Reducing Gadolinium Contrast With Artificial Intelligence.

Gadolinium contrast is an important agent in magnetic resonance imaging (MRI), particularly in neuroimaging where it can help identify blood-brain barrier breakdown from an inflammatory, infectious, or neoplastic process. However, gadolinium contrast has several drawbacks, including nephrogenic systemic fibrosis, gadolinium deposition in the brain and bones, and allergic-like reactions. As computer hardware and technology continues to evolve, machine learning has become a possible solution for eliminating or reducing the dose of gadolinium contrast. This review summarizes the clinical uses of gadolinium contrast, the risks of gadolinium contrast, and state-of-the-art machine learning methods that have been applied to reduce or eliminate gadolinium contrast administration, as well as their current limitations, with a focus on neuroimaging applications. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 1.

Trans Element-Hydrogen Bonds: A Distinctive Difference Between Transition Metals and Main Group Elements.

The change in sign of the interaction force constant between element-hydrogen stretching modes of trans-dihydrides of the d block and p block elements is analyzed for the first time. As the transition metal M approaches group 12, the higher energy symmetric trans-H-M-H vibration νsym approaches the energy of the antisymmetric vibration νasym. Crossing to group 13 elements E, the trans-H-E-H vibration νsym increasingly drops below νasym. This reversal is attributed to the d orbital that participates in the H-M-H bonding but is nonbonding in the H-E-H compounds. DFT calculations are used to probe the energetics of isoelectronic triatomic [H-M-H]n+ and [H-E-H]n- to reveal this trend and also to demonstrate that the magnitude of these interactions (νgap) increases down groups 11, 12, and 14 but remains fairly constant for group 13. They are also used to show that this reversal is seen in the transition state for hydride transfer to CO2 from the model compounds trans-NiH2(porphyrin) and trans-EH2(porphyrin), E = Si and Ge in their singlet states.

Integrative genomic analysis of mouse and human hepatocellular carcinoma.

Cancer genomics has enabled the exhaustive molecular characterization of tumors and exposed hepatocellular carcinoma (HCC) as among the most complex cancers. This complexity is paralleled by dozens of mouse models that generate histologically similar tumors but have not been systematically validated at the molecular level. Accurate models of the molecular pathogenesis of HCC are essential for biomedical progress; therefore we compared genomic and transcriptomic profiles of four separate mouse models [MUP transgenic, TAK1-knockout, carcinogen-driven diethylnitrosamine (DEN), and Stelic Animal Model (STAM)] with those of 987 HCC patients with distinct etiologies. These four models differed substantially in their mutational load, mutational signatures, affected genes and pathways, and transcriptomes. STAM tumors were most molecularly similar to human HCC, with frequent mutations in Ctnnb1, similar pathway alterations, and high transcriptomic similarity to high-grade, proliferative human tumors with poor prognosis. In contrast, TAK1 tumors better reflected the mutational signature of human HCC and were transcriptionally similar to low-grade human tumors. DEN tumors were least similar to human disease and almost universally carried the Braf V637E mutation, which is rarely found in human HCC. Immune analysis revealed that strain-specific MHC-I genotype can influence the molecular makeup of murine tumors. Thus, different mouse models of HCC recapitulate distinct aspects of HCC biology, and their use should be adapted to specific questions based on the molecular features provided here.

Metal Hydride Vibrations: The Trans Effect of the Hydride.

trans-Dihydride complexes are important in many homogeneous catalytic processes. Here vibrational spectroscopy and density functional theory (DFT) methods are used for the first time to reveal that 4d and 5d metals transmit more effectively than the 3d metals influence of the ligand trans to the hydride and also couple the motions of the trans-hydrides more effectively. This property of the metal is linked to higher hydride reactivity. The IR and Raman spectra of trans-FeH2(dppm)2, trans-RuH2(PPh(OEt)2)4, and mer-IrH3(PiPr2CH2pyCH2PiPr2) provide M-H force constants and H-M-H interaction force constants that increase as FeII < RuII < IrIII. DFT methods are used to determine, for the first time, the effect of the metal ion (MnI, ReI, FeII, RuII, OsII, CoIII, RhIII, IrIII, PtIV) and ligands on the gap in wavenumbers between the symmetric νsymH-M-H and antisymmetric νasymH-M-H vibrational modes of hydrides that are mutually trans in d6 octahedral complexes. The magnitude of this gap reflects the degree of coupling of, or interaction between, these modes, and this is shown to be a distinctive property of the metal ion. The more polarizable 4d and 5d metal ions are found to have an average gap of 246 cm-1, while the 3d metals have only 90 cm-1. This has been verified experimentally for 3d, 4d, and 5d transition-metal trans-dihydrides, where both the IR and Raman spectra have been measured: trans-RuH2(PPh(OEt)2)4 (from the literature) and trans-FeH2(PPh2CH2PPh2)2 and mer-IrH3(PiPr2CH2pyCH2PiPr2) (this work). Because the 4d and 5d metal ions tend to be better catalysts for the hydrogenation of substrates with polar bonds, this gap may be a fundamental determinant of the kinetic hydricity of the catalyst. Finding the magnitude of this gap and a new estimate of the large hydride trans-effect (Δνt -235 cm-1) allows us to improve the simple equation reported previously, which allows a better estimate of νM-H.

Retrospective Review of 516 Implantations of Option Inferior Vena Cava Filters at a Single Health Care System.

PURPOSE: To retrospectively evaluate the safety, efficacy, and retrievability of Option inferior vena cava (IVC) filters. MATERIALS AND METHODS: All patients (N = 516; 247 women; mean age, 67.1 y ± 15.1; range, 19.5-101.6 y) who received an Option filter between August 2009 and March 2015 at a single health care system were analyzed. RESULTS: The study duration was 68 months, with median clinical follow-up of 7.1 months (range, 1 d to 61.8 mo). During follow-up, 73 of 83 patients (88.0%) underwent successful filter retrieval, 153 died (including three after successful retrieval), and 293 remained alive with filters in situ. Seventeen cases of breakthrough pulmonary embolism (PE) occurred (3.4%). Among 323 patients with direct filter imaging, there were two cases of tilt > 15°, one case of filter deformity, 16 cases of intracaval migration > 2 cm, and no cases of filter fracture. There were six cases of caval occlusion, nine cases of thrombus trapped inside the filter, and 57 cases of limb penetration on computed tomography scans or radiographs of the IVC. Retrieval failures were attributed to filter tilt or tip embedment in the caval wall (n = 4), complete IVC thrombosis (n = 3), thrombus inside the filter (n = 2), or inability to disengage filter legs (n = 1). Recurrent deep vein thrombosis occurred in 34 patients, including 32 with filters in situ and two whose filters had been removed. CONCLUSIONS: Most Option filters were left in situ for permanent indications. Rates of successful retrieval, device-related complications, and breakthrough PE were similar to those associated with other retrievable filters.

Intramolecular C-H/O-H bond cleavage with water and alcohol using a phosphine-free ruthenium carbene NCN pincer complex.

Transition metal complexes that exhibit metal-ligand cooperative reactivity could be suitable candidates for applications in water splitting. Ideally, the ligands around the metal should not contain oxidizable donor atoms, such as phosphines. With this goal in mind, we report new phosphine-free ruthenium NCN pincer complexes with a central N-heterocyclic carbene donor and methylpyridyl N-donors. Reaction with base generates a neutral, dearomatized alkoxo-amido complex, which has been structurally and spectroscopically characterized. The tert-butoxide ligand facilitates regioselective, intramolecular proton transfer through a CH/OH bond cleavage process occurring at room temperature. Kinetic and thermodynamic data have been obtained by VT NMR experiments; DFT calculations support the observed behavior. Isolation and structural characterization of a doubly dearomatized phosphine complex also strongly supports our mechanistic proposal. The alkoxo-amido complex reacts with water to form a dearomatized ruthenium hydroxide complex, a first step towards phosphine-free metal-ligand cooperative water splitting.

41-Year-Old Male with Sub-Acute Encephalopathy, Seizures, and End Stage Renal Disease: A Unifying Diagnosis and Response to Therapy.

We describe a case of a 41-year-old male with a history of end-stage renal disease, hypertension, epilepsy, ischemic stroke, and traumatic brain injury transferred to our tertiary care center for subacute, progressive cognitive impairment. He was found to have disproportionate brain atrophy, focal seizures, and refractory hypertension. Given suspicion for an underlying genetic etiology, a genetic panel for progressive renal disease was sent, revealing two known pathogenic variants in a gene for a cobalamin metabolism disorder, Cobalamin C deficiency. He was started on targeted metabolic supplementation with subsequent improvement in his cognition. Our case highlights the crucial need to expand diagnostic workup to include genetic and metabolic causes in patients with neurologic disease, atypical features, relevant family history and multi-organ dysfunction.

Perfusion Collateral Index versus Hypoperfusion Intensity Ratio in Assessment of Collaterals in Patients with Acute Ischemic Stroke.

BACKGROUND AND PURPOSE: Perfusion-based collateral indices such as the perfusion collateral index and the hypoperfusion intensity ratio have shown promise in the assessment of collaterals in patients with acute ischemic stroke. We aimed to compare the diagnostic performance of the perfusion collateral index and the hypoperfusion intensity ratio in collateral assessment compared with angiographic collaterals and outcome measures, including final infarct volume, infarct growth, and functional independence. MATERIALS AND METHODS: Consecutive patients with acute ischemic stroke with anterior circulation proximal arterial occlusion who underwent endovascular thrombectomy and had pre- and posttreatment MRI were included. Using pretreatment MR perfusion, we calculated the perfusion collateral index and the hypoperfusion intensity ratio for each patient. The angiographic collaterals obtained from DSA were dichotomized to sufficient (American Society of Interventional and Therapeutic Neuroradiology [ASITN] scale 3-4) versus insufficient (ASITN scale 0-2). The association of collateral status determined by the perfusion collateral index and the hypoperfusion intensity ratio was assessed against angiographic collaterals and outcome measures. RESULTS: A total of 98 patients met the inclusion criteria. Perfusion collateral index values were significantly higher in patients with sufficient angiographic collaterals (P < .001), while there was no significant (P = .46) difference in hypoperfusion intensity ratio values. Among patients with good (mRS 0-2) versus poor (mRS 3-6) functional outcome, the perfusion collateral index of ≥ 62 was present in 72% versus 31% (P = .003), while the hypoperfusion intensity ratio of ≤0.4 was present in 69% versus 56% (P = .52). The perfusion collateral index and the hypoperfusion intensity ratio were both significantly predictive of final infarct volume, but only the perfusion collateral index was significantly (P = .03) associated with infarct growth. CONCLUSIONS: Results show that the perfusion collateral index outperforms the hypoperfusion intensity ratio in the assessment of collateral status, infarct growth, and determination of functional outcomes.

Automated estimation of ischemic core volume on noncontrast-enhanced CT via machine learning.

BACKGROUND: Accurate estimation of ischemic core on baseline imaging has treatment implications in patients with acute ischemic stroke (AIS). Machine learning (ML) algorithms have shown promising results in estimating ischemic core using routine noncontrast computed tomography (NCCT). OBJECTIVE: We used an ML-trained algorithm to quantify ischemic core volume on NCCT in a comparative analysis to pretreatment magnetic resonance imaging (MRI) diffusion-weighted imaging (DWI) in patients with AIS. METHODS: Patients with AIS who had both pretreatment NCCT and MRI were enrolled. An automatic segmentation ML approach was applied using Brainomix software (Oxford, UK) to segment the ischemic voxels and calculate ischemic core volume on NCCT. Ischemic core volume was also calculated on baseline MRI DWI. Comparative analysis was performed using Bland-Altman plots and Pearson correlation. RESULTS: A total of 72 patients were included. The time-to-stroke onset time was 134.2/89.5 minutes (mean/median). The time difference between NCCT and MRI was 64.8/44.5 minutes (mean/median). In patients who presented within 1 hour from stroke onset, the ischemic core volumes were significantly (p = 0.005) underestimated by ML-NCCT. In patients presented beyond 1 hour, the ML-NCCT estimated ischemic core volumes approximated those obtained by MRI-DWI and with significant correlation (r = 0.56, p < 0.001). CONCLUSION: The ischemic core volumes calculated by the described ML approach on NCCT approximate those obtained by MRI in patients with AIS who present beyond 1 hour from stroke onset.

MR Angiography in Assessment of Collaterals in Patients with Acute Ischemic Stroke: A Comparative Analysis with Digital Subtraction Angiography.

Collateral status has prognostic and treatment implications in acute ischemic stroke (AIS) patients. Unlike CTA, grading collaterals on MRA is not well studied. We aimed to evaluate the accuracy of assessing collaterals on pretreatment MRA in AIS patients against DSA. AIS patients with anterior circulation proximal arterial occlusion with baseline MRA and subsequent endovascular treatment were included. MRA collaterals were evaluated by two neuroradiologists independently using the Tan and Maas scoring systems. DSA collaterals were evaluated by using the American Society of Interventional and Therapeutic Neuroradiology grading system and were used as the reference for comparative analysis against MRA. A total of 104 patients met the inclusion criteria (59 female, age (mean ± SD): 70.8 ± 18.1). The inter-rater agreement (k) for collateral scoring was 0.49, 95% CI 0.37-0.61 for the Tan score and 0.44, 95% CI 0.26-0.62 for the Maas score. Total number (%) of sufficient vs. insufficient collaterals based on DSA was 49 (47%) and 55 (53%) respectively. Using the Tan score, 45% of patients with sufficient collaterals and 64% with insufficient collaterals were correctly identified in comparison to DSA, resulting in a poor agreement (0.09, 95% CI 0.1-0.28). Using the Maas score, only 4% of patients with sufficient collaterals and 93% with insufficient collaterals were correctly identified against DSA, resulting in poor agreement (0.03, 95% CI 0.06-0.13). Pretreatment MRA in AIS patients has limited concordance with DSA when grading collaterals using the Tan and Maas scoring systems.

Renal-function change after transjugular intra-hepatic portosystemic shunt placement and its relationship with survival: a single-center experience.

BACKGROUND: The effect of transjugular intra-hepatic portosystemic shunt (TIPS) placement on renal function and the correlation of post-TIPS Cr with mortality remain unclear. This study aimed to assess the effect of TIPS placement on renal function and to examine the relationship between post-TIPS Cr and mortality risk. METHODS: A total of 593 patients who underwent de novo TIPS placement between 2004 and 2017 at a single institution were included in the study. The pre-TIPS Cr level (T0; within 7 days before TIPS placement) and post-TIPS Cr levels, at 1-2 days (T1), 5-12 days (T2), and 15-40 days (T3), were collected. Predictors of Cr change after TIPS placement and the 1-year mortality rate were analysed using multivariable linear-regression and Cox proportional-hazards models, respectively. RESULTS: Overall, 21.4% of patients (n = 127) had elevated baseline Cr (≥1.5 mg/dL; mean, 2.51 ± 1.49 mg/dL) and 78.6% (n = 466) had normal baseline Cr (<1.5 mg/dL; mean, 0.92 ± 0.26 mg/dL). Patients with elevated pre-TIPS Cr demonstrated a decrease in post-TIPS Cr (difference, -0.60 mg/dL), whereas patients with normal baseline Cr exhibited no change (difference, <0.01 mg/dL). The 30-day, 90-day, and 1-year mortality rates were 13%, 20%, and 32%, respectively. Variceal bleeding as a TIPS-placement indication (hazard ratio = 1.731; P = 0.036), higher T0 Cr (hazard ratio = 1.834; P = 0.012), and higher T3 Cr (hazard ratio = 3.524; P < 0.001) were associated with higher 1-year mortality risk. CONCLUSION: TIPS placement improved renal function in patients with baseline renal dysfunction and the post-TIPS Cr level was a strong predictor of 1-year mortality risk.

Gynecologic tumor board: a radiologist's guide to vulvar and vaginal malignancies.

Primary vulvar and vaginal cancers are rare female genital tract malignancies which are staged using the 2009 International Federation of Gynecology and Obstetrics (FIGO) staging. These cancers account for approximately 2,700 deaths annually in the USA. The most common histologic subtype of both vulvar and vaginal cancers is squamous cell carcinoma, with an increasing role of the human papillomavirus (HPV) in a significant number of these tumors. Lymph node involvement is the hallmark of FIGO stage 3 vulvar cancer while pelvic sidewall involvement is the hallmark of FIGO stage 3 vaginal cancer. Imaging techniques include computed tomography (CT), positron emission tomography (PET)-CT, magnetic resonance imaging (MRI), and PET-MRI. MRI is the imaging modality of choice for preoperative clinical staging of nodal and metastatic involvement while PET-CT is helpful with assessing response to neoadjuvant treatment and for guiding patient management. Determining the pretreatment extent of disease has become more important due to modern tailored operative approaches and use of neoadjuvant chemoradiation therapy to reduce surgical morbidity. Moreover, imaging is used to determine the full extent of disease for radiation planning and for evaluating treatment response. Understanding the relevant anatomy of the vulva and vaginal regions and the associated lymphatic pathways is helpful to recognize the potential routes of spread and to correctly identify the appropriate FIGO stage. The purpose of this article is to review the clinical features, pathology, and current treatment strategies for vulvar and vaginal malignancies and to identify multimodality diagnostic imaging features of these gynecologic cancers, in conjunction with its respective 2009 FIGO staging system guidelines.

Increasing metadata coverage of SRA BioSample entries using deep learning-based named entity recognition.

High-quality metadata annotations for data hosted in large public repositories are essential for research reproducibility and for conducting fast, powerful and scalable meta-analyses. Currently, a majority of sequencing samples in the National Center for Biotechnology Information's Sequence Read Archive (SRA) are missing metadata across several categories. In an effort to improve the metadata coverage of these samples, we leveraged almost 44 million attribute-value pairs from SRA BioSample to train a scalable, recurrent neural network that predicts missing metadata via named entity recognition (NER). The network was first trained to classify short text phrases according to 11 metadata categories and achieved an overall accuracy and area under the receiver operating characteristic curve of 85.2% and 0.977, respectively. We then applied our classifier to predict 11 metadata categories from the longer TITLE attribute of samples, evaluating performance on a set of samples withheld from model training. Prediction accuracies were high when extracting sample Genus/Species (94.85%), Condition/Disease (95.65%) and Strain (82.03%) from TITLEs, with lower accuracies and lack of predictions for other categories highlighting multiple issues with the current metadata annotations in BioSample. These results indicate the utility of recurrent neural networks for NER-based metadata prediction and the potential for models such as the one presented here to increase metadata coverage in BioSample while minimizing the need for manual curation. Database URL: https://github.com/cartercompbio/PredictMEE.

IRE1α and IGF signaling predict resistance to an endoplasmic reticulum stress-inducing drug in glioblastoma cells.

To date current therapies of glioblastoma multiforme (GBM) are largely ineffective. The induction of apoptosis by an unresolvable unfolded protein response (UPR) represents a potential new therapeutic strategy. Here we tested 12ADT, a sarcoendoplasmic reticulum Ca2+ ATPase (SERCA) inhibitor, on a panel of unselected patient-derived neurosphere-forming cells and found that GBM cells can be distinguished into "responder" and "non-responder". By RNASeq analysis we found that the non-responder phenotype is significantly linked with the expression of UPR genes, and in particular ERN1 (IRE1) and ATF4. We also identified two additional genes selectively overexpressed among non-responders, IGFBP3 and IGFBP5. CRISPR-mediated deletion of the ERN1, IGFBP3, IGFBP5 signature genes in the U251 human GBM cell line increased responsiveness to 12ADT. Remarkably, >65% of GBM cases in The Cancer Genome Atlas express the non-responder (ERN1, IGFBP3, IGFBP5) gene signature. Thus, elevated levels of IRE1α and IGFBPs predict a poor response to drugs inducing unresolvable UPR and possibly other forms of chemotherapy helping in a better stratification GBM patients.

Quantitative Translation of Dog-to-Human Aging by Conserved Remodeling of the DNA Methylome.

All mammals progress through similar physiological stages throughout life, from early development to puberty, aging, and death. Yet, the extent to which this conserved physiology reflects underlying genomic events is unclear. Here, we map the common methylation changes experienced by mammalian genomes as they age, focusing on comparison of humans with dogs, an emerging model of aging. Using oligo-capture sequencing, we characterize methylomes of 104 Labrador retrievers spanning a 16-year age range, achieving >150× coverage within mammalian syntenic blocks. Comparison with human methylomes reveals a nonlinear relationship that translates dog-to-human years and aligns the timing of major physiological milestones between the two species, with extension to mice. Conserved changes center on developmental gene networks, which are sufficient to translate age and the effects of anti-aging interventions across multiple mammals. These results establish methylation not only as a diagnostic age readout but also as a cross-species translator of physiological aging milestones.

Extracting allelic read counts from 250,000 human sequencing runs in Sequence Read Archive.

The Sequence Read Archive (SRA) contains over one million publicly available sequencing runs from various studies using a variety of sequencing library strategies. These data inherently contain information about underlying genomic sequence variants which we exploit to extract allelic read counts on an unprecedented scale. We reprocessed over 250,000 human sequencing runs (>1000 TB data worth of raw sequence data) into a single unified dataset of allelic read counts for nearly 300,000 variants of biomedical relevance curated by NCBI dbSNP, where germline variants were detected in a median of 912 sequencing runs, and somatic variants were detected in a median of 4,876 sequencing runs, suggesting that this dataset facilitates identification of sequencing runs that harbor variants of interest. Allelic read counts obtained using a targeted alignment were very similar to read counts obtained from whole-genome alignment. Analyzing allelic read count data for matched DNA and RNA samples from tumors, we find that RNA-seq can also recover variants identified by Whole Exome Sequencing (WXS), suggesting that reprocessed allelic read counts can support variant detection across different library strategies in SRA. This study provides a rich database of known human variants across SRA samples that can support future meta-analyses of human sequence variation.

PNN' & P2NN' ligands via reductive amination with phosphine aldehydes: synthesis and base-metal coordination chemistry.

Novel PNN' & P2NN' ligands based on 2-aminopyridine (APyPNN-R) R = Ph (1a), Cy (1b), iBu (1c), 8-aminoquinoline (AQPNN-R) R = Ph (2a), Cy (2b), iBu (2c), iPr (2d), and 2-picolylamine (P2NN-R) R = Ph (3a), Cy (3b), iBu (3c), have been synthesized via a versatile, one-pot, single-step, reductive amination of tertiary phosphine acetaldehydes with the amine by reaction with STAB (where STAB is sodium(triacetoxy)borohydride). Ligands 1b and 1c bridge between paramagnetic Co(ii) and form dimeric complexes Co2Cl4(APyPNN-R)2 (4 and 5) when reacted with cobalt dichloride. Ligands 2a-c coordinate in a tridentate fashion forming chelate complexes MCl2(AQPNN-R) M = Co(ii) (6-8), and, for 2d, the Fe(ii) complex FeCl2(AQPNN-iPr) (9). A solution magnetic susceptibility value for 9 of 3.9µB is consistent with a monomer-dimer equilibrium. The synthesis of the dimeric complex [FeCl2(AQPNN-Ph)]2 (10) using 2a as well as solid state magnetic susceptibility measurements on 9 and 10 confirm this phenomenon. Ligand 3a coordinates to Fe(ii) in an interesting tetradentate fashion despite bearing a tertiary amine moiety giving octahedral FeCl2(P2NN') (11). All of the metal complexes have been characterized by elemental analysis, paramagnetic 1H NMR spectroscopy, solution magnetic susceptibility, and single crystal X-ray diffraction (XRD).

Evaluation and accurate diagnoses of pediatric diseases using artificial intelligence.

Artificial intelligence (AI)-based methods have emerged as powerful tools to transform medical care. Although machine learning classifiers (MLCs) have already demonstrated strong performance in image-based diagnoses, analysis of diverse and massive electronic health record (EHR) data remains challenging. Here, we show that MLCs can query EHRs in a manner similar to the hypothetico-deductive reasoning used by physicians and unearth associations that previous statistical methods have not found. Our model applies an automated natural language processing system using deep learning techniques to extract clinically relevant information from EHRs. In total, 101.6 million data points from 1,362,559 pediatric patient visits presenting to a major referral center were analyzed to train and validate the framework. Our model demonstrates high diagnostic accuracy across multiple organ systems and is comparable to experienced pediatricians in diagnosing common childhood diseases. Our study provides a proof of concept for implementing an AI-based system as a means to aid physicians in tackling large amounts of data, augmenting diagnostic evaluations, and to provide clinical decision support in cases of diagnostic uncertainty or complexity. Although this impact may be most evident in areas where healthcare providers are in relative shortage, the benefits of such an AI system are likely to be universal.

Extracellular vesicles produced in B cells deliver tumor suppressor miR-335 to breast cancer cells disrupting oncogenic programming in vitro and in vivo.

The successful implementation of miRNA (miR) therapies in humans will ultimately rely on the use of vehicles with improved cellular delivery capability. Here we tested a new system that leverages extracellular vesicles (EVs) laden with a tumor suppressor miRNA (miR-335) produced in B cells by plasmid DNA induction (iEVs). We demonstrate that iEVs-335 efficiently and durably restored the endogenous miR-335 pool in human triple negative breast cancer cells, downregulated the expression of the miR-335 target gene SOX4 transcription factor, and markedly inhibited tumor growth in vivo. Remarkably, iEVs-335 mediated transcriptional effects that persisted in tumors after 60 days post orthotopic implantation. Genome-wide RNASeq analysis of cancer cells treated in vitro with iEVs-335 showed the regulation of a discrete number of genes only, without broad transcriptome perturbations. This new technology may be ideally suited for therapies aimed to restore tumor suppressor miRNAs in cancer cells, disrupting the oncogenic program established after escape from miRNA control.