Diagnostic performance of diffusion-weighted imaging combined with dynamic contrast-enhanced magnetic resonance imaging for prostate cancer: a systematic review and meta-analysis

Abstract

Background

The diagnostic performance of diffusion-weighted imaging (DWI) combined with dynamic contrast-enhanced (DCE)-magnetic resonance imaging (MRI) for the detection of prostate cancer (PCa) has not been studied systematically to date.

Purpose

To investigate the value of DWI combined with DCE-MRI quantitative analysis in the diagnosis of PCa.

Material and Methods

A systematic search was conducted through PubMed, MEDLINE, the Cochrane Library, and EMBASE databases without any restriction to language up to 10 December 2019. Studies that used a combination of DWI and DCE-MRI for diagnosing PCa were included.

Results

Nine studies with 778 participants were included. The combination of DWI and DCE-MRI provide accurate performance in diagnosing PCa with pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and diagnostic odds ratios of 0.79 (95% confidence interval [CI] = 0.76–0.81), 0.85 (95% CI = 0.83–0.86), 6.58 (95% CI = 3.93–11.00), 0.24 (95% CI = 0.17–0.34), and 36.43 (95% CI = 14.41–92.12), respectively. The pooled area under the summary receiver operating characteristic curve was 0.9268. Moreover, 1.5-T MR scanners demonstrated a slightly better performance than 3.0-T scanners.

Conclusion

Combined DCE-MRI and DWI could demonstrate a highly accurate area under the curve, sensitivity, and specificity for detecting PCa. More studies with large sample sizes are warranted to confirm these results.

Keywords

Prostate cancer diagnostic performance diffusion-weighted imaging dynamic contrast-enhanced magnetic resonance imaging meta-analysis

Introduction

Prostate cancer (PCa) is the most common type of cancer in older men and is considered the second leading cause of cancer-related mortality (1). Clinical management of PCa remains challenging as it can be easily misdiagnosed due to the absence of symptoms, especially during the early stage of the disease (2). Magnetic resonance imaging (MRI) has been regarded as one of the best imaging methods for the diagnosis of PCa as it allows unique anatomic assessment of the prostate and has the best soft-tissue resolution. MRI can be used from initial detection until planning the treatment of the disease (3). However, the specificity and the diagnostic accuracy of traditional MRI method in detecting PCa should be improved.

Although the sequence of T2-weighted (T2W) imaging lacks sufficient sensitivity and specificity, it has been the mainstay sequence for detecting tumors by MRI (4). In recent years, T2W imaging combined with diffusion-weighted imaging (DWI) (5) and dynamic contrast-enhanced (DCE) (6) imaging have significantly improved the sensitivity and specificity of MRI in tumor localization when compared with the use of T2W imaging alone (7,8). DWI reflects the Brownian movement of water in the living tissues by imaging the diffusion characteristics and DCE-MRI analyzes tumor angiogenesis by observing the diffusion of contrast agents into the extravascular space over time. Several recent studies (9 –11) have demonstrated that the additional functional MRI techniques, DWI and DCE, have provided high contrast resolution morphologic imaging, metabolic information, and direct depiction of tumor vascularity, significantly improving the diagnostic accuracy of PCa by MRI (12,13). However, the diagnostic performance of DWI combined with DCE-MRI for PCa has not been studied systematically to date. Therefore, the aim of the present systematic review and meta-analysis was to investigate the value of DWI combined with DCE-MRI quantitative analysis in diagnosing PCa.

Material and Methods

The present systematic review and meta-analysis was performed according to the 2009 preferred reporting items for systematic reviews and meta-analysis (PRISMA) statement (14).

Literature search

The individual and joint keywords of “dynamic contrast-enhanced,” “DCE,” “diffusion-weighted imaging,” “DWI,” and “prostate cancer” were used for searching the studies in PubMed, MEDLINE, the Cochrane Library, and EMBASE databases without any restriction to language up to 10 December 2019. To include more potentially relevant studies, the bibliographies of all relevant studies and reviews were identified; Google Scholar was also searched for studies that cited relevant references.

Eligibility criteria

Studies were considered eligible if they met the following criteria: (1) patients with histologically confirmed diagnosis of PCa; (2) detection using MRI combined with DWI and DCE; (3) pathologically confirmed PCa as study group and non-cancer biopsy as control group; (4) the study provided sufficient raw data to construct a 2 × 2 contingency table to perform MRI sequences; (5) studies published in English; and (6) the most inclusive publication was included in this meta-analysis.

Case reports, letters, review articles, studies conducted in animal models or experiments in vitro, studies not published in English, and studies for which data are not available were excluded from the analysis.

Data extraction

All eligible studies were assessed independently by two reviewers. Consensus was reached on all items by discussion with a third reviewer. All the information from relevant studies was extracted by using a standardized form and consensus was reached on all items by the two reviewers above. The following information was extracted from each study: the author and year of publication; country; study design; sample size; patient characteristics (e.g. age and nation); imaging characteristics including field strength; type of coil; MR sequences, and features, the true-negative (TN), false-negative (FN), true-positive (TP), and false-positive (FP) results of DWI and DCE.

Quality assessment of studies

The quality assessment of each included study was independently performed and cross-checked by two investigators using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool (15). The QUADAS-2 contains four domains: (i) patient selection, which describes the selection of patients; (ii) index test, which describes the conduction and interpretation of the results; (iii) reference standard, which describes the conduction of reference standard and interpretation of the results; and (iv) flow and timing, which describes the flow of patient inclusion and exclusion and the interval between the index test.

Statistical analysis

The individual sensitivity and specificity of DWI and DCE were calculated for each study by TN, FN, TP, and FP results. The data from each study were combined using a random-effects model to calculate the pooled sensitivity and pooled specificity across all the included studies and expressed with corresponding 95% confidence intervals (CI). The summary receiver operating characteristic (sROC) curves were computed for DWI and DCE, and the area under the curve (AUC) and associated standard error (SE) and 95% CIs were also used to assess each technique. The Der Simonian-Laird random-effects method was employed for determining the AUC of the sROC curves due to the presence of heterogeneity. The 95% CIs of the pooled metrics were compared to assess the relative performance of the technique.

The I² statistics, the standard heterogeneity test, were used to assess the consistency of the effect sizes, which subsequently indicates the percentage of the variability in effect estimates because of true between-study variance rather than within-study variance. Heterogeneity was defined as low, moderate, and high according to the values of I² by 25%, 50%, and 75%, respectively (16). Publication bias was assessed by Begg’s rank correlation (17) and Egger’s (18) weighted regression methods (P < 0.05 was considered to be statistically significant publication bias). Review Manager (version 5.3, The Cochrane Collaboration, Oxford, UK) and STATA 15.0 (Stata Corporation, College Station, TX, USA) were used for statistical analyses. The Begg and Egger tests were assessed by STATA 15.0 (Stata Corporation, College Station, TX, USA). A P value < 0.05 was considered significant for all analyses.

Results

Study selection

Search strategy yielded 506 potentially relevant studies and 120 of these were excluded due to overlapping. Of the remaining 386 studies, 343 studies were excluded by screening the titles or abstracts. Ultimately, eight articles (nine studies) (9 –11,19 –23) were included for data extraction and meta-analysis after reading the full texts. A flow chart of study selection process was shown in Fig. 1.

Fig. 1.

Flow chart of the study selection process.

Study characteristics

In total, eight articles (nine studies) with 778 participants were finally included in this meta-analysis and the characteristics of the included studies and participants were summarized in Table 1. The sample size of the studies included was in the range of 15–235 and the studies were published between 2006 and 2019. Two studies were conducted in Japan (20,21) and two in Canada (19,23), one in China (11), one in Spain (9), one in Ireland (10), and one in the Republic of Korea (22). One article was divided into two studies as the author provided results from two independent radiologists.

Table 1.

Characteristics of study participants.

Studies included	Country	Age (years) (median (range))	Study design	Reference standard	Slice thickness (mm)	Field strength (T)	Type of coil	Features (s/mm²)	Readers (n)	Method of analysis	Control
Kozlowski et al., 2006	Canada	60.30 (48–71)	Prospective	Prostatic biopsy	5	1.5	PPC	0–600	2	Visually	Normal PZ
Kitajima et al., 2010	Japan	69.00 (56–84)	Retrospective	Prostatic biopsy	3	3	PPC	0–1000	2	Visually	Normal PZ
Iwazawa et al., 2011	Japan	68.80 (41–86)	Retrospective	Prostatic biopsy	5	1.5	PPC	0–1000	2	Visually	Non-PCa patients
Vilanova et al., 2011	Spain	63.50 (43–87)	Retrospective	Prostatic biopsy	5	1.5	PPC	0–1000	2	Non-visually	Normal PZ
Oon et al., 2014	Ireland	64.00*	Retrospective	Prostatic biopsy	5	1.5	PPC	NA	5	Visually	Normal PZ
Cha et al., 2014	Republic of Korea	63.50 (51–73)	Retrospective	Prostatic biopsy	3	3	PPC	0–1000	2	Visually	Non-PCa patients
Sabouri et al., 2018	Canada	66.47 (58–80)	Prospective	Prostatic biopsy	4	3	PPC	0–1000	2	Visually	Normal PZ
Ma et al., 2019	China	74.00 (59–85)^b72.00 (52–96)^c	Retrospective	Prostatic biopsy	3	3	PPC	700	2	Visually	Non-PCa patients

Values are given as median (range).

NA, not available; PCa, prostate cancer; PPC, pelvic phased-array coil; PZ, peripheral zone; TZ, transition zone.

*Mean age.

^†Patients with PCa.

^‡Patients without PCa.

Assessment of study quality and risk of bias

Two reviewers independently assessed the quality of all included studies by QUADAS-2 items. For all four domains of the QUADAS-2, almost each item had unclear risks due to missing information and none of the item were judged as high risk. The quality assessment results are presented in Fig. 2.

Fig. 2.

Risk of bias assessment of included studies.

Diagnostic accuracy

The pooled sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), and diagnostic odds ratios (DOR) for diagnostic accuracy of MRI combined with DWI and DCE in diagnosing PCa were 0.79 (95% CI = 0.76–0.81), 0.85 (95% CI = 0.83–0.86), 6.58 (95% CI = 3.93–11.00), 0.24 (95% CI = 0.17–0.34), and 36.43 (95% CI = 14.41–92.12), respectively. More detailed results were presented in Figs. 3 –8. As shown in Fig. 9, the pooled area under the SROC curve was 0.9268 with a standard error of 0.0267. The forest plots suggest that heterogeneity was high with almost all the I² values exceeded 80%.

Fig. 3.

Summary sensitivity analysis of the included studies.

Fig. 4.

Summary specificity analysis of the included studies.

Fig. 5.

Summary of positive likelihood ratio of the included studies.

Fig. 6.

Summary of negative likelihood ratio of the included studies.

Fig. 7.

Summary of diagnostic odds ratios of the included studies.

Fig. 8.

Summary of pooled receiver operating characteristic plane.

Fig. 9.

Summary of pooled area under the summary receiver operating characteristic curve.

Subgroup analysis

Table 2 presented the results of subgroups to investigate the influence of study design, prostate regions for sampling, and technical details on pooled sensitivity, specificity, PLR, NLR, and DOR. Two studies reported the data on samples from peripheral zones and the pooled results revealed that the peripheral zone had a slightly higher sensitivity and lower specificity. The included studies were categorized into 1.5-T and 3.0-T groups according to the field strength. The pooled sensitivity, specificity, PLR, NLR, and DOR for the 1.5-T group were lower than that of the 3.0-T group. When the included studies were divided by the b-values of 1000 s/mm² or < 1000 s/mm², the results for the b = 1000 group was similar to that of other studies. The pooled results for the studies grouped by study design and country showed no significant change when compared to the results of all other studies. As the studies were categorized by the controls, the results of studies that used normal peripheral zone as controls showed a slightly better pooled sensitivity, specificity, PLR, and NLR. As shown in Table 2, the heterogeneity for sensitivity, specificity, PLR, NLR, and DOR was decreased to a moderate level (I² < 75%) by dividing the studies into various groups.

Table 2.

Sensitivity analyses for subgroups of studies.

Studies	Studies (n)	Sensitivity (95% CI), I²	Specificity (95% CI), I²	PLR (95% CI), I²	NLR (95%CI), I²	Diagnostic OR (95% CI), I²
All studies	9	0.79 (0.76–0.81), 85.30%	0.85 (0.83–0.86), 92.30%	6.58 (3.93–11.00), 91.10%	0.24 (0.17–0.34), 81.20%	36.43 (14.41–92.12), 89.70%
Prostate regions
Peripheral zone	2	0.84 (0.79–0.88), 67.90%	0.79 (0.75–0.82), 32.80%	4.21 (2.76–6.41), 58.80%	0.19 (0.10–0.34), 71.70%	23.20 (8.50–63.35), 76.10%
Field strength (T)
1.5	3	0.89 (0.84–0.92), 88.90%	0.93 (0.91–0.92), 83.00%	11.16 (4.73–26.33), 83.40%	0.12 (0.05–0.27), 80.20%	106.36 (28.23–400.72), 80.60%
3.0	6	0.74 (0.70–0.77), 52.10%	0.83 (0.81–0.84), 90.60%	4.70 (2.85–7.76), 86.60%	0.32 (0.23–0.43), 68.40%	18.95 (7.32–49.05), 82.90%
b-value (s/mm²)
b = 1000	7	0.79 (0.75–0.82), 86.70%	0.85 (0.84–0.87), 93.70%	8.76 (4.72–16.27), 92.20%	0.22 (0.15–0.32), 76.80%	58.16 (19.45–173.92), 90.20%
b < 1000	2	NA	NA	NA	NA	NA
Study design
Retrospective	7	0.77 (0.74–0.80), 56.00%	0.85 (0.83–0.87), 93.30%	7.37 (4.33–12.54), 90.40%	0.25 (0.19–0.32), 58.40%	37.39 (15.72–88.94), 86.50%
Prospective	2	NA	NA	NA	NA	NA
Country
Asian	5	0.78 (0.74–0.81), 68.10%	0.84 (0.82–0.86), 93.20%	6.49 (3.84–10.97), 91.50%	0.22 (0.15–0.33), 71.40%	32.31 (12.09–86.34), 89.30%
Others	4	0.81 (0.76–0.86), 92.60%	0.92 (0.88–0.95), 91.20%	9.46 (1.52–58.72), 93.40%	0.23 (0.10–0.56), 90.90%	56.42 (3.73–852.37), 92.60%

CI, confidence interval; NA, not available; NLR, negative likelihood ratio; OR, odds ratio; PLR, positive likelihood ratio.

Publication bias

No potential publication bias was observed among the included trials (P value of the analysis > 0.05) according to Begg’s rank correlation analysis and Egger’s weighted regression analysis. The potential publication bias was presented in detailed in Table S1.

Discussion

To the best of our knowledge, this is the first meta-analysis study to examine the accuracy of combined DWI and DCE-MRI for detecting PCa. In the present study, the results of nine studies on the combination of DWI and DCE-MRI provided accurate performance for diagnosing PCa with pooled sensitivity and specificity of 0.79 (95% CI = 0.76–0.81) and 0.85 (95% CI = 0.83–0.86), and the pooled area under the SROC curve was 0.9268.

Another meta-analysis (24) addressing the detection of PCa by T2-MRI showed pooled a sensitivity and specificity of 0.75 and 0.60, respectively. The sensitivity of 0.79 in the present study was significantly lower than that obtained by T2-MRI. Another meta-analysis (25) that focused on detection of PCa by DCE-MRI revealed significantly better AUC by DCE-MRI (0.82–0.86) and DWI (0.84–0.88) than T2W imaging (0.68–0.77). The AUC in the present study outperformed the DCE-MRI, DWI, and T2W imaging. A most recent meta-analysis (26) focused on the combination of DCE-MRI, DWI, and T2W imaging for detecting the extracapsular extension in patients with PCa, and the results showed no significant differences in the performance when compared with the use of T2W imaging alone or with additional functional MRI. In addition, the differences were evaluated using higher field strength (3.0 T or 1.5 T). However, in the present study, the higher MR scanner with 3.0 T did not provided superior sensitivity for detecting PCa. In general, the B0 inhomogeneity at 3.0 T is more severe than B1 inhomogeneity. The issue of B0 inhomogeneity is particularly problematic for DWI, whereas B0 distortions cause piling up of signals that can obscure lesions. This might be a partial reason for the inferior sensitivity of 3.0 T in detecting PCa than 1.5 T. Similarly, another meta-analysis (27) that addressed a similar topic showed comparable diagnostic values of PCa with 1.5-T and 3.0-T MR scanners. Recently, functional MRI techniques have provided high contrast resolution morphologic imaging, metabolic information and direct depiction of tumor vascularity, the multiparametric MRI methods, such as combined DCE-MRI and DWI, or combined DCE-MRI, DWI and T2W imaging, have largely improved the sensitivity and specificity of detection and determination of tumor staging. The present study provided accurate diagnostic performance in detecting PCa using the combination of DWI and DCE-MRI, and it might be possible to avoid unnecessary surgeries or biopsy procedures for inadequate observation of malignancies by combining with DCE-MRI and DWI. However, a cut-off value that can be used to classify malignant/benign lesions could not be determined based on the present meta-analysis. In the future, more studies that aim to explore the cut-off values and focus on the variations in b-values, dynamic phase, and pathological characteristics of the lesions are needed. Furthermore, various studies (28,29) have demonstrated that DCE-MRI has been successful in diagnosing and staging PCa in the peripheral zone. However, there appears to be some overlap in the enhancement patterns between the tumor and the transitional zone (30). In the present study, although it is limited by the number of included studies, the results showed a slightly higher sensitivity and lower specificity in the peripheral zone than others. Future research should not only concentrate on improving tumor detection but also on enhancing the local staging and treatment response assessment. Due to the inclusion of early studies, DCE-MRI and DWI showed promising results in detecting recurrent diseases after external beam radiotherapy as well as after radical prostatectomy (10,22).

The present study has some limitations. First, only nine studies were included in the current meta-analysis and most of these have limited sample sizes. The smaller sample size might reduce the credibility and stability of the results and, moreover, limit the conduction of more subgroup or sensitivity analyses. Second, most of the studies did not match the participants by age. Therefore, the differences in mean age of the participants of each study varied largely, causing heterogeneity and reduction in the stability of the results. Third, the heterogeneity in the analysis methods remained extremely high and this might be due to different type of controls or stages of PCa included in the studies. Analysis of more subgroups decreases the heterogeneity to a moderate level, and so the results should be further validated in the future. Other than the performance of DCE-MRI and DWI sequences, the technical factors that might influence the results of perfusion imaging, such as the rate of contrast media injection and patients’ hemodynamic status should also be observed. However, due to inclusion of limited number of studies, the effect of heterogeneity cannot be assessed. More importantly, it might not be possible to ensure that all parameters are kept constant. Finally, potential language bias might exist as our literature search included articles published in English only.

In conclusion, combined DCE-MRI and DWI produced highly accurate AUC, sensitivity, and specificity in the detection of PCa. More studies are needed to further assess and compare the combination of DCE-MRI and DWI, and combination of DCE-MRI, DWI, and T2W imaging in the future.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received the following financial support for the research, authorship, and/or publication of this article: The work was supported by the Chongqing Health and Wellness Commission General Program (2017MSXM105) and The China National Cancer Center Climbing Fund (NCC201822872).

ORCID iD

Jiuquan Zhang

References

Torre

Bray

Siegel

, et al. Global cancer statistics, 2012. CA Cancer J Clin 2015; 65:87–108.

Matoso

Epstein

JI.

Grading of prostate cancer: Past, present, and future. Curr Urol Rep 2016; 17:25.

Mendhiratta

Taneja

Rosenkrantz

AB.

The role of MRI in prostate cancer diagnosis and management. Future Oncol 2016; 12:2431–2443.

Turkbey

Albert

Kurdziel

, et al. Imaging localized prostate cancer: Current approaches and new developments. AJR Am J Roentgenol 2009; 192:1471–1480.

Kim

Park

Kim

Diffusion-weighted MRI at 3 T for the evaluation of prostate cancer. AJR Am J Roentgenol 2010; 194:1461–1469.

Verma

Turkbey

Muradyan

, et al. Overview of dynamic contrast-enhanced MRI in prostate cancer diagnosis and management. AJR Am J Roentgenol 2012; 198:1277–1288.

Hoeks

Futterer

Somford

, et al. [ multiparametric MRI for prostate cancer screening]. Ned Tijdschr Geneeskd 2009; 153:B487.

Borre

[ Prostate cancer - screening and early detection]. Ugeskr Laeger 2018; 180

Vilanova

Barcelo-Vidal

Comet

, et al. Usefulness of prebiopsy multifunctional and morphologic MRI combined with free-to-total prostate-specific antigen ratio in the detection of prostate cancer. AJR Am J Roentgenol 2011; 196:W715–722.

10.

Oon

Power

Kelly

, et al. The accuracy of magnetic resonance imaging in prostate cancer staging: A single-institution experience. Ir J Med Sci 2015; 184:313–317.

11.

Sheng

, et al. Application evaluation of DCE-MRI combined with quantitative analysis of DWI for the diagnosis of prostate cancer. Oncol Lett 2019; 17:3077–3084.

12.

Pinto

Totaro

Calarco

, et al. Imaging in prostate cancer diagnosis: Present role and future perspectives. Urol Int 2011; 86:373–382.

13.

Pinto

Totaro

Palermo

, et al. Imaging in prostate cancer staging: Present role and future perspectives. Urol Int 2012; 88:125–136.

14.

Moher

Liberati

Tetzlaff

, et al. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. BMJ 2009; 339:b2535.

15.

Whiting

Weswood

Rutjes

, et al. Evaluation of QUADAS, a tool for the quality assessment of diagnostic accuracy studies. BMC Med Res Methodol 2006; 6:9.

16.

Higgins

Thompson

Deeks

, et al. Measuring inconsistency in meta-analyses. BMJ 2003; 327:557–560.

17.

Begg

Mazumdar

Operating characteristics of a rank correlation test for publication bias. Biometrics 1994; 50:1088–1101.

18.

Egger

Davey Smith

Schneider

, et al. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997; 315:629–634.

19.

Kozlowski

Chang

Jones

, et al. Combined diffusion-weighted and dynamic contrast-enhanced MRI for prostate cancer diagnosis–correlation with biopsy and histopathology. J Magn Reson Imaging 2006; 24:108–113.

20.

Kitajima

Kaji

Fukabori

, et al. Prostate cancer detection with 3 T MRI: Comparison of diffusion-weighted imaging and dynamic contrast-enhanced MRI in combination with T2-weighted imaging. J Magn Reson Imaging 2010; 31:625–631.

21.

Iwazawa

Mitani

Sassa

, et al. Prostate cancer detection with MRI: Is dynamic contrast-enhanced imaging necessary in addition to diffusion-weighted imaging? Diagn Interv Radiol 2011; 17:243–248.

22.

Cha

Kim

Park

, et al. Evaluation of suspected soft tissue lesion in the prostate bed after radical prostatectomy using 3T multiparametric magnetic resonance imaging. Magn Reson Imaging 2015; 33:407–412.

23.

Sabouri

Chang

Goldenberg

, et al. Comparing diagnostic accuracy of luminal water imaging with diffusion-weighted and dynamic contrast-enhanced MRI in prostate cancer: A quantitative MRI study. NMR Biomed 2019; 32:e4048.

24.

Fusco

Sansone

Granata

, et al. A systematic review on multiparametric MR imaging in prostate cancer detection. Infect Agent Cancer 2017; 12:57.

25.

Tan

Hobbs

Wei

, et al. Dynamic contrast-enhanced MRI for the detection of prostate cancer: Meta-analysis. AJR Am J Roentgenol 2015; 204:W439–448.

26.

Zhang

Liu

Chen

, et al. Accuracy of multiparametric magnetic resonance imaging for detecting extracapsular extension in prostate cancer: A systematic review and meta-analysis. Br J Radiol 2019; 92:20190480.

27.

Chen

Sutedjo

Wang

, et al. Prostate cancer magnetic resonance spectroscopy imaging at 1.5 and 3.0 T: A meta-analysis. Technol Cancer Res Treat 2016; 15:625–631.

28.

Turnbull

Buckley

Turnbull

, et al. Differentiation of prostatic carcinoma and benign prostatic hyperplasia: Correlation between dynamic Gd-DTPA-enhanced MR imaging and histopathology. J Magn Reson Imaging 1999; 9:311–316.

29.

Barentsz

Engelbrecht

Jager

, et al. Fast dynamic gadolinium-enhanced MR imaging of urinary bladder and prostate cancer. J Magn Reson Imaging 1999; 10:295–304.

30.

Padhani

Gapinski

Macvicar

, et al. Dynamic contrast enhanced MRI of prostate cancer: Correlation with morphology and tumour stage, histological grade and PSA. Clin Radiol 2000; 55:99–109.