Diffusion-weighted MRI in abdominal oncology: Clinical applications

Abstract

Diffusion-weighted magnetic resonance imaging (DWI) provides image contrast that is different from that obtained by conventional magnetic resonance techniques. Although previously, DWI has been used to evaluate various diseases of the central nervous system, several technical advances have expanded the clinical applications of DWI beyond the central nervous system. As a result, many reports have been published on the use of DWI in abdominal diseases. Particularly, abdominal DWI has now being focused on evaluation of patients with abdominal cancer. DWI can be used for pretreatment tumor detection, characterization including predicting tumor response to therapy, monitoring tumor response during therapy, and follow-up study after treatment to detect possible tumor recurrence.

Key Words: Diffusion weighted magnetic resonance imaging; Abdominal neoplasms

INTRODUCTION

Diffusion-weighted magnetic resonance imaging (DWI) has enabled us to obtain additional information derived from the microscopic motion of water protons, which is not possible using conventional magnetic resonance imaging (MRI). Previously, DWI has been used to evaluate various diseases of the central nervous system. The most established clinical application of DWI for the central nervous system is evaluation of acute stroke[1].

DWI has many advantages. First, it is completely noninvasive, does not require exposure to ionizing radiation or injection of contrast material, and does not cause patient discomfort. Second, because it is derived from a well-established MRI technique, DWI does not require expert technicians with sophisticated technical skills or expensive equipment, such as a cyclotron that is required for positron emission tomography. Another advantage of DWI is that it can be added easily to a routine MRI protocol because it requires only a very short prolongation of examination time[2].

Recently, several technical advances have expanded the clinical applications of DWI beyond the central nervous system, and many studies have been published on the use of DWI in abdominal diseases. Particularly, abdominal DWI has now being focused on evaluating patients with abdominal cancer[3-9]. In this article, the application of DWI in abdominal oncology is described.

HOW TO INTERPRET DWI

DWI can offer qualitative and quantitative information that can be helpful for tumor assessment (Figure 1). The former assesses visual differences in signal intensity between a tumor and its surrounding normal tissue, and the latter enables calculation of values (apparent diffusion coefficient, ADC) obtained from DWI, such as a computed tomography value.

Figure 1 MRI of a patient with gallbladder carcinoma. A: Axial T2-weighted MRI of a patient with gallbladder carcinoma shows a mass (arrow) protruding into the gallbladder lumen; B: Corresponding axial DW image shows high intensity (arrow); C: Corresponding color fusion image of T2-weighted image and DW image shows gallbladder carcinoma (arrow). On color fusion images, the red area corresponds to high signal intensity on DW images and blue correspond to low intensity; D: Corresponding ADC map shows low intensity (arrow).

Qualitative assessment in DWI

Visual assessment of relative tissue signal intensity on DWI is being used for tumor detection and characterization[3]. Tumors generally tend to block diffusion more than the tissue from which they originate and show relative high signal intensity on DWI (Figure 1B); however, some normal organs, such as the spleen, adrenal gland and seminal vesicle, also show high signal intensity on DWI. Moreover, DWI has a pitfall known as “T2 shine-through”. DWI is obtained by adding a diffusion-weighting gradient (known as an MPG) to T2-weighted images, the basic sequence of conventional MRI. Thus, because DWI shows signal intensity that depends on diffusion and T2 signal intensity, a region with a high T2 signal retains the high signal on DWI, and may be mistaken for restricted diffusion. Therefore, special care must be taken with these pitfalls in diagnosing with DWI. DWI is usually interpreted by superimposing DWI and conventional morphological T2-weighted images because DWI cannot show minute morphological structures (Figure 1C).

Quantitative analysis in DWI

Quantitative tumor assessment is possible by calculating ADC after performing DWI with changed parameters (known as b values). ADC values in various malignant lesions generally tend to decrease, probably due to increased tissue cellularity or cell density, because the latter correlates with malignancy (Figure 1D). In addition to the cellular membranes, intracellular cytoskeleton, organelles, matrix fibers and soluble macromolecules contribute to diffusion restrictions in tumors[10]; therefore, ADC values are expected to reflect histopathological tissue characteristics. ADC is calculated for each pixel of the image and is displayed as a map. By setting regions of interest within tumors on these maps, ADCs of the tumor can be measured.

CLINICAL APPLICATIONS OF DWI IN ABDOMINAL ONCOLOGY

Tumor detection and characterization

Tumors generally tend to show relative high signal intensity on DWI. Using qualitative assessment, Nasu et al[11] have shown that DWI is superior to superparamagnetic iron oxide (SPIO)-enhanced MRI in detecting liver metastases, which had been the best available examination technique. They have reported that the sensitivity and specificity of DWI was 82% and 94%, respectively. Koh et al[12] also have reported that the sensitivity and specificity of DWI for detecting liver metastases was 78% and 95%. Thus, qualitative assessment with DWI has superior ability for assessing liver metastasis.

In colorectal tumors, Ichikawa et al[7] have shown that DWI has high sensitivity and specificity for detecting tumors, and several authors have shown that DWI has high sensitivity and specificity for detecting tumors even in the pancreatico-biliary system[6,9].

In quantitative assessment of DWI, ADC measurement has the potential to differentiate benign and malignant liver tumors. In many studies, malignancy has a lower ADC value than benignity. Taouli et al[13] have shown that metastatic liver tumors have the lowest ADC in malignant and benign focal lesions of the liver, and have revealed a significant difference between benign and malignant lesions. Chan et al[14] have shown that DWI can be used to distinguish between hepatic abscess and cystic or necrotic malignant liver tumor; ADC of abscess cavities has a lower value than that of cystic or necrotic malignant liver tumors. Also in abdominal tumors other than in the liver, ADCs of malignant lesions have shown lower values[9,15,16]. However, most studies have reported that ADC measurement has no clear threshold to discriminate malignant and benign tumors because of substantial overlapping[9,13,15-18].

Predicting and monitoring response to therapy

Conventional criteria using morphological images have been used to evaluate antitumor therapy; however, measuring tumor size is often not adequate when tumors are treated with cytotoxic therapy and molecular targeting agents, because changes in tumor size after therapy with these drugs are not expected[4,19]; therefore, a new method for evaluating tumor response is required that can precisely reflects the clinical outcome, earlier than conventional imaging modalities.

The ability of DWI to predict therapy outcome has been shown in many clinical studies. Several authors have reported that tumors with low pretreatment ADC values show a better response to various therapies than those with high ADC[20-24]. However, studies of areas other than the abdomen have addressed that the relationship between pretreatment ADC and prognosis yield, with different results: patients suffering from a tumor with high pretreatment ADC show better long-term post-treatment prognosis than those with low ADC[25,26].

Many researchers have reported that DWI has the potential for evaluating tumor response during treatment. The results of animal studies have proved that ADC increases can be depicted in those responding to treatment[27]. In clinical studies, researchers have reported that an early increase in the ADC value after starting therapy suggests a better treatment outcome[20,28-32].

Monitoring response to therapy by visual assessment of DWI has been reported in brain tumors and bone metastasis, but not in the abdominal region[21,33]. Studies on bone metastasis have revealed that the treatment response after therapy could be assessed as a decrease in signal intensity[33]. Several authors have shown that tumors demonstrate an increase in ADC after treatment before a change in tumor size occurs, which heralds later diminution of the tumor size[18,22,34-38]. Chen et al[34] have reported that patients with hepatocellular carcinoma show a significant rise in ADC value when they respond to treatment. Koh et al[22] also have reported that patients with colorectal hepatic metastases show an increase in ADC, at least in those who show a partial response to treatment, but not in non-responders. A decrease in ADC during follow-up suggests tumor recurrence[27].

FUTURE DEVELOPMENT

Several studies have indicated that DWI may be useful for tumor staging, including lymph node and distant metastases[21,39-42]. For tumor staging, whole-body imaging is desirable. Takahara et al have shown that whole-body DWI is promising[43-45] using their method to examine the whole body by composite construction of segmented imaging. The images are processed using maximum intensity projection and 3D display[43-45]. More clinical research on this technique is needed because their study was preliminary.

The most important issue regarding DWI is non-standardization among MRI manufactures and researchers. Substantial differences in the ADC values of the same normal and diseased organs have been presented[5] by researchers using a different imaging technique; therefore, standardization of the imaging protocol is fundamental.

Currently, spatial resolution of DWI is not high enough. In order to compensate for such limited resolution, qualitative assessment might need superimposition of DWI on corresponding T2-weighted images, and quantitative assessment may require meticulous ADC measurements for small lesions. Utilization of high-field MRI may be able to solve the issue of limited spatial resolution.

CONCLUSION

DWI is a promising imaging technique to evaluate abdominal tumors. This technique can be used for pretreatment tumor detection, characterization including predicting tumor response to therapy, monitoring tumor response during therapy, and follow-up study after treatment to detect possible tumor recurrence. Standardization of the imaging protocol and large clinical trials regarding the usefulness of DWI are needed.