Timing the Landmark Events in the Evolution of Clear Cell Renal Cell Cancer: TRACERx Renal.

Clear cell renal cell carcinoma (ccRCC) is characterized by near-universal loss of the short arm of chromosome 3, deleting several tumor suppressor genes. We analyzed whole genomes from 95 biopsies across 33 patients with clear cell renal cell carcinoma. We find hotspots of point mutations in the 5' UTR of TERT, targeting a MYC-MAX-MAD1 repressor associated with telomere lengthening. The most common structural abnormality generates simultaneous 3p loss and 5q gain (36% patients), typically through chromothripsis. This event occurs in childhood or adolescence, generally as the initiating event that precedes emergence of the tumor's most recent common ancestor by years to decades. Similar genomic changes drive inherited ccRCC. Modeling differences in age incidence between inherited and sporadic cancers suggests that the number of cells with 3p loss capable of initiating sporadic tumors is no more than a few hundred. Early development of ccRCC follows well-defined evolutionary trajectories, offering opportunity for early intervention.

Global analysis of shared T cell specificities in human non-small cell lung cancer enables HLA inference and antigen discovery.

To identify disease-relevant T cell receptors (TCRs) with shared antigen specificity, we analyzed 778,938 TCRß chain sequences from 178 non-small cell lung cancer patients using the GLIPH2 (grouping of lymphocyte interactions with paratope hotspots 2) algorithm. We identified over 66,000 shared specificity groups, of which 435 were clonally expanded and enriched in tumors compared to adjacent lung. The antigenic epitopes of one such tumor-enriched specificity group were identified using a yeast peptide-HLA A∗02:01 display library. These included a peptide from the epithelial protein TMEM161A, which is overexpressed in tumors and cross-reactive epitopes from Epstein-Barr virus and E. coli. Our findings suggest that this cross-reactivity may underlie the presence of virus-specific T cells in tumor infiltrates and that pathogen cross-reactivity may be a feature of multiple cancers. The approach and analytical pipelines generated in this work, as well as the specificity groups defined here, present a resource for understanding the T cell response in cancer.

Structural variants drive context-dependent oncogene activation in cancer.

Higher-order chromatin structure is important for the regulation of genes by distal regulatory sequences1,2. Structural variants (SVs) that alter three-dimensional (3D) genome organization can lead to enhancer-promoter rewiring and human disease, particularly in the context of cancer3. However, only a small minority of SVs are associated with altered gene expression4,5, and it remains unclear why certain SVs lead to changes in distal gene expression and others do not. To address these questions, we used a combination of genomic profiling and genome engineering to identify sites of recurrent changes in 3D genome structure in cancer and determine the effects of specific rearrangements on oncogene activation. By analysing Hi-C data from 92 cancer cell lines and patient samples, we identified loci affected by recurrent alterations to 3D genome structure, including oncogenes such as MYC, TERT and CCND1. By using CRISPR-Cas9 genome engineering to generate de novo SVs, we show that oncogene activity can be predicted by using 'activity-by-contact' models that consider partner region chromatin contacts and enhancer activity. However, activity-by-contact models are only predictive of specific subsets of genes in the genome, suggesting that different classes of genes engage in distinct modes of regulation by distal regulatory elements. These results indicate that SVs that alter 3D genome organization are widespread in cancer genomes and begin to illustrate predictive rules for the consequences of SVs on oncogene activation.

The life history of 21 breast cancers.

Cancer evolves dynamically as clonal expansions supersede one another driven by shifting selective pressures, mutational processes, and disrupted cancer genes. These processes mark the genome, such that a cancer's life history is encrypted in the somatic mutations present. We developed algorithms to decipher this narrative and applied them to 21 breast cancers. Mutational processes evolve across a cancer's lifespan, with many emerging late but contributing extensive genetic variation. Subclonal diversification is prominent, and most mutations are found in just a fraction of tumor cells. Every tumor has a dominant subclonal lineage, representing more than 50% of tumor cells. Minimal expansion of these subclones occurs until many hundreds to thousands of mutations have accumulated, implying the existence of long-lived, quiescent cell lineages capable of substantial proliferation upon acquisition of enabling genomic changes. Expansion of the dominant subclone to an appreciable mass may therefore represent the final rate-limiting step in a breast cancer's development, triggering diagnosis.

Massive genomic rearrangement acquired in a single catastrophic event during cancer development.

Cancer is driven by somatically acquired point mutations and chromosomal rearrangements, conventionally thought to accumulate gradually over time. Using next-generation sequencing, we characterize a phenomenon, which we term chromothripsis, whereby tens to hundreds of genomic rearrangements occur in a one-off cellular crisis. Rearrangements involving one or a few chromosomes crisscross back and forth across involved regions, generating frequent oscillations between two copy number states. These genomic hallmarks are highly improbable if rearrangements accumulate over time and instead imply that nearly all occur during a single cellular catastrophe. The stamp of chromothripsis can be seen in at least 2%-3% of all cancers, across many subtypes, and is present in ∼25% of bone cancers. We find that one, or indeed more than one, cancer-causing lesion can emerge out of the genomic crisis. This phenomenon has important implications for the origins of genomic remodeling and temporal emergence of cancer.

Patterns of structural variants within TP53 introns and relocation of the TP53 promoter: a commentary†.

Gene disruption from double-strand DNA breaks within introns is a mechanism of inactivating the tumor suppressor TP53. This occurs more frequently in osteosarcoma and biliary adenocarcinoma compared with other cancer types. The patterns of intron breakpoints within TP53 do not correlate with prevalence, intron length, or overall genome-wide levels of rearrangements. Therefore, these breakpoints appear to be selected for reasons other than to disrupt TP53. A recent article published by Saba et al in The Journal of Pathology illustrates a benefit to having breakpoints within intron 1 using high-quality matched genomic and transcriptomic osteosarcoma sequencing data as well as in vitro validation. The authors describe how the rearrangement results in relocation of the TP53 promoter region to regions upstream of genes that encode members of cartilage, growth plate development, osteoclast formation, and other TP53-related pathways. The upregulation of these genes by the TP53 promoter are gain-of-function events that are likely to promote tumor development and growth. Therefore, this article presents a potential new paradigm in which a single mutation would result in both the loss of a tumor suppressor and the gain of an oncogenic program. © 2024 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.

Epigenomic and transcriptomic profiling of solitary fibrous tumors identifies site-specific patterns and candidate genes regulated by DNA methylation.

Solitary fibrous tumor (SFT) is a rare mesenchymal neoplasm which can arise at any anatomic site and is characterized by recurrent NAB2::STAT6 fusions and metastatic progression in 10-30%. The cell of origin has not been identified. Despite some progress in understanding the contribution of heterogeneous fusion types and secondary mutations to SFT biology, epigenetic alterations in extrameningeal SFT remain largely unexplored, and most sarcoma research to date has focused on the use of methylation profiling for tumor classification. We interrogated genome-wide DNA methylation in 79 SFTs to identify informative epigenetic changes. RNA-seq data from targeted panels and data from the Cancer Genome Atlas (TCGA) were used for orthogonal validation of selected findings. In unsupervised clustering analysis, the top 500 most variable CpGs segregated SFTs by primary anatomic site. Differentially methylated genes (DMGs) associated with primary SFT site included EGFR, TBX15, multiple HOX genes and their cofactors EBF1, EBF3, and PBX1, as well as RUNX1 and MEIS1. Of the 20 DMGs that were interrogated on the RNA-seq panel, twelve were significantly differentially expressed according to site. However, with the exception of TBX15, most of these also showed differential expression according to NAB2::STAT6 fusion type, suggesting that the fusion oncogene contributes to transcriptional regulation of these genes. Transcriptomic data confirmed an inverse correlation between gene methylation and the expression of TBX15 in both SFT and TCGA sarcomas. TBX15 also showed differential mRNA expression and 5' UTR methylation between tumors located in different anatomic sites in TCGA data. In all analyses, TBX15 methylation and mRNA expression retained the strongest association with tissue of origin in SFT and other sarcomas, suggesting a possible marker to distinguish metastatic tumors from new primaries without genomic profiling. Epigenetic signatures may further help to identify SFT progenitor cells at different anatomic sites.

Lenalidomide promotes the development of TP53-mutated therapy-related myeloid neoplasms.

There is a growing body of evidence that therapy-related myeloid neoplasms (t-MNs) with driver gene mutations arise in the background of clonal hematopoiesis (CH) under the positive selective pressure of chemo- and radiation therapies. Uncovering the exposure relationships that provide selective advantage to specific CH mutations is critical to understanding the pathogenesis and etiology of t-MNs. In a systematic analysis of 416 patients with t-MN and detailed prior exposure history, we found that TP53 mutations were significantly associated with prior treatment with thalidomide analogs, specifically lenalidomide. We demonstrated experimentally that lenalidomide treatment provides a selective advantage to Trp53-mutant hematopoietic stem and progenitor cells (HSPCs) in vitro and in vivo, the effect of which was specific to Trp53-mutant HSPCs and was not observed in HSPCs with other CH mutations. Because of the differences in CK1α degradation, pomalidomide treatment did not provide an equivalent level of selective advantage to Trp53-mutant HSPCs, providing a biological rationale for its use in patients at high risk for t-MN. These findings highlight the role of lenalidomide treatment in promoting TP53-mutated t-MNs and offer a potential alternative strategy to mitigate the risk of t-MN development.

Analysis of Immune Intratumor Heterogeneity Highlights Immunoregulatory and Coinhibitory Lymphocytes as Hallmarks of Recurrence in Stage I Non-Small Cell Lung Cancer.

Our understanding of the molecular mechanisms underlying postsurgical recurrence of non-small cell lung cancer (NSCLC) is rudimentary. Molecular and T cell repertoire intratumor heterogeneity (ITH) have been reported to be associated with postsurgical relapse; however, how ITH at the cellular level impacts survival is largely unknown. Here we report the analysis of 2880 multispectral images representing 14.2% to 27% of tumor areas from 33 patients with stage I NSCLC, including 17 cases (relapsed within 3 years after surgery) and 16 controls (without recurrence ≥5 years after surgery) using multiplex immunofluorescence. Spatial analysis was conducted to quantify the minimum distance between different cell types and immune cell infiltration around malignant cells. Immune ITH was defined as the variance of immune cells from 3 intratumor regions. We found that tumors from patients having relapsed display different immune biology compared with nonrecurrent tumors, with a higher percentage of tumor cells and macrophages expressing PD-L1 (P =.031 and P =.024, respectively), along with an increase in regulatory T cells (Treg) (P =.018), antigen-experienced T cells (P =.025), and effector-memory T cells (P =.041). Spatial analysis revealed that a higher level of infiltration of PD-L1+ macrophages (CD68+PD-L1+) or antigen-experienced cytotoxic T cells (CD3+CD8+PD-1+) in the tumor was associated with poor overall survival (P =.021 and P =.006, respectively). A higher degree of Treg ITH was associated with inferior recurrence-free survival regardless of tumor mutational burden (P =.022), neoantigen burden (P =.021), genomic ITH (P =.012) and T cell repertoire ITH (P =.001). Using multiregion multiplex immunofluorescence, we characterized ITH at the immune cell level along with whole exome and T cell repertoire sequencing from the same tumor regions. This approach highlights the role of immunoregulatory and coinhibitory signals as well as their spatial distribution and ITH that define the hallmarks of tumor relapse of stage I NSCLC.

Gut microbiome features associated with liver fibrosis in Hispanics, a population at high risk for fatty liver disease.

BACKGROUND AND AIMS: Hispanics are disproportionately affected by NAFLD, liver fibrosis, cirrhosis, and HCC. Preventive strategies and noninvasive means to identify those in this population at high risk for liver fibrosis, are urgently needed. We aimed to characterize the gut microbiome signatures and related biological functions associated with liver fibrosis in Hispanics and identify environmental and genetic factors affecting them. APPROACH AND RESULTS: Subjects of the population-based Cameron County Hispanic Cohort (CCHC; n = 217) were screened by vibration-controlled transient elastography (FibroScan). Among them, 144 (66.7%) had steatosis and 28 (13.0%) had liver fibrosis. The gut microbiome of subjects with liver fibrosis was enriched with immunogenic commensals (e.g., Prevotella copri, Holdemanella, Clostridiaceae 1) and depleted of Bacteroides caccae, Parabacteroides distasonis, Enterobacter, and Marinifilaceae. The liver fibrosis-associated metagenome was characterized by changes in the urea cycle, L-citrulline biosynthesis and creatinine degradation pathways, and altered synthesis of B vitamins and lipoic acid. These metagenomic changes strongly correlated with the depletion of Parabacteroides distasonis and enrichment of Prevotella and Holdemanella. Liver fibrosis was also associated with depletion of bacterial pathways related to L-fucose biosynthesis. Alcohol consumption, even moderate, was associated with high Prevotella abundance. The single-nucleotide polymorphisms rs3769502 and rs7573751 in the NCK adaptor protein 2 (NCK2) gene positively associated with high Prevotella abundance. CONCLUSION: Hispanics with liver fibrosis display microbiome profiles and associated functional changes that may promote oxidative stress and a proinflammatory environment. These microbiome signatures, together with NCK2 polymorphisms, may have utility in risk modeling and disease prevention in this high-risk population.

Clonal dynamics and clinical implications of postremission clonal hematopoiesis in acute myeloid leukemia.

Although clonal hematopoiesis (CH) can precede the development of acute myeloid leukemia (AML), it can also persist after achieving remission. Long-term clonal dynamics and clinical implications of persistent CH are not well understood. Here, we studied the prevalence, dynamics, and clinical implications of postremission CH in 164 AML patients who attained complete remission after induction chemotherapies. Postremission CH was identified in 79 (48%) patients. Postremission CH persisted long term in 91% of the trackable patients despite treatment with various types of consolidation and maintenance therapies. Postremission CH was eradicated in 20 out of 21 (95%) patients who underwent allogeneic stem cell transplant. Although patients with postremission CH as a group had comparable hematopoiesis with those without it, patients with persistent TET2 mutations showed significant neutropenia long term. Postremission CH had little impact on relapse risk, nonrelapse mortality, and incidence of atherosclerotic cardiovascular disease, although the clinical impact of post-CR CH was heterogeneous among different mutations. These data suggest that although residual clonal hematopoietic stem cells are generally resistant to consolidation and maintenance therapies, they retain the ability to maintain normal hematopoiesis and have little impact on clinical outcomes. Larger study is needed to dissect the gene-specific heterogeneity.

Integrated multi-omic analysis of low-grade ovarian serous carcinoma collected from short and long-term survivors.

BACKGROUND: Low-grade serous ovarian cancer (LGSOC) is a rare disease that occurs more frequently in younger women than those with high-grade disease. The current treatment is suboptimal and a better understanding of the molecular pathogenesis of this disease is required. In this study, we compared the proteogenomic analyses of LGSOCs from short- and long-term survivors (defined as < 40 and > 60 months, respectively). Our goal was to identify novel mutations, proteins, and mRNA transcripts that are dysregulated in LGSOC, particularly in short-term survivors. METHODS: Initially, targeted sequencing of 409 cancer-related genes was performed on 22 LGSOC and 6 serous borderline ovarian tumor samples. Subsequently, whole-genome sequencing analysis was performed on 14 LGSOC samples (7 long-term survivors and 7 short-term survivors) with matched normal tissue samples. RNA sequencing (RNA-seq), quantitative proteomics, and phosphoproteomic analyses were also performed. RESULTS: We identified single-nucleotide variants (SNVs) (range: 5688-14,833 per sample), insertion and deletion variants (indels) (range: 880-1065), and regions with copy number variants (CNVs) (range: 62-335) among the 14 LGSOC samples. Among all SNVs and indels, 2637 mutation sites were found in the exonic regions. The allele frequencies of the detected variants were low (median12%). The identified recurrent nonsynonymous missense mutations included KRAS, NRAS, EIF1AX, UBR5, and DNM3 mutations. Mutations in DNM3 and UBR5 have not previously been reported in LGSOC. For the two samples, somatic DNM3 nonsynonymous missense mutations in the exonic region were validated using Sanger sequencing. The third sample contained two missense mutations in the intronic region of DNM3, leading to a frameshift mutation detected in RNA transcripts in the RNA-seq data. Among the 14 LGSOC samples, 7754 proteins and 9733 phosphosites were detected by global proteomic analysis. Some of these proteins and signaling pathways, such as BST1, TBXAS1, MPEG1, HBA1, and phosphorylated ASAP1, are potential therapeutic targets. CONCLUSIONS: This is the first study to use whole-genome sequencing to detect somatic mutations in LGSOCs with matched normal tissues. We detected and validated novel mutations in DNM3, which were present in 3 of the 14 samples analyzed. Additionally, we identified novel indels, regions with CNVs, dysregulated mRNA, dysregulated proteins, and phosphosites that are more prevalent in short-term survivors. This integrated proteogenomic analysis can guide research into the pathogenesis and treatment of LGSOC.

Statistical tests for intra-tumour clonal co-occurrence and exclusivity.

Tumour progression is an evolutionary process in which different clones evolve over time, leading to intra-tumour heterogeneity. Interactions between clones can affect tumour evolution and hence disease progression and treatment outcome. Intra-tumoural pairs of mutations that are overrepresented in a co-occurring or clonally exclusive fashion over a cohort of patient samples may be suggestive of a synergistic effect between the different clones carrying these mutations. We therefore developed a novel statistical testing framework, called GeneAccord, to identify such gene pairs that are altered in distinct subclones of the same tumour. We analysed our framework for calibration and power. By comparing its performance to baseline methods, we demonstrate that to control type I errors, it is essential to account for the evolutionary dependencies among clones. In applying GeneAccord to the single-cell sequencing of a cohort of 123 acute myeloid leukaemia patients, we find 1 clonally co-occurring and 8 clonally exclusive gene pairs. The clonally exclusive pairs mostly involve genes of the key signalling pathways.

Genomic profiling of multifocal intrahepatic cholangiocarcinoma reveals intraindividual concordance of genetic alterations.

In multifocal intrahepatic cholangiocarcinoma (IHC), intrahepatic metastases (IM) represent a contraindication to surgical resection, whereas satellite nodules (SN) do not. However, no consensus criteria exist to distinguish IM from SN. The purpose of this study was to determine genetic alterations and clonal relationships in surgically resected multifocal IHC. Next-generation sequencing of 34 spatially separated IHC tumors was performed using a targeted panel of 201 cancer-associated genes. Proposed definitions in the literature were applied of SN located in the same liver segment and ≤2 cm from the primary tumor; and IM located in a different liver segment and/or >2 cm from the primary tumor. Somatic point mutations concordant across tumors from individual patients included BAP1, SMARCA4 and IDH1. Small insertions and deletions (indels) present at the same genome positions among all tumors from individuals included indels in DNA repair genes, CHEK1, ERCC5, ATR and MSH6. Copy number alterations were also similar between all tumors in each patient. In this cohort of multifocal IHC, genomic profiles were concordant across all tumors in each patient, suggesting a common progenitor cell origin, regardless of the location of tumors in the liver. The decision to perform surgery should not be based upon a perceived distinction between IM and SN.

High-throughput single-cell DNA sequencing of acute myeloid leukemia tumors with droplet microfluidics.

To enable the characterization of genetic heterogeneity in tumor cell populations, we developed a novel microfluidic approach that barcodes amplified genomic DNA from thousands of individual cancer cells confined to droplets. The barcodes are then used to reassemble the genetic profiles of cells from next-generation sequencing data. By using this approach, we sequenced longitudinally collected acute myeloid leukemia (AML) tumor populations from two patients and genotyped up to 62 disease relevant loci across more than 16,000 individual cells. Targeted single-cell sequencing was able to sensitively identify cells harboring pathogenic mutations during complete remission and uncovered complex clonal evolution within AML tumors that was not observable with bulk sequencing. We anticipate that this approach will make feasible the routine analysis of AML heterogeneity, leading to improved stratification and therapy selection for the disease.

PRDM16s transforms megakaryocyte-erythroid progenitors into myeloid leukemia-initiating cells.

Oncogenic mutations confer on cells the ability to propagate indefinitely, but whether oncogenes alter the cell fate of these cells is unknown. Here, we show that the transcriptional regulator PRDM16s causes oncogenic fate conversion by transforming cells fated to form platelets and erythrocytes into myeloid leukemia stem cells (LSCs). Prdm16s expression in megakaryocyte-erythroid progenitors (MEPs), which normally lack the potential to generate granulomonocytic cells, caused AML by converting MEPs into LSCs. Prdm16s blocked megakaryocytic/erythroid potential by interacting with super enhancers and activating myeloid master regulators, including PU.1. A CRISPR dropout screen confirmed that PU.1 is required for Prdm16s-induced leukemia. Ablating PU.1 attenuated leukemogenesis and reinstated the megakaryocytic/erythroid potential of leukemic MEPs in mouse models and human AML with PRDM16 rearrangement. Thus, oncogenic PRDM16 s expression gives MEPs an LSC fate by activating myeloid gene regulatory networks.

Enhancer reprogramming in PRC2-deficient malignant peripheral nerve sheath tumors induces a targetable de-differentiated state.

Malignant peripheral nerve sheath tumors (MPNSTs) are soft tissue sarcomas that frequently harbor genetic alterations in polycomb repressor complex 2 (PRC2) components-SUZ12 and EED. Here, we show that PRC2 loss confers a dedifferentiated early neural-crest phenotype which is exclusive to PRC2-mutant MPNSTs and not a feature of neurofibromas. Neural crest phenotype in PRC2 mutant MPNSTs was validated via cross-species comparative analysis using spontaneous and transgenic MPNST models. Systematic chromatin state profiling of the MPNST cells showed extensive epigenomic reprogramming or chromatin states associated with PRC2 loss and identified gains of active enhancer states/super-enhancers on early neural crest regulators in PRC2-mutant conditions around genomic loci that harbored repressed/poised states in PRC2-WT MPNST cells. Consistently, inverse correlation between H3K27me3 loss and H3K27Ac gain was noted in MPNSTs. Epigenetic editing experiments established functional roles for enhancer gains on DLX5-a key regulator of neural crest phenotype. Consistently, blockade of enhancer activity by bromodomain inhibitors specifically suppressed this neural crest phenotype and tumor burden in PRC2-mutant PDXs. Together, these findings reveal accumulation of dedifferentiated neural crest like state in PRC2-mutant MPNSTs that can be targeted by enhancer blockade.

Somatic Mutations in Circulating Cell-Free DNA and Risk for Hepatocellular Carcinoma in Hispanics.

Hispanics are disproportionally affected by liver fibrosis and hepatocellular carcinoma (HCC). Advanced liver fibrosis is a major risk factor for HCC development. We aimed at identifying somatic mutations in plasma cell-free DNA (cfDNA) of Hispanics with HCC and Hispanics with advanced liver fibrosis but no HCC. Targeted sequencing of over 262 cancer-associated genes identified nonsynonymous mutations in 22 of the 27 HCC patients. Mutations were detected in known HCC-associated genes (e.g., CTNNB1, TP53, NFE2L2, and ARID1A). No difference in cfDNA concentrations was observed between patients with mutations and those without detectable mutations. HCC patients with higher cfDNA concentrations or higher number of mutations had a shorter overall survival (p < 0.001 and p = 0.045). Nonsynonymous mutations were also identified in 17 of the 51 subjects with advanced liver fibrosis. KMT2C was the most commonly mutated gene. Nine genes were mutated in both subjects with advanced fibrosis and HCC patients. Again, no significant difference in cfDNA concentrations was observed between subjects with mutations and those without detectable mutations. Furthermore, higher cfDNA concentrations and higher number of mutations correlated with a death outcome in subjects with advanced fibrosis. In conclusion, cfDNA features are promising non-invasive markers for HCC risk prediction and overall survival.

Clinical implications of cancer gene mutations in patients with chronic lymphocytic leukemia treated with lenalidomide.

Lenalidomide is clinically active in chronic lymphocytic leukemia (CLL), but its effectiveness in the context of the CLL mutational landscape is unknown. We performed targeted capture sequencing of 295 cancer genes in specimens from 102 CLL patients with treatment-naïve disease (TN patients) and 186 CLL patients with relapsed/refractory disease (R/R patients) who received lenalidomide-based therapy at our institution. The most frequently mutated gene was SF3B1 (15%), followed by NOTCH1 (14%) and TP53 (14%), with R/R patients having significantly more TP53 mutations than did TN patients. Among all lenalidomide-treated patients, del(17p) (P ≤ .001), del(11q) (P = .032), and complex karyotype (P = .022), along with mutations in TP53 (P ≤ .001), KRAS (P = .034), and DDX3X (P ≤ .001), were associated with worse overall response (OR). R/R patients with SF3B1 and MGA mutations had significantly worse OR (P = .025 and .035, respectively). TN and R/R patients with del(17p) and TP53 mutations had worse overall survival (OS) and progression-free survival (PFS). In R/R patients, complex karyotype and SF3B1 mutations were associated with worse OS and PFS; DDX3X mutations were associated with worse PFS only. Weibull regression multivariate analysis revealed that TP53 aberrations (del(17p), TP53 mutation, or both), along with complex karyotype and SF3B1 mutations, were associated with worse OS in the R/R cohort. Taken together, cancer gene mutations in CLL contribute to the already comprehensive risk stratification and add to prognosis and response to treatment. The related trials were registered at www.clinicaltrials.gov as #NCT00267059, #NCT00535873, #NCT00759603, #NCT01446133, and #NCT01002755.

Applying Artificial Intelligence to Address the Knowledge Gaps in Cancer Care.

BACKGROUND: Rapid advances in science challenge the timely adoption of evidence-based care in community settings. To bridge the gap between what is possible and what is practiced, we researched approaches to developing an artificial intelligence (AI) application that can provide real-time patient-specific decision support. MATERIALS AND METHODS: The Oncology Expert Advisor (OEA) was designed to simulate peer-to-peer consultation with three core functions: patient history summarization, treatment options recommendation, and management advisory. Machine-learning algorithms were trained to construct a dynamic summary of patients cancer history and to suggest approved therapy or investigative trial options. All patient data used were retrospectively accrued. Ground truth was established for approximately 1,000 unique patients. The full Medline database of more than 23 million published abstracts was used as the literature corpus. RESULTS: OEA's accuracies of searching disparate sources within electronic medical records to extract complex clinical concepts from unstructured text documents varied, with F1 scores of 90%-96% for non-time-dependent concepts (e.g., diagnosis) and F1 scores of 63%-65% for time-dependent concepts (e.g., therapy history timeline). Based on constructed patient profiles, OEA suggests approved therapy options linked to supporting evidence (99.9% recall; 88% precision), and screens for eligible clinical trials on ClinicalTrials.gov (97.9% recall; 96.9% precision). CONCLUSION: Our results demonstrated technical feasibility of an AI-powered application to construct longitudinal patient profiles in context and to suggest evidence-based treatment and trial options. Our experience highlighted the necessity of collaboration across clinical and AI domains, and the requirement of clinical expertise throughout the process, from design to training to testing. IMPLICATIONS FOR PRACTICE: Artificial intelligence (AI)-powered digital advisors such as the Oncology Expert Advisor have the potential to augment the capacity and update the knowledge base of practicing oncologists. By constructing dynamic patient profiles from disparate data sources and organizing and vetting vast literature for relevance to a specific patient, such AI applications could empower oncologists to consider all therapy options based on the latest scientific evidence for their patients, and help them spend less time on information "hunting and gathering" and more time with the patients. However, realization of this will require not only AI technology maturation but also active participation and leadership by clincial experts.