The Case Against Heavy PETing

Abstract

“Do you know the difference between education and experience? Education is what you get when you read the fine print; experience is what you get when you don’t.” — Pete Seeger Rarely have such major changes in practice been based on so much published data with so little in the way of guidance as to how to actually apply those data, as with the use of positron emission tomography (PET). Hundreds of publications suggest that the availability of 18-fluorodeoxyglucose (FDG) PET scans has revolutionized the management and improved the outcome of patients with lymphomas and other cancers. FDG-PET has also been proposed as a replacement for current methods of lymphoma staging and response assessment and as a guide to modifying management during and after therapy. A positive scan after one or more cycles of therapy seems to be the strongest predictor of poor patient outcome. However, we need to examine the fine print before adopting the widespread use of PET. In lymphoma, computed tomography–PET (CT-PET) scans clearly identify more lesions than other imaging studies; however, stage is changed in only 10% to 30% of patients, and treatment is altered in even fewer patients, with no data to suggest that prognosis is improved based on this information. PET scans are imperfect with many false-negative and falsepositive results. The reasons for false-negative PET scans include technical problems, uncontrolled diabetes, and variability of FDG avidity among lymphoma subtypes. The lengthy list of causes of false positives, which may occur in one quarter of patients, includes inflammation, infection, tumor necrosis, and scanning too soon after chemotherapy or radiotherapy. Fortunately, the problem of standardization has been resolved, for now, by the International Harmonization Project in Lymphoma; its recommendations for the conduct of scans and the definition of a positive scan were incorporated into the revised International Response Criteria. Two articles in this issue of Journal of Clinical Oncology support the judicious use of FDG-PET in lymphoma. Zinzani et al performed serial FDG-PET scans on patients with diffuse large B-cell lymphoma, Hodgkin’s lymphoma, or follicular lymphoma in first complete remission at 6, 12, 18, and 24 months and annually thereafter. Scans were interpreted as positive, negative, or inconclusive. The patients with inconclusive scans underwent biopsy, particularly when there was a sufficient index of suspicion, such as if a scan remained positive 1 to 2 months later or if there were other signs suggesting active disease with additional follow-up; the likelihood of a positive biopsy also correlated with the pretreatment risk group. Their conclusion that FDG-PET was useful for follow-up is the large print version. However, the likelihood of relapse became insignificant after 12 months for Hodgkin’s lymphoma and after 18 months for diffuse large B-cell lymphoma, although there was the expected continuous risk for relapse with a follicular histology. Importantly, patient risk before treatment correlated with systemic relapse. Only one third of patients with an inconclusive PET had a suspicious CT scan with a biopsy-proven 33% false-positive rate. For patients with a positive PET and a negative CT, the false-positive rate was 42%. FDG-PET identified unsuspected early relapse in only 10% of patients with Hodgkin’s lymphoma, which was assisted by clinical risk and CT findings. What cannot be determined from these data are the cost effectiveness of the PET scan, compared with repeating CT scans a few months later, and whether identifying unsuspected relapses a few months earlier made a difference in clinical outcome. Indeed, it is the patient or the physician who suggests the possibility of relapse on clinical grounds 80% of the time, before it is detected by routine imaging studies. Certainly, routine scanning in the indolent histologies is not justified. FDG-PET scans are also being overused for interim assessment during treatment; however, the limited available data do not support that changing therapy on the basis of such information impacts favorably on patient outcome. The second article in this issue by Terasawa et al reports a systematic review of 13 studies that included 360 patients with advanced-stage Hodgkin’s lymphoma and 311 patients with diffuse large B-cell lymphoma. The generalizability of the data is questionable because only 10% of Hodgkin’s patients were in an unfavorable risk group, and patients with diffuse large B-cell lymphoma were heterogeneous with respect to risk of treatment failure and therapeutic interventions; in addition, there was variability in the number of chemotherapy cycles after which the PET was performed, when during the cycle the PET scan was performed, and equipment; some studies had methodologic problems; and a relatively small number of case series often from the same investigators were included. The authors appropriately conclude that the evidence does not support interim scanning outside of a clinical trial. Although patients often request interim scans, a negative scan does not guarantee a positive clinical outcome, and a positive scan does not ensure ultimate doom but clearly engenders anxiety in the patient and caregivers. So how does the clinician navigate the numerous articles published on FDG-PET and apply the information to daily practice? First, limit generalizations; not all lymphomas are FDG avid. Just because the test is more sensitive than other available technologies, such as CT or magnetic resonance imaging scans, does not mean that it provides information that should be acted on, and acting on such information does not necessarily translate into improved patient outcome. Most importantly, it is mandatory to incorporate clinical information in scan interpretation and decision making; a PET scan does not replace a biopsy. JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 27 NUMBER 11 APRIL 1