Evaluation of Liver Cancer With Magnetic Resonance Imaging (MRI)

The incidence of hepatocellular carcinoma (HCC) has recently increased in the United States. Although imaging plays a major role in HCC screening and staging, the possibility of predicting HCC tumor grade, aggressiveness, angiogenesis and hypoxia with imaging are unmet needs. In addition, new antiangiogenic drugs now available to treat advanced HCC necessitate the use of new imaging criteria beyond size. The investigators would like to develop and validate non-invasive magnetic resonance imaging (MRI) methods based on advanced diffusion-weighted imaging (DWI), MR Elastography, BOLD (blood oxygen level dependent) MRI and perfusion-weighted imaging (PWI, using gadolinium contrast) to be used as non-invasive markers of major histopathologic features of HCC, and to predict and assess early response of HCC to systemic therapy. The investigators also would like to develop quality control tools to improve the quality and decrease variability of quantitative MRI metrics. These techniques combined could represent non-invasive correlates of histologic findings in HCC, could enable individualized therapy, and provide prognosis in patients with HCC.

The incidence of hepatocellular carcinoma (HCC) has recently increased in the US mostly due to an increase in chronic hepatitis C infection. Angiogenesis is critical for the growth and metastatic progression of HCC. With the development of new antiangiogenic drugs such as sorafenib, imaging methods to predict and assess therapeutic response beyond changes in size become critical. However, validated imaging methods to predict and assess early HCC response to targeted agents are lacking. In this study, the investigators would like to develop quantitative MRI methods interrogating different features of HCC tumor biology and pathology, including tumor cellularity, grade, angiogenesis and hypoxia. The investigators propose a multiparametric approach combining advanced DWI (IVIM: intravoxel incoherent motion diffusion measuring perfusion fraction and true diffusion coefficient), DCE-MRI (dynamic contrast-enhanced MRI, which measures arterial and portal flow, mean transit time, blood volume and distribution volume), and BOLD MRI using oxygen or carbogen challenge. This protocol will be performed in patients with HCC undergoing hepatic resection. Routine and advanced histopathologic methods will be performed (tumor grade, CK19 expression, presence of microvascular invasion, VEGF expression, microvessel density, HIF 1-alpha expression). MRI metrics will be correlated with histopathologic metrics. The first portion of the proposal involves the development of a QC algorithm assessing MR data quality and test-retest. The investigators will propose solutions to improve data acquisition and processing. The last 2 years of the study will be dedicated to a prospective randomized study comparing Yttrium 90 radioembolization to sorafenib, assessing the role of baseline MRI metrics and early changes (at 2 weeks) in these metrics as markers of tumor response and time to progression in patients with unresectable HCC.

Magnetic Resonance Imaging is a radiation free non invasive technique using magnetic radiofrequency waves to image the body. In this study, the research team would like to investigate the possibility of providing functional information on aggressiveness, vascularity and oxygen uptake in liver cancer tumors.

Tumor mean transit time (MTT) of contrast agent. Perfusion/flow measured with dynamic contrast-enhanced imaging using gadolinium contrast

Tumor distribution volume (DV) of contrast agent. Perfusion/flow measured with dynamic contrast-enhanced imaging using gadolinium contrast

Oxygen uptake measured with T2* and T1-weighted imaging. Oxygen uptake (% change pre and post O2 administration) calculated by Liver ΔR2*=100 x (R2* post O2-R2* pre O2)/R2* pre O2. The healthy participants breathed 100% medical O2 through a mask for 10 min., and were imaged before and after O2 administration with the MRI methods that are sensitive to oxygen uptake in tumors.

Tumor diffusion measured with diffusion-weighted imaging sequence. In diffusion weighted MR imaging (DWI), the signal is proportional to the Brownian motion diffusion of free water protons in tissues. Deposition of collagen in tissue (as in fibrotic disease), or cellularity in tumors act as impediments to free water diffusion. Using different mathematical models, the degree of diffusion can be quantified from the MRI signal, to provide information on diffusion restriction due to disease. From mono exponential fit of diffusion signal, one can obtain the apparent diffusion coefficient (ADC). However, this coefficient reflects free water proton diffusion, as well as transport of water protons in the capillary vessels (capillary perfusion).

Tumor diffusion measured with diffusion-weighted imaging sequence. To separate the diffusion effect from capillary perfusion, a bi-exponential model is used, which provides 3 coefficients: one is the true diffusion coefficient D, reflecting free water proton diffusion.

Tumor diffusion measured with diffusion-weighted imaging sequence. To separate the diffusion effect from capillary perfusion, a bi-exponential model is used, which provides 3 coefficients: one is the pseudo-diffusion coefficient D*, affected by free diffusion and capillary perfusion.

Tumor diffusion measured with diffusion-weighted imaging sequence. To separate the diffusion effect from capillary perfusion, a bi-exponential model is used, which provides 3 coefficients: one is the perfusion fraction PF, which reflects how much the diffusion-weighted signal is affected by capillary perfusion. PF is a measure of vascularity in the tissue.

Perfusion/flow measured with dynamic contrast-enhanced imaging using gadolinium contrast. Extravascular extracellular volume fraction ve (%) - represents the portion of tissue occupied by the extravascular extracellular volume (interstitial space), in which MRI contrast agent can distribute.

Tumor response to treatment is evaluated clinically by radiologists according to RECIST and modified RECIST criteria, by which the diameter of the tumor portion that enhances (lights up on imaging) after administration of gadolinium contrast agent is measured before and after treatment. The response is not reported as diameter or diameter difference in mm, but rather as a qualitative variable: complete response, partial response, stable disease and progressive disease. Complete response means no enhancing tumor regions after treatment (i.e. complete tumor necrosis, no more vascular regions of the tumor that take up contrast), partial response is a decrease in the diameter of the enhancing region, stable disease is unchanged diameter, and progressive disease is an increase in the diameter of the enhancing region after treatment.

Inclusion Criteria: Study group Patients diagnosed with HCC, who will undergo resection or transplantation within 6 months, as part of routine clinical care and patients diagnosed with unresectable HCC 18 years of age and older Patient is able to give informed consent for this study Control group Healthy volunteers 18 years of age and older Subject is able to give informed consent for this study Exclusion Criteria: Age less than 18 years Unable or unwilling to give informed consent Contra-indications to MRI: Electrical implants such as cardiac pacemakers or perfusion pumps Ferromagnetic implants such as aneurysm clips, surgical clips, prostheses, artificial hearts, valves with steel parts, metal fragments, shrapnel, tattoos near the eye, or steel implants Ferromagnetic objects such as jewelry or metal clips in clothing Pregnant subjects Pre-existing medical conditions including a likelihood of developing seizures or claustrophobic reactions