Diffusion changes precede size reduction in neoadjuvant treatment of breast cancer

Abstract

Traditionally, tumor response has been assessed via tumor size measurements during the course of a treatment. However, changes in these morphologically based measures occur relatively late in the course of a treatment. Alternative biomarkers are currently being evaluated to enable an earlier assessment of treatment to facilitate early cessation and cost savings. Diffusion-weighted imaging (DWI) has been identified by preclinical studies to be a likely alternative to tumor size measurements.

In this study, 10 patients were examined prior to and after the first and second chemotherapy cycle time points. Longest diameter tumor measurements and apparent diffusion coefficients (ADCs) were recorded at each exam. An increase in the mean (normalized) ADC was noted as early as the first cycle time point. However, a reduction in the mean (normalized) longest diameter was only noted at the second cycle time point. Significant alterations from the baseline value were noted for ADC at the first (P=.005) and second cycle time points (P=.004). Longest diameter measurements only achieved a borderline significance at the second time point (P=.057).

These results indicate that DWI may provide a suitable biomarker capable of providing an indication of response to treatment prior to tumor size measurements.

Introduction

Traditionally, in oncology, response to treatment has been assessed via tumor size measurements. For breast cancer patients, these measurements are undertaken by a variety of techniques including clinical palpation, X-ray mammography, ultrasound and magnetic resonance imaging. Unfortunately, assessment of treatment response via this approach is considered a late event since functional changes occur prior to alterations in size [1], [2], [3]; additionally, tumor size assessments are usually undertaken in the clinic halfway through a course of treatment. Biomarkers that can provide an earlier indication of response are urgently required. Ultimately, these methods will enable cessation of ineffective treatments, thereby minimizing unnecessary toxicity and cost of a failing treatment.

In diffusion-weighted imaging (DWI), the MR signal is attenuated due to the thermally driven motion of water molecules affording an insight into tissue microstructure [3], [4]. The apparent diffusion coefficient (ADC) provides a quantifiable measure of the signal attenuation and, therefore, the molecular motion of water. The ADC is high in tissues with few obstacles to the motion of water, whereas ADC is low in tissues with many obstacles. Indeed, ADC has been demonstrated to be inversely correlated to cellular density [5], [6], [7], [8]. It is widely held that following successful treatment, tumor ADC values will be increased, reflecting a reduction in cellular density and barriers to water motion. A number of recent preclinical animal and cell models have indicated the potential usefulness of DWI in predicting response to a number of different treatments including radiotherapy [9], chemotherapy [10], [11], [12], [13], photodynamic therapy [14] and gene therapy [10]. Two general points arise from these papers. Firstly, increases in ADC values were noted early, hours to days, post treatment and prior to tumor size changes. Secondly, the authors believed that the increases in ADC values were associated with treatment-induced tumor regression such as necrosis, reduced cellular density and reduced barriers to water motion. Encouraging as these results are, to date, few papers demonstrating similar results in human studies have been published. However, DWI has been utilized to assess response to treatment in brain tumors [3], [4], [15], metastatic liver lesions [16] and rectal carcinomas [17], [18], [19], [20]. It should be noted that in the papers examined, the treatments were of a cytolytic nature, reflecting a more traditional treatment approach. Currently, a whole host of new drugs are undergoing trials prior to general clinical use. Some of these drugs have different actions including cytostatic and antiangiogenic actions whereby cell membranes and cellular densities remain unchanged. Consequently, a treatment-induced increase in ADC following successful treatment may not be anticipated with these drugs. Equally, not all treatments are drug based. Jacobs et al. [21] recently demonstrated elevated ADC values in uterine fibroids treated with either embolization or MRI-guided, focused ultrasound thermal ablation.

This work tests the hypothesis that diffusion changes occur prior to tumor size changes, as evidenced by alterations in ADC and longest diameter measurements, during the course of treatment in a cohort of patients with invasive breast cancer undergoing neoadjuvant chemotherapy.