Imaging in Oncology: Evaluation of Bone Metastases in Lung Cancer

Poonam Malhotra, MD, and Claudia G. Berman, MD


August 15, 2002


Clinical PET scanning using 18F-fluoro-2-deoxy-D-glucose (FDG) is based on the observation that malignant tumors have an accelerated glycolytic rate. FDG is a glucose analog that is transported across capillary membranes in the direction of a concentration gradient. However, unlike glucose, FDG is trapped in the cell and accumulates at a rate proportional to glucose utilization.[1] Since bone has low glucose utilization rates compared with the brain, heart, and striated muscle, whole-body FDGPET images of primary or metastatic skeletal neoplasms demonstrate striking focal abnormalities that have high contrast compared with the background normal bone FDG uptake pattern.[2]

FDG-PET and technetium-99m methylene diphosphonate (99mTc MDP) bone scans use different mechanisms to detect bone involvement by tumor. It is possible to image tumor metabolism directly with PET, and early studies have demonstrated a high sensitivity for the detection of skeletal metastases, although the detection of osteoblastic metastases is not as impressive.[3] FDG-PET shows a lower sensitivity for sclerotic bone metastases, especially in breast[4] and prostate cancer,[5] which may be due to the acellular nature of sclerotic lesions and hence lower volumes of viable tumor tissue within the lesion.[4] Therefore, it may be pertinent in such patients to perform both FDG-PET and a bone scan to evaluate fully the distribution of skeletal metastases.

99mTc MDP used in bone scans is absorbed onto bone surfaces, and its uptake depends on both local blood flow and osteoblastic activity. Nearly all metastases are accompanied by osteoblastic reaction and hence are easily detected on bone scans. However, predominantly lytic lesions (eg, myeloma) demonstrate poor or absent uptake on bone scans.[3] More aggressive osteolytic lesions have an outstripped blood supply that render them relatively hypoxic. Increased FDG uptake through activation of the glycolytic pathway is observed in these.[4] FDG-PET is superior to bone scan in the detection of purely osteolytic disease and in the early detection of malignancy when only the bone marrow is involved.[4,6]

In a study by Kao et al,[7] the sensitivity of FDG-PET was shown to be lower than that of bone scan. However, its specificity in the detection of malignant bone metastases was better, suggesting that a focus of tumor sufficient to stimulate increased osseous uptake of bone tracer might be below the required tumor mass needed for FDG accumulation in these instances. Studies evaluating the sensitivity of bone scans and FDG-PET scans in detecting bone metastases in lung, breast, and prostate cancer suggest that only a small percentage of bony metastases are associated with increased glycolysis, which may explain the lower sensitivity of FDGPET in these cases.[7]

An FDG-PET scan helps to differentiate benign from malignant causes of increased uptake on bone scan studies. The difference in mechanism of FDG and 18F fluoride uptake can help to identify bone tumors successfully responding to treatment in which a persistent osteoblastic response, which results in increased uptake on bone scintigraphy, may indicate normal bone repair once the tumor has been suppressed. This picture should be associated with decreased FDG uptake as opposed to tumor progression, where there should be increased FDG uptake. Uptake of FDG is assumed to occur predominantly within the tumor cells rather than as a result of bone reaction to tumor. Therefore, PET has the potential to monitor response to therapy.[3]

In an evaluation of 110 patients with non-small cell lung cancer, Bury et al[8] found that the sensitivity for detecting osseous involvement was similar for both FDG-PET and 99mTc MDP but that PET yielded better specificity. In this study, the positive and negative predictive values for FDG-PET were 90% (95% confidence interval [CI], 69%-98%) and 98% (95% CI, 92%-99%), respectively. In contrast, bone scanning revealed a positive and negative predictive value of 35% (95% CI, 23%-49%) and 96% (95% CI, 88%-99%), respectively. In a similar study evaluating 145 patients on a lesion-by-lesion basis,[9] the sensitivity and specificity of FDG-PET were 87.5% and 97.3%, respectively, compared to a sensitivity of 81.2% and a sensitivity of 74.3% for bone scans with 99mTc MDP.

In a study evaluating the detection of extrathoracic metastases in lung cancer using whole-body PET, skeletal metastases were detected in 13% of patients, and approximately 75% of these patients were asymptomatic.[10] Bone pain is usually considered a good indicator of skeletal metastases. However, up to 40% of patients with proven bone metastases are asymptomatic. For this reason it is probably useful to examine the skeletal system radiographically in patients with lung cancer.[5]

It is worthy of note that studies on FDG-PET do not unanimously show a greater sensitivity than bone scans. However, FDG-PET scans have a higher specificity and accuracy when compared to bone scans in the evaluation of bone metastases. Lytic lesions are better detected by FDG-PET, and since these are linked with a reduced survival, earlier detection of these provides earlier treatment options for this group. FDG-PET scans can also determine the presence of active metastatic disease following treatment and are useful in monitoring therapy.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: