Biology subjects

Van Allen, E.

Publications and source records attributed to Van Allen, E..

25 records · Page 2Linked to original sources

bioRxiv2021

Molecular features of exceptional response to neoadjuvant anti-androgen therapy in high-risk localized prostate cancer

High-risk localized prostate cancer (HRLPC) is associated with a substantial risk of recurrence and prostate cancer-specific mortality1. Recent clinical trials have shown that intensifying anti-androgen therapies administered prior to prostatectomy can induce pathologic complete responses (pCR) or minimal residual disease (MRD) (<5 mm), together termed exceptional response, although the molecular determinants of these clinical outcomes are largely unknown. Here, we performed whole exome (WES) and whole transcriptome sequencing (RNA-seq) on pre-treatment multi-regional tumor biopsies from exceptional responders (ER: pCR and MRD patients) and non-responders (NR: pathologic T3 or lymph node positive disease) treated with intensive anti-androgen therapies prior to prostatectomy. SPOP mutation and SPOPL copy number loss were exclusively observed in ER, while TP53 mutation and PTEN copy number loss were exclusively observed in NR. These alterations were clonal in all tumor phylogenies per patient. Additionally, transcriptional programs involving androgen signaling and TGF{beta} signaling were enriched in ER and NR, respectively. The presence of these alterations in routine biopsies from patients with HRLPC may inform the prospective identification of responders to neoadjuvant anti-androgen therapies to improve clinical outcomes and stratify other patients to alternative biologically informed treatment strategies.

cancer biology↗

bioRxiv2020

Biologically informed deep neural network for prostate cancer classification and discovery

Determination of molecular features that mediate clinically aggressive phenotypes in prostate cancer (PrCa) remains a major biological and clinical challenge. Here, we developed a biologically informed deep learning model (P-NET) to stratify PrCa patients by treatment resistance state and evaluate molecular drivers of treatment resistance for therapeutic targeting through complete model interpretability. Using a molecular cohort of 1,238 prostate cancers, we demonstrated that P-NET can predict cancer state using molecular data that is superior to other modeling approaches. Moreover, the biological interpretability within P-NET revealed established and novel molecularly altered candidates, such as MDM4 and FGFR1, that were implicated in predicting advanced disease and validated in vitro. Broadly, biologically informed fully interpretable neural networks enable preclinical discovery and clinical prediction in prostate cancer and may have general applicability across cancer types.

bioinformatics↗

bioRxiv2020

Subtype Heterogeneity and Epigenetic Convergence in Neuroendocrine Prostate Cancer

Neuroendocrine carcinomas (NEC) are tumors expressing markers of neuronal differentiation that can arise at different anatomic sites but have strong histological and clinical similarities. Here we report the chromatin landscapes of a range of human NECs and show convergence to the activation of a common epigenetic program. With a particular focus on treatment emergent neuroendocrine prostate cancer (NEPC), we analyzed cell lines, patient-derived xenograft (PDX) models and human clinical samples to show the existence of two distinct NEPC subtypes based on the expression of the neuronal transcription factors ASCL1 and NEUROD1. While in cell lines and PDX models these subtypes are mutually exclusive, single cell analysis of human clinical samples exhibit a more complex tumor structure with subtypes coexisting as separate sub-populations within the same tumor. These tumor sub-populations differ genetically and epigenetically contributing to intra- and inter-tumoral heterogeneity in human metastases. Overall our results provide a deeper understanding of the shared clinicopathological characteristics shown by NECs. Furthermore, the intratumoral heterogeneity of human NEPCs suggests the requirement of simultaneous targeting of coexisting tumor populations as a therapeutic strategy.

cancer biology↗

bioRxiv2020

Systematic auditing is essential to debiasing machine learning in biology

Representational biases that are common in biological data can inflate prediction performance and confound our understanding of how and what machine learning (ML) models learn from large complicated datasets. However, auditing for these biases is not a common practice in ML in the life sciences. Here, we devise a systematic auditing framework and harness it to audit three different ML applications of significant therapeutic interest: prediction frameworks of protein-protein interactions, drug-target bioactivity, and MHC-peptide binding. Through this, we identify unrecognized biases that hinder the ML process and result in low model generalizability. Ultimately, we show that, when there is insufficient signal in the training data, ML models are likely to learn primarily from representational biases.

systems biology↗

bioRxiv2018

Scaling computational genomics to millions of individuals with GPUs

Current genomics methods were designed to handle tens to thousands of samples, but will soon need to scale to millions to keep up with the pace of data and hypothesis generation in biomedical science. Moreover, costs associated with processing these growing datasets will become prohibitive without improving the computational efficiency and scalability of methods. Here, we show that recently developed machine-learning libraries (TensorFlow and PyTorch) facilitate implementation of genomics methods for GPUs and significantly accelerate computations. To demonstrate this, we re-implemented methods for two commonly performed computational genomics tasks: QTL mapping and Bayesian non-negative matrix factorization. Our implementations ran > 200 times faster than current CPU-based versions, and these analyses are [~]5-10 fold cheaper on GPUs due to the vastly shorter runtimes. We anticipate that the accessibility of these libraries, and the improvements in run-time will lead to a transition to GPU-based implementations for a wide range of computational genomics methods.

genomics↗

bioRxiv2018

Inherited DNA Repair Defects in Colorectal Cancer

Colorectal cancer (CRC) heritability has been estimated to be around 30%. However, mutations in the known CRC susceptibility genes explain CRC risk in under 10% of the cases. Germline mutations in DNA-repair genes (DRGs) have recently been reported in CRC but their contribution to CRC risk is largely unknown. We evaluated the gene-level germline mutation enrichment of 40 DRGs in 680 unselected CRC individuals compared to 27728 ancestry-matched cancer-free adults. Significant findings were then examined in independent cohorts of 1661 unselected CRC cases and 1456 early-onset CRC cases. Of 680 individuals in the discovery set, 31 (4.56%) individuals harbored germline pathogenic mutations in known CRC susceptibility genes while another 33 (4.85%) individuals had DRG mutations that have not been previously associated with CRC risk. Germline pathogenic mutations in ATM and PALB2 were enriched in both the discovery (OR= 2.81; P= 0.035 and OR= 4.91; P= 0.024, respectively) and validation sets (OR= 2.97; Adjusted P= 0.0013 and OR= 3.42; Adjusted P= 0.034, for ATM and PALB2 respectively). Biallelic loss of ATM was evident in all cases with matched tumor profiling. CRC cases also had higher rates of actionable mutations in the HR pathway that can substantially increase the risk of developing cancers other than CRC. Our analysis provides evidence for ATM and PALB2 as CRC risk genes, underscoring the importance of the homologous recombination pathway in CRC. In addition, we identified frequent complete homologous recombination deficiency in CRC tumors, representing a unique opportunity to explore targeted therapeutic interventions such as PARPi.

genetics↗

bioRxiv2016

Somatic Mutations and Neoepitope Homology in Melanomas Treated with CTLA-4 Blockade

Immune checkpoint inhibitors are promising treatments for patients with a variety of malignancies. Toward understanding the determinants of response to immune checkpoint inhibitors, it was previously demonstrated that somatic mutation burden is associated with benefit and a hypothesis was posited that neoantigen homology to pathogens may in part explain the link between somatic mutations and response. To further examine this hypothesis, we reanalyzed cancer exome data obtained from a previously published study of 64 melanoma patients treated with CTLA-4 blockade and a new dataset of RNA-Seq data from 24 of those patients. We found that the predictive accuracy does not increase as analysis narrows from somatic mutation burden to predicted MHC Class I neoantigens, expressed neoantigens, or homology to pathogens. Further, the association between somatic mutation burden and response is only found when examining samples obtained prior to treatment. Neoantigen and expressed neoantigen burden are also associated with response, but neither is more predictive than somatic mutation burden. Neither the previously-described tetrapeptide signature nor an updated method to evaluate neoepitope homology to pathogens were more predictive than mutation burden.

cancer biology↗