Search

Mismatch repair deficiency (d-MMR)/microsatellite instability (MSI), KRAS, and BRAF mutational status are crucial for treating advanced colorectal cancer patients. Traditional methods like immunohistochemistry or polymerase chain reaction (PCR) can be challenged by artificial intelligence (AI) based on whole slide images (WSI) to predict tumor status. In this systematic review, we evaluated the role of AI in predicting MSI status, KRAS, and BRAF mutations in colorectal cancer. Studies published in PubMed up to June 2023 were included (n = 17), and we reported the risk of bias and the performance for each study. Some studies were impacted by the reduced number of slides included in the data set and the lack of external validation cohorts. Deep learning models for the d-MMR/MSI status showed a good performance in training cohorts (mean AUC = 0.89, [0.74–0.97]) but slightly less than expected in the validation cohort when available (mean AUC = 0.82, [0.63–0.98]). Contrary to the MSI status, the prediction of KRAS and BRAF mutations was less explored with a less robust methodology. The performance was lower, with a maximum of 0.77 in the training cohort, 0.58 in the validation cohort for KRAS, and 0.82 AUC in the training cohort for BRAF.

Accurate testing for epidermal growth factor receptor (EGFR) variants is essential for informing treatment decisions in non-small cell lung cancer (NSCLC). Automated diagnostic workflows may allow more streamlined initiation of targeted treatments, where appropriate, while comprehensive variant analysis is ongoing. FACILITATE, a real-world, prospective, multicenter, European study, evaluated performance and analytical turnaround time of the Idylla™ EGFR Mutation Test compared with local reference methods. Sixteen sites obtained formalin-fixed paraffin-embedded biopsy samples with ≥ 10% neoplastic cells from patients with NSCLC. Consecutive 5 μm sections from patient samples were tested for clinically relevant NSCLC-associated EGFR variants using the Idylla™ EGFR Mutation Test and local reference methods; performance (concordance) and analytical turnaround time were compared. Between January 2019 and November 2020, 1,474 parallel analyses were conducted. Overall percentage agreement was 97.7% [n = 1,418; 95% confidence interval (CI): 96.8–98.3], positive agreement, 87.4% (n = 182; 95% CI: 81.8–91.4) and negative agreement, 99.2% (n = 1,236; 95% CI: 98.5–99.6). There were 38 (2.6%) discordant cases. Ninety percent of results were returned with an analytical turnaround time of within 1 week using the Idylla™ EGFR Mutation Test versus ∼22 days using reference methods. The Idylla™ EGFR Mutation Test performed well versus local methods and had shorter analytical turnaround time. The Idylla™ EGFR Mutation Test can thus support application of personalized medicine in NSCLC.

Abstract Background Artificial intelligence (AI) has the potential to transform our healthcare systems significantly. New AI technologies based on machine learning approaches should play a key role in clinical decision-making in the future. However, their implementation in health care settings remains limited, mostly due to a lack of robust validation procedures. There is a need to develop reliable assessment frameworks for the clinical validation of AI. We present here an approach for assessing AI for predicting treatment response in triple-negative breast cancer (TNBC), using real-world data and molecular -omics data from clinical data warehouses and biobanks. Methods The European “ITFoC (Information Technology for the Future Of Cancer)” consortium designed a framework for the clinical validation of AI technologies for predicting treatment response in oncology. Results This framework is based on seven key steps specifying: (1) the intended use of AI, (2) the target population, (3) the timing of AI evaluation, (4) the datasets used for evaluation, (5) the procedures used for ensuring data safety (including data quality, privacy and security), (6) the metrics used for measuring performance, and (7) the procedures used to ensure that the AI is explainable. This framework forms the basis of a validation platform that we are building for the “ITFoC Challenge”. This community-wide competition will make it possible to assess and compare AI algorithms for predicting the response to TNBC treatments with external real-world datasets. Conclusions The predictive performance and safety of AI technologies must be assessed in a robust, unbiased and transparent manner before their implementation in healthcare settings. We believe that the consideration of the ITFoC consortium will contribute to the safe transfer and implementation of AI in clinical settings, in the context of precision oncology and personalized care.

Abstract Circulating tumor DNA (ctDNA) has become an attractive biomarker in human oncology, and its use may be informative in canine cancer. Thus, we used droplet digital PCR or PCR for antigen receptor rearrangement, to explore tumor-specific point mutations, copy number alterations, and chromosomal rearrangements in the plasma of cancer-affected dogs. We detected ctDNA in 21/23 (91.3%) of histiocytic sarcoma (HS), 2/8 (25%) of oral melanoma, and 12/13 (92.3%) of lymphoma cases. The utility of ctDNA in diagnosing HS was explored in 133 dogs, including 49 with HS, and the screening of recurrent PTPN11 mutations in plasma had a specificity of 98.8% and a sensitivity between 42.8 and 77% according to the clinical presentation of HS. Sensitivity was greater in visceral forms and especially related to pulmonary location. Follow-up of four dogs by targeting lymphoma-specific antigen receptor rearrangement in plasma showed that minimal residual disease detection was concordant with clinical evaluation and treatment response. Thus, our study shows that ctDNA is detectable in the plasma of cancer-affected dogs and is a promising biomarker for diagnosis and clinical follow-up. ctDNA detection appears to be useful in comparative oncology research due to growing interest in the study of natural canine tumors and exploration of new therapies.

Background: A large yet heterogeneous body of literature exists suggesting that endothelial dysfunction appears early in type 1 diabetes, due to hyperglycemia-induced oxidative stress. The latter may also affect vascular smooth muscles (VSM) function, a layer albeit less frequently considered in that pathology. This meta-analysis aims at evaluating the extent, and the contributing risk factors, of early endothelial dysfunction, and of the possible concomitant VSM dysfunction, in type 1 diabetes.Methods: PubMed, Web of Sciences, Cochrane Library databases were screened from their respective inceptions until October 2019. We included studies comparing vasodilatory capacity depending or not on endothelium (i.e., endothelial function or VSM function, respectively) in patients with uncomplicated type 1 diabetes and healthy controls.Results: Fifty-eight articles studying endothelium-dependent function, among which 21 studies also assessed VSM, were included. Global analyses revealed an impairment of standardized mean difference (SMD) (Cohen's d) of endothelial function: −0.61 (95% CI: −0.79, −0.44) but also of VSM SMD: −0.32 (95% CI: −0.57, −0.07). The type of stimuli used (i.e., exercise, occlusion-reperfusion, pharmacological substances, heat) did not influence the impairment of the vasodilatory capacity. Endothelial dysfunction appeared more pronounced within macrovascular than microvascular beds. The latter was particularly altered in cases of poor glycemic control [HbA1c > 67 mmol/mol (8.3%)].Conclusions: This meta-analysis not only corroborates the presence of an early impairment of endothelial function, even in response to physiological stimuli like exercise, but also highlights a VSM dysfunction in children and adults with type 1 diabetes. Endothelial dysfunction seems to be more pronounced in large than small vessels, fostering the debate on their relative temporal appearance.

Mismatch repair deficiency (d-MMR)/microsatellite instability (MSI), KRAS, and BRAF mutational status are crucial for treating advanced colorectal cancer patients. Traditional methods like immunohistochemistry or polymerase chain reaction (PCR) can be challenged by artificial intelligence (AI) based on whole slide images (WSI) to predict tumor status. In this systematic review, we evaluated the role of AI in predicting MSI status, KRAS, and BRAF mutations in colorectal cancer. Studies published in PubMed up to June 2023 were included (n = 17), and we reported the risk of bias and the performance for each study. Some studies were impacted by the reduced number of slides included in the data set and the lack of external validation cohorts. Deep learning models for the d-MMR/MSI status showed a good performance in training cohorts (mean AUC = 0.89, [0.74–0.97]) but slightly less than expected in the validation cohort when available (mean AUC = 0.82, [0.63–0.98]). Contrary to the MSI status, the prediction of KRAS and BRAF mutations was less explored with a less robust methodology. The performance was lower, with a maximum of 0.77 in the training cohort, 0.58 in the validation cohort for KRAS, and 0.82 AUC in the training cohort for BRAF. [ABSTRACT FROM AUTHOR]

Radio bursts are sensitive tracers of non-thermal electron populations in the solar corona. They are produced by electron beams and shock waves propagating through the corona and the heliosphere, and by trapped electron populations in coronal mass ejections (CMEs) and in quiescent active regions. Combining space-borne and ground-based radio spectrographs allows one to track disturbances between the low corona, near or at the sites of particle acceleration, and the spacecraft. Radio observations are, therefore, a significant tool in probing the solar origin of heliospheric disturbances, which is a central research topic as witnessed by the Parker Solar Probe and Solar Orbiter missions. The full scientific return of these projects needs vigorous ground-based support, which at radio wavelengths covers altitudes up to about a solar radius above the photosphere. Besides research in solar and heliospheric physics, monitoring solar radio bursts also supports space weather services. On occasion, radio bursts can themselves be a space weather hazard. The Nançay radio astronomy station in central France has a long tradition of monitoring radio emission at decimetre-to-meter wavelengths. This article describes the radio spectrograph ORFEES (Observations Radiospectrographiques pour FEDOME et l’Etude des Eruptions Solaires). It observes the whole-Sun flux density between 144 and 1004 MHz, pertaining to regions between the low corona and about half a solar radius above the photosphere. ORFEES results from a partnership between Observatoire de Paris and the French Air Force, which operates the experimental space weather service FEDOME. The primary use of the instrument at the Paris Observatory is astrophysical observation. Low-resolution data with rapid availability are presently produced for the French Air Force. Similar information can be made available to a broader range of space weather service providers. This article gives an overview of the instrument design and access to the data and shows a few illustrative observations.