Panitumumab labeled with Auger electron-emitting 197gHg/197mHg: cytotoxicity on human breast cancer cells and tumour and normal tissue uptake in mice with EGFR-overexpressing breast cancer xenografts Authors: Arthur C K Chu, Felix Ho, Conrad Chan, Zhongli Cai, Yumeela Ganga-Sah, Parmissa Randhawa, Caterina F Ramogida, Valery Radchenko, Raymond M Reilly Published: 2026 ABSTRACT AbstractBackground197gHg and 197mHg are attractive radionuclides for radioimmunotherapy of cancer due to their abundant Auger electron emissions (23.2 and 19.4 electrons/decay, respectively). Our aim was to study the cytotoxicity and DNA-damaging properties of panitumumab labeled with 197gHg/197mHg on a panel of human breast cancer cells expressing different levels of epidermal growth factor receptor (EGFR) (104 to 106 EGFR/cell) and evaluate the tumour and normal tissue uptake of these radioimmunoconjugates in mice with EGFR-overexpressing MDA-MB-468 human breast cancer xenografts (106 EGFR/cell).ResultsPanitumumab was conjugated to 2-[4,7,10-tris(2-amino-2-oxoethyl)-1,4,7,10-tetrazacyclododec-1-yl]acetamide (TCMC) or a sulphur-rich (NS4) bifunctional chelator to complex 197gHg/197mHg or directly labeled with 197gHg/197mHg through endogenous mercury binding sites. All radioimmunoconjugates bound specifically (87–91%) to EGFR on MDA-MB-468 cells. The radioimmunoconjugates were bound, internalized and imported into the nucleus of breast cancer cells dependent on their EGFR expression. Absorbed doses in the nucleus from the Auger electron emissions of 197gHg/197mHg were correlated with EGFR expression level. The radioimmunoconjugates were most cytotoxic to MDA-MB-468 cells reducing their clonogenic survival to < 5% while unlabeled immunoconjugates did not decrease survival. Blocking EGFR with excess panitumumab restored clonogenic survival to ~ 60%. DNA double-strand breaks were highest in MDA-MB-468 cells and were decreased by blocking EGFR. Subcutaneous MDA-MB-468 tumours in athymic (nude) mice were imaged by SPECT/CT at 1 d and 3 d post-injection of the radioimmunoconjugates. Tumour uptake on images and in biodistribution studies was decreased by co-administering an excess of panitumumab. High kidney uptake was observed.ConclusionsPanitumumab was complexed to 197gHg/197mHg after conjugation to a bifunctional chelator or labeled directly with 197gHg/197mHg. These radioimmunoconjugates bound specifically to EGFR on human breast cancer cells and caused DNA double-strand breaks that decreased their clonogenic survival. The radioimmunoconjugates localized in EGFR-overexpressing MDA-MB-468 tumours in mice but there was high kidney uptake, suggesting in vivo release of some 197gHg/197mHg. Competition between complexation of 197gHg/197mHg to a bifunctional chelator and binding to endogenous mercury binding sites in panitumumab presents challenges for radiolabeling and may explain the apparent release of some 197gHg/197mHg in vivo. Alternative labeling strategies such as a 2-step method involving conjugation of the preformed 197gHg/197mHg bifunctional chelator to panitumumab could be explored to address these challenges.Supplementary InformationThe online version contains supplementary material available at 10.1186/s41181-026-00449-9.Keywords: Auger electrons, 197gHg, 197mHg, Panitumumab, Epidermal growth factor receptors (EGFR), Breast cancer, Radioimmunotherapy, Cellular dosimetry, SPECT Introduction Radioimmunotherapy (RIT) employs monoclonal antibodies (mAbs) directed towards overexpressed tumour-associated receptors that are labeled with radionuclides emitting α- or β-particles or Auger electrons (AE) (Larson et al. 2015). AE are particularly attractive for RIT since although they have low energy (< 25 keV), this energy is deposited over nanometer distances (< 0.5 μm) resulting in high linear energy transfer (LET = 4–26 keV/μm) that is potent for inflicting lethal DNA double-strand breaks (DSB) in cancer cells, especially when AE are released near the cell nucleus (Ku et al. 2019). AE may further cause oxidative damage to the cell membrane that kills cancer cells (Paillas et al. 2016). High LET radiation such as AE may directly cause DNA DSB as well as damage DNA indirectly by forming reactive oxygen species (ROS) (Aghevlian et al. 2017). Thus, high LET radiation is less susceptible to tumour resistance mediated by hypoxia (Zahavi and Weiner 2020). AE are very precise for killing cancer cells, since they require specific binding, internalization and ideally, routing of RICs to the nucleus for their cytotoxic effects (Ku et al. 2019). Finally, an advantage of AE-emitting radionuclides is that they decay directly to a stable element, in contrast to α-particle emitters (e.g. 225Ac) that decay through a chain of daughter radionuclides that emit α- or β-particles (Aghevlian et al. 2017). Redistribution of these radioactive daughters may cause toxicity to normal tissues (Miederer et al. 2004). Finally, many AE-emitting radionuclides also emit γ-photons that allow imaging by single-photon emission computed tomography (SPECT) creating an opportunity for theranostic tumor imaging and RIT. We previously reported that the anti-epidermal growth factor receptor (EGFR) mAb panitumumab (Vectbix®, Amgen, Thousand Oaks, CA, USA) labeled with the AE-emitter, 111In ([111In]In-DOTA-panitumumab) was cytotoxic in vitro to EGFR-overexpressing PANC-1 human pancreatic cancer cells and inhibited the growth of PANC-1 tumours in vivo in non-obese diabetic severe combined immunodeficiency (NOD/SCID) mice, significantly prolonging their survival (Aghevlian et al. 2020). 111In (t1/2 = 67 h) decays by electron capture (EC) to 111Cd emitting 7.9 AE and a total of 6.9 keV of AE energy/decay. EGFR binding promoted internalization of [111In]In-DOTA-panitumumab into cancer cells and a nuclear translocation sequence (NLS) in the transmembrane domain of the EGFR (Hsu and Hung 2007) mediated uptake from the cytoplasm into the cell nucleus where the AE emitted by 111In were most damaging to DNA and lethal (Ku et al. 2019). Recently, we further reported that [111In]In-DOTA-panitumumab was cytotoxic in vitro to EGFR-overexpressing MDA-MB-468 and MDA-MB-231 human triple-negative breast cancer (TNBC) cells. MDA-MB-231 tumours in the mammary fat pad (MFP) of NOD/SCID mice were imaged by SPECT/CT with [111In]In-DOTA-panitumumab (Facca et al. 2022). MDA-MB-231/LM2-4 cells are a metastatic variant of MDA-MB-231 cells (Munoz et al. 2006). In NOD-Rag1−/−IL2Rgnull (NRG) mice with MDA-MB-231/LM2-4 human TNBC tumours in the MFP, adjuvant RIT with [111In]In-DOTA-panitumumab combined with surgical resection of the primary tumour in the MFP and X-ray irradiation (XRT) of the surgical margins significantly improved survival compared to surgery or XRT alone (Facca et al. 2023). Despite these encouraging results, mice with MDA-MB-231/LM2-4 tumors ultimately died of metastatic disease, suggesting the need for a more potent AE-emitter than 111In. Potential candidates are 197gHg and 197mHg which emit more abundent AE/decay and AE energy/decay than 111In (Ku et al. 2019). Metastable 197mHg (t1/2 = 23.9 h) decays to 197Au by electron capture (EC; 9%) or by isomeric transition (IT; 91%) to 197gHg [Supppementary Information, Fig. S1] emitting 19.4 AE and 13.5 keV of AE energy/decay. 197gHg (t1/2 = 64.1 h) decays by EC (97%) to 197Au emitting 23.2 AE and 16.1 keV of AE energy/decay (Ku et al. 2019) [Fig. S1]. The total number of AE/decay and AE energy/decay for 197gHg and 197mHg are 5.4-fold and 4.3-fold greater than 111In, respectively. 197gHg and 197mHg are co-produced from a natural gold (197Au) target in a cyclotron by the 197Au(p,n) 197g/mHg reaction (Randhawa et al. 2021). In addition to their AE emissions, 197gHg emits a 77 keV γ-photon (19%) and 197mHg emits a 134 keV γ-photon (33%) that may be imaged by SPECT. Recently, we reported the synthesis of a novel sulfur-rich macrocyclic (NS4) chelator that efficiently complexes 197gHg/197mHg (Randhawa et al. 2023) and its bifunctional analogue NS4-Tetrazine (NS4-Tz) that may be conjugated to mAbs via inverse electron demand Diels–Alder (IEDDA) “click” chemistry (Randhawa et al. 2026). We describe here for the first time complexation of 197gHg/197mHg to panitumumab via the bifunctional chelators (BFCs) NS4-Tz or commercially available 2-[4,7,10-tris(2-amino-2-oxoethyl)-1,4,7,10-tetrazacyclododec-1-yl]acetamide (TCMC) analogue p-SCN-Bn-TCMC. We further describe our finding that panitumumab can be directly labeled with 197gHg/197mHg without conjugation of a BFC through endogenous mercury binding sites. The cell binding, internalization and nuclear importation of the RICs were studied in EGFR-positive human BC cells and the radiation absorbed doses in the nucleus estimated. The induction of DNA DSB and cytotoxicity of the RICs in vitro on BC cells was studied. Finally, we compared the tumour and normal tissue uptake of the RICs in CD1 athymic (nude) mice with subcutaneous (s.c.) MDA-MB-468 tumours by SPECT/CT imaging and biodistribution studies. This work provides insight into the potential of 197gHg/197mHg-labeled panitumumab for theranostic SPECT/CT imaging and AE RIT of TNBC which overexpresses EGFR in 60–90% of cases (Changavi et al. 2015). Methods Cell culture and human BC xenografts MDA-MB-468, MDA-MB-231 and MCF-7 human BC cells expressing 1.3 × 106, 1.3 × 105, and 1.5 × 104 EGFR/cell, respectively (Reilly et al. 2000b) were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). All cells were cultured at 37 °C and 5% CO2 in Dulbecco’s Modified Eagle Medium (DMEM, Sigma-Aldrich, St. Louis, MO) supplemented with 10% fetal bovine serum (FBS, Gibco Thermo Fisher Scientific, Mississauga, ON, Canada) and 1% penicillin–streptomycin (Thermo Fisher Scientific). Production of 197gHg/197mHg 197gHg (t1/2 = 64.1 h) and 197mHg (t1/2 = 23.8 h) were co-produced by the 197Au(p,n)197gHg/197mHg reaction using an isotopically pure (100%) natural gold (197Au) target in a TR13 (13 MeV) cyclotron at TRIUMF (Vancouver, BC, Canada) (Randhawa et al. 2021). The target was dissolved in aqua regia and 197gHg/197mHg were transferred to 6 M HCl and separated from 197Au on an LN resin (Chen et al. 2023). The 6 M HCl was evaporated under vacuum and [197gHg/197mHg]HgCl2 recovered into Milli-Q® water (Millipore, Burlington, MA, USA). [197gHg/197mHg]HgCl2 was shipped from Vancouver to Toronto. At the time of receipt and conduct of experiments, there was < 10% 197mHg and the predominant radionuclide was 197gHg. The activity and concentration on receipt were approximately 115 MBq and 1,000 MBq/mL. Conjugation of panitumumab with BFC and labeling with 197gHg/197mHg Panitumumab was conjugated to TCMC by reaction with the benzylisothiocyanate ester (p-SCN-Bn-TCMC) as shown in Scheme 1 and described in the SI. Panitumumab was conjugated to a 1,4,7,10-tetrathia-13-azacyclopentadecane (NS4) BFC (Randhawa et al. 2026) via click chemistry using the inverse-electron-demand Diels–Alder (IEDDA) cycloaddition reaction between transcyclooctene (TCO)-modified or TCO-PEG4-panitumumab and a tetrazine group on the NS4 BFC (Scheme 2) as described in the SI. Panitumumab was labeled with 197gHg/197mHg after conjugation to TCMC or the NS4 BFC or labeled directly with 197gHg/197mHg (Scheme 1 or 2). Approximately 10 μg of TCMC-panitumumab (0.1 mg/mL in 1 M NaOAc buffer, pH 5.5), NS4-TCO-panitumumab or NS4-TCO-PEG4-panitumumab (0.1 mg/mL in 1 M NH4OAc, pH 7.0) were incubated with 2.5 MBq (5—10 μL) of [197gHg/197mHg]HgCl2 (TRIUMF) for 2 h at 42 °C. Alternatively, 10 μg of panitumumab (0.1 mg/mL in 1 M NaOAc buffer, pH 5.5 or 1 M NH4OAc, pH 7.0) was incubated for 2 h at 42 °C with [197gHg/197mHg]HgCl2 without conjugation to a BFC. The RICs were purified from free 197gHg/197mHg and exchanged into PBS, pH 7.4 by ultrafiltration on an Amicon Ultra 0.5 device (MWCO = 30 kDa; MilliporeSigma) centrifuged at 8,000 × g for 3 min, repeated 6 times. This method was validated to assure complete removal of all radiochemical impurities by loading [197gHg/197mHg]HgCl2 or [197gHg/197mHg]Hg-NS4 BFC in 1 M NaOAc buffer, pH 5.5 onto the device, buffer-exchanging into PBS, pH 6.0 and centrifuging at 8,000 × g for 3 min, repeated 6 times. The activity filtered by the device (MW < 30 kDa), remaining in the retentate (MW > 30 kDa) or adhering to the filter at each centrifugation step was measured in a radioisotope dose calibrator (CRC-15R, Capintec Inc., Ramsey, NJ, USA). The final radiochemical purity (RCP) of the RICs was further assessed by size-exclusion HPLC on a BioSep SEC-S4000 column eluted with 0.1 M NaH2PO4 at a flow rate of 0.65 mL/min with UV detection at 280 nm and radioactivity detection (FlowStar 2 LB 514). The percent of 197gHg/197mHg strongly-bound to panitumumab was determined by instant thin layer-silica gel chromatography (ITLC-SG) developed in 50 mM meso-2,3-dimercaptosuccinic acid (DMSA; Sigma-Aldrich, St. Louis, MO), pH 4.5. In this system, 197gHg/197mHg transchelated to DMSA migrates to the solvent front (Rf = 1.0), while 197gHg/197mHg strongly bound to panitumumab remains at the origin (Rf = 0.0). For ITLC-SG, a 5 μL aliquot of RICs was diluted with 5 μL of 50 mM DMSA and a 2 μL sample was spotted on the ITLC-SG strip and the strip developed in 50 mM DMSA The strip was cut into two sections at Rf = 0.5 and each section measured in a γ-counter (PerkinElmer Model 1480, Waltham, MA, USA). Radiolabeling of panitumumab with 197gHg/197mHg via a direct complexation through thiol groups or b complexation by TCMC BFC Radiolabeling of panitumumab with 197gHg /197mHg via a conjugation of NS4 BFC without or b with a PEG4 spacer Binding to EGFR-positive MDA-MB-468 cells Binding of [197gHg/197mHg]Hg-TCMC-panitumumab, [197gHg/197mHg]Hg-NS4-TCO-panitumumab, [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab or [197gHg/197mHg]Hg-panitumumab to MDA-MB-468 human TNBC cells (1.3 × 106 EGFR/cell) (Reilly et al. 2000b) was determined. Briefly, 1 × 106 MDA-MB-468 cells were suspended in 200 μL of growth medium in 1.5 mL Eppendorf tubes and incubated for 2.5 h at 4 °C with 30 nmoles/L of [197gHg/197mHg]Hg-TCMC-panitumumab or [197gHg/197mHg]Hg-panitumumab (all RICs: 0.05 MBq/μg; 7.5 × 109 MBq/mole) in the absence [total binding; TB] or presence [non-specific binding; NSB] of a 500-fold molar excess of panitumumab. Similarly, 40 nmoles/L of [197gHg/197mHg]Hg-NS4-TCO-panitumumab or [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab (all RICs: 0.05 MBq/μg; 7.5 × 109 MBq/mole) were incubated for 3 h at 4 °C with 1 × 106 MDA-MB-468 cells in the absence or presence of a 100-fold excess of panitumumab. The cells were then rinsed with 500 µL of PBS, pH 7.4 and centrifuged at 2,000 rpm for 5 min in an Eppendorf Model 5422 centrifuge. The supernatant was removed and the cells rinsed again with 500 µL of PBS, pH 7.4, then centrifuged at 2,000 rpm for 5 min, and the supernatant removed. Activity in the cell pellets was measured in a γ-counter (PerkinElmer Model 1480). The percent SB to EGFR on MDA-MB-468 cells was calculated by subtracting the NSB from TB. Subcellular distribution and cellular dosimetry The subcellular distribution of the RICs in MDA-MB-468, MDA-MB-231 and MCF-7 cells was determined as previously reported (Facca et al. 2022). Briefly, 1 × 105 cells in 1 mL of growth medium were seeded into wells in a 24-well plate and cultured overnight. The medium was removed and replaced with 200 μL of fresh medium containing 20 nmoles/L or 40 nmoles/L of [197gHg/197mHg]Hg-TCMC-panitumumab, [197gHg/197mHg]Hg-NS4-TCO-panitumumab, [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab or [197gHg/197mHg]Hg-panitumumab (all RICs: 0.05 MBq/μg; 7.5 × 109 MBq/mole) or an equivalent activity (59 kBq) of [197gHg/197mHg]HgCl2 then incubated for 16–22 h at 37 °C and 5% CO2. The non-cell bound activity was removed and the cells rinsed twice with 500 μL of PBS, pH 7.4. These rinses contained activity in the medium (M) and were collected and pooled. The cells were then rinsed with 200 μL of 0.2 M NaOAc with 0.5 M NaCl, pH 2.5 for 10 min on ice, then twice with cold PBS, pH 7.4 to displace activity on the cell surface which was collected and pooled. The cells were then lysed overnight with 200 μL of ice-cold Nuclei EZ lysis buffer (Sigma-Aldrich) at 4 °C and the lysate recovered into 1.5 mL tubes using cell scrapers (Sarstedt, Nümbrecht, Germany). The lysate was centrifuged at 5,000 × g for 5 min, then rinsed twice with 200 µL PBS, pH 7.4. The supernatant containing the cytoplasmic activity was collected and separated from the pellets containing nuclear activity. Activity in the medium, on the cell surface, in the cytoplasm or nucleus were measured in a γ-counter (PerkinElmer Model 1480). The radiation absorbed dose in the nucleus of BC cells incubated with the RICs or free [197gHg/197mHg]HgCl2 was estimated based on the subcellular distribution and monolayer S-values calculated using previously reported methods (Brown et al. 2025; Cai et al. 2023, 2017) and described in the SI. Only emissions from 197gHg were considered for dosimetry estimates since this was the predominant radionuclide in subcellular fractionation and clonogenic survival assays due to the faster decay of 197mHg than 197gHg and 91.4% of 197mHg decayed into 197gHg in transit from TRIUMF (Vancouver) to the University of Toronto. Clonogenic survival assays The clonogenic survival (CS) of MDA-MB-468, MDA-MB-231 and MCF-7 cells treated in vitro with the RICs was determined. Briefly, 1 × 105 cells in 1 mL of growth medium were seeded into wells in 24-well plates (Sarstedt, Nümbrecht, Germany) and cultured overnight. The medium was removed and the cells were then treated with 200 μL of fresh medium containing 29.4 or 58.8 kBq (20 or 40 nmoles/L) of [197gHg/197mHg]Hg-TCMC-panitumumab, [197gHg/197mHg]Hg-NS4-TCO-panitumumab, [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab or [197gHg/197mHg]Hg-panitumumab (all RICs: 0.05 MBq/μg; 7.5 × 109 MBq/mole) or an equivalent activity of [197gHg/197mHg]HgCl2 for 16–22 h at 37 °C and 5% CO2. Controls consisted of untreated cells or cells treated with unlabeled TCMC-panitumumab or NS4-TCO-panitumumab. To further determine the EGFR-specific cytotoxicity of [197Hg]Hg-panitumumab, EGFR on MDA-MB-468 cells were blocked by incubation for 16 h with a 100-fold excess of panitumumab prior to treatment with these RICs. The treatment medium was then removed and the cells detached by incubation with 0.25% trypsin–EDTA (Sigma-Aldrich, St. Louis, MO) for 5 min at 37 °C. Approximately 1 × 104 cells were then seeded in 3 mL of medium into wells in a 6-well plate (Sarstedt, Nümbrecht, Germany) and cultured at 37 °C and 5% CO2 for 7–14 d. The medium was removed and the surviving colonies stained with methylene blue in 1:1 water:ethanol and colonies with ≥ 50 cells counted. The plating efficiency (PE) was calculated by dividing the number of colonies by the number of cells seeded. The surviving fraction (SF) was calculated by dividing the PE for treated cells by that of untreated cells. DNA double-strand breaks (DSB) DNA DSB in BC cells treated with RICs were assessed by immunofluorescence microscopy for phosphorylated histone-2AX (γ-H2AX) (Cai et al. 2009). Briefly, 1 × 105 cells were exposed to 40 nmoles/L of, [197gHg/197mHg]Hg-NS4-TCO-panitumumab, [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab or [197gHg/197mHg]Hg-panitumumab (all RICs: 0.05 MBq/μg; 7.5 × 109 MBq/mole), unlabeled panitumumab or NS4-TCO-panitumumab or an equivalent activity of [197gHg/197mHg]HgCl2 or were untreated. Additionally, MDA-MB-468 cells were treated with [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab with prior blocking of EGFR with a 100-fold excess of panitumumab. Cells were then seeded in 1 mL of medium containing the RICs or control treatments onto glass coverslips (Ted Pella, Redding, CA, USA) in wells in a 24-well plate (Sarstedt, Nümbrecht, Germany) and incubated overnight at 37 °C and 5% CO2. The medium was removed and the cells were rinsed twice with 500 μL of PBS, pH 7.4. The cells were then immunostained for γ-H2AX as previously reported (Cai et al. 2009). The cover slips containing stained cells were sandwiched onto microscope slides with 12 µL of Vectashield® mounting medium containing 4',6-diamidino-2-phenylindole (DAPI) to counterstain the nucleus. The slides were imaged on a LSM 700 fluorescence microscope (Carl Zeiss MicroImaging, Inc., Thornwood, NY, USA) with a Plan-Apochromat 40 × /1.3 Oil DIC M27 objective. The integrated density of γH2AX foci per nucleus area was quantified for treated and control cells as previously reported (Cai et al. 2009). SPECT/CT imaging and biodistribution studies CD-1 athymic (nude) mice were inoculated subcutaneously (s.c.) with 107 MDA-MB-468 cells in PBS on the left flank (hind leg) to establish tumors. Tumors were allowed to grow to 5 mm in diameter (4 weeks). Mice were then injected intravenously (i.v.) in the lateral tail vein with 130 µL of PBS containing 60 µg (0.20–3.58 MBq) of [197gHg/197mHg]Hg-panitumumab (5.0 × 108 MBq/mole – 8.9 × 109 MBq/mole) alone (n = 4) or combined with a 10–30-fold excess (n = 3) of panitumumab to block EGFR, or with 60 µg (0.30–2.43 MBq) of [197gHg/197mHg]Hg-NS4-TCO-panitumumab (7.5 × 108 MBq/mole – 6.0 × 109 MBq/mole) alone (n = 3) or combined with a tenfold excess of panitumumab (n = 3) or free [197gHg/197mHg]HgCl2 (n = 4). At 1 d and 3 d post-injection (p.i.), mice were imaged by SPECT/CT on a trimodality (PET/SPECT/CT) system (Mediso) using previously reported parameters (Brown et al. 2025). At 3 d p.i., mice were sacrificed under aneasthesia and the tumour and samples of blood and normal tissues collected, weighed and measured in a γ-counter (Wizard 3; PerkinElmer, Waltham, MA). Uptake was expressed as percent injected dose per gram (%ID/g). All animal studies were conducted under a protocol (AUP 282.34) approved by the Animal Care Committee at the University Health Network following Canadian Council on Animal Care (CCAC) guidelines. Statistical analysis Data were presented as mean ± SD. Statistical significance were analyzed by Prism Ver. 10.4.0 software (GraphPad) using one-way ANOVA and unpaired t-tests (P < 0.05). Results Conjugation of panitumumab to TCMC or NS4 BFC The number of TCMC conjugated per panitumumab when reacted at a panitumumab concentration of 5-20 mg/mL in 0.1 M NaHCO3 buffer, pH 8.5 using a fivefold, tenfold, 20-fold or 40-fold excess of p-SCN-Bn-TCMC are shown in the Supplementary Information (Table S1). The conjugation conditions selected for subsequent experiments were reaction of panitumumab (10 mg/mL) with a tenfold molar excess of p-SCN-Bn-TCMC, which resulted in 2.3 ± 0.6 TCMC per panitumumab. The number of TCO groups introduced when panitumumab was reacted at a concentration of 30 mg/mL in PBS, pH 9.5 using a 50-fold, 75-fold or 100-fold excess of NHS-TCO or with a tenfold excess of NHS-PEG4-TCO are shown in Table S2. A 100-fold molar excess of TCO-NHS or tenfold excess of NHS-TCO-PEG4 were selected for reaction with tetrazine-NS4 BFC, resulting in 7.5 ± 1.1 TCO or 8.8 ± 0.2 TCO-PEG4 per panitumumab, respectively. Under these conditions, reaction of TCO-panitumumab with a 15-fold or 30-fold excess of tetrazine-NS4 resulted in 4.8 ± 1.4 or 4.1 ± 0.1 BFC per panitumumab, respectively while reaction of TCO-PEG4-panitumumab with a 15-fold excess of tetrazine-NS4 resulted in conjugation of 6.7 ± 1.0 BFC per panitumumab (Table S2). Labeling of panitumumab with 197gHg/197mHg Panitumumab was labeled with 197gHg/197mHg at a specific activity = 0.05–0.13 MBq/μg (molar activity: 7.5 × 109–1.9 × 1010 MBq/mole) after conjugation to TCMC or the NS4 BFC or directly labeled with 197gHg/197mHg, then purified by ultrafiltration on an Amicon Ultra 0.5 device (MWCO = 30 kDa). Validation studies showed that 76 ± 3% (mean ± SD; n = 3) of uncomplexed [197gHg/197mHg]HgCl2 was filtered by the Amicon device, while 24 ± 3% adhered to the filter, but no activity was present in the retentate. Unconjugated [197gHg/197mHg]Hg-NS4 BFC was almost completely adherent to the filter (99 ± 2%; n = 3) with < 1% filtered by the device or present in the retentate. In contrast, 36 ± 9%, 36 ± 1%, 25 ± 2% and 53 ± 5% of activity were recovered in the retentate for [197gHg/197mHg]Hg-TCMC-panitumumab, [197gHg/197mHg]Hg-NS4-TCO-panitumumab, [197gHg/197mHg]Hg-NS4-TCO-PEG4- and [197gHg/197mHg]Hg-panitumumab, respectively. Since the device completely removed all radiochemical impurities by filtration or adherence to the filter, the RICs recovered in the retentate were considered 100% radiochemically pure (RCP). RCP was confirmed by SE-HPLC (Suplementary Information, Fig. S2). The percent 197gHg/197mHg strongly bound to panitumumab in each RIC was determined by ITLC-SG developed in 50 mM DMSA (Table 1). Table 1Percent of 197gHg/197mHg strongly-bound to panitumumab aAgentStrongly-bound 197gHg/197mHg b[197gHg/197mHg]Hg-TCMC-panitumumab66.8 ± 2.5%[197gHg/197mHg]Hg-NS4-TCO-panitumumab52.8 ± 2.9%[197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab50.8 ± 4.3%[197gHg/197mHg]Hg-panitumumab61.9 ± 2.8%[197/gHg/197mHg]HgCl20.4 ± 0.2%aActivity remaining at the origin in ITLC-SG developed in 50 mM DMSAbMean + SD (n = 3) Percent of 197gHg/197mHg strongly-bound to panitumumab a aActivity remaining at the origin in ITLC-SG developed in 50 mM DMSA bMean + SD (n = 3) Binding to EGFR-positive MDA-MB-468 cells The TB to MDA-MB-468 cells decreased by tenfold when the RICs were co-incubated with a 50-fold excess of panitumumab, indicating that binding was EGFR-specific (Fig. 1). SB was determined by subtracting NSB, i.e. binding in the presence of an excess of panitumumab, from TB. The percent SB (mean ± SD) was 87.3 ± 0.7% (n = 3), 91.0 ± 2.0% (n = 3), 87.8 ± 0.2% (n = 3), and 89.0 ± 1.0% (n = 3) for incubation of MDA-MB-468 cells with [197gHg/197mHg]Hg-TCMC-panitumumab, [197gHg/197mHg]Hg-NS4-TCO-panitumumab, directly-labeled [197gHg/197mHg]Hg-panitumumab or [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab, respectively (all RICs: 0.05 MBq/μg; 7.5 × 109 MBq/mole). Binding to EGFR-positive MDA-MB-468 human BC cells of a [197gHg/197mHg]Hg-TCMC-panitumumab, b [197gHg/197mHg]Hg-panitumumab, c [197gHg/197mHg]Hg-NS4-TCO-panitumumab or d [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab in the absence (TB) or presence (NSB) of a 100–500-fold excess panitumumab. SB was obtained by subtracting NSB from TB. Values shown are mean ± SD (n = 3) Subcellular distribution and cellular dosimetry There was significantly greater activity on the cell surface, in the cytoplasm or in the nucleus of MDA-MB-468 cells (1.3 × 106 EGFR/cell) vs. MDA-MB-231 or MCF-7 cells (1.3 × 105 and 1.5 × 104 EGFR/cell, respectively) when incubated with 20 nmoles/L (29.4 kBq) of [197gHg/197mHg]Hg-TCMC-panitumumab (0.05 MBq/μg; 7.5 × 109 MBq/mole) for 16–22 h at 37 °C (Fig. 2a). Uptake in all subcellular compartments was significantly greater for MCF-7 cells than MDA-MB-231 cells incubated with [197gH/197mHg]Hg-TCMC-panitumumab. A similar subcellular distribution was observed for MDA-MB-468, MDA-MB-231 and MCF-7 cells incubated with [197gHg/197mHg]Hg-panitumumab (0.05 MBq/μg; 7.5 × 109 MBq/mole; Fig. 2b). Incubation with 40 nmoles/L (58.8 kBq) of [197gHg/197mHg]Hg-NS4-TCO-panitumumab (Fig. 2c) or [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab (Fig. 2d) (0.05 MBq/μg; 7.5 × 109 MBq/mole) revealed significantly higher activity on the cell surface for MDA-MB-468 than MDA-MB-231 or MCF-7 cells. Cytoplasmic and nucleus activity for [197gHg/197mHg]Hg-NS4-TCO-panitumumab were significantly greater for MDA-MB-468 cells than MDA-MB-231 cells, but not significantly different than MCF-7 cells (Fig. 2c). However, nucleus activity was significantly higher for MCF-7 cells than MDA-MB-231 cells. Cytoplasmic and nucleus activity for [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab were significantly greater for MCF-7 than MDA-MB-231 cells. Incubation of MDA-MB-468, MDA-MB-231 or MCF-7 cells with [197gHg/197mHg]HgCl2 (58.8 kBq) (Fig. 2e) revealed a very different subcellular distribution than the RICs. In particular, there was higher cell surface activity on MCF-7 cells than MDA-MB-231 or MDA-MB-468 cells and there was no relationship between EGFR expression of BC cells and cell surface activity. Further, cytoplasmic and nucleus activity in MCF-7 cells incubated with [197gHg/197mHg]HgCl2 were high and significantly greater than MDA-MB-468 or MDA-MB-231 cells. Activity on the surface, internalized into the cytoplasm or in the nucleus of MDA-MB-468, MDA-MB-231 or MCF-7 human BC cells with high, moderate or low EGFR expression after incubation for 16–22 h at 37 °C with 20 nmoles/L (29.4 kBq) of a [197gHg/197mHg]Hg-TCMC-panitumumab or b directly labeled [197gHg/197mHg]Hg-panitumumab, or 40 nmoles/L (58.8 kBq) of c [197gHg/197mHg]Hg-NS4-TCO-panitumumab or d [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab or e [197gHg/197mHg]HgCl2 (58.8 kBq). Values shown are mean ± SD (n = 3). Statistical significance: ***P < 0.001; **P < 0.01; *P < 0.05; Not significant: ns. Note different y-axis scale for panel e than other panels Based on the time-integrated subcellular distribution of activity in MDA-MB-468, MDA-MB-231 or MCF-7 cells incubated with the RICs or [197gHg/197mHg]HgCl2 (Supplementary Information, Table S3) the absorbed doses in the nucleus were estimated (Table 2). Only emissions from 197gHg were considered to estimate doses as this was the predominant radionuclide due to the decay of 197mHg in transit from TRIUMF (Vancouver) to the University of Toronto (< 10% 197mHg). These doses ranged from 0.08 to 0.10 Gy/kBq for MDA-MB-468 cells, 0.01 to 0.02 Gy/kBq for MDA-MB-231 cells and 0.03 to 0.05 Gy/kBq for MCF-7 cells treated with the RICs. The doses in the nucleus of MDA-MB-468, MDA-MB-231 or MCF-7 cells treated with [197gHg/197mHg]HgCl2 were 0.053, 0.040 and 0.174 Gy/Bq, respectively. Activity in the nucleus was the main contributor to the total absorbed doses in the nucleus for all cells incubated with RICs or [197gHg/197mHg]HgCl2. Table 2Absorbed doses (Gy) in the nucleus of EGFR-positive human BC cellsAmountaTotal absorbed dose (Gy) in the nucleus(% contribution of nucleus activity to total nucleus dose)b[197gHg]Hg AgentcMDA-MB-468cMDA-MB-231cMCF-7c[197gHg]Hg-TCMC-Panitumumabd20 nmols (29.4 kBq)3.0 ± 0.5 (51%)0.41 ± 0.02 (62%)1.1 ± 0.3 (57%)[197gHg]Hg-Panitumumabd20 nmols (29.4 kBq)2.9 ± 0.3 (57%)0.3 ± 0.1 (54%)1.5 ± 0.2 (66%)[197gHg]Hg-NS4-TCO-Panitumumabe40 nmols (58.8 kBq)4.7 ± 0.1 (70%)1.4 ± 0.3 (78%)2.9 ± 0.6 (84%)[197gHg]Hg-NS4-TCO-PEG4-Panitumumabe40nmols (58.8 kBq)4.8 ± 0.4 (68%)1.2 ± 0.2 (86%)1.9 ± 0.1 (84%)[197gHg]HgCl2fTrace (58.8 kBq)3.1 ± 0.4 (77%)2.3 ± 0.3 (89%)10.2 ± 0.9 (86%)aIncubated for 19 h with 1 × 105 cells cultured overnight in wells in a 24-well plate. Only the dose from 197gHg was considered due to < 10% of 197mHg remaining due to faster decay of 197mHg than 197gHg in transit from TRIUMF to the University of TorontobDoses were estimated by S-values calculated using MCNP6 and time-integrated activities as described in the SIcFor each 197gHg agent, the doses in the nucleus were significantly different for all 3 cell lines (P < 0.05)dFor the same cell line, the dose from [197gHg]Hg-TCMC-Panitumumab was not significantly different than [197/gHg]Hg-Panitumumab (P > 0.1)eFor MDA-MB-468 or MDA-MB-231 cells, the dose from [197gHg]Hg-NS4-TCO-Panitumumab was not significantly different from [197gHg]Hg-NS4-TCO-PEG4-Panitumumab (P > 0.5). For MCF-7 cells, the dose from [197gHg]Hg-NS4-TCO-Panitumumab was significantly higher than [197gHg]Hg-NS4-TCO-PEG4-Panitumumab (P = 0.048)fFor the same cell line, the dose from [197/gHg]HgCl2 was significantly different than [197gHg]Hg-NS4-TCO-Panitumumab or [197gHg]Hg-NS4-TCO-PEG4-Panitumumab (P < 0.02) Absorbed doses (Gy) in the nucleus of EGFR-positive human BC cells aIncubated for 19 h with 1 × 105 cells cultured overnight in wells in a 24-well plate. Only the dose from 197gHg was considered due to < 10% of 197mHg remaining due to faster decay of 197mHg than 197gHg in transit from TRIUMF to the University of Toronto bDoses were estimated by S-values calculated using MCNP6 and time-integrated activities as described in the SI cFor each 197gHg agent, the doses in the nucleus were significantly different for all 3 cell lines (P < 0.05) dFor the same cell line, the dose from [197gHg]Hg-TCMC-Panitumumab was not significantly different than [197/gHg]Hg-Panitumumab (P > 0.1) eFor MDA-MB-468 or MDA-MB-231 cells, the dose from [197gHg]Hg-NS4-TCO-Panitumumab was not significantly different from [197gHg]Hg-NS4-TCO-PEG4-Panitumumab (P > 0.5). For MCF-7 cells, the dose from [197gHg]Hg-NS4-TCO-Panitumumab was significantly higher than [197gHg]Hg-NS4-TCO-PEG4-Panitumumab (P = 0.048) fFor the same cell line, the dose from [197/gHg]HgCl2 was significantly different than [197gHg]Hg-NS4-TCO-Panitumumab or [197gHg]Hg-NS4-TCO-PEG4-Panitumumab (P < 0.02) Clonogenic survival assays Treatment of MDA-MB-468 cells (1.3 × 106 EGFR/cell) with 20 nmoles/L of [197gHg/197mHg]Hg-TCMC-panitumumab or [197gHg/197mHg]Hg-panitumumab (all RICs: 0.05 MBq/μg; 7.5 × 109 MBq/mole) decreased the SF to 0.05 ± 0.04 and 0.02 ± 0.02, respectively, while exposure to 20 nmoles/L of unlabeled panitumumab resulted in a SF of 0.94 ± 0.11 (P < 0.001; Fig. 3a). Blocking EGFR with a 100-fold excess of panitumumab prior to treatment with [197gHg/197mHg]Hg-panitumumab significantly increased the SF of MDA-MB-468 cells to 0.57 ± 0.21 (P < 0.001), demonstrating that cytotoxicity was EGFR-dependent. [197gHg/197mHg]Hg-TCMC-panitumumab and [197gHg/197mHg]Hg-panitumumab were less cytotoxic to MDA-MB-231 cells (1.3 × 105 EGFR/cell) only slightly decreasing their SF to 0.91 ± 0.10 and 0.96 ± 0.19, respectively (P > 0.05; Fig. 3b). Compared to unlabeled panitumumab (SF = 1.14 ± 0.21) only MDA-MB-231 cells treated with [197gHg/197mHg]Hg-TCMC-panitumumab showed a significantly decreased SF (P < 0.05). [197gHg/197mHg]Hg-TCMC-panitumumab and [197gHg/197mHg]Hg-panitumumab were less cytotoxic to MCF-7 cells than MDA-MB-468 cells, decreasing their SF to 0.72 ± 013 and 0.79 ± 0.38, respectively (Fig. 3c). [197gHg/197mHg]Hg-TCMC-panitumumab was not more cytotoxic to MCF-7 cells than [197gHg/197mHg]Hg-panitumumab (P > 0.05). Compared to unlabeled panitumumab (SF = 1.09 ± 0.37) only [197gHg/197mHg]Hg-TCMC-panitumumab significantly decreased the SF of MCF-7 cells (P < 0.05). Clonogenic survival of a MDA-MB-468 cells, b MDA-MB-231 cells or c MCF-7 cells treated with 20 nmoles/L (0.05 MBq/μg; 7.5 × 109 MBq/mole) of [197gHg/197mHg]Hg-TCMC-panitumumab, [197gHg/197mHg]Hg-panitumumab, unlabeled panitumumab (20 nmoles/L). For one of the treatment arms, EGFR were blocked on MDA-MB-468 cells with a 100-fold excess of panitumumab prior to treatment with [197gHg/197mHg]Hg-panitumumab. Values shown are the mean ± SD (n = 9). Significant differences: ***P < 0.001;**P < 0.01; *P < 0.05.; Not significant: ns Treatment of MDA-MB-468 cells with 40 nmoles/L of [197gHg/197mHg]Hg-NS4-TCO-panitumumab or [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab (all RICs: 0.05 MBq/μg; 7.5 × 109 MBq/mole) decreased the SF to 0.05 ± 0.031 and 0.03 ± 0.02, respectively, which was significantly lower than the SF of these cells exposed to 40 nmoles/L of unlabeled NS4-TCO-panitumumab which was 1.39 ± 0.38 (P < 0.001; Fig. 4a). [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab was more cytotoxic than [197gHg/197mHg]Hg-NS4-TCO-panitumumab to MDA-MB-468 cells (P < 0.01). Incubation of MDA-MB-468 cells with an equivalent activity of [197gHg/197mHg]HgCl2 decreased the SF to 0.22 ± 0.10, but this was significantly greater than cells treated with [197gHg/197mHg]Hg-NS4-TCO-panitumumab or [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab (P < 0.001). [197Hg]Hg-NS4-TCO-panitumumab or [197Hg]Hg-NS4-TCO-PEG4-panitumumab were less cytotoxic to MDA-MB-231 cells than MDA-MB-468 cells, decreasing their SF to 0.62 ± 0.09 and 0.76 ± 0.07, respectively (Fig. 4b). Compared to unlabeled NS4-TCO-panitumumab (SF = 0.84 ± 0.09), only [197gHg/197mHg]Hg-NS4-TCO-panitumumab significantly decreased the SF of MDA-MB-231 cells (P < 0.01). [197gHg/197mHg]HgCl2 was significantly more cytotoxic to MDA-MB-231 cells than [197gHg/197mHg]Hg-NS4-TCO-panitumumab (SF = 0.54 ± 0.08; P < 0.05) or [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab (P < 0.001). [197gHg/197mHg]Hg-NS4-TCO-panitumumab or [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab were less cytotoxic to MCF-7 cells than MDA-MB-468 cells, decreasing their SF to 0.48 ± 0.26 and 0.61 ± 0.15, respectively (Fig. 4c). Compared to unlabeled NS4-panitumumab (SF = 0.90 ± 0.10) only [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab significantly decreased the SF of MCF-7 cells (P < 0.05). [197gHg/197mHg]Hg-NS4-TCO-panitumumab non-significantly decreased the SF of MCF-7 cells (P = 0.06). [197gHg/197mHg]HgCl2 was the most cytotoxic agent on MCF-7 cells, decreasing the SF to 0.10 ± 0.08. There was no correlation between the cytotoxicity of [197gHg/197mHg]HgCl2 and EGFR expression for any of the three BC cell types. Clonogenic survival of a MDA-MB-468 cells, b MDA-MB-231 cells or c MCF-7 cells treated with 40 nmoles/L (0.05 MBq/μg; 7.5 × 109 MBq/mole) of [197gHg/197mHg]Hg-NS4-TCO-panitumumab, [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab, unlabeled NS4-TCO-panitmumab (40 nmoles/L) or an equivalent activity of [197gHg]HgCl2. Values shown are mean ± SD (n = 3–9). Significant differences: ***P < 0.001;**P < 0.01; *P < 0.05. Not significant: ns DNA double-strand breaks (DSB) Treatment of MDA-MB-468 cells with 40 nmoles/L of [197gHg/197mHg]Hg-panitumumab (0.05 MBq/μg; 7.5 × 109 MBq/mole) significantly increased γ-H2AX foci representing sites of DNA DSB in the nucleus, compared to untreated cells (Fig. 5) but blocking EGFR with excess panitumumab decreased γ-H2AX foci (P = 0.06). This suggests that DNA damage caused by the RICs was EGFR-dependent. There were no significant differences between the density of γ-H2AX foci in the nucleus of MDA-MB-468 cells in which EGFR were blocked treated with [197gHg/197mHg]Hg-panitumumab compared to cells treated with unlabeled panitumumab or untreated cells. Immunofluorescence images and quantification of γ-H2AX foci in MDA-MB-231 and MCF-7 cells treated with [197gHg/197mHg]Hg-panitumumab, unlabeled panitumumab or untreated cells are shown in the SI (Fig. S3 and S4). For these cells, there were no significant differences in the density of γ-H2AX foci after treatment with RICs or unlabeled panitumumab compared to untreated cells. Treatment of MDA-MB-468 cells with 40 nmoles/L of [197gHg/197mHg]Hg-NS4-TCO-panitumumab or [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab (all RICs: 0.05 MBq/μg; 7.5 × 109 MBq/mole) increased γ-H2AX foci compared to cells treated with unlabeled NS4-TCO-panitumumab or untreated cells (Fig. 6). Images and quantification of γ-H2AX foci for MDA-MB-231 or MCF-7 cells treated with these RICs are shown in the Supporting Information (Fig. S5 and S6). There were no significant differences in γ-H2AX foci for MCF-7 cells treated with these RICs, unlabeled NS4-TCO-panitumumab or [197gHg/197mHg]Cl2 compared to untreated cells. a Immunofluorescence microscopy for γ-H2AX in MDA-MB-468 cells treated with 40 nmoles/L (0.05 MBq/μg; 7.5 × 109 MBq/mole) [197gHg/197mHg]Hg-panitumumab, [197gHg/197mHg]Hg-panitumumab with pre-incubation with a tenfold excess of panitumumab, unlabeled panitumumab or in untreated cells. Images were adjusted to the same intensity. b Integrated density of γ-H2AX foci in the nucleus. Values shown are the mean ± SD (n = 5). Significant differences: *P < 0.05 a Immunofluorescence microscopy for γ-H2AX (left panels) in MDA-MB-468 cells treated with 40 nmoles/L (0.05 MBq/μg; 7.5 × 109 MBq/mole) of [197gHg/197mHg]Hg-NS4-TCO-panitumumab, [197gHg/197mHg]Hg-NS4-TCO-PEG4-panitumumab, unlabeled NS4-TCO-panitumumab, [197gHg/197mHg]HgCl2 or in untreated cells. Images were adjusted to the same intensity. Right panels show the nucleus stained with DAPI. b Integrated density of γ-H2AX foci in the nucleus. Values shown are mean ± SD (n = 5). For cells treated with unlabeled NS4-panitumumab, the integrated density of γ-H2AX foci was < 1. Significant differences: *P < 0.05 SPECT/CT imaging and biodistribution studies SPECT/CT images of CD1 athymic (nude) mice with s.c. MDA-MB-468 human BC xenografts (Fig. 7) injected i.v. (tail vein) with 60 μg (0.20–3.58 MBq) of [197gHg/197mHg]Hg-panitumumab (5.0 × 108 MBq/mole – 8.9 × 109 MBq/mole) or 60 μg (0.30–2.43 MBq) of [197gHg/197mHg]Hg-NS4-panitumumab (7.5 × 108 MBq/mole – 6.0 × 109 MBq/mole) revealed tumour uptake at 1 d p.i. and 3 d p.i. Blocking EGFR by co-administration of a 10–30 fold excess of panitumumab appeared to decrease tumour uptake on the images at 1 d p.i. and especially at 3 d p.i. of [197gHg/197mHg]Hg-panitumumab. There was high kidney uptake of all RICs on the images at 1 d and 3 d p.i. In contrast, mice injected with [197gHg/197mHg]HgCl2 did not show tumour uptake at 1 d or 3 d p.i.. The kidneys were most prominent on the images of mice injected with [197gHg/197mHg]HgCl2. The tumour and kidney uptake of [197Hg]Hg-panitumumab or [197Hg]Hg-NS4-panitumumab or [197Hg]HgCl2 at 3 d p.i. in biodistribution studies are shown in Fig. 8. Complete biodistribution results are shown in the Supplementary Information (Fig. S7). Tumour uptake of [197gHg/197mHg]Hg-panitumumab or [197gHg/197mHg]NS4-panitumumab were significantly greater than [197gHg/197mHg]HgCl2 (P < 0.01 and P < 0.0001, respectively) and appeared to be decreased by blocking EGFR by co-administering a 10–30-fold excess of panitumumab (P = 0.082 and P = 0.055, respectively; Fig. 3a). Kidney uptake was highest for mice injected with [197gHg/197mHg]HgCl2 (Fig. 3b). Kidney uptake of [197gHg/197mHg]Hg-panitumumab was significantly lower than [197gHg/197mHg]HgCl2 (P = 0.02).. Blocking EGFR did not significantly decrease the kidney uptake of [197gHg/197mHg]Hg-panitumumab (P = 0.06) or [197gHg/197mHg]Hg-NS4-panitumumab (P = 0.224). μ a Tumour uptake and b kidney uptake at 3 d p.i. of [197gHg/197mHg]Hg-panitumumab or [197gHg/197mHg]NS4-panitumumab alone or combined with a 10–30 fold excess of panitumumab to block EGFR or [197gHg/197mHg]HgCl2 in mice with s.c. MDA-MB-468 human TNBC tumours (n = 3–4). Significant differences: *P < 0.05, ***P < 0.001. Not significant: ns or P-values shown Complete biodistribution in all tissues is shown in the SI (Fig. S7) Discussion In this study, we constructed for the first time RICs composed of panitumumab labeled with the AE-emitters, 197gHg/197mHg and determined their cytotoxicity in vitro on a panel of human BC cells and their tumour and normal tissue uptake in vivo in CD1 athymic (nude) mice with s.c. EGFR-overexpressing MDA-MB-468 human BC xenografts. Labeling of panitumumab with 197gHg/197mHg was achieved after conjugation to TCMC or a NS4 BFC or by direct labeling of endogenous mercury binding sites on panitumumab. Panitumumab was conjugated to TCMC by reaction of p-SCN-Bn-TCMC with the ε-amino group on lysines which forms a thiourea bond. Randhawa et al. (Randhawa et al. 2024) reported that p-SCN-Bn-TCMC efficiently complexed 197gHg/197mHg and this radiometal complex was stable in vitro in serum and to challenge with glutathione (GSH). Alternatively, panitumumab was conjugated to the tetrazine-functionalized NS4 BFC (Randhawa et al. 2023) via an IEDDA cycloaddition (click chemistry) reaction with TCO-modified panitumumab. Randhawa et al. (Randhawa et al. 2026) previously reported conjugation of the tetrazine-NS4 BFC to the anti-human epidermal growth factor receptor-2 (HER2) mAb, trastuzumab and labeling of these immunoconjugates with 197gHg/197mHg. In an unconjugated form, the NS4 BFC efficiently complexed 197gHg/197mHg with 87–99% incorporation obtained after incubation at 37 °C for 1 h, depending on the BFC concentration (10–4 to 10–6 M). The complex was stable in vitro to challenge with glutathione (GSH) and in serum. Similar to our study, the tetrazine-NS4 BFC was conjugated via an IEDDA cycloaddition reaction with TCO-modified trastuzumab. Randhawa et al. measured the labeling efficiency (LE) of the NS4-trastuzumab conjugates by ITLC-SG in 50 mM DMSA. The LE reported was 47 ± 17%, but an almost identical LE (48 ± 10%) was obtained for trastuzumab not conjugated to the BFC. However, here we interpret the results of ITLC-SG in 50 mM DMSA as a measure of the strength of binding of 197gHg/197mHg rather than LE, since DMSA is a strong mercury chelator (George et al. 2004). Transchelation of 197gHg/197mHg from panitumumab to DMSA causes migration of activity to the solvent front on the ITLC-SG strip, which we believe may be incorrectly interpreted as uncomplexed 197gHg/197mHg. Only 0.4% of [197gHg/197mHg]Cl2 mixed with DMSA remains at the origin of the ITLC-SG strip. Similar to Rhandhawa et al. (Randhawa et al. 2026), we found that panitumumab could be labeled with 197gHg/197mHg without conjugation to a BFC. Moreover, both panitumumab labeled with 197gHg/197mHg after BFC conjugation or labeled directly with 197gHg/197mHg showed a similar proportion of strongly-bound 197gHg/197mHg determined by ITLC-SG in 50 mM DMSA, ranging from 51 to 67%. Direct complexation of 197gHg/197mHg by antibodies may be explained by endogenous mercury binding sites analogous to those present in albumin which binds mercury through cysteine or cystine residues (Lau and Sarkar 1979; Song et al. 2021). The competition between these endogenous mercury binding sites and the BFC introduced into the antibodies poses a challenge to labeling with 197gHg/197mHg, potentially resulting in complexation to sites that have different metal-binding affinities. Moreover, the results of ITLC-SG in 50 mM DMSA demonstrated that panitumumab labeled directly with 197gHg/197mHg exhibited two populations of radiometal binding sites—one that is more labile and displaceable by DMSA and one that is stable and not displaceable by DMSA. Cysteine binds Hg2+ very strongly with a reported stability constant (LogKML) = 40 (Stary and Kratzer 1988), while amino acids such as glycine, histidine and aspartic acid bind Hg2+ more weakly (logKML = 13.9, 15.7 and 14.9, respectively) (Foti et al. 2009). The logKML for binding of Hg2+ to the NS4 BFC is ≥ 20 (Randhawa et al. 2023). The stability constant for binding of Hg2+ to TCMC is not known. It is possible that strongly-bound 197gHg/197mHg may be complexed either to the BFC or to cysteine amino acids in panitumumab, while 197gHg/197mHg that is weakly bound may be complexed by glycine, histidine or aspartic acid amino acids or other endogenous mercury binding sites. To our knowledge, our study is the first to report direct binding of 197gHg/197mHg to panitumumab and the first to identify that there are two populations of direct mercury binding sites—one population that is stable and one that is more labile. Both TCMC or NS4 RICs labeled with 197gHg/197mHg or panitumumab directly labeled with 197gHg/197mHg exhibited high specific binding (87–91%) to EGFR on MDA-MB-468 cells. Further, all RICs showed cell-surface binding to MDA-MB-468, MDA-MB-231 and MCF-7 cells that was correlated with their EGFR expression. In addition, the RICs were internalized and imported into the nucleus of BC cells. Binding of RICs to the EGFR internalizes and routes them to the cell nucleus due to a NLS in the transmembrane domain of the EGFR (Hsu and Hung 2007). This NLS has been shown to mediate the nuclear uptake of [111In]In-DTPA-EGF (Reilly et al. 2000a) and [111In]In-DOTA-panitumumab (Facca et al. 2022) in EGFR-overexpressing human BC cells. In contrast, the binding, internalization and nuclear uptake of [197gHg/197mHg]HgCl2 not conjugated to panitumumab were not correlated with the EGFR expression of BC cells. The highest nuclear uptake of [197gHg/197mHg]HgCl2 was found in MCF-7 cells which have the lowest EGFR expression. Nuclear uptake of [197gHg/197mHg]HgCl2 in MCF-7 cells may be due to passive diffusion into these cells (Maniero et al. 2018) and subsequent binding to estrogen receptors (ER) in the cytoplasm since mercury is proposed to act as a metalloestrogen which translocates mercury to the cell nucleus (Martin et al. 2003). We found lower uptake of [197gHg/197mHg]HgCl2 in the cytoplasm and nucleus of MDA-MB-231 cells that are ER-negative, but uptake of non-radioactive mercury into these subcellular compartments has been reported for MDA-MB-231 and MCF-7 cells (Maniero et al. 2018). The results of subcellular fractionation studies were used to determine the time-integrated activity in subcellular compartments and estimate the absorbed doses in the nucleus of BC cells from the RICs or [197gHg/197mHg]HgCl2 by applying the monolayer S-values calculated by MCNP6 software following our previously reported method (Brown et al. 2025; Cai et al. 2023, 2017). Only emissions from 197gHg were considered for dosimetry estimates as this was the predominant radionuclide in subcellular fractionation and clonogenic survival assays due to the decay of 197mHg in transit from TRIUMF (Vancouver) to the University of Toronto (<10% 197mHg). Due to the subcellular range of AE emitted by 197gHg, the self-dose S-values of nucleus-to-nucleus (N → N) were much higher than the S-values for cell surface-to-nucleus (CS → N) or cytoplasm-to-nucleus (Cy → N), thus 197gHg in the nucleus was the major contributor to the dose in the nucleus for all cells and RICs or [197gHg/197mHg]HgCl2. The absorbed dose in the nucleus for MDA-MB-468 cells with high EGFR expression incubated with RICs was greater than MDA-MB-231 or MCF-7 cells with moderate or low EGFR expression, respectively. However, there was no correlation between the dose in the nucleus for cells incubated with [197gHg/197mHg]HgCl2 and the EGFR expression of BC cells. The decrease in SF of BC cells exposed to the RICs or [197gHg/197mHg]HgCl2 in clonogenic survival assays and density of DNA DSBs measured by immunofluorescence for γ-H2AX agreed with the estimated absorbed doses in the nucleus. Unlabeled panitumumab did not decrease the SF, indicating that the AE emissions of 197gHg/197mHg were required for cytotoxicity. The RICs were most cytotoxic to MDA-MB-468 cells with high EGFR expression and these cells exhibited the greatest density of DNA DSBs, while RICs were much less cytotoxic and caused fewer DNA DSBs in MDA-MB-231 and MCF-7 cells that have moderate or low EGFR expression, respectively. The low cytotoxicity of RICs on MDA-MB-231 cells may be due to the low SA used for these assays (0.05 MBq/μg; 7.5 × 109 MBq/mole) combined with the radioresistance of these cells (Cai et al. 2008). We previously reported higher cytotoxicity for 20 nmols/L of [111In]In-DOTA-panitumumab (SA = 1.7 MBq/mg; 3.4 × 1010 MBq/mole) incubated with MDA-MB-231 cells (SF = 0.6) than here for 197gHg/197mHg-labeled RICs (SF = 0.9–0.96) (Facca et al. 2022). However, incubation of MDA-MB-231 cells with 40 nmoles/L of 197gHg/197mHg-labeled RICs decreased their SF further to 0.6–0.8. Blocking EGFR with panitumumab increased the SF of MDA-MB-468 cells treated with [197gHg/197mHg]Hg-panitumumab and decreased the density of DNA DSBs demonstrating that cytotoxicity and DNA damage were EGFR-dependent. [197gHg/197mHg]HgCl2 was cytotoxic to BC cells which was correlated with nuclear uptake of activity and absorbed dose in the nucleus, with MCF-7 cells exhibiting the greatest decrease in SF followed by MDA-MB-231 then MDA-MB-468 cells. Radiation resistance of MDA-MB-231 cells compared to MDA-MB-468 cells may also explain the higher SF of these cells (Cai et al. 2008). Since DNA DSB were lower in MCF-7 cells treated with [197gHg/197mHg]HgCl2 than in MDA-MB-468 cells but the SF was lower for MCF-7 cells, the chemical toxicity of mercury could also contribute to the cytotoxicity on MCF-7 cells. SPECT/CT imaging at 1 or 3 d p.i of [197gHg/197mHg]-NS4-panitumumab or [197gHg/197mHg]Hg-panitumumab revealed tumour uptake in CD1 athymic (nude) mice with s.c. EGFR-overexpressing MDA-MB-468 BC xenografts which decreased by co-administration of excess panitumumab to block EGFR. In contrast, almost no tumour uptake was seen on images of mice injected with [197gHg/197mHg]HgCl2. However, there was high kidney uptake in mice receiving RICs and especially in mice injected with [197gHg/197mHg]HgCl2. Biodistribution studies confirmed these findings. Our results were similar to those reported for mice with s.c. HER2-positive SK-OV-3 xenografts injected with [197gHg/197mHg]Hg-NS4-TCO-trastuzumab (Randhawa et al. 2026). Kidney uptake could be explained by in vivo release of 197gHg/197mHg from the RICs. Mercury accumulates in the kidneys (Park and Zheng 2012). In the plasma, mercury binds to albumin which is taken up by the liver and transchelated to glutathione which re-enters the circulation and is subsequently secreted by the renal tubules (Barfuss et al. 2022). Since the NS4 chelator forms complexes with 197gHg/197mHg that are stable in serum in vitro, kidney uptake of 197gHg/197mHg-labeled NS4 panitumumab RICs or NS4-trastuzumab RICs (Randhawa et al. 2026) may be due to release of some 197gHg/197mHg from endogenous mercury binding sites. A 2-step labeling method in which NS4 BFC is first complexed to 197gHg/197mHg and subsequently conjugated to mAbs may provide more stable complexation by avoiding binding to endogenous mercury binding sites, decreasing kidney uptake. However, preliminary studies of this 2-step method for labeling panitumumab with 197gHg/197mHg yielded > twofold lower SA (< 0.02 MBq/μg) than a 1-step labeling method which was not feasible for cytotoxicity and in vivo imaging or biodistribution studies. Further studies will be required to optimize this 2-step approach for labeling panitumumab with 197gHg/197mHg. Conclusions Panitumumab was conjugated to TCMC or a NS4 BFC for complexing 197gHg/197mHg or labeled directly with 197gHg/197mHg through endogenous mercury binding sites. The RICs bound specifically to EGFR on MDA-MB-468 human BC cells. The RICs were internalized and routed to the nucleus of MDA-MB-468, MDA-MB-231 or MCF-7 human BC cells. Emission of AE by 197gHg/197mHg caused DNA DSB that decreased the clonogenic survival of BC cells, especially MDA-MB-468 cells which exhibit high EGFR expression. The RICs localized in s.c. MDA-MB-468 human BC xenografts in nude mice but there was high kidney uptake that may be due to in vivo release of 197gHg/197mHg from the RICs. Future studies could explore a 2-step approach to labeling panitumumab with 197gHg/197mHg to avoid labeling endogenous mercury binding sites. Electronic Supplementary Material Below is the link to the electronic supplementary material. Supplementary Material Acknowledgements The authors appreciate the contribution to this study by the TRIUMF TR13 cyclotron team (David Provost, Toni Epp, Ryley Morgan and Spencer Staiger) and the TRIUMF Life Sciences production team (Dr. Qing Miao) for the production and separation of 197gHg/197mHg. The authors also acknowledge the assistance with SPECT/CT imaging provided at the STTARR Innovation Centre at the University Health Network. We also acknowledge the Centre for Pharmaceutical Oncology instrument facility at the Leslie Dan Faculty of Pharmacy, University of Toronto and the expert assistance provided by Dr. Azza Al-Mahrouki. Abbreviations Auger electrons Analysis of variance Breast cancer Bifunctional chelator Canadian council on animal care Clonogenic survival Meso-2,3-dimercaptosuccinic acid Dimethyl sulfoxide Deoxyribonucleic acids DNA double-stranded breaks Epidermal growth factor receptor Estrogen receptor Fetal bovine serum Human epidermal growth factor receptor 2 Inverse electron demand diels–alder Instant thin layer-silica gel chromatography Labeling efficiency Linear energy transfer Monte carlo N-particle Mammary fat pad Nuclear translocation sequence Sodium acetate Ammonium acetate Non-specific binding NS4-tetrazine Phosphate buffered saline Plating efficiency Polyethylene glycol Radiochemical purity Retention factor Radioimmunoconjugates Radioimmunotherapy Reactive oxygen species Specific activity Specific binding Standard devivation Single photon emission computed tomography/Computed tomography Survival fraction Total binding 2-[4,7,10-Tris(2-amino-2-oxoethyl)-1,4,7,10-tetrazacyclododec-1—yl]acetamide Transcyclooctene X-ray irradiation Author contributions ACKC.: Investigation, methodology, formal analysis, writing—original draft, writing—review and editing. FH and CC.: Investigation, methodology, formal analysis, writing—review and editing. ZC: Conceptualization, investigation, methodology, formal analysis, data curation, writing—original draft, writing—review and editing. YG-S.: Resources, writing—review and editing. PR: Resources, writing—review and editing. CFR.: Resources, supervision, writing—review and editing. VR: Resources, funding acquisition, writing—review and editing. RMR: Conceptualizaton, formal analysis, supervision, writing—review and editing, funding acquisition, project administration. Funding Supported by a grant (178266) from the Canadian Institutes of Health Research to VR and RMR and a sponsored research agreement between Orano Support and the University of Toronto to RMR. The production of 197gHg/197mHg was supported by a grant (NFRFT-2022–00269) from the New Frontiers in Research Fund-Transformation (NFRF-T) program to RMR, CFR, and VR. ACKC was a scholar of the Strategic Training in Transdisciplinary Radiation Science for the 21st Century (STARS21) program at the University Health Network and received an MDS Nordion Scholarship in Radiopharmaceutical Sciences from the University of Toronto. Data availability All data generated or analyzed during this study are included in this published article [and its supporting information files]. Declarations Ethics approval and consent to participate This study was performed in accordance with Canadian Council on Animal Care (CCAC) guidelines and an Animal Use Protocol (AUP 282.34) approved by the institutional Animal Care Committee at the University Health Network. Consent for publication Not applicable. Competing interests The authors declare that they have no competing interests. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material Data Availability Statement All data generated or analyzed during this study are included in this published article [and its supporting information files].