| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Universtiy Hospitals of Geneva, Switzerland
In an article in the December 2004 Journal of the American Medical Association [1], a prospective multicenter investigation of the International Breast MR Consortium conducted at 14 university hospitals in North America and Europe, including 821 patients, determined the accuracy of breast MRI for the detection of breast cancer—the first cause of cancer in women—in patients with suspicious clinical or mammographic findings. The results and conclusion of this study are the following: Although the positive predictive value of MRI is greater than that of mammography (72.4% vs 52.8%, respectively), MRI does not obviate subsequent tissue sampling, because its specificity is only moderate (67.7%) independent of breast density, tumor type, and menopausal status.
An article in the August 2004 British Journal of Radiology [2], a systematic review of radiological imaging for hepatocellular carcinoma— the third leading cause of cancer-related death worldwide—in cirrhotic patients, based on 997 articles published between 1995 and 2001, concluded that the evidence for choosing the best imaging technique for characterizing nodules in cirrhotic patients is inadequate.
In many circumstances and for most suspicious lesions, few scientific data clearly show that imaging-based findings can allow a radiologist to invariably or confidently respond to this simple daily practice question: Is the lesion a cancer or not?
Imaging is not as good as we would like it to be. Fortunately, when combining clinical and biologic data with morphologic information provided by imaging, a high diagnostic probability can be given in most circumstances.
Are there new horizons in oncologic imaging? The development of the exciting field of molecular imaging, based on nuclear imaging as well as MRI, may offer new opportunities [3].
So far, for almost all types of tumors, a histopathologic diagnosis is considered mandatory before starting therapy. It includes tumor architecture, immunohistochemical staining, tumor grading, and hormonal receptor evaluation in some cases. Histopathologic diagnosis is one goal of tumor tissue sampling. Another emerging goal of tumor tissue sampling is to assess new markers of tumors that are coming from molecular biology (proteomic and genomic profiling)—that is, the cellular and molecular basis of malignant transformation [4]. Molecular biology can, first, improve subclassification of tumors; second, derive prognostic markers of the behavior of an individual tumor for a more rational basis for treatment; and, third, identify potential molecular therapeutic targets.
A thorough histopathologic and biologic tumor analysis relies on adequate tissue sampling. Most tumors not accessible by endoscopy can be biopsied by percutaneous imaging guidance. Several publications have shown that percutaneous multisampling coaxial core needle biopsy by imaging guidance is an accurate, safe, and outpatient procedure that allows representative tumor sampling to be obtained in many circumstances.
The future of oncologic imaging may rely on new imaging technology, such as molecular imaging, but, from now on, multisampling biopsy with imaging-guidance expertise can allow a radiologist to be involved in a multidisciplinary oncology team to contribute to the moving field of oncologic diagnosis.
 Top References |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
