Bone scan index of the jaw: a new approach for evaluating early-stage medication-related osteonecrosis

Objectives A computer-aided diagnosis of bone scintigraphy (BS) using a bone scan index (BSI) has been shown to enhance diagnostic accuracy and reproducibility of bone metastases. However, no study has applied BSI to the diagnosis of osteonecrosis of the jaw (ONJ). Medication-related ONJ (MRONJ) is an important side effect of antiresorptive drugs (ARD) prescribed for bone metastases of advanced cancer. Since the early detection of MRONJ has been considered important for the patient care, the aim of this study was to validate a diagnostic ability of BSI for early-stage MRONJ.

Methods A total of 85 BS from 70 cancer patients treated with ARD were evaluated retrospectively. The types of cancer were prostate (53%), kidney (11%), breast (11%), lung (7%), and others (17%). Regarding ARD, 77% of the patients had bisphosphonate, 14% had denosumab, and 9% had both (non-simultaneous). MRONJ legions were staged according to the American Association of Oral and Maxillofacial Surgeons staging system. All patients underwent Tc-99m methylene diphosphonate BS. Tracer uptakes in the jaw were analyzed semiquantitatively by the following 3 parameters. The first was BSI of the jaw relating to MRONJ (BSIJ_MRONJ). Software BONENAVI (FUJIFILM RI Pharma, Co. Ltd, Tokyo, Japan; EXINIbone, EXINI Diagnostics AB, Lund, Sweden) could detect abnormal intensities based on the learning with an artificial neural network system, and calculate each regional BSI (rBSI), which was defined by the fraction of abnormality to the entire skeleton (%). Using the regions detected by the software, MRONJ-related regions were manually and retrospectively selected according to both dental records and serial scintigraphic images, and the sum of mandibular and/or maxillary rBSI was referred as BSIJ_MRONJ. The second was BSIJ_max. Among the regions detected by the software, only the largest one in the jaw was manually selected and the rBSI was referred as BSIJ_max. The third was conventional semiquantitative parameter, uptake ratio (UR), which was calculated as a ratio of the maximum jaw count-to-average count of the forehead. Receiver operating characteristic (ROC) analysis was performed. The significance of high BSIJ_max for the early-stage MRONJ screening was calculated by Fisher’s exact test.

Results A total of 24 (34%) patients developed MRONJ, and their average BSIJ_MRONJ and BSIJ_max were almost same at a half year before the diagnosis of stage 2 MRONJ (0.10±0.07% and 0.09±0.05%). The BSIJ_max was significantly higher in patients who developed MRONJ than those who did not (0.05±0.04%, p = 0.003). The average BSIJ_max of patients who developed MRONJ increased to 0.17±0.09% around the timing of stage2 diagnosis. The ROC analysis showed sensitivity and specificity of 78% and 91%, respectively, using the cutoff value of 0.10% around the diagnosis of stage 2 MRONJ. Using the cutoff value of BSIJ_max a half year before the diagnosis of stage 2 MRONJ, the sensitivity and specificity for predicting MRONJ were 45% and 91%, respectively. The BSIJ_max 蠅0.10% were significantly more often observed in patients who subsequently developed stage 2 MRONJ a half year after BS, compared to patients who did not (p = 0.002, odds ratio = 8.6). ROC analysis of UR also showed similar diagnostic ability; namely, optimal cutoff value of UR, sensitivity, and specificity were 6.0, 89%, and 70%, respectively, around the diagnosis of stage 2 MRONJ. When diagnostic threshold of 6.0 was used to predict stage 2 MRONJ, the sensitivity and specificity were 80% and 70%, respectively. The merit of BSIJ_max was high reproducibility compared with UR for which manual region of interest setting was always required.

Conclusions The BSIJ using quantitative bone scan provided a new approach for evaluating and screening early-stage MRONJ. For predicting occurrence of MRONJ, the threshold of BSIJ_max = 0.10% may be used in patients treated with ARD, and differential diagnosis of MRONJ is recommended.