A Novel Approach To Monitor The Efficacy Of Anti-tumor In Animal Models:Combining Functional MRI And Texture Analysis

Objective:Our study aimed to establish two kinds of functional magnetic resonance imaging (fMRI) technical proposals-- multi b-value diffusion weighted imaging (DWI) and dynamic contrast-enhanced MRI (DCE-MRI) for 4T1 breast tumor homograft-bearing mice, and evaluate and monitor the early anti-tumor therapy response of bevacizumab and paclitaxel by the combination of multi b-value DWI, DCE-MRI and Texture Analysis. To explore the anatomy and cellular molecular level changes in tumor after treatment.Methods:The 18 homograft tumor models were implanted subcutaneously with 4T1 breast cancer cell line. The multi b-value DWI and DCE-MRI examinations were performed on a GE Discovery MR750 3.0 T horizontal bore superconducting magnet, and coupled with a 35mm diameter small animal coil.The therapy was initiated after tumors reached approximately 150 mm3 in volume. Then these 4T1 breast tumor homograft-bearing mice were randomized into untreated group (n=6) and three experimental groups:control, paclitaxel monotherapy and combination therapy with antiangiogenic bevacizumab and paclitaxel (6 mice in each group). All of the mice were treated with intraperitoneal injection every three days. Sterile saline was used to control group with the volume of 100ul, and the dose of 10 mg/kg was used to paclitaxel monotherapy group. In combination therapy group, the mice were tested with same dose of 10 mg/kg each. The whole treatment process was lasted for 15days. The growth trend of tumor in each group was observed. All of the mice were scanned immediately prior to the treatment and 15 days after the initiation of the treatment. The relevant parameters were measured after MRI examinations. ADCslow (pure molecular diffusion), ADCfast (perfusion related diffusion), and f (perfusion fraction) were obtained from a bi-exponential IVIM model of multi b-value DWI by using GE Functool ADW 4.5 software system. Pharmacokinetic parameters of Ktrans(transfer rate constant), Kep(reverse rate constant), Ve(extravascular extracellular volume fraction), fPV(fraction of plasma volume), AUC90(area under curve 90s) and AUC180(area under curve 180s) were obtained from the Toft two-compartment model of DCE-MRI by using GE Cinetool ADW 4.5 software system. There were 6 mice and 18 mice were sacrificed after the first scanning and last scanning, respectively. The tumor tissues from these three groups were subjected to histopathological analysis of vascularization.The texture parameters were obtained by an advanced research software algorithm—TexRAD, an image-histogram technique invented by University of Sussex (United Kingdom), from the axial T2 weighted images about all animals, and the regions of interest (ROI) were defined as the tumor outline in the largest cross-sectional images performed by an experienced radiologist (8 years of experiences in imaging analysis) with manually delineating. The ROI areas were selected by different spatial scale filter (SSF) values from 0 to 6 mm to extract MR texture features. The SSFs of 0 and 2 reflect fine texture scales, and 3,4,5 SSFs reflect medium texture scales, and an SSF of 6 reflects coarse texture scale. The heterogeneity of these tissues were indicated by those following histogram parameters:mean intensity, standard deviation of the gray-level histogram distribution (SD), entropy (irregularity of gray level distribution), mean value of positive pixels (MPP), kurtosis (flatness of the histogram), and skewness (asymmetry of the histogram). These quantitative parameters were related to the tumor histological features, such as blood and oxygen supply, necrosis, and fibrosis.Quantitative parameters as described above were acquired from the functional MRI and analyzed in SPSS 20.0. The data under paclitaxel monotherapy and combination therapy were compared with the control by use of analysis of variance (ANOVA). The correlations between MRI parameters and pathological features data were analyzed by means of linear regression.Differences in textural feature values between before and after treatment within control group, paclitaxel monotherapy group and combination therapy group were tested using the Mann-Whitney U test respectively.All of the tests were two-tailed. The level of P values less than 0.05 were considered to be statistical significant.Results:Multi b-value DWI was acquired with 11 grading b values using axial spin-echo sequence (0,20,50,100,200,400,600,800,1000,1200,1500 s/mm2), and calculated ADCfast and f in small b value (b<200 mm2/s), the ADCslow was obtained by higher b-value. The DCE-MRI was followed by a 200-phase dynamic series of T1WI 2D FSPGR images with identical geometry and a temporal resolution of 3 seconds, and the quantitative or semi quantitative parameters including Ktrans, Kep, Ve, fPV, AUC90 and AUC180 were calculated according to Toft two-compartment model.The baseline tumor volumes in control, paclitaxel monotherapy and combination therapy groups were 192.4±47.7 mm3,263.7±82.8 mm3 and 195.3±85.2 mm3 respectively, and there were not significant different between them (P=0.26). Similarly, the growth of 4T1-tumor in these three groups had no conspicuous difference on day 7 after therapy (control, paclitaxel, paclitaxel unite bevacizumab:156.5±48.7%,119.3± 42.0% and 118.7±48.0%, P=0.60). However, after 15 days therapy, the measuring results showed that tumors in the blank control group were significantly larger compared to the combination therapy group, the tumor volumes reached 652.5±142.8 mm3 with no therapy and the tumors only 416.2±157.5 mm3 in paclitaxel unite bevacizumab conjoint therapy conversely (P=0.018). Meanwhile, the volumes of paclitaxel group was 521.2± 129.0 mm3. Accordingly, no obvious difference were found between control and paclitaxel monotherapy group (P=0.177), and the distinction between the two treatment groups was less intuitive (P=0.055).The multi b-value diffusion weighted imaging (DWI) after all the treatments showed increasing trends of ADCslow value in these three groups, especially a distinct increase in the combination therapy group (control:42.17±19.0%, paclitaxel:53.74±24.16%, combined treat group:118.84±47.59%). There were significant difference between control and combining treatment groups (P=0.001), and the same difference was reflected in the two therapeutic groups (P=0.008). Regrettably, no conspicuous difference was found between control and paclitaxel monotherapy groups (P=0.269). Even more remarkably, the perfusion fraction (f) values showed the opposite situation. The growth trends of f values were showed in control and paclitaxel groups (control:36.72±17.47%, paclitaxel:52.24±36.35%), while the bevacizumab and paclitaxel combination agent group showed a decrease (-25.12±47.39%) on day 15 after initiation of therapy. Meanwhile, the statistical differences between control and combination therapy groups, as well as these two therapeutic groups were highly significant (P=0.013, P=0.005 respectively). Similarly, there was no significant difference in the f values between control and paclitaxel monotherapy groups (P=0.671).A comparative analysis of the DCE-MRI results between before and after anti-tumor therapy in the three groups exhibited some significant difference. The transfer rate constant (Ktrans) vaules in the two therapeutic groups showed a significant decrease, however, the control group showed an increase (paclitaxel:-28.8±20.3%, combined treat group:-55.42±30.43%, control:127.37±76.7%) on day 15 after treatment. Accordingly, the statistic results were very similar to DWI findings. There were significant differences between control and combining treatment groups (P=0.003), along with the two therapeutic groups (P=0.044). Although no statistical significance was detected in the Ktrans values between control and paclitaxel monotherapy groups (P=0.219). Furthermore, there was no significant difference in other parameters between these three groups.Analysis of tumor texture in pre-, mid-and post-treatment in these three groups respectively, to examine microstructure change and therapy response, has revealed that the entropy values were unceasingly increasing with or without therapy in the three groups and all the changes had significant statistical difference within group (P<0.01, under all the SSF values, from 0 to 6mm). In addition, the MPP, mean intensity and SD values were all showing the same increasing tendency only in bevacizumab and paclitaxel combination therapy group for medium and coarse features (SSF=4,5,6), and these differences had statistics significance (PMPP<0.05, Pmean<0.05, PSD<0.03 respectively).There was no difference in the mean, SD, entropy, MPP, skewness and kurtosis among the three groups before treatment. With the implement of various handling, compared to pretherapy, the mean and MPP values under fine and medium features by using SSF of 0,2 and 3mm demonstrated significant differences among different groups at posttreatment (Pmean<0.05 and PMPP<0.05). However, the change of other parameters was not remarkable.Histological analysis of 4T1 allograft tumors showed that the combined medication treatment caused significant tumor suppression and CD31 immunostaining had a higher specificity of new vessels, compared to CD34. The quantitative analysis of the microvessel density (MVD) was assessed by CD31 revealed that a obviously decrease in combination therapy group after 15 days treatment, which was in deep contrast with other two groups (combined treat group:-17.61±23.16% VS control:31.39±30.41%/ paclitaxel:30.12±27.65%). Admittedly, these detection results also had significant statistical differences (combination therapy VS control/paclitaxel:P=0.007, P=0.006). Moreover, the same changing trends of MVD in control and paclitaxel monotherapy groups did not cause significant difference (P=0.907). The average optical density of VEGF also showed the same changes among these groups. Through the treatment of bevacizumab and paclitaxel, the VEGF average optical density emerged a decrease (-13.50±57.25%) in combination therapy group, and control and paclitaxel monotherapy groups demonstrated a increase tendency on the contrary (14.20±44.41%,27.50±96.19% respectively).In order to further clarify our research, association study was taken among the above results. It contained the comparisons of MVD versus DWI/DCE-MRI, DWI versus DCE-MRI, and texture analysis versus DWI/DCE-MRI. The correlation coefficients’r’ of the percentage change of MVD versus Ktrans was 0.612 (P=0.012), MVD versus ADCslow was-0.810 (P=0.001), MVD versus perfusion fraction (f) was 0.580 (P=0.019), Kep versus ADCfast was -0.593 (P=0.016), ADC slow versus entropy was -0.503 (P=0.047), and ADCslow versus MPP was 0.603 (P=0.013). In addition, MVD was also positively correlated with the expression of VEGF (r=0.563, P=0.023).Conclusion:In this study, we aimed to explore a practical approach to monitor and assess the efficacy of neoadjuvant chemotherapy. It proved the multi b-value DWI and DCE-MRI have irreplaceable advantages in this respect. The two fMRI techniques can quantitative observation the perfusion and microstructure characteristics in tumor, such as blood perfusion, cell membrane integrity and cell density. Texture analysis can highlight the tumor anatomy and biological heterogeneity, which can reflect the metabolism, oxygenation, and new blood vessel formation within tumor in functional and cellular molecular levels, and greatly verify and enlarge the image information provided by functional magnetic resonance. The three reciprocal causation and inextricably linked, can be used as a strong potential molecular imaging technology to evaluate the efficacy of oncotherapy.