Value of detecting bone marrow involvement in Hodgkin lymphoma ‐ Response to Adams and Kwee

Abstract

We thank Adams and Kwee for the recent correspondence, and appreciate their review of our article (Pedersen et al, 2019). In the following we will reply to the criticisms they raised about our work. Adams and Kwee find it inappropriate to compare the prognosis of patients without bone marrow involvement (stage I–IV disease) to that of patients with bone marrow involvement (stage IV disease). In fact, to characterize this as the aim of our work is incorrect, because we did not ‘a priori’ regard the focal (unior multifocal) skeletal lesion as evidence of bone marrow infiltration, but rather as a possible indication of bone lesions. This is an uncommon presentation feature of classic Hodgkin lymphoma (cHL), which was first recognized after the introduction of fluorodeoxyglucose positron emission tomography/computed tomography (FDG PET/CT)-based diagnostic imaging in the pre-therapeutic staging of cHL. Hence, we wanted specifically to examine the different patterns of skeletal uptake observed in staging FDG PET/CT, and evaluate the presence/absence of high diffuse bone marrow uptake (BMU) compared to the presence/absence of unifocal and/or multifocal bone lesions on survival. Therefore, we found it appropriate to divide the patients into four groups (1. low diffuse BMU, 2. high diffuse BMU, 3. unifocal bone lesion and 4. multifocal bone lesions), which were then compared with regard to survival (3-year PFS was 80%, 87%, 69% and 51% respectively). The No lesion group was a result of the pooling of groups 1 (low diffuse BMU) and 2 (high diffuse BMU) due to similar PFS rates (fig 3 in Pedersen et al, 2019). We then compared the remaining groups with regard to survival. However, as correctly pointed out by Adams and Kwee, tumour stage is the most important prognostic factor in Hodgkin lymphoma. Therefore, when comparing a group of stage I–IV with a group of stage IV patients, the latter group predictably showed a poorer outcome. As correctly pointed out by Adams and Kwee, this is further supported by the fact that there was no difference in survival between stage IV patients with focal bone lesions and the remaining stage IV patients. Adams and Kwee conclude that determination of bone marrow status in cHL is a clinically negligible issue, because of the lack of prognostic impact of bone marrow involvement in patients with advanced-stage disease, and the low incidence of bone marrow involvement in patients with early-stage disease. We, however, find it of great interest that there was no significant difference in survival among patients with unifocal versus those with multifocal bone lesions. Skeletal tumour burden thus appears to be of minor importance, suggesting instead a possible more aggressive tumour biology (tumour microenvironmental factors?). Our results therefore emphasize that unifocal and multifocal bone lesions should be regarded as equally important predictors of adverse outcome, suggesting that the presence of even minor skeletal involvement is predictive of particularly aggressive disease. This notion is further underscored by the observation that focal bone lesions were associated with equally poor outcome regardless of whether ABVD (adriamycin, bleomycin, vincristine, dacarbazine)) or BEACOPP (bleomycin, etoposide, adriamycin, cyclophosphamide, vincristine, procarbazine, prednisone)-like chemotherapy was used as first-line treatment. Adams and Kwee would have appreciated information regarding how many of our patients with presumed extramedullary early-stage disease were upstaged to advanced-stage disease. We agree that this would have been valuable information. However, in both institutions participating in this study, FDG PET/CT is performed as part of the initial workup of newly diagnosed cHL patients whose stage is therefore not conclusively determined before all information (including FDG PET/CT) has been collected. Therefore, no information is available on how many patients were presumed to have early-stage disease before FDG PET/CT was performed. According to Adams and Kwee, our suggestion that diffusely increased BMU may reflect a non-malignant inflammatory bone marrow environment is incorrect. This is based on their previous analysis (Adams et al, 2016), where they found a significant inverse correlation between BMU and haemoglobin level. This study should, of course, have been cited in our article, particularly considering that some of its results seem to agree with ours. In fact, we also found a correlation between low haemoglobin level and increased BMU. On the basis of this observation (low haemoglobin), we find it reasonable to assume that the increased BMU is suggestive of a non-malignant bone marrow reaction. Adam and Kwee’s rejection of our statement that increased BMU may be indicative of bone marrow inflammation is also based on another parameter reported in their previous study (Adams et al., 2016), where they examined the correlation between BMU and bone marrow biopsy markers (cellularity, myeloid/erythroid ratio, fibrosis, reactive Tand B-lymphocytes). They found no correlation between correspondence