Physiological distribution and dosimetry of Al¹⁸F-NOTA-Pentixafor in humans

Abstract

Objective

This study aims to quantify the biodistribution of Al¹⁸F-NOTA-Pentixafor (^{1 8} F-CXCR4) using PET/CT imaging, establish reference ranges of organ radiotracer uptake, characterize uptake patterns in benign lesions, and evaluate radiation dosimetry.

Methods

In this retrospective study, we analyzed ^{1 8} F-CXCR4 PET/CT data from 51 patients, with radiotracer uptake in organs and benign lesions quantified using SUVmax and SUVmean. Biodistribution differences according to sex, age, and gallbladder excretion status were evaluated. In addition, three postmenopausal females underwent serial whole-body PET/CT imaging during the rapid distribution, equilibrium, and clearance phases of ^{1 8} F-CXCR4, and time–activity curves were used to calculate organ-absorbed doses based on the ICRP adult female model.

Results

Physiological ^{1 8} F-CXCR4 radiotracer uptake was highest in the urinary tract, biliary system, and uterus (SUVmax ≥ 4.6), moderate in the myocardium, lungs, and pancreas (SUVmax 1.30–3.16), and lowest in the brain, spinal cord, and skeletal muscle (SUVmax ≤ 0.84). No significant sex differences were found for either SUVmax or SUVmean (P > 0.05), except for higher pancreatic SUVmean in males (P = 0.01) and higher lung and blood-pool SUVmean in females (P = 0.02). Younger patients exhibited higher biliary SUVmax than older patients (P = 0.04). Although fasted patients showed increased gallbladder tracer accumulation, the difference was not statistically significant (P = 0.06). Benign lesions demonstrated a wide range of radiotracer uptake (SUVmax 1.25–6.22). The effective dose of ^{1 8} F-CXCR4 was estimated to be 2.02E-02 mSv/MBq, with the kidneys receiving the highest absorbed dose.

Conclusion

¹⁸F-CXCR4 exhibits a stable physiological biodistribution profile minimally influenced by sex or age, thereby enabling the establishment of reliable diagnostic reference ranges. Analysis of CXCR4 expression in benign lesions supports basic research and diagnosis/treatment of related diseases, while confirming dosimetric safety.

Trial registration

The trial is retrospectively registered at the Chinese Clinical Trial Registry (ChiCTR) (registration number: ChiCTR2400090595, registration date: 2024–10-09).

The C-X-C chemokine receptor type 4 (CXCR4), a G protein-coupled receptor on hematopoietic stem cells, immune cells, and endothelial cells, regulates cell migration, proliferation, and survival [1, 2]. Its primary ligand, CXCL12 (also known as stromal cell-derived factor 1, SDF-1), drives these functions via the CXCR4–CXCL12 axis, controlling immune cell trafficking, hematopoiesis, and pathological processes like cancer and inflammation [3]. In oncology, CXCR4 over-expression promotes tumor progression, metastasis, and therapy resistance, making it a key target for nuclear medicine diagnostics and therapeutics [4, 5].

CXCR4 expression is elevated in tumors such as indolent lymphomas, aldosterone-producing adenoma, and multiple myeloma, enhancing tumor cell survival and microenvironment interactions [6–10]. In indolent lymphomas, including chronic lymphocytic leukemia (CLL) and Waldenström’s macroglobulinemia (WM), CXCR4 over-expression is linked to poor prognosis due to CXCL12-mediated apoptosis resistance [11–13]. In WM, CXCR4 mutations, present in 30–40% of patients, correlate with increased bone marrow involvement and resistance to therapies like ibrutinib [14, 15].

Given that the typically low metabolic activity of indolent lymphomas can lead to underestimated tumor involvement on ¹⁸ F-FDG PET, ⁶⁸ Ga-Pentixafor PET is the preferred imaging modality for assessing CXCR4 expression in these cases [16–18]. Al ¹⁸ F-NOTA-Pentixafor (further referred to as ¹⁸F-CXCR4) is a ¹⁸ F-labeled tracer targeting CXCR4. This tracer uses ¹⁸ F, which has a longer half-life of 110 minutes and a low positron range, resulting in improved image resolution and clinical utility. It also enhances the visualization of CXCR4-positive lesions across various tumor types. [19, 20].

Imaging with ¹⁸ F-CXCR4 poses challenges due to limited understanding of its distribution across different organs and tumor types. High radiotracer activity in the spleen, a common site for lymphoma, can complicate distinguishing between malignant and normal tissues [21]. Moreover, CXCR4 expression in benign inflammatory conditions increases the risk of false-positive findings [22]. At the same time, limited pharmacokinetic and dosimetry data for ¹⁸ F-CXCR4 clinical protocol optimization and future radioligand therapy studies [23]. Therefore, defining normal biodistribution patterns and characterizing radiation exposure are essential for accurate lesion identification and understanding the role of CXCR4 in tumor-microenvironment interactions.

This study utilizes ¹⁸ F-CXCR4 PET/CT imaging and dosimetry assessments to evaluate CXCR4 expression in healthy normal organs. It lays the groundwork for basic research and diagnostic applications related to CXCR4-related diseases. The study establishes reference thresholds for the physiological distribution of CXCR4 and the uptake of radiotracers in benign lesions, while also quantifying radiation exposure to ensure safety.

Methods

Patients

This study, approved by the Ethics Committee of Sichuan Cancer Hospital (SCCSMC-01–202-87) and registered with the Chinese Clinical Trial Registry (ChiCTR2400090595), retrospectively analyzed ¹⁸F-CXCR4 PET/CT data from 51 patients examined between February 2024 and June 2025. Eligible patients were aged 18 years or older, had tumors pathologically confirmed by pathology, and provided written informed consent. Exclusions included the absence of pathological confirmation, suboptimal image quality due to tracer extravasation, motion, metal artifacts, or technical issues and incomplete clinical records.

Three postmenopausal female patients undergoing dosimetry scans participated in the study with institutional ethical approval and provided written informed consent after fully understanding the study objectives and potential risks.

Radiosynthesis and quality control

The radiosynthesis process has been described previously [24]. Quality assurance of ¹⁸F-CXCR4 was performed using High-Performance Liquid Chromatography (HPLC, Shimadzu LC-15, Suzhou, China) to ensure a radiochemical purity greater than 99%. Additional evaluations of appearance, color and pH were conform to pharmacopeia standards.

¹⁸F-CXCR4 PET/CT scanning

Patients did not need to fast, but fasting duration and food type were recorded. Each received an intravenous injection of ¹⁸F-CXCR4, rested for 60 minutes, and voided their bladder before scanning. PET/CT scans (Siemens, Germany) covered the cranial roof to mid-thigh. Low-dose non-contrast CT (1.3–1.5 mSv; 140 keV, 42 mAs, 5 mm slice thickness, 0.8 pitch) was followed by a PET scan (3D-Flow-Motion, 2 minutes/bed position). Data were reconstructed using ordered subset expectation maximization (three iterations, 21 subsets). Focal diagnostic CT was performed as needed. Patients were instructed to report any abnormalities within 30 minutes post-examination.

Image analysis and quantification

Two board-certified nuclear medicine physicians independently evaluated the images, resolving discrepancies through consultation. Tracer uptake was quantified using Standardized Uptake Value (SUVmax and SUVmean). Organs with SUVmax values > 1.88, derived from the mean SUVmax of the thoracic aorta across all patients, were considered to exhibit increased ¹⁸F-CXCR4 uptake, whereas those with SUVmax ≤1.88 were considered negative. Regions of interest (ROIs), measuring 1–2 cm in diameter, were placed within the parenchyma of organs unaffected by tumors. The mean SUV of bilateral ROIs was recorded for paired organs.

In benign lesions, ROIs were manually delineated to encompass areas of increased ¹⁸F-CXCR4 tracer uptake, defined as uptake higher than the surrounding normal tissue. These lesions were confirmed by pathological analysis or a combination of clinical history, imaging, and follow-up. Malignant lesions were analyzed by selecting the hottest lesion, defined as the one with the highest SUVmax, for quantitative assessment. A three-dimensional volume of interest (VOI) was manually drawn around the entire CXCR4-avid area based on PET/CT fusion images. SUVmax and SUVmean values were recorded for the selected lesion. Tumor-to-background ratios (TBRs) were calculated using blood pool activity derived from the thoracic aorta as the reference tissue. Additionally, patients were categorized into fasting (≥8 hours) and non-fasting ( < 8 hours) groups to assess gallbladder tracer uptake.

Radiation dosimetry

Three postmenopausal female patients underwent whole-body PET/CT imaging at different time points to evaluate the dosimetric properties of ¹⁸F-CXCR4. The mean injected activity was 6.7 mCi (247.9 MBq). Injected activity was normalized to body weight, and relative time-activity curves (TACs) were generated to assess tracer kinetics. Whole-body PET maximum intensity projection (MIP) images were used to visualize tracer distribution. The analysis was conducted across three distinct phases: the rapid distribution phase, equilibrium phase, and clearance phase. Due to the exploratory nature of the study, imaging time points varied based on patient tolerance and equipment scheduling. Data comparability was ensured through normalization and linear interpolation. Specific imaging time points were as follows: Patient 1 was scanned at 30, 60, and 90 minutes post-injection; Patient 2 at 30, 60, 90, and 120 minutes; and Patient 3 at 15, 60, and 180 minutes. Data from Patient 3 defined the rapid distribution phase at 15 minutes and Patients 1 and 2 at 30 minutes, characterizing initial tracer uptake. The equilibrium phase included data from all patients at 60 minutes and Patients 1 and 2 at 90 minutes, reflecting stable tracer distribution. The clearance phase was indicated by data from Patient 2 at 120 minutes and Patient 3 at 180 minutes, showing tracer elimination.

Organ segmentation was performed using the nnU-Net framework (dataset: TotalSegmentator total 3 mm, 1559 subjects, version 2.04) [25], focusing on the CT modality for automated delineation of organs. The resulting segmentation masks were mapped onto corresponding PET images to derive SUV for the kidneys, bladder wall, uterus, pancreas, spleen, liver, gallbladder wall, heart wall, and red bone marrow (defined by volume-of-interest [VOIs] encompassing the L2–L4 vertebral bodies). This process was executed by a nuclear medicine physicist.

SUVmax values for each VOI at multiple time points were extracted from PET images and fitted using a custom single-exponential decay function implemented in Python (version 3.11). Time-integrated activity coefficients (TIACs) were subsequently calculated through analytical integration of the fitted TACs.

Absorbed dose coefficients (mGy/MBq) and effective dose coefficients (mSv/MBq) were calculated using MIRDcalc software with appropriate S-values for the adult female reference model [26, 27]. Effective doses were determined according to the International Commission on Radiological Protection (ICRP) guidelines for adult females.

Statistical analysis

Statistical analyses were performed using GraphPad Prism software (version 10.1.2). Quantitative data are expressed as mean ± standard deviation (SD). Differences in SUV between gender and age groups were compared using the independent samples t-test, with Welch’s correction applied for unequal variances. The Mann-Whitney U test was used to evaluate the effect of fasting status on gallbladder SUV. p < 0.05 was considered statistically significant. Descriptive analyses were conducted for dosimetry data.

Results

Patients

This study enrolled 51 patients (30 females, 21 males; median age: 56 years, range: 22–77 years), with baseline characteristics summarized in Table 1. All patients underwent ¹⁸F-CXCR4 PET/CT imaging for accurate staging and treatment response assessment.

Table 1.

Demographic and clinical characteristics of patients

Description of patients (n = 51)	No.
Age (years)
Median	56
Gender
Female	30
Male	21
Tumor types
Multiple myeloma	5
Lymphoma	39
Renal carcinoma	1
Liver cancer	2
Prostate tumor	1
Adrenal tumor	1
Aldosterone-producing adenoma	2
Imaging purposes
Initial diagnosis	38
Efficacy evaluation	13

¹⁸F-CXCR4 PET/CT scanning

Each patient received an intravenous injection of 7.1 ± 1.8 mCi (263 ± 67 MBq) of ^{1 8} F-CXCR4, with a radiochemical purity > 99% and specific activity > 20 GBq/μmol. All ^{1 8} F-CXCR4 PET/CT scans were completed, yielding high-quality diagnostic images. No notable abnormalities were reported by patients within 30 minutes post-examination, indicating excellent tolerability. Among the 51 enrolled patients, 38 underwent ^{1 8} F-CXCR4 PET/CT staging to inform treatment planning. Of the 13 patients evaluated for treatment response, 11 achieved remission, whereas 2 showed disease progression on ^{1 8} F-CXCR4 PET/CT. All patients benefited from clinically useful information provided by ¹⁸F-CXCR4 PET/CT, which enabled individualized diagnostic and therapeutic strategies.

Physiological distribution of¹⁸F-CXCR4

The variability of ¹⁸F-CXCR4 tracer distribution in major organs is summarized in Table 2 and Fig. 1. Physiological distribution of ¹⁸F-CXCR4 was observed in the urinary tract, renal cortex, biliary system, uterus, Waldeyer’s ring, and spleen (as shown in Fig. 2). The highest tracer uptake levels (SUVmax ≥ 4.6) were observed in the urinary tract, biliary system, and uterus. Moderate to high ^{1 8} F-CXCR4 tracer uptake was observed in the myocardium, lungs, pancreas, parotid glands, submandibular glands, thyroid, colon, esophagus, small intestine, bone marrow, liver, and adrenal glands, with organ-specific SUVmax values ranging from 1.30 to 3.16. Tracer uptake was minimal to absent in the brain, spinal cord, and skeletal muscle.

Table 2.

Physiological biodistribution of ¹⁸F-CXCR4

Organ	SUVmax ± SD	SUVmean ± SD
Brain	0.18±0.06	0.10±0.03
Waldeyer’s ring	4.55±1.28	2.64±0.76
Parotid gland	1.42±0.37	1.21±0.33
Submandibular glands	1.65±0.46	1.42±0.43
Thyroid	1.78±0.52	1.59±0.47
Lungs(right upper lobe)	1.35±0.29	0.91±0.21
Spinal cord	0.78±0.24	0.55±0.21
Myocardium	1.30±0.41	1.12±0.32
Esophagus	2.13±0.45	1.78±0.37
Blood pool	1.88±0.27	1.64±0.23
Liver	2.19±0.35	1.40±0.25
Biliary system	11.57±8.34	7.32±5.71
Pancreas	1.42±0.29	1.09±0.23
Spleen	3.90±0.87	3.09±0.70
Bone marrow(L2–L4)	2.39±0.61	1.77±0.48
Kidney cortex	2.56±0.45	2.19±0.40
Adrenal gland	3.16±1.00	2.47±0.64
Colon	1.90±0.51	1.15±0.32
Muscles	0.84±0.20	0.58±0.14
Intestine	2.15±0.55	1.37±0.38
Uterus	4.57±1.03	2.99±0.72
Breasts	1.14±0.50	0.82±0.41
Ovaries	2.39±0.63	1.95±0.54
Prostate gland	1.92±0.65	1.41±0.42
Seminal vesicle	2.06±0.49	1.62±0.38
Testes	1.55±0.38	1.22±0.26

Fig. 1.

The distribution of ¹⁸F-CXCR4 tracer uptake varied across different organs. The highest tracer uptake was observed in the urinary tract, biliary system, and uterus. Additionally, varying levels of tracer uptake were noted in the myocardium, lungs, pancreas, parotid glands, submandibular glands, thyroid, colon, esophagus, small intestine, bone marrow, liver, and adrenal glands

Fig. 2.

Typical physiological biodistribution of ¹⁸ F-CXCR4 in male (A) and female (B). The urinary tract and biliary system show intense radiotracer uptake

In analyzing SUVmax and SUVmean differences between males and females, no significant overall differences were found (p > 0.05), and the organ-specific variations are summarized in Table 3. However, some variations in ¹⁸F-CXCR4 tracer distribution were noted. The SUVmean for the male pancreas was higher (1.54 ± 0.29) than for females (1.34 ± 0.26, p = 0.01). In contrast, females had a slightly higher lung parenchyma SUVmean (0.86 ± 0.21) than males (0.83 ± 0.21, p = 0.04). Additionally, both SUVmax (1.94 ± 0.26 for females vs. 1.79 ± 0.27 for males, p = 0.05) and SUVmean (1.70 ± 0.23 for females vs. 1.55 ± 0.21 for males, p = 0.02) for the blood pool were higher in females.

Table 3.

Effect of sex on physiological ¹⁸ F-CXCR4 activity

Organ	SUVmax ± SD		t(t’)	P values	SUVmean ± SD		t(t’)	P values
	Male	Female			Male	Female
	(n = 21)	(n = 30)			(n = 21)	(n = 30)
Brain	0.19±0.06	0.18±0.05	0.9	0.37	0.11±0.03	0.09±0.03	1.23	0.23
Waldeyer’s ring	4.23±1.09	4.74±1.36	−1.25	0.22	2.41±0.65	2.78±0.80	−1.50	0.14
Parotid gland	1.39±0.30	1.43±0.41	−0.39	0.70	1.16±0.23	1.24±0.39	−0.90	0.37
Submandibular glands	1.66±0.60	1.64±0.36	0.18	0.86	1.43±0.55	1.42±0.35	0.05	0.96
Thyroid	1.79±0.41	1.77±0.59	0.12	0.91	1.57±0.36	1.60±0.54	−0.18	0.86
Lungs(right upper lobe)	1.28±0.28	1.40±0.28	−1.49	0.14	0.83±0.21	0.86±0.21	−2.11	0.04
Spinal cord	0.79±0.24	0.76±0.25	0.42	0.68	0.55±0.22	0.55±0.21	0.08	0.94
Myocardium	1.32±0.52	1.29±0.32	0.26	0.80	1.11±0.37	1.12±0.29	−0.17	0.87
Esophagus	2.04±0.48	2.20±0.42	−1.27	0.21	1.72±0.39	1.82±0.35	−1.01	0.32
Blood pool	1.79±0.27	1.94±0.26	−2.03	0.05	1.55±0.21	1.70±0.23	−2.52	0.02
Liver	2.28±0.40	2.12±0.30	1.64	0.11	1.43±0.29	1.38±0.22	0.66	0.52
Biliary system	12.64±8.43	10.82±8.34	0.76	0.45	7.66±5.36	7.27±6.04	0.24	0.81
Pancreas	1.54±0.29	1.34±0.26	2.56	0.01	1.14±0.22	1.05±0.23	1.46	0.15
Spleen	3.79±0.74	3.98±0.95	−0.74	0.47	2.97±0.62	3.18±0.74	−1.03	0.31
Bone marrow	2.44±0.58	2.36±0.64	0.42	0.68	1.83±0.41	1.73±0.52	0.63	0.53
Kidney cortex	2.50±0.47	2.60±0.44	−0.8	0.43	2.10±0.39	2.25±0.40	−1.33	0.19
Adrenal gland	3.33±0.99	3.03±1.00	1.04	0.31	2.50±0.68	2.45±0.63	0.31	0.76
Colon	2.06±0.55	1.79±0.46	1.87	0.07	1.24±0.34	1.08±0.28	1.78	0.08
Muscles	0.83±0.21	0.84±0.20	−0.19	0.85	0.54±0.11	0.61±0.15	−1.86	0.07
Intestine	2.13±0.47	2.17±0.60	−0.25	0.80	1.36±0.31	1.38±0.40	−0.16	0.88

Note. The P-value in bold in the table indicates that the difference is statistically significant (p≤0.05)

The median age of participants was 56 years, and patients were accordingly divided into an older group (≥56 years) and a younger group ( < 56 years). Overall, no significant age-related differences in ^{1 8} F-CXCR4 tracer uptake were observed across most organs, as summarized in Table 4. Nevertheless, several minor variations were noted. The blood pool SUVmax was slightly higher in the older group than in the younger group (1.91 ± 0.25 vs. 1.86 ± 0.29, p = 0.05). In addition, the younger group demonstrated higher biliary SUVmax (14.02 ± 10.56 vs. 9.22 ± 4.50, p = 0.04), whereas the older group showed marginally higher pancreatic SUVmean (1.15 ± 0.22 vs. 1.02 ± 0.22, p = 0.03).

Table 4.

Effect of age on physiological ¹⁸ F-CXCR4 activity

Organ	SUVmax ± SD		t(t’)	P values	SUVmean ± SD		t(t’)	P values
	＜56 years	≥56 years			＜56 years	≥56 years
	(n = 25)	(n = 26)			(n = 25)	(n = 26)
Brain	0.19±0.05	0.18±0.06	0.56	0.58	0.11±0.03	0.10±0.04	0.58	0.56
Waldeyer’s ring	4.53±1.11	4.56±1.41	−0.08	0.94	2.65±0.72	2.63±0.80	0.06	0.96
Parotid gland	1.37±0.36	1.47±0.37	−0.9	0.37	1.17±0.32	1.25±0.35	−0.82	0.42
Submandibular glands	1.56±0.43	1.73±0.49	−1.27	0.21	1.32±0.39	1.53±0.46	−1.73	0.09
Thyroid	1.70±0.63	1.33±0.26	1.06	0.30	1.52±0.55	1.66±0.37	−1.06	0.30
Lungs(right upper lobe)	1.36±0.31	1.40±0.28	−0.38	0.70	0.91±0.24	0.91±0.19	0.02	0.99
Spinal cord	0.78±0.26	0.77±0.23	0.06	0.96	0.55±0.25	0.54±0.18	0.14	0.89
Myocardium	1.34±0.51	1.26±0.28	0.74	0.47	1.15±0.40	1.08±0.23	0.73	0.47
Esophagus	2.05±0.47	2.21±0.42	−1.24	0.22	1.70±0.37	1.86±0.35	−1.6	0.12
Blood pool	1.86±0.29	1.91±0.25	−0.7	0.05	1.62±0.27	1.66±0.20	−0.68	0.5
Liver	2.16±0.33	2.21±0.37	−0.49	0.63	1.37±0.22	1.42±0.27	−0.76	0.45
Biliary system	14.02±10.56	9.22±4.50	2.09	0.04	8.83±7.26	6.07±3.29	1.74	0.09
Pancreas	1.36±0.31	1.48±0.27	−1.52	0.13	1.02±0.22	1.15±0.22	−2.2	0.03
Spleen	3.92±0.86	3.89±0.89	0.08	0.94	3.15±0.67	3.05±0.73	0.462	0.65
Bone marrow	2.54±0.58	2.25±0.61	1.53	0.13	1.85±0.44	1.70±0.50	1.01	0.32
Kidney cortex	2.55±0.47	2.56±0.45	−0.1	0.92	2.12±0.28	2.26±0.41	−1.22	0.23
Adrenal gland	3.15±1.04	3.17±0.98	−0.06	0.95	2.47±0.68	2.47±0.62	−0.04	0.97
Colon	1.92±0.55	1.88±0.48	0.27	0.79	1.15±0.33	1.14±0.31	0.08	0.93
Muscles	0.79±0.20	0.88±0.20	−1.47	0.15	0.54±0.13	0.62±0.14	−1.85	0.07
Intestine	2.04±0.59	2.26±0.49	−1.42	0.16	1.30±0.43	1.44±0.32	−1.27	0.21

Note.The P-value in bold in the table indicates that the difference is statistically significant (p≤0.05)

Effect of fasting on CXCR4 PET imaging

During the study, researchers discovered a difference in the tracer distribution of the ¹⁸F-CXCR4 imaging agent in the gallbladder. The 13 patients were categorized into two groups based on their dietary habits: the fed group, shown in Fig. 3, and the fasted group, illustrated in Fig. 4. The findings indicate a trend of higher gallbladder tracer distribution of the imaging agent in the fasted group compared to the fed group, suggesting that the fasting state may lead to increased tracer accumulation in the gallbladder. Although this difference in mean SUV did not reach statistical significance (p > 0.05), further details are presented in Table 5. The primary foods consumed by the fed group included pork, rice, eggs, and milk. Additionally, it was observed that gallbladder tracer uptake was uneven, as depicted in Fig. 5.

Fig. 3.

Patients in the fed group had lower tracer uptake in their gallbladder (arrows) on ¹⁸F-CXCR4 PET/CT imaging

Fig. 4.

The gallbladder tracer uptake was significantly increased in the fasted group (arrows) on ¹⁸F-CXCR4 PET/CT imaging

Table 5.

Effect of feeding status on gallbladder ¹⁸F-CXCR4 activity

	Fasted group (n = 5)	Fed group (n = 8)	P values
SUVmax ± SD	13.97 ±8.52	7.16±4.07	0.06
SUVmean ± SD	8.52±4.67	4.37±2.54	0.08

Fig. 5.

Heterogeneous gallbladder radiotracer uptake was observed on ¹⁸ F-CXCR4 PET/CT imaging (arrows)

Tracer distribution in benign lesions

Benign lesions showed varying degrees of ¹⁸F-CXCR4 tracer uptake (SUVmax 1.5–6.6; see Table 6). Among 51 patients, 3 had osteophytes and inflammation (Fig. 6), 1 femoral head necrosis (Fig. 7), 4 rib fractures (Fig. 8; 1 old trauma, 1 pathological, 1 recent trauma, 1 unclear history), 2 thymic tracer distribution (Fig. 9), 2 arthritis, 16 lung inflammation, 8 sinusitis, 6 tonsillitis, 6 renal cysts, and 6 hepatic cysts. Nine patients presented with lymph nodes exhibiting increased ^{1 8} F-CXCR4 tracer uptake. These nodes were interpreted as benign based on an integrated assessment. Three showed interval regression in size and tracer activity on 6–10-month follow-up imaging, findings consistent with reactive changes. The remaining nodes were located in regions not typically involved by the primary malignancy and demonstrated low-level tracer activity with non-suspicious CT morphology, in concordance with the patients’ clinical histories.

Table 6.

Patterns of ¹⁸F-CXCR4 uptake in benign lesions

Localization	Number of lesions	Number of patients	SUVmax (± SD)	SUVmean (± SD)
Arthritis	2	2	3.49±0.83	1.69±0.39
Fracture	5	4	3.41±2.21	2.14±1.27
Thymus	2	2	5.65±3.30	3.29±3.30
sinusitis	8	8	2.03±1.13	1.31±0.75
tonsillitis	6	6	6.22±0.94	3.71±0.55
renal cysts	9	6	1.31±0.49	1.23±0.51
hepatic cysts	8	6	1.25±0.33	0.99±0.28
lung inflammation	-	16	3.87±1.51	2.16±0.98
Femoral head necrosis	1	1	4.12	2.1
Inflammatory osteophyte	5	3	4.94±1.79	2.93±0.83
Inflammatory lymph nodes	30	9	4.38±1.21	2.73±0.74

Fig. 6.

The right part of the L4 vertebral body was locally elevated with slight swelling of the surrounding soft tissues, and the tracer uptake of ¹⁸F-CXCR4 was increased (arrows)

Fig. 7.

Necrosis of the left femoral head and slightly increased tracer uptake (arrows) on ¹⁸F-CXCR4 PET/CT imaging

Fig. 8.

Focal radiotracer uptake was observed in the right third rib corresponding to a prior fracture (arrows) on ¹⁸ F-CXCR4 PET/CT imaging

Fig. 9.

Enhanced strip tracer uptake is observed in the thymus region (arrows) on ¹⁸F-CXCR4 PET/CT imaging

Tumor uptake and target-to-background contrast

Among baseline patients, most individuals demonstrated measurable ¹⁸F-CXCR4–avid lesions, with lesion activity clearly exceeding blood-pool levels, resulting in strong tumor-to-background contrast. Similar patterns were observed in the post-treatment cohort, in which detectable lesions maintained distinct radiotracer activity relative to background despite variable activity levels. Overall, malignant lesions exhibited tracer activity substantially above physiologic background, supporting the use of tumor-to-blood pool ratios as an informative semi-quantitative metric for assessing CXCR4-avid disease (Table 7). A subset of patients did not demonstrate abnormal radiotracer distribution on imaging. Among baseline patients, 11 had a history of biopsy or tumor resection, and no residual lesions were identified on ¹⁸ F-CXCR4 PET/CT at the time of imaging. In addition, 3 post-treatment patients with MALT lymphoma showed no abnormal radiotracer distribution on follow-up PET/CT examinations.

Table 7.

Tumor SUV metrics and tumor-to-blood pool ratios (TBR) in baseline and post-treatment patients

Metrics	Baseline (n = 38)	Post-treatment (n = 13)
Patients with detectable lesions, n (%)	27 (71.1%)	10 (76.9%)
SUVmax range of tumor	2.14–24.23	1.37–14.89
SUVmax of tumor*	7.83	7.24
Blood-pool SUVmax*	1.86	1.88
TBR	4.21	3.85

Note.Values marked with * indicate mean values

Radiation dosimetry

The biodistribution of ¹⁸F-CXCR4 was evaluated in major organs of three female patients, with a mean injected activity of 6.7 mCi (247.9 MBq). Whole-body MIP images from a representative patient (Patient 2), shown in Fig. 10, demonstrate significant radiotracer distribution in the kidneys and urinary bladder, with diminishing activity in non-target tissues over time. TIACs for Patient 2, illustrated in Fig. 11, quantify the radiotracer’s temporal behavior in key organs, forming the basis for dosimetric calculations. Organ-specific absorbed dose coefficients, detailed in Table 8, include contributions from positrons and gamma radiation.

Fig. 10.

MIP images of whole-body ¹⁸F-CXCR4 PET/CT scans at 30, 60, 90, and 120 minutes post-injection for Patient 2, illustrating the temporal distribution of the radiotracer

Fig. 11.

Time-integrated activity coefficients (TIACs) derived from fitted time-activity curves (TACs) for selected organs in Patient 2, illustrating the temporal integration of radiotracer activity

Table 8.

Absorbed organ dose coefficients and absorbed organ doses

Target Organ	β (mGy/MBq)	γ	Absorb Dose
		(mGy/MBq)	(mGy/MBq)
Adrenals	5.05E-02	4.61E-02	5.11E-02
Bone marrow （red marrow）	2.83E-03	5.86E-03	5.89E-03
Brain	2.00E-03	2.31E-03	4.31E-03
Breasts	1.98E-03	3.54E-03	5.52E-03
Colon	1.73E-03	1.78E-02	1.97E-02
Gallbladder wall	3.00E-03	2.77E-02	3.07E-02
Heart wall	2.01E-03	6.31E-03	8.31E-03
Kidneys	3.11E-01	1.00E-01	4.12E-01
Liver	2.16E-03	1.59E-02	1.80E-02
Lungs	2.03E-03	5.37E-03	7.40E-03
Muscle	2.02E-03	6.96E-03	8.97E-03
Ovaries	2.12E-03	2.70E-02	2.92E-02
Pancreas	2.47E-03	3.29E-02	3.53E-02
Small intestine	2.28E-02	2.51E-02	2.68E-02
Spleen	2.06E-03	2.07E-02	2.28E-02
Stomach	1.41E-03	1.71E-02	1.85E-02
Thymus	2.01E-03	3.97E-03	5.98E-03
Thyroid	1.99E-03	3.26E-03	5.25E-03
Ureters	3.06E-03	2.83E-02	3.14E-02
Urinary bladder wall	2.62E-02	9.20E-02	1.18E-01
Uterus	2.37E-03	4.79E-02	5.03E-02
Whole body target	-	-	1.20E-02

Effective dose coefficient (mSv/MBq): 2.02E-02

The kidneys exhibited the highest absorbed dose (0.412 mGy/MBq), driven primarily by positron emissions (0.311 mGy/MBq), followed by the urinary bladder wall (0.118 mGy/MBq). The whole-body absorbed dose coefficient was 1.20E-02 mGy/MBq, and the effective dose coefficient was 2.02E-02 mSv/MBq, supporting the suitability of ¹⁸F-CXCR4 for diagnostic nuclear medicine imaging.

Discussion

This study quantifies the physiological biodistribution of CXCR4 in humans, establishing ¹⁸F-CXCR4 tracer uptake thresholds for benign lesions to enhance diagnostic accuracy in PET/CT imaging. This work provides a molecular imaging foundation for precise target selection and the development of integrated theranostic strategies for CXCR4-related diseases.

Physiological biodistribution of ¹⁸F-CXCR4 showed prominent tracer uptake above blood pool levels in the biliary system, uterus, Waldeyer’s ring, spleen, adrenal gland, kidney cortex, bone marrow, ovaries, liver, intestine, esophagus, seminal vesicles, colon, and prostate. Tracer uptake near or below the blood pool was observed in the thyroid, submandibular glands, pancreas, parotid gland, lungs, myocardium, breasts, and testes, with no tracer uptake in the brain, spinal cord, or muscles. Gender significantly influenced tracer uptake only in the pancreas (SUVmean higher in males) and lungs/blood pool (SUVmean and SUVmax higher in females). Age-related differences were minimal, with slightly higher SUVmax in the blood pool and SUVmean in the pancreas in older individuals, and slightly higher biliary system SUVmax in younger individuals; other organ tracer yptake showed no significant age dependence.

This physiological distribution pattern reflects both organ-specific CXCR4 expression and probe pharmacokinetics. High tracer uptake in immune organs like the spleen (SUVmax: 3.90 ± 0.87) and Waldeyer’s ring (SUVmax: 4.55 ± 1.28) correlated positively with circulating T/B lymphocyte CXCR4 levels [28]. This aligns mechanistically with the CXCL12–CXCR4 axis mediating immune cell chemotaxis to lymphoid tissues, consistent with observations in animal models [12]. A key anatomical finding was the use of the L2–L4 vertebral bodies for bone marrow assessment, as previously recommended [29], due to uniform marrow distribution, clear anatomical landmarks, and consistent morphology, minimizing partial volume effects for broad clinical utility.

Excretion pathways significantly contributed to tracer distribution: urinary tract and biliary system accumulation was primarily tracer-excretion driven. In contrast, the tracer uptake observed in the renal cortex reflected active, CXCR4receptor-specific binding of the radioligand [30]. Addressing a specific gap in prior literature [29], we found significantly higher ¹⁸F-CXCR4 gallbladder tracer uptake in fasting subjects versus fed (SUVmax 13.97 ± 8.52 vs. 7.16 ± 4.07). Because gallbladder activity primarily reflects biliary excretion rather than CXCR4-specific radiotracer uptake, excessive gallbladder radiotracer retention under fasting conditions may obscure evaluation of the biliary region and reduce interpretability. These findings indicate that performing ¹⁸F-CXCR4 PET/CT after a standard meal is advisable to minimize gallbladder radiotracer accumulation and to improve consistency and clarity in hepatobiliary imaging.

Beyond excretion, uterine tracer uptake (SUVmax: 4.57 ± 1.03) was prominent. CXCR4 expression here is dynamically regulated by pregnancy and microenvironmental factors (hypoxia, hormones), functioning in endometrial remodeling and immune recruitment [31, 32]. While endometrial CXCR4 is HIF1A-regulated and upregulated during menstruation (potentially aiding repair), tracer uptake was stable in our predominantly postmenopausal cohort, possibly reflecting hormone decline and atrophy. Conversely, baseline breast tracer uptake was low (SUVmax: 1.14 ± 0.50), but hormonal stimulation could modulate tracer uptake via CXCL12/CXCR4 axis activation or immune microenvironment changes, warranting further investigation into specific hormone mechanisms (e.g., estrogen, gonadotropins) and clinical relevance.

Glandular tracer uptake varied: submandibular, thyroid, and adrenal glands showed moderate tracer distribution (SUVmax: 1.14–3.16), with adrenal fluctuations (SUVmax: 3.16 ± 1.00) potentially influenced by hormonal changes, as suggested by dexamethasone-induced suppression in murine models [33]. Testes, prostate, seminal vesicles, esophagus, and liver exhibited only mild tracer uptake (SUVmax: 1.55–2.19), consistent with lower tissue CXCR4 expression. Critically, negligible tracer uptake in the brain (SUVmax: 0.18 ± 0.06), spinal cord (SUVmax: 0.78 ± 0.24), and skeletal muscle (SUVmax: 0.84 ± 0.20), coupled with low tracer uptake in breasts, myocardium, pancreas, parotid, and lung parenchyma (SUVmax ≤ 1.5), provides a high target-to-background ratio. This establishes ¹⁸F-CXCR4 PET/CT as particularly suitable for detecting CXCR4-high diseases in these regions (e.g., gliomas, lymphomas) [4, 34, 35].

Importantly, our study provides novel insights into the benign lesion tracer distribution of ¹⁸F-CXCR4. The tracer uptake was observed in contexts such as rib fractures, arthritis, osteophyte-induced inflammation, sinusitis, tonsillitis, and lung inflammation, underscoring CXCR4‘s role in wound healing and lymphocyte recruitment. Mechanistically, tracer uptake at fracture sites likely results from SDF-1 release post-injury, which recruits CXCR4-positive monocytes/macrophages and upregulates CXCR4 on mesenchymal stem cells, creating a feedback loop for tracer accumulation [36]. Notably, tracer distribution in rib fractures varies by healing stage, with higher tracer uptake in acute and subacute phases due to active SDF-1-mediated cell recruitment and lower tracer distribution in chronic phases as healing advances and cellular activity declines. In inflammatory processes, including sinusitis, tonsillitis, lung inflammation, and arthritis, immune and stromal cells upregulate CXCR4 via NF-κB pathways, facilitating cell recruitment and tissue remodeling [37]. Combined with patient history, thymic tracer activity may reflect post-treatment regeneration or immune stress responses. Interestingly, cystic lesions, such as renal cysts and hepatic cysts, exhibited mild ¹⁸F-CXCR4 tracer uptake, potentially reflecting low-level cellular activity or microenvironmental factors not fully attributable to inflammation.

We supplemented prior literature by formally evaluating gender and age effects on ¹⁸F-CXCR4 tracer distribution. Benet et al. [38] reported that CXCR4 expression in bone marrow plasma cells increases with age, suggesting age-related effects on ¹⁸F-CXCR4 biodistribution. A 56-year age cut-off was selected to ensure balanced group sizes. While most organs showed no significant differences (Tables 3 and 4, p > 0.05), a higher lung-to-blood pool tracer uptake was observed in females across the overall study cohort. This sex-related difference could potentially be influenced by hormonal regulation of the CXCL12–CXCR4 axis, such as estrogen-mediated modulation of CXCL12 secretion; however, this interpretation remains speculative and requires further investigation. Reduced biliary tracer uptake with age might reflect declining metabolism or bile composition changes, while increased pancreatic tracer uptake in older adults could be associated with subclinical inflammation or fibrosis. These findings underscore the need for age- and gender-adjusted tracer uptake thresholds to ensure accurate image interpretation.

The dosimetry analysis of ¹⁸F-CXCR4 reveals a radiation exposure profile suitable for diagnostic nuclear medicine, with an effective dose coefficient of 2.02E-02 mSv/MBq, comparable to fluorine-18 radiotracers such as ¹⁸F-FDG (2.00E-02 mSv/MBq) [39]. The kidneys exhibited the highest absorbed dose (0.412 mGy/MBq), predominantly from beta radiation (0.311 mGy/MBq), followed by the urinary bladder wall (0.118 mGy/MBq), as detailed in Table 7. This pronounced renal tracer accumulation is likely driven by the radiotracer’s affinity for CXCR4 receptors expressed in renal tissue and its excretion through the urinary system, as supported by the TIACs for Patient 2 (Fig. 11). However, additional factors may contribute to the elevated renal dose. The use of automated organ delineation in dosimetric calculations could introduce errors, potentially overestimating tracer activity in the kidneys due to inaccuracies in contouring complex renal anatomy. Furthermore, the study cohort consisted of postmenopausal females, who may exhibit reduced renal function, potentially prolonging radiotracer retention and increasing absorbed doses [40].

This study observed relatively elevated physiological ¹⁸F-CXCR4 tracer uptake in the uterus. Because the dosimetric calculations were derived from three postmenopausal females, this physiological uptake could theoretically introduce some uncertainty into the estimation of uterus-related organ absorbed doses. However, the uterus carries a very low weighting in standard ICRP effective dose models, and therefore, this physiological uptake is not expected to have a noticeable impact on the overall effective dose assessment.

This preliminary study evaluated the dosimetric properties of ^{1 8} F-CXCR4, confirming its safety for PET imaging and establishing a foundation for the development of CXCR4-targeted therapeutic radioligands, such as those based on ^{1 7 7} Lu. To improve the accuracy of dosimetric estimates, future studies should prioritize manual organ delineation, particularly for critical organs such as the kidneys, to enhance precision in radiation dose calculations. Additionally, optimizing radiotracer pharmacokinetics through chemical modifications or co-administration of protective agents, such as amino acid infusions, could minimize renal retention and reduce potential nephrotoxicity risks, especially in patients with compromised renal function. These advancements in dosimetry methodology will enhance the safety and efficacy of ^{1 8} F-CXCR4 for diagnostic imaging.

The study is limited by its retrospective design and small sample size, which may introduce selection bias and limit the generalizability of findings. Future prospective, multicenter studies with larger and more diverse cohorts are essential to validate these results and explore whether age- or sex-related physiological differences, such as variations in renal clearance, influence radiotracer distribution.

Conclusion

In conclusion, this investigation exhibits a distinctive in vivo biodistribution pattern for ¹⁸ F-CXCR4, with modest influence from gender or age, defining diagnostic reference standards, with elevated gallbladder radiotracer accumulation in fasting patients, underscoring the importance of a standardized imaging protocol. Evaluation of CXCR4 distribution in non-target organs and select benign conditions lays the groundwork for advancing basic research and developing integrated diagnostic-therapeutic strategies for CXCR4-related diseases, while quantitative evaluation of the imaging agent’s radiation dosimetry supports its clinical safety.

Abbreviations

ICRP

International commission on radiological protection

CXCR4

C-X-C chemokine receptor type 4

CXCL12

C-X-C motif chemokine ligand 12

SDF-1

Stromal cell-derived factor 1

CLL

Chronic lymphocytic leukemia

WM

Waldenström’s macroglobulinemia

PET

Positron emission tomography

^{1 8} F-CXCR4

Al¹⁸F-NOTA-Pentixafor

¹⁸F-FDG

¹⁸F-fluorodeoxyglucose

CT

Computed tomography

ROI

Regions of interest

TACs

Time-activity curves

TIACs

Time-integrated activity coefficients

MIP

Maximum intensity projection

VOIs

Volume-of-interest

SUV

Standardized uptake value

SD

Standard deviation

SUV

Standardized tracer distribution values

Author contributions

Xinyang Li, Yu He, Jingkai Yi, Dan Wang, Kailin Qi, Yutang Yao, Hao Lu, Shirong Chen and Meng Zhao contributed to the conception and design of the study, conceived the manuscript, performed data analysis and interpretation. Xinyang Li carried out the manuscript preparation. Xiao Jiang, Ying Kou, Yingchun Li and Ping Wu contributed to the conception and design of the study, conceived the manuscript, performed data analysis and interpretation, and revised the manuscript. Zhen Cao was responsible for the calculation of radiation dosimetry and related data collection. Zhuzhong Cheng, as the corresponding author, contributed to the conception and design of the study, conceived the manuscript, performed data analysis and interpretation, revised the manuscript, and oversaw the overall coordination and correspondence for the study.

Funding

Nuclear Technology R&D Project under the 14th Five-Year Plan of the State Administration of Science, Technology and Industry for National Defense (HNK202323(36)02), Isotope and Drug Innovation Fund Program (TWSCX-2023-CXJJ-9), Sichuan Provincial Clinical Key Specialty Construction Project (2024GHYXP002), Sichuan Cancer Hospital Outstanding Youth Funding (YB 2023022), Noncommunicable Chronic Diseases-National Science and Technology Major Project (2024ZD0525900), Precision Treatment Regimens and Translational Applications for Lymphoma (2024YFFK0067), and Sichuan Medical and Health Promotion Association Research Project (KY2022SJ0260).

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Sichuan Cancer Hospital (approval number: SCCSMC-01–202-87) and conducted in compliance with Chinese regulations. The study was registered in the Chinese Clinical Trial Registry (ChiCTR2400090595) on October 9, 2024. All participants provided written informed consent prior to enrollment. The research was conducted in accordance with the principles of the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.