PET/CT for Monitoring Therapy Response in Breast Cancer Patients with Bone Metastases

Editor: We read with great interest the paper by Dr Tateishi and colleagues (1) that appeared in the April 2008 issue of Radiology. This is one of a few studies (2) that monitor therapy response in metastatic bone lesions and correlate the morphologic and metabolic patterns with positron emission tomographic (PET)/computed tomographic (CT) imaging. The authors reported that, after treatment, increased attenuation of metastatic bone lesions and decreased standardized uptake value (SUV) were potential predictors of incidence of progression and that decreased SUV was a predictor of response duration in patients with metastatic breast cancer. Although the reported findings are potentially of great clinical impact, we think that some methodological and clinical issues require further clarification. First, the authors reported that patients with decreased fluorine 18 fluorodeoxyglucose (FDG) uptake in the target lesion compared with uptake seen on the baseline images were considered to be responders. It has been noted in similar studies (3,4) that metastatic breast cancer patients with predominantly sclerotic disease showed fewer lesions at FDG PET imaging than at bone scintigraphy, as well as lower FDG uptake (in terms of SUV) than lytic metastases. Also, in prostate cancer, similar findings have been seen on FDG (4,5) and 18F fluorocholine (FCH) (2,5,6) PET/CT images. In a recent study, our group showed that even in progressive disease no FCH uptake was detected in metastatic sclerotic lesions with an attenuation greater than 825 HU (6). However, it is not clear if the decreased tracer uptake in the sclerotic lesions is due to therapy response or lower sensitivity of PET imaging (3,4). In our opinion, reduction in SUV may reflect the response of therapy, but the cause remains somewhat unclear, which warrants further histopathologic evaluation. Second, metastatic lesions in the skeleton without corresponding morphologic changes on CT scans might reflect “early marrow-based” metastases (7). This is one of the important issues that is not clearly mentioned and followed by the authors. Finally, in Dr Tateishi and colleagues’ study, the lesion that exhibited the most substantial uptake was selected as the only target lesion. However, flat or long bones may present varying patterns of bone metastases and different therapy responses (5,8), which is not pointed out. In conclusion, we think that the value of PET imaging in the assessment of metastatic sclerotic lesions needs further histopathologic investigation. Future studies should clarify whether such lesions are clinically relevant when compared with metabolically active bone metastases. If proved, this would emphasize once more the advantage of metabolic (ie, PET) over morphologic (ie, CT) imaging for therapy monitoring.