Routine surveillance scans are not indicated [for patients with large cell lymphomas]. They are expensive, potentially toxic, and anxiety-provoking. They also have false-positive results, increase cancer risk, and, importantly, do not impact patient outcomes.— Bruce D. Cheson, MD
Tweet this quote
There is no role for routine imaging as a means of following patients with large cell lymphoma, according to Bruce D. Cheson, MD, Deputy Chief of Hematology-Oncology and Professor of Medicine at Georgetown University Hospital, Lombardi Comprehensive Cancer Center, Washington, DC. “Routine surveillance scans are not indicated. They are expensive, potentially toxic, and anxiety-provoking. They also have false-positive results, increase cancer risk, and, importantly, do not impact patient outcomes,” he declared.
At the 2016 Pan Pacific Lymphoma Conference, Dr. Cheson made the case for eliminating surveillance positron-emission tomography (PET) or PET plus computed tomography (CT), except in unusual circumstances, and he addressed some clinical conundrums in using PET for staging.1
Staging Tools
For staging of lymphomas, PET-CT is the standard for FDG (18-fluoro-deoxyglucose)–avid lymphomas; CT is indicated for non–FDG-avid histologies and is useful in a limited number of circumstances. Contrast-enhanced CT is reserved for measuring nodal size, radiation planning, distinguishing bowel from nodes, and assessing compression/thrombosis of central/mediastinal vessels if required at staging, said Dr. Cheson. “However, in general, you don’t need contrast-enhanced CT,” he added. “Numerous studies show they are no better than regular CT scans in most situations.”
In practice, many patients have a separate contrast-enhanced CT before a PET-CT. If not, when a contrast-enhanced CT is required at staging, it should be combined with PET-CT at a single visit (with the PET scan done first). Since full-dose contrast-enhanced CT involves additional radiation, it should be used judiciously, Dr. Cheson reminded listeners.
PET Imaging in Lymphoma
- PET-CT has become the gold standard for staging and restaging FDG-avid lymphomas.
- Routine surveillance scans are not indicated for patients with large cell lymphomas.
A bone marrow biopsy still remains useful in staging some, but not all, subtypes. Bone marrow biopsies are not indicated for patients with Hodgkin lymphoma who undergo PET-CT. For patients with diffuse large B-cell lymphoma, bone marrow biopsy should be performed only if the PET scan is negative and identification of discordant histology is useful for patient management, according to Dr. Cheson.
“Patients don’t like to have biopsies, and there have been studies in large cell and Hodgkin lymphomas showing that PET scans are more sensitive and specific than bone marrow biopsies,” stated Dr. Cheson. “For Hodgkin lymphoma and most cases of large cell lymphoma, you can eliminate bone marrow biopsy if you do a PET scan. The ability for PET to pick up bone marrow involvement in follicular lymphoma, however, is not good enough.”
For most lymphomas, the need to perform PET scans generally stops with the end-of-treatment scan.
Posttreatment PET: One and Done
Although one would think PET surveillance might identify recurrences sooner, which would prompt the use of salvage therapy and therefore improve outcomes, this idea has not been supported by data. In fact, 80% of recurrences are detected by patients or their clinicians, according to Dr. Cheson. Scans are associated with false-positive results, and therefore unnecessary invasive procedures, and the practice is not cost-effective, he added.
Studies evaluating restaging in Hodgkin lymphoma have recorded positive predictive values in the range of 85% to 100%, according to Dr. Cheson. “This means if the scan shows disease, there’s more likely to be disease there,” he added. Negative predictive values have generally been 90% to 100%. “This means if the scan is negative, there is no disease there, and the patient is likely to be cured,” he noted.
In non-Hodgkin lymphoma, however, the positive predictive values range widely, from 43% to 93%, because of the presence of inflammation (as a result of rituximab [Rituxan] treatment). The negative predictive value, however, is still high, generally 85% to 100%, according to Dr. Cheson. “If you can predict that well from the posttreatment assessment, that’s one reason not to do a lot of scans after that,” he noted.
No Impact on Outcomes
The utility of posttherapy surveillance scans in diffuse large B-cell lymphoma was recently explored in a collaborative study between the Mayo Clinic and investigators from Lyon, France.2 For 680 patients treated with chemoimmunotherapy, 20% of patients ultimately relapsed, and 64% of relapses were identified before a scheduled visit. Surveillance imaging identified asymptomatic relapses in 4 of 112 patients (1.8%). Among those who relapsed, over 80% had signs or symptoms. “Whether patients relapsed before a planned surveillance visit, or the relapse was discovered during the visit, overall survival was identical,” he said.
In a European study of 1,221 patients with diffuse large B-cell lymphoma in first complete remission, in which one group underwent routine scans and the other was followed without scans, overall survival curves were identical.3 Approximately 90% of patients were alive at 8 years.3
Finally, in a study of 55 patients with transformed indolent non-Hodgkin lymphoma (67% underwent autologous stem cell transplant), 180 surveillance scans were performed; 153 were true negatives, 4 were false positives, 7 were indeterminate, 15 were true positives, and 1 was a false negative.4 Specificity was 94%, and sensitivity was 83%. The negative predictive value was 98%, but the positive predictive value was 63%. “All the diffuse large B-cell lymphoma relapses were symptomatic. Not a single asymptomatic relapse was picked up by a scan,” revealed Dr. Cheson.
Not Cost-Effective
Dr. Cheson indicated that routine surveillance scans are not cost-effective. In a recent decision-analysis simulation of three surveillance strategies for diffuse large B-cell lymphoma, imaging provided no benefit in terms of incremental effectiveness or quality-adjusted life years gained but did add expense.5 The cost for a routine follow-up (no scans) over 2 years was $38,280, vs $41,590 for biannual CT scans and $42,550 for biannual FDG-PET-CT.
Since the estimated overall survival for these three groups was similar, over a patient’s lifetime, the surveillance scans would cost “hundreds of thousands of dollars, for no benefit,” Dr. Cheson explained. Finally, routine scans provoke anxiety and pose a risk for unnecessary radiation, he added.
Current Recommendations and Future Tools
Dr. Cheson discouraged surveillance scans for patients achieving remission, especially in diffuse large B-cell lymphoma and Hodgkin lymphoma, although a repeat study may be considered following an equivocal finding posttreatment, he said. “Once you get that scan 6 to 8 weeks posttreatment, that’s enough, unless you need to evaluate something equivocal.” For other situations, “judicious use” of follow-up scans may be considered (eg, in patients with indolent non-Hodgkin lymphoma who have residual intra-abdominal or retroperitoneal disease).
Future technologies may obviate the need for PET scans, such as circulating tumor (ct) DNA (ie, “liquid biopsy”). Detection of ctDNA on day 1 of cycle 3 in a cohort of patients with diffuse large B-cell lymphoma predicted relapse in a study by Roschewski et al.6 Disease progression occurred in only about 20% of ctDNA-negative patients, compared with more than 50% of the ctDNA-positive group (P < .0001). In addition, a study from the National Cancer Institute evaluating the clonoSEQ ctDNA assay showed that progression-free survival was approximately 80% for patients with low levels of ctDNA but less than 50% for those with high levels of ctDNA (P = .027).7
“There are potentially superior methods of surveillance, such as liquid biopsy, that may provide a suitable alternative to PET in the future,” Dr. Cheson concluded. ■
Disclosure: Dr. Cheson reported no potential conflicts of interest.
References
