can facilitate efficient NGS CNV visualization and interpretation in both research and clinical settings.

AVAILABILITY AND IMPLEMENTATION

The source code and documentation including a tutorial can be accessed at https://github.com/rghu/reconCNV as well as a Docker image at https://hub.docker.com/repository/docker/raghuc1990/reconcnv.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Cell-free DNA captures tumor heterogeneity and driver alterations in rapid autopsies with pre-treated metastatic cancer

In patients with metastatic cancer, spatial heterogeneity of somatic alterations may lead to incomplete assessment of a cancer's mutational profile when analyzing a single tumor biopsy. In this study, we perform sequencing of cell-free DNA (cfDNA) and distinct metastatic tissue samples from ten rapid autopsy cases with pre-treated metastatic cancer. We show that levels of heterogeneity in genetic biomarkers vary between patients but that gene expression signatures representative of the tumor microenvironment are more consistent. Across nine patients with plasma samples available, we are able to detect 62/62 truncal and 47/121 non-truncal point mutations in cfDNA. We observe that mutation clonality in cfDNA is correlated with the number of metastatic lesions in which the mutation is detected and use this result to derive a clonality threshold to classify truncal and non-truncal driver alterations with reasonable specificity. In contrast, mutation truncality is more often incorrectly assigned when studying single tissue samples. Our results demonstrate the utility of a single cfDNA sample relative to that of single tissue samples when treating patients with metastatic cancer.

Distinct mutational backgrounds and clonal architectures implicated prognostic discrepancies in small-cell carcinomas of the esophagus and lung

Small-cell carcinoma of the esophagus (SCCE) is a rare and aggressive cancer. Although several consistent genomic changes were observed previously between SCCE and small-cell lung cancer (SCLC), detailed mutational landscapes revealing discrepancies in genetic underpinnings of tumorigenesis between these two cancers are scarce, and little attention has been paid to answer whether these genetic alterations were related to the prognosis. Herein by performing whole-exome sequencing of 48 SCCE and 64 SCLC tumor samples, respectively we have shown that the number of driver mutations in SCCE was significantly lower than in SCLC (p = 0.0042). In SCCE, 46% of recurrent driver mutations were clonal, which occurred at an early stage during tumorigenesis, while 16 driver mutations were found clonal in SCLC. NOTCH1/3, PIK3CA, and ATM were specifically clonal in SCCE, while TP53 was clonal in SCLC. The total number of clonal mutations differed between two cancers and presented lower in SCCE compared to SCLC (p = 0.0036). Moreover, overall survival (OS) was shorter in patients with higher numbers of clonal mutations for both cancers. In summary, SCCE showed distinct mutational background and clonal architecture compared with SCLC. Organ-specific clonal events revealed different molecular mechanisms underlying tumorigenesis, tumor development, patients’ prognosis, and possible variations in therapeutic outcomes to candidate treatments.

Molecular and morphological changes induced by ivosidenib correlate with efficacy in mutant-IDH1 cholangiocarcinoma

Background: IDH1 mutations occur in approximately 13% of intrahepatic cholangiocarcinomas (IHCCs). The oral, targeted, mutant IDH1 (mIDH1) inhibitor ivosidenib (AG-120) suppresses production of the oncometabolite D-2-hydroxyglutarate, promoting disease stabilization and improved progression-free survival (PFS) in mIDH1 IHCC. Materials & methods: Harnessing matched baseline and on-treatment biopsies, we investigate the potential mechanisms underlying ivosidenib's efficacy. Results: mIDH1 inhibition leads to decreased cytoplasm and expression of hepatocyte lineage markers in patients with prolonged PFS. These findings are accompanied by downregulation of biliary fate, cell cycle progression and AKT pathway activity. Conclusion: Ivosidenib stimulates a hepatocyte differentiation program in mIDH1 IHCC, a phenotype associated with clinical benefit. mIDH1 inhibition could be a paradigm for differentiation-based therapy in solid tumors. Clinical trial registration: NCT02073994 (ClinicalTrials.gov).