DYNAMIC CONTRAST ENHANCED 1 H-MAGNETIC RESONANCE IMAGING IN ASSESSMENT OF SKELETAL MUSCLE AND FIBROSARCOMA PERFUSION

Background. The detection of neoplastic transformation and prediction of the thera­ peutic response are very important for effective cancer therapy. Current assessment of tumor treatment efficacy relies on evaluating changes in the tumor size or in treated b y 5­fluorouracil tu mors compared with control tumors suggest a shift in tumor metabolism from glycolysis to oxidation and/or a decrease in cell density. Conclusions. Method of dynamic contrast enhanced 1 H-MRI in combination with other NMR methods, positron emission tomography, histology, etc. should prove useful in assessment of neoplastic transformation of tumor capillaries and efficacy of chemo therapy.

Background. The detection of neoplastic transformation and prediction of the thera peutic response are very important for effective cancer therapy. Current assessment of tumor treatment efficacy relies on evaluating changes in the tumor size or volume, weeks to months after the assumption of a therapeutic protocol. The tissue perfusion is one of the most important parameter to estimate the neoplastic progress and the efficacy of antitumor therapy. 1 H-magnetic resonance imaging (MRI) is an effective tool that provides distinctive information related to structural, cellular, apoptotic, and necrotic changes in tumor tissue. The technique can be used widely for tumor detection and monitoring of the response to treatment.
Methods. Dynamic contrast enhanced 1 H-MRI was used for the assessment of tumor perfusion parameters in normal muscle and in subcutaneous Radiation Infused Fibrosarcoma -1 (RIF1) developed under the skin in C3H mice. Gadolinium (20 mM) was used as a capillaries perfusion tracer. Therapy of RIF1 was administered by a single intraperitoneal injection of 5fluorouracil (150 mg/kg). MRI experiments were performed before and 3 days after the treatment.
Results. Dynamic contrast enhanced 1 H-MRI has shown a much lower perfusion rate in RIF1 tumor compared to skeletal muscle. 5fluorouracil caused a significant decrease in subcutaneous RIF1 volume on days 2 and 3 posttreatment, as well as an increase in tumor inflow measured by Dynamic contrast enhanced 1 H-MRI. An increase in tumor tissue perfusion correlated with an increase in tissue apparent diffusion coefficient and total Na + concentration following 5fluorouracil chemotherapy reflect an increase in extracellular space and vasodilatation. On the other hand, as it was shown in our previous publications, the lower intracellular Na + concentration and glucose uptake

INTRODUCTION
Biological tissues hemodynamics plays an important role in normal and tumor tissues including an incretion and excretion of antitumor drugs. Several methods have been proposed for the quantification of tumor perfusion, such as a clearance of xenobio tics, singlephoton emission computerized tomography, positron emission tomography, and diffusion weighted 1 HMRI (DWI) [23]. Dynamic contrast enhanced (DCE) 1 H-magnetic resonance imaging (MRI) reflects perfusion in microvessels. This method has a good potential value in characterizing muscle pathological changes, differentia ting between malignant and benign sarcomas, monitoring response to chemotherapy [20,22,24] and predicting clinical efficacy of vascular disrupting agents with good ima ging histopathology correlation [12].
5-Fluorouracil (5FU) belongs to the antimetabolite and pyrimidine analog family of antitumor medication [21]. It is involved in blocking the action of thymidylate synthase and thus stopping the production of DNA [1]. It was shown that RIF-1 tumor volumes decreased in 5FU-treated mice [4]. 23 Na and 1 H MRI experiments show that this 5FU effect was correlated with an increase in both total tissue Na + level and water apparent diffusion coefficient (ADC) whereas intracellular Na + level and glucose uptake in treated tumors were lower compared with control tumors [2]. We hypothesized that a shift of energetic metabolism to oxidative phosphorylation has to be related to a better supply of oxygen to the treated tumor tissue. However, the possible role of tissue perfusion in 5FU antitumor effects was not clarified yet.
The goal of this paper was to estimate the possibility of applying DCE method in monitoring of blood perfusion in normal muscle tissue and radiation infused fibrosarcoma (RIF-1) treated with effective chemotherapeutic agent.

RIF-1 Tumor Model.
All animal studies were approved by the Indiana University Institutional Animal Care and Use Committee. RIF1 tumor cells were grown in monolayers using minimum essential medium (MEM; Mediatech, Herdon, VA, USA) supplemented with 10% fetal bovine serum, 10 mM HEPES, and 1% penicillin under a 5% CO 2 and 95% O 2 atmosphere at 37 °C. The tumor cells were passaged between in vitro and in vivo states according to the protocol [2].
Male C3H/HeN mice (Harlan, Indianapolis, IN, USA), approximately 6 weeks old and weighing 18 to 20 g, were inoculated in the right or left thigh (region of m. biceps femoris and m. gluteus maximus) with a subcutaneous (sc) injection of ∼2×10 6 cells in 0.10 to 0.15 mL volume of Hank's balanced salt solution. Animals were anesthetized with an intraperitoneal (ip) injection of 50 mg/kg ketamine, 5 mg/kg acepromazine, and 0.25 mg/kg atropine. The tumors were allowed to grow for 2 to 3 weeks to a volume of 0.7 to 1.0 cm 3 before performing the first MRI experiment. Tumor growth was monitored by caliper measurement. Tumor volume was calculated from three orthogonal diameters (x, y, and z) using the formula (π/6)xyz. Ten tumorbearing mice were treated with a single dose of 5FU (150 mg/kg, ip; SigmaAldrich, St. Louis, MO, USA) for tumor volume estimation. Nine animals served as untreated controls.
In Vivo MRI Experiments. All in vivo MRI experiments were performed on a 9.4T, 31cm horizontal bore system (Varian, Palo Alto, CA, USA) equipped with a 12cmdia meter shielded gradient set capable of up to 40 G/cm in three directions. A loopgap resonator (inner diameter = 30 mm, depth = 25 mm) tuned to 400 MHz for 1 H was used to collect DCE 1 HMRI of the muscle and tumor. The animals were anesthetized with 1-1.5% isoflurane delivered in medical air at 1.0-1.5 L/min using a mouse nose mask connected to a gas anesthesia machine (Vetland, Louisville, KY, USA). Animals were positioned on top of a custom-designed plastic cradle with the tuned loop-gap resonator attached to it [2]. The right or left thigh with subcutaneous tumor was positioned inside the resonator, and the animal was held in place with a tape. A detachable cylindrical phantom (6.5 mm diameter and 23 mm length) consisting of distillate water was also placed inside the resonator to serve as a water MRI standard. Warm air was blown through the magnet bore to maintain the rectal animal temperature during the MRI experiments (1-1.5 h) at 32-36 °C. The temperature was monitored with a fiber optic probe (FISO Technologies, Inc., Quebec, Canada). The magnet was shimmed to less than 100 Hz line width at half height of the 1 H water signal. The MRI experiments were performed prior to treatment with 5FU (150 mg/kg) and on day 3 after treatment.
DWI. Multislice DWI was collected using a modified spinecho sequence and the following parameters: repetition time (TR) = 1000 ms, echo time (TE) = 1000 ms/21 ms, δ = 6 ms, ∆ = 11 ms, matrix size = 256×128, field of view (FOV) = 80 mm×80 mm, number of slices = 12, slice thickness = 0.5 mm, slice gap =1.5 mm. DCE 1 H MRI. After collecting a baseline of DWI, 0.2 mmol/kg of GadoliniumDOTA was manually injected over a 30 s interval through a 26-gauge catheter placed in the tail vein. Contrast agent GadoliniumDOTA is commonly used for DCE MRI showing appro priate changes of tissue perfusion without damages of capillaries [16]. All bolus injections were performed by the same investigator. DCE 1 H-MRI was obtained using a gra dientecho sequence and the following parameters: TR/TE = 10 ms/3.1 ms, matrix size = 256×128, FOV = 64 mm×64 mm, number of slices = 1, and slice thickness = 4 mm. 200 images were collected over approximately 13 min, with 4.5 s acquisition time per image. Data analysis. 1 H images were reconstructed using the Image Browser software (Varian, Palo Alto, CA, USA). PSIPLOT software was used to analyze DCE 1 H-MRI data. The kinetics of contrast agent uptake was estimated by measuring the area under the curve over the first 60 s after the contrast agent arrival, as well as by fitting the DCE 1 H-MRI signal intensity versus time data to a triexponential function [5].
Tumor volume changes are presented as the mean ± standard error of mean (SEM) and represent the range across a cohort of animals. Statistical analyses of the data were performed by ANOVA (Statistica/v. 5.1 program). P ≤ 0.05 was used to define statistical significance in tumor volume changes.
All animal studies were approved by the Indiana University Institutional Animal Care and Use Committee.

RESULTS and DISCUSSION
Effect of 5FU on Tumor Growth. The mean tumor volumes of control and treated mice are shown in Fig 1. Before treatment, both groups had similar tumor volumes (0.72 ± 0.03 cm 3 for control group, 0.77 ± 0.05 cm 3 for treated group). Baseline of tumor volume in both control and 5FU treated groups was taken as 100%. In 5FU-treated animals, the mean tumor volume decreased by 38% (P ≤ 0.05 vs pretreatment) on day 2 and by 54% (P ≤ 0.05 vs pretreatment) on day 3 after treatment. The mean tumor volumes in control animals grew by 14% on day 2 and by 18% on day 3 and were larger (P ≤ 0.05) than the tumor volumes 2 and 3 days after treatment.   Fig. 2 shows the representative 1 H-MRI images before (images 1 and 2) and throughout (images 3-9) accumulation of perfusion tracer Gadolinium by a mouse C3H muscle (marked by dotted red line) and by subcutaneous RIF-1 (dotted yellow line). Gadolinium (20 mM) was delivered to animal through the tail vein. To enhance the diffe rence of MRI signals after Gadolinium treatment the baseline image 2 was extracted from each image 3-9 and marked as images 3′-9′. Delivering of the perfusion tracer to tissue microvessels is accompanied by an increase in brightness of the image. It is visib le that DCE in the muscle started much earlier (images 4 and 4′) than in subcutaneous RIF-1 (images 7 and 7′). The difference between DCE in muscle and tumor disap peared in images 9 and 9′ only.
The representative DCE 1 H-MRI signal intensity vs. time curves in the mouse muscle (m. biceps femoris and m. gluteus maximus) and subcutaneous RIF-1 are presented in Fig. 3. The rate of perfusion was calculated as a slope of the tracer permeability or outflow curves and was calculated as tg of angle α. The permeability rate following very next to the Gadoilinium bolus had a biexponential shape. During this period, the rate of perfusion in muscle was much higher compared to tumor: 19.1 and 8.14 (in muscle) and 8.14 and 2.14 (in tumor). After this period, the equilibrium of tracer inflow and outflow was observed, but the duration of this step was much shorter in muscle (130 c) than in tumor (430 c). During 450-1500 s (muscle) and 750-1500 s (tumor) the tracer washout rates were equal (tg α = -2.48) in both tissues.
The DCE 1 H-MRI in tumor vs. muscle ratio are presented in Fig. 4. Both traces were monitored on the same mouse before (pink dots) and 3 days after (dark blue dots) treatment with 5FU. Light blue and yellow lines represent linearized data points for before and after treatment cases, respectively. Baselines (before Gadolinium bolus) of DCE in tumor vs. muscle ratios were normalized to one. We assumed that DCE in normal musc le is relatively stable in both before and after treatment circumstances. Thus, all diffe ren ces presented in the Fig. 4 are caused mainly by DCE changes in the tumor. was delivered through the tail vein. To enhance the difference of MRI signals after the Gadolinium injection the baseline image 2 was extracted from each images 3-9 and marked as images 3′-9′. The reference, distillate water, was placed in plastic tube next to the mouse back thigh Рис. 2. Динамічноконтрастні 1 НМРТ зображення до (1 і 2) і впродовж (3-9) акумулювання перфузійного маркера ґадолінію скелетними м'язами (помічені червоною переривчастою лінією) і підшкірною радіаційноіндукованою фібросаркомою (жовта переривчаста лінія), прищепленою на задній лапці миші лінії С3Н. Ґадоліній (20 мМ) вводили через хвостову вену. Для покращення візуалізації МРТсигналу після введення ґадолінію базове зображення (2) "екстрагували" з кожного із зображень 3-9 і позначали як 3′-9′. Для стандарту використовували дистильовану воду, яку поміщали у пластиковій трубці біля задньої лапки миші The response to the Gadolinium injection was different before and after treatment. During the first 100-150 s period following the Gadolinium bolus, the inflow of tracer in the untrea ted tumor decreased by 40% (tg α = 0.18) while after treatment the inflow increased by 20% (tg α = 3.49). During the next 150 s period, which corresponds to the equilibrium period (see Fig. 3), the rate of perfusion in the treated tumor was still higher compared to the untreated tumor. Thus, these data show that after treatment with 5FU the rate of tumor permeability is higher. ∼ 300 s after the Gadolinium injection the DCE of tumor to muscle ratio became equal for both cases. During the next period up to 1500 s, representing mostly balance between inflow and washout, the DCE ratio of tumor vs. muscle was slightly higher in before treatment (tg α = 0.31) compared to after treatment case (tg α = 0.16). An early detecting and monitoring of the tumor development and treatment efficacy are very important in chemotherapy. In this work, the capillary tissue perfusion in ske letal muscle and subcutaneous fibrosarcoma were compared to evaluate the possibility to use DCE 1 H-MRI in the study of possible mechanisms of effective treatment of tumors. The Gadolinium inflow represents mostly a development of perfusion system in tissue whereas slow outflow most likely reflects a steadystate of inflow and washout of the contrast agent. The inflow was sufficiently slower in tumor tissue than that of the nearby normal muscle tissue (Fig. 3). The common explanation of this effect invokes an unregularly formatted blood vessels and a leaking of tumor capillaries [15]. Some contribution to the tumor vasoconstriction may be provided also an increase in number of the actively proliferated tumor cells creating pressure on the capillaries. Some tissue damages such as CCl 4 intoxication also lead to a decrease in inflow kinetics while outflow components did not show any significant difference [5].
In our previous studies, we showed that 5FU and cyclophosphamide significantly decreased subcutaneous RIF1 tumor volumes developed in C3H mice [3,4]. Li et al. [11] and Lemeir et al. [10] also reported a similar effect of 5FU therapy. At the same time points, in vivo MRI measurements showed an increase in both total tissue Na + NMR signal intensity and ADC in 5FU treated tumors while intracellular Na + signal intensity and glycolysis were significantly lower compared with control tumors. The correlated increases in tissue perfusion, total tissue Na + signal intensity and water ADC following chemotherapy reflect an increase in the extracellular space, vasodilatation, while the lower intracellular Na + signal intensity and glucose uptake in treated tumors compared with control tumors suggest a shift in tumor metabolism from glycolysis to oxidation and/or a decrease in cell density [4]. It was shown that 5FU therapy leads to an increase in the ATP level, ATP/P i ratio and pH i [6,19]. An increase in the extracellular space may be an outcome of the apoptotic and/or necrotic death of tumor cells [8]. The apoptotic mechanism includes cells shrinking, blebbing of cellular membranes and developing of apoptotic bodies, while the necrotic transformation is associated with water drainage, fibrosis, cells swelling and tissue inflammation. However, both mechanisms finally lead to endocytosis and phagocytosis of tumor cells. An increase in tumor blood circulation create better oxygen supply and a higher energetic status of leaving tumor cells [7,11,14,18]. Posttherapeutic reoxygenation may occur due to a number of reasons such as a decrease in distances between capillaries and their occlusion, as well as an increase in total vascularization of the tumor [17]. Therefore, an increase in both a diffusion, measured by ADC, and a perfusion, measured by DCE 1 HMRI, may be a promising tool to estimate early therapeutic effects prior to changes in tumor volume [9,13].

CONCLUSIONS
Method of dynamic contrast enhanced 1 H-MRI in combination with other NMR methods, positron emission tomography, histology, etc. should prove useful in the assessment of neoplastic transformations of tumor capillaries and efficacy of chemotherapy.