Diffusion-weighted imaging in rectal carcinoma patients without and after chemoradiotherapy: A comparative study with histology

Abstract

Diffusion-weighted imaging (DWI) can be used to quantitatively assess functional parameters in rectal carcinoma that are relevant for prognosis and treatment response assessment. However, there is no consensus on the histopathological background underlying the findings derived from DWI. The aim of this study was to perform a comparison of DWI and histologic parameters in two groups of rectal carcinoma patients without (n = 12) and after (n = 9) neoadjuvant chemoradiotherapy (CRT). The intravoxel incoherent motion (IVIM) model was used to calculate the diffusion coefficient D and the perfusion fraction f in rectal carcinoma, the adjacent rectum and fat in the two patient groups. Immunohistological analysis was performed to assess the cellularity, vascular area fraction and vessel diameter for comparison and correlation. Out of 36 correlations between parameters from DWI and histology, four were found to be significant. In rectal carcinoma of patients without CRT, the diffusion D and the perfusion f correlated with the vascular area fraction, respectively, which could not be found in the group of patients who received CRT. Further correlations were found for the rectum and fat. Histological evaluation revealed significant differences between the tissues on the microscopic level concerning the cellular and vascular environment that influence diffusion and perfusion. In conclusion, DWI produces valuable biomarkers for diffusion and perfusion in rectal carcinoma and adjacent tissues that are highly dependent of the underlying cellular microenvironment influenced by structural and functional changes as well as the administered treatment, and consequently can be beyond histological ascertainability.

Introduction

Diffusion-weighted MRI (DWI) is a technique that exploits differences in the extracellular movement of water protons to discriminate between tissues of varying properties depending on the respective cellular microenvironment. For oncologic imaging, DWI has shown promise for identification of malignant tumor and is frequently used as reliable tool for assessment of tumor response [1], [2]. It has been shown previously that diffusion values correlate inversely with tissue cellularity in many tumors such as in brain [3], but not in others, e.g. in multiple myeloma [4]. For rectal carcinoma, recent studies showed that DWI was useful to identify tumor in the rectum and differentiate between cancer and residual scar tissue after chemoradiotherapy (CRT) [5], [6].

Modern approaches to cancer therapy, e.g. anti-angiogenic and anti-tumor effects of therapy require a therapy-adapted response monitoring [7]. Currently monitoring is mostly based on the change in size of the tumor as assessed according to the response evaluation criteria in solid tumors (RECIST) [8], [9]. As therapy may, however, result in constant tumor size but modified tissue composition, it is expected therefore that functional techniques such as DWI, that reflect these alterations in tissue micro- rather than macrostructure may replace the current standards [10]. To support the use of DWI parameters in decision-making about pharmaceuticals, it is important to link DWI derived parameters to underlying pathophysiological processes both before and after therapeutic interventions. In patients with rectal carcinoma receiving neoadjuvant CRT, a significant increase of the apparent diffusion coefficient (ADC) was reported [11], while others contradictorily showed a decrease in ADC [12], [13].

The ADC derived from DWI reflects functional processes in tumor tissue including a combined measure of the molecular movement of water (diffusion) and microcirculation of blood in the capillaries (perfusion). Le Bihan et al. proposed the intravoxel incoherent motion (IVIM) model to separate perfusion and diffusion effects [14], which can be applied to oncologic imaging [15], [16]. In brief, the IVIM theory exploits the differences in signal decay between the blood and tissue compartments, i.e. the blood magnetization dominates the signal decay at low b-values enabling a differentiation between the perfusion fraction f and perfusion-free diffusion D [14]. Recently, the IVIM model was used for the differentiation between chronic pancreatitis and pancreatic carcinoma [15] or healthy pancreatic tissue and pancreatic carcinoma by the perfusion fraction f, respectively [16]. To the best of our knowledge, no correlation for parameters of cellularity and vascularization has been performed for rectal carcinoma between DWI and histological properties. Therefore, we applied the IVIM model to patients with rectal carcinoma with and without preceding CRT for correlation of DWI values of diffusion and perfusion with postoperative histological parameters including cellularity, vascular area fraction and vessel diameter.