Zoledronate

After acclimatization, all animals received antiresorptive treatment of zoledronate (0.05 mg/kg) intravenously once a week for the 12 subsequent weeks. The weekly zoledronate injections and all examinations of the oral cavity were performed under deep sedation. The treatment went ahead as planned, without any incidences or complications.
The first surgical intervention was performed after the above pretreatment of 12 weeks at ARI. All surgical procedures were performed under general anesthesia according to our standardized protocol [15 (link)]. The first molar teeth of the mandible and the maxilla were extracted unilaterally (Fig. 1) to determine whether any differences occurred between the upper and lower jaw. On the contralateral side, no teeth were extracted, which was therefore used as the control side. All tooth extractions were performed without incidences or complications, apart from some small root remnants.Fig. 1

TF group, intervention side: first surgical intervention after 12 weeks of antiresorptive pretreatment. Left maxilla of a Göttingen minipig a before and b after extraction of the first molar. c Exposed bone in the extracted area 8 weeks later

After surgery, antiresorptive treatment was continued for the subsequent 8 weeks (0.05 mg/kg intravenously once weekly). The weekly Zoledronate injections and examinations of the oral cavity were again performed under deep sedation, without any incidences or complications. Regular examinations for clinical signs of oral lesions were performed throughout the study and photo-documented (Fig. 1). Ten days before death, the animals were randomly assigned to two groups. Fluorescent-labeling of the bone with tetracycline (25 mg/kg intravenously once a day) was performed in half of the animals (TF group, n = 4), whereas the other half received no fluorochrome labeling (AF group, n = 4).
The second surgical intervention was performed 8 weeks after the first surgical intervention and after a continuation of the antiresorptive treatment for another 8 weeks. All surgical procedures were performed under general anesthesia according to our standardized protocol [15 (link)]. Following in vivo sub-periosteal preparation of the MRONJ regions (and of the control side), detailed high-resolution photographic documentation was performed on macroscopic necrosis signs and on non-diseased areas (with and without illumination by fluorescence with the VELscope Vx system® (blue excitation light: spectrum 400–460 nm and a green filter with an emission open from 460 nm)) (Fig. 2). This enabled subsequent planned data analysis, correlating in vivo and in vitro macroscopic and histological results.Fig. 2

TF group, intervention side: a second surgical intervention 8 weeks after first surgical intervention. Necrosis and viable areas following in vivo sub-periosteal preparation b without and c with fluorescence light

Extraction of data from the clinical data warehouse (CDW) of a single institute with a high-volume hematology hospital and a dental hospital, data of a total of 1434 patients who were diagnosed with multiple myeloma (ICD-10 C90, C900 with confirmed diagnosis) and treated with bisphosphonates (intravenous pamidronate or zoledronate) in periods January 2009 through December 2019 were obtained, then assessed for eligibility. Patients who deceased within one year of diagnosis, were treated with bisphosphonates for less than one year, had a previous history of treated osteoporosis, were ever treated with other class agents (e.g., denosumab), or had insufficient radiographic/clinical data were excluded from the study. Medical charts, including dental records, were retrospectively reviewed, then plain anteroposterior and lateral view femur radiographs with supplementary bone scan images were thoroughly reviewed by three orthopedic surgeons for 644 eligible patients who met the criteria (Figure 1).
The American Society for Bone and Mineral Research (ASBMR) task force 2013 revised definition of AFF [22 (link)] was used to evaluate the presence of AFFs or incomplete AFFs. The staging system suggested in the 2009, 2014 updated position paper from the American Association of Oral and Maxillofacial Surgeons [19 ,23 ] was used for the diagnosis of MRONJ in this study, when all patients were referred to the dental clinic for an oral health assessment with or prior to the initiation of the bisphosphonate therapy. Chance-corrected kappa coefficient of inter-observer variability for diagnosing AFF was 0.88.
For independent variables, age, sex, body weight (kg), body mass index (BMI, kg/m²), total doses of bisphosphonate used, follow-up period, and treatment periods including a complete history of prescribed doses were collected. By reviewing the medical records, a considerable portion of the population had a prescription history of switching pamidronate or zoledronate to one another. Adjustments were necessary before arithmetically unifying the total dose to compare the accumulative dose of the two bisphosphonates into one variable. To adjust the relatively powerful potency of zoledronate over pamidronate, 100-times weight was applied to accumulative zoledronate doses in milligrams [24 (link),25 (link),26 (link)] for the addition with pamidronate dose. After the simple sum of total accumulative dose (mg) of the two bisphosphonates was calculated, these per BMI (kg/m²) and per body weight (kg) as well as potency-weighted accumulative dose (mg) per body weight (kg) were also assessed as variables for each complication.
Incidences of AFF and MRONJ in the study population were initially calculated, then risk factor analyses were carried out to select a variable for subsequent cutoff titration to obtain the safety limit dose. Primary outcomes were separately recorded for surgical intervention on the femur as one endpoint and prominent radiologic evidence of impending or apparent AFF as another. Univariate logistic regression for all dependent variables, then multiple logistic regression were conducted to search for the risk factors and set a variable for subsequent analysis using the receiver operating characteristics (ROC) curve analysis to titrate cutoff values for the safety limit of accumulative dosages. The area under the curves (AUC) was again tested by Pearson chi-square to obtain odds ratios for each complication. p-values under 0.05 were considered statistically significant. All statistical analyses were conducted using SPSS (version 26.0, SPSS Inc., Chicago, IL, USA).

293T cells at 2*10⁴ cells/100 μL (DMEM, 10% FCS) per well were cultured in triplicates in 96 well-plate flat bottom with or without 25 μM zoledronate (SIGMA) overnight. The next day, cells were washed twice with PBS, and Vγ9Vδ2 T cells expanded from PBMCs at 2*10⁴ cells/100 μL per well were added and cultured for 4 hours. After 4 hours, supernatants were frozen at −20°C until human INFγ assay ELISA (Invitrogen, EHIFNG) could be performed as per the manufacturer’s instructions. For the CD107a assay, 293T cells were seeded as above-mentioned. Vγ9Vδ2 T cells expanded from PBMCs were also added as above-mentioned but along with anti-CD107a-PE (BD Pharmingen) conjugated antibody and cultured for 4 hours. After 4 hours, the cells were collected from the wells as triplicates and washed once with PBS. After which cells were treated with anti-human Vδ2-FITC (Beckman Coulter) conjugated antibody for 20 mins and washed once, followed by analysis at FACSCalibur (BD) for the percentage of Vδ2-FITC and CD107a-PE population.

DC culture is examined. Human bone marrow-derived mononuclear cells (MNC) were obtained from Lonza (2M-125C). Hematopoietic stem cells were isolated from MNC by magnetic separation, using the CD34 microbead kit Ultrapure human (Miltenyibiotec, Bergisch Gladbach, Germany 130-100-453). GM-CSF (PeproTech, Cranbury, NJ, USA, AF-300-03-1000), SCF (PeproTech, AF-300-07-1000), and FLT-3ligand (PeproTech, AF-300-19-1000) were used for proliferation, lasting about 14 days. Thereafter, GM-CSF and IL-4 (PeproTech, AF-200-04-1000) were used to differentiate cells into DC for one week. At the end of the differentiation process, DC was pulsed with WT1 protein for antigen recognition and maturation was induced with zoledronate. After maturation, CDW is complete.
T cell culture is examined. Human peripheral blood mononuclear cells (PBMC) were obtained from Lonza (CC-2702). Naïve T cells were isolated from PBMC by magnetic separation using the Pan T cell isolation kit human (Miltenyibiotec, 130-096-535). IL-2 (PeproTech, 200-02) was used for proliferation, lasting about 10~14 days. TransAct (Miltenyibiotec, 130-111-160) and CDW were used for activation on the first and seventh days of culture.
These cells were routinely grown in HyClone RPMI 1640 media (Sigma-Aldrich, Burlington, MA, USA, SH30255.01) supplemented with 10% Fetal bovine serum (Gibco, Carlsbad, CA, USA, 10099-141) and 1% Gentamicin (Gibco, 15710-064) at 37 °C in a humidified atmosphere containing 5% CO₂.