Skip to main content
NCBI home page
Neoplasia (New York, N.Y.) logoLink to Neoplasia (New York, N.Y.)
. 2004 Mar;6(2):136–142. doi: 10.1593/neo.03349

Pretreatment Prediction of Brain Tumors' Response to Radiation Therapy Using High b-Value Diffusion-Weighted MRI1

Yael Mardor *,**, Yiftach Roth *,†, Aharon Ocherashvilli *, Roberto Spiegelmann ‡,§, Thomas Tichler , Dianne Daniels *, Stephan E Maier #, Ouzi Nissim ‡, Zvi Ram ‡,§, Jacob Baram §,, Arie Orenstein *, Raphael Pfeffer §,
PMCID: PMC1502089  PMID: 15140402

Abstract

Diffusion-weighted magnetic resonance imaging (DWMRI) is sensitive to tissues' biophysical characteristics, including apparent diffusion coefficients (ADCs) and volume fractions of water in different populations. In this work, we evaluate the clinical efficacy of DWMRI and high diffusion-weighted magnetic resonance imaging (HDWMRI), acquired up to b = 4000 sec/mm2 to amplify sensitivity to water diffusion properties, in pretreatment prediction of brain tumors' response to radiotherapy. Twelve patients with 20 brain lesions were studied. Six ring-enhancing lesions were excluded due to their distinct diffusion characteristics. Conventional and DWMRI were acquired on a 0.5-T MRI. Response to therapy was determined from relative changes in tumor volumes calculated from contrast-enhanced T1-weighted MRI, acquired before and a mean of 46 days after beginning therapy. ADCs and a diffusion index, RD, reflecting tissue viability based on water diffusion were calculated from DWMRIs. Pretreatment values of ADC and RD were found to correlate significantly with later tumor response/nonresponse (r = 0.76, P < .002 and r = 0.77, P < .001). This correlation implies that tumors with low pretreatment diffusion values, indicating high viability, will respond better to radiotherapy than tumors with high diffusion values, indicating necrosis. These results demonstrate the feasibility of using DWMRI for pretreatment prediction of response to therapy in patients with brain tumors undergoing radiotherapy.

Keywords: Diffusion-weighted MRI, high b-value, response to treatment, radiation, brain tumors

Introduction

Diffusion-weighted magnetic resonance imaging (DWMRI) enables noninvasive characterization of biological tissues based on their water diffusion properties. It is widely accepted that water in biological tissues is described in terms of a fast equilibration between two main components [1,2], a population of low-mobility/slow-diffusing water molecules (either bound to macromolecules or confined within the cell membrane), and a population of high-mobility water molecules, which is mostly extracellular. It has been shown in vitro that the diffusion of water molecules in the intracellular compartment is an order of magnitude smaller than in the extracellular space. The acquisition of high diffusion-weighted magnetic resonance imaging (HDWMRI), up to b = 4000 sec/mm2 in our case, amplifies the sensitivity to water diffusion properties and enables the separation of signals arising from the two populations [3,4]. Hence, DWMRI should be sensitive to several physiological and morphological characteristics of the tissues, which are associated with the diffusion of the low- and high-mobility water populations.

Previous studies in animal models [5–7] and human brain tumors [8,9] have demonstrated the ability of DWMRI to distinguish solid viable tumors from cystic and necrotic regions and have shown that water diffusion in tumors is correlated with tumor cellularity [10–13]. Several studies investigating DWMRI in tumors following various antitumor therapies [13–25] have found increased apparent diffusion coefficients (ADCs) of water that significantly preceded later tumor regression or decelerated growth, thus enabling early detection of tumor response. Most of these studies have used conventional DWMRI, up to a b-value of about 1000 sec/mm2.

Several MR methods have been suggested recently as having potential to predict tumor response to treatment. Contrast-enhanced MRI has been shown to be able to reveal distinct tumor patterns that can serve as a predictor of response to chemotherapy in human breast cancer [26]. P-31 MR spectroscopy was shown in a preliminary study to be a feasible method in predicting the response of head and neck cancers to radiation therapy [27]. This method, however, has a low sensitivity and is generally limited to large and preferably superficial tumors. Recent diffusion-weighted MR studies suggested that the initial ADC can serve as a predictive parameter for chemosensitivity of primary rat mammary tumors [7] and for chemoradiation/chemotherapy response [28,29] in patients with rectal cancer.

Our group has previously shown that the capacity of HDWMRI to obtain more information on the low-mobility water population enhances the sensitivity, especially at early stages, to relatively small treatment effects [30]. In a following study, we demonstrated the application of HDWMRI for pretreatment prediction of treatment outcome in an animal model [31] (Roth et al., Pretreatment Prediction and Early Monitoring of Tumor Response to Therapy Using High b-Value Diffusion-Weighted MRI, submitted for publication).

In this work, we studied the correlation between pretreatment water diffusion characteristics and later tumor response to treatment in patients with brain tumors undergoing radiation therapy. The results suggest that diffusion characteristics, related to the overall viability of the tumor, may be used for noninvasive prediction of treatment outcome prior to initiation of treatment.

Materials and Methods

Patients and Treatment

Twelve patients with 20 brain lesions were studied. Three patients had gliomas (grades III and IV) and nine patients had brain metastasis (six breast cancers, two lung cancers, and one prostate cancer). All patients received standard fractionated radiation therapy. The high-grade glioma patients received fractions of 2 Gy/day for a total of 54 to 60 Gy. The patients with metastases received either 3 Gy/day for a total of 30 Gy, or 2 Gy/day for a total of 40 Gy. All patients underwent MR scans before treatment and at regular intervals thereafter.

Equipment and Software

Data were acquired on a General Electric (Waukesha, WI) 0.5-T interventional MRI machine [Signa SP/i (special proceeding/interventional)] at the Chaim Sheba Medical Center using the Line Scan DWMRI (LSDI [32]) software. The standard GE head coil was used for data acquisition. Image analysis was performed using the Interactive Data Language (IDL), version 3.6.1, of Research Systems Inc. (Berkshire, UK) and InStat GraphPad (San Diego, CA) version 3.05 software packages.

Diffusion-Weighted MR Method

Diffusion-weighted images are usually obtained by acquiring conventional T2-weighted images with the addition of diffusion-weighting gradients that filter out the signal from high-mobility water molecules and sensitize the MR images to molecular diffusion/mobility [33]. Hence, regions of accumulated liquids or severe necrosis appear dark in DWMRI. At low diffusion-weighting values, most of the signals from the tissues are present in the images. At high diffusion-weighting values, most of the signals are filtered out, and the signal remaining in the image originates mostly from low-mobility molecules.

In this method, the normalized intensity of the water signal is given by:

I/I0=exp[−bADC] (1)

where I and I0 denote the signal intensities in the presence and absence of diffusion-weighting gradients, ADC is the molecular ADC, and b is the diffusion weighting factor, which is expressed in units of seconds per square millimeter.

By varying b (i.e., varying the intensity, duration, and/or separation time of the gradients), a diffusion curve can be obtained in which ln(I/I0) is plotted as a function of b (Figure 1).

Figure 1.

Figure 1

Open in a new tab

Mean signal intensity as a function of the diffusion factor b, for necrotic tissue, viable tissue, and noise level.

In order to quantify the diffusion characteristics of the tissue as reflected in the diffusion curve, we define a diffusion index, RD, which is the normalized summation over the curve:

RD=−∑j=0mln⁡(IjI0) (2)

The summation is over m data points of the diffusion curve. The division by I0 reduces the T2 effect. Therefore, small RD indicates slow signal decay as a function of b, implying more viable tissue. Regions with high RD correspond to more necrotic tissues.

Data Acquisition

Line scan DW images, gadolinium contrast-enhanced spin-echo T1-weighted MR images, and fast spin-echo T2-weighted MR images were used to monitor the patients before and at regular intervals following treatment. All images were acquired with 5-mm slices, two signal averages, and a 22 x 16.5-cm field of view. T2-weighted MR images were acquired with a 256 x 128 matrix, TR = 3000 milliseconds, and TE = 19/95 milliseconds. T1-weighted MR images were acquired with a 256 x 128 matrix, TR = 500 milliseconds, and TE = 14.5 milliseconds. DWMR images were acquired with a 128 x 64 matrix, b = 5 and 1000 sec/mm2, δ = 31 milliseconds, Δ = 51 milliseconds, TR = 2907 milliseconds, and TE = 105.2 milliseconds. Diffusion curves were calculated from additional DWMR images acquired at 14 b-values ranging from 15 to 4000 sec/mm2, δ = 53 milliseconds, Δ = 73 milliseconds, TR = 4964 milliseconds, TE = 149.8 milliseconds, and one signal average.

In normal white matter, the diffusion of the water molecules is anisotropic and data must be acquired in at least three orthogonal directions and then averaged in order to obtain isotropic diffusion coefficients. In this work, data were acquired using a monodirectional diffusion scheme (described in detail in Ref. [32]) with all three gradients turned on at the same time, due to long acquisition times at 0.5 T. This measurement should suffice due to the natural isotropy of cancer tumors and the reasonably reproducible head orientation.

The duration of a conventional DWMRI measurement at b = 5 and 1000 sec/mm2 was 20 seconds per average per slice. In this study, we used two averages; therefore, the scan lasted 40 seconds per slice. The number of slices varied from patient to patient, chosen in a manner that covered the entire tumor with an extra slice in each direction. The duration of a 14 b-value scan was 3 minutes and 47 seconds per slice. In order to save scan time, only three to four slices, chosen in the center of the tumor, were scanned. The overall diffusion scan time was, therefore, of the order of 20 minutes. In addition, precontrast and postcontrast T1-weighted and T2-weighted images were acquired. Thus, the overall scan time of an average exam reached 40 to 50 minutes.

Assessment of Tumor Response

Tumor volumes were calculated from the contrast-enhanced T1-weighted images. A region of interest (ROI) was defined over the entire apparent tumor in each slice and the number of pixels was counted. Tumor volumes in cubic centimeters were calculated prior to treatment and 46 days, on average, posttreatment. The relative change in tumor volume was defined as the ratio between the final volume and the initial volume.

Responding tumors were defined as tumors that decreased to 50% or less of their original volume. The rest were defined as stable/nonresponding tumors.

Analysis of DWMRI Data

ADCs were calculated on a pixel-by-pixel basis using the LSDI images acquired at b = 5 and 1000 sec/mm2 (Eq. (1)) to form ADC maps. Examples are shown in Figures 2D and 3D. Diffusion curves were obtained by plotting the logarithmic signal intensity in the diffusion-weighted image (I) normalized to the signal intensity of the image acquired at b = 15 mm2/sec (I0), as a function of b (Eq. (2)), on a pixel-by-pixel basis. RD was then calculated by summing over the 14 images in the curve to form RD maps (Figures 2C and 3C). ROIs were defined over the entire apparent tumor in the contrast-enhanced T1-weighted images and copied to the diffusion maps. Mean ADC and RD values were calculated for each tumor. Error bars were defined as the standard deviations of ROIs chosen in the contralateral normal tissue. The diffusion parameters were then correlated with response/nonresponse, as determined by the relative change in tumor volume measured, on average, 46 days after initiation of treatment. The error bars were incorporated in the calculation of the correlation parameters.

Figure 2.

Figure 2

Open in a new tab

Axial images of a patient with breast cancer metastasis tumor (arrow). The tumor appears dark in the diffusions maps, suggesting viable tissue. (A) Contrast-enhanced T1-weighted image; (B) T2-weighted image; (C) RD map; and (D) ADC map.

Figure 3.

Figure 3

Open in a new tab

Axial images of a patient with a GBM (arrow). The tumor appears bright in the diffusions maps, suggesting necrotic tissue. (A) Contrastenhanced T1-weighted image; (B) T2-weighted image; (C) RD map; and (D) ADC map.

Results

Ring Enhancement Lesions

Six lesions (five lung cancer metastases and one breast cancer metastasis) were defined as ring enhancement lesions (Table 1). They all appeared on T1-weighted MRI as cystic lesions surrounded by a thin enhancing rim. An example is shown in Figure 4. The ADC values of these lesions were significantly higher than all other lesions (Mann-Whitney, two tail, P < .0001) and they responded well to therapy. There was a clear correlation between the volume of the lesions and their mean ADC values, up to a value of 2.4 x 10-3 mm2/sec, where it plateaus (Figure 5). These lesions were not included in the study.

Table 1.

Diffusion and Volume Parameters of the Six Ring Enhancement Tumors Not Included in the Study.

Lesion Number Tumor Type Initial Tumor Volume (mm2) ADC x 10-3 (mm2/sec) (b = 1000 sec/mm2) Relative Change in Tumor Volume
1 Lung 6983 2.2 0.7
2 Lung 2635 2.4 0.4
3 Lung 1079 2.4 0.3
4 Lung 250 1.4 0
5 Lung 356 1.9 0
6 Breast 330 1.7 0.7
Open in a new tab

ADC values were calculated from images acquired at b = 5 and 1000 sec/mm2. Volumes were calculated from contrast-enhanced T1-weighted images.

Figure 4.

Figure 4

Open in a new tab

Axial images of a patient with small cell lung cancer metastasis. The tumor is defined as a ring enhancement tumor. (A) Contrast-enhanced T1-weighted image; (B) T2-weighted image; (C) RD map; and (D) ADC map.

Figure 5.

Figure 5

Open in a new tab

Plot of pretreatment mean ADC values as a function of lesion volume for ring enhancement tumors. There is a clear correlation up to a value of 2.4 x 10-3 mm2/sec, where the ADC value plateaus.

Pretreatment Prediction of Response

The tumors included in the study covered a wide range of tissue viability [RD: 14.4–25.4; ADC: (0.7–1.3) x 10-3 mm2/sec] and in tumor response (0.08–1.42). The pretreatment values of the diffusion parameters, ADC and RD, as well as the relative changes in tumor volumes 46 days, on average, after initiation of treatment are listed in Table 2 for all 14 tumors.

Table 2.

Diffusion and Volume Parameters of the 14 Brain Tumors Included the Study.

Lesion Number Tumor Type ADC x 10-3 (mm2/sec) (b = 1000 sec/mm2) RD Relative Change in Tumor Volume Time of Measurement of Volume Change
1 GBM 1.06 ± 0.09 20.0 ± 1.6 1.0 40
2 Breast 0.69 ± 0.05 14.4 ± 1.5 2.4 44
3 Breast 0.79 ± 0.08 15.7 ± 1.3 0.4 44
4 Glioma III 1.08 ± 0.08 21.7 ± 0.9 1.3 54
5 Glioma III 1.17 ± 0.05 25.0 ± 0.9 1.1 50
6 Breast 0.73 ± 0.13 15.2 ± 2.3 0.4 47
7 Prostate 0.78 ± 0.07 14.7 ± 0.8 1.4 54
8 Breast 1.08 ± 0.06 20.1 ± 1.3 0.6 51
9 Breast 1.33 ± 0.08 25.4 ± 1.6 0.8 51
10 Breast 1.24 ± 0.08 24.9 ± 2.5 1.4 51
11 Breast 1.19 ± 0.11 20.5 ± 2.0 1.0 51
12 Breast 0.99 ± 0.07 19.8 ± 1.6 0.5 30
13 Breast 1.24 ± 0.06 21.9 ± 1.4 0.9 30
14 Breast 0.92 ± 0.07 17.8 ± 1.4 0.1 52
Open in a new tab

ADC values were calculated from images acquired at b = 5 and 1000 sec/mm2. RD values were calculated from images acquired at 14 b-values ranging from b = 15 to 4000 sec/mm2. The relative change in tumor volume was defined as the ratio between the final tumor volume (measured 46 days, on average, after treatment initiation) and the pretreatment volume as determined from contrast-enhanced T1-weighted MR images.

Viable tissues are associated with low water mobility; therefore, viable tumors appear dark on ADC and RD maps, whereas necrotic tumors appear bright. Example of diffusion maps of tumors with different water diffusion characteristics are shown in Figures 2 and 3.

The potential of using pretreatment diffusion characteristics to predict tumor response to therapy was studied by correlating the tumor diffusion parameters, ADC and RD (reflecting tissue viability), measured prior to initiation of treatment, with the relative change in tumor volume, measured, on average, 46 days after initiation of treatment.

The positive correlation between pretreatment values of ADC and later tumor response is considered very significant (P < .002, r = 0.76, Pearson correlation), as presented in Figure 6.

Figure 6.

Figure 6

Open in a new tab

Plot of tumor pretreatment mean ADC values as a function of the relative change in tumor volume (ratio between final volume and initial volume) 46 days, on average, after initiation of treatment. The significant correlation demonstrates the potential of this method for pretreatment prediction of response to therapy.

The positive correlation between pretreatment values of RD (calculated from HDWMRI) and later tumor response is considered very significant (P < .001, r = 0.77, Pearson correlation), as presented in Figure 7.

Figure 7.

Figure 7

Open in a new tab

Plot of tumors pretreatment mean RD values as a function of the relative change in tumor volume (ratio between final volume and initial volume) 46 days, on average, after initiation of treatment. The significant correlation demonstrates the potential of this method for pretreatment prediction of response to therapy.

In order to determine the contribution of the high b-value data to the prediction of response, RD was recalculated using b-values up to 940 sec/mm2 (conventional values). The positive correlation between the pretreatment values of this parameter and later tumor response is similar to the correlation found for the ADC (P < .002, r = 0.76, Pearson correlation).

A comparison between the diffusion parameters of responding and stable/nonresponding tumors using a two-tailed P value Mann-Whitney test resulted in P < .0007 for both RD and ADC, which is considered extremely significant. The 95% confidence ranges were 14.1 to 18.5 (responding) and 20.5 to 24.4 (stable/nonresponding) for RD and (0.7–0.9) x 10-3 mm2/sec (responding) and (1.1–1.3) x 10-3 mm2/sec (stable/nonresponding) for ADC.

The positive correlation between the pretreatment values of the two diffusion parameters (ADC and RD) is considered extremely significant (P < .0001, r = 0.95, Pearson correlation).

Discussion

The metastases defined as ring enhancement lesions were notincludedinthestudy. Theiruniqueradiologicalappearance and high response to treatment are consistent with previous studies [26]. These lesions had no massive tumor but the enhancing rim surrounding the cystic content. Because the partial volume of the viable rim decreases as the total lesion volume increases, it is reasonable to assume that their diffusion characteristics are dominated by their volumes. This effect is reflected in the correlation observed between the lesion volume and the ADC value. Therefore, the diffusion characteristics of these lesions do not reflect their overall viability; thus, they were not included in the study.

The tumors included in the study are spread over a wide range of tissue viabilities and tumor responses, enabling us to study the correlation between pretreatment values of the diffusion parameters and treatment outcomes over a wide range of lesions.

Previous studies have shown that regions appearing dark in DWMR images are associated with more necrotic tissues [10–13,15,23]. Hence, high ADC values and diffusion curves that decay quickly with b (high RD values) are typical of low cellularity and necrotic tissues, whereas low ADC values and curves that decay slowly with b (low RD values) may indicate more viable tissues with higher cellularity. Therefore, the mean ADC and RD values calculated for each tumor (by choosing ROIs over the entire tumor and averaging over these ROIs) reflect tumor cellularity/viability. The positive correlation between the values of the pretreatment diffusion parameters and later response/nonresponse to therapy implies that cellular/viable tumors respond better to radiation therapy than necrotic tumors.

A possible explanation may be related to the fact that the viable cancer cells adjacent to necrotic regions may experience hypoxic conditions and therefore may be less sensitive to radiation therapy [34,35].

To the best knowledge of the authors, the only published clinical studies [28,29] that show a correlation between pretreatment ADCs and response to therapy consist of patients with rectal carcinoma treated with chemoradiation. The correlations presented in these papers are consistent with the data presented in our study with brain tumors in the sense that low ADC tumors respond better to treatment than high ADC tumors.

In two recent studies [30,31] we have demonstrated the additional sensitivity gained by using high b-value DWMRI, both for pretreatment prediction of response to therapy (animal study and chemotherapy) and for early posttreatment detection of response to therapy (clinical study, brain tumors, and radiosurgery/radiation therapy).

The current study demonstrates a similar predictive power for the parameters calculated from conventional and high b-value data. The high correlation between the ADC values and the RD values suggests that in this case, most of the information is contained already in the ADC data (i.e., in the conventional DWMRIs). Further studies on higher-field MR systems should be carried out in order to determine the value of HDWMRI in the pretreatment prediction of response to therapy.

This study was performed using a relatively low magnetic field of 0.5 T and relatively weak diffusion gradients of 1 G/cm. The data were acquired using the LSDI sequence. Although this approach is less sensitive to motion and susceptibility artifacts, it yields relatively low signal-to-noise data per unit scan time and requires longer scan times. This combination results in low signal-to-noise ratio (SNR), especially in the high diffusion-weighted data. In order to compensate for the low SNR, we acquired a large number [14] of b-values, resulting in better SNR for the calculated diffusion index (Eq. (2)) but relatively long scan times. In addition, we defined a diffusion index, RD, calculated by summing over the normalized signal intensities of the diffusion curves. In a previous publication [30], we have used a similar diffusion index, R, calculated from the ADCs and volume fractions of the two water populations, obtained by fitting the HDWMRI data to a biexponential function. In the current study, R and RD were found to be highly correlated, but due to the higher sensitivity of R to low SNR, the predictive power of RD was greater (data not shown). Once using a higher field magnet with better SNR, R may become advantageous over RD due to the separation of the two water components (obtained using the biexponential fit) and because it is protocol-independent.

With single-shot diffusion-weighted echo-planar imaging (EPI) [36], which is available on most new MR machines, data of several slices could be obtained without increasing scan times. This type of sequence would be more appropriate for routine clinical use rather than the methodology presented here due to its availability and short acquisition times, although some susceptibility artifacts may be formed in boundary areas. A combination of short scan times and reduced sensitivity to susceptibility variations is offered by novel diffusion imaging techniques, such as slab scan diffusion imaging [37] or single-shot EPI with sensitivity-encoded single shot (SENSE) [38].

In summary, this study demonstrates the feasibility of using conventional and high b-value DWMRI as a noninvasive tool for pretreatment prediction of response to radiation therapy in patients with malignant brain tumors. We are currently extending this study to higher magnetic fields and stronger gradient intensities in order to establish the application of this method for clinical use, thus enabling optimization of the management of malignancy.

Acknowledgements

We thank Asher Gotsmann for many fruitful discussions. We thank Cipora Podhorzer and Avishai Goldblat for their dedicated help in scanning the patients.

Footnotes

1

This research was supported by the Israel Science Foundation, the Israel Cancer Research Fund, the Adams Super Center for Brain Studies at Tel-Aviv University, the Izmel Program of the Israel Ministry of Industry and Commerce, and NIH R01 NS39335.

References

  • 1.Niendorf T, Dijkhuizen RM, Norris DG, van Lookeren Campagne M, Nicolay K. Biexponential diffusion attenuation in various states of brain tissue: implications for diffusion-weighted imaging. Magn Reson Med. 1996;36:847–857. doi: 10.1002/mrm.1910360607. [DOI] [PubMed] [Google Scholar]
  • 2.Mulkern RV, Gudbjartsson H, Westin CF, Zengingonul HP, Gartner W, Guttmann CR, Robertson R, Kyriakos W, Schwartz R, Holtzman D, Jolesz FA, Maier SE. Multi-component apparent diffusion coefficients in human brain. NMR Biomed. 1999;12:51–62. doi: 10.1002/(sici)1099-1492(199902)12:1<51::aid-nbm546>3.0.co;2-e. [DOI] [PubMed] [Google Scholar]
  • 3.van Zijl PCM, Moonen CTW, Faustino PJ, Pekar J, Kaplan O, Cohen JS. Complete separation of intracellular and extracellular information in NMR spectra by diffusion weighting. Proc Natl Acad Sci USA. 1991;88:3228–3232. doi: 10.1073/pnas.88.8.3228. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Pilatus U, Shim H, Artemov D, Davis D, van Zijl PC, Glickson JD. Intracellular volume and apparent diffusion constants of perfused cancer cell cultures, as measured by NMR. Magn Reson Med. 1997;37:825–832. doi: 10.1002/mrm.1910370605. [DOI] [PubMed] [Google Scholar]
  • 5.Maier CF, Paran Y, Bendel P, Rutt BK, Degani H. Quantitative diffusion imaging in implanted human breast tumors. Magn Res Med. 1997;37:576–581. doi: 10.1002/mrm.1910370417. [DOI] [PubMed] [Google Scholar]
  • 6.Eis M, Els T, Hoehn-Berlage M. High resolution quantitative relaxation and diffusion MRI of three different experimental brain tumors in rat. Magn Reson Med. 1995;34:835–844. doi: 10.1002/mrm.1910340608. [DOI] [PubMed] [Google Scholar]
  • 7.Lemaire L, Howe FA, Rodrigues LM, Griffiths JR. Assessment of induced rat mammary tumor response to chemotherapy using the apparent diffusion coefficient of tissue water as determined by diffusion-weighted 1H-NMR spectroscopy in vivo. Magn Res Materials in Phys Biol Med. 1999;8:20–26. doi: 10.1007/BF02590631. [DOI] [PubMed] [Google Scholar]
  • 8.Tsuruda JS, Chew WM, Moseley ME, Norman D. Diffusion-weighted MR imaging of the brain: value of differentiating between extraaxial cysts and epidermoid tumors. Am J Neuroradiol. 1990;11:925–931. [PMC free article] [PubMed] [Google Scholar]
  • 9.Brunberg JA, Chenevert TL, McKeever PE, Ross DA, Junck LR, Muraszko KM, Dauser R, Pipe JG, Betley AT. In vivo MR determination of water diffusion coefficients and diffusion anisotropy: correlation with structural alteration in gliomas of cerebral hemispheres. Am J Neuroradiol. 1995;16:361–371. [PMC free article] [PubMed] [Google Scholar]
  • 10.Sugahara T, Korogi Y, Kochi M, Ikushima I, Shigematsu Y, Hirai T, Okuda T, Liang L, Ge Y, Komohara Y, Ushio Y, Takahashi M. Usefulness of diffusion-weighted MRI with echo-planar technique in the evaluation of cellularity in gliomas. J Magn Reson Imaging. 1999;9:53–60. doi: 10.1002/(sici)1522-2586(199901)9:1<53::aid-jmri7>3.0.co;2-2. [DOI] [PubMed] [Google Scholar]
  • 11.Lyng H, Haraldseth O, Rofstad EK. Measurement of cell density and necrotic fraction in human melanoma xenografts by diffusion-weighted magnetic resonance imaging. Magn Reson Med. 2000;43:828–836. doi: 10.1002/1522-2594(200006)43:6<828::aid-mrm8>3.0.co;2-p. [DOI] [PubMed] [Google Scholar]
  • 12.Gupta RK, Cloughesy TF, Sinha U, Garakian J, Lazareff J, Rubino G, Rubino L, Becker DP, Vinters HV, Alger JR. Relationships between choline magnetic resonance spectroscopy, apparent diffusion coefficient and quantitative histopathology in human glioma. J Neuro-Oncol. 2000;50(3):215–226. doi: 10.1023/a:1006431120031. [DOI] [PubMed] [Google Scholar]
  • 13.Chenevert TL, McKeever PE, Ross BD. Monitoring of early response of experimental brain tumors to therapy using diffusion MRI. Clin Cancer Res. 1997;3:1457–1466. [PubMed] [Google Scholar]
  • 14.Kauppinen RA. Monitoring cytotoxic tumor treatment response by diffusion magnetic resonance imaging and proton spectroscopy. NMR Biomed. 2002;15:6–17. doi: 10.1002/nbm.742. [DOI] [PubMed] [Google Scholar]
  • 15.Chenevert TL, Stegman LD, Taylor JMG, Robertson PL, Greenberg HS, Rehemtulla A, Ross BD. Diffusion magnetic resonance imaging: an early surrogate marker of therapeutic efficacy in brain tumors. J Natl Cancer Inst. 2000;92:2029–2036. doi: 10.1093/jnci/92.24.2029. [DOI] [PubMed] [Google Scholar]
  • 16.Hakumaki JM, Poptani H, Puumalainen AM, Loimas S, Paljarvi LA, Yla-Herttuala S, Kauppinen RA. Quantitative 1H nuclear magnetic resonance diffusion spectroscopy of BT4C rat glioma during thymidine kinase-mediated gene therapy in vivo: identification of apoptotic response. Cancer Res. 1998;58:3791–3799. [PubMed] [Google Scholar]
  • 17.Poptani H, Puumalainen AM, Grohn OH, Loimas S, Kainulainen R, Yla-Herttuala S, Kauppinen RA. Monitoring thymidine kinase and ganciclovir-induced changes in rat malignant glioma in vivo by nuclear magnetic resonance imaging. Cancer Gene Ther. 1998;5:101–109. [PubMed] [Google Scholar]
  • 18.Stegman LD, Rehemtulla A, Hamstra DA, Rice DJ, Jonas SJ, Stout KL, Chenevert TL, Ross BD. Diffusion MRI detects early events in the response of a glioma model to the yeast cytosine deaminase gene therapy strategy. Gene Ther. 2000;7:1005–1010. doi: 10.1038/sj.gt.3301199. [DOI] [PubMed] [Google Scholar]
  • 19.Galons JP, Altbach MI, Paine-Murrieta GD, Taylor CW, Gillies RJ. Early increases in breast tumor xenograft water mobility in response to paclitaxel therapy detected by non-invasive diffusion magnetic resonance imaging. Neoplasia. 1999;1:113–117. doi: 10.1038/sj.neo.7900009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Winsborrow BG, Grondey H, Savoie H, Fyfe CA, Dolphin D. Magnetic resonance imaging evaluation of photodynamic therapy-induced hemorrhagic necrosis in the murine M1 tumor model. Photochem Photobiol. 1997;66:847–852. doi: 10.1111/j.1751-1097.1997.tb03236.x. [DOI] [PubMed] [Google Scholar]
  • 21.Chinnaiyan AM, Prasad U, Shankar S, Hamstra DA, Shanaiah M, Chenevert TL, Ross BD, Rehemtulla A. Combined effect of tumor necrosis factor related apoptosis inducing ligand and ionizing radiation in breast cancer therapy. Proc Natl Acad Sci. 2000;97:1754–1759. doi: 10.1073/pnas.030545097. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Zhao M, Pipe JG, Bonnet J, Evelhoch JL. Early detection of treatment response by diffusion-weighted 1H-NMR spectroscopy in a murine tumor in vivo. Br J Cancer. 1996;73:61–64. doi: 10.1038/bjc.1996.11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Geschwind JF, Artemov D, Abraham S, Omdal D, Arepally A, Venbrux A, Lund G. Chemoembolization of liver tumor in a rabbit model: assessment of tumor cell death with diffusion-weighted MR imaging and histologic analysis. J Vasc Interv Radiol. 2000;11:1245–1255. doi: 10.1016/s1051-0443(07)61299-8. [DOI] [PubMed] [Google Scholar]
  • 24.Jennings D, Hatton BN, Guo J, Galons JP, Trouard TP, Raghunand N, Marshall J, Gillies RJ. Early response of prostate carcinoma xenografts to docetaxel chemotherapy monitored with diffusion MRI. Neoplasia. 2002;4:255–262. doi: 10.1038/sj.neo.7900225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Mardor Y, Roth Y, Lidar Z, Jonas T, Pfeffer R, Maier SE, Faibel M, Nass D, Hadani M, Orenstein A, Cohen JS, Ram Z. Monitoring response to convection-enhanced Taxol delivery in brain tumor patients using diffusion-weighted magnetic resonance imaging. Cancer Res. 2001;61:4971–4973. [PubMed] [Google Scholar]
  • 26.Esserman LJ, Kaplan E, Partridge S, Tripathy D, Rugo H, Park J, Hwang S, Kuerer H, Sudilovsky D, Lu Y, Hylton N. MRI phenotype is associated with response to doxorubicin and cyclophosphamide neoadjuvant chemotherapy in stage III breast cancer. Ann Surg Oncol. 2001;8:549–559. doi: 10.1007/s10434-001-0549-8. [DOI] [PubMed] [Google Scholar]
  • 27.Shukla-Dave A, Poptani H, Loevner LA, Mancuso A, Serrai H, Rosenthal M, Kilger A, Nelson DS, Zakian K, Arias-Mendoza F, Rijpkema M, Koutcher JA, Brown TR, Heershcap A, Glickson JD. Prediction of treatment response of head and neck cancers with P-31 MR spectroscopy from pretreatment relative phosphomonoester levels. Acad Radiol. 2002;9:688–694. doi: 10.1016/s1076-6332(03)80314-8. [DOI] [PubMed] [Google Scholar]
  • 28.Dzik-Jurasz A, Domenig C, George M, Wolber J, Pedhani A, Brown G, Doran S. Diffusion MRI for prediction of response of rectal cancer to chemoradiation. Lancet. 2002;360:307–308. doi: 10.1016/S0140-6736(02)09520-X. [DOI] [PubMed] [Google Scholar]
  • 29.Hein PA, Kremser C, Judmaier W, Griebel J, Rudisch A, Pfeiffer KP, Hug EB, Lukas P, De Vries AF. Diffusion-weighted MRI—a new parameter for advanced rectal carcinoma? RoFo Fortschr Geb Rontgenstr neuen Bildgeb Verfahr. 2003;175:381–386. doi: 10.1055/s-2003-37836. [DOI] [PubMed] [Google Scholar]
  • 30.Mardor Y, Pfeffer R, Spiegelmann R, Roth Y, Maier SE, Nissim O, Berger R, Glicksman A, Baram J, Orenstein A, Cohen JS, Tichler T. Early detection of response to radiation therapy in patients with brain malignancies using standard and high b-value diffusion-weighted MRI. J Clinic Oncol. 2003;21:1094–1100. doi: 10.1200/JCO.2003.05.069. [DOI] [PubMed] [Google Scholar]
  • 31.Roth Y, Orenstein A, Ruiz-Cabello J, Maier SE, Cohen JS, Mardor Y. Pre-treatment prediction and early monitoring of tumor response to therapy using diffusion-weighted MRI. Proc Int Soc Magn Reson Med. 2002;10:2002. [Google Scholar]
  • 32.Gudbjartsson H, Maier SE, Mulkern RV, Morocz IA, Patz S, Jolesz FA. Line scan diffusion imaging. Magn Reson Med. 1996;36:509–519. doi: 10.1002/mrm.1910360403. [DOI] [PubMed] [Google Scholar]
  • 33.Stejskal EO, Tanner JE. Spin diffusion: spin echoes in the presence of a time-dependent field gradient. J Chem Phys. 1965;42:288–292. [Google Scholar]
  • 34.Thomlinson RH, Gray LH. The histological structure of some human lung cancers and the possible implications for radiotherapy. Br J Cancer. 1955;9:539–549. doi: 10.1038/bjc.1955.55. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Coleman CN. Hypoxia in tumors: a paradigm for the approach to biochemical and physiologic heterogeneity. J Natl Cancer Inst. 1988;80:310–317. doi: 10.1093/jnci/80.5.310. [DOI] [PubMed] [Google Scholar]
  • 36.Turner R, Le Bihan F, Maier J, Vavrec R, Hedges LK, Pekar J. Echo-planar imaging of intravoxel incoherent motions. Radiology. 1990;177:407–414. doi: 10.1148/radiology.177.2.2217777. [DOI] [PubMed] [Google Scholar]
  • 37.Maier SE. Slab scan diffusion imaging. Magn Reson Med. 2001;46:1136–1143. doi: 10.1002/mrm.1310. [DOI] [PubMed] [Google Scholar]
  • 38.Bammer R, Auer M, Keeling SL, Augustin M, Stables LA, Prokesch RW, Stollberger R, Moseley ME, Fazekas F. Diffusion tensor imaging using single-shot SENSE-EPI. Magn Reson Med. 2002;48:128–136. doi: 10.1002/mrm.10184. [DOI] [PubMed] [Google Scholar]

Articles from Neoplasia (New York, N.Y.) are provided here courtesy of Neoplasia Press

RESOURCES