Combined Radiomic and Visual Assessment for Improved Detection of Lung Adenocarcinoma Invasiveness on Computed Tomography Scans: A Multi-Institutional Study

Moving from conventional to adaptive risk stratification for oropharyngeal cancer

Oropharyngeal cancer (OPC) poses a complex therapeutic dilemma for patients and oncologists alike, made worse by the epidemic increase in new cases associated with the oncogenic human papillomavirus (HPV). In a counterintuitive manner, the very thing which gives patients hope, the high response rate of HPV-associated OPC to conventional chemo-radiation strategies, has become one of the biggest challenges for the field as a whole. It has now become clear that for ~30-40% of patients, treatment intensity could be reduced without losing therapeutic efficacy, yet substantially diminishing the acute and lifelong morbidity resulting from conventional chemotherapy and radiation. At the same time, conventional approaches to de-escalation at a population (selected or unselected) level are hampered by a simple fact: we lack patient-specific information from individual tumors that can predict responsiveness. This results in a problematic tradeoff between the deleterious impact of de-escalation on patients with aggressive, treatment-refractory disease and the beneficial reduction in treatment-related morbidity for patients with treatment-responsive disease. True precision oncology approaches require a constant, iterative interrogation of solid tumors prior to and especially during cancer treatment in order to tailor treatment intensity to tumor biology. Whereas this approach can be deployed in hematologic diseases with some success, our ability to extend it to solid cancers with regional metastasis has been extremely limited in the curative intent setting. New developments in metabolic imaging and quantitative interrogation of circulating DNA, tumor exosomes and whole circulating tumor cells, however, provide renewed opportunities to adapt and individualize even conventional chemo-radiation strategies to diseases with highly variable biology such as OPC. In this review, we discuss opportunities to deploy developing technologies in the context of institutional and cooperative group clinical trials over the coming decade.

The value of intratumoral and peritumoral radiomics features in differentiating early-stage lung invasive adenocarcinoma (≤3 cm) subtypes

Background

The identification of different subtypes of early-stage lung invasive adenocarcinoma before surgery contributes to the precision treatment. Radiomics could be one of the effective and noninvasive identification methods. The value of peritumoral radiomics in predicting the subtypes of early-stage lung invasive adenocarcinoma perhaps clinically useful.

Methods

This retrospective study included 937 lung adenocarcinomas which were randomly divided into the training set (n=655) and testing set (n=282) with a ratio of 7:3. This study used the univariate and multivariate analysis to choose independent clinical predictors. Radiomics features were extracted from 18 regions of interest (1 intratumoral region and 17 peritumoral regions). Independent and conjoint prediction models were constructed based on radiomics and clinical features. The performance of the models was evaluated using receiver operating characteristic (ROC) curves, accuracy (ACC), sensitivity (SEN), and specificity (SPE). Significant differences between areas under the ROC (AUCs) were estimated using in the Delong test.

Results

Patient age, smoking history, carcinoembryonic antigen (CEA), lesion location, length, width and clinic behavior were the independent predictors of differentiating early-stage lung invasive adenocarcinoma (≤3 cm) subtypes. The highest AUC value among the 19 independent models was obtained for the PTV0~+3 radiomics model with 0.849 for the training set and 0.854 for the testing set. As the peritumoral distance increased, the predictive power of the models decreased. The radiomics-clinical conjoint model was statistically significantly different from the other models in the Delong test (P<0.05).

Conclusions

The intratumoral and peritumoral regions contained a wealth of clinical information. The diagnostic efficacy of intra-peritumoral radiomics combined clinical model was further improved, which was particularly important for preoperative staging and treatment decision-making.