Association between Contrast-Enhanced Computed Tomography Radiomic Features, Genomic Alterations and Prognosis in Advanced Lung Adenocarcinoma Patients.

Non-invasive methods to assess mutational status, as well as novel prognostic biomarkers, are warranted to foster therapy personalization of patients with advanced non-small cell lung cancer (NSCLC). This study investigated the association of contrast-enhanced Computed Tomography (CT) radiomic features of lung adenocarcinoma lesions, alone or integrated with clinical parameters, with tumor mutational status (EGFR, KRAS, ALK alterations) and Overall Survival (OS). In total, 261 retrospective and 48 prospective patients were enrolled. A Radiomic Score (RS) was created with LASSO-Logistic regression models to predict mutational status. Radiomic, clinical and clinical-radiomic models were trained on retrospective data and tested (Area Under the Curve, AUC) on prospective data. OS prediction models were trained and tested on retrospective data with internal cross-validation (C-index). RS significantly predicted each alteration at training (radiomic and clinical-radiomic AUC 0.95-0.98); validation performance was good for EGFR (AUC 0.86), moderate for KRAS and ALK (AUC 0.61-0.65). RS was also associated with OS at univariate and multivariable analysis, in the latter with stage and type of treatment. The validation C-index was 0.63, 0.79, and 0.80 for clinical, radiomic, and clinical-radiomic models. The study supports the potential role of CT radiomics for non-invasive identification of gene alterations and prognosis prediction in patients with advanced lung adenocarcinoma, to be confirmed with independent studies.

Application of nnU-Net for Automatic Segmentation of Lung Lesions on CT Images and Its Implication for Radiomic Models.

Radiomics investigates the predictive role of quantitative parameters calculated from radiological images. In oncology, tumour segmentation constitutes a crucial step of the radiomic workflow. Manual segmentation is time-consuming and prone to inter-observer variability. In this study, a state-of-the-art deep-learning network for automatic segmentation (nnU-Net) was applied to computed tomography images of lung tumour patients, and its impact on the performance of survival radiomic models was assessed. In total, 899 patients were included, from two proprietary and one public datasets. Different network architectures (2D, 3D) were trained and tested on different combinations of the datasets. Automatic segmentations were compared to reference manual segmentations performed by physicians using the DICE similarity coefficient. Subsequently, the accuracy of radiomic models for survival classification based on either manual or automatic segmentations were compared, considering both hand-crafted and deep-learning features. The best agreement between automatic and manual contours (DICE = 0.78 ± 0.12) was achieved averaging 2D and 3D predictions and applying customised post-processing. The accuracy of the survival classifier (ranging between 0.65 and 0.78) was not statistically different when using manual versus automatic contours, both with hand-crafted and deep features. These results support the promising role nnU-Net can play in automatic segmentation, accelerating the radiomic workflow without impairing the models' accuracy. Further investigations on different clinical endpoints and populations are encouraged to confirm and generalise these findings.

Integrating Biological and Radiological Data in a Structured Repository: a Data Model Applied to the COSMOS Case Study.

Integrating the information coming from biological samples with digital data, such as medical images, has gained prominence with the advent of precision medicine. Research in this field faces an ever-increasing amount of data to manage and, as a consequence, the need to structure these data in a functional and standardized fashion to promote and facilitate cooperation among institutions. Inspired by the Minimum Information About BIobank data Sharing (MIABIS), we propose an extended data model which aims to standardize data collections where both biological and digital samples are involved. In the proposed model, strong emphasis is given to the cause-effect relationships among factors as these are frequently encountered in clinical workflows. To test the data model in a realistic context, we consider the Continuous Observation of SMOking Subjects (COSMOS) dataset as case study, consisting of 10 consecutive years of lung cancer screening and follow-up on more than 5000 subjects. The structure of the COSMOS database, implemented to facilitate the process of data retrieval, is therefore presented along with a description of data that we hope to share in a public repository for lung cancer screening research.

A segmentation tool for pulmonary nodules in lung cancer screening: Testing and clinical usage.

PURPOSE: With the future goal of defining a large dataset based on low-dose CT with labelled pulmonary lesions for lung cancer screening (LCS) research, the aim of this work is to propose and evaluate into a clinical context a tool for semi-automatic segmentation able to facilitate the process of labels collection from a LCS study (COSMOS, Continuous Observation of SMOking Subjects). METHODS: Considering a preliminary set of manual annotations, a segmentation model based on a 2D-Unet was trained from scratch. Contour quality of the final 2D-Unet was assessed on an internal test set of manual annotations and on a subset of the public available LIDC dataset used as external test set. The tool for semi-automatic segmentation was then designed integrating the tested model into a Graphical User Interface. According to the opinion of two clinical users, the percentage of lesions properly contoured through the tool was quantified (Acceptance Rate, AR). The variability between segmentations derived by the two readers was estimated as mean percentage of difference (MPD) between the two sets of volumes and comparing the likelihood of malignancy derived from Volume Doubling Time (VDT). RESULTS: Performance in test sets were found similar (DICE ~ 0.75(0.15)). Accordingly, a good mean AR (80.1%) resulted from the two readers. Variability in terms of MPD was equal to 23.6% while 2.7% was the VDTs percentage of disagreement. CONCLUSIONS: A semi-automatic segmentation tool was developed and its applicability evaluated into a clinical context demonstrating the efficacy of the tool in facilitating the collection of labelled data.

Pediatric Rhabdomyosarcomas: Three-Dimensional Radiological Assessments after Induction Chemotherapy Predict Survival Better than One-Dimensional and Two-Dimensional Measurements.

Radiological response to neoadjuvant chemotherapy is currently used to assess the efficacy of treatment in pediatric patients with rhabdomyosarcoma (RMS), but the association between early tumor response on imaging and survival is still controversial. The aim of this study was to investigate the prognostic value of assessing radiological response after induction therapy in pediatric RMS, comparing four different methods. This retrospective, two-center study was conducted on 66 non-metastatic RMS patients. Two radiologists measured tumor size on pre- and post-treatment magnetic resonance (MR) or computed tomography (CT) images using four methods: considering maximal diameter with the 1D-RECIST (Response Evaluation Criteria in Solid Tumors); multiplying the two maximal diameters with the 2D-WHO (World Health Organization); multiplying the three maximal diameters with the 3D-EpSSG (European pediatric Soft tissue sarcoma Study Group); obtaining a software-assisted volume assessment with the 3D-Osirix. Each patient was classified as a responder or non-responder based on the proposed thresholds for each method. Tumor response was compared with survival using Kaplan-Meier plots, the log-rank test, and Cox's regression. Agreement between methods and observers (weighted-κ) was also calculated. The 5-year event-free survival (5yr-EFS) calculated with the Kaplan-Meier plots was significantly longer for responders than for non-responders with all the methods, but the 3D assessments differentiated between the two groups better than the 1D-RECIST or 2D-WHO (p1D-RECIST = 0.018, p2D-WHO = 0.007, p3D-EpSSG and p3D-Osirix < 0.0001). Comparing the 5yr-EFS of responders and non-responders also produced adjusted hazard ratios of 3.57 (p = 0.0158) for the 1D-RECIST, 5.05 for the 2D-WHO (p = 0.0042), 14.40 for the 3D-EpSSG (p < 0.0001) and 11.60 for the 3D-Osirix (p < 0.0001), indicating that the volumetric measurements were significantly more strongly associated with EFS. Inter-method agreement was excellent between the 3D-EpSSG and the 3D-Osirix (κ = 0.98), and moderate for the other comparisons (0.5 < κ < 0.8). The 1D-RECIST and the 2D-WHO tended to underestimate response to treatment. Inter-observer agreement was excellent with all methods (κ > 0.8) except for the 2D-WHO (κ = 0.7). In conclusion, early tumor response was confirmed as a significant prognostic factor in RMS, and the 3D-EpSSG and 3D-Osirix methods predicted response to treatment better than the 1D-RECIST or 2D-WHO measurements.

Association between baseline tumour burden and outcome in patients with cancer treated with next-generation immunoncology agents.

BACKGROUND: Baseline tumour burden is a prognostic factor for patients with melanoma and non-small-cell lung cancer treated with immunotherapy. However, no data are available on its role in other solid tumours, nor for treatment with next-generation immunoncology agents (NGIOs). METHODS: We reviewed data of patients with any solid tumour consecutively treated at our institution from August 2014 to March 2019, who received ≥1 dose of immune checkpoint inhibitor and/or NGIO within phase 1 trials. Baseline tumour burden was calculated as ∑i Response Evaluation Criteria in Solid Tumours 1.1 baseline target lesions (baseline tumour size [BTS]) or as sum of all measurable baseline lesions (total tumour burden [TTB]); the impact of both parameters on treatment outcomes was investigated. RESULTS: One hundred fifty patients were included in the analysis. Median BTS and TTB were 79 mm and 212 mm, respectively. Objective response rate was found significantly associated with BTS (p < 0.001) and TTB quartiles (p = 0.006), with response rates progressively increasing with decreasing tumour burden quartiles. Both progression-free survival (PFS) (p = 0.001) and overall survival (OS) (p < 0.001) were significantly associated with BTS quartiles, with 26% of the patients progression-free and 56% alive at 12 months in the lower BTS quartile, compared with 3% and 24%, respectively, in the upper quartile. TTB was also significantly associated with OS (P = 0.01) and borderline-significant for PFS (p = 0.07). Multivariate analysis confirmed that baseline burden, also considered as continuous variable, is independently associated with PFS and OS, when assessed with BTS (p = 0.001 and p < 0.001) and TTB (p = 0.007 and p < 0.001). CONCLUSIONS: Lower baseline tumour burden is associated with better outcomes in patients with cancer treated with novel immunotherapies.

Inter-observer agreement on the morphology of screening-detected lung cancer: beyond pulmonary nodules and masses.

OBJECTIVES: Pulmonary nodules and masses are the typical presentations of lung cancer. However, a spectrum of focal opacities cannot be defined as either "pulmonary nodule" or "mass," despite representing cancer. We aimed to assess the morphology of screening-detected lung cancers at low-dose computed tomography LDTC and to evaluate inter-observer agreement in their classification. METHODS: Four radiologists with different experiences in thoracic imaging retrospectively reviewed 273 screening-detected lung cancers. Readers were asked to assess if morphology at the time of diagnosis was consistent with the Fleischner Society definition of pulmonary "nodule" or "mass." Cancers not consistent were defined as "non-nodular/non-mass" (NN/NM) and sub-classified as follows: associated with cystic airspaces, stripe-like, scar-like, endobronchial, or not otherwise defined (NOD). Inter-observer agreement was evaluated using Cohen's K statistic among pairs of readers and modified Fleiss' kappa statistic for overall agreement. RESULTS: Two hundred forty-one of the 273 (88%) lesions were defined as pulmonary nodule or mass by complete agreement, while 20/273 (7.3%) were defined as NN/NM. Six (2.2%) of 273 were sub-classified as lesions associated with cystic airspace, six (2.2%) as scar-like, five (1.8%) as endobronchial, and one (0.7%) as NOD by complete agreement. The concordance in defining morphology was excellent (261/273; 96%, 95%CI 92-98%; k 0.85, 95%CI 0.75-0.92) and also in the sub-classification (18/20; 90%, 95%CI 68-99%, k 0.93, 95%CI 0.86-1.00). There was incomplete agreement regarding lesion morphology in 4.4% (12/273) of cases. CONCLUSIONS: A non-negligible percentage of screening-detected lung cancers has a NN/NM appearance at LDCT. The concordance in defining lesion morphology was excellent. The awareness of various presentations can avoid missed or delayed diagnosis. KEY POINTS: • A non-negligible percentage of screening-detected lung cancers have neither nodular nor mass appearance at low-dose CT. • The awareness of various LDCT presentations of lung cancer can avoid missed or delayed diagnosis. • Optimal protocol management in CT screening should take into consideration lung nodules as well as various other focal abnormalities.

Signal intensity change on unenhanced T1-weighted images in dentate nucleus and globus pallidus after multiple administrations of gadoxetate disodium: an intraindividual comparative study.

PURPOSE: To investigate whether there is an increased signal intensity (SI) of dentate nucleus (DN) and globus pallidus (GP) on unenhanced T1-weighted magnetic resonance imaging (MRI), in patients who had undergone multiple administrations of gadoxetate disodium. MATERIALS AND METHODS: We retrospectevely included stage III melanoma patients, who had been previously enrolled in a trial of adjuvant therapy and who had undergone whole-body contrast-enhanced MRIs with gadoxetate disodium every three months for their follow-up. The SI ratios of DN-to-pons and GP-to-thalamus on unenhanced T1-weighted images were calculated. The difference in SI ratios between the first and the last MRI examinations was assessed and a linear mixed model was performed to detect how SI ratios varied with the number of administrations. RESULTS: Eighteen patients were included in our study. The number of gadoxetate disodium administrations ranged from 2 to 18. Paired t-test did not show any significant difference in DN-to-pons (p=0.21) and GP-to-thalamus (p=0.09) SI ratios by the end of the study. DN-to-pons SI ratio and GP-to-thalamus SI ratio did not significantly increase with increasing the number of administrations (p=0.14 and p=0.06, respectively). CONCLUSION: Multiple administrations of gadoxetate disodium are not associated with increased SI in DN and GP in the brain. KEY POINTS: • Gadolinium may deposit in the human brain after multiple GBCA administrations. • Gadolinium deposition is associated with increased T1W signal intensity • Increase in signal intensity is most apparent within the DN and GP • Multiple administrations of gadoxetate disodium do not increase T1W signal.

Spectrum of early lung cancer presentation in low-dose screening CT: a pictorial review.

The typical presentation of early stage lung cancers on low-dose CT screening are non-calcified pulmonary nodules. However, there is a wide spectrum of unusual focal abnormalities that can be early presentations of lung cancer. These abnormalities include, for example, cancers associated with 'cystic airspaces' or scar-like cancers. The detection of lung cancer with low-dose CT can be affected by the absence of intravenous contrast medium. As a consequence, endobronchial and central lesions can be difficult to recognize, raising the potential for missed cancers. Focal lesions arising within pre-existing lung disease, such as lung fibrosis or apical scars, can also be early lung cancer manifestations and deserve particular consideration as recognition of these lesions may be hindered by the underlying disease. Furthermore, the unpredictable growth rate of lung cancer, which ranges from indolent to aggressive cancers, necessitates attention to the wide spectrum of progression in lung cancer appearance on serial low-dose CT scans. In this pictorial review we discuss the spectrum of early lung cancer presentation in low-dose CT screening, highlighting typical as well as unusual radiological features and the varied growth rates of early lung cancer. Teaching Points • There is a wide spectrum of early presentations of lung cancer on LDCT. • Low radiation dose and the absence of contrast medium injection can affect lung cancer detection. • Lung cancer growth shows various behaviours, ranging from indolent to aggressive cancers. • Familiarity with LDCT technique can improve CT screening effectiveness and avoid missed diagnosis.