e13580
Background: Malignant Pleural mesothelioma (MPM) is a highly aggressive cancer of the pleural surface and represents 80-90% of mesothelioma diagnosis. MPM is broadly subclassified into three ...histological subgroups (epithelioid, sarcomatoid, biphasic), however tissue heterogeneity has resulted in diagnostic challenges and suboptimal patient care. There are currently no specific histological markers of high/low-risk MPM patients, which is critical in predicting patient prognosis. Methods: Owkin developed MesoNet, a deep learning model that predicts overall survival (OS) of MPM patients from whole slide images (WSI) and trained on the French MESOBANK and TCGA cohorts (Courtiol et al, 2019). In this study, we sought to validate MesoNet’s performance on an independent cohort of 127 WSI stained with haematoxylin/eosin from MPM patients collected at the University of Pittsburgh as part of the National Mesothelioma Virtual Bank (funding by U24OH009077). Patient demographics, survival data, expertly curated pathology annotations were also collected. Results: Our analyses showed that MesoNet predicted OS as risk score based on WSI, which validated high-risk MPM patients exhibited poorer OS, as compared to low-risk patients. Analyses on histological subtypes revealed sarcomatoid and biphasic patients were overrepresented in high-risk groups, as compared to epithelioid patients, which correlates with observed OS data. Notably, histological features associated with high-risk patients revealed tumor pleomorphism and anaplastic nuclear features, whereas low-risk tiles appear to be enriched in tumor infiltrating lymphocytes (TILs) with accompanying stromal proliferation and dense fibrosis. Conclusions: Collectively, our studies validate MesoNet performance on an independent cohort and identify new features related to MPM risk groups, which may inform future treatment stratification and personalization of immunotherapies.
Although the majority of patients with acute myeloid leukemia (AML) initially respond to chemotherapy, many of them subsequently relapse, and the mechanistic basis for AML persistence following ...chemotherapy has not been determined. Recurrent somatic mutations in DNA methyltransferase 3A (DNMT3A), most frequently at arginine 882 (DNMT3A
), have been observed in AML and in individuals with clonal hematopoiesis in the absence of leukemic transformation. Patients with DNMT3A
AML have an inferior outcome when treated with standard-dose daunorubicin-based induction chemotherapy, suggesting that DNMT3A
cells persist and drive relapse. We found that Dnmt3a mutations induced hematopoietic stem cell expansion, cooperated with mutations in the FMS-like tyrosine kinase 3 gene (Flt3
) and the nucleophosmin gene (Npm1
) to induce AML in vivo, and promoted resistance to anthracycline chemotherapy. In patients with AML, the presence of DNMT3A
mutations predicts minimal residual disease, underscoring their role in AML chemoresistance. DNMT3A
cells showed impaired nucleosome eviction and chromatin remodeling in response to anthracycline treatment, which resulted from attenuated recruitment of histone chaperone SPT-16 following anthracycline exposure. This defect led to an inability to sense and repair DNA torsional stress, which resulted in increased mutagenesis. Our findings identify a crucial role for DNMT3A
mutations in driving AML chemoresistance and highlight the importance of chromatin remodeling in response to cytotoxic chemotherapy.
Mutations in epigenetic modifiers and signaling factors often co-occur in myeloid malignancies, including TET2 and NRAS mutations. Concurrent Tet2 loss and NrasG12D expression in hematopoietic cells ...induced myeloid transformation, with a fully penetrant, lethal chronic myelomonocytic leukemia (CMML), which was serially transplantable. Tet2 loss and Nras mutation cooperatively led to decrease in negative regulators of mitogen-activated protein kinase (MAPK) activation, including Spry2, thereby causing synergistic activation of MAPK signaling by epigenetic silencing. Tet2/Nras double-mutant leukemia showed preferential sensitivity to MAPK kinase (MEK) inhibition in both mouse model and patient samples. These data provide insights into how epigenetic and signaling mutations cooperate in myeloid transformation and provide a rationale for mechanism-based therapy in CMML patients with these high-risk genetic lesions.
Display omitted
•Tet2−/− and NrasG12D alleles cooperate to induce myeloid transformation•Tet2 loss and Nras mutation synergistically activate Ras signaling•Tet2−/− and NrasG12D cooperatively suppress Dusp/Spry regulators including Spry2•TET2/NRAS co-mutant leukemia is highly sensitive to MEK inhibition
Kunimoto et al. show that concurrent Tet2 loss and NrasG12D expression in hematopoietic cells induces fully penetrant, lethal chronic myelomonocytic leukemia by decreasing negative regulators of MAPK activation. Mouse and human TET2/NRAS double-mutant leukemia show preferential sensitivity to MEK inhibition.
Abstract
Today, pathology imaging is one of the most common and inexpensive diagnostic/prognostic tools used in oncology, while more sophisticated methods such as next generation sequencing (NGS) ...remain relatively expensive and not routinely used in a clinical setting. Deep convolutional neural networks (CNNs) have emerged as an important image analysis technology enhancing the workflow of pathologists and improving the prediction of patient prognosis and response to treatment. Recently, a few attempts have been made to predict molecular features from tissue imaging using CNNs. While these preliminary results are encouraging, there have been no systematic attempts to link Whole Slide Images (WSIs) to transcriptomic profiles. In this study, we developed a cutting-edge deep learning model named HE2RNA, specifically customized for the direct prediction of gene expression from H&E-stained WSIs without need for annotation from pathologists. Our model was trained and tested on 8,725 patients from 28 different cancer types available at The Cancer Genome Atlas (TCGA).
HE2RNA accurately predicted the expression of six gene signatures related to well known cancer hallmarks (angiogenesis, hypoxia, DNA repair, cell cycle and immunity) and performed particularly well for signalling pathways involved in immune cell activation. This indicates that suitably designed deep learning models can recognize subtle structures in tissue imaging and relate them to molecular portraits.
Moreover, HE2RNA is designed to generate a spatial representation (virtual map) of any well-predicted gene expression overlaying the H&E slide. Such a virtual map was validated on a double-stained H&E/CD3 slide obtained from an independent hepatocellular carcinoma sample. This spatialization could be useful in augmenting the pathologists' workflow by providing a virtual multiplexed staining for each H&E slide while overcoming the technical issues associated with immunohistochemistry.
Various important prognostic factors, such as microsatellite instability (MSI), are derived from molecular features. Microsatellite instability refers to the hypermutability of short repetitive genomic sequences caused by impaired DNA mismatch repair. These mutations frequently observed in gastric and colorectal cancer are associated with better response to immunotherapy. We show that the transcriptomic representation learned by our model can be used to improve the performance of MSI status prediction for small datasets of WSI. This type of setting is common since large databases of matched RNA-Seq profiles and WSI are widely available, while databases of matched MSI status and WSI are more scarce. In the future, such technologies could therefore facilitate universal screening of molecular biomarkers and improved identification of patients that could benefit from new therapeutic strategies.
Citation Format: Elodie Pronier, Benoît Schmauch, Alberto Romagnoni, Charlie Saillard, Pascale Maillé, Julien Calderaro, Meriem Sefta, Sylvain Toldo, Mikhail Zaslavskiy, Thomas Clozel, Matahi Moarii, Pierre Courtiol, Gilles Wainrib. HE2RNA: A deep learning model for transcriptomic learning from digital pathology abstract. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 2105.
Somatic mutations in IDH1/IDH2 and TET2 result in impaired TET2-mediated conversion of 5-methylcytosine (5mC) to 5-hydroxymethylcytosine (5hmC). The observation that WT1 inactivating mutations ...anticorrelate with TET2/IDH1/IDH2 mutations in acute myeloid leukemia (AML) led us to hypothesize that WT1 mutations may impact TET2 function. WT1 mutant AML patients have reduced 5hmC levels similar to TET2/IDH1/IDH2 mutant AML. These mutations are characterized by convergent, site-specific alterations in DNA hydroxymethylation, which drive differential gene expression more than alterations in DNA promoter methylation. WT1 overexpression increases global levels of 5hmC, and WT1 silencing reduced 5hmC levels. WT1 physically interacts with TET2 and TET3, and WT1 loss of function results in a similar hematopoietic differentiation phenotype as observed with TET2 deficiency. These data provide a role for WT1 in regulating DNA hydroxymethylation and suggest that TET2 IDH1/IDH2 and WT1 mutations define an AML subtype defined by dysregulated DNA hydroxymethylation.
Genetic studies have identified recurrent somatic mutations in Acute Myeloid Leukemia (AML) patients, including in WT1 (Wilms' tumor gene 1). WT1 encodes for a zinc finger transcription factor that ...can be either targeted by nonsense/frameshift mutations or, alternatively, overexpressed in AML patients. The molecular mechanisms by which WT1 mutations contribute to leukemogenesis have not been fully elucidated. We recently showed that WT1 acts as a co-factor for TET2-mediated enzymatic function and placement of 5-hydroxymethylcytosine. Here, we show that with time Wt1 haploinsufficiency enhances stem cell self-renewal and cooperates with FLT3-ITD mutations to induce leukemogenesis.
In order to assess the impact of Wt1 loss/haploinsufficiency on hematopoiesis and leukemogenesis in vivo, we generated a hematopoietic-specific conditional Wt1 knockout mice model (Mx1-Cre+ Wt1fl/fl). Studies of steady state hematopoiesis showed that Wt1 loss or haploinsufficiency was not associated with any significant differences in blood parameters compared to control counterparts. Lineage-Sca-1+c-Kit+ from Wt1fl/fl or Wt1fl/+ animals were not able to serially replate in CFU assays compare to control. We did note that homozygous Wt1-deletion induces an increase absolute number of GM colonies. Wt1-deletion led to a reduction in the proportion of ST-HSC percentage and an increase in GMP and MPPs in BM of primary non-competitive transplanted animals (Figure 1). These results showed loss of Wt1 alters myeloid differentiation but does not markedly impact steady state hematopoiesis.
However, when we transplanted Wt1fl/+, but not Wt1fl/fl cells into secondary recipients, we enhanced engraftment of BM cells over control counterparts (Figure 2) and progression to T-ALL. This disease was transplantable and characterized by increased percentages of LT-HSC and MP in the BM of the recipient mice. This led us to hypothesize that WT1 haploinsufficiency, could induce age-dependent changes in hematopoietic stem cell function. Consistent with this proposal, a subset of WT1-haploinsufficient primary elderly mice (18-22 months old) developed hepatomegaly due to liver-specific myeloproliferation and extramedullary hematopoiesis. Moreover, we observed that Wt1fl/+, but not Wt1fl/fl cells, from mice 6-8 months after initial deletion showed a competitive advantage in transplantation assays and increased serial replating. By contrast, loss of one copy of Wt1 did not induce these phenotypes in young mice analyzed 4-6 weeks post deletion (Figure 3) or in mice in which we deleted one copy of Wt1 at 6-8 months of age. These data suggest that Wt1 haploinsufficiency can induce a progressive increase in self-renewal consistent with progressive remodeling of the Wt1-mutant clone.
Given this observation, we hypothesized additional genetic events were required to induce leukemogenesis. Activating mutations in the FLT3 (Fms- like tyrosine kinase 3) gene commonly co-occur with WT1 mutations. To assess if these genes cooperate in leukemogenesis, we crossed Wt1fl/fl mice with Flt3-ITD knock-in mice and induced Wt1-deletion. In vivo competitive assays and serial transplantation experiments showed that Flt3-ITD-Wt1-depleted cells engrafted better in recipient mice than Flt3-ITD cells (Figure 4). Wt1fl/+ FLT3m/m mice developed fully penetrant AML with short latency (5-7 weeks post birth). This disease was characterized by increased percentages of MP and GMP in the BM of these mice compared to controls (Figure 5). Prompted by the known influence of genetic background on the phenotypes resulting from gene-targeting experiments, we bred Wt1fl/+ FLT3m/+ into the mouse strain 129S1/SvImJ. Wt1fl/+ FLT3m/m mouse developed a similar AML with a similar penetrance but a longer latency (7-9 weeks). We also noticed that Wt1fl/+ FLT3m/+ micedeveloped AML in this strain (6-7 months post birth) compare to mice with a C57BL/6 background.
Taken together, our results demonstrate that alterations in WT1 function have a dose- and time-dependent impact on hematopoiesis. We hypothesize that Wt1-haploinsufficiency allows for hematopoietic stem/progenitor cells to accumulate epigenetic changes and secondary mutations (such as FLT3-ITD), which then promote leukemic transformation. This work also demonstrates that AML mutations can induce age-dependent, progressive alterations in hematopoiesis, which contribute to transformation.
Display omitted
Levine:Qiagen: Equity Ownership; Roche: Research Funding; Roche: Research Funding; Celgene: Research Funding; Qiagen: Equity Ownership; Celgene: Research Funding.
There is a pressing need to develop novel, mechanism-based therapeutic approaches that can be used to improve therapies for genetically defined tumor subtypes. Chan and colleagues have demonstrated ...recently that BCL-2 inhibitors can target IDH1/2 mutant cancers through a mutant-specific dependency in metabolic regulation.
There is a pressing need to develop novel, mechanism-based therapeutic approaches that can be used to improve therapies for genetically defined tumor subtypes. Chan and colleagues have demonstrated recently that BCL-2 inhibitors can target IDH1/2-mutant cancers through a mutant-specific dependency in metabolic regulation.