This study demonstrated that remineralization occurred in approximately half of the patients with jawbone destruction following radiation therapy or chemotherapy, which is consistent with earlier observations in the vertebrae [12, 14, 20]. However, our research extended these previous findings by targeting the jawbone and including participants with a wider age range and varied tumor histology encompassing invasive and metastatic bone destruction. This wider focus facilitated the identification of additional factors influencing remineralization. Our findings revealed that age, tumor histology, site of jaw destruction and type of bone destruction were significantly associated with remineralization. These add to sex, tumor histology, and radiation dose, which have been reported in previous studies [13,14,15,16,17,18,19].

Several previous studies have reported a significant association between the type of primary tumor and bone remineralization [12,13,14, 20]. Our study found a similar trend for the tumor primary site, but this association was not statistically significant. A statistically significant difference was observed between tumor histology and remineralization, suggesting that histology-specific factors may influence remineralization. In addition, the pattern of bone destruction—classified as either invasive or metastatic—was also significantly associated with remineralization. These results support the possibility that the factors promoting remineralization are inherent to the tumor itself and not its original location.

Reports regarding the relationship between radiation and remineralization have been inconsistent; some studies have reported a significant association [12, 20, 22] while others have not detected such a correlation [14, 26, 27]. The radiation dose applied to the vertebrae in those studies ranged from 8 to 40 Gy, corresponding to the less than 60 Gy category in our study. No significant difference in the remineralization of the jawbone was detected for this category.

Further, our study suggested that remineralization occurred more frequently in patients who had not undergone radiation therapy than in those who had. The group without radiation therapy received chemotherapy alone, and remineralization was observed even in the absence of radiation exposure. Chemotherapy agents and radiation dose have diverse effects on bone metabolism and the inflammatory cascade, and their precise impact on remineralization remains incompletely understood [28, 29]. Furthermore, factors derived from tumor cell death may contribute to the induction of bone formation [30], suggesting that bone regeneration is regulated through the complex interplay between therapeutic modalities, tumor cells, and normal cellular responses.

CT values have been widely used to measure bone mineral density at remineralized sites [22]. However, bone mineral density varies within the jawbone depending on the anatomical location and specific jaw regions. To account for this variability, this study employed the mean ratio of CT values at the bone destruction or remineralization sites to those at the reference sites for comparison with clinical factors.

Previous research has reported that bone remineralization is often evident approximately 3 months after treatment [31], with subsequent CT values at remineralized sites exceeding those at reference sites [12]. We confirmed remineralization at approximately 3 months, with the CT value ratio of the remineralized sites to the reference sites reaching 1.02. The CT value ratio at the remineralized sites markedly increased to exceed the reference site value of 1.7 by 12 months.

Our study also demonstrated that patient age and sex significantly influenced the 12-month CT value ratios. Patients younger than 18 years and male patients presented with significantly lower CT values at the remineralized sites. Previous studies did not include this age group, making this the first study to establish that age plays a role in remineralization, although the number of patients younger than 18 years was limited. Some reports have shown no significant association between sex and remineralization [20, 26]. These reports included patients taking bone-modifying agents, which may have confounded earlier results. From a biological rationale, delayed bone remodeling due to cellular senescence as well as hormonal differences between sexes may account for the observed changes in CT values [32, 33]. These factors suggest that physiological bone metabolism and growth-related processes should be considered when interpreting remineralization outcomes.

Bone-modifying agents, such as bisphosphonates, affect CT values [13, 14, 20]. These agents are commonly prescribed to prevent fractures in patients with osteoporosis and bone metastases in cancers such as breast cancer, where bone metastases frequently occur [6, 8,9,10]. However, these medications carry the risk of medication-related osteonecrosis of the jaw, which can influence bone remineralization after metastasis. To eliminate this potential confounder, our study excluded patients taking bone-modifying agents. This may also explain the relatively few patients with breast cancer in our cohort.

Previous research has suggested that the primary tumor type may influence CT values after remineralization [14]. However, our study did not find statistically significant associations between the primary tumor type and CT values. Unlike earlier investigations that focused solely on bone metastases, our study analyzed metastatic bone destruction and invasion of tumors adjacent to the jawbone. This difference may have contributed to the significant differences observed.

The morphology of the remineralized jawbone is remarkably varied, echoing the diversity observed in vertebral remineralization [17, 18]. We classified remineralization morphologies as excessive, similar, or insufficient compared with the reference sites and found no associations with clinical factors. The initial morphology after remineralization persisted over time for most cases, and no changes were detected. However, some patients, especially those younger than 18 years, initially had excessive remineralization. This was followed by a shift toward normal bone morphology indicating active bone-remodeling processes. Our study revealed that most patients retained their remineralization morphology for a few years, whereas those who eventually regained their original bone shape required significantly longer than is typical for fracture healing. This prolonged process may be influenced by various factors, such as the size of the bone destruction site, presence of tumor cells, and effects of radiation and chemotherapy [29, 34].

PET/CT images further enriched our findings, capturing metabolic activity associated with bone remineralization. These findings suggest that bone remineralization occurs as a localized mass of bone formation, which is unlike the gradual healing processes observed during cyst removal. In addition, previous studies have reported that remineralization occurs through the calcification and ossification of collagen fibers aggregating within fibrous tissue that replaces dead metastatic cancer cells after irradiation; this remineralization is thought to correspond to the callus formation process during fracture healing [35]. The 18F-FDG accumulation observed at remineralized sites in this study supports the notion that the mechanism of bone remineralization may be similar to that of fracture healing, consistent with previous reports of 18F-FDG accumulation at fracture sites on PET/CT images [36, 37]. It should be emphasized that 18F-FDG reflects metabolic activity rather than bone formation itself [38]; only two patients in our cohort underwent PET/CT during the remineralization period, which limits the strength of interpretation of the study results. Nonetheless, the ability to visualize localized metabolic activity preceding morphological remineralization provides valuable insight into the cellular processes that may shape the eventual bone morphology; we believe this observation warrants further investigation.

This study had several limitations. These include the relatively small sample size, as the study was conducted at a single institution. A multicenter collaboration is recommended to increase the sample size, given the low incidence of jawbone metastases (1–8% of all bone metastases) [6, 8]. Furthermore, the retrospective nature of the study introduced variability in CT scans, as they were performed using different protocols and equipment across multiple facilities [39]. Nonetheless, the use of reference sites likely mitigated the impact of this variability.

In summary, this study provides valuable clinical insights into the factors influencing jawbone remineralization after tumor-induced destruction. We demonstrate that age, sex, tumor histology, the site and type of bone destruction, and radiation dose affect the occurrence and morphology of remineralization. Recognizing these factors may help clinicians to predict the course of bone healing and to optimize jawbone surveillance in individual patients. Furthermore, the observation of metabolic activity corresponding to remineralization on PET/CT may support recovery monitoring and inform decisions on post-treatment interventions, including the necessity and timing of surgical reconstruction or prosthetic rehabilitation. Overall, these findings may contribute to treatment planning and enhance clinical decision-making for patients with jawbone destruction.