Sunday, April 5, 2015
In assessing treatment response in oncologic patients, computed tomography (CT) and magnetic resonance imaging (MRI) remain the workhorses. CT is routinely used to survey the extent of disease throughout the chest, abdomen and pelvis; MRI is particularly useful for evaluating hepatobiliary tumors including primary hepatocellular carcinomas (HCC) and liver metastases. Response Evaluation Criteria In Solid Tumors (RECIST) (1) remains the standard for therapy evaluation in most clinical trials, but new methods have been developed to predict and evaluate tumor response to novel targeted therapies. Although CT and MRI techniques are generally viewed as modalities for anatomic imaging, their potential for functional imaging of tumor response is a very exciting area of research.
Advances in Computed Tomography
RECIST and similar anatomic-based criteria rely on the detection of changes in tumor dimensions, an approach that is ideal for assessment of response to cytotoxic chemotherapies that shrink tumors. With some recently developed novel therapies that are designed to more precisely target specific pathways in cancer cells, tumor shrinkage may be minimal, despite an obvious therapeutic benefit.
In the treatment of gastrointestinal stromal tumors (GIST) with imatinib, a tyrosine kinase inhibitor, alternative criteria were developed that include changes in tumor attenuation. Choi et al (2) showed that decreases in tumor density on CT predicted a clinical response and prolonged survival for metastatic GIST patients treated with imatinib. The Choi Criteria is one of several recently developed response criteria that extend beyond simple tumor size measurements and take into account the mechanism of action for specific therapies.
For example, at Memorial Sloan Kettering Cancer Center (MSK), we have developed immune-related response criteria for melanoma patients treated with immunotherapeutic agents. (3) Our diagnostic radiologists are familiar with the newer response criteria that are critical to the optimal management of oncologic patients.
More recently, we have investigated the use of texture analysis of CT images for assessment of treatment response in patients with hepatopancreatobiliary disease. Although changes in density can be observed for liver metastases from colorectal cancer (CRC) during the course of chemotherapy, changes in tumor heterogeneity, as measured with texture analysis, are often more apparent. This is an emerging method to measure changes in tumor appearances, using established image processing tools to quantitatively describe the homogeneity or heterogeneity of gray-scale images. Also in ongoing studies, we are investigating the prognostic role of texture analysis in tumor response (Figure 1).
Texture analysis also has potential for evaluation of the liver parenchyma during the course of chemotherapy (Figure 2). A retrospective analysis of chemotherapy-treated patients undergoing major hepatic resection demonstrated differences in pre-hepatectomy liver parenchyma texture between those who experienced post-hepatectomy liver failure and those who did not. Additional studies exploring the applications of texture analysis in CT imaging of cholangiocarcinomas and pancreatic cancers are ongoing.
Advances in Magnetic Resonance Imaging
MRI is the modality of choice for the characterization of focal liver lesions. It also provides superior soft tissue contrast compared to CT and offers unique contrast mechanisms based on tissue properties that are not evaluable by CT, for example, the motion of water molecules by diffusion weighted imaging (DWI), (4) or the uptake in functioning hepatocytes of a hepatobiliary contrast agent such as gadoxetic acid. (5)
The motion of water is inherently more limited within tumors because the cells are dividing more rapidly and packed more closely together than in most organs, including the liver. During the course of treatment with embolization or chemotherapy, water motion often increases as the tumor cell membranes break down—a physiologic change that can be measured by DWI through their apparent diffusion coefficients (Figure 3). This technique has shown early promise in the evaluation of cancer treatment responses for a number of different tumors, including HCC and CRC liver metastases. We are currently investigating the utility of DWI for the assessment of treatment in patients with metastatic pancreatic neuroendocrine tumors (IRB#12-058).
The physiology underlying changes in tumor density on CT is a decrease in the uptake of intravenous (IV) iodinated contrast. Tumor uptake of IV contrast can also be evaluated by MRI using dynamic contrast-enhanced (DCE)-MRI. (6) Differences in the uptake of contrast measured by DCE-MRI may predict the response for intrahepatic cholangiocarcinoma (ICC) when treated with hepatic arterial infusion pump chemotherapy. (7) Studies to explore the utility of DCE-MRI in patients with unresectable ICC treated with a combination of systemic gemcitabine, oxaliplatin and pump floxuridine (FUDR) are ongoing (IRB#13-066).
Members of the Department of Radiology at MSK are committed to pushing the limits of what CT and MRI can provide for our patients, and to expanding our capability for precision medicine.Back to top