BBI Contributing Writer

CHICAGO The late-November/early-December annual meeting of the Radiological Society of North America (RSNA; Oak Brook, Illinois) once again reflected the clinical and economic importance of medical imaging in the world's healthcare systems. This 90th annual assembly of radiologists, supplier representatives and others opened the doors of McCormick Place.

The organization's name has become a misnomer, as attendance by participants outside North America has increased substantially over the last 10 years. For example, the number of pre-registered North American professionals was up 7% from the prior year, while non-North American professionals had increased by 11%. Abstract submissions to this annual meeting reflected the same trend. Total professional registration at the meeting's close was up 4% from a year ago to about 26,200. Total attendance was about 59,600, reflecting a 3% increase. These totals assure the RSNA annual gathering its place as the world's largest international medical meeting.

Image-guided cancer therapy

As noted in these pages a year ago (BBI, January 2004), minimally invasive methods for treating cancers continue to proliferate in the hands of interventional radiologists for important clinical and economic reasons. Patients and their clinicians seek improved quality and length of life while radiologists seek expansion of their therapeutic responsibilities. This winter's RSNA meeting amply demonstrates the continued progress being made along these lines.

Conventional cancer therapies notably chemotherapy, surgery and classical radiation therapy are less than ideal. Practitioners want therapies that are less invasive, less painful, less costly, quicker to accurately assess tumor response, able to be applied repetitively, and hopefully more effective by some measure. Highly precise, image-guided approaches are now known to offer many of these patient and societal benefits while at the same time expanding radiology's therapeutic role. Sessions dealing with ablation, chemoembolization, focused ultrasound and intensity-modulated radiation therapy (IMRT) nicely captured these benefits and current methodological shortcomings.

A special focus session on image-guided methods was especially revealing. Damian Dupuy, MD, of Brown University (Providence, Rhode Island), reviewed tumor ablation methods via radio frequency, microwaves and cryotherapy. He noted that while ablative technologies are still developmental, their improved quality of life and cost-effectiveness have already been conclusively demonstrated. These methods are applicable to treating localized tumors to achieve palliation. Widespread adoption will require insurance reimbursement, which is not currently available.

Michael Soulen, MD, of the University of Pennsylvania (Philadelphia) discussed chemoembolization methods achieved through targeted vascular delivery of microspheres, magnetically targeted carriers or drug-eluting beads. The delivered agents can be classical or genomic. He said he favors bead-based methods while acknowledging that, at least so far, some cancer cells remain following therapy. Clare Tempany, MD, of Brigham & Women's Hospital (Boston), described the use of focused ultrasound in combination with real-time thermometry and magnetic resonance imaging to treat uterine fibroids and breast cancer. Results in treating breast cancer are especially promising. In addition, early work is under way in treating cancers of the liver, brain and prostate. James Hevezi, PhD, of the Cancer Therapy & Research Center (San Antonio), summarized IMRT and brachytherapy applications of radiation. IMRT has become feasible owing to the precision of time-dependent imaging guidance. On the other hand, brachytherapy requires only initial accurate targeting, as the radiating seeds tend to remain at the target site. Hevezi said he is enthusiastic about brachytherapy for prostate cancer (now a standard treatment) and for many other anatomical sites.

Additional presentations in this field were noteworthy. More than 70 contributions at the meeting were dedicated to the use of radio frequency ablation (RFA) to treat various cancers. Most of the work to date has been in RFA in the liver. L. Solbiati, MD, of the Busto Arsizio General Hospital (Busto Arsizio, Italy), and colleagues reported on treatment of liver colorectal metastases. They noted constantly increasing survival of treated patients with this condition and very low morbidity rates. The three-year survival rate (59.7%) has equaled that of the best surgical series reported in the literature. D.S. Lu, MD, and coworkers found that percutaneous RFA is an effective bridge to transplantation for patients with hepatocellular carcinoma. Riccardo Lencioni, MD, of the University of Pisa (Italy) presented a total of five papers describing survival rates for the treatment of liver metastases from colorectal and breast cancers, local tumor control rates for the treatment of certain non-small-cell lung cancers, and survival rates for the treatment of lung cancer and primary liver cancer.

Kemal Tuncali, MD, and co-authors at Brigham and Women's Hospital reported satisfying results in treating liver cancer with cryotherapy, guided and monitored by MRI. Atsuhiro Nakatsuka, MD, of the Mie University School of Medicine (Japan) and co-workers noted that hepatocellular carcinoma that has metastasized to bone could be successfully treated by RFA performed under CT fluoroscopic guidance. S.B. Solomon, MD, of Johns Hopkins University Hospital and associates reported intermediate results of their RFA treatment of renal masses. They found that this treatment is a promising minimally invasive procedure for patients with solid renal tumors and high-risk medical conditions. Christiane Marx, MD, of the Friedrich-Schiller-University Jena (Jena, Germany) and co-authors demonstrated success in treating breast cancer with cryotherapy under ultrasound guidance. Tumors smaller than 16 mm in diameter were completely destroyed, with a 12% risk of leaving ductal carcinoma in situ in the surrounding tissue. Dupuy and colleagues at Brown University (Providence, Rhode Island) reported on the use of image-guided RFA in treating lung cancer patients. They performed 155 procedures over six years on 126 medically inoperable patients. The outpatient RFA procedure was found to effectively control small tumors. While the procedure does not necessarily extend patient lives, it can improve the quality of those lives. RFA is particularly beneficial for patients who have a lung cancer recurrence or re-growth and cannot have additional surgery or radiation therapy. Finally, Feng Wu, MD, and colleagues at the University of Medical Sciences in China found that focused-ultrasound ablation is safe and effective in the treatment of liver tumors up to 14 centimeters in diameter. They suggest that the technique should be considered as a palliative treatment in patients with advanced-stage hepatocellular carcinomas.

Breast imaging

The lingering unmet needs in breast screening and diagnosis stimulate continuing research, method development and clinical assessment. Judging by many presentations at the RSNA meeting, the areas of greatest focus are augmentations of X-ray mammography achieved by other modalities and by computer-aided detection (CAD). As a prime example, ultrasound examination has long been applied as a secondary screen for suspicious X-ray findings.

MRI has recently shown considerable promise in augmenting classical mammography. G.M. Newstead, MD, and co-workers at the University of Chicago explored the effect of MRI on the clinical management of women with newly diagnosed breast cancer. They found that pretreatment MRI offers more accurate assessment of tumor size than mammography and ultrasound, particularly for those patients with certain forms of cancer. The International Breast Magnetic Resonance Consortium (IBMC) seeks a screening modality that can improve detection in high-risk women, including those with dense breast tissue. High-risk women need to be screened at a younger age when they are more likely to have dense breast tissue, a circumstance not favorable to X-ray mammography.

Constance Dobbins Lehman, MD, PhD, of the University of Washington Medical Center (Seattle) reported on the results of an IBMC trial in which researchers at 13 sites studied 367 high-risk women. MRI had a 1.1% diagnostic yield, while mammography exhibited a 0.3% diagnostic yield. However, Lehman noted a lack of evidence that MRI is an effective complement to mammography in average-risk women. Since 1999 the utilization of breast MRI exams has grown at about 40% per year in the U.S. It has been used for cancer staging, therapeutic monitoring and screening high-risk populations. It is now clear that MRI is more sensitive than X-ray mammography, but MRI is by no means a clear choice for the general population, especially since MRI of the breast is reimbursed at about $800 while X-ray mammography earns about $83. Another methodology was reported by M.E. Lobrano, MD, and co-workers at the East Jefferson General Hospital (Metairie, Louisiana). They sought to establish the utility of positron emission mammography (PEM) in distinguishing benign and malignant abnormalities. PEM showed promise in this regard, and it may also play a role in identifying unsuspected additional sites of disease.

CAD has been explored by numerous researchers, and is now being adopted in clinical practice, in attempts to improve the screening effectiveness of X-ray mammography. An estimated 25% to 30% of all screening mammography exams in the U.S. are now interpreted with the help of some form of CAD. Findings have been mixed to date, with some investigators convinced of its value and cost-effectiveness and others quite skeptical. Not to be ignored is the perception by many mammographers that CAD somehow threatens their professional image and ultimately their livelihood. Murray Rebner, MD, and his colleagues at William Beaumont Hospital (Royal Oak, Michigan) conducted a retrospective study to determine CAD's ability to detect pure ductal carcinoma in situ (DCIS), a pre-cancerous condition in which cells lining the milk ducts are cancerous. CAD helped find 91.4% of the lesions, performance which is similar to other published studies of CAD sensitivity for invasive cancers.

CAD also is a key component of a new breast imaging modality, full-field digital breast tomosynthesis, reported as works-in-progress by several suppliers at the 2004 meeting. A tomosynthesis scan involves multiple low-dose cone-beam X-ray projections through the breast at discrete angles spread over an arc. The projections intersect with each other at various points and allow the software to confirm areas of suspected masses from more than one angle, with the aim of improving cancer-detection efficiency. The acquired images can be reconstructed for 3-D viewing. The technology has the potential to eliminate overlapping tissue that can obscure lesions, and it may provide more diagnostic information. Groups at the University of Chicago and the University of Michigan (Ann Arbor, Michigan) reported lesion detection sensitivities of 80% and 86%, respectively, and fewer false positives occurred with 3-D views than with 2-D studies.

Molecular imaging

Molecular imaging has come a long way at a rapid pace and now rightly is characterized as a clinical and research discipline in its own right. It also is a topic that radiologists, most comfortable in their anatomical worlds, are just beginning to understand and appreciate. Michael Phelps, PhD, of the UCLA School of Medicine (Los Angeles), recognized as an inventor of the positron emission tomography (PET) scanner, was invited to give this year's "New Horizons" lecture, a presentation titled "Molecular Imaging: from Nanotechnology to Patients." Phelps' discussion ranged widely from systems biology to microfluidic chips to drug development to pre-clinical testing to Main Street clinical imaging. His fundamental message to his clinical colleagues was that the radiology community must begin to consider and understand the biological foundations of disease to complement its current understanding of structural aspects. Today's molecular imaging tools include PET, functional MRI (fMRI), optical methods and hybrid approaches such as PET/CT and PET/MRI.

Hybrid methods have become especially popular in human clinical applications. For example, PET/CT was not available in 2000, yet it now accounts for about 85% of the sales of PET imagers. U.S. PET procedure volume will be about 900,000 studies in 2004, up 35% from 2003. The unique value of these functional imaging techniques is now widely recognized. Phelps cited two important examples of PET's clinical applications: pre-symptomatic detection of Alzheimer's and Huntington's diseases, and positive influences on the treatment plans of as many as 50% of all cancer patients. He also strongly encouraged radiologists to learn about existing industrial applications in drug development, including small-animal imaging in preclinical studies and in real-time in vivo drug metabolism studies in human subjects.

PET and single photon emission computed tomography (SPECT) imaging both have gained additional traction in the diagnosis of Alzheimer's disease. A report by medical student Amitha Rao at the University of Kansas School of Medicine (Kansas City, Kansas) demonstrated that both modalities were substantially superior to the current diagnostic standard, the Mini-Mental Status Exam. On the other hand, Dr. Florian Vogt and co-workers at University Hospital of Essen (Essen, Germany) found that whole-body MRI and PET/CT imaging demonstrated disappointing performance in staging the cancers of their patients with malignant melanoma of the skin. Some patients initially staged as negative for lymph node and distant metastases eventually proved to have been positive at the time of initial examination, suggesting that follow-up examinations seem to be mandatory.

CT colonography

The assault on gastroenterologist incomes marshaled by the emergence of CT colonography (a.k.a. virtual colonoscopy) as a potential alternative to optical colonoscopy seemingly has abated with the report of several desultory findings at the opening CT colonography session moderated by Dr. C. Daniel Johnson of the Mayo Clinic (Rochester, Minnesota). One study of 614 patients at Duke University Medical Center (Durham, North Carolina) found that detection sensitivities for 1 cm polyps was 48% for air contrast barium enema, 59% for CT colonography and 98% for optical colonoscopy. For polyps greater than 6 mm the results were 41%, 55% and 99%, respectively. So perhaps CT colonography should be more appropriately compared to air contrast barium enema and not optical colonoscopy. However, while its sensitivity appears to be better, CT colonography reportedly incurs a greater risk of colon perforation (associated with air insufflation) than air contrast barium enema: an incidence of 0.066% vs. about 0.01%. Johnson noted, "This session shows that for CT colonography, we're just not there yet."