Does Surveillance Imaging After Treatment for Diffuse Large B-Cell Lymphoma Really Work?
Menée aux Etats-Unis, cette étude analyse le rapport coût-efficacité de trois stratégies de surveillance en routine de patients asymptomatiques atteints d'un lymphome diffus à grandes cellules B en rémission après une immunochimothérapie de première ligne
The late Dr Jane Weeks, a pioneer in cancer-focused outcomes research, was fond of asking the question “Don't you want to know if what we are doing really works?” In the report accompanying this editorial, Huntington et al1 present a rigorous cost-effectiveness analysis addressing such a question in hematologic oncology: Is post-therapy surveillance imaging worthwhile for diffuse large B-cell lymphoma (DLBCL)? Using a decision-analytic Markov model, they calculated quality-adjusted life-years (QALYs), lifetime costs, and incremental cost-effectiveness ratios (ICERs) for surveillance strategies versus routine clinical follow-up without imaging for asymptomatic patients with DLBCL in remission. Neither strategy they modeled—biannual computed tomography (CT) scans for 2 years or biannual positron emission tomography (PET)/CT scans for 2 years—was associated with a substantial increase in survival. The potential benefit of surveillance imaging was even less impressive after quality-of-life adjustments, and ICERs were > $160,000 per QALY (well beyond current willingness-to-pay thresholds).
Although the debate regarding imaging after treatment for aggressive lymphoma is not new,2–5 there have been no randomized controlled trials. Several observational studies have suggested that routine imaging can capture a subset of patients with asymptomatic relapsed disease,6–8 but none have definitively shown early detection to make a difference in terms of survival. For example, in an analysis of 108 patients with relapsed aggressive non-Hodgkin lymphoma, Liedtke et al9 found patients to be approximately 4 times more likely to have lower-risk disease when relapse was detected with routine imaging; however, despite an increase of 11% in 5-year median overall survival for relapses detected with routine imaging, the difference was not statistically significant (P = .13). In addition, it is possible that differences in survival observed for patients diagnosed through imaging alone may actually reflect lead-time bias, such that they include patients with less aggressive disease who would have lived the same amount of time even if the disease were detected and treated later. [...]
Journal of Clinical Oncology , éditorial, 2015