The role and limitations of imaging technologies are constantly evolving, presenting a challenge for all medical practitioners but especially for specialists who work closely with radiologists in caring for their patients. Among the plethora of technologic sessions offered at the 89th Scientific Assembly and Annual Meeting of the Radiological Society of North America, held in Chicago in December 2003, was a minicourse addressing the use of positron emission tomography (PET) in one field of specialty, oncology.

The session, chaired by Barry A. Siegel, MD, Mallinckrodt Institute of Radiology, St Louis, and moderated by R. Edward Coleman, MD, Duke University Medical Center, Durham, NC, included information on (1) how to explain the limitations of PET with regard to variable uptake and retention of fluorodeoxyglucose (FDG) and when PET scans can be helpful for clinical care; (2) the basics of patient preparation and interpretation of images; and (3) the role of PET in the diagnosis and staging of tumors.

What Oncologists Need to Know

The first speaker, Anthony F. Shields, MD, PhD, Karmanos Cancer Institute, Detroit, Mich, focused on the fact that PET provides images of the physiology that complement the anatomy and thus can play a role in the assessment of cancers of the lung, head and neck, colon, esophagus, lymph glands, skin, breast, and thyroid. The only tracer routinely used in oncology cases is FDG, which normally exhibits retention in certain organs such as the brain, heart, bowel, and kidney. Oncologists should be trained about the limitations of PET in this regard so they will be able to understand the images obtained.

In diagnosis, the role of PET is to reveal solitary pulmonary nodules (for which the specificity and sensitivity have been shown to be 93% and 86%, respectively). However, the FDG uptake varies with different tumors (and is highest with the most common ones, involving the lungs, for instance) and can miss small lesions (<1 cm). Thus, a negative scan does not prove that a patient is free of cancer.

If oncologists are made aware of these limitations, they can plan treatment accordingly. Even though PET may not be a perfect diagnostic tool, it can direct a biopsy and can detect additional lesions in the staging of cancer. Furthermore, its use can help prevent the performance of unnecessary thoracotomies and surgeries.

Oncologists who want to assess a patient's response to treatment may be eager to obtain a PET scan, but radiologists must explain that the time for a response to become evident after treatment will vary. "This greatly depends on  the tumor type being imaged and the treatment employed," Shields explained in a later interview. "For example, in a patient with gastrointestinal stromal tumors treated with imatinib one can often see dramatic evidence of response by PET in 24 hours. On the other hand, in lung cancer treated with radiation and chemotherapy, one often has to wait 12 weeks for the treatment response to become manifest."

Assessment of response to therapy has become more and more important in order to save on the expense of unnecessary chemotherapy or radiation, and the use of imaging in particular as an assessment tool becomes more important as the therapeutic choices for patients increase and improve.

Two final points made by Shields were that radiologists should be careful to clarify the language they use on reports, because some patients will read them, and that radiologists and clinicians should review PET images together so that the case history is considered and artifacts can be recognized.

PET Basics for Oncology Studies

Next to address the audience was Coleman, who began by explaining patient-preparation goals. When FDG is administered, the patient must be at basal glucose and insulin levels, because the FDG competes with serum glucose for accumulation. Increased concentrations of glucose and FDG occur within tumors.

Further patient preparation includes no caloric intake for 4 hours prior to the scan and no diabetic or glucose-control medication within 4 hours. No limit on prior activity is necessary, although the patient should refrain from extremely strenuous exercise such as power-lifting or marathon running. The standard interval between administration of FDG and imaging is 45 minutes to 1 hour, but accuracy increases with the length of the interval (up to 2 hours) because it allows for more FDG uptake.

With regard to reading the PET scans, it is beneficial to look first at the maximum-intensity projection (MIP) image; this provides an excellent overview of the imaging results. It is also helpful to review the abnormalities seen on computed tomography in relation to the PET scan.

There are certain pitfalls and artifacts of which radiologists and oncologists must be aware as they interpret a patient's images. For example, malignancy may simulate a benign process and vice versa; in addition, hot spots may occur at the injection site and obscure the local anatomy, and uneven distribution of FDG may occur. Patient motion or breathing also can cause technical artifacts, so careful attention must be given to the comfort and positioning of the patient.

Variability of FDG uptake can be caused by postsurgical inflammation or infection, post-therapy changes, or chronic conditions (involving the thyroid, for instance). Because of the resulting artifacts, one of the primary problems associated with PET diagnostically is that malignancy cannot be totally excluded.

Cancer Diagnosis and Staging

The final speaker in this session, Farrokh Dehdashti, MD (St Louis), focused on the benefits of PET for the diagnosis and staging of cancers. PET technology is advantageous for detection of occult lesions when the primary lesion is unknownfor example, as a complement to mammography. It also can be used to differentiate benign from malignant lesions (as in the breast, for instance) and thus can eliminate unnecessary surgery or biopsy. Dehdashti explained that if a PET study is positive, then there likely is a malignancy. However, if the PET study is negative, follow-up CT should be performed.

PET may be superior to CT or MRI postsurgically in the detection of recurrent disease. PET findings may also redirect investigative or treatment efforts; for example, they may help pinpoint the most efficacious radiation port. PET has also been found to be more specific and sensitive than CT or MRI in the evaluation of patients with suspected recurrent or metastatic colon disease; overall survival has been shown to be better than with conventional imaging because of improved selection of patients for surgery with use of PET. In cases of lymphoma, FDG-PET in comparison with gallium scanning has been shown to have 20% higher sensitivity per patient and to offer better staging of lesions, and PET has been shown to be much more accurate than conventional imaging for the detection of stage IV disease in esophageal cancer.

In summary, Dehdashti enumerated five aspects in which PET has greater efficacy than conventional imaging in oncology: diagnosis of selected cancers; initial staging; detection of recurrence; treatment selection; and cost-effectiveness.

Seleen Street Collins is a contributing writer for Decisions in Imaging Economics.