Background 18F-FDG-PET/CT is an established method for staging patients with newly diagnosed Hodgkin Lymphoma (HL). It has a high ability to detect skeletal and bone marrow involvement. It has been discussed whether diffusely increased bone marrow uptake (BMU) of 18F-FDG indicates bone marrow involvement or not. In addition, the importance of unifocal versus multifocal bone lesions on survival is under debate. Thus, the aim of our study was to examine the prognostic value of BMU and compare it with that of uni- and multifocal lesions.
Materials and methods 217 patients from Aarhus and Uppsala university hospitals (mean age 42, range 8-83) with newly diagnosed HL were included. BMU was calculated as SUVmax in the vertebral column (L3/L4) divided by liver SUVmax. In case of focal 18F-FDG-uptake in bone, patients were categorized as having uni- or multifocal lesions. For survival analysis, patients were divided into four groups: Patients with BMU below the median and no focal lesions (lowBMU), patients with BMU higher than the median and no focal lesions (highBMU), patients with a single bone lesion (unifocal), and patients with multifocal bone lesions (multifocal).
Results Median BMU was 1.15 (range 0.52–5.56). 42/217 (19.4%) patients had either unifocal (21/217) or multifocal lesions (21/217). With a median follow-up time of 41 months, 3-year PFS was 87% (lowBMU), 88% (highBMU), 67% (unifocal), and 58% (multifocal). The presence of focal bone lesions (uni- or multifocal) was associated with a significantly inferior PFS (log-rank p=0.00046), but no significant difference in PFS between unifocal and multifocal lesions was observed (log-rank p=0.35). Multivariate analysis (Cox regression) on a subset of patients (n=183), showed that presence of bone lesions, age, and leukocyte count were independent predictors of PFS whereas gender, haemoglobin, albumin, and sedimentation rate were not.
Conclusion Diffusely increased BMU at initial staging was not associated with a poor prognosis. The presence of focal bone lesions was associated with significantly inferior prognosis, regardless of whether they were uni- or multifocal, and should therefore be considered equally important risk factors. Whether this bone lesion-associated clinical picture of HL also correlates to specific biological features, remains to be clarified.