Prognostic Implications of Imaging-Based Bone Marrow Assessment in Lymphoma: ¹⁸F-FDG PET, MR Imaging, or ¹⁸F-FDG PET/MR Imaging?

TO THE EDITOR: We read with interest the recent article by Berthet et al. (1), who investigated the prognostic implications of ¹⁸F-FDG PET–based bone marrow assessment in diffuse large B-cell lymphoma (DLBCL). Their well-designed retrospective study included 133 patients with newly diagnosed DLBCL, of whom 32 were positive for bone marrow involvement according to ¹⁸F-FDG PET whereas only 8 were positive according to bone marrow biopsy. In a multivariate analysis, Berthet et al. showed that only the International Prognostic Index (IPI) and the ¹⁸F-FDG PET bone marrow status were independent predictors of progression-free survival (P = 0.005 and P = 0.02, respectively), whereas only the IPI remained an independent predictor of overall survival (P = 0.004). Almost simultaneously, another study on the same subject was published by Khan et al. (2). In their retrospective study that included 130 patients with newly diagnosed DLBCL, 35 were judged to have marrow involvement; of these, 33 were identified by ¹⁸F-FDG PET and 14 by bone marrow biopsy. Cases with bone marrow deposits identified by ¹⁸F-FDG PET but not by biopsy had progression-free and overall survival similar to Ann Arbor stage IV disease without involved bone marrow (2). Both studies suggest that ¹⁸F-FDG PET–based bone marrow assessment in newly diagnosed DLBCL may have prognostic implications and that the importance of ¹⁸F-FDG PET bone marrow status may overshadow that of the bone marrow biopsy result in this context (1,2).

Although ¹⁸F-FDG PET is a powerful method for evaluation of the bone marrow, it is a pity that neither Berthet et al. (1) nor Khan et al. (2) make any mention of the role of MR imaging in this setting. Back in 1997, Tsunoda et al. (3) had already reported on the prognostic value of bone marrow MR imaging in lymphoma. In their study, Tsunoda et al. retrospectively investigated a mixed population consisting of 56 patients with newly diagnosed low-, intermediate-, and high-grade non-Hodgkin lymphoma (n = 48) and Hodgkin lymphoma (n = 8). At the time of diagnosis, all patients underwent masked bone marrow biopsy of the posterior iliac crest and MR imaging of the femoral bone marrow at 1.5 T. The findings of the biopsy were negative in 39 patients, of whom 12 had positive results on MR imaging. Patients were followed for 1–58 mo after the MR imaging examination, with a median of 17 mo. Interestingly, patients with a positive MR imaging result but a negative biopsy result had a significantly shorter overall survival than did those for whom both MR imaging and biopsy were negative (P = 0.016). Tsunoda et al. concluded that abnormal MR imaging findings for the femoral bone marrow are associated with a significantly poorer survival in patients with lymphoma, regardless of histologic findings in the bone marrow. Since 1997, MR imaging has made a giant leap forward; nowadays, a high-quality MR imaging examination of the bone marrow in the entire body (i.e., from cranial vertex to toes) can be routinely obtained in less than half an hour. Recent data have shown that the sensitivity of whole-body MR imaging for the detection of lymphomatous bone marrow involvement equals that of ¹⁸F-FDG PET (4). Even more interestingly, preliminary data from our ongoing prospective study on the value of whole-body MR imaging in DLBCL patients with a negative masked bone marrow biopsy show that disease relapse or progression and death occur more frequently if whole-body bone marrow MR imaging findings are positive. Thus, although more prospective research is warranted and a comparison with established prognostic stratification models such as the IPI should be done, both older and more recent data indicate that bone marrow MR imaging findings may have prognostic implications in lymphoma, independently of (masked) bone marrow biopsy results.

In conclusion, both ¹⁸F-FDG PET and MR imaging play a major clinical role in the evaluation of bone marrow diseases, including lymphomatous bone marrow involvement. Given this background information, one may wonder which of the two should be used as a noninvasive bone marrow biomarker of prognosis in lymphoma. ¹⁸F-FDG PET/MR imaging will both answer this question and relieve us from the difficult decision of choosing between them.

• © 2013 by the Society of Nuclear Medicine and Molecular Imaging, Inc.