Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Padhani, A. R.
Right arrow Articles by Husband, J. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Padhani, A. R.
Right arrow Articles by Husband, J. E.
(Radiology. 2001;218:365-374.)
© RSNA, 2001

Genitourinary Imaging

Effects of Androgen Deprivation on Prostatic Morphology and Vascular Permeability Evaluated with MR Imaging1

Anwar R. Padhani, MRCP, FRCR, A. David MacVicar, MRCP, FRCR, Connie J. Gapinski, MD, PhD 2, David P. Dearnaley, MA, MD, FRCP, FRCR, Geoffrey J. M. Parker, PhD 3, John Suckling, PhD 4, Martin O. Leach, PhD and Janet E. Husband, FRCP, FRCR

1 From the CRC Clinical Magnetic Resonance Research Group and Academic Department of Radiotherapy (D.P.D.), Institute of Cancer Research and Royal Marsden NHS Trust, Sutton, Surrey, England. From the 1998 RSNA scientific assembly. Received Jan 7, 2000; revision requested Mar 2; revision received May 30; accepted Jun 28. Supported by CRC grant no. SP1780-0103, the Bob Champion Cancer Trust, and the NHS Executive. Address correspondence to A.R.P., Paul Strickland Cancer Center, Mount Vernon Hospital, Rickmansworth Rd, Northwood, Middlesex HA6 2RN, United Kingdom (e-mail: anwar_padhani@fsnet.co.uk).   The views expressed in this article are those of the authors and not necessarily those of the NHS Executive.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
REFERENCES
 
PURPOSE: To assess magnetic resonance (MR) measures of vascular permeability of prostate cancer treated with androgen deprivation and to correlate these with morphologic appearances and serum prostate-specific antigen (PSA) levels.

MATERIALS AND METHODS: MR examinations in 56 consecutive patients with prostate cancer were performed before and after luteinizing hormone–releasing hormone analog treatment. T2-weighted and contrast medium–enhanced T1-weighted MR images were obtained. Pre- and posttreatment comparisons of morphologic features, glandular volume, and enhancement-related parameters (capillary permeability, leakage space, gadolinium accumulation) were made.

RESULTS: Fifty-five tumors were seen before treatment; 42, after treatment. Signal intensity in the peripheral zone and seminal vesicles decreased on T2-weighted images in 42 (75%) and 25 (45%) patients, respectively. Median volume in tumor decreased by 65% (95% CI: 55%, 76%); in central gland, by 30% (95% CI: 25%, 35%). Reductions in tumor permeability (P < .001) and changes in washout patterns were observed (P < .001). Tumor permeability reductions coincided with a decrease in serum PSA levels in 91% of patients. A weak correlation between tumor permeability and volume change was seen (r = 0.55, P = .04). Reductions in peripheral zone (P < .001) and central gland (P = .009) permeability were noted.

CONCLUSION: Androgen deprivation decreases tumor volume and vascular permeability and impairs detection of prostate cancers. Use of MR estimates of permeability may be an additional way of assessing prostatic tumor response to antiandrogen treatment.

Index terms: Hormones • Magnetic resonance (MR), contrast enhancement, 84.12143 • Prostate neoplasms, 844.32 • Prostate neoplasms, MR, 84.121411, 84.121412, 84.12143 • Prostate neoplasms, therapy


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
 DISCUSSION
REFERENCES
 
The choice of appropriate treatment for patients with prostate cancer remains controversial. The most commonly offered treatments include observation only, radical prostatectomy, radiation therapy, hormone ablation treatment, or a combination (1–4). Treatment selection is guided by patient age and general condition, tumor stage and histologic grade, serum prostate-specific antigen (PSA) levels, and patient and physician preferences. Hormone ablation treatment is the preferred choice for those patients with advanced disease, but it is also used in patients before radiation therapy or prostatectomy (4–10). Hormone ablation before surgery results in substantial reductions in the serum PSA levels, prostatic and tumor volume, and rate of positive surgical margins but does not affect long-term outcome (11–15). The latter finding is in contrast to that of treatment before radiation therapy, where long-term results are improved with the use of combination treatment (7,8,10,16), and findings in animal models suggest a synergistic effect between androgen ablation and radiation therapy (17,18). The response of patients to hormonal treatment can be assessed by means of digital rectal examination, changes in serum PSA levels, and transrectal ultrasonography (US) (19,20). Recently, magnetic resonance (MR) imaging has been used to assess the response of prostate cancer treated with androgen deprivation (21,22). Imaging methods are able to demonstrate a reduction in the volume of the prostate gland but are limited in objective assessments of tumor response. This limitation is due to morphologic changes occurring within the prostate gland on treatment that impair tumor visibility.

Dynamic contrast medium–enhanced MR imaging can be used to noninvasively assess functional aspects of tissue microcirculation (23). With optimal data collection, MR sequences can be designed to be sensitive to tissue perfusion and blood volume (so-called T2* methods) and/or capillary wall permeability and extracellular leakage space (so-called T1 methods). Vascular endothelial growth factor (VEGF) also called vascular permeability factor is a recognized stimulus of tumor neoangiogenesis and a potent tissue permeability factor (24). Prostatic VEGF production requires continual stimulation by androgens (25). Findings in cell and animal studies (26) suggest that androgen withdrawal results in down-regulation of VEGF production and vascular regression prior to tumor cell death. The aims of this study were therefore (a) to assess MR measures of permeability of normal prostatic tissue and cancer treated with androgen deprivation and (b) to correlate these with morphologic appearances and serum PSA levels.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
REFERENCES
 
Patients and Treatment
A prospective study was performed in 56 consecutive untreated patients with biopsy-proved prostate cancer who were undergoing diagnostic pelvic MR imaging for staging purposes. The examinations were performed between April 1996 and January 1998. Our institutional committees on clinical research and ethics approved the study, and patient consent was obtained. The median age of the patients was 69 years (range, 51–77 years). The histologic diagnosis of prostate cancer was made by means of core biopsy in 47 patients and from transurethral resection of the prostate, or TURP, specimens in nine. Histologic features were classified into five patterns according to the degree of architectural differentiation of the tumor. The Gleason (sum) score accounted for tumor heterogeneity, which is the sum of the two dominant grade numbers (27). The median Gleason score was 6 (range, 4–9). The pretreatment MR imaging stages were localized disease (stages 1 and 2) in 22 patients and advanced disease in 34. Further patient details are given in Table 1.


View this table:
[in this window]
[in a new window]

  		TABLE 1. MR Stage and Histologic Grade
 
All patients were treated with an antiandrogen (Cyprostat [cyproterone acetate]; Schering Health, Burgess Hill, United Kingdom) in combination with a luteinizing hormone–releasing hormone analog (LH-RHA) (Zoladex [goserelin acetate]; Zeneca, Wilmslow, United Kingdom). Cyproterone acetate (100 mg, orally administered three times daily) was administered for 1 week before and 3 weeks after commencing LH-RHA treatment to prevent possible tumor flare. LH-RHA therapy was administered monthly (3.6 mg, subcutaneous depot injection). Fifteen patients were treated with androgen deprivation alone, and 41 underwent androgen suppression for 3–6 months prior to radical external-beam radiation therapy. No patients underwent prostatectomy. Lymph node status was determined at MR imaging by using criteria established by Vinnicombe et al (28). Bone metastases were demonstrated on technetium 99m diphosphonate bone scans.

MR Imaging
MR imaging was performed before and after hormonal treatment. The median interval between the first (pretreatment) and second (posttreatment) MR examinations was 119 days (range, 49–140 days). The median interval from the beginning of hormonal treatment to the second MR examination was 107 days (range, 47–128 days). All examinations were performed with a 1.5-T magnet (Vision; Siemens Medical Systems, Erlangen, Germany) by using a circularly polarized pelvic phased-array coil. Patients received a bowel relaxant, either hyoscine butylbromide 20 mg (Buscopan; Boehringer Ingelheim, Bracknell, United Kingdom) or glucagon 1 mg (GlucaGen; Nova Nordisk, Crawley, United Kingdom), before imaging.

The imaging protocol comprised routine T1- and T2-weighted imaging through the whole pelvis for the purposes of lymph node staging. Thereafter, small–field-of-view T2-weighted turbo spin-echo sequences (29) were performed in the transverse and coronal planes for local tumor staging (repetition time msec/echo time msec, 3,500–4,000/120; echo train length, 15; 4–5-mm-thick contiguous sections; four signal averages; matrix size, 256 x 150; field of view, 18 [transverse] or 16 cm [coronal]). Images were inspected for the presence of a peripheral zone abnormality consistent with cancer. When no tumor was visible, an anatomic level with an image that demonstrated normal anatomic detail or the central gland was chosen. At this position, a series of quantifiable dynamic contrast-enhanced images were obtained.

The dynamic enhanced protocol included either a single-section spoiled gradient-echo fast low-angle shot (FLASH) (44 examinations) or a five-section saturation recovery turbo FLASH (51 examinations) sequence (30,31). A single proton-density–weighted measurement was acquired as a reference before the series of T1-weighted gradient-echo images was acquired. T1-weighted images were obtained sequentially every 9–10 seconds for 6.3–7.0 minutes. The parameters for the T1-weighted FLASH sequence were as follows: 35/5; flip angle, 70°; one signal acquired; matrix size, 256 x 192; field of view, 25 cm; 10-mm section thickness; acquisition time, 10 seconds; single transverse section; 42 time points; and total imaging time, 7 minutes. For the T1-weighted turbo FLASH sequence, the parameters were as follows: recovery time, 150 msec; 11.7/4.4; flip angle, 20°; one signal acquired; matrix size, 128 x 128; field of view, 20 cm; 8-mm section thickness; acquisition time, 9 seconds; five sections; 42 time points; and total imaging time, 6.3 minutes.

The parameters were altered for proton-density–weighted FLASH (350/5; flip angle, 20°; acquisition time, 1.5 minutes, otherwise as before) and turbo FLASH (recovery time, 10 seconds; 11.7/4.4; acquisition time, 58 seconds, otherwise as before) sequences. Gadopentetate dimeglumine (Magnevist; Schering Health Care, Burgess Hill, United Kingdom) was injected intravenously as a bolus through a peripherally placed canula after the third baseline data point was obtained (0.1 mmol per kilogram of body weight injected within 10 seconds and followed by a 20-mL flush of normal saline).

Image Review and Analysis
Morphologic assessment.—The diagnostic turbo spin-echo T2-weighted images were retrospectively reviewed by two radiologists (A.D.M., J.E.H.) in consensus. The reviewers knew that the patients were being evaluated for prostate cancer but were unaware of the histologic grade, serum PSA levels, or results of dynamic enhanced studies. On the pretreatment turbo spin-echo T2-weighted MR images, the reviewers were asked to identify and stage the primary tumor by using standard 1997 TNM definitions (32). They also noted the signal intensity of the tumor compared with that of the normal peripheral zone of the patient. With the posttreatment images, they were asked to comment on the visibility of the tumor and any change in contrast between the tumor and peripheral zone. Morphologic assessment of the central gland and peripheral zone on pre- and posttreatment images was also performed. In nine patients, the diagnosis of prostate cancer had been made at transurethral resection of the prostate; the central gland of these patients was classified as involved by tumor by default. With the section positions corresponding to the those imaged with the dynamic enhanced sequence, diagrams were completed to indicate (a) the site and outline of the tumor and (b) the areas of normal-appearing peripheral zone and central gland.

On the T2-weighted images, an irregular mass of low signal intensity in the peripheral zone was considered to represent tumor. When an obvious malignant peripheral zone tumor was contiguous with homogeneous low signal intensity in the central gland, the finding was interpreted as probable central glandular involvement. The criteria for extraprostatic spread required one of the following features: disruption of the prostatic capsule by low-signal-intensity tumor, extension of tumor in the periprostatic fat contiguous with low-signal-intensity tumor in the gland, or obliteration of the neurovascular bundle (33). A bulge in the contour of the gland that was large enough to reach the puborectalis muscle was interpreted as probable capsular invasion by tumor. A homogeneous high-signal-intensity peripheral zone was considered normal. Benign prostatic hyperplasia, which arises from the transitional zone, was considered part of the central gland of the prostate (central and transitional zones). The MR imaging features of benign prostatic hyperplasia were uniform areas of low signal intensity (glandular benign prostatic hyperplasia) or areas of nodular whorls with low signal intensity.

Volume measurements.—The prostatic and central glandular volumes were estimated by using the prolate ellipse formula (volume = width x length x anteroposterior diameter x 0.52). When a discrete tumor nodule was seen, its volume was similarly estimated. When multifocal lesions were seen, an easily measurable lesion was chosen as the marker lesion. If a measured lesion was not visible following treatment, its volume was presumed to be zero.

Analysis of dynamic enhanced images.—The dynamic contrast-enhanced FLASH and turbo FLASH images were analyzed at an independent workstation (UltraSPARC 2; Sun Microsystems, Mountain View, Calif) by a radiologist not involved in the hard-copy evaluation of the T2-weighted images. Specialized software designed to quantitatively analyze and display contrast-enhanced dynamic MR data sets was used (34). Each dynamic data set required approximately 20 minutes for analysis. Up to four regions of interest (ROIs) were placed by one author (A.R.P.) on a single section for each set of time-course images (when five-section turbo FLASH images had been obtained [51 examinations], a single section that passed through the center of the tumor was analyzed). By using the information from the diagrammatic review of the turbo spin-echo images, small circular ROIs were placed on the normal peripheral zone and the most enhancing part of the central gland. By using the mean gradient mapping facility of the software, a circular ROI was placed on the fastest enhancing area within the tumor outline. Care was taken when the ROIs were traced to avoid nonenhancing areas. ROIs were manually placed at identical locations on pre- and posttreatment images.

For each ROI, time–signal intensity curves normalized to baseline values were generated, and signal intensity and modeling parameters were derived. We used four time-and–signal intensity parameters to help describe the following important features of the tissue enhancement curve (30).

1. Onset time was defined as the period between the bolus injection of contrast agent (taken as the middle of the fourth data point on the dynamic enhanced images) and the point on the enhancement curve at which the signal intensity first exceeded 10% of its maximum signal intensity (see later).

2. Mean gradient was defined as the mean rate of change of the relative signal intensity between the 10% and 90% points of maximum enhancement.

3. Maximum signal intensity was defined as the peak level of signal intensity during the dynamic measurement period.

4. Washout score was defined as type A when a monophasic increase in signal intensity was seen throughout the observation period, as type B if the peak signal intensity was achieved within the first 2 minutes and sustained or if there was a late decrease in signal intensity (washout), or as type C when an early peak of enhancement was seen and followed immediately by a decrease in signal intensity (35).

Quantitative modeling parameters, including the permeability–surface area product per unit volume of tissue Kp, leakage space as a percentage of unit volume of tissue v1, and maximum tissue gadolinium accumulation (tissue contrast medium concentration) were then calculated by applying a multicompartment model analysis to the tissue time–gadolinium concentration curve. The images obtained with the dynamic enhancement protocol were used to derive measurements of contrast agent concentration in vivo.

The longitudinal relaxation rate, T1, of the water protons at each time point in the dynamic enhanced sequence was obtained from the ratio of each T1-weighted image to the baseline proton-density–weighted image (30). Once the T1 of each pixel within each image is calculated, the concentration of contrast agent can be obtained (36). The model we used assumes four compartments within the body to which the contrast agent has access (37,38). These are the whole-body vasculature, whole-body extracellular space, kidneys (ultimate washout pathway), and lesion leakage space (effectively, lesion extracellular space). The permeability–surface area product Kp refers to the transfer constant between blood plasma and the extracellular space, and leakage space v1 refers to the volume of the extravascular-extracellular space. For further information on these modeling techniques, the readers are referred to articles by Parker et al (30,31,34). Note that the term "permeability–surface area product" has recently been replaced by the term "transfer constant Ktrans)," and the term "leakage space" has been replaced by "extravascular-extracellular space ve" (39).

Statistical Methods
The enhancement parameters of the different tissues before and after treatment were compared by using descriptive statistics and nonparametric methods (Wilcoxon rank sum test; Arcus Quickstat Biomedical, Cambridge, United Kingdom). Correlation of MR enhancement parameters and baseline serum PSA levels was performed by using linear regression analysis. Categorical data were analyzed by using the {chi}2 test. A probability value of less than 5% (P < .05) was considered statistically significant.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
 
The serum PSA level was reduced in all patients from a median of 16.6 ng/mL (range, 1.0–98.5 ng/mL) before hormonal treatment to 1.0 ng/mL (range, 0.3–56.4 ng/mL) after hormonal treatment (median reduction, 92%; range, 28%–99%) (Fig 1). Three patients had serum PSA levels in the normal range (<4.0 ng/mL) at diagnosis.



View larger version (40K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Graph shows changes in serum PSA levels in 56 patients treated with androgen deprivation. Reduction in PSA level was seen in every patient (median decrease, 92%; range, 28%-99%).

 
Morphologic Changes
In 56 patients, 55 tumors were visible on T2-weighted images before treatment (multiple lesions in 22 patients). The signal intensity of the carcinoma compared with that of the adjacent peripheral zone before and after hormonal therapy is shown in Table 2 and in Figure 2. The tumor-to–peripheral zone contrast decreased in 42 (76%) patients, increased in one (2%), and was not changed in 12 (22%). As a result, 42 tumors were visible after hormonal treatment. Tumor margins were clearly delineated, which enabled tumor volume measurement in 41 patients before treatment but in 27 after treatment.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Signal Intensity of Prostate Cancer versus Adjacent Peripheral Zone on T2-weighted Images before and after Androgen Deprivation
 


View larger version (184K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a. Morphologic changes after androgen deprivation in a 62-year-old man with prostate cancer (Gleason grade 2 + 2). (a) Transverse T2-weighted turbo spin-echo MR image (3,600/120) before treatment (PSA level, 11.5 ng/mL) shows a low-signal-intensity mass, which is compatible with tumor (straight solid arrow), in the left peripheral zone with homogenous intermediate to high signal intensity (curved arrow). The central gland is morphologically normal (open arrow). (b) Transverse T2-weighted turbo spin-echo image (3,500/120) obtained after 121 days of androgen deprivation (PSA level, 0.4 ng/mL) shows that glandular volume has reduced by 40%. The whole gland has lower signal intensity, with poor zonal differentiation, and the primary tumor is not visible.

 


View larger version (177K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b. Morphologic changes after androgen deprivation in a 62-year-old man with prostate cancer (Gleason grade 2 + 2). (a) Transverse T2-weighted turbo spin-echo MR image (3,600/120) before treatment (PSA level, 11.5 ng/mL) shows a low-signal-intensity mass, which is compatible with tumor (straight solid arrow), in the left peripheral zone with homogenous intermediate to high signal intensity (curved arrow). The central gland is morphologically normal (open arrow). (b) Transverse T2-weighted turbo spin-echo image (3,500/120) obtained after 121 days of androgen deprivation (PSA level, 0.4 ng/mL) shows that glandular volume has reduced by 40%. The whole gland has lower signal intensity, with poor zonal differentiation, and the primary tumor is not visible.

 
The central gland showed a decrease in signal intensity on T2-weighted images with treatment in 18 patients, did not alter in 28 patients, increased in one, and was not assessed in the nine patients who had prostate carcinoma diagnosed at transurethral resection of the prostate. In addition, a more homogeneous central gland with lowered signal intensity was noted in 19 (40%) of 47 of these patients after hormone treatment. The signal intensity of the seminal vesicles was compared with that of adjacent pelvic fat. The signal intensity of the seminal vesicles decreased in 25 (45%) patients, increased in one (2%), and remained unaltered in 30 (54%). A decrease in the size of the seminal vesicles was also noted in 30 (54%) patients.

The posttreatment volumes in the whole prostate and central gland (median, 24.7 mL; range, 7.0–99.8 mL and median, 15.3 mL; range, 2.7–83.9 mL, respectively) were significantly smaller than the pretreatment values (median, 42 mL; range, 15.2–131.7 mL and median 22.2 mL; range, 6.3–105.1 mL, respectively; P < .001). The tumor volume was also reduced with treatment (before treatment: median, 3.0 mL; range, 0.6–19.9 mL; after treatment: median, 0.6 mL; range, 0–13.7 mL; P < .001) (Fig 3). One patient showed an increase in tumor volume of 12%, and in 14 patients, tumor volume could not be measured and was assumed to be zero. Thus, the median reduction in the volumes of the whole prostate, central gland, and tumor were 36% (95% CI: 33%, 40%), 30% (95% CI: 25%, 35%) and 65% (95% CI: 55%, 76%), respectively.



View larger version (33K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Graph shows prostate tumor volume changes in 41 patients treated with androgen deprivation. Reduction in tumor volume was seen in 40 patients. One patient showed an increase in tumor volume of 12%, and in 14 patients, the tumor was not measurable after treatment and the volume was assumed to be zero.

 
Contrast-enhanced Parameter Changes
Pre- and/or posttreatment dynamic contrast-enhanced MR images were analyzed in 38 patients. The following events caused contrast-enhanced images of 18 patients to be excluded from analysis (in two patients, more than one event occurred): patient or internal organ movement (rectum) in eight patients, a poor contrast medium bolus in eight patients, no contrast medium administration in four patients, and technical failure in two patients (poor signal-to-noise ratio in one and unexpected machine gain changes in another).

In 38 patients in whom pretreatment dynamic MR images were available, we noted 36 tumors, 28 regions of normal peripheral zone, and 22 regions of normal central gland in the dynamic section plane. Enhancement parameter differences in pre- and posttreatment groups and individual patient changes were analyzed.

Tumor.—Group differences are presented in Table 3, and changes observed in individual patients are summarized in Table 4 (32 observations). The mean size of the regions of interest was 10 pixels (95% CI: 7, 13) before treatment and 9 pixels (95% CI: 7, 10) after treatment.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Comparison of Enhancement Parameters in 36 Prostate Tumors Treated with Androgen Deprivation
 

View this table:
[in this window]
[in a new window]

  		TABLE 4. Changes in Quantitative Enhancement Parameters in Prostatic Tissue after Androgen Deprivation
 
There were no differences in the onset times between the pre- and posttreatment groups (Table 3). The mean gradient and maximal amplitude of enhancement after treatment were significantly less than the pretreatment values. Marked differences in washout patterns were also noted: Before treatment, type B or C washout patterns were seen in 34 (94%) tumors, but after treatment the majority of tumors (n = 28 [78%]) showed a type A pattern. No patients had a type C pattern of washout following treatment.

We also noted significant differences in the capillary permeability and maximum tissue gadolinium accumulation, with no alteration in the leakage space. Significant reductions in tumor capillary permeability were observed (median, 56%; 95% CI: 42%, 64%) (Table 4; Figs 4, 5). A reduction in tumor permeability was seen in 29 of 32 cases and coincided with a reduction in serum PSA levels. In three cases, reductions in serum PSA levels and tumor permeability were discordant (Fig 6). No direct correlation between the magnitude of reduction of serum PSA level and tumor permeability was observed (r = 0.17; P > .05). No direct correlation was observed between tumor volume and permeability change (r = 0.20; P > .05) (Fig 7). However, when the 10 patients with unmeasurable tumors were excluded (ie, those in whom tumor volume responses were assumed to be 100%), a statistically significant correlation was seen (r = 0.55; P = .04) (Fig 7).



View larger version (40K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Graph shows changes in tumor permeability in 32 patients treated with androgen deprivation. Reduction in permeability was seen in 29 tumors (median, 56%; range, 21%-93%); three showed no change (±10%) in tumor permeability, and two had an increase (12% and 15%).

 


View larger version (172K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a. Permeability changes after androgen deprivation in 62-year-old man with prostate cancer (Gleason grade 3 + 4). (a) Transverse T2-weighted turbo spin-echo MR image (3,600/120) before treatment (PSA level, 6.0 ng/mL) shows a low-signal-intensity mass in the right peripheral zone (straight arrow). Peripheral zone in the left side of the gland appears normal (curved arrow). (b) Time-gadopentetate dimeglumine concentration curves from ROIs in the tumor ({lozenge}), central gland ({bullet}), and peripheral zone ({blacksquare}) before LH-RHA treatment. Calculated contrast concentrations are discontinuous data points, according to tissue type, and corresponding curves are the model fit to the data points. Calculated kinetic parameters—permeability, leakage space, and maximum contrast agent concentration—were as follows: in tumor, 0.52 min-1, 67%, and 0.41 mmol/kg; in the central gland, 0.81 min-1, 74%, and 0.47 mmol/kg; and in the peripheral zone, 0.15 min-1, 23%, and 0.15 mmol/kg. (c) Permeability map superimposed on transverse T1-weighted FLASH MR image (35/5; flip angle, 70°) before treatment (maximum permeability depicted = 1 min-1). Higher levels of capillary permeability (orange and yellow) are seen in the tumor and central gland. (d) Transverse T2-weighted turbo spin-echo MR image (3,600/120) after 123 days of androgen deprivation (PSA level, 1.2 ng/mL). Section position relative to surrounding bone and soft tissue is not the same because prostate glandular volume has reduced by 46%. Tumor (straight arrow) and normal peripheral zone (curved arrow) are still visible. Morphologically, the tumor was judged as stable disease. (e) Time-gadopentetate dimeglumine concentration curves and corresponding model fits after LH-RHA treatment. Calculated kinetic parameters—permeability, leakage space, and maximum contrast agent concentration—were as follows: in tumor ({lozenge}), 0.31 per minute, 56%, and 0.34 mmol/kg; in the central gland ({bullet}), 0.31 per minute, 58%, and 0.35 mmol/kg; and in the peripheral zone ({blacksquare}), 0.04 per minute, 51%, and 0.13 mmol/kg. (f) Permeability map superimposed on anatomic T1-weighted FLASH MR image (35 msec/5 msec; flip angle, 70°) after treatment (maximum permeability depicted = 1 min-1). Permeability has decreased in all prostate tissues (deep blue and red).

 


View larger version (40K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b. Permeability changes after androgen deprivation in 62-year-old man with prostate cancer (Gleason grade 3 + 4). (a) Transverse T2-weighted turbo spin-echo MR image (3,600/120) before treatment (PSA level, 6.0 ng/mL) shows a low-signal-intensity mass in the right peripheral zone (straight arrow). Peripheral zone in the left side of the gland appears normal (curved arrow). (b) Time-gadopentetate dimeglumine concentration curves from ROIs in the tumor ({lozenge}), central gland ({bullet}), and peripheral zone ({blacksquare}) before LH-RHA treatment. Calculated contrast concentrations are discontinuous data points, according to tissue type, and corresponding curves are the model fit to the data points. Calculated kinetic parameters—permeability, leakage space, and maximum contrast agent concentration—were as follows: in tumor, 0.52 min-1, 67%, and 0.41 mmol/kg; in the central gland, 0.81 min-1, 74%, and 0.47 mmol/kg; and in the peripheral zone, 0.15 min-1, 23%, and 0.15 mmol/kg. (c) Permeability map superimposed on transverse T1-weighted FLASH MR image (35/5; flip angle, 70°) before treatment (maximum permeability depicted = 1 min-1). Higher levels of capillary permeability (orange and yellow) are seen in the tumor and central gland. (d) Transverse T2-weighted turbo spin-echo MR image (3,600/120) after 123 days of androgen deprivation (PSA level, 1.2 ng/mL). Section position relative to surrounding bone and soft tissue is not the same because prostate glandular volume has reduced by 46%. Tumor (straight arrow) and normal peripheral zone (curved arrow) are still visible. Morphologically, the tumor was judged as stable disease. (e) Time-gadopentetate dimeglumine concentration curves and corresponding model fits after LH-RHA treatment. Calculated kinetic parameters—permeability, leakage space, and maximum contrast agent concentration—were as follows: in tumor ({lozenge}), 0.31 per minute, 56%, and 0.34 mmol/kg; in the central gland ({bullet}), 0.31 per minute, 58%, and 0.35 mmol/kg; and in the peripheral zone ({blacksquare}), 0.04 per minute, 51%, and 0.13 mmol/kg. (f) Permeability map superimposed on anatomic T1-weighted FLASH MR image (35 msec/5 msec; flip angle, 70°) after treatment (maximum permeability depicted = 1 min-1). Permeability has decreased in all prostate tissues (deep blue and red).

 


View larger version (147K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5c. Permeability changes after androgen deprivation in 62-year-old man with prostate cancer (Gleason grade 3 + 4). (a) Transverse T2-weighted turbo spin-echo MR image (3,600/120) before treatment (PSA level, 6.0 ng/mL) shows a low-signal-intensity mass in the right peripheral zone (straight arrow). Peripheral zone in the left side of the gland appears normal (curved arrow). (b) Time-gadopentetate dimeglumine concentration curves from ROIs in the tumor ({lozenge}), central gland ({bullet}), and peripheral zone ({blacksquare}) before LH-RHA treatment. Calculated contrast concentrations are discontinuous data points, according to tissue type, and corresponding curves are the model fit to the data points. Calculated kinetic parameters—permeability, leakage space, and maximum contrast agent concentration—were as follows: in tumor, 0.52 min-1, 67%, and 0.41 mmol/kg; in the central gland, 0.81 min-1, 74%, and 0.47 mmol/kg; and in the peripheral zone, 0.15 min-1, 23%, and 0.15 mmol/kg. (c) Permeability map superimposed on transverse T1-weighted FLASH MR image (35/5; flip angle, 70°) before treatment (maximum permeability depicted = 1 min-1). Higher levels of capillary permeability (orange and yellow) are seen in the tumor and central gland. (d) Transverse T2-weighted turbo spin-echo MR image (3,600/120) after 123 days of androgen deprivation (PSA level, 1.2 ng/mL). Section position relative to surrounding bone and soft tissue is not the same because prostate glandular volume has reduced by 46%. Tumor (straight arrow) and normal peripheral zone (curved arrow) are still visible. Morphologically, the tumor was judged as stable disease. (e) Time-gadopentetate dimeglumine concentration curves and corresponding model fits after LH-RHA treatment. Calculated kinetic parameters—permeability, leakage space, and maximum contrast agent concentration—were as follows: in tumor ({lozenge}), 0.31 per minute, 56%, and 0.34 mmol/kg; in the central gland ({bullet}), 0.31 per minute, 58%, and 0.35 mmol/kg; and in the peripheral zone ({blacksquare}), 0.04 per minute, 51%, and 0.13 mmol/kg. (f) Permeability map superimposed on anatomic T1-weighted FLASH MR image (35 msec/5 msec; flip angle, 70°) after treatment (maximum permeability depicted = 1 min-1). Permeability has decreased in all prostate tissues (deep blue and red).

 


View larger version (171K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5d. Permeability changes after androgen deprivation in 62-year-old man with prostate cancer (Gleason grade 3 + 4). (a) Transverse T2-weighted turbo spin-echo MR image (3,600/120) before treatment (PSA level, 6.0 ng/mL) shows a low-signal-intensity mass in the right peripheral zone (straight arrow). Peripheral zone in the left side of the gland appears normal (curved arrow). (b) Time-gadopentetate dimeglumine concentration curves from ROIs in the tumor ({lozenge}), central gland ({bullet}), and peripheral zone ({blacksquare}) before LH-RHA treatment. Calculated contrast concentrations are discontinuous data points, according to tissue type, and corresponding curves are the model fit to the data points. Calculated kinetic parameters—permeability, leakage space, and maximum contrast agent concentration—were as follows: in tumor, 0.52 min-1, 67%, and 0.41 mmol/kg; in the central gland, 0.81 min-1, 74%, and 0.47 mmol/kg; and in the peripheral zone, 0.15 min-1, 23%, and 0.15 mmol/kg. (c) Permeability map superimposed on transverse T1-weighted FLASH MR image (35/5; flip angle, 70°) before treatment (maximum permeability depicted = 1 min-1). Higher levels of capillary permeability (orange and yellow) are seen in the tumor and central gland. (d) Transverse T2-weighted turbo spin-echo MR image (3,600/120) after 123 days of androgen deprivation (PSA level, 1.2 ng/mL). Section position relative to surrounding bone and soft tissue is not the same because prostate glandular volume has reduced by 46%. Tumor (straight arrow) and normal peripheral zone (curved arrow) are still visible. Morphologically, the tumor was judged as stable disease. (e) Time-gadopentetate dimeglumine concentration curves and corresponding model fits after LH-RHA treatment. Calculated kinetic parameters—permeability, leakage space, and maximum contrast agent concentration—were as follows: in tumor ({lozenge}), 0.31 per minute, 56%, and 0.34 mmol/kg; in the central gland ({bullet}), 0.31 per minute, 58%, and 0.35 mmol/kg; and in the peripheral zone ({blacksquare}), 0.04 per minute, 51%, and 0.13 mmol/kg. (f) Permeability map superimposed on anatomic T1-weighted FLASH MR image (35 msec/5 msec; flip angle, 70°) after treatment (maximum permeability depicted = 1 min-1). Permeability has decreased in all prostate tissues (deep blue and red).

 


View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5e. Permeability changes after androgen deprivation in 62-year-old man with prostate cancer (Gleason grade 3 + 4). (a) Transverse T2-weighted turbo spin-echo MR image (3,600/120) before treatment (PSA level, 6.0 ng/mL) shows a low-signal-intensity mass in the right peripheral zone (straight arrow). Peripheral zone in the left side of the gland appears normal (curved arrow). (b) Time-gadopentetate dimeglumine concentration curves from ROIs in the tumor ({lozenge}), central gland ({bullet}), and peripheral zone ({blacksquare}) before LH-RHA treatment. Calculated contrast concentrations are discontinuous data points, according to tissue type, and corresponding curves are the model fit to the data points. Calculated kinetic parameters—permeability, leakage space, and maximum contrast agent concentration—were as follows: in tumor, 0.52 min-1, 67%, and 0.41 mmol/kg; in the central gland, 0.81 min-1, 74%, and 0.47 mmol/kg; and in the peripheral zone, 0.15 min-1, 23%, and 0.15 mmol/kg. (c) Permeability map superimposed on transverse T1-weighted FLASH MR image (35/5; flip angle, 70°) before treatment (maximum permeability depicted = 1 min-1). Higher levels of capillary permeability (orange and yellow) are seen in the tumor and central gland. (d) Transverse T2-weighted turbo spin-echo MR image (3,600/120) after 123 days of androgen deprivation (PSA level, 1.2 ng/mL). Section position relative to surrounding bone and soft tissue is not the same because prostate glandular volume has reduced by 46%. Tumor (straight arrow) and normal peripheral zone (curved arrow) are still visible. Morphologically, the tumor was judged as stable disease. (e) Time-gadopentetate dimeglumine concentration curves and corresponding model fits after LH-RHA treatment. Calculated kinetic parameters—permeability, leakage space, and maximum contrast agent concentration—were as follows: in tumor ({lozenge}), 0.31 per minute, 56%, and 0.34 mmol/kg; in the central gland ({bullet}), 0.31 per minute, 58%, and 0.35 mmol/kg; and in the peripheral zone ({blacksquare}), 0.04 per minute, 51%, and 0.13 mmol/kg. (f) Permeability map superimposed on anatomic T1-weighted FLASH MR image (35 msec/5 msec; flip angle, 70°) after treatment (maximum permeability depicted = 1 min-1). Permeability has decreased in all prostate tissues (deep blue and red).

 


View larger version (149K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5f. Permeability changes after androgen deprivation in 62-year-old man with prostate cancer (Gleason grade 3 + 4). (a) Transverse T2-weighted turbo spin-echo MR image (3,600/120) before treatment (PSA level, 6.0 ng/mL) shows a low-signal-intensity mass in the right peripheral zone (straight arrow). Peripheral zone in the left side of the gland appears normal (curved arrow). (b) Time-gadopentetate dimeglumine concentration curves from ROIs in the tumor ({lozenge}), central gland ({bullet}), and peripheral zone ({blacksquare}) before LH-RHA treatment. Calculated contrast concentrations are discontinuous data points, according to tissue type, and corresponding curves are the model fit to the data points. Calculated kinetic parameters—permeability, leakage space, and maximum contrast agent concentration—were as follows: in tumor, 0.52 min-1, 67%, and 0.41 mmol/kg; in the central gland, 0.81 min-1, 74%, and 0.47 mmol/kg; and in the peripheral zone, 0.15 min-1, 23%, and 0.15 mmol/kg. (c) Permeability map superimposed on transverse T1-weighted FLASH MR image (35/5; flip angle, 70°) before treatment (maximum permeability depicted = 1 min-1). Higher levels of capillary permeability (orange and yellow) are seen in the tumor and central gland. (d) Transverse T2-weighted turbo spin-echo MR image (3,600/120) after 123 days of androgen deprivation (PSA level, 1.2 ng/mL). Section position relative to surrounding bone and soft tissue is not the same because prostate glandular volume has reduced by 46%. Tumor (straight arrow) and normal peripheral zone (curved arrow) are still visible. Morphologically, the tumor was judged as stable disease. (e) Time-gadopentetate dimeglumine concentration curves and corresponding model fits after LH-RHA treatment. Calculated kinetic parameters—permeability, leakage space, and maximum contrast agent concentration—were as follows: in tumor ({lozenge}), 0.31 per minute, 56%, and 0.34 mmol/kg; in the central gland ({bullet}), 0.31 per minute, 58%, and 0.35 mmol/kg; and in the peripheral zone ({blacksquare}), 0.04 per minute, 51%, and 0.13 mmol/kg. (f) Permeability map superimposed on anatomic T1-weighted FLASH MR image (35 msec/5 msec; flip angle, 70°) after treatment (maximum permeability depicted = 1 min-1). Permeability has decreased in all prostate tissues (deep blue and red).

 


View larger version (23K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6. Scatterplot shows reductions in serum PSA levels and tumor permeability in 32 patients. Reduction in PSA level is seen in 29 (91%) patients with an associated reduction in tumor permeability (exceptions within oval). No correlation in the magnitude of change in PSA level and permeability is seen (r = 0.17; P > .05).

 


View larger version (26K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7. Scatterplot shows reductions in tumor volume and permeability in 24 patients. Positive correlation (r = 0.55; P = .04) between reductions in tumor volume and permeability is seen when inaccessible tumors are excluded (within ellipse). (These were presumed to have complete response, ie, 100% volume reduction.)

 
Peripheral zone.—The mean sizes of the regions of interest were 19 pixels (95% CI: 13, 24) before treatment (28 observations) and 14 pixels (95% CI: 11, 17) after treatment (27 observations). There were no differences in signal intensity parameters or washout patterns between the pre- and posttreatment groups. However, capillary permeability was reduced in the posttreatment group (median, 31%; 95% CI: 10%, 50%). This reduction was accompanied by an increase in the leakage space (median, 46%; 95% CI: 10%, 100%) (Table 4). In general, the decrease in permeability of the tumor was greater than that of the peripheral zone in the same patients (noted in 16 of 25 patients). The mean decrease in peripheral zone permeability was 20% (95% CI: 2%, 40%), compared with 50% in tumors (95% CI: 30%, 60%; P = .08).

Central gland.—The mean size of the regions of interest was 30 pixels (95% CI: 18, 43) before treatment (22 observations) and 22 pixels (95% CI: 14, 30) after treatment (21 observations). Marked changes in the washout patterns were noted. Before treatment, a type C pattern was seen in 12 patients, but this pattern was observed in only two after treatment (P = .005). A reduction in capillary permeability and an increase in the leakage space were also observed (Table 4).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
REFERENCES
 
Morphologic Changes
The effects of hormonal treatment on prostate gland and tumor volumes can be evaluated with estimations of serum PSA levels, digital rectal examination, transrectal US, and MR imaging (11,19–22,40). Clinical evaluations and imaging studies all show substantial reductions in glandular size and tumor volume. Shearer et al (19) reported that, in the majority of patients, the glandular volume stabilizes at 13 weeks, although larger prostates require up to 6 months to obtain maximum volume reductions. Kojima et al (40) reported that the maximum effect of endocrine treatment occurs at 3–4 months. A 10%–52% reduction in whole prostate volume has been widely reported (6,19–22,41). Glandular volume reduction has a direct treatment benefit because it reduces the radiation therapy target volume and reduces irradiation of the bladder and rectum (16,42,43). In agreement with these findings, our findings show a reduction in volume of the prostate gland of 36% after a median 119 days of treatment.

It is known that LH-RHAs can ameliorate symptoms related to benign prostatic hyperplasia (44), and previous MR imaging findings (21,22) have shown reductions of 30% in the volume of the central gland. Our study findings confirm this percentage (mean, 30%; 95% CI: 25%, 35%). Reductions in tumor volume have also been reported (20–22) at transrectal US and MR imaging, with reductions of 20%–97%. We have shown a similar tumor volume reduction of 65% (95% CI: 55%, 76%). It is likely that the calculated reduction in tumor volume is overestimated because fewer tumors were visible and measurable after treatment (55 visible and 41 measurable before treatment vs 42 visible and 28 measurable after treatment); when tumors were not visible, their volume was assumed to be zero (21,22).

In keeping with the MR findings of Chen et al (21) and Nakashima et al (22), we noted that the central gland decreased in signal intensity (38%) and became more homogeneous (40%) with treatment. We also noted that androgen deprivation resulted in a decrease in the size of the seminal vesicle (54%), and a reduction in signal intensity was also noted (21,45). Hormonal ablation also reduced the number of tumors that were detectable at MR imaging. Fifty-five tumors were visible on T2-weighted images before treatment, and 42 tumors were visible after treatment. This occurred because the peripheral zone decreased in signal intensity, reducing tumor–peripheral zone contrast (42 patients). Chen et al (21) and Nakashima et al (22) also reported a similar reduction in the number of tumors detectable at MR imaging after treatment.

The morphologic appearances we observed can be explained by the distinctive histologic changes that occur in patients treated with LH-RHAs (46,47). Histologic findings in prostate glands removed after androgen deprivation show marked glandular shrinkage. The histologic LH-RHA effect is characterized by a reduction in glandular size and density, compression of the glandular lumina, and increased periglandular fibrous tissue. Nuclei are often rounded and without large nucleoli. The glandular cytoplasm is scant and pale and may be vacuolated. There is little cellular degeneration or necrosis observed with LH-RHA treatment (46). Squamous metaplasia and cellular degeneration that is observed after high-dose estrogen treatment is not a prominent feature (46,48). Tissue shrinkage, glandular atrophy, and fibrosis would be expected to result in a smaller, darker prostate gland with poor zonal differentiation on T2-weighted images.

Permeability Changes
Dynamic contrast-enhanced MR imaging can be used to noninvasively assess functional aspects of the microcirculation of tissues. Such assessments may be qualitative or quantitative. Qualitative descriptors are derived from tissue time–signal intensity curves, and quantitative measurements (eg, permeability–surface area product and lesion leakage space) are derived from modelling tissue time-and–contrast agent concentration changes. Quantitative measurements have the advantages of being independent of imager variation and imaging routines and thus allow valid comparisons of measurements acquired serially and between different imaging sites.

It is generally recognized that dynamic contrast-enhanced MR imaging can be used to characterize tissues and that the onset time and rate of enhancement are valuable parameters in the differentiation of malignant from benign lesions. Such studies (49–54) have been undertaken to evaluate tumors in the breast, bladder, bone, and prostate. This technique has been used to monitor the response of breast, cervix, bladder, and bone tumors to chemotherapy and radiation therapy (55–58), but we are not aware of any reports of MR permeability changes in prostate cancer treated with androgen deprivation.

A decrease in permeability occurred in all prostatic tissues: tumor (median, 56%), central gland (median, 40%), and peripheral zone (median, 31%). This finding is consistent with recent observations made by Okihara et al (59), who showed a decrease in tumor vascular resistance at Doppler US in 11 patients with prostate cancer treated with castration. These findings may be explained by the fibrotic histologic changes in patients treated with androgen deprivation (46,47). Fibrosis is known to decrease tissue capillary permeability to contrast medium and is the basis for the distinction that can be made between postoperative fibrosis and tumor recurrence at dynamic contrast-enhanced MR imaging (60,61).

In addition, the reduction in permeability of all prostatic tissues with androgen deprivation may relate to down-regulation of VEGF production. VEGF is strong stimulus of tumor neoangiogenesis and a potent tissue permeability factor (25). Recently, it has been shown (62,63) that VEGF is produced in abundance by the secretory epithelium of normal, hyperplastic, and tumorous prostate glands. The physiologic role of VEGF in the prostate gland is poorly understood, and target cells may include cells other than vascular endothelium.

With respect to the vasculature, it is clear that VEGF is required for vascular homeostasis and maintains the high fraction of immature vessels within the prostate gland. Immature vessels (those without investing pericytes and/or smooth muscle cells) are highly dependent on exogenous survival factors, including VEGF (64). In the prostate, VEGF production requires continual stimulation by androgens (26); VEGF expression in androgen-dependent cell lines is down-regulated with androgen withdrawal, and prostate tumors from these cell lines undergo vascular regression prior to tumor cell death (27).

Findings from a recent histologic study by Matsushima et al (65) appear to contradict this view. They showed that the maximum intratumoral microvascular density does not appear to differ in patients treated with neoadjuvant hormonal deprivation compared with those who did not and thus suggested that tumor angiogenesis may not be influenced by hormonal manipulation. These findings did show that androgen ablation treatment decreased proliferative activity and enhanced apoptosis of prostate cancer cells. Therefore, it is possible that the lack of a demonstrable difference in intratumoral microvascular density occurred as a direct result of the concomitant decrease in tumor size (a decrease in the number of vessels and size of tumor may result in no change in intratumoral microvascular density). Furthermore, these findings could not be used to distinguish immature from mature vessels, and the selective ablation of immature vessels by androgen deprivation described by Benjamin et al (64) would therefore not have been observed.

In 91% of our patients, we observed that a reduction in tumor permeability coincided with a reduction in serum PSA levels (Fig 6). In three cases, the reductions in PSA level and tumor permeability were discordant. There was no direct correlation between the magnitude of reduction of serum PSA level and tumor permeability. It should be noted that the serum PSA level is not cancer specific; rather, it is only prostate specific. In this study, we did not normalize individual PSA levels with estimations of the volume of the prostate gland because this would have been inappropriate for patients with lymph node disease and those with metastases (16 patients) who would have had extraprostatic sources of PSA production. The PSA level is a reliable indicator of response to hormonal treatment, but because the expression of PSA is regulated by androgens (66–68), the reduction of serum PSA level is in part due to down-regulation of the PSA gene independent of cell death (69).

A weak correlation was seen between the change in tumor permeability and volume (r = 0.55; P = .04) when nonassessable tumors after treatment are excluded (Fig 7). This finding lends weight to the hypothesis that MR permeability can be used to assess the vascular effect of LH-RHA that is independent of the mechanisms involved in alterations of serum PSA levels and tumor volume. Thus, tumor size, MR permeability, and measurements of serum PSA levels may be alternate means of monitoring prostatic tumor response to androgen deprivation.

Study Limitations
A limitation of our study was that no direct histologic confirmation was obtained by using prostatectomy specimens. Neoadjuvant androgen deprivation has been shown to improve the results of radical radiation therapy (7–10) and is routinely used at our institution. Initial hormone treatment does not benefit patients undergoing total prostatectomy (13–15), and a surgical treatment option would not have been feasible in 36 of our patients who had T2 disease with metastases, T3 disease, or T4 disease. Histologic correlation of MR findings is also recognized as an imperfect standard for a number of reasons. These include errors in the registration of the location of the imaging sections with histologic section specimens, inaccuracies resulting from tissue shrinkage secondary to fixation, and partial volume averaging effects (70–72).

We did not use a balloon-inflated endorectal coil for MR examinations, because this method is more likely to produce artifacts compared with use of pelvic phased-array coils (73); these artifacts include near-field flare and other coil-related artifacts. Rectal movements are also common in patients examined with endorectal coils and can be observed in most patients on careful review. The previous artifacts would have increased the failure rate of our dynamic contrast-enhanced MR evaluations. Furthermore, prostate gland distortion commonly occurs, and the anterior part of the gland is poorly shown when endorectal examinations are performed.

Other limitations relate to the lack of exact registration on the pre- and posttreatment dynamic contrast-enhanced MR images. When the studies were performed, we did not have automated registration software, and an experienced radiologist was asked to visually inspect the anatomic images to determine the section position of the dynamic contrast-enhanced MR image. Similarly, the anatomic placement of ROIs on pre- and posttreatment MR images may not have been exactly the same due to glandular shrinkage and changes in the signal intensity of the prostate gland. This source of error was minimized by evaluating anatomic images at consensus review, by evaluating the pre- and posttreatment dynamic contrast-enhanced MR data at the same sitting (to reduce intraobserver variability), and by recording the exact sites of the ROIs diagrammatically. Confirmation of ROI statistics was obtained by visually inspecting appropriate parametric maps for consistency.

It should also be recognized that the kinetic model we used to derive the tumor permeability–surface area product assumes that the amount of contrast agent leaving the blood vessels due to leakiness is small compared with the arterial supply of contrast agent. In tumors, this may is not necessarily the case since permeability is often high, and, therefore, the measured permeability may also reflect local perfusion to a (variable, possibly small) degree (39).

We therefore conclude that MR imaging demonstrates reductions in tumor volume and vascular permeability in prostate cancers responding to androgen deprivation. Hormonal treatment also reduces the number of tumors detectable at MR imaging. Morphologic and vascular permeability changes are seen in both tumor and normal glandular tissues. The positive correlation of permeability and volume changes and the lack of correlation with changes in serum PSA level suggest the presence of independent effects of androgen deprivation treatment (morphologic, biochemical, and vascular). MR quantification of capillary permeability may therefore allow an alternate means of monitoring tumor response, even when morphologic changes are difficult to discern; however, no special advantage over the change in PSA level was noted.


   FOOTNOTES
 
2 Current address: Methodist Hospital, Indianapolis, Ind. Back

3 Current address: NMR Research Unit, Institute of Neurology, London, England. Back

4 Current address: Clinical Age Research Unit, King’s College Hospital, School of Medicine and Dentistry, London, England. Back

Abbreviations: FLASH = fast low-angle shot, LH-RHA = luteinizing hormone–releasing hormone analog, PSA = prostate-specific antigen, ROI = region of interest, VEGF = vascular endothelial growth factor

Author contributions: Guarantor of integrity of entire study, J.E.H.; study concepts and design, A.R.P., J.E.H.; definition of intellectual content, A.R.P., J.E.H., M.O.L., D.P.D.; literature research, A.R.P., D.P.D.; clinical studies, J.E.H., A.R.P., A.D.M.; data acquisition, C.J.G.; data analysis, A.R.P., C.J.G., G.J.M.P., J.S.; statistical analysis, A.R.P., manuscript preparation, A.R.P., J.E.H.; manuscript editing, M.O.L., A.D.M.; manuscript review, A.R.P., J.E.H., D.P.D.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
REFERENCES
 

  1. Chodak GW, Thisted RA, Gerber GS, et al. Results of conservative management of clinically localized prostate cancer. N Engl J Med 1994; 330:242-248.[Abstract/Free Full Text]
  2. Van Poppel H, de Ridder D, Goethuys H, Elgamal AA, Baert L. Surgical treatment of carcinoma of the prostate. In: Petrovich Z, Baert L, Brady LW, eds. Carcinoma of the prostate: innovations in management. New York, NY: Springer-Verlag, 1996.
  3. Zagars GK, von Eschenbach AC, Johnson DE, Oswald MJ. The role of radiation therapy in stages A2 and B adenocarcinoma of the prostate. Int J Radiat Oncol Biol Phys 1988; 14:701-709.[Medline]
  4. Fair WR, Aprikian A, Sogani P, Reuter V, Whitmore WF. The role of neoadjuvant hormonal manipulation in localized prostatic cancer. Cancer 1993; 71(3 suppl):1031-1038.[Medline]
  5. Macfarlane MT, Abi-Aad A, Stein A, Danella J, Belldegrun A, deKernion JB. Neoadjuvant hormonal deprivation in patients with locally advanced prostate cancer. J Urol 1993; 150:132-134.[Medline]
  6. Andros EA, Danesghari F, Crawford ED. Neoadjuvant hormonal therapy in stage C adenocarcinoma of the prostate. Clin Invest Med 1993; 16:510-515.[Medline]
  7. Laverdiere J, Gomez JL, Cusan L, et al. Beneficial effect of combination hormonal therapy administered prior and following external beam radiation therapy in localized prostate cancer. Int J Radiat Oncol Biol Phys 1997; 37:247-252.[Medline]
  8. Pilepich MV, Winter K, Roach M, et al. Phase III Radiation Oncology Group (RTOG) 86-10 of androgen deprivation before and during radiotherapy in locally advanced carcinoma of the prostate (abstr). In: Proceedings of the American Society of Clinical Oncology. Philadelphia, Pa: WB Saunders, 1998; 17:308a.
  9. Pilepich MV, Krall JM, al-Sarraf M, et al. Androgen deprivation with radiation therapy compared with radiation therapy alone for locally advanced prostatic carcinoma: a randomized comparative trial of the Radiation Therapy Oncology Group. Urology 1995; 45:616-623.[Medline]
  10. Porter A, Ethliali M, Manji M, et al. A phase III randomised trial to evaluate the efficacy of neoadjuvant therapy prior to curative radiotherapy in locally advanced prostate cancer patients: a Canadian Urologic Oncology Group study (abstr). In: Proceedings of the American Society of Clinical Oncology. Philadelphia, Pa: WB Saunders, 1998; 17:315a.
  11. Soloway MS, Sharifi R, Wajsman Z, McLeod D, Wood DP, Jr, Puras-Baez A. Randomized prospective study comparing radical prostatectomy alone versus radical prostatectomy proceeded by androgen blockade in clinical stage B2 (T2bNxM0) prostate cancer: the Lupron Depot Neoadjuvant Prostate Cancer Study Group. J Urol 1995; 154:424-428.[Medline]
  12. Labrie F, Cusan L, Gomez JL, et al. Neoadjuvant hormonal therapy: the Canadian experience. Urology 1997; 49(suppl 3A):56-64.[Medline]
  13. Goldenberg SL, Klotz LH, Srigley J, et al. Randomized, prospective, controlled study comparing radical prostatectomy alone and neoadjuvant androgen withdrawal in the treatment of localized prostate cancer: Canadian Urologic Oncology Group. J Urol 1996; 156:873-877.[Medline]
  14. Fair WR, Cookson MS, Stroumbakis N, et al. The indications, rationale, and results of neoadjuvant androgen deprivation in the treatment of prostatic cancer: Memorial Sloan-Kettering Cancer Center results. Urology 1997; 49(suppl 3A):46-55.
  15. Witjes WP, Schulman CC, Debruyne FM. Preliminary results of a prospective randomized study comparing radical prostatectomy versus radical prostatectomy associated with neoadjuvant hormonal combination therapy in T2-3 N0 M0 prostatic carcinoma: the European Study Group on Neoadjuvant Treatment of Prostate Cancer. Urology 1997; 49(suppl 3A):65-69.
  16. Zelefsky MJ, Leibel SA, Burman CM, et al. Neoadjuvant hormonal therapy improves the therapeutic ratio in patients with bulky prostatic cancer treated with three-dimensional conformal radiation therapy. Int J Radiat Oncol Biol Phys 1994; 29:755-761.[Medline]
  17. Zietman AL, Prince EA, Nakfoor BM, Park JJ. Androgen deprivation and radiation therapy: sequencing studies using the Shionogi in vivo tumor system. Int J Radiat Oncol Biol Phys 1997; 38:1067-1070.[Medline]
  18. Joon DL, Hasegawa M, Sikes C, et al. Supraadditive apoptotic response of R3327-G rat prostate tumors to androgen ablation and radiation. Int J Radiat Oncol Biol Phys 1997; 38:1071-1077.[Medline]
  19. Shearer RJ, Davies JH, Gelister JS, Dearnaley DP. Hormonal cytoreduction and radiotherapy for carcinoma of the prostate. Br J Urol 1992; 69:521-524.[Medline]
  20. Pinault S, Tetu B, Gagnon J, Monfette G, Dupont A, Labrie F. Transrectal ultrasound evaluation of local prostate cancer in patients treated with LHRH agonist and in combination with flutamide. Urology 1992; 39:254-261.[Medline]
  21. Chen M, Hricak H, Kalbhen CL, et al. Hormonal ablation of prostatic cancer: effects on prostate morphology, tumor detection, and staging by endorectal coil MR imaging. AJR Am J Roentgenol 1996; 166:1157-1163.[Abstract/Free Full Text]
  22. Nakashima J, Imai Y, Tachibana M, Baba S, Hiramatsu K, Murai M. Effects of endocrine therapy on the primary lesion in patients with prostate carcinoma as evaluated by endorectal magnetic resonance imaging. Cancer 1997; 80:237-241.[Medline]
  23. Padhani AR. Dynamic contrast enhanced MRI in human tumors. Br J Radiol 1999; 72:427-431.[Medline]
  24. Dvorak HF, Brown LF, Detmar M, Dvorak AM. Vascular permeability factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis. Am J Pathol 1995; 146:1029-1039.[Abstract]
  25. Haggstrom S, Lissbrant IF, Bergh A, Damber JE. Testosterone induces vascular endothelial growth factor synthesis in the ventral prostate in castrated rats. J Urol 1999; 161:1620-1625.[Medline]
  26. Jain RK, Safabakhsh N, Sckell A, et al. Endothelial cell death, angiogenesis, and microvascular function after castration in an androgen-dependent tumor: role of vascular endothelial growth factor. Proc Natl Acad Sci USA 1998; 95:10820-10825.[Abstract/Free Full Text]
  27. Gleason DF. Histologic grading of prostate cancer: a perspective. Hum Pathol 1992; 23:273-279.[Medline]
  28. Vinnicombe SJ, Norman AR, Nicolson V, Husband JE. Normal pelvic lymph nodes: evaluation with CT after bipedal lymphangiography. Radiology 1995; 194:349-355.[Abstract/Free Full Text]
  29. Brown MA, Semelka RC. MR imaging abbreviations, definitions, and descriptions: a review. Radiology 1999; 213:647-662.[Free Full Text]
  30. Parker GJM, Suckling J, Tanner SF, et al. Probing tumor microvessel density by measurement, analysis and display of contrast agent uptake kinetics. J Magn Reson Imaging 1997; 7:564-574.[Medline]
  31. Parker GJ, Baustert I, Tanner SF, Leach MO. Improving image quality and T(1) measurements using saturation recovery turboFLASH with an approximate K-space normalisation filter. Magn Reson Imaging 2000; 18:157-167.[Medline]
  32. Sobin LH, Witteking CH, eds. TNM classification of malignant tumours 5th ed. New York, NY: Wiley-Liss, 1997; 170-173.
  33. Outwater EK, Petersen RO, Siegelman ES, Gomella LG, Chernesky CE, Mitchell DG. Prostate carcinoma: assessment of diagnostic criteria for capsular penetration on endorectal coil MR images. Radiology 1994; 193:333-339.[Abstract/Free Full Text]
  34. Parker GJM, Suckling J, Tanner SF, Padhani AR, Husband JE, Leach MO. MRIW: parametric analysis software for contrast-enhanced dynamic MR imaging in cancer. RadioGraphics 1998; 18:497-506.[Abstract]
  35. Kuhl CK, Mielcareck P, Klaschik S, et al. Dynamic breast MR imaging: are signal intensity time course data useful for differential diagnosis of enhancing lesions?. Radiology 1999; 211:101-110.[Abstract/Free Full Text]