Laurie L. Carr

PURPOSE: We conducted a phase II study to assess the efficacy of continuous dosing of sunitinib in patients with flurodeoxyglucose positron emission tomography (FDG-PET)-avid, iodine-refractory well-differentiated thyroid carcinoma (WDTC) and medullary thyroid cancer (MTC) and to assess for early response per FDG-PET. EXPERIMENTAL DESIGN: Patients had metastatic, iodine-refractory WDTC or MTC with FDG-PET-avid disease. Sunitinib was administered at 37.5 mg daily on a continuous basis. The primary end point was response rate per Response Evaluation Criteria in Solid Tumors (RECIST). Secondary end points included toxicity, overall survival, and time to progression. We conducted an exploratory analysis of FDG-PET response after 7 days of treatment. RESULTS: Thirty-five patients were enrolled (7 MTC, 28 WDTC), and 33 patients were evaluable for disease response. The primary end point, objective response rate per RECIST, was 11 patients (31%; 95% confidence interval, 16-47%). There were 1 complete response (3%), 10 partial responses (28%), and 16 patients (46%) with stable disease. Progressive disease was seen in 6 patients (17%). The median time to progression was 12.8 months (95% confidence interval, 8.9 months-not reached). Repeat FDG-PET was done on 22 patients. The median percent change in average standardized uptake values was -11.7%, -13.9%, and 8.6% for patients with RECIST response, stable disease, and progressive disease, respectively. Differences between response categories were statistically significant (P = 0.03). The most common toxicities seen included fatigue (11%), neutropenia (34%), hand/foot syndrome (17%), diarrhea (17%), and leukopenia (31%). One patient on anticoagulation died of gastrointestinal bleeding. CONCLUSION: Continuous administration of sunitinib was effective in patients with iodine-refractory WDTC and MTC. Further study is warranted.

Stimulation of mature peripheral T cells by TCR engagement results in activation of signals that drive induction of cytokine gene expression and clonal expansion. However, under some conditions, engagement of the TCR leads instead to apoptosis. Recent studies demonstrate that TCR-stimulated apoptosis requires expression of CD95 ligand on activated T cells followed by an interaction between CD95 ligand and the CD95 receptor also expressed on this population. The experiments reported in this study were designed to address the signaling events triggered by TCR engagement that are important for regulating CD95 ligand gene expression. To approach this, we generated a luciferase reporter construct containing elements of the CD95 ligand promoter. Using a previously described mutant of the Jurkat T cell line, we show that proximal signaling events dependent on the presence of the CD45 tyrosine phosphatase are required for TCR-stimulated CD95 ligand expression. Transient transfection studies demonstrate further that TCR-stimulated activation of the Ras signaling pathway is required for optimal activation of CD95 ligand. Next, in an effort to determine critical transcription factors that regulate CD95 ligand expression, we demonstrate a cyclosporin A-sensitive nuclear factor-AT response element in the promoter region of this gene that is critical for optimal CD95 ligand reporter activity in stimulated T cells. Together, these studies begin a dissection of the biochemical events that lead to expression of CD95 ligand, a required step for TCR-induced apoptosis.

PURPOSE: Computed tomography (CT) is an effective method for detecting and characterizing lung nodules in vivo. With the growing use of chest CT, the detection frequency of lung nodules is increasing. Noninvasive methods to distinguish malignant from benign nodules have the potential to decrease the clinical burden, risk, and cost involved in follow-up procedures on the large number of false-positive lesions detected. This study examined the benefit of including perinodular parenchymal features in machine learning (ML) tools for pulmonary nodule assessment. METHODS: Lung nodule cases with pathology confirmed diagnosis (74 malignant, 289 benign) were used to extract quantitative imaging characteristics from computed tomography scans of the nodule and perinodular parenchyma tissue. A ML tool development pipeline was employed using k-medoids clustering and information theory to determine efficient predictor sets for different amounts of parenchyma inclusion and build an artificial neural network classifier. The resulting ML tool was validated using an independent cohort (50 malignant, 50 benign). RESULTS: The inclusion of parenchymal imaging features improved the performance of the ML tool over exclusively nodular features (P < 0.01). The best performing ML tool included features derived from nodule diameter-based surrounding parenchyma tissue quartile bands. We demonstrate similar high-performance values on the independent validation cohort (AUC-ROC = 0.965). A comparison using the independent validation cohort with the Fleischner pulmonary nodule follow-up guidelines demonstrated a theoretical reduction in recommended follow-up imaging and procedures. CONCLUSIONS: Radiomic features extracted from the parenchyma surrounding lung nodules contain valid signals with spatial relevance for the task of lung cancer risk classification. Through standardization of feature extraction regions from the parenchyma, ML tool validation performance of 100% sensitivity and 96% specificity was achieved.