Introduction

Bone metastases frequently occur in patients diagnosed with prostate and lung cancer, leading to severe complications and reduced survival rates1,2,3. Nuclear medicine plays an important role in early tumor metastasis detection, using highly sensitive agents. Fluorine 18 (18F)-labeled fluorodeoxyglucose (18F-FDG) has been proven to have a good performance in detecting bone metastases of different cancers. While 18F-FDG PET/CT is widely used for detecting bone metastases, its sensitivity in identifying osteoblastic lesions is limited4. Physiological bone marrow uptake and coexisting hematologic disorders can also affect diagnostic accuracy.

Ibandronate (Ibandronic acid, IBA) is a third-generation bisphosphonate clinically used to prevent osteoporosis in participants suffering from bone metastases5. 68Ga-DOTA-IBA has been successfully synthesized via a simple radiolabeling method in a previous study6. Pre-clinical study and biodistribution of 68Ga-DOTA-IBA have shown its high radiochemical purity, stability, and low toxicity6. Moreover, 68Ga-DOTA-IBA exhibits a high tumor-to-background ratio in bone metastases imaging, supporting its use as a novel PET tracer7,8.

This study aims to evaluate the diagnostic efficacy of 68Ga-DOTA-IBA in detecting bone metastases compared to 18F-FDG.

Participants and methods

Participants

This prospective study was conducted in our hospital. All participants were consecutively enrolled from October 2022 to September 2023. Sample size calculations are summarized in Supplementary Material (1) STARD checklist can be viewed in Supplementary Material (2) This study was approved by the ethics committee of the affiliated hospital of Southwest Medical University and registered on the Clinical Trial Registry (Clinical ethics registration number: ChiCTR2200064487). All participants had clear indications for 18F-FDG PET/CT for initial staging or restaging after therapy. Before undergoing 68Ga-DOTA-IBA, participants were asked to sign a consent agreement to receive the examination. Inclusion criteria included (a) pathologic confirmed primary disease, (b) at least one comparable bone metastatic lesion, and (c) agreement to undergo 68Ga-DOTA-IBA examination. Exclusion criteria included (a) previous treatment of bone metastases with bisphosphonate analog, (b) any treatment capable of stimulating bone marrow metabolism, and (c) any history of non-solid tumors or hematologic diseases.

Preparation of 18F-FDG and 68Ga-DOTA-IBA

Following a standard protocol, 18F-FDG was synthesized using an FDG-N synthesis module (PET Science & Technology, China). 68Ga was eluted from a 68Ge/68Ga generator (ITG 101, Eckert & Ziegler, Germany). DOTA-IBA was synthesized in our laboratory and provided by Shanghai New Drug Development Company, with detection performed using PET/CT (United Imaging 780, Shanghai United Imaging Medical Technology Co., Ltd.). All other equipment, software, and reagents were obtained from commercial sources.

68Ga-DOTA-IBA was prepared as previously described. Briefly, a mixture of 10 µg (10 µL) DOTA-IBA, 1 mL of sodium acetate, and 4 mL of 68GaCl3 was incubated at 95 °C for 15 min after the pH was adjusted to 4–5. The solution was then cooled and sterilized through filtration.

Radiochemical quality control was conducted via thin-layer chromatography, using a 0.1 M sodium citrate solution as the solvent. The radiochemical purity of 68Ga-DOTA-IBA was confirmed to be > 97%.

PET/CT imaging acquisition

Participants fasted for at least 6 h before imaging to ensure normal plasma glucose levels (< 8.3 mmol/L). After intravenous injection of 18F-FDG (0.1 mCi/kg), with residual activity in the empty syringe kept below 0.5 mCi, participants were instructed to drink water and rest silently. Imaging was performed 45–60 min post-injection. Whole-body CT scans (head to mid-thigh) were acquired using tube voltage: 120 kV; current: 100 mAs; slice thickness: 5.0 mm. Paired 68Ga-DOTA-IBA scans were conducted within 7 days of the 18F-FDG scans. The 68Ga-DOTA-IBA dose was calculated as 0.1 mCi/kg, with residual activity in the empty syringe below 0.5 mCi. A total acquisition of 5–6 beds at 3 min/position with a subset number 33, an iteration number 3, and a 512 \(\:\times\:\) 512 matrix were used to reconstruct PET images using CT images for attenuation correction.

Imaging review

Initially, three board-certified nuclear medicine physicians (T.X. and D.P.) with more than 5 years of experience reached a consensus on bone metastases based on CT or MRI findings. Subsequently, T.X. and D.P., blinded to each other’s assessments, independently reviewed 18F-FDG and 68Ga-DOTA-IBA images. Any discrepancies were resolved through consensus with a third nuclear medicine expert (Y.C.). Follow-up imaging 3 months later served as the reference standard for lesion confirmation. Lesions were categorized into five groups based on anatomical location: limbs, vertebras, pelvis, ribs, and skulls. Lesions were classified as positive if activity was significantly increased and markedly higher than the adjacent background; or negative: if activity was diffusely mild or indistinguishable from surrounding normal tissues.

Maximum standardized uptake values (SUVmax) for 18F-FDG and 68Ga-DOTA-IBA were automatically calculated by the software after manually placing a volume of interest (VOI) over the targeted lesion. If discrepancies arose between the two primary reviewers (T.X. and D.P.), a consensus was reached after a third review by Y.C. If there were multiple bone metastases at one site, the average value was used, which was defined as averaging the SUVmax of all lesions (no more than 5 lesions at one location) or averaging the SUVmax of 5 biggest lesions ranked by their longer length (more than 5 lesions at one location). The length of each selected lesion for SUVmax comparison was recorded. Besides, the morphological bone changes of each selected were also recorded as osteolytic, osteoblastic, and mixed.

Statistical analysis

Radiotracer uptake, semi-quantified by SUVmax, was compared using the paired t or Mann-Whitney U tests. Detection rates between the two scans were compared using the McNemar test. Two-tailed P < 0.05 was considered statistically significant. All statistical analysis and graph building were conducted using SPSS software (version 26.0; IBM) and GraphPad (version 9.5.0; PRISM).

Results

Participants

A total of 46 participants (28 men and 18 women; median age: 64 years [interquartile range: 53–68 years]) were consecutively enrolled in our hospital between October 2022 and September 2023. All participants had a confirmed diagnosis through biopsy or surgery. The characteristics of the participants are summarized in Table 1, and the study flow diagram is presented in Fig. 1. The interval between the two imaging examinations was 3.1 ± 1.2 days. Among these participants, 22 of them had lung cancer and 12 of them had prostate cancer. The number of participants suffered from nasopharyngeal, intestinal, and neuroendocrine cancer was 4, 3 and 3 respectively. Others included 1 renal carcinoma and 1 spleen hemangiosarcoma. Among all participants, 23 had osteoblastic bone lesions as major metastatic type, 13 had an osteolytic type, and 10 had a mixed type.

Table 1 Characteristics of participants.
Full size table
Fig. 1
figure 1

Flow diagram shows participants selection for this study. 18F-FDG, fluorine 18 (18F)-labeled fluorodeoxyglucose; 68Ga-DOTA-IBA, gallium 68 (68Ga)-labeled DOTA-conjugate ibandronic acid.

Full size image

Detecting bone metastases from different sites

Morphological imaging identified 1103 bone lesions among the 46 participants. Tables 2 and 3 show the diagnostic sensitivity and radiotracer uptake comparison between 18F-FDG and 68Ga-DOTA-IBA.

Table 2 Comparisons of tracer uptake (SUVmax) of lesions from different sites.
Full size table
Table 3 Comparison of diagnostic performance of lesions from different sites.
Full size table

Limb lesions

  • 153 bone metastases were detected in 26 participants.

  • Detection rates 18F-FDG vs. 68Ga-DOTA-IBA): 64.1% (98/153) vs. 73.2% (112/153) (P = 0.024).

  • SUVmax 18F-FDG vs. 68Ga-DOTA-IBA): 4.1 (1.8–15.2) vs. 8.5 (2.7–29.2) (P = 0.004).

Vertebral lesions

  • 485 vertebral metastases were found in 37 participants.

  • Detection rates 18F-FDG vs. 68Ga-DOTA-IBA): 67.4% (327/485) vs. 78.1% (379/485) (P < 0.001).

  • SUVmax 18F-FDG vs. 68Ga-DOTA-IBA): 5.2 (2.3–14.9) vs. 9.8 (3.7–31.4) (P < 0.001).

Rib lesions

  • 209 metastatic rib lesions were identified in 31 participants.

  • Detection rates 18F-FDG vs. 68Ga-DOTA-IBA): 62.2% (130/209) vs. 86.6% (181/209) (P  < 0.001).

  • SUVmax 18F-FDG vs. 68Ga-DOTA-IBA): 2.6 (1.2–8.9) vs. 5.8 (2.5–21.0) (P < 0.001).

Pelvic lesions

  • 206 pelvic metastases were found in 30 participants.

  • Detection rates 18F-FDG vs. 68Ga-DOTA-IBA): 68.9% (142/206) vs. 78.6% (162/206) (P= 0.019).

  • SUVmax 18F-FDG vs. 68Ga-DOTA-IBA): 5.0 (1.6–12.8) vs. 8.4 (3.5–23.0) (P= 0.001).

Skull lesions

  • 50 skull metastases were detected in 19 participants.

  • Detection rates 18F-FDG vs. 68Ga-DOTA-IBA): 38.0% (19/50) vs. 80.0% (40/50) (P < 0.001).

  • SUVmax 18F-FDG vs. 68Ga-DOTA-IBA): 3.1 (1.2–8.9) vs. 5.8 (2.5–21.0) (P = 0.005).

Detection of bone metastases by different types

Among all participants involved, 23(50%) of them had osteoblastic bone lesions as major metastatic type, 13(28%) for osteolytic type and 10(22%) for mixed type. The diagnostic performance for different lesion types is summarized in Table 4, while the comparison of radiotracer uptake (SUVmax) in osteoblastic and osteolytic lesions is presented in Fig. 2. Representative imaging is shown in Fig. 3.

Table 4 Comparison of diagnostic performance detecting lesions of different types.
Full size table
Fig. 2
figure 2

Comparison of radiotracer uptake of lesions with different types. *Paired samples t test (both P < 0.001), **Mann–Whitney U test (P < 0.001). The number of osteoblastic and osteolytic bone metastases to compare was 249 and 164, respectively. SUVmax was displayed with mean and 95% CI. 68Ga-DOTA-IBA, gallium 68 (68Ga)-labeled DOTA-conjugate ibandronic acid; 18F-FDG, fluorine 18 (18F)-labeled fluorodeoxyglucose; SUVmax, maximum standardized uptake value.

Full size image
Fig. 3
figure 3

Images of a 69-year-old man diagnosed with prostate carcinoma who had a history of endocrine therapy for 2 years. (a) The maximum intensity projection (MIP) image of 18F-FDG showed moderate radiotracer uptake in a thoracic vertebra. (e) The maximum intensity projection (MIP) image of 68Ga-DOTA-IBA demonstrated more lesions in the cervical vertebra and pelvis. (b) An osteoblastic lesion in the C3 vertebra right attachment (curved arrows) was negative on 18F-FDG (SUVmax=2.0) and positive on 68Ga-DOTA-IBA (SUVmax=8.7). (c) Another osteoblastic lesion in the T10 vertebra left side (arrowheads) was both positive on 18F-FDG (SUVmax=5.2) and positive on 68Ga-DOTA-IBA (SUVmax=35.0). (d) Osteoblastic lesions in bilateral ilium (dotted arrows and arrows) were negative on 18F-FDG (left: SUVmax=1.4, right: SUVmax=1.6) and positive on 68Ga-DOTA-IBA (left: SUVmax=7.6, right: SUVmax=8.7). Follow-up imaging 3 months later confirmed the presence of these bone metastases. 68Ga-DOTA-IBA, gallium 68 (68Ga)-labeled DOTA-conjugate ibandronic acid; 18F-FDG, fluorine 18 (18F)-labeled fluorodeoxyglucose.

Full size image

Osteoblastic lesions

  • 635 lesions were identified.

  • Detection rates 18F-FDG vs. 68Ga-DOTA-IBA): 51.5% (327/635) vs. 83.3% (529/635) (P < 0.001).

  • SUVmax 18F-FDG vs. 68Ga-DOTA-IBA): 3.96 (95% CI 3.57–4.35) vs. 7.88 (95% CI 7.09–8.66) (P < 0.001).

Osteolytic lesions

  • 226 lesions were identified.

  • Detection rates 18F-FDG vs. 68Ga-DOTA-IBA): 79.6% (180/226) vs. 81.9% (185/226) (P = 0.533).

  • SUVmax 18F-FDG vs. 68Ga-DOTA-IBA): 6.28 (95% CI 5.71–6.85) vs. 8.40 (95% CI 7.48–9.32) (P < 0.001).

Additionally, the radiotracer uptake of 18F-FDG by osteoblastic lesions 3.96 (95% CI 3.57–4.35) was lower than that of 18F-FDG uptake by osteolytic lesions 6.28 (95% CI 5.71–6.85) (P < 0.001, Mann–Whitney U test).

Detection of bone metastases in lung and prostate cancer

Among the involved 46 participants, 22 participants were diagnosed with lung cancer and 12 participants with prostate cancer. Comparisons of diagnostic performance detecting lesions of lung or prostate cancer were summarized in Table 5. The comparison of radiotracer uptake using SUVmax of 50 lesions from lung cancer participants and 60 lesions from prostate cancer participants was shown in Fig. 4. Images of two representative participants are shown in Figs. 5 and 6.

Table 5 Comparison of diagnostic performance in lung and prostate cancer.
Full size table
Fig. 4
figure 4

Comparison of radiotracer uptake of bone metastases from different diseases.*Paired samples t test (Both P < 0.001), **Mann–Whitney U test (P = 0.024). The number of bone lesions from lung cancer and prostate cancer to compare SUVmax was 50 and 60, respectively. SUVmax was expressed with mean and 95% CI. 68Ga-DOTA-IBA, gallium 68 (68Ga)-labeled DOTA-conjugate ibandronic acid; 18F-FDG, fluorine 18 (18F)-labeled fluorodeoxyglucose; SUVmax, maximum standardized uptake value.

Full size image
Fig. 5
figure 5

Images of a 46-year-old woman diagnosed with lung carcinoma who underwent 18F-FDG examinations for initial staging. (a) The maximum intensity projection (MIP) image of 18F-FDG showed moderate-low radiotracer uptake in the right pelvis (arrows, SUVmax=3.9) and left head of the femur (arrowheads, SUVmax=2.4). (b) The maximum intensity projection (MIP) image of 68Ga-DOTA-IBA demonstrated obvious radiotracer uptake in the right pelvis (arrows, SUVmax=13.6) and left head of the femur (arrowheads, SUVmax=9.2). (c,d) Axial fusion images of a mixed type of lesion showed that it is negative on 18F-FDG and positive on 68Ga-DOTA-IBA (curved arrow, SUVmax=2.5 vs. 8.0). CT showed osteogenic bone changes at the corresponding sites. Follow-up imaging 3 months later confirmed the presence of these bone metastases. 68Ga-DOTA-IBA, gallium 68 (68Ga)-labeled DOTA-conjugate ibandronic acid; 18F-FDG, fluorine 18 (18F)-labeled fluorodeoxyglucose.

Full size image
Fig. 6
figure 6

Images of a 52-year-old woman diagnosed with lung carcinoma treated by surgery who underwent 18F-FDG examinations for restaging. (a) The maximum intensity projection (MIP) image of 18F-FDG showed obvious radiotracer uptake in multiple vertebras (arrowheads, mean SUVmax=5.7) and right ribs (arrows, SUVmax=6.2). (b) The maximum intensity projection (MIP) image of 68Ga-DOTA-IBA also demonstrated obvious radiotracer uptake in multiple vertebras (arrowheads, mean SUVmax=19.4) and right ribs (arrows, SUVmax=12.0). (c,d) Axial fusion images of an osteoblastic lesion showed that it is negative on 18F-FDG and positive on 68Ga-DOTA-IBA (curved arrow, SUVmax=2.9 vs. 14.7). CT showed osteogenic bone changes at the corresponding sites. Follow-up imaging 3 months later confirmed the presence of these bone metastases. 68Ga-DOTA-IBA, gallium 68 (68Ga)-labeled DOTA-conjugate ibandronic acid; 18F-FDG, fluorine 18 (18F)-labeled fluorodeoxyglucose.

Full size image

Lung cancer

  • 303 metastatic lesions were identified.

  • Detection rates 18F-FDG vs. 68Ga-DOTA-IBA): 78.2% (237/303) vs. 87.1% (264/303) (P < 0.001).

  • SUVmax 18F-FDG vs. 68Ga-DOTA-IBA): 6.01 (95% CI: 4.84–7.19) vs. 9.10 (95% CI: 7.57–10.63) (P < 0.001).

Prostate cancer

  • 593 metastatic lesions were identified.

  • Detection rates 18F-FDG vs. 68Ga-DOTA-IBA) 55.0% (326/593) vs. 84.7% (502/593) (P < 0.001).

  • SUVmax 18F-FDG vs. 68Ga-DOTA-IBA): 4.29 (95% CI: 3.51–5.07) vs. 10.44 (95% CI: 8.57–12.30) (P < 0.001).

Additionally, the radiotracer uptake of 18F-FDG by lesions from lung cancer 6.01 (95% CI 4.84–7.19) was higher than 18F-FDG uptake by lesions from prostate cancer 4.29 (95% CI 3.51–5.07) (P = 0.024, Mann–Whitney U test).

Discussion

Early and precise diagnosis of bone metastases is crucial for patient management. Although various bone-seeking imaging agents are available, 18F-FDG is commonly used for initial staging and restaging. Our study supports the rationale for using 68Ga-DOTA-IBA to improve the management of patients with bone metastases, considering the therapeutic efficacy of 177Lu-DOTA-IBA. (7) In addition, it has been shown that 18F-NaF is a response predictor in nasopharyngeal, breast, and prostate cancer, including radiotherapy, endocrine therapy or 223RaCl9,10,11,12,13. 68Ga-DOTA-IBA may also have prognostic value for precision therapy.

Recent studies have shown that 68Ga-DOTA-IBA and 18F-NaF have comparable diagnostic efficacy and are superior to bone scan, and they are commonly used imaging agents for targeting bone metastases14,15. 68Ga-labeled fibroblast activation protein (FAP) inhibitor (FAPI) is an emerging imaging agent that may be superior to 18F-FDG in the diagnosis of bone metastases in different tumors and may have prognostic significance irrespective of the primary lesion16,17,18,19. However, a recent meta-analysis found that there may be no significant difference between 18F-FDG and 68Ga-FAPI in diagnosing bone metastases20. This may be because some benign lesions have 68Ga-FAPI uptake21. On the other hand, there are fewer studies of 68Ga-FAPI in diagnosing bone metastases. Prostate-specific membrane antigen (PSMA) is an imaging agent targeting prostate cancer and has a high accuracy for prostate cancer with bone metastases22. A recent meta-analysis found that 68Ga-PSMA PET/CT was superior to bone scan for diagnosing bone metastases23. There is a need to investigate 68Ga-DOTA-IBA versus these emerging imaging agents in the future as necessary.

A previous study reported that bone metastasis was most frequently detected by 18F-FDG PET in vertebras (74%), followed by pelvic bones (70%), and ribs (65%)24. Our findings align with this, as 37 (80.4%) participants had lesions in the vertebrae, 31 (67%) in the ribs, and 30 (65%) in the pelvis. The lower detection rate of 18F-FDG in the skull is likely due to physiological uptake in the brain25. Our study found a 38% sensitivity in detecting skull metastases, similar to a previous report of 42.9%26.

The sensitivity of 18F-FDG PET to detect bone metastases from lung cancer is 78.2% in our study which is lower than the pooled sensitivity previously reported in two meta-analyses27,28. This discrepancy may be due to the small sample size and the heterogeneity of participants in our study.

Although previous studies have demonstrated the potential of 18F-FDG for detecting bone metastases across different cancer types29,30. It shows limited diagnostic efficacy in prostate cancer due to the low metabolism of prostate cancer cells. In our study, there are 12 participants diagnosed with prostate cancer, and 2 of them were treated by androgen deprivation therapy. Our results are consistent with previous studies. Shreve’s study demonstrated a sensitivity of 65% in 22 untreated prostate cancer participants31. Pietrzak reported a low SUVmax of bone metastases from 27 participants32. Moreover, diagnostic efficacy could be reduced to 18% for those treated with androgen deprivation therapy33.

Another reason why 18F-FDG is poor in detecting bone metastases for prostate cancer may attributed to the osteoblastic type. In our study, 18F-FDG showed lower sensitivity for osteoblastic lesions (51.5%) compared to osteolytic lesions (79.6%), with significantly lower radiotracer uptake (3.96 vs. 6.28). Cook et al. reported lower metabolic activity in osteoblastic metastases (0.95) compared to osteolytic metastases (6.77) in 23 breast cancer participants34. Nakai’s study reported a visualization rate of 55.6% vs. 100%, analyzing 18 blastic and 10 lytic sites35. Previous studies also reported similar results in other cancers36,37,38.

Bisphosphonates are the standard treatment for tumor patients after diagnosis of bone metastases. The effect of bisphosphonates on their radiolabeled analogs remains unclear. Previous studies have suggested that bisphosphonate therapy does not affect the diagnostic efficacy of bone scan39,40,41. Thus, bisphosphonates may also not affect the diagnostic efficacy of 68Ga-DOTA-IBA, but further studies are needed.

Although our study indicates that 68Ga-DOTA-IBA PET may outperform 18F-FDG in detecting bone metastases, 18F-FDG remains valuable for providing information about primary lesions. Additionally, we excluded participants who had received bisphosphonates before the examination, as it remains uncertain whether these treatments affect diagnostic efficacy. Other limitations of our study include the small sample size and the absence of breast cancer participants, despite breast cancer being a common cause of bone metastases. Moreover, limb lesion analysis was incomplete because 18F-FDG PET scans were limited to the head-to-mid-thigh region, and hand and forearm imaging was missing due to participant positioning. Lastly, we only analyzed lesions with visible morphological changes. We did not record false-positive lesions or calculate specificity, as it would be unethical to perform lesion-by-lesion biopsies or delay treatment for confirmation.

Conclusions

68Ga-DOTA-IBA PET may outperform 18F-FDG in bone metastase detection, especially for those with osteoblastic bone destruction or those from prostate cancer.