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


     

DOI: 10.1148/radiol.2381052345
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 Sala, E.
Right arrow Articles by Hricak, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sala, E.
Right arrow Articles by Hricak, H.
(Radiology 2006;238:176-183.)
© RSNA, 2006

Genitourinary Imaging

Endorectal MR Imaging before Salvage Prostatectomy: Tumor Localization and Staging1

Evis Sala, MD, PhD, FRCR, Steven C. Eberhardt, MD, Oguz Akin, MD, Chaya S. Moskowitz, PhD, Chinyere N. Onyebuchi, MPH, Kentaro Kuroiwa, MD, Nicole Ishill, MS, Michael J. Zelefsky, MD, James A. Eastham, MD, FACS and Hedvig Hricak, MD, PhD

1 From the Departments of Radiology (E.S., S.C.E., O.A., C.N.O., H.H.), Epidemiology and Biostatistics (C.S.M., N.I.), Pathology (K.K.), Radiation Oncology (M.J.Z.), and Urology (J.A.E.), Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10021. Received February 28, 2005; revision requested April 5; revision received April 25; accepted May 19; final version accepted, June 13. Supported by NIH grant R01 CA76423. Address correspondence to E.S. (e-mail: es220{at}radiol.cam.ac.uk).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Purpose: To evaluate retrospectively the accuracy of endorectal magnetic resonance (MR) imaging for the depiction of tumor, extracapsular extension (ECE), and seminal vesicle invasion (SVI) before salvage prostatectomy in patients with locally recurrent prostate cancer after radiation therapy, by using pathologic analysis as the reference standard.

Materials and Methods: The Institutional Review Board granted exempt status for this HIPAA-compliant study, with a waiver of informed consent. Forty-five consecutive patients (age range, 43–76 years) were identified who underwent salvage radical prostatectomy for prostate cancer at Memorial Sloan-Kettering Cancer Center between December 1, 1998, and October 31, 2004, and who underwent endorectal MR imaging prior to surgery. Tumor localization and determination of local stage with MR imaging were performed independently by two radiologists. Interpretations were compared to pathologic findings from surgical specimens. Interrater variability was estimated with the {kappa} statistic. Areas under the receiver operating characteristic curve (AUCs) were used to assess the accuracy of endorectal MR imaging in tumor detection and determination of ECE and SVI.

Results: Findings of histologic examination showed that tumor was present in all patients. For tumor detection, the AUC value for reader 1 was 0.75 (95% confidence interval [CI]: 0.67, 0.84), whereas the AUC value for reader 2 was 0.61 (95% CI: 0.52, 0.71). The AUC values for prediction of ECE were 0.87 (95% CI: 0.80, 0.94) for reader 1 and 0.76 (95% CI: 0.67, 0.85) for reader 2. The AUC values for prediction of SVI were 0.76 (95% CI: 0.62, 0.90) for reader 1 and 0.70 (95% CI: 0.56, 0.85) for reader 2. For all variables, the {kappa} statistics used to assess interrater agreement between readers were fair (0.45, 0.52, and 0.47 for tumor location, ECE, and SVI, respectively).

Conclusion: Endorectal MR imaging following radiation therapy can help identify tumor sites and depict ECE and SVI with reasonable accuracy in patients with recurrent prostate cancer.

© RSNA, 2006


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Prostate cancer is the most common malignancy and the second leading cause of cancer death in American men. An estimated 232 090 men will receive a diagnosis of prostate cancer in the United States in 2005 (1). In 2004, 86% of reported cases of prostate cancer were of a local or regional stage at diagnosis (2), and most patients underwent some type of local therapy with curative intent (3–5). The data on treatment distribution show that approximately 55% of patients with newly diagnosed prostate cancer receive radiation therapy to the prostate (6,7). At 5 years, 30%–50% of these patients will have biochemical recurrence of the disease (6,7). Although in some patients the treatment will fail because of distant disease, others will have local recurrence only and will be potential candidates for additional local therapy, including salvage radical prostatectomy.

Detection of local recurrence after failed radiation therapy is a clinical challenge, because an increase in the prostate-specific antigen (PSA) level is not a reliable variable for differentiating local from distant recurrence (8,9). Clinical criteria for salvage radical prostatectomy after failed radiation therapy include locally confined, biopsy-proved recurrence of cancer with no evidence of metastasis and an otherwise healthy patient with a life expectancy of at least 10 years who would have been a candidate for surgery at the time of the original diagnosis. In properly selected cases, salvage radical prostatectomy represents a chance for cure, but because this procedure has historically been associated with substantial short- and long-term morbidity and with difficulties in the evaluation of local disease extent, its use has been limited. Recent clinical reports (9–11) of salvage radical prostatectomy series have suggested that the surgical time, the estimated blood loss, and the average hospital stay approach those for standard radical prostatectomy. Nevertheless, despite efforts to select patients with clinically localized prostate cancer, in previous series (11–14) the rate of organ-confined disease at pathologic analysis has been only 25%–42%, and the rate of seminal vesicle invasion (SVI) has been 25%–63%.

In patients being considered for salvage prostatectomy, computed tomography (CT) is helpful in the assessment of lymph node and bone metastases, but its role in local tumor staging has not been evaluated (12,14). To our knowledge, there is no published study addressing the role of endorectal magnetic resonance (MR) imaging in the evaluation of local tumor extent prior to salvage prostatectomy. Thus, the purpose of our study was to evaluate retrospectively the accuracy of endorectal MR imaging for the depiction of tumor, extracapsular extension (ECE), and SVI before salvage prostatectomy in patients with locally recurrent prostate cancer after radiation therapy, by using pathologic analysis as the reference standard.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Patient Characteristics
This was a retrospective, single-institutional cross-sectional study. The Institutional Review Board granted exempt status for the study, which complied with the Health Insurance Portability and Accountability Act, with a waiver of informed consent. We identified 45 consecutive patients who underwent salvage radical prostatectomy for prostate cancer at our institution between December 1, 1998, and October 31, 2004, and who underwent endorectal MR imaging before surgery. The median patient age was 62 years (range, 43–76 years). The median Gleason score (Table 1) at initial diagnosis was 7 (range, 4–8), and the median baseline serum PSA level (before radiation therapy) was 6.7 ng/mL (range, 0.55–26.00 ng/mL). Radiation therapy was performed at our institution for 16 of 45 patients. All patients treated at our institution underwent conformal external-beam radiation therapy (median dose, 8100 cGy; range, 4500–8640 cGy). Of the remaining 29 patients who received radiation therapy outside our institution, nine underwent brachytherapy, three underwent combination therapy, and 17 underwent external radiation therapy. All patients had biopsy-proved recurrence and no clinical evidence of distant metastases. Seventeen (32%) patients received chemotherapy or hormonal therapy before salvage prostatectomy. All patients had documented biochemical failure before surgery (median PSA level before MR imaging, 3.57 ng/mL; range, 0.27–12.44 ng/mL). The median time from radiation therapy to surgery was 54.2 months.


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

  		Table 1. Patient Characteristics

 
Surgery and Postoperative Evaluation
The surgical technique for salvage radical prostatectomy with pelvic lymph node dissection was similar to that of standard radical prostatectomy (9). There were no major intraoperative or postoperative complications and no patient required rehospitalization within 30 days of surgery.

MR Imaging Technique
All patients underwent preoperative endorectal MR imaging of the pelvis. Nineteen of 45 patients underwent MR imaging before the biopsy, whereas 26 patients did so after the tumor recurrence was proved at biopsy (mean, 59 days after biopsy; range, 98 days before biopsy to 535 days after biopsy). Imaging was performed by using a 1.5-T whole-body MR imager (Signa; GE Medical Systems, Milwaukee, Wis). The MR imaging technique has been previously described (15,16). The patients were examined in supine position by using the body coil for excitation and a pelvic phased-array coil (GE Medical Systems) in combination with a balloon-covered expandable endorectal coil (Medrad, Pittsburgh, Pa) for signal reception. Transverse spin-echo T1-weighted images were obtained from the aortic bifurcation to the symphysis pubis by using the following parameters: repetition time msec/echo time msec, 700/8; section thickness, 5 mm; intersection gap, 1 mm; field of view, 24 cm; matrix, 256 x 192; frequency encoding, transverse; and number of signals acquired, one. Thin-section transverse and coronal T2-weighted fast spin-echo images of the prostate and seminal vesicles were obtained by using the following parameters: 5000/96 (effective); echo train length, 16; section thickness, 3 mm; intersection gap, 0 mm; field of view, 14 cm; matrix, 256 x 192; frequency encoding, anteroposterior; and number of signals acquired, three.

MR Imaging Analysis
Only presalvage prostatectomy endorectal MR images were analyzed retrospectively and separately by two radiologists (E.S., O.A.) who were unaware of the clinical and surgical and/or histologic findings. Each had completed a body imaging fellowship and had read more than 500 prostate MR images. Prior to image interpretation, the readers met with other coauthors of the manuscript and decided which MR diagnostic criteria to use in the diagnosis of ECE and SVI and designed the data collection form. They independently completed data sheets for each case and specified suspected tumor locations according to quadrant (divided left vs right and anterior vs posterior).

The criterion for tumor detection was a low-signal-intensity region within the transition zone and/or peripheral zone of the gland on T2-weighted images. Only one lesion per patient was recorded. Sites where ECE was suspected were recorded by using similar location designations. Criteria for ECE on endorectal MR images included capsular irregularity, bulging of the capsule, capsular retraction, obliteration of the recto-prostatic angle, and asymmetry or direct involvement of the neurovascular bundles (16). The presence or lack of SVI was recorded for the left and right sides. Criteria for SVI were a focal low-signal-intensity mass or diffuse enlargement with low signal intensity and loss of the perceptible vesical wall at both T1- and T2-weighted sequences. A five-point scoring system was used to evaluate all the findings (score of 1 = tumor definitely absent, score of 2 = tumor probably absent, score of 3 = tumor possibly present, score of 4 = tumor probably present, and score of 5 = tumor definitely present). There were no specific criteria associated with each score; the scoring system simply related to the reader's overall impression on the basis of features observed for the specific finding being evaluated. A lymph node was considered suspicious for malignancy if it measured 7 mm or longer in the short-axis diameter.

Changes associated with prior pelvic irradiation were recorded (17); these included changes in the appearance of the zonal anatomy of the prostate. Conspicuity of zonal anatomy as observed on T2-weighted images was recorded as either preserved or indistinct. Also noted on T2-weighted images were increases in signal intensity and loss of normal thickness of the levator ani muscles and increases in signal intensity in the adjacent organs, such as the walls of the bladder and rectum. Bone marrow changes, observed as an increase in signal intensity, were evaluated on T1-weighted images.

Pathologic Assessment and Comparison with MR Images
All prostatectomy specimens were stained with India ink (green dye on right, blue dye on left) and fixed in 10% formalin for 36 hours. The distal 5 mm of the apex was amputated and coned. The remainder of the gland was serially sectioned from apex to base to obtain transverse slices at 3-mm intervals (transverse step sections), and the slices were entirely submitted for paraffin embedding as whole mounts. The seminal vesicles were amputated and submitted separately. After paraffin embedding, microsections were placed on glass slides and were stained with hematoxylin-eosin. The transverse pathologic step sections were numbered consecutively from the apex to the base, and the tumor areas were mapped in each section with a marker. The pathologic stage and surgical Gleason score were determined for each patient. The assessment was performed by a pathologist (K.K.) with 5 years of experience in prostate cancer pathology.

After the completion of the MR imaging readings, these transverse pathologic step sections were then matched with the transverse T2-weighted MR images. On the basis of these matches, the presence or absence of cancer, the presence or absence of ECE, and the presence or absence of SVI were determined for each suspicious lesion marked on the MR images. This was performed by one of the authors (H.H.), who has 23 years of experience in prostate MR imaging.

Statistical Analysis
Two of the authors (C.S.M., N.I.) were responsible in consensus for the statistical analysis. The receiver operating characteristic (ROC) curve and the area under the ROC curve (AUC) were estimated with empirical nonparametric estimates. The quadrant was the unit of analysis in the assessment of the accuracy of endorectal MR images in detection and localization of tumor and ECE, whereas the side was the unit of analysis in the assessment of the accuracy of endorectal MR images in detection and localization of SVI. For the estimated AUCs, confidence intervals (CIs) that accounted for the multiple observations per patient were calculated by using methods described by Obuchowski (18). Interrater variability in tumor localization and assessment of ECE and SVI was estimated with the {kappa} statistic. As suggested previously (19), a {kappa} value of 0.75–1.0 indicates good agreement, a {kappa} value of 0.40–0.75 indicates fair to good agreement, and a {kappa} value of less than 0.40 indicates poor agreement. The ROC analysis was conducted by using S-Plus (version 6.2 for Windows; Insightful, Seattle, Wash) software. All other analyses were performed with STATA (version 8.0 for Windows; Stata, College Station, Tex) software.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Tumor Location
At histopathologic examination, all patients had tumor in at least one quadrant of the prostate gland. Nine patients had tumor in a single quadrant, 12 had tumor in two quadrants, seven had tumor in three quadrants, and 16 had tumor in all four quadrants. Twenty-three of 45 patients had organ-confined tumor on the basis of pathologic analysis. Nineteen (42%) patients had ECE. Of these, 12 had ECE in one quadrant, six had ECE in two quadrants, and one had ECE in all four quadrants. Thirteen patients had SVI on at least one side, and six had SVI on both sides.

ROC Analysis
In detection and localization of cancer, the AUC value was 0.75 (95% CI: 0.67, 0.84) for reader 1 and 0.61 (95% CI: 0.52, 0.71) for reader 2 (Fig 1). The {kappa} statistic for assessment of interrater agreement was 0.45, which indicated fair agreement between readers.



View larger version (30K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1: ROC curve summarizes the accuracy of endorectal MR imaging in tumor detection and localization.

 
In detection and localization of ECE, the AUC value was 0.87 (95% CI: 0.80, 0.94) for reader 1 and 0.76 (95% CI: 0.67, 0.85) for reader 2 (Fig 2). The {kappa} statistic for assessment of interrater agreement was 0.52, which indicated fair agreement between readers.



View larger version (30K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2: ROC curve summarizes the accuracy of endorectal MR imaging in detection and localization of ECE.

 
In detection and localization of SVI, the AUC value was 0.76 (95% CI: 0.62, 0.90) for reader 1 and 0.70 (95% CI: 0.56, 0.85) for reader 2 (Fig 3). The {kappa} statistic for assessment of interrater agreement was 0.47, which indicated fair agreement between readers.



View larger version (30K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3: ROC curve summarizes the accuracy of endorectal MR imaging in detection and localization of SVI.

 
Sensitivity and Specificity
We dichotomized the five-point scoring system to assess the sensitivity and specificity of endorectal MR imaging in tumor detection and diagnosis of ECE (Fig 4) and SVI (Fig 5). Two cutoff points were explored: cutoff point 1 assumed that values 1 and 2 indicated absence of tumor and values 3–5 indicated presence of tumor, whereas cutoff point 2 assumed that values 1–3 indicated absence of tumor and values 4 and 5 indicated presence of tumor. Table 2 shows the sensitivity and specificity for each reader for tumor location, ECE, and SVI with use of the two cutoff points. In detection of tumor location, sensitivities at cutoff points 1 and 2 were 76% and 58%, respectively, for reader 1 and 55% and 36%, respectively, for reader 2. At cutoff point 1, the specificity for reader 2 was lower than that for reader 1 (65% vs 73%), whereas at cutoff point 2 both readers had the same specificity of 81%. Both readers were able to evaluate ECE and SVI with good specificity (Table 2).



View larger version (119K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4a: Clinical stage T2b prostate cancer (Gleason score, 7; PSA level, 9.13 ng/mL) in a 57-year-old patient. (a, b) Transverse T2-weighted MR images show a large focal low-signal-intensity area in the right posterior peripheral zone consistent with recurrent tumor (T). Note the capsular irregularity (arrow in a) and asymmetry of the neurovascular bundle (arrow in b) compatible with ECE. (c, d) Corresponding histopathologic whole-mount sections confirm established ECE at the right posterior base (marked in red). (Hematoxylin-eosin stain.)

 


View larger version (120K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4b: Clinical stage T2b prostate cancer (Gleason score, 7; PSA level, 9.13 ng/mL) in a 57-year-old patient. (a, b) Transverse T2-weighted MR images show a large focal low-signal-intensity area in the right posterior peripheral zone consistent with recurrent tumor (T). Note the capsular irregularity (arrow in a) and asymmetry of the neurovascular bundle (arrow in b) compatible with ECE. (c, d) Corresponding histopathologic whole-mount sections confirm established ECE at the right posterior base (marked in red). (Hematoxylin-eosin stain.)

 


View larger version (103K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4c: Clinical stage T2b prostate cancer (Gleason score, 7; PSA level, 9.13 ng/mL) in a 57-year-old patient. (a, b) Transverse T2-weighted MR images show a large focal low-signal-intensity area in the right posterior peripheral zone consistent with recurrent tumor (T). Note the capsular irregularity (arrow in a) and asymmetry of the neurovascular bundle (arrow in b) compatible with ECE. (c, d) Corresponding histopathologic whole-mount sections confirm established ECE at the right posterior base (marked in red). (Hematoxylin-eosin stain.)

 


View larger version (79K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4d: Clinical stage T2b prostate cancer (Gleason score, 7; PSA level, 9.13 ng/mL) in a 57-year-old patient. (a, b) Transverse T2-weighted MR images show a large focal low-signal-intensity area in the right posterior peripheral zone consistent with recurrent tumor (T). Note the capsular irregularity (arrow in a) and asymmetry of the neurovascular bundle (arrow in b) compatible with ECE. (c, d) Corresponding histopathologic whole-mount sections confirm established ECE at the right posterior base (marked in red). (Hematoxylin-eosin stain.)

 


View larger version (100K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a: Clinical stage T1c prostate cancer (Gleason score, 7; PSA level, 4.35 ng/mL) in a 57-year-old patient. (a–c) Transverse T2-weighted MR images show a large focal low-signal-intensity area in the left base consistent with recurrent tumor (T). Note low-signal-intensity area (arrow) outside the capsule compatible with ECE (not shown on pathologic map) and a low-signal-intensity area in the left seminal vesicle (T) indicative of SVI. (d) Coronal T2-weighted image demonstrates the location of the tumor (T) just superior to brachytherapy seeds (arrows) and direct extension into the left seminal vesicle (LSV). (e, f) Corresponding histopathologic whole-mount sections confirm established left SVI. (Hematoxylin-eosin stain.)

 


View larger version (107K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b: Clinical stage T1c prostate cancer (Gleason score, 7; PSA level, 4.35 ng/mL) in a 57-year-old patient. (a–c) Transverse T2-weighted MR images show a large focal low-signal-intensity area in the left base consistent with recurrent tumor (T). Note low-signal-intensity area (arrow) outside the capsule compatible with ECE (not shown on pathologic map) and a low-signal-intensity area in the left seminal vesicle (T) indicative of SVI. (d) Coronal T2-weighted image demonstrates the location of the tumor (T) just superior to brachytherapy seeds (arrows) and direct extension into the left seminal vesicle (LSV). (e, f) Corresponding histopathologic whole-mount sections confirm established left SVI. (Hematoxylin-eosin stain.)

 


View larger version (109K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5c: Clinical stage T1c prostate cancer (Gleason score, 7; PSA level, 4.35 ng/mL) in a 57-year-old patient. (a–c) Transverse T2-weighted MR images show a large focal low-signal-intensity area in the left base consistent with recurrent tumor (T). Note low-signal-intensity area (arrow) outside the capsule compatible with ECE (not shown on pathologic map) and a low-signal-intensity area in the left seminal vesicle (T) indicative of SVI. (d) Coronal T2-weighted image demonstrates the location of the tumor (T) just superior to brachytherapy seeds (arrows) and direct extension into the left seminal vesicle (LSV). (e, f) Corresponding histopathologic whole-mount sections confirm established left SVI. (Hematoxylin-eosin stain.)

 


View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5d: Clinical stage T1c prostate cancer (Gleason score, 7; PSA level, 4.35 ng/mL) in a 57-year-old patient. (a–c) Transverse T2-weighted MR images show a large focal low-signal-intensity area in the left base consistent with recurrent tumor (T). Note low-signal-intensity area (arrow) outside the capsule compatible with ECE (not shown on pathologic map) and a low-signal-intensity area in the left seminal vesicle (T) indicative of SVI. (d) Coronal T2-weighted image demonstrates the location of the tumor (T) just superior to brachytherapy seeds (arrows) and direct extension into the left seminal vesicle (LSV). (e, f) Corresponding histopathologic whole-mount sections confirm established left SVI. (Hematoxylin-eosin stain.)

 


View larger version (108K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5e: Clinical stage T1c prostate cancer (Gleason score, 7; PSA level, 4.35 ng/mL) in a 57-year-old patient. (a–c) Transverse T2-weighted MR images show a large focal low-signal-intensity area in the left base consistent with recurrent tumor (T). Note low-signal-intensity area (arrow) outside the capsule compatible with ECE (not shown on pathologic map) and a low-signal-intensity area in the left seminal vesicle (T) indicative of SVI. (d) Coronal T2-weighted image demonstrates the location of the tumor (T) just superior to brachytherapy seeds (arrows) and direct extension into the left seminal vesicle (LSV). (e, f) Corresponding histopathologic whole-mount sections confirm established left SVI. (Hematoxylin-eosin stain.)

 


View larger version (85K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5f: Clinical stage T1c prostate cancer (Gleason score, 7; PSA level, 4.35 ng/mL) in a 57-year-old patient. (a–c) Transverse T2-weighted MR images show a large focal low-signal-intensity area in the left base consistent with recurrent tumor (T). Note low-signal-intensity area (arrow) outside the capsule compatible with ECE (not shown on pathologic map) and a low-signal-intensity area in the left seminal vesicle (T) indicative of SVI. (d) Coronal T2-weighted image demonstrates the location of the tumor (T) just superior to brachytherapy seeds (arrows) and direct extension into the left seminal vesicle (LSV). (e, f) Corresponding histopathologic whole-mount sections confirm established left SVI. (Hematoxylin-eosin stain.)

 

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

  		Table 2. Sensitivity and Specificity for Tumor Location, ECE, and SVI at Two Cutoff Points

 
MR Imaging Accuracy in Patient Treatment
Because the decision whether to proceed to surgery depends only on the detection of ECE or SVI in an individual patient and not on the location of either of these features, we also analyzed the data on ECE and SVI at the patient level, ignoring the location of tumor. For this analysis, we dichotomized the five-point scale for ECE and SVI at cutoff point 1 and classified a judgment at any modality (either of the two radiologists or pathologic analysis) as positive for SVI and/or ECE if the modality indicated SVI or ECE was present anywhere in the prostate. A sensitivity of 62% (eight of 13) (95% CI: 33%, 84%) and a specificity of 97% (31 of 32) (95% CI: 80%, 100%) for detection of SVI at the patient level and a sensitivity of 89% (17 of 19) (95% CI: 65%, 98%) and a specificity of 50% (13 of 26) (95% CI: 31%, 96%) for detection of ECE at the patient level were recorded for reader 1. A sensitivity of 38% (five of 13) (95% CI: 16%, 67%) and a specificity of 94% (30 of 32) (95% CI: 77%, 99%) for detection of SVI at the patient level and a sensitivity of 84% (16 of 19) (95% CI: 60%, 95%) and a specificity of 46% (12 of 26) (95% CI: 28%, 66%) for detection of ECE at the patient level were recorded for reader 2.

Other Findings and Subset Analysis
Eight (18%) of 45 patients had lymph node involvement at pathologic analysis. Readers 1 and 2 both indicated that only three (7%) of 45 patients had lymph node involvement; both readers identified the same three patients. Only one of these three patients had lymph node involvement at pathologic analysis. Both readers missed lymph node involvement in seven patients. All involved nodes that were missed by both readers measured less than 5 mm.

All patients had radiation therapy changes (ie, diffuse low signal intensity on T2-weighted images) to the prostate gland on endorectal MR images, 71% (n = 32) had radiation therapy changes to bone (ie, increased signal intensity on T2-weighted images), and 60% (n = 27) had radiation therapy changes to muscle. Zonal anatomy was preserved in 60% of the patients and was indistinct in 40%. An ROC analysis was performed that was restricted to the subset of patients in whom zonal anatomy was preserved. For tumor localization, the AUC value for reader 1 was 0.72 (95% CI: 0.60, 0.83) and the AUC value for reader 2 was 0.60 (95% CI: 0.48, 0.72). For ECE detection, the AUC value for reader 1 was 0.85 (95% CI: 0.75, 0.94) and the AUC value for reader 2 was 0.75 (95% CI: 0.62, 0.88). For detection and localization of SVI, the AUC value for reader 1 was 0.73 (95% CI: 0.51, 0.96) and the AUC value for reader 2 was 0.74 (95% CI: 0.51, 0.98). The results of this analysis of the subset of patients in whom zonal anatomy was preserved are similar to the results for all subjects, which suggests that there is no relationship between radiation therapy changes to zonal anatomy and the accuracy of endorectal MR imaging.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
Curative treatment options following failure of radiation therapy for localized prostate cancer are limited. Additional regimens of radiation are risky owing to potential injury to the rectum, urethra, or bladder (20). Hormonal therapy provides limited local tumor control and is not curative (9,21). Salvage radical prostatectomy can result in prolonged disease-free survival, but its routine use has not been widely accepted. A major reason for the lack of acceptance of salvage prostatectomy has been the inaccuracy of local staging and a tendency for clinical understaging prior to surgery.

Clinical tools that are available in the assessment of potential candidates for salvage prostatectomy are limited. Serum PSA levels less than 10 ng/mL have been associated with improved cancer control outcomes after salvage radical prostatectomy (22). Digital rectal examination and radiographic imaging, including CT or transrectal ultrasonography, add little to the assessment of patients in this setting. Additional methods of preoperative assessment are needed to improve patient selection for additional local therapy after radiation failure.

The MR imaging technique used in our study was similar to that used for pretreatment assessment in patients who have not received radiation. Endorectal coil MR imaging is generally considered the most accurate imaging method for local staging of prostate cancer (15,16,23–25). After pelvic radiation, however, assessment of intraprostatic tumor location and parameters that help determine local stage, specifically ECE and SVI, has been believed to be hindered by tissue changes related to radiation therapy. It has been reported that a diffuse decrease in T2-weighted signal intensity within the gland and loss or indistinctness of zonal anatomy make it more difficult to distinguish cancer lesions, which are hypointense both before and after radiation, from benign prostatic tissue (26). Our data, however, did not support that observation. Our study findings showed the accuracies for tumor detection and localization as estimated with AUC, sensitivity, and specificity were similar to those obtained in studies (15,25,27,28) of untreated patients, in which sensitivities ranged from 67% to 88%. For both readers, the sensitivities and specificities for determination of ECE and SVI were similar to staging accuracy in a series of patients who had not received radiation (15,28). Although both readers were considered experienced specialists in interpretation, we observed a fair amount of interobserver variability, as shown with {kappa} values of 0.45–0.52 for tumor localization, detection of ECE, and detection of SVI. That prostate MR image interpretation is subject to substantial interobserver variability has been an ongoing issue (25,29).

Although our study was limited by a relatively small sample size, to our knowledge, it represents the only series to date with preoperative endorectal MR imaging and pathologic correlation. When two studies (13,30) that collected data over a period of 20 years are excluded, our study size (45 patients) was larger than that of any other study on salvage prostatectomy (mean number of patients, 21; range, five to 43) reported in the literature (14).

Another limitation of this study was that all subjects included here were known to have biopsy-proved recurrence. For this reason, we did not report measures of accuracy such as sensitivity, specificity, and AUC for detection of tumor at the patient level (ie, identification of the presence of tumor in the patient, regardless of its location). Instead, we followed an approach described by Obuchowski et al (31) in which we divided the prostate into distinct regions and reported measures of accuracy for localization of the tumor (detection within each region). Because finding tumor in one quadrant may influence finding tumor in another quadrant, this approach introduced bias in the form of intrapatient correlation. By using the statistical methodology described by Obuchowski (18), we adjusted the estimates of the AUCs and their CIs for this correlation. We emphasize, however, that these numbers may well differ from the accuracy one would find in detection of tumor at the patient level.

A final limitation of our study was a potential for verification bias, because for the past 2 years at our institution endorectal MR imaging has been performed in every patient considered for salvage prostatectomy. The results of endorectal MR imaging may therefore have contributed to the decision to perform or cancel a salvage prostatectomy. It is possible that this bias led to inflated estimates of AUC, sensitivity, and specificity, as well as to better selection for organ-confined disease at surgery.

In summary, findings of preoperative endorectal MR imaging showed substantial interobserver variability, even among readers with similar experience in endorectal MR image interpretation. Nevertheless, the results suggest that the accuracy of endorectal MR imaging in tumor evaluation prior to salvage prostatectomy is similar to the accuracy of endorectal MR imaging in pretreatment staging. Therefore, we propose that the information obtained with endorectal MR imaging should be incorporated into the criteria used to select patients for salvage radical prostatectomy and may serve as an aid in surgical planning.


   FOOTNOTES
 

Abbreviations: AUC = area under the ROC curve • CI = confidence interval • ECE = extracapsular extension • PSA = prostate-specific antigen • ROC = receiver operating characteristic • SVI = seminal vesicle invasion

Authors stated no financial relationship to disclose.

Author contributions: Guarantors of integrity of entire study, E.S., H.H.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, E.S.; clinical studies, E.S., S.C.E., O.A., K.K., J.A.E.; statistical analysis, C.S.M., N.I.; and manuscript editing, all authors


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. CA Cancer J Clin 2005;55:10–30.[Abstract/Free Full Text]
  2. American Cancer Society: Cancer facts and figures 2004. Atlanta, Ga: American Cancer Society, 2004.
  3. Stephenson RA, Stanford JL. Population-based prostate cancer trends in the United States: patterns of change in the era of prostate-specific antigen. World J Urol 1997;15:331–335.[CrossRef][Medline]
  4. Kaminski JM, Hanlon AL, Horwitz EM, Pinover WH, Mitra RK, Hanks GE. Relationship between prostate volume, prostate-specific antigen nadir, and biochemical control. Int J Radiat Oncol Biol Phys 2002;52:888–892.[CrossRef][Medline]
  5. Zelefsky MJ, Fuks Z, Hunt M, et al. High dose radiation delivered by intensity modulated conformal radiotherapy improves the outcome of localized prostate cancer. J Urol 2001;166:876–881.[CrossRef][Medline]
  6. Shipley WU, Thames HD, Sandler HM, et al. Radiation therapy for clinically localized prostate cancer: a multi-institutional pooled analysis. JAMA 1999;281:1598–1604.[Abstract/Free Full Text]
  7. Lee WR, Moughan J, Owen JB, Zelefsky MJ. The 1999 patterns of care study of radiotherapy in localized prostate carcinoma: a comprehensive survey of prostate brachytherapy in the United States. Cancer 2003;98:1987–1994.[CrossRef][Medline]
  8. Boccon-Gibod L, Djavan WB, Hammerer P, et al. Management of prostate-specific antigen relapse in prostate cancer: a European Consensus. Int J Clin Pract 2004;58:382–390.[CrossRef][Medline]
  9. Stephenson AJ, Scardino PT, Bianco FJ, Eastham JA. Salvage therapy for locally recurrent prostate cancer after external beam radiotherapy. Curr Treat Options Oncol 2004;5:357–365.[Medline]
  10. Shekarriz B, Upadhyay J, Pontes JE. Salvage radical prostatectomy. Urol Clin North Am 2001;28:545–553.[CrossRef][Medline]
  11. Vaidya A, Soloway MS. Salvage radical prostatectomy for radiorecurrent prostate cancer: morbidity revisited. J Urol 2000;164:1998–2001.[CrossRef][Medline]
  12. Tefilli MV, Gheiler EL, Tiguert R, et al. Salvage surgery or salvage radiotherapy for locally recurrent prostate cancer. Urology 1998;52:224–229.[CrossRef][Medline]
  13. Lerner SE, Blute ML, Zincke H. Critical evaluation of salvage surgery for radio-recurrent/resistant prostate cancer. J Urol 1995;154:1103–1109.[CrossRef][Medline]
  14. Chen BT, Wood DP Jr. Salvage prostatectomy in patients who have failed radiation therapy or cryotherapy as primary treatment for prostate cancer. Urology 2003;62(suppl 1):69–78.
  15. Claus FG, Hricak H, Hattery RR. Pretreatment evaluation of prostate cancer: role of MR imaging and 1H MR spectroscopy. RadioGraphics 2004;24(suppl 1):S167–S180.[Abstract/Free Full Text]
  16. Wang L, Mullerad M, Chen HN, et al. Prostate cancer: incremental value of endorectal MR imaging findings for prediction of extracapsular extension. Radiology 2004;232:133–139.[Abstract/Free Full Text]
  17. Sugimura K, Carrington BM, Quivey JM, Hricak H. Postirradiation changes in the pelvis: assessment with MR imaging. Radiology 1990;175:805–813.[Abstract/Free Full Text]
  18. Obuchowski NA. Nonparametric analysis of clustered ROC curve data. Biometrics 1997;53:567–578.[CrossRef][Medline]
  19. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159–174.[CrossRef][Medline]
  20. Grado GL, Collins JM, Kriegshauser JS, et al. Salvage brachytherapy for localized prostate cancer after radiotherapy failure. Urology 1999;53:2–10.[CrossRef][Medline]
  21. Moul JW. Variables in predicting survival based on treating "PSA-only" relapse. Urol Oncol 2003;21:292–304.[Medline]
  22. Rogers E, Ohori M, Kassabian VS, Wheeler TM, Scardino PT. Salvage radical prostatectomy: outcome measured by serum prostate specific antigen levels. J Urol 1995;153:104–110.[CrossRef][Medline]
  23. Hricak H, White S, Vigneron D, et al. Carcinoma of the prostate gland: MR imaging with pelvic phased-array coils versus integrated endorectal–pelvic phased-array coils. Radiology 1994;193:703–709.[Abstract/Free Full Text]
  24. Coakley FV, Qayyum A, Kurhanewicz J. Magnetic resonance imaging and spectroscopic imaging of prostate cancer. J Urol 2003;170(6 pt 2):S69–S76.[CrossRef][Medline]
  25. Mullerad M, Hricak H, Wang L, Chen HN, Kattan MW, Scardino PT. Prostate cancer: detection of extracapsular extension by genitourinary and general body radiologists at MR imaging. Radiology 2004;232:140–146.[Abstract/Free Full Text]
  26. Nudell DM, Wefer AE, Hricak H, Carroll PR. Imaging for recurrent prostate cancer. Radiol Clin North Am 2000;38:213–229.[CrossRef][Medline]
  27. Ikonen S, Karkkainen P, Kivisaari L, et al. Magnetic resonance imaging of clinically localized prostatic cancer. J Urol 1998;159:915–919.[CrossRef][Medline]
  28. Scheidler J, Hricak H, Vigneron DB, et al. Prostate cancer: localization with three-dimensional proton MR spectroscopic imaging—clinicopathologic study. Radiology 1999;213:473–480.[Abstract/Free Full Text]
  29. Tempany CM, Zhou X, Zerhouni EA, et al. Staging of prostate cancer: results of Radiology Diagnostic Oncology Group project comparison of three MR imaging techniques. Radiology 1994;192:47–54.[Abstract/Free Full Text]
  30. Amling CL, Lerner SE, Martin SK, Slezak JM, Blute ML, Zincke H. Deoxyribonucleic acid ploidy and serum prostate specific antigen predict outcome following salvage prostatectomy for radiation refractory prostate cancer. J Urol 1999;161:857–863.[CrossRef][Medline]
  31. Obuchowski NA, Lieber ML, Powell KA. Data analysis for detection and localization of multiple abnormalities with application to mammography. Acad Radiol 2000;7:516–525.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 RadiologyHome page
H. Hricak, P. L. Choyke, S. C. Eberhardt, S. A. Leibel, and P. T. Scardino
Imaging Prostate Cancer: A Multidisciplinary Perspective
Radiology, April 1, 2007; 243(1): 28 - 53.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
O. Rouviere, E. Sala, and H. Hricak
MR Assessment of Recurrent Prostate Cancer after Radiation Therapy
Radiology, February 1, 2007; 242(2): 635 - 637.
[Full Text] [PDF]

Home page
 RadiologyHome page
F. G. Sommer, P. T. Scardino, H. Hricak, M. W. Kattan, and L. Wang
Radiology Research and Rock Soup
Radiology, February 1, 2007; 242(2): 637 - 639.
[Full Text] [PDF]

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 Sala, E.
Right arrow Articles by Hricak, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sala, E.
Right arrow Articles by Hricak, H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE