Integration of quantitative DCE-MRI and ADC mapping to monitor treatment response in human breast cancer: initial results

Abstract

Purpose

The objective of this study was to assess changes in the water apparent diffusion coefficient (ADC) and in pharmacokinetic parameters obtained from the fast-exchange regime (FXR) modeling of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) during neoadjuvant chemotherapy in breast cancer.

Materials and Methods

Eleven patients with locally advanced breast cancer underwent MRI examination prior to and after chemotherapy but prior to surgery. A 1.5-T scanner was used to obtain T₁, ADC and DCE-MRI data. DCE-MRI data were analyzed by the FXR model returning estimates of K^trans (volume transfer constant), v_e (extravascular extracellular volume fraction) and τ_i (average intracellular water lifetime). Histogram and correlation analyses assessed parameter changes post-treatment.

Results

Significant (P<.05) changes or trends towards significance (P<.10) were seen in all parameters except τ_i, although there was qualitative reduction in τ_i values post-treatment. In particular, there was reduction (P<.035) in voxels with K^trans values in the range 0.2–0.5 min⁻¹ and a decrease (P<.05) in voxels with ADC values in the range 0.99×10⁻³ to 1.35×10⁻³ mm²/s. ADC and v_e were negatively correlated (r=−.60, P<.02). Parameters sensitive to water distribution and geometry (T₁, v_e, τ_i and ADC) correlated with a multivariable linear regression model.

Conclusion

The analysis presented here is sensitive to longitudinal changes in breast tumor status; K^trans and ADC are most sensitive to these changes. Relationships between parameters provide information on water distribution and geometry in the tumor environment.

Introduction

Neoadjuvant chemotherapy is currently the standard of care for women with locally advanced breast cancer [1], [2], [3], [4]. Neoadjuvant therapy reduces tumor burden, allowing a less extensive surgical procedure and early initiation of systemic therapy in patients at high risk for distant and local failure. Various metrics of therapeutic response to chemotherapy are actively being investigated, but these methods generally require invasive procedures such as repeat biopsy [5], [6], [7], [8]. The development of noninvasive methods of tissue characterization that could be applied early in the course of treatment to assess response and to modify subsequent treatment would allow clinicians to tailor therapy on an individual basis based on each patient's response to a particular agent or to a combination of agents. Although X-ray mammography and ultrasound imaging play critical roles in the detection and the diagnosis of breast cancer, there are currently no adequate radiological methods for assessing tumor response to treatments. Conventional magnetic resonance imaging (MRI) of the breast has proven less successful than expected [9], but more specialized methods, including dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and diffusion-weighted magnetic resonance imaging (DW-MRI), have advanced to the point where they provide quantitative measurements of tissue properties that are highly relevant for assessing tumor progression and/or responses [10].

DCE-MRI involves the serial acquisition of MR images of a tissue of interest (e.g., a tumor locus) before, during and after an intravenous injection of a contrast agent (CA). As the CA enters into the tissue under investigation, the T₁ and T₂ values of tissue water decrease to an extent determined mostly by the concentration of the agent. By considering a set of images acquired before, during and after CA infusion, a region of interest (ROI) or individual voxels will display a characteristic signal intensity time course that can be related to CA concentration. By fitting DCE-MRI data to an appropriate pharmacokinetic model, physiological parameters that relate to, for example, tissue perfusion, microvascular vessel wall permeability and extracellular volume fraction, can be extracted [11]. In DCE-MRI, lesions often display characteristic enhancement patterns that have been shown to change reliably following treatments and are related to the extent and the integrity of the tumor vasculature [12], [13]. Thus, there is considerable and continuing interest in developing new and improved methods to obtain these parameter values accurately and precisely.

In the analysis of DCE-MRI data, linear dependence between the longitudinal relaxation rate constant R₁ (=1/T₁) and the concentration of CA in the tissue C_t is typically assumed [11], [14]. While true for a homogeneous solution, the use of a linear relationship for tissue assumes that transcytolemmal water exchange is effectively infinitely fast — what is commonly referred to as the fast-exchange limit (FXL) of the nuclear magnetic resonance time scale. Analyses of DCE-MRI data reported in the literature almost exclusively assume that water exchange is fast. However, several recent contributions have demonstrated that when CA concentration is high, this assumption may be violated [14], [15]. By incorporating the pharmacokinetic theory of Kety [16] into a two-site exchange model, “fast-exchange regime” (FXR) formalism is obtained [14], [15]. Initial applications of the FXR method on human DCE-MRI data suggest that assuming the FXL can underestimate the volume transfer constant (K^trans) and extravascular–extracellular volume fractions (v_e) by values up to 300% [15], [17].

The microscopic thermally induced random motion of water molecules is referred to as self-diffusion or Brownian motion. The rate of water diffusion is described by the apparent diffusion coefficient (ADC); in a system of small compartments (cells) separated by semipermeable barriers (cell membranes), the ADC may largely depend on the separation of barriers. In particular, the ADC has been shown to correlate with tissue cellularity; in addition, it has been shown that the exposure of tumors to both chemotherapy and radiotherapy leads to measurable increases in water diffusion in cases of favorable treatment response [18]. Preliminary studies in humans have shown that ADCs in both normal tissues and benign lesions have ADCs significantly higher than those found in malignant breast lesions [19], [20]. Furthermore, recent results in rodents have shown a twofold increase in ADC values in tumors following treatment [21].

This pilot study was designed with three goals in mind. The first goal is to determine changes that can be measured with the application of FXR DCE-MRI analysis to longitudinal studies of breast cancer patients receiving neoadjuvant chemotherapy. The second goal is to extend available data on the ADC mapping of breast cancer response to treatment. This is the first effort to combine both techniques to study the in vivo response of human breast cancer to treatment. The third goal is to perform an exploratory analysis on the relationships between measured parameters and pathologic response at the time of definitive surgery.