MR imaging confirms early promise in bladder cancer

Urinary bladder carcinoma is the second most common malignant tumor in the urogenital tract. This cancer causes 5000 deaths each year in Germany and 10,400 in the U.S., affecting men more often than women. The patient population is predominantly elderly, with a mean age of 70 years old.

Transitional cell carcinoma accounts for 90% to 95% of bladder cancers. Etiological factors include occupational exposure to chemicals such as aniline dyes and chlorinated hydrocarbons, cigarette smoking, analgesic abuse (phenacitin), and pelvic irradiation.^1-3 The prognosis depends on several factors, especially local staging and tumor grading. Accurate preoperative staging is the most important factor in determining patient management.⁴

Pelvic MRI is now considered the best imaging technique for staging urinary bladder carcinoma. Several groups have shown the high accuracy of MRI for local tumor staging. Their reported sensitivities and specificities range from 70% to 95% in the differentiation of locally advanced tumor.^5-10 The efficiency of MRI in delineating initial bladder wall infiltration and tumor vascularity, using fast gadolinium-enhanced techniques, has also been proven.^11-13

Dynamic contrast-enhanced MRI as a means of monitoring chemotherapy response has also been assessed.¹⁴ Gadolinium-assisted MRI demonstrated a high accuracy (95%), sensitivity (93%), and specificity (100%) in distinguishing responders from nonresponders. A separate investigation into the role of MRI in distinguishing residual or recurrent tumor from radiation changes documented negative predictive values of 100% and 93% for tumor recurrence at four and 12 months, respectively.¹⁵

Virtual cystoscopy can be performed with MRI, as with CT. One study reported sensitivities of 100% for tumors larger than 1 cm and 88.9% for tumors smaller than 1 cm.¹⁶ Another study of MR-based virtual cystography demonstrated an accuracy of 85% when differentiating superficial carcinomas from invasive carcinomas. An accuracy of 82% was recorded for the distinction of organ-defined tumors from non-organ-defined tumors.¹⁷

MR IMAGING ON TRIAL

The urologic center at Allgemeines Krankenhaus Harburg is known for the surgical treatment of patients with bladder and prostate cancer. We worked with our colleagues at this center to perform a prospective study into MRI for bladder cancer.

The study included 68 patients undergoing cystectomy and 54 patients undergoing transurethral resection (TUR). Imaging was performed on a 1.5T scanner (Philips Medical Systems, Gyroscan ACS-NT). Our strategy included a T1-weighted turbo spin-echo axial sequence for pelvic lymph node staging. The bladder was imaged using ultraturbo spin-echo technique, fat-suppressed MRI, and multiplanar gadolinium-enhanced sequences. MRI findings were correlated with cystectomy and TUR results for T and N staging (see table).

MRI revealed a sensitivity of 96.1% and specificity of 68.8% when differentiating the muscle invasion stage T2 in 68 patients with cystectomy (positive predictive value: 90.9%; negative predictive value: 81.6%; accuracy: 89.7%) (Figure 1). The modality also demonstrated a sensitivity of 96.7% and specificity of 78.9% for the detection of T3b extravesical tumors (PPV: 78.4%; NPV: 96.7%; accuracy: 86.8%) (Figures 2 to 4).

Lymph node metastases were present in 16.2% of patients. MRI demonstrated a sensitivity of 36.4% and specificity of 94.7% for detection of the N-positive stage (PPV: 57.1%; NPV: 88.5%; accuracy: 85.3%). In the TUR group, MRI had a sensitivity of 88% and specificity of 74.2% for the differentiation of invasive carcinoma (PPV: 66.7%; NPV: 92%; accuracy: 79.6%).

We overestimated the tumor stage in the cystectomy group, classifying 13 patients as T2 when histology showed T1. MRI led us to stage one patient as T2, in contrast to the histological diagnosis of Ta and multifocal Tis.

We also underestimated the number of metastases during lymph node staging, detecting just four of the 11 positive patients. Three patients with 6 to 8-mm para-iliacal lymph nodes whom we classified as negative turned out to be positive on lymphadenectomy, while we overlooked a 1.1-cm positive lymph node in one patient. In two other cases, we classified a 1.2-cm and a 1.1-cm lymph node as positive, though these were negative on histology. TUR underestimated the tumor stage in the cystectomy group in 17% of cases.

WIDER VIEW

Urinary bladder carcinoma is seen frequently, with increasing incidence. Seventy percent of patients present with superficial tumors, while 30% have carcinomas invading the muscle. Superficial tumors are most often treated with local endoscopic resection, either with or without chemotherapy. Invasive carcinomas are treated with radical cystectomy, radiation therapy, and chemotherapy.

Significant improvements in the clinical outcome of bladder cancer patients over the past 25 years can be attributed to advances in surgical technique. Nonetheless, higher stage local tumors and lymph node metastases still lead to increased tumor recurrence and higher mortality. The prognosis of bladder carcinoma and patient management are dependent on local tumor extension and tumor grading, frequency of tumor recurrence, and clinical parameters such as age and risk factors.⁴ MRI has proven itself to be the best imaging method for diagnosing local tumor extension. Its success in differentiating invasive carcinoma is due to high tissue contrast, multiplanar reformatting, contrast strategies, and progress in coil technology.^11-13

CT is less reliable for local tumor staging. One study documented the overall accuracy of CT as 54.9%, with understaging reported as 39% and overstaging as 6.1%.¹⁸ The authors concluded that CT tended to significantly understage advanced tumor. Imaging findings failed to alter surgical management in nearly all cases. Another group reported a sensitivity of 89%, specificity of 95%, and accuracy of 93% when using CT to detect extravesical tumor extension. Their study was performed on a multislice CT system and involved dynamic and delayed scans.¹⁹

We do not consider CT to be substantially helpful in local tumor and lymph node staging of bladder carcinoma. The only exception is the staging of retroperitoneal lymph nodes. We will be looking for studies comparing MSCT and MRI in larger patient populations, though it is unlikely that CT will replace MRI as our first-line imaging strategy. MRI offers superior soft-tissue contrast and can visualize perivesical infiltration better than CT, especially with fat-suppressed techniques. Its main limitations involve differentiation of bladder wall layers and distinction of subtle inflammation around bladder carcinomas from muscle invasion or perivesical tumor infiltration.

Our results confirm the diagnostic validity of MRI in bladder cancer, though our specificity in the detection of muscle invasion was low (68.8%). Our imaging strategy was not as sophisticated as that adopted by the leading groups. We did not, for example, perform dynamic contrast-enhanced imaging.

From a clinical point of view, discrepancy sometimes occurs among the guidelines, the low number of bladder MR examinations, and evidence of MR's accuracy. German Society of Urology guidelines for the diagnosis and treatment of urinary bladder carcinoma classify both MRI and CT as only minimally useful in assessing locally advanced tumor.²⁰ Yet clinical staging and TUR often underestimate tumor extension. Clinical staging has proven to be accurate in 25% to 50% of invasive carcinomas.⁴ Transabdominal ultrasound is also inaccurate for local staging. Intravesical ultrasound is more promising, but it cannot assess deeply invasive tumors and lymph nodes. Cystoscopic examination remains the basis for diagnosis, with TUR-guided biopsy necessary for histology and tumor grading.

MRI is the ideal technique to increase the accuracy of local staging in bladder carcinoma. Nevertheless, the clinical value of MRI is reduced by the fact that it is not very helpful in the lower tumor stages. MRI in most studies is not accurate enough to rule out lymph node metastases. Many false-negative cases in our study were caused by lymph nodes that had not increased in size. Blood-pool contrast agents aid lymph node staging by providing signal parameters that detect tumor infiltration in normal-sized lymph nodes. These agents should also help identify lymph nodes that are larger only because of inflammation.

MRI is the most accurate imaging procedure for local tumor staging of urinary bladder carcinoma. Differentiation of the prognosis-relevant locally advanced tumor is possible with high accuracy, but specificity in the detection of the early muscle invasive stage is reduced.

It is not practical to begin staging all patients with bladder cancer with MRI. Such a strategy would incur significant costs. MRI is best reserved for evaluating locally advanced and recurrent tumor.

DR. KUHN and DR. HUTTMANN are radiologists, and PROF. DR. GROSS-FENGELS is head of the radiology department, all at the Allgemeines Krankenhaus Harburg, Eissendorfer Pferdeweg in Hamburg, Germany. PROF. DR. FISCH is head of the urology department in the Hamburg Urology Center, Asklepios Klinik Harburg.

References1. Hall RR. Superficial bladder cancer. Br Med J 1994;308(6933):910-913.
2. See www.cancer.org/downloads/STT/CAFF2006PWSecured.pdf3. Rockall AG, Newman-Sanders AP, al-Kutoubi MA, Vale JA. Haematuria. Postgrad Med J 1997;73(857):129-136.
4. Macvicar AD. Bladder cancer staging. BJU Int 2000;86(Suppl 1):111-122.
5. Amendola MA, Glazer GM, Grossman HB, et al. Staging of bladder carcinoma: MRI-CT-surgical correlation. AJR 1986;146(6):1179-1183.
6. Neuerburg JM, Bohndorf K, Sohn M, et al. Urinary bladder neoplasms: evaluation with contrast-enhanced MR imaging. Radiology 1989;172(3):739-743.
7. Scattoni V, Da Pozzo LF, Colombo R, et al. Dynamic gadolinium-enhanced magnetic resonance imaging in staging of superficial bladder cancer. J Urol 1996;155(5):1594-1599.
8. Barentsz JO, Jager GJ, van Vierzen PB, et al. Staging urinary bladder cancer after transurethral biopsy: value of fast dynamic contrast-enhanced MR imaging. Radiology 1996;201(1):185-193.
9. Barentsz JO, Jager GJ, Witjes JA, Ruijs JH. Primary staging of urinary bladder carcinoma: the role of MRI and a comparison with CT. Europ Radiol 1996;6(2):129-133.
10. Tekes A, Kamel IR, Imam K, et al. MR imaging features of transitional cell carcinoma of the urinary bladder. AJR 2003;180(3):771-777.
11. Barentsz JO, Engelbrecht M, Jager GJ, et al. Fast dynamic gadolinium-enhanced MR imaging of urinary bladder and prostate cancer. J Magn Reson Imag 1999;10(3):295-304.
12. Barentsz JO, Engelbrecht M, Witjes JA, et al. MR imaging of the male pelvis. Europ Radiol 1999;9(9)1722-1736.
13. Barentsz JO, Jager GJ, Witjes JA. MR imaging of the urinary bladder. Magn Reson Imaging Clin N Am 2000;8(4):853-867.
14. Barentsz JO, Berger-Hartog O, Witjes JA, et al. Evaluation of chemotherapy in advanced urinary bladder cancer with fast dynamic contrast-enhanced MR imaging. Radiology 1998;207(3):791-797.
15. Dobson MJ, Carrington BM, Collins CD, et al. The assessment of irradiated bladder carcinoma using dynamic contrast-enhanced MR-imaging. Clin Radiol 2001;56(2):94-99.
16. Bernhardt TM, Schmidt H, Philipp C, et al. Diagnostic potential of virtual cystoscopy of the bladder: MRI vs.CT. Preliminary report. Europ Radiol 2003;13(2):305-312.
17. Tekes A, Kamel I, Imam K, et al. Dynamic MRI of bladder cancer: evaluation of staging accuracy. AJR 2005;184(1):121-127.
18. Paik M, Scolieri J, Brown S, et al. Limitations of computerized tomography in staging invasive bladder cancer before radical cystectomy. J Urol 2000;163(6):1693-1696.
19. Kim JK, Park SY, Ahn HJ, et al. Bladder cancer: analysis of multi-detector row helical CT enhancement pattern and accuracy in tumor detection and perivesical staging. Radiology 2004;231(3):725-731.
20. Leitlinien zur Diagnostik und Therapie des Harnblasenkarzinoms Urologe(A) 1998;37:440-457 Springer Verlag. (Guidelines of the German Society of Urology in association with the German Cancer Society)