- Email: [email protected]
MR Cystography for Bladder Tumor Detection
IMAGING
97
behavior. The collective immunohistochemical expression of suppressor genes, growth factor, and proliferation activity marker has not been previously studied in this disease. Formalin-fixed, paraffin-embedded archival tissues from 12 cases of nested variant of urothelial carcinoma were stained with monoclonal antibodies to p21, p27, p53, EGF-R, and bcl-2, as well as the proliferation marker MIB-1. The area of predominant immunoreactivity was also evaluated. The pattern of immunostaining was compared with the clinical parameters. p21 was positive in 10 of 12 cases and located at the deepest portion of the tumor in 5 of 10 positive cases. Immunoreactivity for p27 was seen in 11 of 12 cases and limited to the superficial portion of the tumor in 9 of 11 positive cases. Only 3 and 2 of 12 cases were positive for p53 and bcl-2, respectively. MIB- 1 immunoreactivity ranged from 2 to 35% of the neoplastic cells, with most tumors showing a proliferation index of ⬎15%. Follow-up ranged from 3 to 30 months (mean, 17.6 mo). All patients except one were alive, although three patients developed metastases. Nested variant of urothelial carcinoma is a deceptively benign-appearing neoplasm with potential of deep invasion and metastases. Immunohistochemically, nested variant of urothelial carcinoma shares some features with high-risk conventional urothelial carcinomas, such as loss of p27 expression and high proliferation index. Nevertheless, p53, bcl-2, or EGF-r immunoreactivity is not frequently seen. Editorial Comment: Mixed histology urothelial tumors represent an important subset of bladder cancers. My clinical impression is that we are seeing more of such tumors, whether from increasing recognition by pathologists or potentially cell type selection pressures from intravesical therapies. The authors describe the nested variant, which is deceptively benign appearing, and the micropapillary variant, which may be a form of glandular differentiation. Both of these variants imply a more aggressive tumor. Small cell, sarcomatoid and plasmacytoid differentiation also must be recognized. The pathologist must describe these variants, and urologists must have an appreciation of their significance. James E. Montie, M.D.
IMAGING
MR Cystography for Bladder Tumor Detection A. Beer, B. Saar, N. Zantl, T. M. Link, R. Roggel, S.-L. Hwang, H. Schwaibold and E. J. Rummeny, Departments of Clinical Radiology and Urology, Klinikum Rechts der Isar, Technische Universitaet Munich, Munich, Germany Eur Radiol, 14: 2311–2319, 2004 The purpose of the study was to assess the diagnostic performance of MR cystography with virtual cystoscopic and multiplanar reconstructions for detection of malignant bladder tumors. Thirty-two patients with 43 bladder tumors previously confirmed by cystoscopy (mean size 2.5 cm; 0.4 –9.1 cm;) were examined at 1.5 T with a three-dimensional T2-weighted turbo spin echo sequence (TR ⫽ 2911 ms, TE ⫽ 500 ms, echo train length 256). Virtual cystoscopic reconstructions (VC) and multiplanar reconstructions (MPR) were obtained and analyzed separately by three radiologists without knowledge of the tumor location. Intraoperative or cystoscopic findings served as standard of reference. Sensitivities and specificities were calculated using a receiver-operating characteristic analysis with five levels of confidence. Area-under-curve values were similar for MPRs (0.952), VC (0.932) and the combination of both methods (0.954). Optimal sensitivity was 92.3% for MPR and 90.7% for VC, respectively, with a specificity of 91.1% for MPR and 90.4% for VC. The combination of MPR and VC resulted in a sensitivity of 90.7% and specificity of 94.0%. MR cystography is a promising, completely non-invasive technique for the detection of bladder lesions with a high diagnostic performance. Editorial Comment: This article assesses the diagnostic performance of magnetic resonance (MR) cystography with virtual cystoscopy and multiplanar reconstructions to evaluate for malignant bladder tumors. The authors demonstrate beautiful multiplanar reconstructions and corresponding virtual cystoscopy images, some with cystoscopic correlation. Prior studies have looked at virtual cystoscopy using computerized tomography data sets. There has been little information on the role of MR and virtual cystoscopy. Although preliminary, this study of 32 patients with 43 bladder tumors shows promising results. This MR cystography study was performed without ionizing radiation, intravenous contrast or CO2 insufflation in the bladder. Intravenous contrast with gadolinium has shown to be beneficial in assessing the depth of bladder wall involvement and perivesical extent of the tumor. Also, it is
98
IMAGING noteworthy that virtual cystoscopy will not be beneficial in patients who have flat tumors. Additionally, carcinoma in situ cannot be detected on cystography because these images give no information on the color of the mucosa. Urologists may wish to use MR cystography imaging on a limited basis, perhaps in conjunction with or interwoven between cystoscopy examinations, to increase the time interval between interventional procedures. MR cystography may lead to a decrease in the number of followup cystoscopic examinations and an increase in the interval between cystoscopies. Cary Siegel, M.D.
97
behavior. The collective immunohistochemical expression of suppressor genes, growth factor, and proliferation activity marker has not been previously studied in this disease. Formalin-fixed, paraffin-embedded archival tissues from 12 cases of nested variant of urothelial carcinoma were stained with monoclonal antibodies to p21, p27, p53, EGF-R, and bcl-2, as well as the proliferation marker MIB-1. The area of predominant immunoreactivity was also evaluated. The pattern of immunostaining was compared with the clinical parameters. p21 was positive in 10 of 12 cases and located at the deepest portion of the tumor in 5 of 10 positive cases. Immunoreactivity for p27 was seen in 11 of 12 cases and limited to the superficial portion of the tumor in 9 of 11 positive cases. Only 3 and 2 of 12 cases were positive for p53 and bcl-2, respectively. MIB- 1 immunoreactivity ranged from 2 to 35% of the neoplastic cells, with most tumors showing a proliferation index of ⬎15%. Follow-up ranged from 3 to 30 months (mean, 17.6 mo). All patients except one were alive, although three patients developed metastases. Nested variant of urothelial carcinoma is a deceptively benign-appearing neoplasm with potential of deep invasion and metastases. Immunohistochemically, nested variant of urothelial carcinoma shares some features with high-risk conventional urothelial carcinomas, such as loss of p27 expression and high proliferation index. Nevertheless, p53, bcl-2, or EGF-r immunoreactivity is not frequently seen. Editorial Comment: Mixed histology urothelial tumors represent an important subset of bladder cancers. My clinical impression is that we are seeing more of such tumors, whether from increasing recognition by pathologists or potentially cell type selection pressures from intravesical therapies. The authors describe the nested variant, which is deceptively benign appearing, and the micropapillary variant, which may be a form of glandular differentiation. Both of these variants imply a more aggressive tumor. Small cell, sarcomatoid and plasmacytoid differentiation also must be recognized. The pathologist must describe these variants, and urologists must have an appreciation of their significance. James E. Montie, M.D.
IMAGING
MR Cystography for Bladder Tumor Detection A. Beer, B. Saar, N. Zantl, T. M. Link, R. Roggel, S.-L. Hwang, H. Schwaibold and E. J. Rummeny, Departments of Clinical Radiology and Urology, Klinikum Rechts der Isar, Technische Universitaet Munich, Munich, Germany Eur Radiol, 14: 2311–2319, 2004 The purpose of the study was to assess the diagnostic performance of MR cystography with virtual cystoscopic and multiplanar reconstructions for detection of malignant bladder tumors. Thirty-two patients with 43 bladder tumors previously confirmed by cystoscopy (mean size 2.5 cm; 0.4 –9.1 cm;) were examined at 1.5 T with a three-dimensional T2-weighted turbo spin echo sequence (TR ⫽ 2911 ms, TE ⫽ 500 ms, echo train length 256). Virtual cystoscopic reconstructions (VC) and multiplanar reconstructions (MPR) were obtained and analyzed separately by three radiologists without knowledge of the tumor location. Intraoperative or cystoscopic findings served as standard of reference. Sensitivities and specificities were calculated using a receiver-operating characteristic analysis with five levels of confidence. Area-under-curve values were similar for MPRs (0.952), VC (0.932) and the combination of both methods (0.954). Optimal sensitivity was 92.3% for MPR and 90.7% for VC, respectively, with a specificity of 91.1% for MPR and 90.4% for VC. The combination of MPR and VC resulted in a sensitivity of 90.7% and specificity of 94.0%. MR cystography is a promising, completely non-invasive technique for the detection of bladder lesions with a high diagnostic performance. Editorial Comment: This article assesses the diagnostic performance of magnetic resonance (MR) cystography with virtual cystoscopy and multiplanar reconstructions to evaluate for malignant bladder tumors. The authors demonstrate beautiful multiplanar reconstructions and corresponding virtual cystoscopy images, some with cystoscopic correlation. Prior studies have looked at virtual cystoscopy using computerized tomography data sets. There has been little information on the role of MR and virtual cystoscopy. Although preliminary, this study of 32 patients with 43 bladder tumors shows promising results. This MR cystography study was performed without ionizing radiation, intravenous contrast or CO2 insufflation in the bladder. Intravenous contrast with gadolinium has shown to be beneficial in assessing the depth of bladder wall involvement and perivesical extent of the tumor. Also, it is
98
IMAGING noteworthy that virtual cystoscopy will not be beneficial in patients who have flat tumors. Additionally, carcinoma in situ cannot be detected on cystography because these images give no information on the color of the mucosa. Urologists may wish to use MR cystography imaging on a limited basis, perhaps in conjunction with or interwoven between cystoscopy examinations, to increase the time interval between interventional procedures. MR cystography may lead to a decrease in the number of followup cystoscopic examinations and an increase in the interval between cystoscopies. Cary Siegel, M.D.