METhodological RadiomICs Score (METRICS): a quality scoring tool for radiomics research endorsed by EuSoMII.

PURPOSE: To propose a new quality scoring tool, METhodological RadiomICs Score (METRICS), to assess and improve research quality of radiomics studies. METHODS: We conducted an online modified Delphi study with a group of international experts. It was performed in three consecutive stages: Stage#1, item preparation; Stage#2, panel discussion among EuSoMII Auditing Group members to identify the items to be voted; and Stage#3, four rounds of the modified Delphi exercise by panelists to determine the items eligible for the METRICS and their weights. The consensus threshold was 75%. Based on the median ranks derived from expert panel opinion and their rank-sum based conversion to importance scores, the category and item weights were calculated. RESULT: In total, 59 panelists from 19 countries participated in selection and ranking of the items and categories. Final METRICS tool included 30 items within 9 categories. According to their weights, the categories were in descending order of importance: study design, imaging data, image processing and feature extraction, metrics and comparison, testing, feature processing, preparation for modeling, segmentation, and open science. A web application and a repository were developed to streamline the calculation of the METRICS score and to collect feedback from the radiomics community. CONCLUSION: In this work, we developed a scoring tool for assessing the methodological quality of the radiomics research, with a large international panel and a modified Delphi protocol. With its conditional format to cover methodological variations, it provides a well-constructed framework for the key methodological concepts to assess the quality of radiomic research papers. CRITICAL RELEVANCE STATEMENT: A quality assessment tool, METhodological RadiomICs Score (METRICS), is made available by a large group of international domain experts, with transparent methodology, aiming at evaluating and improving research quality in radiomics and machine learning. KEY POINTS: • A methodological scoring tool, METRICS, was developed for assessing the quality of radiomics research, with a large international expert panel and a modified Delphi protocol. • The proposed scoring tool presents expert opinion-based importance weights of categories and items with a transparent methodology for the first time. • METRICS accounts for varying use cases, from handcrafted radiomics to entirely deep learning-based pipelines. • A web application has been developed to help with the calculation of the METRICS score ( https://metricsscore.github.io/metrics/METRICS.html ) and a repository created to collect feedback from the radiomics community ( https://github.com/metricsscore/metrics ).

O-RADS MRI scoring system has the potential to reduce the frequency of avoidable adnexal surgery.

OBJECTIVE: To assess the potential impact of the O-RADS MRI score on the decision-making process for the management of adnexal masses. METHODS: EURAD database (prospective, European observational, multicenter study) was queried to identify asymptomatic women without history of infertility included between March 1st and March 31st 2018, with available surgical pathology or clinical findings at 2-year clinical follow-up. Blinded to final diagnosis, we stratified patients into five categories according to the O-RADS MRI score (absent i.e. non adnexal, benign, probably benign, indeterminate, probably malignant). Prospective management was compared to theoretical management according to the score established as following: those with presumed benign masses (scored O-RADS MRI 2 or 3) (follow-up recommended) and those with presumed malignant masses (scored O-RADS MRI 4 or 5) (surgery recommended). RESULTS: The accuracy of the score for assessing the origin of the mass was of 97.2 % (564/580, CI95% 0.96-0.98) and was of 92.0 % (484/526) for categorizing lesions with a negative predictive value of 98.1 % (415/423, CI95% 0.96-0.99). Theoretical management using the score would have spared surgery in 229 patients (87.1 %, 229/263) with benign lesions and malignancy would have been missed in 6 borderline and 2 invasive cases. In patients with a presumed benign mass using O-RADS MRI score, recommending surgery for lesions >= 100 mm would miss only 4/77 (4.8 %) malignant adnexal tumors instead of 8 (50 % decrease). CONCLUSION: The use of O-RADS MRI scoring system could drastically reduce the number of asymptomatic patients undergoing avoidable surgery.

Longitudinal Study of Advanced Non-Small Cell Lung Cancer with Initial Durable Clinical Benefit to Immunotherapy: Strategies for Anti-PD-1/PD-L1 Continuation beyond Progression.

BACKGROUND AND AIM: A better understanding of resistance to checkpoint inhibitors is essential to define subsequent treatments in advanced non-small cell lung cancer. By characterizing clinical and radiological features of progression after anti-programmed death-1/programmed death ligand-1 (anti-PD-1/PD-L1), we aimed to define therapeutic strategies in patients with initial durable clinical benefit. PATIENTS AND METHODS: This monocentric, retrospective study included patients who presented progressive disease (PD) according to RECIST 1.1 criteria after anti-PD-1/PD-L1 monotherapy. Patients were classified into two groups, "primary resistance" and "Progressive Disease (PD) after Durable Clinical Benefit (DCB)", according to the Society of Immunotherapy of Cancer classification. We compared the post-progression survival (PPS) of both groups and analyzed the patterns of progression. An exploratory analysis was performed using the tumor growth rate (TGR) to assess the global growth kinetics of cancer and the persistent benefit of immunotherapy beyond PD after DCB. RESULTS: A total of 148 patients were included; 105 of them presented "primary resistance" and 43 "PD after DCB". The median PPS was 5.2 months (95% CI: 2.6-6.5) for primary resistance (p < 0.0001) vs. 21.3 months (95% CI: 18.5-36.3) for "PD after DCB", and the multivariable hazard ratio was 0.14 (95% CI: 0.07-0.30). The oligoprogression pattern was frequent in the "PD after DCB" group (76.7%) and occurred mostly in pre-existing lesions (72.1%). TGR deceleration suggested a persistent benefit of PD-1/PD-L1 blockade in 44.2% of cases. CONCLUSIONS: PD after DCB is an independent factor of longer post-progression survival with specific patterns that prompt to contemplate loco-regional treatments. TGR is a promising tool to assess the residual benefit of immunotherapy and justify the continuation of immunotherapy in addition to radiotherapy or surgery.

A European Society of Oncologic Imaging (ESOI) survey on the radiological assessment of response to oncologic treatments in clinical practice.

OBJECTIVES: To present the results of a survey on the assessment of treatment response with imaging in oncologic patient, in routine clinical practice. The survey was promoted by the European Society of Oncologic Imaging to gather information for the development of reporting models and recommendations. METHODS: The survey was launched on the European Society of Oncologic Imaging website and was available for 3 weeks. It consisted of 5 sections, including 24 questions related to the following topics: demographic and professional information, methods for lesion measurement, how to deal with diminutive lesions, how to report baseline and follow-up examinations, which previous studies should be used for comparison, and role of RECIST 1.1 criteria in the daily clinical practice. RESULTS: A total of 286 responses were received. Most responders followed the RECIST 1.1 recommendations for the measurement of target lesions and lymph nodes and for the assessment of tumor response. To assess response, 48.6% used previous and/or best response study in addition to baseline, 25.2% included the evaluation of all main time points, and 35% used as the reference only the previous study. A considerable number of responders used RECIST 1.1 criteria in daily clinical practice (41.6%) or thought that they should be always applied (60.8%). CONCLUSION: Since standardized criteria are mainly a prerogative of clinical trials, in daily routine, reporting strategies are left to radiologists and oncologists, which may issue local and diversified recommendations. The survey emphasizes the need for more generally applicable rules for response assessment in clinical practice. CRITICAL RELEVANCE STATEMENT: Compared to clinical trials which use specific criteria to evaluate response to oncological treatments, the free narrative report usually adopted in daily clinical practice may lack clarity and useful information, and therefore, more structured approaches are needed. KEY POINTS: · Most radiologists consider standardized reporting strategies essential for an objective assessment of tumor response in clinical practice. · Radiologists increasingly rely on RECIST 1.1 in their daily clinical practice. · Treatment response evaluation should require a complete analysis of all imaging time points and not only of the last.

Best imaging signs identified by radiomics could outperform the model: application to differentiating lung carcinoid tumors from atypical hamartomas.

OBJECTIVES: Lung carcinoids and atypical hamartomas may be difficult to differentiate but require different treatment. The aim was to differentiate these tumors using contrast-enhanced CT semantic and radiomics criteria. METHODS: Between November 2009 and June 2020, consecutives patient operated for hamartomas or carcinoids with contrast-enhanced chest-CT were retrospectively reviewed. Semantic criteria were recorded and radiomics features were extracted from 3D segmentations using Pyradiomics. Reproducible and non-redundant radiomics features were used to training a random forest algorithm with cross-validation. A validation-set from another institution was used to evaluate of the radiomics signature, the 3D 'median' attenuation feature (3D-median) alone and the mean value from 2D-ROIs. RESULTS: Seventy-three patients (median 58 years [43‒70]) were analyzed (16 hamartomas; 57 carcinoids). The radiomics signature predicted hamartomas vs carcinoids on the external dataset (22 hamartomas; 32 carcinoids) with an AUC = 0.76. The 3D-median was the most important in the model. Density thresholds < 10 HU to predict hamartoma and > 60 HU to predict carcinoids were chosen for their high specificity > 0.90. On the external dataset, sensitivity and specificity of the 3D-median and 2D-ROIs were, respectively, 0.23, 1.00 and 0.13, 1.00 < 10 HU; 0.63, 0.95 and 0.69, 0.91 > 60 HU. The 3D-median was more reproducible than 2D-ROIs (ICC = 0.97 95% CI [0.95‒0.99]; bias: 3 ± 7 HU limits of agreement (LoA) [- 10‒16] vs. ICC = 0.90 95% CI [0.85‒0.94]; bias: - 0.7 ± 21 HU LoA [- 4‒40], respectively). CONCLUSIONS: A radiomics signature can distinguish hamartomas from carcinoids with an AUC = 0.76. Median density < 10 HU and > 60 HU on 3D or 2D-ROIs may be useful in clinical practice to diagnose these tumors with confidence, but 3D is more reproducible. CRITICAL RELEVANCE STATEMENT: Radiomic features help to identify the most discriminating imaging signs using random forest. 'Median' attenuation value (Hounsfield units), extracted from 3D-segmentations on contrast-enhanced chest-CTs, could distinguish carcinoids from atypical hamartomas (AUC = 0.85), was reproducible (ICC = 0.97), and generalized to an external dataset. KEY POINTS: • 3D-'Median' was the best feature to differentiate carcinoids from atypical hamartomas (AUC = 0.85). • 3D-'Median' feature is reproducible (ICC = 0.97) and was generalized to an external dataset. • Radiomics signature from 3D-segmentations differentiated carcinoids from atypical hamartomas with an AUC = 0.76. • 2D-ROI value reached similar performance to 3D-'median' but was less reproducible (ICC = 0.90).

Designing clinical trials based on modern imaging and metastasis-directed treatments in patients with oligometastatic breast cancer: a consensus recommendation from the EORTC Imaging and Breast Cancer Groups.

Breast cancer remains the most common cause of cancer death among women. Despite its considerable histological and molecular heterogeneity, those characteristics are not distinguished in most definitions of oligometastatic disease and clinical trials of oligometastatic breast cancer. After an exhaustive review of the literature covering all aspects of oligometastatic breast cancer, 35 experts from the European Organisation for Research and Treatment of Cancer Imaging and Breast Cancer Groups elaborated a Delphi questionnaire aimed at offering consensus recommendations, including oligometastatic breast cancer definition, optimal diagnostic pathways, and clinical trials required to evaluate the effect of diagnostic imaging strategies and metastasis-directed therapies. The main recommendations are the introduction of modern imaging methods in metastatic screening for an earlier diagnosis of oligometastatic breast cancer and the development of prospective trials also considering the histological and molecular complexity of breast cancer. Strategies for the randomisation of imaging methods and therapeutic approaches in different subsets of patients are also addressed.

Comparative performances of machine learning algorithms in radiomics and impacting factors.

There are no current recommendations on which machine learning (ML) algorithms should be used in radiomics. The objective was to compare performances of ML algorithms in radiomics when applied to different clinical questions to determine whether some strategies could give the best and most stable performances regardless of datasets. This study compares the performances of nine feature selection algorithms combined with fourteen binary classification algorithms on ten datasets. These datasets included radiomics features and clinical diagnosis for binary clinical classifications including COVID-19 pneumonia or sarcopenia on CT, head and neck, orbital or uterine lesions on MRI. For each dataset, a train-test split was created. Each of the 126 (9 × 14) combinations of feature selection algorithms and classification algorithms was trained and tuned using a ten-fold cross validation, then AUC was computed. This procedure was repeated three times per dataset. Best overall performances were obtained with JMI and JMIM as feature selection algorithms and random forest and linear regression models as classification algorithms. The choice of the classification algorithm was the factor explaining most of the performance variation (10% of total variance). The choice of the feature selection algorithm explained only 2% of variation, while the train-test split explained 9%.

Strategies to safely rule out pulmonary embolism in COVID-19 outpatients: a multicenter retrospective study.

OBJECTIVES: The objective was to define a safe strategy to exclude pulmonary embolism (PE) in COVID-19 outpatients, without performing CT pulmonary angiogram (CTPA). METHODS: COVID-19 outpatients from 15 university hospitals who underwent a CTPA were retrospectively evaluated. D-Dimers, variables of the revised Geneva and Wells scores, as well as laboratory findings and clinical characteristics related to COVID-19 pneumonia, were collected. CTPA reports were reviewed for the presence of PE and the extent of COVID-19 disease. PE rule-out strategies were based solely on D-Dimer tests using different thresholds, the revised Geneva and Wells scores, and a COVID-19 PE prediction model built on our dataset were compared. The area under the receiver operating characteristics curve (AUC), failure rate, and efficiency were calculated. RESULTS: In total, 1369 patients were included of whom 124 were PE positive (9.1%). Failure rate and efficiency of D-Dimer > 500 µg/l were 0.9% (95%CI, 0.2-4.8%) and 10.1% (8.5-11.9%), respectively, increasing to 1.0% (0.2-5.3%) and 16.4% (14.4-18.7%), respectively, for an age-adjusted D-Dimer level. D-dimer > 1000 µg/l led to an unacceptable failure rate to 8.1% (4.4-14.5%). The best performances of the revised Geneva and Wells scores were obtained using the age-adjusted D-Dimer level. They had the same failure rate of 1.0% (0.2-5.3%) for efficiency of 16.8% (14.7-19.1%), and 16.9% (14.8-19.2%) respectively. The developed COVID-19 PE prediction model had an AUC of 0.609 (0.594-0.623) with an efficiency of 20.5% (18.4-22.8%) when its failure was set to 0.8%. CONCLUSIONS: The strategy to safely exclude PE in COVID-19 outpatients should not differ from that used in non-COVID-19 patients. The added value of the COVID-19 PE prediction model is minor. KEY POINTS: • D-dimer level remains the most important predictor of pulmonary embolism in COVID-19 patients. • The AUCs of the revised Geneva and Wells scores using an age-adjusted D-dimer threshold were 0.587 (95%CI, 0.572 to 0.603) and 0.588 (95%CI, 0.572 to 0.603). • The AUC of COVID-19-specific strategy to rule out pulmonary embolism ranged from 0.513 (95%CI: 0.503 to 0.522) to 0.609 (95%CI: 0.594 to 0.623).

Artificial intelligence in musculoskeletal oncology imaging: A critical review of current applications.

Artificial intelligence (AI) is increasingly being studied in musculoskeletal oncology imaging. AI has been applied to both primary and secondary bone tumors and assessed for various predictive tasks that include detection, segmentation, classification, and prognosis. Still, in the field of clinical research, further efforts are needed to improve AI reproducibility and reach an acceptable level of evidence in musculoskeletal oncology. This review describes the basic principles of the most common AI techniques, including machine learning, deep learning and radiomics. Then, recent developments and current results of AI in the field of musculoskeletal oncology are presented. Finally, limitations and future perspectives of AI in this field are discussed.

MRI Evaluation of Uterine Masses for Risk of Leiomyosarcoma: A Consensus Statement.

Laparoscopic myomectomy, a common gynecologic operation in premenopausal women, has become heavily regulated since 2014 following the dissemination of unsuspected uterine leiomyosarcoma (LMS) throughout the pelvis of a physician treated for symptomatic leiomyoma. Research since that time suggests a higher prevalence than previously suspected of uterine LMS in resected masses presumed to represent leiomyoma, as high as one in 770 women (0.13%). Though rare, the dissemination of an aggressive malignant neoplasm due to noncontained electromechanical morcellation in laparoscopic myomectomy is a devastating outcome. Gynecologic surgeons' desire for an evidence-based, noninvasive evaluation for LMS is driven by a clear need to avoid such harms while maintaining the availability of minimally invasive surgery for symptomatic leiomyoma. Laparoscopic gynecologists could rely upon the distinction of higher-risk uterine masses preoperatively to plan oncologic surgery (ie, potential hysterectomy) for patients with elevated risk for LMS and, conversely, to safely offer women with no or minimal indicators of elevated risk the fertility-preserving laparoscopic myomectomy. MRI evaluation for LMS may potentially serve this purpose in symptomatic women with leiomyomas. This evidence review and consensus statement defines imaging and disease-related terms to allow more uniform and reliable interpretation and identifies the highest priorities for future research on LMS evaluation.

Attitudes and perceptions of radiologists towards online (virtual) oncologic multidisciplinary team meetings during the COVID-19 pandemic-a survey of the European Society of Oncologic Imaging (ESOI).

OBJECTIVES: To explore radiologists' opinions regarding the shift from in-person oncologic multidisciplinary team meetings (MDTMs) to online MDTMs. To assess the perceived impact of online MDTMs, and to evaluate clinical and technical aspects of online meetings. METHODS: An online questionnaire including 24 questions was e-mailed to all European Society of Oncologic Imaging (ESOI) members. Questions targeted the structure and efficacy of online MDTMs, including benefits and limitations. RESULTS: A total of 204 radiologists responded to the survey. Responses were evaluated using descriptive statistical analysis. The majority (157/204; 77%) reported a shift to online MDTMs at the start of the pandemic. For the most part, this transition had a positive effect on maintaining and improving attendance. The majority of participants reported that online MDTMs provide the same clinical standard as in-person meetings, and that interdisciplinary discussion and review of imaging data were not hindered. Seventy three of 204 (35.8%) participants favour reverting to in-person MDTs, once safe to do so, while 7/204 (3.4%) prefer a continuation of online MDTMs. The majority (124/204, 60.8%) prefer a combination of physical and online MDTMs. CONCLUSIONS: Online MDTMs are a viable alternative to in-person meetings enabling continued timely high-quality provision of care with maintained coordination between specialties. They were accepted by the majority of surveyed radiologists who also favoured their continuation after the pandemic, preferably in combination with in-person meetings. An awareness of communication issues particular to online meetings is important. Training, improved software, and availability of support are essential to overcome technical and IT difficulties reported by participants. KEY POINTS: • Majority of surveyed radiologists reported shift from in-person to online oncologic MDT meetings during the COVID-19 pandemic. • The shift to online MDTMs was feasible and generally accepted by the radiologists surveyed with the majority reporting that online MDTMs provide the same clinical standard as in-person meetings. • Most would favour the return to in-person MDTMs but would also accept the continued use of online MDTMs following the end of the current pandemic.

Standardised lesion segmentation for imaging biomarker quantitation: a consensus recommendation from ESR and EORTC.

BACKGROUND: Lesion/tissue segmentation on digital medical images enables biomarker extraction, image-guided therapy delivery, treatment response measurement, and training/validation for developing artificial intelligence algorithms and workflows. To ensure data reproducibility, criteria for standardised segmentation are critical but currently unavailable. METHODS: A modified Delphi process initiated by the European Imaging Biomarker Alliance (EIBALL) of the European Society of Radiology (ESR) and the European Organisation for Research and Treatment of Cancer (EORTC) Imaging Group was undertaken. Three multidisciplinary task forces addressed modality and image acquisition, segmentation methodology itself, and standards and logistics. Devised survey questions were fed via a facilitator to expert participants. The 58 respondents to Round 1 were invited to participate in Rounds 2-4. Subsequent rounds were informed by responses of previous rounds. RESULTS/CONCLUSIONS: Items with ≥ 75% consensus are considered a recommendation. These include system performance certification, thresholds for image signal-to-noise, contrast-to-noise and tumour-to-background ratios, spatial resolution, and artefact levels. Direct, iterative, and machine or deep learning reconstruction methods, use of a mixture of CE marked and verified research tools were agreed and use of specified reference standards and validation processes considered essential. Operator training and refreshment were considered mandatory for clinical trials and clinical research. Items with a 60-74% agreement require reporting (site-specific accreditation for clinical research, minimal pixel number within lesion segmented, use of post-reconstruction algorithms, operator training refreshment for clinical practice). Items with ≤ 60% agreement are outside current recommendations for segmentation (frequency of system performance tests, use of only CE-marked tools, board certification of operators, frequency of operator refresher training). Recommendations by anatomical area are also specified.

Extraction of Tumor Response Criteria in Semi-Structured Imaging Report.

In this study, we extracted information from 6,376 french CT scan semi-structured text reports evaluating the cancer treatment response using the RECIST methodology. We evaluated the performance against manual annotation of 100 reports and measured the evolution of the presence of information over time. The results show high performances of the extraction as well as trends.

Nivolumab, nivolumab-ipilimumab, and VEGFR-tyrosine kinase inhibitors as first-line treatment for metastatic clear-cell renal cell carcinoma (BIONIKK): a biomarker-driven, open-label, non-comparative, randomised, phase 2 trial.

BACKGROUND: We previously reported a 35-gene expression classifier identifying four clear-cell renal cell carcinoma groups (ccrcc1 to ccrcc4) with different tumour microenvironments and sensitivities to sunitinib in metastatic clear-cell renal cell carcinoma. Efficacy profiles might differ with nivolumab and nivolumab-ipilimumab. We therefore aimed to evaluate treatment efficacy and tolerability of nivolumab, nivolumab-ipilimumab, and VEGFR-tyrosine kinase inhibitors (VEGFR-TKIs) in patients according to tumour molecular groups. METHODS: This biomarker-driven, open-label, non-comparative, randomised, phase 2 trial included patients from 15 university hospitals or expert cancer centres in France. Eligible patients were aged 18 years or older, had an Eastern Cooperative Oncology Group performance status of 0-2, and had previously untreated metastatic clear-cell renal cell carcinoma. Patients were randomly assigned (1:1) using permuted blocks of varying sizes to receive either nivolumab or nivolumab-ipilimumab (ccrcc1 and ccrcc4 groups), or either a VEGFR-TKI or nivolumab-ipilimumab (ccrcc2 and ccrcc3 groups). Patients assigned to nivolumab-ipilimumab received intravenous nivolumab 3 mg/kg plus ipilimumab 1 mg/kg every 3 weeks for four doses followed by intravenous nivolumab 240 mg every 2 weeks. Patients assigned to nivolumab received intravenous nivolumab 240 mg every 2 weeks. Patients assigned to VEGFR-TKIs received oral sunitinib (50 mg/day for 4 weeks every 6 weeks) or oral pazopanib (800 mg daily continuously). The primary endpoint was the objective response rate by investigator assessment per Response Evaluation Criteria in Solid Tumors version 1.1. The primary endpoint and safety were assessed in the population who received at least one dose of study drug. This trial is registered with ClinicalTrials.gov, NCT02960906, and with the EU Clinical Trials Register, EudraCT 2016-003099-28, and is closed to enrolment. FINDINGS: Between June 28, 2017, and July 18, 2019, 303 patients were screened for eligibility, 202 of whom were randomly assigned to treatment (61 to nivolumab, 101 to nivolumab-ipilimumab, 40 to a VEGFR-TKI). In the nivolumab group, two patients were excluded due to a serious adverse event before the first study dose and one patient was excluded from analyses due to incorrect diagnosis. Median follow-up was 18·0 months (IQR 17·6-18·4). In the ccrcc1 group, objective responses were seen in 12 (29%; 95% CI 16-45) of 42 patients with nivolumab and 16 (39%; 24-55) of 41 patients with nivolumab-ipilimumab (odds ratio [OR] 0·63 [95% CI 0·25-1·56]). In the ccrcc4 group, objective responses were seen in seven (44%; 95% CI 20-70) of 16 patients with nivolumab and nine (50% 26-74) of 18 patients with nivolumab-ipilimumab (OR 0·78 [95% CI 0·20-3·01]). In the ccrcc2 group, objective responses were seen in 18 (50%; 95% CI 33-67) of 36 patients with a VEGFR-TKI and 19 (51%; 34-68) of 37 patients with nivolumab-ipilimumab (OR 0·95 [95% CI 0·38-2·37]). In the ccrcc3 group, no objective responses were seen in the four patients who received a VEGFR-TKI, and in one (20%; 95% CI 1-72) of five patients who received nivolumab-ipilimumab. The most common treatment-related grade 3-4 adverse events were hepatic failure and lipase increase (two [3%] of 58 for both) with nivolumab, lipase increase and hepatobiliary disorders (six [6%] of 101 for both) with nivolumab-ipilimumab, and hypertension (six [15%] of 40) with a VEGFR-TKI. Serious treatment-related adverse events occurred in two (3%) patients in the nivolumab group, 38 (38%) in the nivolumab-ipilimumab group, and ten (25%) patients in the VEGFR-TKI group. Three deaths were treatment-related: one due to fulminant hepatitis with nivolumab-ipilimumab, one death from heart failure with sunitinib, and one due to thrombotic microangiopathy with sunitinib. INTERPRETATION: We demonstrate the feasibility and positive effect of a prospective patient selection based on tumour molecular phenotype to choose the most efficacious treatment between nivolumab with or without ipilimumab and a VEGFR-TKI in the first-line treatment of metastatic clear-cell renal cell carcinoma. FUNDING: Bristol Myers Squibb, ARTIC.

CHAIMELEON Project: Creation of a Pan-European Repository of Health Imaging Data for the Development of AI-Powered Cancer Management Tools.

The CHAIMELEON project aims to set up a pan-European repository of health imaging data, tools and methodologies, with the ambition to set a standard and provide resources for future AI experimentation for cancer management. The project is a 4 year long, EU-funded project tackling some of the most ambitious research in the fields of biomedical imaging, artificial intelligence and cancer treatment, addressing the four types of cancer that currently have the highest prevalence worldwide: lung, breast, prostate and colorectal. To allow this, clinical partners and external collaborators will populate the repository with multimodality (MR, CT, PET/CT) imaging and related clinical data. Subsequently, AI developers will enable a multimodal analytical data engine facilitating the interpretation, extraction and exploitation of the information stored at the repository. The development and implementation of AI-powered pipelines will enable advancement towards automating data deidentification, curation, annotation, integrity securing and image harmonization. By the end of the project, the usability and performance of the repository as a tool fostering AI experimentation will be technically validated, including a validation subphase by world-class European AI developers, participating in Open Challenges to the AI Community. Upon successful validation of the repository, a set of selected AI tools will undergo early in-silico validation in observational clinical studies coordinated by leading experts in the partner hospitals. Tool performance will be assessed, including external independent validation on hallmark clinical decisions in response to some of the currently most important clinical end points in cancer. The project brings together a consortium of 18 European partners including hospitals, universities, R&D centers and private research companies, constituting an ecosystem of infrastructures, biobanks, AI/in-silico experimentation and cloud computing technologies in oncology.

Validation of a deep learning segmentation algorithm to quantify the skeletal muscle index and sarcopenia in metastatic renal carcinoma.

OBJECTIVES: To validate a deep learning (DL) algorithm for measurement of skeletal muscular index (SMI) and prediction of overall survival in oncology populations. METHODS: A retrospective single-center observational study included patients with metastatic renal cell carcinoma between 2007 and 2019. A set of 37 patients was used for technical validation of the algorithm, comparing manual vs DL-based evaluations. Segmentations were compared using mean Dice similarity coefficient (DSC), SMI using concordance correlation coefficient (CCC) and Bland-Altman plots. Overall survivals (OS) were compared using log-rank (Kaplan-Meier) and Mann-Whitney tests. Generalizability of the prognostic value was tested in an independent validation population (N = 87). RESULTS: Differences between two manual segmentations (DSC = 0.91, CCC = 0.98 for areas) or manual vs. automated segmentation (DSC = 0.90, CCC = 0.98 for areas, CCC = 0.97 for SMI) had the same order of magnitude. Bland-Altman plots showed a mean difference of -3.33 cm2 [95%CI: -15.98, 9.1] between two manual segmentations, and -3.28 cm2 [95% CI: -14.77, 8.21] for manual vs. automated segmentations. With each method, 20/37 (56%) patients were classified as sarcopenic. Sarcopenic vs. non-sarcopenic groups had statistically different survival curves with median OS of 6.0 vs. 12.5 (p = 0.008) and 6.0 vs. 13.9 (p = 0.014) months respectively for manual and DL methods. In the independent validation population, sarcopenic patients according to DL had a lower OS (10.7 vs. 17.3 months, p = 0.033). CONCLUSION: A DL algorithm allowed accurate estimation of SMI compared to manual reference standard. The DL-calculated SMI demonstrated a prognostic value in terms of OS. KEY POINTS: • A deep learning algorithm allows accurate estimation of skeletal muscle index compared to a manual reference standard with a concordance correlation coefficient of 0.97. • Sarcopenic patients according to SMI thresholds after segmentation by the deep learning algorithm had statistically significantly lower overall survival compared to non-sarcopenic patients.

Adjusting D-dimer to Lung Disease Extent to Exclude Pulmonary Embolism in COVID-19 Patients (Co-LEAD).

OBJECTIVE: D-dimer measurement is a safe tool to exclude pulmonary embolism (PE), but its specificity decreases in coronavirus disease 2019 (COVID-19) patients. Our aim was to derive a new algorithm with a specific D-dimer threshold for COVID-19 patients. METHODS: We conducted a French multicenter, retrospective cohort study among 774 COVID-19 patients with suspected PE. D-dimer threshold adjusted to extent of lung damage found on computed tomography (CT) was derived in a patient set (n = 337), and its safety assessed in an independent validation set (n = 337). RESULTS: According to receiver operating characteristic curves, in the derivation set, D-dimer safely excluded PE, with one false negative, when using a 900 ng/mL threshold when lung damage extent was <50% and 1,700 ng/mL when lung damage extent was ≥50%. In the derivation set, the algorithm sensitivity was 98.2% (95% confidence interval [CI]: 94.7-100.0) and its specificity 28.4% (95% CI: 24.1-32.3). The negative likelihood ratio (NLR) was 0.06 (95% CI: 0.01-0.44) and the area under the curve (AUC) was 0.63 (95% CI: 0.60-0.67). In the validation set, sensitivity and specificity were 96.7% (95% CI: 88.7-99.6) and 39.2% (95% CI: 32.2-46.1), respectively. The NLR was 0.08 (95% CI; 0.02-0.33), and the AUC did not differ from that of the derivation set (0.68, 95% CI: 0.64-0.72, p = 0.097). Using the Co-LEAD algorithm, 76 among 250 (30.4%) COVID-19 patients with suspected PE could have been managed without CT pulmonary angiography (CTPA) and 88 patients would have required two CTs. CONCLUSION: The Co-LEAD algorithm could safely exclude PE, and could reduce the use of CTPA in COVID-19 patients. Further prospective studies need to validate this strategy.