No abstract

Background Poorly differentiated thyroid cancer (PDTC) is a rare, follicular cell-derived neoplasm with unfavorable prognosis. The oncocytic variant of PDTC may be associated with even more adverse outcome than classical PDTC cases, but its specific molecular features are largely unknown. Our aim was to explore the immune-related gene expression profile of oncocytic and classical PDTC, in correlation with clinical and pathological characteristics (including PD-L1 expression) and outcome, and in comparison with a control group of well differentiated follicular carcinomas (WDFC), including conventional follicular carcinomas (FTC) and Hürthle cell carcinomas (HCC). Methods A retrospective series of 48 PDTC and 24 WDFC was analyzed by means of NanoString technology employing nCounter PanCancer Immune Profiling panel. Gene expression data were validated using quantitative real time PCR. Results Oncocytic PDTC showed a specific immune-related gene expression profile, with higher expression of LAIR2, CD274, DEFB1, IRAK1, CAMP, LCN2, LY96, and APOE, and lower expression of NOD1, as compared to conventional PDTCs. This molecular signature was associated with increased intra-tumoral lymphocytic infiltration, PD-L1 expression and adverse outcome. Three of these genes, CD274, DEFB1, IRAK1, as well as PD-L1 expression, were also the hallmarks of HCCs as compared to FTCs. By contrast, the panel of genes differentially regulated in PDTCs as compared to WDFCs was unrelated to the oncocytic phenotype. Conclusions Our results revealed a distinctive immune-related gene expression profile of oncocytic PDTC and confirmed a more aggressive outcome in this cancer subtype. These findings may provide guidance when exploring novel immunotherapeutic options for oncocytic PDTC patients.

No abstract

experienced certified senior radiologist. Lesions <1cm, inflammatory and indeterminate lesions were excluded from delineation. A total of 47 radiomic features including shape, first-order and texture features were extracted from the segmented tumour using PyRadiomics. No preprocessing of the images was performed. Highly correlated features (r>0.85) were removed from further analysis. Least Absolute Shrinkage and Selection Operator (LASSO) feature selection was performed to find informative features that could predict either best overall response or overall survival. Univariate logistic regression and cox proportional hazard models were then used for an initial assessment of the potential of these features in predicting response and survival respectively. Result: Sixteen patients with evaluable best overall response (partial response n¼9, progressive disease n¼7) were selected for the initial discovery-cohort. Mean age was 68 years with 63% adenocarcinoma histology. From the 47 features extracted, 32 were highly correlated to each other and were removed from further analysis. For predicting best overall response, LASSO selected 5 features with univariate logistic regression suggesting that tumour surface area to volume ratio might be informative (p¼0.057, AUC of 0.83 (95% CI 0.61-1.0)). With respect to overall survival, LASSO selected 3 features with univariate cox regression suggesting the first-order feature skewness might be predictive (HR ¼ 0.27, 95% CI 0.08-0.88, p¼0.03). When split on the median skewness value the Kaplan-Meier plot showed a significant survival difference between high and low risk patients (p¼0.007). Conclusion: Radiomic features extracted from baseline contrast-enhanced CT scans may have the potential to predict response and survival in patients treated with first-line pembrolizumab in advanced NSCLC. We emphasize the exploratory nature of these results given the very limited number of patients in the study. We are expanding this discovery cohort to further investigate and validate these results. Updated results will be presented at the meeting.