Targeting guanine nucleotide exchange factors for novel cancer drug discovery.

INTRODUCTION: Guanine nucleotide exchange factors (GEFs) regulate the activation of small GTPases (G proteins) of the Ras superfamily proteins controlling cellular functions. Ras superfamily proteins act as 'molecular switches' that are turned 'ON' by guanine exchange. There are five major groups of Ras family GTPases: Ras, Ran, Rho, Rab and Arf, with a variety of different GEFs regulating their GTP loading. GEFs have been implicated in various diseases including cancer. This makes GEFs attractive targets to modulate signaling networks controlled by small GTPases. AREAS COVERED: In this review, the roles and mechanisms of GEFs in malignancy are outlined. The mechanism of guanine exchange activity by GEFs on a small GTPase is illustrated. Then, some examples of GEFs that are significant in cancer are presented with a discussion on recent progress in therapeutic targeting efforts using a variety of approaches. EXPERT OPINION: Recently, GEFs have emerged as potential therapeutic targets for novel cancer drug development. Targeting small GTPases is challenging; thus, targeting their activation by GEFs is a promising strategy. Most GEF-targeted drugs are still in preclinical development. A deeper biological understanding of the underlying mechanisms of GEF activity and utilizing advanced technology are necessary to enhance drug discovery for GEFs in cancer.

TGFß in pancreas and colorectal cancer: opportunities to overcome therapeutic resistance.

TGFß is a pleiotropic signaling pathway, which plays a pivotal role in regulating a multitude of cellular functions. TGFß has a dual role in cell regulation where it induces growth inhibition and cell death; however, it can switch to a growth-promoting state under cancerous conditions. TGFß is upregulated in CRC and pancreatic cancer, altering the tumor microenvironment, immune system, and promoting a mesenchymal state. The upregulation of TGFß in certain cancers leads to resistance to immunotherapy, and attempts to inhibit TGFß expression have led to reduced therapeutic resistance when combined with chemo- and immunotherapy. Here, we review the current TGFß inhibitor drugs in clinical trials for pancreatic and colorectal cancer, with the goal of uncovering advances in improving clinical efficacy for TGFß combinational treatments in patients. Furthermore, we discuss the relevance of alterations in TGFß signaling and germline variants in the context of personalizing treatment for patients who show lack of response to current therapeutics.

Adaptation of a Personalized Electronic Care Planning Tool for Cancer Follow-up Care: Formative Study.

BACKGROUND: Most patients diagnosed with colorectal cancer will survive for at least 5 years; thus, engaging patients to optimize their health will likely improve outcomes. Clinical guidelines recommend patients receive a comprehensive care plan (CP) when transitioning from active treatment to survivorship, which includes support for ongoing symptoms and recommended healthy behaviors. Yet, cancer care providers find this guideline difficult to implement. Future directions for survivorship care planning include enhancing information technology support for developing personalized CPs, using CPs to facilitate self-management, and assessing CPs in clinical settings. OBJECTIVE: We aimed to develop an electronic tool for colorectal cancer follow-up care (CFC) planning. METHODS: Incorporating inputs from health care professionals and patient stakeholders is fundamental to the successful integration of any tool into the clinical workflow. Thus, we followed the Integrate, Design, Assess, and Share (IDEAS) framework to adapt an existing application for stroke care planning (COMPASS-CP) to meet the needs of colorectal cancer survivors (COMPASS-CP CFC). Constructs from the Consolidated Framework for Implementation Research (CFIR) guided our approach. We completed this work in 3 phases: (1) gathering qualitative feedback from stakeholders about the follow-up CP generation design and workflow; (2) adapting algorithms and resource data sources needed to generate a follow-up CP; and (3) optimizing the usability of the adapted prototype of COMPASS-CP CFC. We also quantitatively measured usability (target average score ≥70; range 0-100), acceptability, appropriateness, and feasibility. RESULTS: In the first phase, health care professionals (n=7), and patients and caregivers (n=7) provided qualitative feedback on COMPASS-CP CFC that informed design elements such as selection, interpretation, and clinical usefulness of patient-reported measures. In phase 2, we built a minimal viable product of COMPASS-CP CFC. This tool generated CPs based on the needs identified by patient-completed measures (including validated patient-reported outcomes) and electronic health record data, which were then matched with resources by zip code and preference to support patients' self-management. Elements of the CFIR assessed revealed that most health care professionals believed the tool would serve patients' needs and had advantages. In phase 3, the average System Usability Scale score was above our target score for health care professionals (n=5; mean 71.0, SD 15.2) and patients (n=5; mean 95.5, SD 2.1). Participants also reported high levels of acceptability, appropriateness, and feasibility. Additional CFIR-informed feedback, such as desired format for training, will inform future studies. CONCLUSIONS: The data collected in this study support the initial usability of COMPASS-CP CFC and will inform the next steps for implementation in clinical care. COMPASS-CP CFC has the potential to streamline the implementation of personalized CFC planning to enable systematic access to resources that will support self-management. Future research is needed to test the impact of COMPASS-CP CFC on patient health outcomes.

Feasibility of Systems Support Mapping to guide patient-driven health self-management in colorectal cancer survivors.

OBJECTIVE: To evaluate feasibility of System Support Mapping (MAP), a systems thinking activity that involves creating a diagram of existing self-management activities (e.g. symptom management, health behaviors) to facilitate autonomous engagement in optimal self-management. DESIGN: One-arm pilot study of MAP in colorectal cancer survivors (NCT03520283). MAIN OUTCOME MEASURES: Feasibility of recruitment and retention (primary outcome), acceptability, and outcome variability over time. RESULTS: We enrolled 24 of 66 cancer survivors approached (36%) and 20 completed follow-up (83%). Key reasons for declining participation included: not interested (n = 18), did not perceive a need (n = 9), and emotional distress/overwhelmed (n = 7). Most participants reported that MAP was acceptable (e.g. 80% liked MAP quite a bit/very much). Exploratory analyses revealed a -4.68 point reduction in fatigue from before to 2 weeks after MAP exceeding a minimally important difference (d = -0.68). There were also improvements in patient autonomy (d = 0.63), self-efficacy (for managing symptoms: d = 0.56, for managing chronic disease: d = 0.44), psychological stress (d = -0.45), anxiety (d = -0.34), sleep disturbance (d = -0.29) and pain (d = -0.32). Qualitative feedback enhanced interpretation of results. CONCLUSIONS: MAP feasibility in colorectal cancer survivors was mixed, predominantly because many patients did not perceive a need for this approach. MAP was acceptable among participants and showed promise for improving health outcomes.

Regulator of Chromosome Condensation (RCC1) a novel therapeutic target in pancreatic ductal adenocarcinoma drives tumor progression via the c-Myc-RCC1-Ran axis.

Pancreatic ductal adenocarcinoma (PDAC) is a highly lethal malignancy with limited therapeutic options. Here we for the first time evaluated the role of regulator of chromosome condensation 1 (RCC1) in PDAC subsistence and drug resistance. RCC1 expression was found to be elevated in PDAC tissues in comparison with normal pancreatic tissues and was linked to poor prognosis. RCC1 silencing in a panel of PDAC cells by RNA interference and CRISPR-Cas9 resulted in reduced cellular proliferation in 2D and 3D cultures. RCC1 KD reduced migratory and clonogenic ability, enhanced apoptosis, and altered cell cycle distribution in human PDAC cells as well as cells isolated from the LSL-Kras G12D/+; LSL-Trp53 R172H/+ ;Pdx1-Cre (KPC) mouse tumors. Subcutaneous cell-derived xenografts show significantly attenuated growth of RCC1 KO tumors. Mechanistically, RCC1 knockdown resulted in disruption of subcellular Ran distribution indicating that stable nuclear Ran localization is critical for PDAC proliferation. Nuclear and cytosolic proteomic analysis revealed altered subcellular proteome in RCC1 KD KPC-tumor-derived cells. Altered cytoplasmic protein pathways include several metabolic pathways and PI3K-Akt signaling pathway. Pathways enriched in altered nuclear proteins include cell cycle, mitosis, and RNA regulation. RNA sequencing of RCC1 KO cells showed widespread transcriptional alterations. Upstream of RCC1, c-Myc activates the RCC1-Ran axis, and RCC1 KO enhances the sensitivity of PDAC cells to c-Myc inhibitors. Finally, RCC1 knockdown resulted in the sensitization of PDAC cells to Gemcitabine. Our results indicate that RCC1 is a potential therapeutic target in PDAC that warrants further clinical investigations.

A transformer-based deep-learning approach for classifying brain metastases into primary organ sites using clinical whole-brain MRI images.

Treatment decisions for brain metastatic disease rely on knowledge of the primary organ site and are currently made with biopsy and histology. Here, we develop a deep-learning approach for accurate non-invasive digital histology with whole-brain magnetic resonance imaging (MRI) data. Contrast-enhanced T1-weighted and fast spoiled gradient echo brain MRI exams (n = 1,582) were preprocessed and input to the proposed deep-learning workflow for tumor segmentation, modality transfer, and primary site classification into one of five classes. Tenfold cross-validation generated an overall area under the receiver operating characteristic curve (AUC) of 0.878 (95% confidence interval [CI]: 0.873,0.883). These data establish that whole-brain imaging features are discriminative enough to allow accurate diagnosis of the primary organ site of malignancy. Our end-to-end deep radiomic approach has great potential for classifying metastatic tumor types from whole-brain MRI images. Further refinement may offer an invaluable clinical tool to expedite primary cancer site identification for precision treatment and improved outcomes.

hDirect-MAP: projection-free single-cell modeling of response to checkpoint immunotherapy.

There is a lack of robust generalizable predictive biomarkers of response to immune checkpoint blockade in multiple types of cancer. We develop hDirect-MAP, an algorithm that maps T cells into a shared high-dimensional (HD) expression space of diverse T cell functional signatures in which cells group by the common T cell phenotypes rather than dimensional reduced features or a distorted view of these features. Using projection-free single-cell modeling, hDirect-MAP first removed a large group of cells that did not contribute to response and then clearly distinguished T cells into response-specific subpopulations that were defined by critical T cell functional markers of strong differential expression patterns. We found that these grouped cells cannot be distinguished by dimensional-reduction algorithms but are blended by diluted expression patterns. Moreover, these identified response-specific T cell subpopulations enabled a generalizable prediction by their HD metrics. Tested using five single-cell RNA-seq or mass cytometry datasets from basal cell carcinoma, squamous cell carcinoma and melanoma, hDirect-MAP demonstrated common response-specific T cell phenotypes that defined a generalizable and accurate predictive biomarker.

Recruitment of KMT2C/MLL3 to DNA Damage Sites Mediates DNA Damage Responses and Regulates PARP Inhibitor Sensitivity in Cancer.

When recruited to promoters, histone 3 lysine 4 (H3K4) methyltransferases KMT2 (KMT2A-D) activate transcription by opening chromatin through H3K4 methylation. Here, we report that KMT2 mutations occur frequently in non-small cell lung cancer (NSCLC) and are associated with high mutation loads and poor survival. KMT2C regulated DNA damage responses (DDR) through direct recruitment to DNA damage sites by Ago2 and small noncoding DNA damage response RNA, where it mediates H3K4 methylation, chromatin relaxation, secondary recruitment of DDR factors, and amplification of DDR signals along chromatin. Furthermore, by disrupting homologous recombination (HR)-mediated DNA repair, KMT2C/D mutations sensitized NSCLC to Poly(ADP-ribose) polymerase inhibitors (PARPi), whose efficacy is unclear in NSCLC due to low BRCA1/2 mutation rates. These results demonstrate a novel, transcription-independent role of KMT2C in DDR and identify high-frequency KMT2C/D mutations as much-needed biomarkers for PARPi therapies in NSCLC and other cancers with infrequent BRCA1/2 mutations. SIGNIFICANCE: This study uncovers a critical role for KMT2C in DDR via direct recruitment to DNA damage sites, identifying high-frequency KMT2C/D mutations as biomarkers for response to PARP inhibition in cancer.

Improved Antitumor Activity of the Fluoropyrimidine Polymer CF10 in Preclinical Colorectal Cancer Models through Distinct Mechanistic and Pharmacologic Properties.

Chemotherapy regimens that include 5-fluorouracil (5-FU) are central to colorectal cancer treatment; however, risk/benefit concerns limit 5-FU's use, necessitating development of improved fluoropyrimidine (FP) drugs. In our study, we evaluated a second-generation nanoscale FP polymer, CF10, for improved antitumor activity. CF10 was more potent than the prototype FP polymer F10 and much more potent than 5-FU in multiple colorectal cancer cell lines including HCT-116, LS174T, SW480, and T84D. CF10 displayed improved stability to exonuclease degradation relative to F10 and reduced susceptibility to thymidine antagonism due to extension of the polymer with arabinosyl cytidine. In colorectal cancer cells, CF10 strongly inhibited thymidylate synthase (TS), induced Top1 cleavage complex formation and caused replication stress, while similar concentrations of 5-FU were ineffective. CF10 was well tolerated in vivo and invoked a reduced inflammatory response relative to 5-FU. Blood chemistry parameters in CF10-treated mice were within normal limits. In vivo, CF10 displayed antitumor activity in several colorectal cancer flank tumor models including HCT-116, HT-29, and CT-26. CF10's antitumor activity was associated with increased plasma levels of FP deoxynucleotide metabolites relative to 5-FU. CF10 significantly reduced tumor growth and improved survival (84.5 days vs. 32 days; P < 0.0001) relative to 5-FU in an orthotopic HCT-116-luc colorectal cancer model that spontaneously metastasized to liver. Improved survival in the orthotopic model correlated with localization of a fluorescent CF10 conjugate to tumor. Together, our preclinical data support an early-phase clinical trial of CF10 for treatment of colorectal cancer.

IGFBP2 promotes tumor progression by inducing alternative polarization of macrophages in pancreatic ductal adenocarcinoma through the STAT3 pathway.

Tumor-associated macrophages (TAMs) represent the M2-like phenotype with potent immunosuppressive activity, and play a pro-tumor role in pancreatic ductal adenocarcinoma (PDAC) biology. In this study, we investigated the role of the insulin-like growth factor binding protein 2 (IGFBP2) as a determinant of TAM polarity. Clinical data revealed that the levels of IGFBP2 correlated with M2 TAMs accumulation and disease progression in human PDAC. In vivo mouse model experiments showed that IGFBP2 promoted an immunosuppressive microenvironment and tumor growth in a macrophage dependent manner. Bioinformatics analysis of PDAC transcriptomes revealed a significant association between IGFBP2 expression and M2 macrophage polarization and signal transducer and activator of transcription 3 (STAT3) activation. Mechanistic investigations demonstrated that IGFBP2 augmented the expression and secretion of IL-10 through STAT3 activation in PDAC cells, which induced TAM polarization toward an M2 phenotype. IGFBP2-polarized M2 macrophages significantly increased Tregs infiltration and impaired antitumor T-cell immunity in a mouse model. Thus, our investigations have illuminated the IGFBP2 signaling pathway that contributes to the macrophage-based immunosuppressive microenvironment in PDAC, suggesting that blocking the IGFBP2 axis constitutes a potential treatment strategy to reset TAM polarization toward an antitumor state in PDAC.

Regorafenib dose-optimisation in patients with refractory metastatic colorectal cancer (ReDOS): a randomised, multicentre, open-label, phase 2 study.

BACKGROUND: Regorafenib confers an overall survival benefit in patients with refractory metastatic colorectal cancer; however, the adverse event profile of regorafenib has limited its use. Despite no supportive evidence, various dosing schedules are used clinically to alleviate toxicities. This study evaluated the safety and activity of two regorafenib dosing schedules. METHODS: In this randomised, multicentre, open-label, phase 2 study done in 39 outpatient cancer centres in the USA, adults aged 18 years or older with histologically or cytologically confirmed advanced or metastatic adenocarcinoma of the colon or rectum that was refractory to previous standard therapy, including EGFR inhibitors if KRAS wild-type, were enrolled. Eligible patients had an Eastern Cooperative Oncology Group performance status of 0-1 and had no previous treatment with regorafenib. Patients were randomly assigned (1:1:1:1) into four groups with two distinct regorafenib dosing strategies and two clobetasol usage plans, stratified by hospital. Regorafenib dosing strategies were a dose-escalation strategy (starting dose 80 mg/day orally with weekly escalation, per 40 mg increment, to 160 mg/day regorafenib) if no significant drug-related adverse events occurred and a standard-dose strategy (160 mg/day orally) for 21 days of a 28-day cycle. Clobetasol usage plans (0·05% clobetasol cream twice daily applied to palms and soles) were either pre-emptive or reactive. After randomisation to the four preplanned groups, using the Pocock and Simon dynamic allocation procedures stratified by the treating hospitals, we formally tested the interaction between the two interventions, dosing strategy and clobetasol usage. Given the absence of a significant interaction (p=0·74), we decided to pool the data for the pre-emptive and reactive treatment with clobetasol and compared the two dosing strategies (dose escalation vs standard dose). The primary endpoint was the proportion of evaluable patients (defined as those who were eligible, consented, and received any protocol treatment) initiating cycle 3 and was analysed per protocol. Superiority for dose escalation was declared if the one-sided p value with Fisher's exact test was less than 0·2. This trial is registered with ClinicalTrials.gov, number NCT02368886. This study is fully accrued but remains active. FINDINGS: Between June 2, 2015, and June 22, 2017, 123 patients were randomly assigned to treatment, of whom 116 (94%) were evaluable. The per-protocol population consisted of 54 patients in the dose-escalation group and 62 in the standard-dose group. At data cutoff on July 24, 2018, median follow-up was 1·18 years (IQR 0·98-1·57). The primary endpoint was met: 23 (43%, 95% CI 29-56) of 54 patients in the dose-escalation group initiated cycle 3 versus 16 (26%, 15-37) of 62 patients in the standard-dose group (one-sided p=0·043). The most common grade 3-4 adverse events were fatigue (seven [13%] patients in the dose-escalation group vs 11 [18%] in the standard-dose group), hand-foot skin reaction (eight [15%] patients vs ten [16%] patients), abdominal pain (nine [17%] patients vs four [6%] patients), and hypertension (four [7%] patients vs nine [15%] patients). 14 patients had at least one drug-related serious adverse event: six patients in the dose-escalation group and eight patients in the standard-dose group. There was one probable treatment-related death in the standard-dose group (myocardial infarction). INTERPRETATION: The dose-escalation dosing strategy represents an alternative approach for optimising regorafenib dosing with comparable activity and lower incidence of adverse events and could be implemented in clinical practice on the basis of these data. FUNDING: Bayer HealthCare Pharmaceuticals.

Tumour-specific amplitude-modulated radiofrequency electromagnetic fields induce differentiation of hepatocellular carcinoma via targeting Cav3.2 T-type voltage-gated calcium channels and Ca2+ influx.

BACKGROUND: Administration of amplitude modulated 27·12 MHz radiofrequency electromagnetic fields (AM RF EMF) by means of a spoon-shaped applicator placed on the patient's tongue is a newly approved treatment for advanced hepatocellular carcinoma (HCC). The mechanism of action of tumour-specific AM RF EMF is largely unknown. METHODS: Whole body and organ-specific human dosimetry analyses were performed. Mice carrying human HCC xenografts were exposed to AM RF EMF using a small animal AM RF EMF exposure system replicating human dosimetry and exposure time. We performed histological analysis of tumours following exposure to AM RF EMF. Using an agnostic genomic approach, we characterized the mechanism of action of AM RF EMF. FINDINGS: Intrabuccal administration results in systemic delivery of athermal AM RF EMF from head to toe at levels lower than those generated by cell phones held close to the body. Tumour shrinkage results from differentiation of HCC cells into quiescent cells with spindle morphology. AM RF EMF targeted antiproliferative effects and cancer stem cell inhibiting effects are mediated by Ca2+ influx through Cav3·2 T-type voltage-gated calcium channels (CACNA1H) resulting in increased intracellular calcium concentration within HCC cells only. INTERPRETATION: Intrabuccally-administered AM RF EMF is a systemic therapy that selectively block the growth of HCC cells. AM RF EMF pronounced inhibitory effects on cancer stem cells may explain the exceptionally long responses observed in several patients with advanced HCC. FUND: Research reported in this publication was supported by the National Cancer Institute's Cancer Centre Support Grant award number P30CA012197 issued to the Wake Forest Baptist Comprehensive Cancer Centre (BP) and by funds from the Charles L. Spurr Professorship Fund (BP). DWG is supported by R01 AA016852 and P50 AA026117.

Oncology Navigation Decreases Time to Treatment in Patients with Pancreatic Malignancy.

BACKGROUND: Care of pancreatic cancer patients has become increasingly complex, which has led to delays in the initiation of therapy. Nurse navigators have been added to care teams, in part, to ameliorate this delay. This study investigated the difference in time from first oncology visit to first treatment date in patients with any pancreatic malignancy before and after the addition of an Oncology Navigator. METHODS: A single-institution database of patients with any pancreatic neoplasm evaluated by a provider in radiation, medical, or surgical oncology between 1 October 2015 and 30 September 2017 was analyzed. After 1 October 2016, an Oncology Navigator met patients at their initial visit and coordinated care throughout treatment. The cohort was divided into two groups: patients evaluated prior to the implementation of an Oncology Navigator and patients evaluated after implementation. Patient demographics and time from first visit to first intervention were compared. RESULTS: Overall, 147 patients with a new diagnosis of pancreatic neoplasm were evaluated; 57 patients were seen prior to the start of the Oncology Navigator program and 79 were evaluated after the navigation program was implemented. On univariate analysis, time from first contact by any provider to intervention was 46 days prior to oncology navigation and 26 days after implementation of oncology navigation (p = 0.005). While controlling for other covariates, employment of the Oncology Navigator decreased the time from first contact by any provider to intervention by almost 16 days (p = 0.009). CONCLUSIONS: Implementing an oncology navigation program significantly decreased time to treatment in patients with pancreatic malignancy.

Diagnosis of thyroid cancer using deep convolutional neural network models applied to sonographic images: a retrospective, multicohort, diagnostic study.

BACKGROUND: The incidence of thyroid cancer is rising steadily because of overdiagnosis and overtreatment conferred by widespread use of sensitive imaging techniques for screening. This overall incidence growth is especially driven by increased diagnosis of indolent and well-differentiated papillary subtype and early-stage thyroid cancer, whereas the incidence of advanced-stage thyroid cancer has increased marginally. Thyroid ultrasound is frequently used to diagnose thyroid cancer. The aim of this study was to use deep convolutional neural network (DCNN) models to improve the diagnostic accuracy of thyroid cancer by analysing sonographic imaging data from clinical ultrasounds. METHODS: We did a retrospective, multicohort, diagnostic study using ultrasound images sets from three hospitals in China. We developed and trained the DCNN model on the training set, 131 731 ultrasound images from 17 627 patients with thyroid cancer and 180 668 images from 25 325 controls from the thyroid imaging database at Tianjin Cancer Hospital. Clinical diagnosis of the training set was made by 16 radiologists from Tianjin Cancer Hospital. Images from anatomical sites that were judged as not having cancer were excluded from the training set and only individuals with suspected thyroid cancer underwent pathological examination to confirm diagnosis. The model's diagnostic performance was validated in an internal validation set from Tianjin Cancer Hospital (8606 images from 1118 patients) and two external datasets in China (the Integrated Traditional Chinese and Western Medicine Hospital, Jilin, 741 images from 154 patients; and the Weihai Municipal Hospital, Shandong, 11 039 images from 1420 patients). All individuals with suspected thyroid cancer after clinical examination in the validation sets had pathological examination. We also compared the specificity and sensitivity of the DCNN model with the performance of six skilled thyroid ultrasound radiologists on the three validation sets. FINDINGS: Between Jan 1, 2012, and March 28, 2018, ultrasound images for the four study cohorts were obtained. The model achieved high performance in identifying thyroid cancer patients in the validation sets tested, with area under the curve values of 0·947 (95% CI 0·935-0·959) for the Tianjin internal validation set, 0·912 (95% CI 0·865-0·958) for the Jilin external validation set, and 0·908 (95% CI 0·891-0·925) for the Weihai external validation set. The DCNN model also showed improved performance in identifying thyroid cancer patients versus skilled radiologists. For the Tianjin internal validation set, sensitivity was 93·4% (95% CI 89·6-96·1) versus 96·9% (93·9-98·6; p=0·003) and specificity was 86·1% (81·1-90·2) versus 59·4% (53·0-65·6; p<0·0001). For the Jilin external validation set, sensitivity was 84·3% (95% CI 73·6-91·9) versus 92·9% (84·1-97·6; p=0·048) and specificity was 86·9% (95% CI 77·8-93·3) versus 57·1% (45·9-67·9; p<0·0001). For the Weihai external validation set, sensitivity was 84·7% (95% CI 77·0-90·7) versus 89·0% (81·9-94·0; p=0·25) and specificity was 87·8% (95% CI 81·6-92·5) versus 68·6% (60·7-75·8; p<0·0001). INTERPRETATION: The DCNN model showed similar sensitivity and improved specificity in identifying patients with thyroid cancer compared with a group of skilled radiologists. The improved technical performance of the DCNN model warrants further investigation as part of randomised clinical trials. FUNDING: The Program for Changjiang Scholars and Innovative Research Team in University in China, and National Natural Science Foundation of China.

Loss of XIST in Breast Cancer Activates MSN-c-Met and Reprograms Microglia via Exosomal miRNA to Promote Brain Metastasis.

Up to 30% of patients with metastatic breast cancer eventually develop brain metastasis, yet the pathologic mechanism behind this development remains poorly understood. Here, we profiled long noncoding RNAs in brain metastatic tumors from patients with breast cancer and found that the X-inactive-specific transcript (XIST) was significantly downregulated in these tissues. XIST expression levels inversely correlated with brain metastasis, but not with bone metastasis in patients. Silencing of XIST preferentially promoted brain metastatic growth of XISThigh cells in our xenograft models. Moreover, knockout of XIST in mice mammary glands accelerated primary tumor growth as well as metastases in the brain. Decreased expression of XIST stimulated epithelial-mesenchymal transition and activated c-Met via MSN-mediated protein stabilization, which resulted in the promotion of stemness in the tumor cells. Loss of XIST also augmented secretion of exosomal miRNA-503, which triggered M1-M2 polarization of microglia. This M1-M2 conversion upregulated immune suppressive cytokines in microglia that suppressed T-cell proliferation. Furthermore, we screened an FDA-approved drug library and identified fludarabine as a synthetic lethal drug for XISTlow breast tumor cells and found that fludarabine blocked brain metastasis in our animal model. Our results indicate that XIST plays a critical role in brain metastasis in breast cancer by affecting both tumor cells and the tumor microenvironment and that the XIST-mediated pathway may serve as an effective target for treating brain metastasis.Significance: These findings describe mechanisms of how loss of the lncRNA XIST promotes brain metastasis in breast cancer and identify fludarabine as a potential therapeutic agent that specifically eliminates XISTlow tumor cells in the brain. Cancer Res; 78(15); 4316-30. ©2018 AACR.

Truncated Glioma-Associated Oncogene Homolog 1 (tGLI1) Mediates Mesenchymal Glioblastoma via Transcriptional Activation of CD44.

The molecular pathways driving mesenchymal glioblastoma (GBM) are still not well understood. We report here that truncated glioma-associated oncogene homolog 1 (tGLI1) is a tumor-specific transcription factor that facilitates GBM growth, is enriched in the mesenchymal subtype of GBM and glioma stem cells (GSC), and promotes mesenchymal GSC by upregulating transcription of CD44. In an orthotopic GBM xenograft mouse model, tGLI1-overexpressing tumors grew more aggressively with increased proliferation and angiogenesis compared with control and GLI1-overexpressing xenografts. tGLI1 was highly expressed in GBM clinical specimens but undetectable in normal brains, whereas GLI1 was expressed in both tissues. A tGLI1 activation signature (tGAS) correlated with glioma grade, tumor angiogenesis, and poor overall survival, and GBMs with high tGAS were enriched with mesenchymal GBM/GSC gene signatures. Neurospheres contained increased levels of tGLI1, but not GLI1, compared with the monolayer culture; mesenchymal GSC expressed more tGLI1 than proneural GSC. Ectopic tGLI1 expression enhanced the ability of mesenchymal GSC to yield neurospheres in vitro and to form tumors in mouse brains. Selective tGLI1 knockdown reduced neurosphere formation of GBM cells. tGLI1 bound to and transactivated the promoter of the CD44 gene, a marker and mediator for mesenchymal GSC, leading to its expression. Collectively, these findings advance our understanding of GBM biology by establishing tGLI1 as a novel transcriptional activator of CD44 and a novel mediator of mesenchymal GBM and GSC.Significance: These findings highlight the role of a tumor-specific gain-of-function transcription factor tGLI1 in mesenchymal glioma stem cell maintenance and mesenchymal GBM growth. Cancer Res; 78(10); 2589-600. ©2018 AACR.

Combined inhibition of AKT and HSF1 suppresses breast cancer stem cells and tumor growth.

Breast cancer is the most common cancer in women and the second leading cause of cancer deaths in women. Over 90% of breast cancer deaths are attributable to metastasis. Our lab has recently reported that AKT activates heat shock factor 1 (HSF1), leading to epithelial-to-mesenchymal transition in HER2-positive breast cancer. However, it is unknown whether the AKT-HSF1 pathway plays an important role in other breast cancer subtypes, breast cancer stem cells, or breast cancer growth and metastasis. Herein, we showed AKT and HSF1 to be frequently co-activated in breast cancer cell lines and specimens across different subtypes. Activated AKT (S473) and HSF1 (S326) are strongly associated with shortened time to metastasis. Inhibition of the AKT-HSF1 signaling axis using small molecule inhibitors, HSF1 knockdown or the dominant-negative HSF1 mutant (S326A) reduced the growth of metastatic breast cancer cells and breast cancer stem cells. The combination of small molecule inhibitors targeting AKT (MK-2206) and HSF1 (KRIBB11) resulted in synergistic killing of breast cancer cells and breast cancer stem cells across different molecular subtypes. Using an orthotopic xenograft mouse model, we found that combined targeting of AKT and HSF1 to significantly reduce tumor growth, induce tumor apoptosis, delay time to metastasis, and prolong host survival. Taken together, our results indicate AKT-HSF1 signaling mediates breast cancer stem cells self-renewal, tumor growth and metastasis, and that dual targeting of AKT and HSF1 resulted in synergistic suppression of breast cancer progression thereby supporting future testing of AKT-HSF1 combination therapy for breast cancer patients.