Reply to H.J.A. Adams and T.C. Kwee.

Abstract

We thank Adams and Kwee for their detailed review of our article and also for their comments. The main objective of this trial was to prospectively determine the prognostic value of interim positron emission tomography (PET) after two cycles of rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone every 2 weeks, and we reported that this procedure has only limited value. Additionally, we demonstrated that the best separation of prognosis between PET-positive and PET-negative patients was achieved by the end-of-treatment PET scan. We agree that an event-free survival of 72% at 2 years for PETnegative patients does not indicate a high prognostic value. However, as discussed in our article, the outcome for all patients (PET-positive and PET-negative) seemed disappointing but can be explained by the fact that involved field radiotherapy or high-dose methotrexate as CNS prophylaxis were given to 25 patients regardless of PET results, and this counted as an event per the protocol. Because of a balanced distribution in both groups and the fact that our findings showed no differences when excluding these patients, we decided to report on the end points that were defined in the protocol and were studied prospectively. In further analysis, we also looked at progression-free survival (PFS), where such events were excluded. In this situation, PFS at 2 years exceeded 88% for the patients having a negative PET scan at the end of treatment. At least for these patients, PET negativity at the end of this trial predicted a low risk of disease progression within 2 years. The role of PET imaging in lymphoma patients for staging and response assessment has been studied extensively in the past, and, clearly, this technology has its limitations, such as the detection of microscopic residual disease or bone marrow infiltration. Of course, patients with indolent lymphomas or aggressive lymphomas treated with noncurative chemotherapy can achieve a negative PETresult at the end of treatment. However, inDLBCL, cure ismost often the aim of chemotherapy in this setting, and the PET has been evaluated as a surrogatemarker for this goal. A PFS of 88% after 2 years for patients having a negative PETresult at the end of treatment is in line with the results of other trials in this situation, and to my knowledge, there are not many other predictive parameters that are superior to these findings. In our view,more concerning than the PET-negative results is the relatively high number of false-positive PET results for the interim PET in this trial, as well as for the end-of-treatment PET.