Comparison of reprojected bone SPECT/CT and planar bone scintigraphy for the detection of bone metastases in breast and prostate cancer

Abstract

<p><strong>Objective: </strong>The aim of this study was to compare reprojected bone SPECT/CT (RBS) against planar bone scintigraphy (BS) in the detection of bone metastases in breast and prostate cancer patients.</p><p><strong>Methods: </strong>Twenty-six breast and 105 prostate cancer patients with high risk for bone metastases underwent 99mTc-HMDP BS and whole-body SPECT/CT, 1.5-T whole-body diffusion-weighted MRI and 18F-NaF or 18F-PSMA-1007 PET/CT within two prospective clinical trials (<a href="http://clinicaltrials.gov/show/NCT01339780" title="See in ClinicalTrials.gov">NCT01339780</a> and <a href="http://clinicaltrials.gov/show/NCT03537391" title="See in ClinicalTrials.gov">NCT03537391</a>). Consensus reading of all imaging modalities and follow-up data were used to define the reference standard diagnosis. The SPECT/CT data were reprojected into anterior and posterior views to produce RBS images. Both BS and RBS images were independently double read by two pairs of experienced nuclear medicine physicians. The findings were validated against the reference standard diagnosis and compared between BS and RBS on the patient, region and lesion levels.</p><p><strong>Results: </strong>All metastatic patients detected by BS were also detected by RBS. In addition, three metastatic patients were missed by BS but detected by RBS. The average patient-level sensitivity of two readers for metastases was 75% for BS and 87% for RBS, and the corresponding specificity was 79% for BS and 39% for RBS. The average region-level sensitivity of two readers was 64% for BS and 69% for RBS, and the corresponding specificity was 96% for BS and 87% for RBS.</p><p><strong>Conclusion: </strong>Whole-body bone SPECT/CT can be reprojected into more familiar anterior and posterior planar images with excellent sensitivity for bone metastases, making additional acquisition of planar BS unnecessary.</p>