Human Epidermal Growth Factor Receptor 2 (HER2), a transmembrane glycoprotein, plays a pivotal role in cellular signaling through the initiation of cascading events upon either hetero- or homo-dimerization. This activation triggers various proteins, including mitogen-activated protein kinase (MAPK), phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K), and protein kinase C (PKC). As a result, the HER2 activation regulates many fundamental cellular processes, including proliferation, survival, and migration, thus being a crucial step in tumor growth and invasion [1].

HER2 amplification and overexpression are prevalent across several cancer types, with the highest incidences noted in breast and gastric cancer, representing 20%–30% of cases in these two cancer types [2–4]. Additionally, research studies have reported HER2 involvement in ovarian cancer [5], endometrial cancer [6], esophagus cancer [7], lung cancer [4], and cervical cancer [8]. Moreover, HER2 mutations have also been identified in bladder cancer (more than 10% of cases) and colorectal cancer (5.8% of cases) [9].

Therefore, therapy targeting HER2 has become an essential component of treatment for HER2-positive cancers, particularly in breast and gastric cancer. Trastuzumab, a humanized monoclonal antibody, is the most common HER2-targeted therapy and has significantly improved the overall survival of patients with HER2-enriched breast cancer [10]. Many other strategies, such as targeting different HER2 domains, antibody-drug- conjugates, or tyrosine kinase inhibitors, have recently been approved for treatments; however, these therapies often entail adverse effects, especially cardiovascular risks [11–13].

For this reason, a thorough assessment of HER2 status is required for selecting therapy candidates and monitoring the treatment responses. Currently, fluorescence in situ hybridization (FISH) and immunohistochemistry (IHC) are the standard assessment tools, but they are invasive and may not always accurately represent the entire tumor. Consequently, evaluating the intra-tumor and the inter-tumor heterogeneity is suboptimal. Molecular imaging, via positron emission tomography (PET), offers a solution to this challenge by providing a systemic, non-invasive assessment method with both temporal and spatial information.

Trastuzumab-based imaging tracers hold great potential due to their high affinity for HER2 and accurate reflection of the antibody’s biodistribution, paving the way for future radioimmunotherapy dosimetry and personalized treatments. In this context, ⁵²Mn, with a prolonged half-life of 5.6 days [14], matches trastuzumab’s in vivo half-life (a few days to 23 days, depending on administered dose [15]), enhancing the tracer’s imaging capabilities. Moreover, ⁵²Mn can be produced from natural chromium by conventional medical cyclotron using 16–6 MeV bombardment energy [16]. We chose DOTAGA-anhydride as a bifunctional chelator owing to its high reactivity, regioselective anhydride-opening reaction, and the ability to conjugate without needing protection for other chelating ligands or producing unwanted side products [17].

In our study, we utilized [⁵²Mn]Mn-DOTAGA-trastuzumab to assess the tumor uptake of HER2-positive and HER2-negative xenografts in preclinical models; melanoma xenografts were included for specificity control. Additionally, we investigated the tumor uptake differences between orthotopic and ectopic HER2-positive tumors.

Given the need for an isotope with a suitable half-life for labelling trastuzumab, many of the radiometals studied still have significant limitations. For instance, ⁸⁹Zr (T_1/2 = 78.4 h) has been used in [⁸⁹Zr]Zr-trastuzumab studies with promising clinical results [22, 23]. Despite efficient labelling using the chelator DFO, in vivo stability of the tracer at later time points is a concern, as free zirconium accumulates in bone, causing unnecessary bone marrow radiation and false-positive bone metastasis. Recent improvements, particularly with DFO* having additional hydroxamic acid, have improved complexation stability in vivo [24, 25]. ⁶⁴Cu (T_1/2 = 12.7 h) has been investigated in [⁶⁴Cu]Cu-trastuzumab studies on patients [26, 27], although its redox activity in vivo suggests that further improvements in tracer integrity are needed [24].

⁵²Mn, a potential candidate for antibody-based imaging, offers a superior radiation profile compared to ⁸⁹Zr and ⁶⁴Cu. With a T_1/2 = 5.6 days and a 29% beta-positive emission with low maximum energy (0.575 MeV), ⁵²Mn provides a short-tissue penetration and an improved resolution [14]. This makes ⁵²Mn an attractive but underexplored option for antibody labelling, with only a few studies investigating its potential, indicating a promising area for further research and development [28–32].

After intravenous injection, the tracer shows an initial high blood uptake due to FcRn-mediated antibody recycling, with moderate liver and spleen uptake driven by FcR-mediated uptake [33] and permeable sinusoidal capillaries [34, 35]. Lung uptake is attributed to the high vascularity [36] and low reflection coefficient of pulmonary tissue, allowing for enhanced tracer uptake [37]. Remarkably, renal activity exceeded typical liver activity for antibody-based tracers, due to the additional carboxylic arm of the tracer (compared to the conventional DOTA chelator), which increases the hydrophilicity of the tracer and consequently elevates the renal excretion [38].

Interestingly, moderate activity was also seen in the ovaries and the lacrimal glands. The physiological expression of HER2 in the ovaries is well-documented [39]. However, since trastuzumab does not cross-react with murine HER2 [40], the ovarian activity may be attributed to non-specific uptake. The lacrimal glands produce protective tears containing immunoglobulins [41], especially under the effect of anesthesia with constant airflow from the breathing mask [42], which causes dry eyes and lens opacity [43]. This was also observed in our scans, resulting in the noticeable tracer uptake in the eyes and lacrimal glands, particularly in SCID mice that lack endogenous immunoglobulins. These findings underscore the advantages of PET imaging in providing a comprehensive whole-body assessment, enabling the detection of unexpected tracer accumulation.

Initial urinary activity in the bladder suggests the rapid excretion of a small portion of unconjugated tracer, as observed in our previous study using unconjugated [⁵²Mn]Mn-DOTAGA [28]; nonetheless, low urinary bladder activity in subsequent scans indicates stable conjugation in vivo. The kidney, pancreas, salivary glands, and joints are frequently identified as the localized sites of free manganese, as demonstrated in our previous study with [⁵²Mn]MnCl₂ [31] and in other related research [44, 45]. Free manganese is strongly reabsorbed by the kidneys, probably via transporters in proximal tubular cells [46]. In the pancreas, voltage-dependent calcium channels assist in manganese uptake [47] and the metal remains bound to pro-carboxypeptidase B [44]. In our current study, in addition to the stable conjugation, the complexation of the tracer seems to be stable in vivo, as suggested by the minimal uptake in the pancreas, salivary glands, and joints without renal reabsorption. Using a similar compound, [⁵²Mn]Mn-DOTAGA-p-SCN-trastuzumab, we reported a high stability (RCP maintained above 90% up to day 10) of the ⁵²Mn-labelled trastuzumab using the DOTAGA chelator [31].

In our study, both HER2-positive tumors demonstrated high tumor-background contrast; however, significantly higher uptake was observed in the orthotopic tumors compared to the ectopic tumors, possibly due to the microenvironment which favors tracer uptake in orthotopic tumors [48, 49]. This phenomenon corresponds to the better microenvironment characterized by higher microvascular density [50] and increased perfusion [49], which facilitate tracer penetration and promote binding to more specific binding sites.

However, the differences in uptake between MDA-MB-HER2+ and MDA-MB-468 cell lines at the same inoculation site, which diminished over time, are likely due to non-specific binding of the tracer. The larger tumor size in the MDA-MB-468 group (Supplementary Figure S1), leading to increased non-specific uptake [51], is partly responsible for this finding. However, tumor size alone does not explain for the dynamic and clearance of uptake, as aggressive melanoma xenografts, although large, showed lower activity compared to HER2-positive xenografts. Furthermore, melanoma xenografts exhibited a similar uptake pattern to HER2-positive xenografts, in contrast to the MDA-MB-468 group, where uptake increased steadily. This disparity can be due to the fact that both MDA-MB-HER2+ (Figure 5A) and the B16F10 melanoma (Figure 5C) xenografts exhibit necrotic features [52], resulting in decreased overall xenograft uptake [53]. Conversely, the stroma-rich features of MDA-MB-468 xenografts facilitate tracer retention, resulting in slower clearance rate and increased activity over time [54]. Furthermore, the HER2-positive xenograft can show limited tracer uptake due to the binding site barrier as specific binding accumulates in the peripheral parts, reducing uptake in the necrotic/hypoxic core located further from the feeding blood vessels [55], as shown in our in vivo images (Figure 3). Whereas MDA-MB-468 tumors with known HER2 phosphorylation at tyrosine Y877 [56] can render the xenografts sensitive to trastuzumab [56, 57].

Our pilot in vitro study demonstrates higher tracer uptake in HER2-positive cells; further blocking studies or direct binding assays are required to confirm specific tracer binding. In vivo, while the tracer retains some specificity that can differentiate the HER2 positivity in initial scans, tumor uptake includes a high proportion of non-specific binding which is responsible for elevated uptake in HER2-negative tumors at later time points. Astudy using the same compound labelled with radioactive indium, [¹¹¹In]In-DOTAGA(anhydride)-trastuzumab, reported a decrease in the tracer immunoreactivity compared to labelling with the parent DOTA chelator [58]. This reduced immunoreactivity may account for the high non-specific binding observed with our tracer.

Recently, BPPA, a novel bispyclen-based chelator, was used to label trastuzumab with ⁵²Mn. The [⁵²Mn]Mn-BPPA-trastuzumab showed superior tumor-to-background ratios and earlier detection of HER2 positivity compared to [⁵²Mn]Mn-DOTAGA-p-SCN-trastuzumab, though its in vivo stability requires improvement [31]. Omweri et al. demonstrated that, with comparable specific activity to [⁵²Mn]Mn-BPPA-trastuzumab and enhanced stability, [⁵²Mn]Mn-Oxo-DO3A-trastuzumab effectively assessed HER2 status and achieved a higher radiochemical purity yield (RPY) compared to using p-SCN-Bn-DOTA as the chelator [32].

Our study has several limitations. First, there were only two mice in the MDA-MB-468 group, which required the assumption of a normal distribution for the two-way ANOVA analysis. Thus, though our pilot study aimed to generate preliminary insights, further research with larger sample sizes is necessary to provide stronger statistical evidence. Second, discrepancies between PET/CT and PET/MRI can arise from variations in attenuation correction maps [59, 60]. In our study, we also observed systemic differences between these two modalities (Supplementary Tables S1, S2). To address this limitation, we used the tumor-to-background ratio for assessment, aiming to ensure more accurate and reliable results. Using the ratios, such as SUV_mean Organ/Muscle, provided reproducible results, as demonstrated by the comparable biodistribution observed in our two breast cancer-bearing groups scanned with different modalities (Supplementary Figure S2). Lastly, we lacked in vitro binding assays to further investigate the immunoreactivity or receptor binding characteristics of the tracer.

Based on our findings, we conclude that higher antibody tracer uptake is expected in orthotopic tumors compared to ectopic tumors. However, due to the suboptimal specificity of the tracer, this difference may result from either higher HER2 expression or a more favorable microenvironment at the orthotopic inoculation site, and further investigation is needed to confirm our findings. Several non-specific factors, such as tumor size, necrotic or stromal-rich properties, and the tumor microenvironment, influence imaging outcomes. Despite the need for improved specificity towards HER2, [⁵²Mn]Mn-DOTAGA(anhydride)-trastuzumab demonstrates a favorable tumor-to-background ratio and relatively good tracer stability. Additionally, the paramagnetic properties of manganese (II) make it a valuable tool for MRI studies [61], especially in the expanding field of hybrid cameras.

The study demonstrates high tumor/non-tumor ratios and the relative stability of [⁵²Mn]Mn-DOTAGA(anhydride)-trastuzumab and highlights the influence of inoculation sites, tumor characteristics, and the microenvironment on tumor tracer uptake. Despite its ease of production, the tracer exhibits noticeable non-specific binding and requires further improvement in immunoreactivity. Nevertheless, ⁵²Mn-labelled trastuzumab has great potential for in vivo HER2 assessment as it allows non-invasive longitudinal characterization of HER2 expression in tumors and whole-body imaging to detect unusual tracer accumulation.