Critical Views on the Prognostic Potential and Interpretation of Bone Marrow ¹⁸F-FDG PET in Diffuse Large B-Cell Lymphoma

TO THE EDITOR: We read with interest the recent article by Cerci et al. (1), who performed a prospective multinational cohort study in 327 patients with newly diagnosed diffuse large B-cell lymphoma (DLBCL) to determine the relative prognostic implications of blind bone marrow biopsy (BMB) and ¹⁸F-FDG PET–based bone marrow status in this disease. We would like to share our views on the roles of BMB and ¹⁸F-FDG PET in the prognostication of DLBCL and the interpretation of nonfocal diffusely increased bone marrow ¹⁸F-FDG uptake in these patients.

Cerci et al. reported that neither BMB nor ¹⁸F-FDG PET alone is a reliable indicator of poor-risk bone marrow disease. Only bone marrow involvement identified by both ¹⁸F-FDG PET and histology was found to indicate a poor prognosis. The authors speculated that blind BMB is more likely to be tumor-positive in cases of more extensive marrow disease. That, rather than bone marrow involvement per se, was thought to be associated with a worse patient outcome. First, we do not agree with the authors’ statement that BMB alone is not a reliable indicator of poor prognosis. Numerous large-scale studies have demonstrated that BMB-based bone marrow status is an independent predictor of outcome (2–4). Second, we believe bone marrow ¹⁸F-FDG PET has not yet convincingly been shown to have prognostic value in DLBCL. Cerci et al. claimed the combination of both positive bone marrow ¹⁸F-FDG PET and BMB findings to be of predictive value, but its incremental value over BMB-based bone marrow status alone was not assessed. In addition, the prognostic value of ¹⁸F-FDG PET–based bone marrow status alone was not reported at all. Therefore, we believe Cerci et al. provided insufficient data to support their conclusion that bone marrow staging by ¹⁸F-FDG PET is important for defining prognosis in DLBCL. On the contrary, our own recently published data indicate that, unlike BMB, ¹⁸F-FDG PET–based bone marrow status has no value at all in predicting either progression-free survival or overall survival in newly diagnosed DLBCL (5). In our study, we also measured metabolic tumor volume and total lesion glycolysis of all ¹⁸F-FDG–avid bone marrow lesions in each patient, but this did not have any prognostic value either (5). This finding contradicts the speculation of Cerci et al. that the amount of tumor burden in the bone marrow may be predictive of survival, at least when ¹⁸F-FDG PET is used for this purpose. Furthermore, risk assessment by BMB may reach beyond that by dichotomizing into groups with and without marrow involvement (6–8). In the study by Cerci et al. the group of patients with positive BMB and negative bone marrow ¹⁸F-FDG PET findings had a favorable outcome but consisted of 6 patients with large-cell low-volume tumor involvement, 2 with small-cell involvement, and only 2 with large-cell high-volume involvement, the last of which is known to have a major adverse impact on patient outcome (6–8). Unfortunately, Cerci et al. did not report how frequently large-cell high-volume involvement was observed in the group of patients who were positive for bone marrow involvement both on ¹⁸F-FDG PET and BMB. However, it is not unlikely that the prevalence of large-cell high-volume bone marrow involvement was higher in this group, enabling an even more detailed risk assessment solely by BMB results and thus reducing the additional benefit of bone marrow ¹⁸F-FDG PET.

Cerci et al. also reported that of their 18 patients who appeared to have nonfocal diffusely increased bone marrow ¹⁸F-FDG uptake, only 4 had histologic evidence of marrow disease on iliac crest biopsy. Although this finding may suggest that this particular bone marrow appearance on ¹⁸F-FDG PET is not associated with or due to lymphomatous bone marrow involvement in most cases, the finding should be interpreted cautiously. First, Cerci et al. did not indicate whether patients who were treated with hematopoietic growth factors (a well-known cause of diffusely increased bone marrow ¹⁸F-FDG uptake (9)) were excluded. Second, the criterion that was used to define the presence of diffusely increased bone marrow ¹⁸F-FDG uptake was not reported. This, in combination with ¹⁸F-FDG interpretation by different observers, may have introduced considerable reader variability that may have affected their results (9). Our own unpublished results indicate that BMB is positive in most cases of treatment-naive non-Hodgkin lymphoma, including DLBCL, when nonfocal diffuse bone marrow ¹⁸F-FDG uptake exceeds liver ¹⁸F-FDG uptake.

In conclusion, it is our opinion that, unlike BMB, the prognostic potential of bone marrow ¹⁸F-FDG PET in DLBCL has not been proven yet. Furthermore, we believe that the phenomenon of nonfocal diffusely increased bone marrow ¹⁸F-FDG uptake in DLBCL should be interpreted within the appropriate clinical context (particularly considering recently administered therapies) and using standardized criteria, to predict the most likely histologic correlate.

• © 2015 by the Society of Nuclear Medicine and Molecular Imaging, Inc.