Detection of prostate cancer: Utility of diffusion-weighted
MR imaging and 3D MR spectroscopic imaging
Ghada K. Gouhar
a,*
, Tamer F. Taha
a, Mohamed N. Allam
ba
Radiodiagnosis Department, Faculty of Medicine, Zagazig University, Sharkia, Egypt
b
Urosurgery Department, Faculty of Medicine, Zagazig University, Sharkia, Egypt
Received 8 July 2010; accepted 18 August 2010 Available online 20 October 2010
KEYWORDS Prostate; Diffusion; Spectroscopy; MR; Cancer
Abstract Purpose: To prospectively compare the mean ADC generated from DWI, the mean line + creatine/citrate ratio generated from 3D MRS and the combined mean ADC and mean cho-line + creatine/citrate ratio in the detection of prostate cancer, and to correlate between the choline + creatine/citrate ratio and the aggressiveness of malignancy determined by Gleason score, with histopathological examination of the excised gland as the reference standard.
Patients and methods: Forty-six patients with biopsy-proved cancer underwent pre-operative MRI at 1.5 T. AxialT1, axial, coronal and sagittalT2-weighted, diffusion-weighted and 3D MRS using a point-resolved spectroscopic sequence (PRESS) were acquired. The mean ADC, mean cho-line + creatine/citrate ratio and combined parameters for malignant lesions are correlated with the pathological results. For each malignant lesion choline + creatine/citrate ratio was correlated with the aggressiveness of malignancy determined by Gleason score. Receiver operating character-istic (ROC) curves were used to determine sensitivity, and specificity of the studied parameters, and Kappa measures of agreement were calculated for prostate cancer detection.
Results: The mean ADC for tumor tissue was 1.0 ± 0.22·103mm2/s (mean ± SD), and was sig-nificantly lower than that for non-tumor tissue 1.44 ± 0.28·103mm2/s (p< 0.001). For MRS study the mean (choline + creatine)/citrate ratio in tumor tissue was 1.98 ± 1.0, and was signifi-cantly higher than that for non-tumor tissue, 0.72 ± 0.39 (p< 0.001). By combining both ADC values and (choline + creatine)/citrate ratio for differentiating malignant from non-malignant
* Corresponding author. Tel.: +20 0124038955.
E-mail addresses:[email protected](G.K. Gouhar),doctor [email protected] (T.F. Taha), mohamednageb1968@yahoo. com(M.N. Allam).
0378-603X 2010 Egyptian Society of Radiology and Nuclear Medicine. Production and hosting by Elsevier B.V.
Peer review under responsibility of Egyptian Society of Radiology and Nuclear Medicine.
doi:10.1016/j.ejrnm.2010.08.004
Production and hosting by Elsevier
Egyptian Society of Radiology and Nuclear Medicine
The Egyptian Journal of Radiology and Nuclear Medicine
www.elsevier.com/locate/ejrnm
www.sciencedirect.com
Open access under CC BY-NC-ND license.
tissues a receiver operating characteristic analysis (ROC) curve showed Area under curve (AUC = 0.93) and was significantly higher than either (choline + creatine)/citrate ratio alone (AUC = 0.86) (p< 0.001) or ADC value alone (AUC = 0.89) (p< 0.001). There is an increasing (choline + creatine)/citrate ratio with increasing Gleason score, however, there is overlap between groups. A greater sensitivity of MRS for tumor detection 85% and 92% was present for tumors with Gleason score 4 + 3 andP4 + 4, respectively, while for tumors with Gleason score 3 + 3 the sensitivity was 63%.
Conclusion: The combination of ADC and (choline + creatine)/citrate ratio is better than each parameter alone in differentiating between tumor and non-tumor prostatic tissue, also MR spectro-scopic imaging findings of prostate tumor (Cho + Cr)/Cit ratio correlate with pathologic Gleason score. The combined parameters offer a promising non-invasive method for the diagnostic workup of prostate cancer.
2010 Egyptian Society of Radiology and Nuclear Medicine. Production and hosting by Elsevier B.V. All rights reserved.
1. Introduction
Despite the recent development in the diagnosis and the treat-ment of prostatic cancer, it continues to be the most common cancer and the third leading cause of death in men(1). The choice of treatment depends on a number of clinical parame-ters and clinical nomograms such as the patient’s age at diag-nosis, the stage and the aggressiveness of the tumor(2), with the highest priority being the differentiation between indolent and aggressive disease, which is based mainly on Gleason grade determined by TRUS-guided biopsy(2). However, US is of low accuracy in detection and localization of prostate can-cer, so a random instead of targeted biopsy is performed.
Random sampling has several disadvantages such as miss-ing of the cancer located outside the routine biopsy site and unnecessary sampling of normal prostatic tissue. Moreover, the site of previously negative biopsy cannot be determined in patient with continuously high Prostatic Surface Antigen (PSA) and the biopsy is repeated(3). Thus a non-invasive tech-nique that allows accurate detection and assessment of the de-gree of aggressiveness of prostatic cancer is needed to make a substantial contribution to the decision-making process for proper treatment selection(2).
The use of MRI as a non-invasive tool for the evaluation and management of prostatic cancer has grown steadily in the past decades(4). It allows functional assessment by differ-ent techniques such as diffusion-weighted images (DWI) and MR spectroscopy (MRS)(5). At 2WI prostate cancer appears as a focus of low-signal intensity relative to bright normal peripheral zone. However, prostate cancer is difficult to detect when it is in the central zone due to high-signal intensity from benign prostatic hypertrophy(6). The accuracy ofT2-weighted MRI in tumor localization is 67–77%(7).
Diffusion-weighted MRI (DWI), a non-invasive technique in which molecular motion of water is measured in biological tissues, is now used in the detection of prostate cancer as an adjunct to T2WI (8). The apparent diffusion coefficient (ADC) calculated from DWI in prostate cancer showed that the mean ADC for malignant prostate is lower than the mean ADC in the non-malignant prostatic tissue(9–11).
MR spectroscopic imaging has shown to provide an incre-mental value to MRI regarding tumor detection and localiza-tion(12), by providing information on relative concentration of metabolites such as citrate (Cit), choline (Cho) and creatine (Cr) within a voxel(13).
In a previous study (14) a positive correlation between ADC values, metabolites ratio (Cit to Cho and Cr) and PSA was observed, and there was a direct relationship between the reduction in (Cit) level and the malignant changes in pros-tate(15).
Thus the purpose of this study is to evaluate the role of the mean ADC with DWMRI and the mean metabolic ratio with
1H MR spectroscopic imaging in the detection of prostatic
cancer as well as the degree of aggressiveness of the lesion with post-operative histopathological examination as the reference standard.
2. Patients and methods
2.1. Patient characteristics
Forty-six men (mean age 61.5 ± 12.2 years) were included in this prospective study who were scheduled for radical retropu-bic prostatectomy or transurethral resection of the prostate due to biopsy-proved cancer, within a mean 23 ± 19 days. A minimum delay of 6 weeks was required between biopsy and MR imaging and MR spectroscopic imaging to minimize biopsy artifacts. Patient exclusion criteria were previous hor-monal therapy, positive lymphadenectomy results, and contra-indications to MR imaging (e.g., cardiac pacemakers, intracranial clips). Informed consent was obtained from all patients.
2.2. MR imaging protocol
MR imaging was performed with a 1.5-T Philips Achieva sys-tem by using a pelvic phased-array coil with the patient in su-pine position. MR imaging examination included conventional, DW imaging and MR spectroscopy imaging.
2.2.1. Conventional images
After localizer images, conventional images were obtained including transverse T1-weighted spin-echo MR images from the aortic bifurcation to the symphysis pubis with the follow-ing parameters: repetition time ms/echo time ms, 520/15 sec-tion thickness, 4 mm; intersecsec-tion gap, 1 mm; field of view, 20 cm; matrix, 256·192 and flip angle 90 deg. Thin-section high-spatial-resolution transverse T2-weighted fast spin-echo MR imagesof the prostate and seminal vesicles were obtained with the following parameters: 3500/90, section thickness,
4 mm; intersection gap, 1 mm; field of view, 25 cm; matrix, 256·192 and flip angle 90 deg. Coronal and sagittal T2-weighted fast spine echo MR imagesof the prostate and seminal vesicles were obtained with the following parameters: 4100– 4500/90, section thickness, 3 mm; intersection gap, 1 mm; field of view, 38 cm; matrix, 256·192 and flip angle 90 deg.
2.2.2. Diffusion-weighted imaging
Diffusion-weighted images were obtained by using single-shot spin-echo echo-planar imaging with a pair of rectangular gra-dient pulses along three orthogonal axes. The imaging param-eters were as follows: 2800/74; field of view, 38 cm2; section thickness, 3 mm; intersection gap, 1 mm. Images were zero-filled to a 256·256 matrix. The orientation and location of these images were prescribed identically to the transverse
T2-weighted prostate images. The b values were 0 and 1000 s/mm2 and flip angle 90 deg. Apparent diffusion coeffi-cient (ADC) maps were automatically calculated by MRI ma-chine software and included in the sequence for the interpretation of DW imaging findings.
2.2.3. Three-dimensional1H MR spectroscopy
2.2.3.1. Data acquisition.Three-dimensional (3D) MR spectro-scopic imaging of the entire prostate was performed by using a section-selected box drawn closely around the prostate and a point-resolved spectroscopic sequence (PRESS). The 3D MR spectroscopic imaging parameters were as follows: 1500/130; number of signals acquired, one; spectral width, 1000 Hz; number of points, 512; field of view, 15· 5.5·5.5 cm3. The MR spectroscopic imaging acquisition voxel volume was 326.1 mm3. Magnetic field homogeneity was optimized for the selected volume by using an automated shimming algo-rithm provided by the manufacturer.
2.2.3.2. Spectroscopic data analysis. Spectral data were pro-cessed by using the manufacturer’s post-processing software package. Processing included zero-filling of the raw data in the superoinferior direction and reconstruction with a four-dimensional Fourier transformation to yield1H MR
spectro-scopic images. Estimates of the areas under the resonances of the metabolite peaks were obtained by integrating a region centered at each peak. The metabolites studied were those res-onating at approximately 3.22 parts per million (ppm) (i.e., Cho),3.02 ppm (i.e., Cr), and 2.62 ppm (i.e., Cit). The normal-ized peak areas for Cho, Cr, and Cit for each tumor voxel were tabulated and used to calculate the following ratios: (Cho + Cr)/Cit, Cho/Cr, and Cho/Cit. In our study we used the metabolic ratio maps of (choline + creatine)/citrate for data management. For all voxels considered suspicious, (Cho + Cr)/Cit was recorded for statistical analysis. The MR spectroscopic imaging assignment of suspicious voxels
was performed by using only (Cho + Cr)/Cit and without ref-erence to the MR imaging findings or knowledge of the results of pathologic evaluation.
2.3. Pathologic evaluation
Prostatectomy specimen was prepared and fixed in 10% for-malin. After paraffin embedding, microslices were placed on glass slides and stained with hematoxylin and eosin. At patho-logic analysis, grading was assigned to the lesion in the speci-men according to the Gleason score.
2.4. Data and statistical analysis
Statistical analysis was performed by using statistical software SPSS version 10. (Cho + Cr)/Cit value for each suspicious voxel within the lesion was tabulated. The mean values of (Cho + Cr)/Cit for each lesion and the total number of suspi-cious voxels in the lesion were recorded for statistical analysis. We used the mean ADC and the mean (choline + creatine)/ citrate ratio for each multivoxel ROI. The mean values of the benign ROIs were compared with the mean values of the malignant ROIs by using pairedttests.
Diffusion-weighted information and 3D MR spectroscopic imaging information were combined by using generalized esti-mating equations with an independent working correlation matrix to account for the correlated data. To obtain predicted probabilities of an ROI being cancerous, the mean ADC and the mean (choline + creatine)/citrate ratio for the ROI were entered into a logistic regression model. On the basis of the re-sults of this model (Table 1), the following equation was ob-tained for the probability of an ROI being cancerous:
6:15þADC6:07MRS1:864
The regression model yielded estimated regression coeffi-cients that weighted the information from these two variables in an optimal way for combining them. The regression analysis also determined whether each variable was significantly associ-ated with the probability of an ROI being cancerous, after adjusting for the other information. Receiver operating char-acteristic curves and the corresponding areas under the recei-ver operating characteristic curves (AUCs) were estimated non-parametrically for the detection of cancer by using mean ADC, mean (choline + creatine)/citrate ratio, and combined mean ADC and mean (choline + creatine)/citrate values for each ROI. In all statistical methods, ap value of less than 0.05 was considered to indicate a significant difference. Be-cause smaller ADC values are associated with cancer, for the receiver operating characteristic analysis we transformed the ADC value by multiplying it by1. The sensitivity and spec-ificity of mean ADC, mean (choline + creatine)/citrate ratio,
Table 1 Regression model for combining ADC and (choline + creatine)/citrate ratio.
B SE Wald df Sig. Exp(B) 95.0% CI for exp(B)
Lower Upper
Step 1a ADC 6.075 1.336 20.684 1 0.000 435.001 31.727 5964.106
MRS 1.864 0.554 11.312 1 0.001 0.155 0.052 0.459
Constant 6.154 1.782 11.923 1 0.001 0.002
a
and combined mean ADC and mean (choline + creatine)/cit-rate were measured and Kappa measures of agreement were calculated.
MR spectroscopic imaging sensitivity for cancer detection was analyzed prospectively by comparing the MR spectro-scopic imaging voxels, which had been designated as suspi-cious for cancer, with the pathologic findings. Sensitivity was calculated for all pathologic lesions as a group and for individual Gleason score groups (3 + 3, 3 + 4, 4 + 3, and
P4 + 4). Then MR spectroscopic imaging analysis tested the hypothesis that MR spectroscopic imaging metabolite ra-tios in true-positive lesions were related to pathologic Glea-son scores. The relationship of GleaGlea-son scores for the individual lesions to the MR spectroscopic imaging (Cho + Cr)/Cit value was assessed. In each lesion, the MR spectroscopic imaging mean ratio was compared with the Gleason score for that lesion.
3. Results
A total of 46 patients underwent conventional MRI, DWI, and MRS for prostatectomy. Twelve patients had a single lesion, nineteen patients had two lesions, nine patients had tree lesions and seven patients had four lesions.
There was an overlap between tumor and non-tumor mean ADC values and the mean (choline + creatine)/citrate ratio (Fig. 1A and B).
The mean ADC for tumor tissue was 1.0 ± 0.22·103
mm2/s (mean ± SD), and was significantly lower than that for non-tumor tissue 1.44 ± 0.28·103mm2/s (p< 0.001)
(Table 2).
All prostatic cancer displayed high-signal intensity on DWI and low-signal intensity on ADC map.
For MRS study the mean (choline + creatine)/citrate ratio in tumor tissue was 1.98 ± 1.0, and was significantly higher than that for non-tumor tissue 0.72 ± 0.39 (p< 0.001) (Table
2).
By combining both ADC values and (choline + creatine)/ citrate ratio for differentiating malignant from non-malignant tissues a receiver operating characteristic analysis (ROC) curve showed Area under curve (AUC) = 0.93, 95% CI: 0.88–0.47 and was significantly higher than either (choline + creatine)/ citrate ratio alone where AUC = 0.86, 95% CI: 0.80–0.92 (p< 0.001) or ADC value alone where AUC = 0.89, 95% CI: 0.83–0.95 (p< 0.001). The latter was more accurate than
(choline + creatine)/citrate ratio alone, although the differ-ence was not significant (p> 0.05) (Fig. 2).
Of 151 lesions that were present in 46 patients included in this study and were identified with pathologic examination, 105 lesions were detected by MRS. Of these 105 lesions de-tected by MRS, 56 lesions had a Gleason score 3 + 3, 26 le-sions had a Gleason score of 3 + 4, 11 lele-sions had a Gleason score 4 + 3 and 12 lesions had a Gleason score
P4 + 4Table 3.
Table 4shows the sensitivity of MRS for tumor detection as graded by Gleason score. A greater sensitivity 85% and 92% was present for tumors with Gleason score 4 + 3 and
Fig. 1 Box plot: (A) the mean ADC, (B) the mean (choline + creatine)/citrate ratio in malignant and benign tissues. Table 2 Mean ADC values (mean ± SD) and mean (cho-line + creatine)/citrate ratio in malignant and benign tissues.
Parameters Mean ttest
ADC (·103mm2/s)
Malignant 1.0 ± 0.22 10.1
Benign 1.44 ± 0.28
(Choline + creatine)/citrate ratio
Malignant 1.98 ± 1.0 6.9
Benign 0.72 ± 0.39
Fig. 2 ROC curve shows mean ADC alone (green line) (AUC = 0.89), (choline + creatine)/citrate ratio alone (red line) (AUC = 0.86), and both ADC and (choline + creatine)/citrate ratio (blue line) (AUC = 0.93).
P4 + 4, respectively, while for tumors with Gleason score 3 + 3 the sensitivity was 63%. This is explained by the fact that the value of mean (choline + creatine)/citrate ratio in-creased by an increase in Gleason score due to markedly ele-vated choline levels associated with marked reduction in citrate levels (high-grade tumors). On the other hand lesions with low Gleason score had low mean (choline + creatine)/cit-rate ratio due to slight elevation in choline levels associated with slight reduction in citrate levels (low-grade tumors) (Fig. 3).
For the discrimination between Gleason score in tumors according to mean (choline + creatine)/citrate ratio there was a significant difference in lesions with low Gleason score (3 + 3) versus lesions with higher Gleason score (4 + 3 or
P4 + 4) (p< 0.001), while lesions with Gleason score (4 + 3) did not differ significantly from those with Gleason
scoreP4 + 4 (p> 0.05) (Table 5). Representative cases are shown inFigs. 4–8.
4. Discussion
Recent advances offer newly developed sequences for MR imaging. DWI has recently received attention as a promising method in the diagnosis of prostate cancer. Currently, it is the only technique that depicts differences in molecular diffu-sion. Restricted diffusion in tumor tissue is attributed to histo-pathological characteristics, such as greater cellular density caused by a high index of neoplastic replication, enlargement of nuclei and hyperchromatosis(16–19). The changes in diffu-sion properties are calculated by the apparent diffudiffu-sion coeffi-cient (ADC) which is affected by both the Brownian motion of water molecules and polluting factors fromT2 tissue signals (20). So cellular changes that inhibit the movement of water molecules result in restricted diffusion and decreased ADC val-ues in tumor tissval-ues(21).
Several studies(22–24)have reported the high performance of DWI in detecting prostate cancer.
In agreement with previous studies (9,25–30) this study showed that the ADC value for malignant tissue was signifi-cantly lower (p< 0.001) than that for non-malignant tissue. It was 1.0 ± 0.22 for tumor tissue and 1.44 ± 0.28 for non-tu-mor tissue. The sensitivity specificity for detecting prostate cancer were 81% and 84%, respectively.However, other stud-ies(15,25,31)showed that the mean ADC values fall within a wide range of values for tumor and non-tumor tissue. They ex-plained the wide variation in ADC values by physiologic fac-tors such as age, tumor size and grade and by technical factors such as variations in acquisition parameters and post-processing techniques.
MRS provides an idea about the relative concentration of chemical metabolites in the prostatic tissue by using a small values of interest (voxels)(1,2).
3D MRS detects the level of metabolites such as citrate, choline and creatine, and making it possible to differentiate tumor from non-tumor tissue. In prostate cancer the reduced
Table 4 Sensitivity of (choline + creatine)/citrate ratio for detection of pathologically proved tumor as graded by Gleason score.
(Cho + Cr)/Cit ratio Gleason score
3 + 3 3 + 4 4 + 3 P4 + 4 Total
True positive 56 26 11 12 105
False negative 33 10 2 1 46
Total 89 36 13 13 151
Sensitivity (%) 63 72 85 92 70
Fig. 3 Error bar shows mean (choline + creatine)/citrate ratio versus Gleason score. *NB: there is an increasing (choline + creatine)/citrate ratio with increasing Gleason score.
Table 5 Significance (Pvalue) of mean (choline + creatine)/ citrate ratio in differentiating the degree of aggressiveness of tumor as regarded by Gleason score.
Gleason score 3 + 3 3 + 4 4 + 3 P4 + 4
3 + 3 NA 0.28 <0.001 <0.001
3 + 4 0.28 NA <0.005 <0.006
4 + 3 <0.001 <0.005 NA 0.12
P4 + 4 <0.001 <0.006 0.12 NA
Table 3 Area under curve (AUC), sensitivity and specificity of studied parameters.
Parameters AUC Sensitivity
(%) Specificity (%) Positive predictive value (PPV) Negative predictive value (NPP) Kappa test ADC 0.89 81 84 92.4 66.1 0.60
(Cho + Cr)/Cit ratio 0.86 71.4 78.3 88.2 54.5 0.44
Combined ADC and (Cho + Cr)/Cit ratio
Fig. 4 Prostate cancer with Gleason score 4 + 4: axialT2WI (a) shows the cancer as an ill-defined hypointense focal lesion at central prostatic zone mainly to the left side. The tumor is hyperintense at DWI (b) and is of low-signal intensity at ADC map (c). 3D MRS at the same level (d), 3D MRS of a voxel at the cancer site (e) shows elevated Cho + Cr/Cit ratio and 3D MRS of a voxel at a normal site (f) shows normal Cho + Cr/Cit ratio.
Fig. 5 Prostate cancer with Gleason score 3 + 4: axialT2WI (a) shows the cancer as a small well-defined hypointense focal lesion at the left prostatic lobe. The tumor is hyperintense at DWI (b) and is of low-signal intensity at ADC map (c). 3D MRS of the same section (d), 3D MRS of a cancerous voxel (e) shows elevated Cho + Cr/Cit ratio and 3D MRS of a normal voxel (f) shows normal Cho + Cr/Cit ratio. A photomicrograph (g) shows Gleason score 3 + 4.
citrate level is due to conversion from citrate-producing to cit-rate-oxidizing metabolism, also the elevated choline level is due to increased cell growth and proliferation of tumor cells as its related phospholipid cell membrane turnover(1,2).
The choline and creatine are used in the spectral analysis in 3D MRS because both peaks are close to each other in the spectral trace and even may be inseparable(2).
In this study we used 3D MRS to evaluate prostatic tissue. By applying (Cho + Cr)/Cit ratio, this study showed that it was possible to differentiate tumor tissue by high (Cho + Cr)/Cit ratio (1.98 ± 1.0) for non-tumor tissue with significantly lower ratio (0.72 ± 0.39) (p< 0.001). The sensi-tivity and specificity for 3D MRS in this study were 71.4% and 78.3%, respectively. These findings agreed with the previ-ous studies(7,32–37)regarding the ability of MRS to differen-tiate tumor from non-tumor tissue.
The strategy of cancer care in the new millennium is direc-ted toward maximizing cancer control while minimizing the risk of complication. The treatment of prostate cancer varied from deferred therapy (watchful waiting), hormonal ablation, radical surgery and different types of radiation therapy (38). The variability in biological aggressiveness, and the biopsy specimen are not accurate predictor of Gleason score and are the most challenging characteristics of prostate cancer(2). Regarding the degree of aggressiveness of prostate cancer, this study showed that there was a trend toward an increasing (Cho + Cr)/Cit ratio in association with higher Gleason score: MRS showed higher sensitivity in detecting tumors of higher degree of aggressiveness and was 92% in Gleason score
P4 + 4, and 85% in Gleason score 4 + 3; however, the sensi-tivity was 63% for Gleason score 3 + 3. There were a signifi-cant difference (p< 0.001) in discriminating between lesions
Fig. 6 Prostate cancer with Gleason score 4 + 5: axial (a) and sagittal (d)T2WI show multiple cancerous well-defined hypointense focal lesions. The lesions are of high-signal intensity at DWI (b) and of low-signal intensity at ADC map (c). 3D MRS of the same section (e) and 3D MRS of a cancerous voxel (f) shows elevated Cho + Cr/Cit ratio. A photomicrograph (g) shows Gleason score 4 + 5.
with low Gleason score 3 + 3 and those with higher Gleason score (4 + 3 orP4 + 4). These findings agreed with the pre-vious studies(39,40) regarding the correlation between MRS (metabolite ratio) and tumor grade.
Previous studies (7,14,28) suggested that the combination of both ADC and MRS has improved the diagnostic accuracy. In consistence with these findings, our study showed that the combination of both ADC and MRS had a higher sensitivity 94.2% and specificity 85% than each parameter alone. The AUC in the combined parameters was 0.93 (Kappa
test = 0.80) which is significantly higher (p< 0.001) than ADC alone where AUC = 0.89 and Kappa test 0.60 or (Cho + Cr)/Cit ratio alone (AUC = 0.86 and Kappa test = 0.44), However, Mazaheri et al.(15) stated that com-bined DWI and MRS was not significantly more accurate than DWI alone.
In conclusion, the diagnostic power of combined DWI and MRS in discrimination between tumor and non-tumor tissue in prostate cancer is higher than a single parameter. The ability of MRS to predict and discriminate between different degrees
Fig. 7 Prostate cancer with Gleason score 4 + 3: axialT2WI (a) shows the cancer as an ill-defined hypointense focal lesion at the right prostatic lobe. The tumor is hyperintense at DWI (b) and is of low-signal intensity at ADC map (c). MRS of the cancerous voxel (d) shows elevated Cho + Cr/Cit ratio. Photomicrographs (e and f) show Gleason score 4 + 3.
of aggressiveness of prostate cancer as graded by Gleason score would warrant the use of MRS in conjugation with
DWI as a promising non-invasive parameters for the diagnos-tic workup of prostate cancer.
Fig. 8 Prostate cancer with Gleason score 4 + 3: axialT2WI (a) shows the cancer as an ill-defined iso- to hypointense focal lesion at central prostatic zone. The tumor is hyperintense at DWI (b) and is of low-signal intensity at ADC map (c). 3D MRS at the same level (d), 3D MRS of a voxel at the cancer site (e) shows elevated Cho + Cr/Cit ratio, and 3D MRS of a voxel at a normal site (f) shows normal Cho + Cr/Cit ratio.
References
(1) Hricak Hedvig, Choyke Peter L, Eberhardt Steven C, Leibel Steven A, Scardino Peter T. Imaging prostate cancer: a multidis-ciplinary perspective. Radiology 2007;243(1):28–53.
(2) Claus Filip G, Hricak Hedvig, Hattery Robert R. Pretreatment evaluation of prostate cancer: role of MR imaging and1HMR spectroscopy. Radiographics 2004;24:167–80.
(3) Choi Young Jun, Kim Jeong Kon, Kim Namkug, Kim Kyoung Won, Choi Eugene K, Kyoung-Sik K. Functional MR imaging of prostate cancer. Radiographics 2007;27:63–77.
(4) Ikonen S, Kivisaari L, Tervahartiala P, Vehmas T, Taari K, Rannikko S. Prostatic MR imaging: accuracy in differentiating cancer from other prostatic disorders. Acta Radiol 2001;42:348–54.
(5) Turkbey Baris, Albert Paul S, Kurdziel Karen, Choyke Peter L. Imaging localized prostate cancer: current approaches and new developments. AJR 2009;192:1471–80.
(6) Fu¨tterer Jurgen J, Heijmink Stijn WTPJ, Scheenen Tom WJ, et al. Prostate cancer localization with dynamic contrast-enhanced MR imaging and proton MR spectroscopic imaging. Radiology 2006;241(2):449–58.
(7) Riches Sophie F, Payne Geoffrey S, Morgan Veronica A, et al. MRI in the detection of prostate cancer: combined apparent diffusion coefficient, metabolite ratio, and vascular parameters. AJR 2009;193:1583–91.
(8) Morgan VA, Kyriazi S, Ashley SE, deSouza NM. Evaluation of the potential of diffusion-weighted imaging in prostate cancer detection. Acta Radiol 2007;48:695–703.
(9) Hosseinzadeh K, Schwarz SD. Endorectal diffusion-weighted imaging in prostate cancer to differentiate malignant and benign peripheral zone tissue. J Magn Reson Imaging 2004;20:654–61. (10) Gibbs P, Pickles MD, Turnbull LW. Diffusion imaging of the
prostate at 3.0 tesla. Invest Radiol 2006;41:185–8.
(11) Shimofusa R, Fujimoto H, Akamata H. Diffusion-weighted imaging of prostate cancer. J Comput Assist Tomogr 2005;29:149–53. (12) Shukla-Dave Amita, Hricak Hedvig, Moskowitz Chaya, et al.
Detection of prostate cancer with MR spectroscopic imaging: an expanded paradigm incorporating polyamines. Radiology 2007;245(2):499–506.
(13) Zakian KL, Sircar K, Hricak H. Correlation of proton MR spectroscopic imaging with Gleason score based on step-section pathologic analysis after radical prostatectomy. Radiology 2005;234:804–14.
(14) Kumar V, Jagannathan NR, Kumar R. Correlation between metabolite ratios and ADC values of prostate in men with increased PSA level. Magn Reson Imaging 2006;24:541–8. (15) Mazaheri Yousef, Shukla-Dave Amita, Hricak Hedvig, et al.
Prostate cancer: identification with combined diffusion-weighted MR imaging and 3D1H MR spectroscopic imaging––correlation with pathologic findings. Radiology 2008;246(2):480–8. (16) Kılıc¸kesmez O¨, Cimilli T,_Inci E, et al. Diffusion-weighted MRI
of urinary bladder and prostate cancers. Diagn Interv Radiol 2009;15:104–10.
(17) Kilickesmez O, Yirik G, Bayramog˘lu S, Cimilli T, Aydin S. Non-breath-hold high b-value diffusion-weighted MRI with parallel imaging technique: apparent diffusion coefficient determination in normal abdominal organs. Diagn Interv Radiol 2008;14:83–7. (18) Bammer R. Basic principles of diffusion-weighted imaging. Eur J
Radiol 2003;45:169–84.
(19) Ren J, Huan Y, Wang H, et al. Diffusion-weighted imaging in normal prostate and differential diagnosis of prostate diseases. Abdom Imaging 2008;33:724–8.
(20) Manenti G, Squillaci E, Roma M, et al. In vivo measurement of the apparent diffusion coefficient in normal and malignant prostatic tissue using thin-slice echo-planar imaging. Radiol Med 2006;111:1124–33.
(21) Choi YJ, Kim JK, Kim N, Kim KW, Choi EK, Cho KS. Functional MR imaging of prostate cancer. Radiographics 2007;27:63–75. (22) Kim CK, Park BK, Han JJ, Kang TW, Lee HM.
Diffusion-weighted imaging of the prostate at 3 T for differentiation of malignant and benign tissue in transition and peripheral zones: preliminary results. J Comput Assist Tomogr 2007;31:449–54. (23) Haider MA, van der Kwast TH, Tanguay J, et al. T2-weighted
and diffusion-weighted MRI for localization of prostate cancer. AJR Am J Roentgenol 2007;189:323–8.
(24) Miao H, Fukatsu H, Ishigaki T. Prostate cancer detection with 3-T MRI: comparison of diffusion-weighted and 3-T2-weighted imaging. Eur J Radiol 2007;61:297–302.
(25) Issa B. In vivo measurement of the apparent diffusion coefficient in normal and malignant prostatic tissues using echo-planar imaging. J Magn Reson Imaging 2002;16:196–200.
(26) Sato C, Naganawa S, Nakamura T, et al. Differentiation of noncancerous tissue and cancer lesions by apparent diffusion coefficient values in transition and peripheral zones of the prostate. J Magn Reson Imaging 2005;21:258–62.
(27) Engelbrecht MR, Huisman HJ, Laheij RJ, et al. Discrimination of prostate cancer from normal peripheral zone and central gland tissue by using dynamic contrast-enhanced MR imaging. Radi-ology 2003;229:248–54.
(28) Reinsberg SA, Payne GS, Riches SF, et al. Combined use of diffusion-weighted MRI and 1H MR spectroscopy to increase accuracy in prostate cancer detection. AJR 2007;188:91–8. (29) deSouza NM, Reinsberg SA, Scurr ED, Brewster JM, Payne GS.
Magnetic resonance imaging in prostate cancer: the value of apparent diffusion coefficients for identifying malignant nodules. Br J Radiol 2007;80:90–5.
(30) Gibbs P, Pickles MD, Turnbull LW. Diffusion imaging of the prostate at 3.0 Tesla. Invest Radiol 2006;41:185–8.
(31) Gibbs P, Tozer DJ, Liney GP, Turnbull LW. Comparison of quantitative T2 mapping and diffusion-weighted imaging in the normal and pathologic prostate. Magn Reson Med 2001;46:1054–8. (32) Carlani M, Mancino S, Bonanno E, Finazzi Agro` E, Simonetti G. Combined morphological, [1H]-MR spectroscopic and contrast-enhanced imaging of human prostate cancer with a 3-Tesla scanner: preliminary experience. Radiol Med 2008;113:670–88. (33) Yuen JS, Thng CH, Tan PH, et al. Endorectal magnetic
resonance imaging and spectroscopy for the detection of tumor foci in men with prior negative transrectal ultrasound prostate biopsy. J Urol 2004;171:1482–6.
(34) Hasumi M, Suzuki K, Oya N, et al. MR spectroscopy as a reliable diagnostic tool for localization of prostate cancer. Anticancer Res 2002;22:1205–8.
(35) Hasumi M, Suzuki K, Taketomi A, et al. The combination of multi-voxel MR spectroscopy with MR imaging improve the diagnostic accuracy for localization of prostate cancer. Antican-cer Res 2003;23:4223–7.
(36) Wefer AE, Hricak H, Vigneron DB, et al. Sextant localization of prostate cancer: comparison of sextant biopsy, magnetic reso-nance imaging and magnetic resoreso-nance spectroscopic imaging with step section histology. J Urol 2000;164:400–4.
(37) Weis J, Ahlstrm H, Hlavcak P, Hoggman M, Ortiz-Nieto F, Bergman A. Two dimensional spectroscopic imaging for pre-treatment evaluation of prostate cancer: comparison with the step-section histology after radical prostatectomy. Magn Reson Imaging 2009;27:87–93.
(38) Gilligan T, Kantoff PW. Chemotherapy for prostate cancer. Urology 2002;60:94–100.
(39) Kurhanewicz J, Vigneron DB. Advances in MR spectroscopy of the prostate. Magn Reson Imaging Clin N Am 2008;16:697–710. (40) Zakian KL, Sircar K, Kleinman S, Shukla-Dave A, Kattan MW, Hricak H. Correlation of proton MR spectroscopic imaging with Gleason score based on step section radical prostatectomy (abstr.). Radiology 2002;225(P):1525.