Al 18 F-AEEA-HER2-BCH Abody Reveals Clearer PET Imaging Than 18 F-FDG in Patients With Breast Cancer: A Pilot Clinical Translation Study

Purpose To construct, validate and preclinically evaluate a novel HER2 target agent Al 18 F-AEEA-HER2-BCH abody (Al 18 F-HER2-BCH), and perform a pilot clinical translational study in HER2-positive breast cancer patients. Methods In preclinical study, the HER2 target specicity were veried using HER2-positive and HER2-negative cells and mice models. In clinical study, we enrolled 14 HER2-positive breast cancer patients (7 newly diagnosed patients and 7 relapsing metastatic patients) and one HER2-negative breast cancer patients. The safety were observed and lesion detection ability was compared with 18 F-FDG in 14 HER2-positive breast cancer patients (NCT04547309). All patients injected with 222±18.5 MBq Al 18 F-HER2-BCH and scanned at 2 h and 4 h, and received standard 18 F-FDG PET/CT scan within 7 d. 18 F-FDG (93 vs 45), especially lymph node (39 vs 23), bone (39 vs 21), and liver metastases (30 vs 3). Compared with 18 F-FDG, Al 18 F-HER2-BCH upstaged 5/7 newly diagnosed breast cancer patients and detected more distant metastases in 4/7 relapsed patients. Conclusions The ﬁ ndings described here demonstrated that Al 18 F-HER2-BCH had high anity and target specicity, and it was feasible to detect HER2-positive lesions in primary and metastatic breast cancer patients.


Introduction
As of 2020, female breast cancer has now surpassed lung cancer as the leading cancer in terms of global incidence, with an estimated 2.3 million new cases, accounting for 11.7% of all cancer-related deaths. It is the fth leading cause of cancer-related mortality worldwide, with a 5-year survival rate of 31.4% 1 2 . HER2 is a transmembrane receptor of the tyrosine kinase family, which overexpress in approximately 15-20% of newly diagnosed breast cancer patients and regulate cell growth, survival, differentiation, angiogenesis and DNA repair 3 4 . Trastuzumab, an anti-HER2 monoclonal antibody, was approved for the treatment of advanced breast cancer by the Food and Drug Administration (FDA) in 1998. Several evidences have demonstrated that trastuzumab therapy is associated with a complete pathological response to 23-26% in patients with HER2 positive advanced breast cancer, while when adding chemotherapy to the trastuzumab therapy protocol, better results were reported 5 . However, HER2-targeted therapy fails in 50-70% of patients with HER2-positive breast cancer even with combination treatments 6 7 . The widely resistance of the biological mechanisms are not completely understood. Therefore, identifying and characterizing in situ biomarkers of HER2 status can help to overcome trastuzumab resistance.
The HER2 expression in tumors is highly heterogeneous and it may change during the treatment.
Pathological examination is an invasive examination, which is not suitable for repeated operation and it is di cult to solve the problem of heterogeneity. The HER2 status is hard to evaluated according to MRI/CT-detected signal intensity 8 9 . 18 F-Fludeoxyglucose (FDG) PET/CT is a valuable imaging modality in the management of patients with breast cancer in stage IIIA or higher [10][11][12][13][14][15][16] . However, the sensitivity of 18 F-FDG for brain and liver lesions is low 17  Fluorine-18 is an ideal PET imaging agent due to its optimal half-life of 110 min. It has been widely used to develop various tracers to image receptors and assess uptake and pathway status in the body 22 .
Moreover, automated methods of radiolabeling have been developed using Current Good Manufacturing Practice (cGMP) guidelines for clinical use to promote clinical application. We added aminoethy lethanolamine (AEEA) and 6-aminocaproic acid (ACP-6) aminocaproic acid at the amino and carboxyl sides of the ZHER2:342 a body structure to construct AEEA-HER2-a body-BCH. In this study, we used the Al 18 F-AEEA-HER2-a body-BCH (hence abbreviated as Al 18 F-HER2-BCH) reagent to assess its safety, feasibility, biodistribution, and tumor targeting ability in HER2-positive breast cancer patients.

Preclinical Study
Cell study HER2 (+) NCI-N87 and HER2 (-) MDA-MB-231 cell lines were chosen for in vitro cell uptake study. 0.5 mL of 74 KBq Al 18 F-HER2-BCH in fresh medium was added into each well and incubating for 10 min, 30 min, 1 h, 2 h, 3 h and 4 h. For the blocking experiment, cells were co-incubated with 50 µg HER2-a body-BCH per well. The Kd of Al 18 F-HER2-BCH was determined by adding different concentrations of radiotracer to N87 cells. The details could be seen in supplemental materials.
Micro-PET study 4.8 MBq of Al 18 F-HER2-BCH was injected into each of BALB/c female mice bearing NCI-N87 and MDA-MB-231 tumors with or without 200 µg HER2-a body-BCH via the tail vein. The mice were anesthetized with 3% (v/v) iso urane and underwent a Super Nova PET/CT scanner (PINGSENG Healthcare (Kunshan) Inc.) with continuous 1% (v/v) iso urane. Micro-PET scans were conducted at 1, 2, and 4 h after injection. The counts of regions of interest (ROI) over tumor, kidney and muscle were collected.
Bio-distribution Mice bearing NCI-N87 and MDA-MB-231 tumors were injected with 0.74 MBq of Al 18 F-HER2-BCH (0.2 ml) with or without 200 µg HER2-a body-BCH via the tail vein. The mice were sacri ced in groups (n = 4) at 0.5, 1, 2, and 4 h postinjection. The heart, liver, lung, kidney, spleen, stomach, bone, muscle, blood, other gastrointestinal organs and tumors were collected and weighed, and the radioactivity was measured with a γ-counter.10 samples of 1.0% injected dose were measured at the same time intervals as a standard. The results were expressed as the percent of injected dose per gram (%ID/g).
Radiotoxicity Al 18 F-HER2-BCH was injected at a dose of 1.55 GBq/kg to each of KM female mice (200 µL, n=7, 20.7 ± 2.2 g, 5 weeks). In the control, 7 mice were injected with 200 µL of saline. The weight and routine orbit blood were monitored for 7 d, and liver function and renal function were detected on the 7th (see supplemental materials for details).

Clinical Study
Patient Enrollment This study was a single-center clinical trial, approved by the Medical Ethics Committee of Peking University Cancer Hospital and American Center for Clinical Trial Management (Ethics Approval License No.2019KT114 and NCT04547309). The patients included in this study had pathologically con rmed HER2-positive (IHC 3+ or IHC 2+/FISH+) breast cancer. Inclusion criteria were as follows: congestive heart failure, severe liver or kidney dysfunction, pregnant or lactating. Ongoing treatment was not an exclusion criterion. All patients provided written informed consent before participating in the study.
Al 18 F-HER2-BCH and 18 F-FDG PET/CT imaging Each patient was injected with 222±18.5 MBq of Al 18 F-HER2-BCH and 500 µg HER2-a body-BCH as our previous study 23 to reduce the non-speci c uptake of liver. For the rst patient, serial whole-body dynamic PET acquisitions were performed immediately after injection (0-40 min dynamic scan, 6 passes) and a static whole-body scan at 2 h and 4 h. The other 13 patients were scanned at 2 h and 4 h. All patients underwent 18 F-FDG (3.7 MBq/Kg) PET/CT scan at 1 h post injection within 7 d, with fasting at least 6 h before imaging.
Imaging was performed on a Biograph mCT Flow 64 scanner (Siemens, Erlangen, Germany), ranging from the apex of the skull to the mid-thigh. The axial eld of view of PET was 21.6 cm. PET was acquired in 3-dimensional mode ow motion (bed entry speed 1 mm/s). Reconstruction was performed on Siemens Company Multimodality Workplace (MMWP) by the TrueX+TOF method; attenuation correction was performed using the unenhanced low-dose CT data.

Image Analysis
The imaging were interpreted by two experienced radiologists. Quanti cation of Al 18 F-HER2-BCH uptake was performed using a volume of interest (VOI) analysis method, and the mean standard uptake (SUVmean) values and the maximum standard uptake (SUVmax) were calculated. A positive PET lesion was de ned as an SUVmax greater than that of the baseline mediastinal blood pool.

Statistics
The Kd value were calculated by One site-Speci c binding respectively with GraphPad Prism 5.0 18 .
Differences between uptake (% IA/10 6 cells, %ID/g, SUVmax) of two tracers were assessed using the wilcoxon signed-rank test included in GraphPad Prism 5.0. Differences with a P-value less than 0.05 were considered statistically signi cant.
The results of quality control were shown in Figure S1. Radio-HPLC analysis of the radiotracers revealed no aggregates, fragments, or radioactive purities. In vitro stability, Al 18 F-HER2-BCH was stable in both saline and 5% HSA within 6 h. In vivo stability, after the mice were injected with the radiotracer for 30 min, the radio-HPLC results of blood showed the main peak (>90%) at retention time of 10.26 min, which was close to that of Al 18 F-HER2-BCH (T=9.85), indicating that Al 18 F-HER2-BCH was stable without signi cant decomposition in 30 min.

Cell uptake and Kd study
As shown in Figure 1A, Al 18 F-HER2-BCH showed signi cantly higher uptake in NCI-N87 cells than in MDA- Bio-distribution study The uptake of Al 18 F-HER2-BCH in the NCI-N87 tumors was obviously higher than in MDA-MB-231 tumors (12.2±1.08% ID/g vs 1.46 ± 0.57 %ID/g, p < 0.0001) at 2 h, and the tumor uptake was blocked to 1.46 ± 0.57 %ID/g by coinjection of 500 µg HER-a body-BCH ( Figure 1D). Figure 1E, Al 18 F-HER2-BCH depicted clearly tumor uptake at 1 h, maintain high uptake until 4 h. Compared with MDA-MB-231 models, Al 18 F-HER2-BCH showed signi cantly higher uptake in the NCI-N87 models during whole imaging. When 500 µg HER2-a body-BCH was coinjected, the uptake in the N87 models markedly decreased further con rmed the speci c uptake. The tumor-to-muscle (T/M) ratios of NCI-N87 was higher than MDA-MB-231 and NCI-N87 block, with T/M values of 8.03, 1.27, 1.41 at 2 h, respectively ( Figure 1C). Pathological examination Figure 1F

Radiotoxicity
No signi cant differences in weight, hematologic markers, liver function tests or renal function tests were observed between the mice injected with Al 18 F-HER2-BCH (1.48 GBq/kg) and those injected with saline within 1 week of growing (p>0.05, Figure S2 and Figure S3). In addition, no lethal or chronic toxicity, hematologic effects, or biochemical effects were observed.

Clinical study in BC Patients
Patient characteristics Between June 2020 and February 2021, 14 HER2-positive breast cancer patients and one HER2-negative breast cancer patient were completed the study protocol. Among them, 7 patients were newly diagnosed and had not received any treatment. Totally 9 primary lesions in 7 newly diagnosed patients were pathologically con rmed. The other 7 patients relapsed during follow-up according to increased tumor markers or other features on imaging examinations.
The optimal time point for Al 18 F-HER2-BCH PET/CT imaging Dynamic imaging showed blood pool activity was initially high, decreased rapidly in the rst 8 min, and was then followed by a slower decrease until 32 min. High uptake was noticeable in both the liver and spleen in the rst 8 min. Uptake in the glands (lacrimal gland, parotid gland, submandibular gland and thyroid gland) and other tissues increased gradually. The tumor could be visualized at 8 min, exceeded the blood pool activity after 30 min, and maintained high accumulation throughout the examination.
For metastatic lymph nodes detection, Al 18 F-HER2-BCH detected more lesions than 18 F-FDG (20 vs 7), and the uptake was higher (SUVmax, 6.42±3.84 vs 5.1±1.06). The average diameter of lymph nodes detected by Al 18 F-HER2-BCH was 0.72 cm, and the smallest one was only 0.29 cm. As depicted in Figure   4, Al 18 F-HER2-BCH more clearly showed the primary tumor and the involved axillary lymph nodes than 18 F-FDG in a HER2-positive primary diffuse breast cancer patient. Puncture pathology showed that the primary focus was HER2 3+, proving that the patient was a HER2-positive patient.
Both images revealed no distant metastasis in the 7 patients.
Moreover, Al 18 F-HER2-BCH changed TNM stages classi ed by 18 F-FDG. One patient changed from T0 to T1, one patient changed from T2 to T3, two patients changed from N1 to N2, and one patient changed from N1 to N3.

Diagnostic performance of 18 F-FDG and Al 18 F-HER2-BCH PET/CT in relapsed patients
Al 18 F-HER2-BCH detected visceral metastases in 7 of 7 patients (100%), while 18 F-FDG detected them in 5 of 7 patients (71.4%), missed 1 patients with liver and bone metastases and 1 patient with liver metastases. In addition, Al 18 F-HER2-BCH detected more bone metastases in 2/5 patients ( Figure 5). Among 7 patients, 93 lesions (19 lymph node metastases, 39 bone metastases, 30 liver metastases, and 5 lung metastases) were selected for analyses of two images. The number of lesions and the semiquantitative parameters were presented in Table 2 and Figure S6.
Diagnostic performance of 18 F-FDG and Al 18 F-HER2-BCH PET/CT in HER2-negative patient As described in Figure 6, 18 F-FDG showed high uptake in the bone metastases in a HER2-negative breast cancer patient with recurrence, while there was relatively low uptake in the same lesion sites of Al 18 F-HER2-BCH. The pathological results of the breast lesions showed HER2 2+, and FISH test showed no ampli cation, proving that the patient was a HER2-negative patient. The HE staining results could be seen in Figure S7.

Discussion
Molecular imaging is a whole-body, noninvasive and dynamic medical imaging technology, that provides detailed images of what is happening at both the cellular and molecular levels inside the human body. HER2 status is an important tumor characteristic for guiding treatment in patients with breast cancer. We have successfully developed HER2-targeted antibody molecular probes, 64 Cu-NOTA-trastuzumab 24 and 124 I-trastuzumab 25 , as PET/CT imaging agents in the clinical imaging of patients with gastric cancer. Due to the slow clearance of radioactivity from antibodies in the blood and normal organs, we explored a small-scaffold HER2 radiotracer ( 68 Ga-NOTA-MAL-MZHER2) and evaluated the sensitivity, speci city and utility for HER2-targeted imaging in patients with advanced gastric cancer (AGC) 23 26 ,and it was feasible to noninvasively detect the HER2 status in AGC patients and enable early detection with a low dose.
To increase the e ciency of 18 F labeling and improve the stability in vivo, we successfully constructed a novel Al 18 F-HER2-BCH that could be e ciently radiosynthesized in an easier process with a favorable nondecayed radiochemical yield (45.2±10.5%) within 30 min. Regarding the overall results of the different imaging modalities among the 14 patients initially assessed, Al 18 F-HER2-BCH upstaged 5/7 patients with newly diagnosed breast cancer and detected more visceral metastases than 18 F-FDG in 4/7 (57.1%) patients with relapsed breast cancer. Combined with the pathological results of the primary lesions, it was veri ed that there was high uptake of Al 18 F-HER2-BCH in the tumor sites with high HER2 expression. In addition, Al 18 F-HER2-BCH showed relatively low tumor uptake in a HER2-negative patient who with HER2 2+ and FISH-, further con rmed the HER2 target speci city.
For the detection of primary focus of newly diagnosed breast cancer patients, Al 18 F-HER2-BCH showed a clearer tumor delineation and a higher tumor-to-background contrast than 18 F-FDG. For the detection of lymph node metastases, Al 18 F-HER2-BCH showed better detection capability than 18 F-FDG (39 vs 21).
Moreover, more lymph nodes smaller than 0.5 cm could be detected by Al 18 F-HER2-BCH, which were easily missed in either 18 F-FDG or other conventional imaging.
Recent advances in breast cancer treatment have showed that, 20-30% developed metastatic disease after the initial diagnosis and treatment of early-stage breast cancer among female breast cancer patients 27 . The four most common metastatic organs associated with breast cancer are bone (50-70%) 28 , lung (17%) 27 , liver (12-20%) and brain (5-20%) 9 . The patients with metastatic breast cancer had a poor prognosis 29 , with a 5-year survival rate of 26% 30 , and a reported median survival of 18 to 24 months. In addition, breast cancer patients with liver metastasis have the worst result, with reduced median survival ranging from 14 to 16 months 31 32 . Metastatic diffusion is sometimes di cult to identify on conventional imaging. When patients are rst diagnosed clinically after show with symptoms, such as bone pain or hepatitis, the optimal intervention time has been missing 32 . The results showed Al 18 F-HER2-BCH had a good detection effect for bone and liver metastatic lesions, which could facilitate early clinical intervention. Compared with 18 F-FDG and conventional imaging, Al 18 F-HER2-BCH is a good supplementary means for the detection of liver metastases. Furthermore, Al 18 F-HER2-BCH showed a lower background in brain than 18 F-FDG, which is more conducive to the observation of brain and skull metastases.
The limitation of the present study is the small number of breast cancer patients with limited diversity in HER2 status. In addition, histopathological con rmation of every detected lesion was not feasible due to ethical and practical reasons. Moreover, we did not analyze the relationship between Al 18 F-HER2-BCH PET imaging and HER2 target treatment in this study. We will further improve the research in a later stage.

Conclusion
Al 18 F-HER2-BCH PET/CT is feasible for detecting HER2-positive lesions in newly diagnosis and metastatic breast cancer patients and shows better detection ability for small/nonnodule primary lesions, small metastatic lymph nodes, and bone and liver metastases. This approach is helpful for monitoring the expression level of HER2 in lesions and facilitates accurate diagnosis and individualized treatment.  Tables   Due to technical limitations, table 1 to 3 is only available   Dynamic imaging study of Al18F-HER2-BCH during 0.5min-40min, static imaging study at 2h and 4 h, and distribution curve of normal tissues. The anterior maximum intensity projection image obtained at 18F-FDG and Al18F-HER2-BCH PET in 7 primary breast cancer patients.  Images of 18F-FDG PET and Al18F-HER2-BCH in 7 relapse and metastatic breast cancer patients. The black arrow showed the lesions detected by both 18F-FDG and Al18F-HER2-BCH, and the red arrow showed the lesions detected only by Al18F-HER2-BCH.