Radiomics-based decision support tool assists radiologists in small lung nodule classification and improves lung cancer early diagnosis

BACKGROUND

Methods to improve stratification of small (≤15 mm) lung nodules are needed. We aimed to develop a radiomics model to assist lung cancer diagnosis.

METHODS

Patients were retrospectively identified using health records from January 2007 to December 2018. The external test set was obtained from the national LIBRA study and a prospective Lung Cancer Screening programme. Radiomics features were extracted from multi-region CT segmentations using TexLab2.0. LASSO regression generated the 5-feature small nodule radiomics-predictive-vector (SN-RPV). K-means clustering was used to split patients into risk groups according to SN-RPV. Model performance was compared to 6 thoracic radiologists. SN-RPV and radiologist risk groups were combined to generate "Safety-Net" and "Early Diagnosis" decision-support tools.

RESULTS

In total, 810 patients with 990 nodules were included. The AUC for malignancy prediction was 0.85 (95% CI: 0.82-0.87), 0.78 (95% CI: 0.70-0.85) and 0.78 (95% CI: 0.59-0.92) for the training, test and external test datasets, respectively. The test set accuracy was 73% (95% CI: 65-81%) and resulted in 66.67% improvements in potentially missed [8/12] or delayed [6/9] cancers, compared to the radiologist with performance closest to the mean of six readers.

CONCLUSIONS

SN-RPV may provide net-benefit in terms of earlier cancer diagnosis.

Metabolic resource overlap impacts competition among phyllosphere bacteria

The phyllosphere is densely colonised by microbial communities, despite sparse and heterogeneously distributed resources. The limitation of resources is expected to drive bacterial competition resulting in exclusion or coexistence based on fitness differences and resource overlap between individual colonisers. We studied the impact of resource competition by determining the effects of different bacterial colonisers on the growth of the model epiphyte Pantoea eucalypti 299R (Pe299R). Resource overlap was predicted based on genome-scale metabolic modelling. By combining results of metabolic modelling and pairwise competitions in the Arabidopsis thaliana phyllosphere and in vitro, we found that ten resources sufficed to explain fitness of Pe299R. An effect of both resource overlap and phylogenetic relationships was found on competition outcomes in vitro as well as in the phyllosphere. However, effects of resource competition were much weaker in the phyllosphere when compared to in vitro experiments. When investigating growth dynamics and reproductive success at the single-cell resolution, resource overlap and phylogenetic relationships are only weakly correlated with epiphytic Pe299R reproductive success, indicating that the leaf's spatial heterogeneity mitigates resource competition. Although the correlation is weak, the presence of competitors led to the development of Pe299R subpopulations that experienced different life histories and cell divisions. In some in planta competitions, Pe299R benefitted from the presence of epiphytes despite high resource overlap to the competitor strain suggesting other factors having stronger effects than resource competition. This study provides fundamental insights into how bacterial communities are shaped in heterogeneous environments and a framework to predict competition outcomes.

Pazopanib in advanced soft tissue sarcomas

Pazopanib is the first and only tyrosine kinase inhibitor currently approved for the treatment of multiple histological subtypes of soft tissue sarcoma (STS). Initially developed as a small molecule inhibitor of vascular endothelial growth factor receptors, preclinical work indicates that pazopanib exerts an anticancer effect through the inhibition of both angiogenic and oncogenic signaling pathways. Following the establishment of optimal dosing and safety profiles in early phase studies and approval for the treatment of advanced renal cell carcinoma, pazopanib was investigated in STS. A landmark phase III randomized study demonstrated improved progression-free survival with pazopanib compared to that with placebo in pretreated patients with STS of various subtypes. The efficacy of pazopanib in specific STS subtypes has been further described in real-world-based case series in both mixed and subtype-specific STS cohorts. At present, there are no clinically validated predictive biomarkers for use in selecting patients with advanced STS for pazopanib therapy, limiting the clinical effectiveness and cost-effectiveness of the drug. In this review, we summarize the preclinical and clinical data for pazopanib, outline the evidence base for its effect in STS and explore reported studies that have investigated putative biomarkers.