John Byrd, MD
Dr. Byrd indicated no relevant conflicts of interest.
Eichhorst BF, Fischer K, Fink AM, et al. Limited clinical relevance of imaging techniques in the follow-up of patients with advanced chronic lymphocytic leukemia: results of a meta-analysis. Blood. 2010. [Epub ahead of print]
Chronic lymphocytic leukemia (CLL) is a disease characterized by blood lymphocytosis, cytopenias, bone marrow infiltration, lymph node enlargement, and hepatosplenomegaly. Prior to the availability of modern imaging studies, the Rai or Binet Staging systems employed a routine blood test and physical exam to predict patient outcomes at diagnosis. When treatment was initiated, response criteria included documentation of partial improvement in cytopenias (if present), resolution of blood, lymph node, and spleen/liver disease for a partial response, and complete resolution of these events with bone marrow clearance for a complete response. Computed tomography (CT) studies of internal lymph nodes, liver, and spleen were not included, as this technology was not yet developed. Over time, CT scans were rapidly adopted for lymphoma staging, and, in the absence of supporting data, the use of CT scans was empirically adopted for staging and therapy response of CLL as well. This empiricism in using expensive CT scans has evolved into a requirement by regulatory agencies for registration of new therapeutic agents for CLL and has become part of response guidelines for clinical trials. In a time of concerns about the costs of health care and unnecessary exposure to radiation by such imaging studies, the question of where the evidence for this mandated imaging approach is derived should be asked.
One small study of our own institution’s retrospective experience with flavopiridol-treated patients demonstrated no difference in using clinical exam and labs versus adding a CT scan assessment when examining progression-free (PFS) and overall survival. In a recent, much larger retrospective study, Dr. Eichorst and colleagues from the German CLL group examined the impact of CT scans in 1,372 patients receiving first-line therapy in clinical trials and demonstrated that these imaging studies added no benefit when following patients for evidence of relapse after treatment. For those patients receiving modern chemoimmunotherapy, no benefit was observed in using CT scans post-therapy to better predict PFS. Finally, the presence of bulky lymph nodes (> 5 cm) was not shown to be predictive of patient outcomes. It was only with chemotherapy-based treatments that CT scans improved prediction of PFS over that of routine clinical response assessment. Collectively, these authors recommend that CT scans not be used for CLL staging or follow-up outside of clinical trials, as supported by the current IWCLL guidelines.
This article provides even more important evidence that CT scans for CLL are not required outside of select circumstances. These include evaluation of symptoms referable to areas not adequately visualized by exam, the rare setting of follow-up post-therapy when the only site of disease is in the abdomen or chest and not measurable by exam, and in those patients participating in clinical trials. It is unclear why the authors did not take their recommendations further to question why these CT scans are needed at all as part of clinical trials. The current requirement for CT scans remains problematic to interpreting new study results, as essentially all prior CLL clinical trials did not include CT scans. More importantly, these imaging tests add significant cost and potential morbidity to the very special patients who volunteer to be part of clinical trials exploring new treatment approaches. Reconsideration of the CT requirement in the setting of implementing detailed lymph node and spleen physical exams might offer an opportunity to match our new clinical trial approaches to methods best supported by evidence-based tumor assessment.