| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Genitourinary Imaging |
1 From the Department of Clinical Medicine and Applied Biotechnology "D. Campanacci" (C. Testa, R.L., C. Tonon, B.B.), Department of Urology (R.S., F.M., E.B., G.M.), Clinical Department of Radiologic and Histopathologic Science, Diagnostic Imaging Section "V. Bollini" (E.S., M.C., R.C.), Pathologic Anatomy Unit, Institute of Oncology "F. Addarii" (B.C., W.F.G.), and Department of Nuclear Medicine, P.E.T. Center (M.F., N.M., P.C., S.F.), Policlinico S. Orsola-Malpighi, University of Bologna, Via Massarenti 9, 40138 Bologna, Italy; and Department of Radiology, University of California-San Francisco, San Francisco, Calif (J.K.). Received June 19, 2006; revision requested August 21; revision received November 9; accepted December 18; final version accepted January 16, 2007. Address correspondence to R.L. (e-mail: raffaele.lodi{at}unibo.it).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEIMPLICATION FOR PATIENT CAREReferences |
|---|
Materials and Methods: The local ethics committee on human research provided approval and a waiver of informed consent for the retrospective study. MR imaging, 3D MR spectroscopy, and 11C-choline PET/CT results were retrospectively reviewed in 26 men with biopsy-proved prostate cancer (mean age, 64 years; range, 51–75 years) who underwent radical prostatectomy. Cancer was identified as areas of nodular low signal intensity on T2-weighted MR images. At 3D MR spectroscopy, choline-plus-creatine–to–citrate and choline-to-creatine ratios were used to distinguish healthy from malignant voxels. At PET/CT, focal uptake was visually assessed, and maximum standardized uptake values (SUVs) were recorded. Agreement between 3D MR spectroscopic and PET/CT results was calculated, and ability of maximum SUV to help localize cancer was assessed with receiver operating characteristic analysis. Significant differences between positive and negative sextants with respect to mean maximum SUV were calculated with a paired t test.
Results: Sensitivity, specificity, and accuracy were, respectively, 55%, 86%, and 67% at PET/CT; 54%, 75%, and 61% at MR imaging; and 81%, 67%, and 76% at 3D MR spectroscopy. The highest sensitivity was obtained when either 3D MR spectroscopic or MR imaging results were positive (88%) at the expense of specificity (53%), while the highest specificity was obtained when results with both techniques were positive (90%) at the expense of sensitivity (48%). Concordance between 3D MR spectroscopic and PET/CT findings was slight (
= 0.139).
Conclusion: In localizing cancer within the prostate, comparable specificity was obtained with either 3D MR spectroscopy and MR imaging or PET/CT; however, PET/CT had lower sensitivity relative to 3D MR spectroscopy alone or combined with MR imaging.
© RSNA, 2007
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEIMPLICATION FOR PATIENT CAREReferences |
|---|
Results of preliminary studies (5–7) have also demonstrated that carbon 11 (11C)-choline positron emission tomography (PET) is a potentially promising new metabolic diagnostic tool for the detection of prostate cancer. Carbon 11–choline PET had better results than fluorine 18 fluorodeoxyglucose PET (8). Thus, the purpose of our study was to retrospectively compare sensitivity and specificity of MR imaging, 3D MR spectroscopic imaging, combined MR imaging and 3D MR spectroscopic imaging, and 11C-choline PET/computed tomography (CT) for sextant localization of intraprostatic tumor by using histologic findings after radical prostatectomy as the reference standard.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEIMPLICATION FOR PATIENT CAREReferences |
|---|
The protocol included all patients who were deemed to be in a condition suitable for surgery (ie, all patients 
75 years of age and in good physical condition) and who had been referred for a first prostate biopsy on the basis of at least one of the following criteria: (a) total prostate-specific antigen level of more than 10 ng/mL; (b) total prostate-specific antigen level of 2.5–10 ng/mL, with percentage of free prostate-specific antigen less than 13% and/or prostate-specific antigen velocity greater than 0.75 ng·mL–1/y; (c) abnormal digital rectal examination findings; and/or (d) suspicious hypoechoic lesion at transrectal US. Exclusion criteria were (a) coexistent clinically proved cancer and (b) neoadjuvant hormonal treatment, including 5-
-reductase inhibitors (to avoid bias from the effects of these drugs on prostate morphology and metabolism) (9).
All eligible patients (n = 50) underwent MR imaging, 3D MR spectroscopic imaging, and 11C-choline PET/CT before the first prostate biopsy. The mean interval between biopsy and the examinations was 14.6 days ± 4.4 (standard deviation) (range, 10–25 days). Diagnosis of prostate cancer was made in all patients by using results of transrectal US–guided biopsy performed by the same operator (R.S., 5 years of experience) who was blinded to MR imaging, 3D MR spectroscopic imaging, and PET/CT results. Biopsy was performed according to a 12-core systematic biopsy scheme (standard sextant scheme, plus laterally directed samples of the prostate apex, middle, and base) (10). Extra biopsy cores were obtained in regions of suspicious digital rectal examination findings and/or hypoechoic areas at transrectal US and for glands larger than 50 cm3. Every core was placed into a container labeled with the location of the biopsy and was reviewed at a central pathology laboratory.
Twenty-three of 50 men were found to have cancer at first biopsy, and three men had cancer at a repeat biopsy performed within 3 months of the first biopsy. Patients who had a positive finding for prostate cancer at first or successive biopsy underwent a retropubic or video laparoscopic radical prostatectomy with lymph node dissection. The mean interval between radical prostatectomy and imaging examination was 45.5 days ± 9.7 (range, 32–65 days). The final study group consisted of 26 (mean age, 64 years; age range, 51–75 years; mean preoperative prostate-specific antigen level, 13.9 ng/mL; range, 2.5–70 ng/mL) of 50 (52%) patients.
MR Imaging and MR Spectroscopy
MR imaging was performed with a 1.5-T unit (Signa; GE Healthcare, Milwaukee, Wis). Images were acquired by using a body coil for excitation and a flexible endorectal coil (Medrad, Pittsburgh, Pa) combined with a pelvic phased-array coil for signal reception. After acquisition of sagittal scout images, transverse fast spin-echo T2-weighted images were obtained from the seminal vesicles to the prostatic apex (repetition time msec/echo time msec, 5000/102; section thickness, 3 mm with no intersection gap; three signals acquired; field of view, 14 cm; acquisition matrix, 256 x 192; no phase wrap). Coronal and sagittal fast spin-echo T2-weighted images were then acquired (with the same acquisition parameters, except for an echo time of 96 msec and field of view of 16 cm). Images were analytically corrected for the reception profiles of the endorectal and pelvic phased-array coils with software (Prostate Analytical Coil Correction; GE Healthcare).
Three-dimensional MR spectroscopic data were acquired by using the PROSE sequence (GE Healthcare), which is a water- and lipid-suppressed double–spin-echo point-resolved spatially localized spectroscopy (PRESS) sequence, with high-bandwidth spectral-spatial 180° refocusing pulses, and with very selective outer voxel suppression pulses (11). A PRESS volume encompassing the prostate was graphically selected by using the T2-weighted transverse images. To avoid inclusion of the air-tissue interface of the rectum in the PRESS volume, the entire prostate gland was not covered in some cases. Regions of the prostate that were not fully covered in the PRESS volume were not included for the correlation with pathologic findings. Water and lipid suppression were achieved with the combined use of dual-band spectral-spatial radiofrequency 180° pulses (12) and very selective outer voxel saturation pulses (13). Voxel saturation bands were automatically placed around the selected region of interest, and additional saturation bands were graphically placed to conform the shape of the rectangular PRESS volume to the shape of the prostate. Three-dimensional MR spectroscopic imaging data were acquired from the prostate by using 1000/130 and 16 x 8 x 8 phase encoding to yield spectra with a nominal resolution of 0.34 cm3 in approximately 17 minutes. The total MR acquisition time was typically about 50 minutes.
PET/CT Imaging
All PET scans were obtained with a scanner (Discovery LS; GE Medical Systems, Waukesha, Wis). Patients fasted at least 6 hours before PET and received an intravenous injection of 370–555 MBq of 11C-choline. Emission data collection started 5 minutes after injection, and data were acquired for two to three bed positions at 5 minutes per position, typically from the upper pelvis through the midthigh. PET images were reconstructed by using CT for attenuation correction. CT tube parameters were 80 mA and 140 kVp; because of multisection technology, four sections of 5-mm thickness were reconstructed for each tube rotation (rotation time, 0.8 second; table speed, 30 mm per gantry rotation). The CT reconstruction section interval was 4.25 mm to match the PET section thickness.
MR Spectroscopic Imaging Data Processing
Three-dimensional MR spectroscopic data were transferred off-line and processed at a workstation (UltraSparc; Sun Microsystems, Mountain View, Calif) with software developed at the University of California, San Francisco. The raw spectral data were first apodized with a 2-Hz Gaussian function and then Fourier transformed in the time domain and in three spatial domains, after zero-filling to 16 data points in the superoinferior and anteroposterior directions. The spectral data were phased by using the phase of residual water and, when necessary, were manually adjusted. Subsequently, the baseline of the spectra was corrected by using a local nonlinear fit to the nonpeak regions of the spectra. The areas of the choline, creatine, and citrate resonances and the peak area of signal-to-noise ratio for these metabolites were calculated. Only metabolites with a peak area signal-to-noise ratio greater than 5 were used for the identification of cancer. For spectral interpretation, data were overlaid on the corresponding transverse T2-weighted MR images and spatially aligned to optimally overlap prostate anatomy by using software developed at the University of California, San Francisco.
MR Spectroscopic Imaging Analysis
Spectroscopic analyses were retrospectively reviewed by one spectroscopist (R.L., 3 years of experience) who was blinded to clinical and pathologic data, except for the knowledge that patients had prostate cancer. Regions of prostate cancer were metabolically identified on the basis of choline-plus-creatine–to–citrate (CC/C) and choline-to-creatine (C/C) peak area ratios. Because the polyamine peak is not resolvable from creatine and choline peaks in healthy tissue, it was incorporated in the choline-plus-creatine peak area. For the same reason, the C/C peak area ratio was only reliable in regions of cancer, where polyamine levels are dramatically reduced or absent. Voxels were classified by comparing CC/C ratio with the mean CC/C ratio of healthy tissue. The mean healthy CC/C ratio value was defined to be 0.22 ± 0.13 on the basis of results recorded at the University of California, San Francisco (14).
Spectroscopic voxels were considered healthy when the CC/C ratio was 2 or fewer standard deviations above the mean healthy value (ie, CC/C ratio 0.35). Voxels with a CC/C ratio of greater than 2, but 3 or fewer, standard deviations above the mean healthy value (ie, 0.35 < CC/C ratio 0.48) were considered equivocal. Voxels with a CC/C ratio of more than 3 standard deviations above the mean healthy value were considered to be malignant. With equivocal results (CC/C ratio > 2 and 3), C/C ratios were evaluated, and voxels with a C/C ratio of 2 or greater were considered to be positive for cancer. The criteria for determining the presence of transition zone cancers were more restrictive than the criteria for peripheral zone cancers because of the known overlap of metabolic ratios in predominately stromal benign prostatic hyperplasia with those in prostate cancer (15). Namely, voxels were considered to be positive for cancer when CC/C ratios were more than 4 standard deviations above the mean healthy value.
For comparison of pathologic findings with the metabolic data, the prostate was divided into sextants according to the following criteria: The base was defined as the upper third, which extended from the vesical margin of the prostate; the midregion was defined as the central third; the apex was defined as the remaining inferior third; and each third then was divided into two sides—that is, the left and right side. CC/C ratios were assessed for each imaging sextant, and a positive result was defined as at least one voxel having an abnormal CC/C ratio in a sextant with pathologically proved cancer.
MR Image Analysis
MR images were retrospectively evaluated by one radiologist (E.S., with 6 years of experience in the interpretation of MR images of the prostate) who was blinded to clinical and pathologic data, except for the knowledge that patients had prostate cancer. Cancer was identified as areas of nodular low signal intensity on T2-weighted images. Sextant assessment was performed as described above. A true-positive MR imaging result was defined as the presence of a region of nodular low signal intensity in a sextant within the peripheral zone and/or transition zone that also had a pathologically proved positive result for cancer.
PET/CT Image Analysis
All PET images were analyzed with dedicated software (eNTEGRA; Elgems, Haifa, Israel), which allowed review of PET, CT, and fused-image data to determine the exact location of intraprostatic focal uptake. PET/CT data were retrospectively reviewed by one nuclear medicine physician (M.F., 3 years of experience) who was blinded to clinical and pathologic data, except for the knowledge that patients had prostate cancer. PET/CT images were first assessed visually, by using transverse, sagittal, and coronal displays. Any abnormal focus of increased 11C-choline uptake in the prostate gland was identified, as previously described (6). Sextant assessment was performed as described above: Focal uptake was visually assessed for each radiologic sextant, and a positive result was defined as presence of focal uptake within a given sextant. Additionally, the maximum standardized uptake value (SUV) was measured for each sextant so we could perform a receiver operating characteristic analysis.
Histologic Review
All histologic assessments were performed by a pathologist (B.C., 9 years of experience) who was blinded to MR imaging, 3D MR spectroscopic imaging, and PET/CT results. After prostatectomy, the prostate was coated with India ink and fixed in 10% buffered formalin. Histologic examination of a step slice of prostate was performed. The gland was sliced from the base to the apex along the longitudinal axis at 4-mm intervals. After paraffin embedding was performed, microslices were cut, placed on glass slides, and stained with hematoxylin-eosin. Presence, location, and grade of prostate cancer, as well as presence of high-grade prostatic intraepithelial neoplasia (HGPIN), benign prostatic hyperplasia, and acute or chronic prostatitis were determined for each sextant and were sketched on a standardized representation of the gland sections to be correlated with findings from the imaging techniques.
Correlation of MR Imaging, 3D MR Spectroscopic, and PET/CT Findings with Histologic Findings
Assessment of the spatial correspondences between findings of the indicated imaging techniques and reference standard findings was performed by a single investigator (R.S.) who was not a reader of MR images, 3D MR spectroscopic images, or PET/CT data. Sextant matching (including both peripheral and transition zones) between transverse MR images, 3D MR spectroscopic images, PET/CT images, and whole prostate histologic findings was performed. A true-positive finding was defined as concordance of a positive finding at MR imaging, 3D MR spectroscopic imaging, or PET/CT in the same sextant of the gland with a pathologically proved positive finding.
Statistical Analysis
Sensitivity, specificity, and accuracy of MR imaging, 3D MR spectroscopic imaging, and PET/CT for sextant localization of prostate cancer were assessed. Agreement between 3D MR spectroscopic imaging and PET/CT findings was assessed by using the Cohen statistic: values less than 0 indicated poor agreement; values of 0–0.20, slight agreement; values of 0.21–0.40, fair agreement; values of 0.41–0.60, moderate agreement; values of 0.61–0.80, substantial agreement; and values of 0.81–1.00, almost perfect agreement (16). The ability of maximum SUV to help localize cancer to a sextant of the prostate was assessed with receiver operating characteristic analysis. Additionally, the mean maximum SUV of sextants classified as negative for cancer according to the reference standard and the mean maximum SUV of sextants classified as positive for cancer according to the reference standard were calculated for each patient to reduce the data to a single pair of observations for each patient. A paired sample t test was then performed to calculate the significant difference between sextants with positive findings and those with negative findings with respect to mean maximum SUV. A P value of less than .05 was considered to indicate a significant difference. Statistical analyses were performed with software (MedCalc, version 7.3.0.1, 2004; MedCalc, Mariakerke, Belgium).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEIMPLICATION FOR PATIENT CAREReferences |
|---|
|
|
|
|
|
|
|
|
Statistical Findings
MR imaging correctly depicted cancer in 52 (54%) of 97 sextants and showed false-negative findings in the remaining 45 sextants (Table 2). Three-dimensional MR spectroscopic imaging correctly depicted cancer in 74 (81%) of 91 sextants and showed false-negative findings in the remaining 17 of 140 sextants analyzed. PET/CT correctly depicted cancer in 53 (55%) of 97 sextants and showed false-negative findings in the remaining 44 sextants.
|
Among the 156 sextants, histologic findings demonstrated HGPIN in 24. (Findings demonstrated HGPIN in 22 of the 140 sextants analyzed at 3D MR spectroscopic imaging.) Ten (42%) of the 24 sextants with HGPIN had positive findings at MR imaging, eight (36%) of 22 had positive findings at 3D MR spectroscopic imaging, and seven (29%) of 24 had positive findings at PET/CT.
The total number of sextants with false-positive findings was 15 at MR imaging, 16 at 3D MR spectroscopic imaging, and eight at PET/CT. Foci of HGPIN were pathologically proved in 10 (67%) of the 15 sextants with false-positive findings at MR imaging, in eight (50%) of the 16 sextants with false-positive findings at 3D MR spectroscopic imaging, and in seven (88%) of eight sextants with false-positive findings at PET/CT. Only a few of the false-positive findings were inconsistent between the imaging techniques. There were only three sextants between MR imaging and 3D MR spectroscopic imaging, one with inconsistencies in false-positive findings between MR imaging and PET/CT and two with inconsistencies in false-positive findings between 3D MR spectroscopic imaging and PET/CT. Among the 140 sextants evaluated by using 3D MR spectroscopic imaging, PET/CT and 3D MR spectroscopic imaging had concordant findings in 77 (55%) sextants (43 positive findings and 34 negative findings), which resulted in an overall slight agreement ( = 0.139).
For 3D MR spectroscopic imaging compared with PET/CT, 3D MR spectroscopic imaging demonstrated the higher sensitivity, while PET/CT demonstrated the higher specificity. When the findings of MR imaging and 3D MR spectroscopic imaging were combined, the best sensitivity was obtained when MR imaging or 3D MR spectroscopy depicted cancer (80 [88%] of 91 sextants) at the cost of lower specificity (26 [53%] of 49 sextants), while the best specificity was obtained when MR imaging and 3D MR spectroscopy depicted cancer (44 [90%] of 49 sextants) at the cost of lower sensitivity (44 [48%] of 91 sextants).
The area under the receiver operating characteristic curve was 0.72 (95% confidence interval: 0.639, 0.786) for PET/CT localization of cancer within the prostate on the basis of the maximum SUV in each sextant (Fig 4). With a maximum SUV threshold greater than 2.9, PET/CT sextant localization of cancer had a sensitivity of 72% and a specificity of 65%.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEIMPLICATION FOR PATIENT CAREReferences |
|---|
Consistent with results of prior studies (4,11), the best results for localizing cancer within the prostate were obtained by combining MR imaging with 3D MR spectroscopic imaging. While the overall accuracy of combined MR imaging and 3D MR spectroscopic imaging for sextant localization of prostate cancer was similar to that of PET/CT, both good sensitivity (88%) and good specificity (90%) could be achieved with the appropriate combination of anatomic and metabolic information. The method used for analyzing 3D MR spectroscopic imaging data was a simplification of that introduced by Jung et al (19) to better classify individual spectroscopic voxels. This approach was based on the use of the C/C peak area ratio in addition to the CC/C ratio; this approach has been used in other intraglandular localization studies (4). This approach was taken to improve the identification of voxels that contained cancer (elevated choline, reduced polyamines) but that still had a high content of citrate due to the presence of healthy tissue in the same voxel. The use of this approach most likely improved the identification of smaller tumors in our study, thereby increasing sensitivity (81% vs 63%) of cancer localization compared with that of a prior publication on combined MR imaging and 3D MR spectroscopic imaging for cancer localization but resulting in an associated small reduction in specificity (67% vs 75%) (4).
Prior published data regarding the diagnostic value of 11C-choline PET for localizing cancer within the prostate are limited (20). Overall, prior clinical studies on 11C-choline PET for assessment of primary prostate cancer have produced controversial results. In our study, PET/CT demonstrated good specificity (86%) but relatively low sensitivity (55%) when 11C-choline uptake was qualitatively assessed by an experienced nuclear medicine physician. When the PET/CT analysis was performed quantitatively by using the maximum SUV for each sextant and a maximum SUV threshold of 2.9 or higher, sensitivity was substantially increased (72%) at the expense of specificity (65%). The overall accuracy of the more quantitative approach (72%) was similar to that obtained with the qualitative interpretation (67%). There was a large overlap of maximum SUVs between sextants with tumors and those without tumors. It has been proposed that high uptake of 11C-choline is not specific for prostate cancer and that areas of prostate cancer may have SUVs that overlap those obtained from benign hyperplastic prostatic tissue (21). Additionally, prostate cancers are characteristically multifocal, and the limited spatial resolution of both combined MR imaging and 3D MR spectroscopic imaging (approximately 7 mm) and PET (approximately 5 mm) will likely hinder the visualization of small foci of prostate cancer. However, results of our study also suggest that in some cases, even large cancers may not demonstrate increased methyl-11C-choline uptake, which suggests other biochemical causes for not visualizing prostate cancer.
All of the studied imaging techniques demonstrated a relatively large number of false-positive findings, which is possibly related to the presence of HGPIN. This could, in principle, stem from the fact that HGPIN seems to be a precursor lesion to adenocarcinoma of the prostate (22) and could give rise to the same metabolic and 11C-choline uptake behavior as prostate cancer at both combined 3D MR spectroscopic imaging and MR imaging and PET/CT. On the other hand, not all voxels containing HGPIN demonstrated abnormal metabolic patterns and increased 11C-choline uptake suggestive of prostate cancer. In fact, only 36% of sextants with HGPIN alone had positive findings at 3D MR spectroscopy, and only 29% had positive findings at PET/CT.
The only slight concordance between PET/CT and 3D MR spectroscopic imaging depiction of prostate cancer could be explained by several factors. The comparison of the cancer localization abilities could be affected by the intrinsic difference in spatial resolution between the techniques and by difficulties inherent in sextant correlation of imaging data. Furthermore, the mechanism of 11C-choline uptake is not completely understood (23) and seems to be influenced by metabolic factors that may be distinct from tumor presence and aggressiveness; however, there is a large amount of published data supporting a correlation between elevated choline and prostate cancer presence and aggressiveness (11,24).
Our imaging findings for tumors arising in the transition zone of the gland are in agreement with those reported in literature. Conventional diagnostic techniques, such as digital rectal examination and transrectal US–guided multiple biopsies, often do not depict cancer within the transition zone (18), and MR imaging is limited in the diagnosis of transition zone tumors because of the overlap of T2-weighted signal intensity between prostate cancer and predominately stromal benign prostatic hyperplasia. Similarly, at 3D MR spectroscopy, there is substantial overlap between the metabolic patterns of prostate cancer and those of predominately stromal benign prostatic hyperplasia (25). Specifically, predominately stromal benign prostatic hyperplasia is associated with low citrate levels and, in some cases, elevated choline. With use of more stringent metabolic criteria for diagnosing cancer in the transition zone (CC/C ratio > 4), 3D MR spectroscopy correctly depicted many transition zone cancers (80%). More important, 11C-choline PET/CT correctly depicted transition zone tumors in all cases. This finding suggests a role for PET/CT in the identification of transition zone tumors, particularly those in the anterior portion of the central gland, which are often missed at transrectal US–guided biopsy and at combined MR imaging and 3D MR spectroscopic imaging.
We are not aware of previously published studies that have evaluated the accuracy of combined MR imaging and 3D MR spectroscopic imaging and PET/CT for sextant tumor localization with step-slice histologic findings of the prostate as the standard of reference. One study (7) involved investigation of the value of MR imaging, single-voxel hydrogen 1 (1H) MR spectroscopy, and PET for diagnosing whether cancer existed on one side of the prostate. Results of this study revealed sensitivities of 100% for PET, 60% for MR imaging, and 65% for single-voxel 1H MR spectroscopy. It is difficult to compare the sensitivity results of our study with those of Yamaguchi et al, because they did not perform a sextant analysis and used single-voxel spectroscopy and not 3D MR spectroscopy. Moreover, no specificity values were reported because spectroscopy was performed only in regions of low T2-weighted signal intensity at MR imaging or regions of biopsy-proved cancer.
Our study had limitations. Due to technical reasons (gland coverage and lipid contamination), 16 sextants had to be eliminated from 3D MR spectroscopic data analysis, and this could have affected the comparison between the imaging techniques. Another limitation of our study was the use of a sextant, rather than a cancer lesion, as the unit of analysis, and our study did not take into account any size criteria in comparing the imaging and pathologic data (26). In addition, our analysis was retrospective and did not automatically take into account patients who underwent MR imaging, 3D MR spectroscopic imaging, and 11C-choline PET/CT but did not undergo radical prostatectomy because transrectal US–guided biopsy results were negative. Finally, because our population did not consist of consecutive patients, this could have influenced the final cohort to be characterized by a rather high volume of disease, with a possible effect on the generalizability of the results.
In conclusion, comparable specificity was obtained by using either combined MR imaging and 3D MR spectroscopic imaging or 11C-choline PET/CT for localizing cancer to a sextant of the prostate; however, 11C-choline PET/CT demonstrated much lower sensitivity relative to 3D MR spectroscopic imaging alone or to combined MR imaging and 3D MR spectroscopic imaging.
|
ADVANCE IN KNOWLEDGE |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEIMPLICATION FOR PATIENT CAREReferences |
|---|
|
IMPLICATION FOR PATIENT CARE |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEIMPLICATION FOR PATIENT CAREReferences |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: CC/C = choline-plus-creatine to citrate • C/C = choline to creatine • HGPIN = high-grade prostatic intraepithelial neoplasia • PRESS = point-resolved spatially localized spectroscopy • SUV = standardized uptake value • 3D = three-dimensional
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, C. Testa, R.L., C. Tonon; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, C. Testa, R.S., R.L., M.F., J.K., F.M., C. Tonon, N.M., R.C., B.B.; clinical studies, R.S., R.L., E.S., B.C., M.F., F.M., E.B., C. Tonon, N.M., P.C., S.F., M.C., W.F.G., G.M.; statistical analysis, C. Testa, M.F., J.K., P.C.; and manuscript editing, C. Testa, R.L., B.C., M.F., J.K., P.C., B.B.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEIMPLICATION FOR PATIENT CAREReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. Turkbey, P. S. Albert, K. Kurdziel, and P. L. Choyke Imaging Localized Prostate Cancer: Current Approaches and New Developments Am. J. Roentgenol., June 1, 2009; 192(6): 1471 - 1480. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. C. Weinreb, J. D. Blume, F. V. Coakley, T. M. Wheeler, J. B. Cormack, C. K. Sotto, H. Cho, A. Kawashima, C. M. Tempany-Afdhal, K. J. Macura, et al. Prostate Cancer: Sextant Localization at MR Imaging and MR Spectroscopic Imaging before Prostatectomy--Results of ACRIN Prospective Multi-institutional Clinicopathologic Study Radiology, April 1, 2009; 251(1): 122 - 133. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. L. Langer, T. H. van der Kwast, A. J. Evans, L. Sun, M. J. Yaffe, J. Trachtenberg, and M. A. Haider Intermixed Normal Tissue within Prostate Cancer: Effect on MR Imaging Measurements of Apparent Diffusion Coefficient and T2--Sparse versus Dense Cancers Radiology, December 1, 2008; 249(3): 900 - 908. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Plathow and W. A. Weber Tumor Cell Metabolism Imaging J. Nucl. Med., June 1, 2008; 49(Suppl_2): 43S - 63S. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. K. Kuhl, F. Traber, J. Gieseke, W. Drahanowsky, N. Morakkabati-Spitz, W. Willinek, M. von Falkenhausen, C. Manka, and H. H. Schild Whole-Body High-Field-Strength (3.0-T) MR Imaging in Clinical Practice * Part II. Technical Considerations and Clinical Applications Radiology, April 1, 2008; 247(1): 16 - 35. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
