Physicians at the Weill Medical College of Cornell and the New York Presbyterian Hospital in New York reported on 23 patients who received PET scans before and after one cycle of treatment and also at the completion of chemotherapy. PET was accurate 87% of the time after one cycle but only 70% of the time after completion of chemotherapy. Sensitivity, i.e., the ability to detect the FDG and hence disease, was also significantly higher after one cycle, at 82% v. 45.5%. In particular, in cases in which results from the first cycle differed from the results following completion of therapy, the results from the first cycle scan were more accurate. Ninety percent of patients with positive 18F-FDG PET results after one cycle experienced disease relapse, while 85% who had negative 18F-FDG PET findings after one cycle remained in remission. The authors concluded that PET was a better predictor of outcome and response to therapy after just one cycle of chemotherapy than after completion.
PET stands for positron emission tomography. Because of its unique ability to measure metabolic activity-or the efficiency of the cells converting food to energy--PET provides accurate, noninvasive detection and staging of many cancers. A radiopharmaceutical, such as 18F-FDG (fluorodeoxyglucose), which includes a radionuclide (a radioactive element) is injected into the patient and gives off signals that are measured by a PET scanner. Because cancer cells are more metabolically active, they show up on the PET images more intensely than normal tissue.
Using 18F-FDG PET to predict outcomes and response to therapy early in treatment is one of the most important new uses of this technology. It is particularly helpful in lymphoma, where tumors can persist after successful treatment. Conventional imaging with CT or MRI has difficulty differentiating between residual masses harboring tumor and those that are scar tissue. Determining at what point during chemotherapy a PET study should be ordered is a major issue in the nuclear medicine community. Debate continues regarding the timing of and access to this procedure.
The study also looked specifically at the results for patients whose diagnosis or original staging placed them in the category of poor-prognosis, and found similar results for this group. According to the study authors, 18F-FDG PET may be especially useful for these patients, since early evidence of persistent disease may mandate innovative intervention such as bone marrow transplantation.
The study reported a statistically significant number of false negatives in patients who had completed treatment, which contributed to its lower value as a predictor at the end of treatment. Six patients who had a false-negative FDG uptake at the completion of treatment ultimately relapsed. The authors noted that this could be the result of small amounts of disease that remained and escaped detection because of the resolution limitations of the coincidence PET scanner used in the study.
In a commentary accompanying the article in JNM, Val Lowe, MD and Gregory Wiseman, MD, of the Department of Radiology at the Mayo Clinic took issue with parts of the study's findings, and elements of the study makeup. While commending the authors' objectives, Lowe and Wisemen noted that the use of a coincidence rather than dedicated full-ring PET camera, which has less sensitivity, could have affected the results, particularly as they related to end-of-cycle findings of disease. They pointed to studies that showed dedicated PET at the completion of therapy had a 91% sensitivity in predicting disease relapse (v. the 45.5% in the Kostakoglu study). Nevertheless, the study authors emphasize the fact that failure to achieve a response can be predicted just after one cycle of therapy at which time second-line therapy or stem cell transplantation can be considered earlier prior to completing the full course of chemotherapy, without patients experiencing unnecessary side effects.
Lowe and Wiseman also expressed concerns about the size and makeup of the study group, noting that there were only 30 patients enrolled (the study report focuses on 23). The fact that it was a small study looking at different diseases, being treated differently, and at different points in their treatments make it difficult to make meaningful comparisons, they felt. According to the American Cancer Society, there will be an estimated 60,900 new cases of lymphoma in 2002, including 7,000 cases of Hodgkin's disease and 53,900 cases of non-Hodgkin's lymphoma. An estimated 25,800 people will die from the disease (24,400 NHL, 1400 HD)
"PET Predicts Prognosis After 1 Cycle of Chemotherapy in Aggressive Lymphoma and Hodgkin's Disease" was written by Lale Kostakoglu, MD, Division of Nuclear Medicine, Department of Radiology; Morton Coleman, MD, and John P. Leonard, MD, Center for Lymphoma and Myeloma, Division of Hematology and Oncology, Department of Medicine; Ichiei Kuji, MD, Holly Zoe, and Stanley J. Goldsmith, MD, Division of Nuclear Medicine, Department of Radiology, Weill Medical College of Cornell University and New York Presbyterian Hospital, New York, New York.
Copies of the article and images related to the study and the commentary are available to media upon request to Karen Lubieniecki at Karenlub@aol.com, (703) 683-0357. Copies of this and past issues of The Journal of Nuclear Medicine are available online at jnm.snmjournals.org. Print copies can be obtained at $15 per copy by contacting the SNM Service Center, Society of Nuclear Medicine, 1850 Samuel Morse Drive, Reston, VA 20190-5315; phone: (703) 326-1186; fax: (703) 708-9015; e-mail: firstname.lastname@example.org. A yearly subscription to the journal is $170. A journal subscription is a member benefit of the Society of Nuclear Medicine.