This phase II trial evaluates the diagnostic performance of contrast-enhanced ultrasound (CEUS) for assessing treatment response in patients undergoing transarterial chemoembolization (TACE) for liver tumors. TACE is a hepatic artery embolization technique involving the injection of a blocking agent and a chemotherapy agent to treat liver cancers. Currently, contrast enhanced magnetic resonance imaging or computed tomography are used to assess disease response 1-2 months after TACE treatment, but ultrasound may be a less expensive, earlier alternative. CEUS is an imaging procedure that uses high-frequency sound waves to generate images of the body after administering Lumason, an imaging agent used to enhance visualization of blood flow on ultrasounds. CEUS is able to be performed during the TACE procedure, making it possible to evaluate treatment response earlier than standard techniques. CEUS may be an effective method to evaluate treatment response more accurately and much earlier than current standard evaluation methods.
PRIMARY OBJECTIVE: I. To evaluate the sensitivity and specificity of CEUS for the evaluation of TACE treatment response in a variety of solid liver tumors (years 1-4). SECONDARY OBJECTIVES: I. To determine the ability of CEUS to identify residual tumor vascularity intraoperatively, thereby enabling immediate retreatment when necessary (years 1-4). II. To explore a variety of advanced imaging approaches to improve on the suboptimal specificity of CEUS for identifying residual viable tumor following TACE (years 1-5). III. To investigate the ability of CEUS obtained prior to TACE to quantitatively assess tumor vascular morphology and predict response to therapy (years 2-5). EXPLORATORY OBJECTIVE: I. Use acquired B-mode in-phase and quadrature (IQ) data for H-scan imaging. OUTLINE: Patients receive sulfur hexafluoride lipid microspheres (Lumason) intravenously (IV) and undergo CEUS 2 weeks prior to TACE, during TACE, 1-2 weeks after TACE, and then 1-2 months after TACE. After completion of study treatment, patients are followed up at 6 months.
Study Type
INTERVENTIONAL
Allocation
NA
Purpose
DIAGNOSTIC
Masking
NONE
Enrollment
266
Given IV
Undergo CEUS
Undergo TACE
Ancillary studies
Sidney Kimmel Cancer Center at Thomas Jefferson University
Philadelphia, Pennsylvania, United States
RECRUITINGRecurrence
Time frame: Up to 6 months
Sensitivity
Will be computed using a reference standard. Variables will be summarized with descriptive statistics, such as means with standard deviations or frequency counts with percentage, across the cohort and within group of interest. Diagnostic accuracy will be compared between hepatocellular carcinoma (HCC) and non-HCC and between all tumor subtypes.
Time frame: Up to 6 months
Specificity
Will be computed using a reference standard. Variables will be summarized with descriptive statistics, such as means with standard deviations or frequency counts with percentage, across the cohort and within group of interest. Diagnostic accuracy will be compared between HCC and non-HCC and between all tumor subtypes.
Time frame: Up to 6 months
Positive predictive value
Will be computed using a reference standard. Variables will be summarized with descriptive statistics, such as means with standard deviations or frequency counts with percentage, across the cohort and within group of interest. Diagnostic accuracy will be compared between HCC and non-HCC and between all tumor subtypes.
Time frame: Up to 6 months
Negative predictive value
Will be computed using a reference standard. Variables will be summarized with descriptive statistics, such as means with standard deviations or frequency counts with percentage, across the cohort and within group of interest. Diagnostic accuracy will be compared between HCC and non-HCC and between all tumor subtypes.
Time frame: Up to 6 months
False discovery rate
Will be computed using a reference standard. Variables will be summarized with descriptive statistics, such as means with standard deviations or frequency counts with percentage, across the cohort and within group of interest. Diagnostic accuracy will be compared between HCC and non-HCC and between all tumor subtypes.
Time frame: Up to 6 months
Residual tumor vacularity
Will use diagnostic summary statistics and generalized estimating equations (GEE) logistic regression modeling. The bedside (interventional radiologist) versus offline (radiologist) reads will be compared using agreement, kappa statistics, and mixed modeling.
Time frame: Up to 6 months
Diagnostic performance for each imaging mode
The diagnostic performance for each reader will be quantified from the volumetric contrast enhanced ultrasound exams and post-processed images. Diagnostic performance for each imaging mode will be compared across all readers using the GEE logistic regression approach. Quantitative H-scan data will be compared between complete and incomplete responders using multiple linear regression or GEE regression modeling, depending on how well assumptions hold for the former.
Time frame: Up to 6 months
Ability of the model to predict binary treatment response
Model performance will be calculated using a leave-one-out cross-validation method to assess the ability of the model to predict binary treatment response. Accuracy, sensitivity and specificity will then be quantified and directly compared between 2 dimensional (D) and 3D datasets.
Time frame: Up to 6 months
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.