CXCR4-Directed Imaging in Solid Tumors

Despite histological evidence in various solid tumor entities, available experience with CXCR4-directed diagnostics and endoradiotherapy mainly focuses on hematologic diseases. With the goal of expanding the application of CXCR4 theranostics to solid tumors, we aimed to elucidate the feasibility of CXCR4-targeted imaging in a variety of such neoplasms. Methods: Nineteen patients with newly diagnosed, treatment-naïve solid tumors including pancreatic adenocarcinoma or neuroendocrine tumor, cholangiocarcinoma, hepatocellular carcinoma, renal cell carcinoma, ovarian cancer, and prostate cancer underwent [68Ga]Pentixafor PET/CT. CXCR4-mediated uptake was assessed both visually and semi-quantitatively by evaluation of maximum standardized uptake values (SUVmax) of both primary tumors and metastases. With physiologic liver uptake as reference, tumor-to-background ratios (TBR) were calculated. [68Ga]Pentixafor findings were further compared to immunohistochemistry and [18F]FDG PET/CT. Results: On [68Ga]Pentixafor PET/CT, 10/19 (52.6%) primary tumors were visually detectable with a median SUVmax of 5.4 (range, 1.7–16.0) and a median TBR of 2.6 (range, 0.8–7.4), respectively. The highest level of radiotracer uptake was identified in a patient with cholangiocarcinoma (SUVmax, 16.0; TBR, 7.4). The relatively low uptake on [68Ga]Pentixafor was also noted in metastases, exhibiting a median SUVmax of 4.5 (range, 2.3–8.8; TBR, 1.7; range, 1.0–4.1). A good correlation between uptake on [68Ga]Pentixafor and histological derived CXCR4 expression was noted (R = 0.62, P < 0.05). In the 3 patients in whom [18F]FDG PET/CT was available, [68Ga]Pentixafor exhibited lower uptake in all lesions. Conclusions: In this cohort of newly diagnosed, treatment-naïve patients with solid malignancies, CXCR4 expression as detected by [68Ga]Pentixafor-PET/CT and immunohistochemistry was rather moderate. Thus, CXCR4-directed imaging may not play a major role in the management of solid tumors in the majority of patients.


INTRODUCTION
C-X-C motif chemokine receptor 4 (CXCR4) is overexpressed in more than 20 tumor types and plays a crucial role in tumor growth, tumor invasiveness, cancer cell-microenvironment interaction, and metastasis (1,2). Notably, the presence of CXCR4 has been linked to unfavorable outcomes in multiple different tumor entities, including hematologic malignancies, breast cancer, renal cell carcinoma, gynecologic malignancies, pancreatic adenocarcinoma, and hepatocellular carcinoma (3).
In solid malignancies, pilot studies have hinted at a role for CXCR4-directed imaging in various selected diseases, such as small cell lung cancer, esophageal adenocarcinoma, and poorly differentiated neuroendocrine neoplasms (21)(22)(23). Of note, Blümel et al. reported on the use of [ 68 Ga]Pentixafor in patients diagnosed with adrenocortical cancer, with 70% of the subjects being potentially suitable for a treatment with [ 177 Lu]/[ 90 Y]Pentixather (24).
On the other hand, Vag et al. could not detect relevant [ 68 Ga]Pentixafor uptake in a heterogenous subset of different solid cancers, including non-small cell lung cancer, malignant melanoma, sarcoma, cancer of unknown primary (CUP), or breast cancer (25). Our group reported on rather discouraging results in a small cohort of patients with malignant pleural mesothelioma (26).
Thus, given these contradictory findings among different solid tumor entities, we aimed to broaden the experience of CXCR4-targeted PET imaging in solid cancers by investigating a subset of different tumors, including cholangiocarcinoma (CCC), hepatocellular carcinoma (HCC), pancreatic adenocarcinoma or neuroendocrine tumor, and ovarian cancer.

MATERIALS AND METHODS
[ 68 Ga]Pentixafor was administered on a compassionate use basis in compliance with §37 of the Declaration of Helsinki and The German Medicinal Products Act, AMG §13.2b. All patients underwent imaging for clinical purposes and gave written and informed consent to the diagnostic procedures. The local institutional review board waived the requirement for additional approval because of the retrospective character of this study.  (8/17, 47.1%) or surgery (n = 9/17, 52.9%) after a median of 4.5 days (range, 1-55 d). Table 1 gives an overview of the clinical information for this patient cohort. 68 Ga]Pentixafor was synthesized as previously described using a fully GMP compliant automated synthesizer (GRP, Scintomics, Fürstenfeldbruck, Germany) (27).

PET Imaging
All PET/CT scans were performed on a dedicated PET/CT scanner (Siemens Biograph mCT 64; Siemens Healthineers, Erlangen, Germany). Before acquisition of [ 18 F]FDG PET, patients fasted for at least 6 h and blood glucose levels were <160 mg/dl. Prior to [ 68 Ga]Pentixafor, patients did not fast. Imaging was performed 60 min after administration of 64-166 MBq (median, 145 MBq) of [ 68 Ga]Pentixafor and 297 (n = 1) or 301 (n = 2) MBq of [ 18 F]FDG, respectively. Spiral CT with (dose modulation with a quality reference of 210 mAs) or without (80mAs) intravenous contrast (120 kV, 512 × 512 matrix, 5 mm slice thickness) including a field of view from the base of the skull to the proximal thighs was acquired. Consecutively, PET emission data were acquired in three-dimensional mode with a 200 × 200 matrix with 2-3 min emission time per bed position. After decay and scatter correction, PET data were reconstructed iteratively with attenuation correction using the algorithm implemented by the manufacturer (Siemens Esoft, Siemens Healthineers, Erlangen, Germany).

Image Analysis
All PET/CT studies were visually assessed by two experienced nuclear medicine physicians (RAW and CL). Lesions were rated as visually detectable if they could be identified by both reviewers on the PET images in a consensus setting.
For derivation of maximum standardized uptake values (SUV max ) of both primary tumors (all patients) and metastatic disease (if present, in n = 7/19, 36.8%), 3-dimensional volumes of interest (VOI) were drawn around the respective lesions. If patients displayed more than 5 metastases (all histologically proven or verified by imaging follow-up), the five lesions with the highest uptake were selected. For calculation of tumor-tobackground ratios (TBR), VOIs with a diameter of 3 cm were placed in normal liver parenchyma and mean SUV were noted. The radiotracer concentration in the VOI was normalized to the injected dose per kilogram of patient's body weight to derive the SUVs (22).
For [ 18 F]FDG-PET/CT, an analogous procedure was carried out.

Statistical Analysis
Descriptive statistics were predominantly utilized. All results are displayed as mean ± SD or as median + range where appropriate. The two-tailed paired Student's t-test was used to check for a correlation between [ 68 Ga]Pentixafor SUV max and histologic CXCR4 expression. A P-value of <0.05 was considered to be statistically significant.
In 16/17 (94.1%), imaging results (SUV max of the primary) could be compared to immunohistological CXCR4 staining, while in 1/17 (5.9%) a SUV max could not be derived as the primary could not be identified (#6, suffering from CUP). A significant correlation between IRS and [ 68 Ga]Pentixafor SUV max of the primary among all tumor entities was detected (R = 0.62, P < 0.05, Figure 1). Figure 2 shows concordant cases of immunohistochemistry and non-invasive imaging in patients suffering from CCC (patient #7) and HCC (patient #17), respectively. Figure 3 demonstrates CXCR4-directed imaging in further selected tumor entities which (with the exception of patient #9 suffering from low-grade ovarian carcinoma) primarily demonstrated moderate to no uptake on CXCR4directed imaging.

DISCUSSION
An extensive body of literature has demonstrated that overexpression of CXCR4 is linked with increased aggressiveness and worse prognosis in solid cancers. Thus, this chemokine receptor is an interesting target in oncology and several therapeutic antagonists have been developed (30,31). A recent phase I trial in women with advanced HER-2 negative metastatic breast cancer investigated a combination regimen of balixafortide (a peptidic CXCR4 antagonist) and eribulin and demonstrated favorable safety and tolerability as well as promising anti-tumor activity (32). Additionally, a first experience with chemokine receptor-directed radioligand therapy in heavily pre-treated hematologic disease has been reported (17)(18)(19).
As an important pre-requisite to CXCR4 directed therapy, [ 68 Ga]Pentixafor PET/CT enables the non-invasive evaluation of receptor expression of all tumor lesions. Marked inter-and intraindividual differences in CXCR4 expression have been noted (33). Further, the receptor presentation on the tumor cell surface seems to be highly dynamic and influenced by a variety of factors including previous therapy (34).
In this study, we aimed to expand the experience with noninvasive imaging of CXCR4 in solid cancers. Previous reports had hinted at a potential role of chemokine-directed imaging in selected entities such as small cell lung cancer, adrenocortical carcinoma, or glioblastoma (21,24,35), while another study by Vag et al. questioned its suitability in other tumors including sarcoma, pancreatic cancer, and breast cancer (25). In the present analysis, additional solid cancers in which CXCR4 expression had been linked to metastasis and inferior outcomes such as renal cell cancer (36), ovarian cancer (37), and CCC (38) were investigated. All patients presented with newly diagnosed, treatment-naïve disease. To exclude potential influence of concomitant therapy on receptor surface expression, biopsies, and/or surgery were performed shortly after PET imaging and prior to treatment The patient presented with negative IHC for CXCR4 derived from a surgical specimen (IRS, 0, Table 1). Magnification of IHC: × 400. initiation. However, only weak to moderate [ 68 Ga]Pentixafor uptake was recorded in the vast majority of patients and almost 80% of lesions could not be visually detected. In addition, only two patients exhibited primary tumor SUV max > 10. Interestingly, both patients suffered from CCC (#7, Figure 2 and #12). These findings were paralleled by immunohistochemistry that also identified relevant CXCR4 expression in few tumor specimens and correlated well with non-invasive imaging results (R = 0.62, Figure 1).
Taken together, the current findings indicate that [ 68 Ga]Pentixafor is unlikely to play a major role in staging and re-staging of most solid tumors, in particular when compared to [ 18 F]FDG. Given the physiologic expression of CXCR4 on hematopoietic stem cells and thus the need for subsequent stem cell support (17,18), endoradiotherapy with radiolabelled Pentixather might also be reserved to very selected cases.
Future efforts for potential applications of CXCR4-directed imaging might focus on the characterization of intra-/interlesional heterogeneity by performing dual-radiotracer studies (in conjunction with [ 18 F]FDG) to visualize different levels of tumor de-differentiation and predict lesions with prognostic relevance. For example, CXCR4-directed PET/CT might help to visualize receptor positive cancer stem cells (and their niche) which are considered to be especially resistant to radiation or chemotherapy (1,39).
Several limitations of the present study have to be considered. It is retrospective and the number of patients is rather small. Further research including a higher number of subjects is definitely warranted to confirm the preliminary findings of the present feasibility study. Histologic validation of imaging results was not available in all cases. Additionally, a variety of different tumor entities were included. However, all patients presented with newly diagnosed disease and were treatment-naïve at the time of imaging. Moreover, in only 3 patients, an additional [ 18 F]FDG PET/CT has been conducted and future studies should evaluate potential tumor heterogeneity in a higher number of subjects. Although the value of [ 68 Ga]Pentixafor PET/CT has been investigated before, our small series adds first experience with other tumor entities such as ovarian cancer and renal cell carcinoma, along with histopathologic proof in the majority of the cases.

CONCLUSIONS
In this cohort of various treatment-naïve solid malignancies, CXCR4 expression as detected by [ 68 Ga]Pentixafor-PET/CT and immunohistochemistry was rather moderate. Thus, CXCR4directed imaging may not play a major role in the management of the majority of solid cancer patients.

DATA AVAILABILITY
The raw data supporting the conclusions of this manuscript will be made available by the authors, without undue reservation, to any qualified researcher.

ETHICS STATEMENT
The local institutional review board of University Würzburg waived the requirement for additional approval because of the retrospective character of this study. All subjects gave written informed consent in accordance with the Declaration of Helsinki.

AUTHOR CONTRIBUTIONS
RW, SK, CL, TH, MP, and SR designed the study, wrote the manuscript, and researched data. MK, AS, H-JW, and AB performed analysis. All authors aided in drafting the manuscript and revised it critically for important intellectual content. All authors read and approved the final manuscript.