PET-MR Guided, Pre-targeted delivery to HER2(+) Breast Cancer Model

Purpose: HER2(+) metastatic breast cancer (mBC) is one of the most aggressive and lethal cancer types among females. While initially effective, targeted therapeutic approaches with trastuzumab and pertuzumab antibodies and antibody-drug conjugates (ADC) lack long-term efficacy against HER2(+) mBC and can cause severe systemic toxicity due to off-target effects. Therefore, the development of novel targeted delivery platforms that minimize toxicity and increase therapeutic efficacy is critical to the treatment of HER2(+) breast cancer (BC). A pretargeting delivery platform can minimize the non-specific accumulation and off-target toxicity caused by traditional one-step delivery method by separating the single delivery step into a pre-targeting step with high-affinity biomarker binding ligand followed by the subsequent delivery step of therapeutic component with fast clearance. Each delivery component is functionalized with bioorthogonal reactive groups that quickly react in situ, forming cross-linked clusters on the cell surface, which facilitates rapid internalization and intracellular delivery of therapeutics. Procedures: We have successfully developed a click chemistry-based pretargeting platform for HER2(+) BC enabling PET-MR image guidance for reduced radiation dose, high sensitivity, and good soft tissue contrast. Radiolabeled trastuzumab and superparamagnetic iron-oxide carriers (uSPIO) were selected as pretargeting and delivery components, respectively. HER2(+) BT-474 cell line and corresponding xenografts were used for in vitro and in vivo studies. Results: An enhanced tumor accumulation as well as tumor-to-organ accumulation ratio was observed in pretargeted mice up to 24 h post uSPIO injection. A 40% local T1 decrease in the pretargeted mice tumor was observed within 4 h, and an overall 15% T1 drop was retained for 24 h post uSPIO injection. Conclusions: Prolonged tumor retention and increased tumor-to-organ accumulation ratio provided a solid foundation for pretargeted image-guided delivery approach for in vivo applications.


Introduction
Breast cancer is the most prevalent cancer type in the female population, and HER2(+) cancer constitutes 20%-30% of total breast cancer cases.[1,2] HER2 receptor overexpression correlates with rapid tumor progression, high grade, and poor prognosis.[3] The discovery of HER2 overexpression has motivated researchers to investigate therapeutic agents that target HER2 receptors.Humanized monoclonal antibody trastuzumab (Tz) has high binding a nity to the HER2 receptor and possesses anti-proliferation activity on HER2(+) cell lines.[4] Combined treatment with Tz and chemotherapeutics and, more recently, with Tz antibody-drug conjugates (ADC), such as ado-trastuzumab (T-DM1), and trastuzumab deruxtecan (T-DXd) have shown an e cient tumor growth inhibition and have become most successful ADCs, used for treatment of patients with HER2(+) BC. [5,6] However, ADCs show long circulation times and ine cient internalization, inhibiting the delivery of drugs into the tumor cells.[6][7][8] Due to off-site accumulation and premature release of the cytotoxic cargo in circulation, these single-step delivery systems may cause systemic toxicity that leads to a narrow therapeutic window.Therefore, implementing a pretargeting step is critical to minimize the off-target effects of the therapeutic carrier delivery component.
Pretargeting image-guided delivery platforms have been intensely studied and steadily evolved for the past three decades.[9,10] The pretargeting strategy circumvents the premature release of cytotoxic cargo from long circulating carriers and non-speci c accumulation of the traditional one-step delivery method by separating it into a pre-targeting of the receptor followed by a subsequent delivery of therapeutic and/or imaging component.[11,12] In the pre-targeting step, a high-a nity ligand speci cally targets the unique tumor microenvironment or receptors overexpressed on the surface of cancer cells without internalization.In the second step, a molecule or nanocarrier component with an optimized drug load and pharmacokinetics is administrated and selectively accumulates in locations, where the pretargeting bioligand is present.Components of each delivery step are conjugated in situ via bioorthogonal reactive groups.If a cell-surface target is bound by delivery components with multiple reactive groups, a web structure is formed on the cell surface that not only improves binding e ciency, but also promotes internalization, which facilitates the delivery of therapeutics.[13,14] Successful implementations of pretargeting strategy have been reported in both preclinical and clinical studies.[15][16][17][18] However, a PET-MR simultaneous imaging platform-based pretargeting strategy highlighting a T 1 MR contrast has not been reported.For fast in situ conjugation between the PET pretargeting component and MR active delivery component, we have used bio-orthogonal click chemistry as a backbone of our pretargeting delivery system.[19] Trans-Cyclooctene (TCO) and tetrazine (Tt) inverse-electron-demand Diels − Alder reaction is one of the fastest click chemistry reaction with a reaction rate of .[20,21] The system uses commercially available TCO and Tt linkers that we have successfully implemented in pretargeting strategy .[22,23] In this study, we synthesized a panel of distinct MR reporters based on the new superparamagnetic iron oxide nanoparticles, uSPIO construct, each bearing different reactive groups and/or uorophores and used for control and experiment groups.We rst con rmed the in vitro binding of uSPIO-Tt-Rhodamine to trastuzumab-TCO pretargeted BT-474 cells.Then, the pretargeting strategy was studied in vivo in mice bearing HER2 + tumors and both the pretargeting component trastuzumab-TCO-89 Zr and delivery component uSPIO-tetrazine were monitored with PET-MR imaging.Colocalization of PET and MR signal was only detected in the tumor region of pretargeted mice, in agreement with the initial hypothesis of our pretargeting model.
Synthesis of uSPIO-Rhodamine (Scheme 1a).Hydrophilic uSPIO with surface primary amine groups was synthesized as described in our previous report.[24] For uorescence labeling, NHS-Rhodamine (Rhod) (0.185 mg, 25 molar equiv.to uSPIO) was dissolved in 10 L of dry DMSO, then added to 2 mL of 1 mg/mL uSPIO aqueous solution at pH = 7.4 and mixed in the dark under room temperature for 1.5 h.The crude mixture was passed through a PD-10 column (equilibrated by pH = 7.4 1×PBS) to eliminate any unbound rhodamine.The collected eluent with the product was further puri ed and concentrated by 1×PBS using a 30 kDa MWCO centrifugal lter unit until no more pink color was present in the ltrate.
The nal uSPIO-Rhod product was stored at a concentration of 1mg/mL in 1×PBS at 4ºC.Synthesis of uSPIO-Tetrazine (Scheme 1b).Tetrazine-PEG 4 -NHS (0.2 mg, 30 molar equiv.to uSPIO) (Tt) was dissolved in 10 L of anhydrous DMSO and then added to a uSPIO solution (2 mL of 1.0 mg/mL in 1×PBS).The reaction was stirred in the dark under room temperature for 1.5 h.The crude mixture was passed through a PD-10 desalting column to eliminate unreacted Tt.The product was further puri ed and concentrated using a 30 kDa MWCO centrifugal lter unit until no more pink color in the ltrate was observed.The nal uSPIO-Tt product was stored at a 1.0 mg/mL concentration in 1× PBS at 4ºC.Synthesis of uSPIO-Tt-Rhodamine (Scheme 1c).NHS-Rhodamine (0.185 mg, 25 molar equiv.to uSPIO) was dissolved in 10 L of anhydrous DMSO, then added to 2 mL of 1 mg/mL uSPIO-Tt in 1×PBS.The reaction was stirred in the dark at room temperature for 1.5 h.The crude mixture was passed through a PD-10 desalting column to eliminate excess Rhod.The collected solution was further puri ed and concentrated as described above.The nal uSPIO-Tt-Rhod product was stored at a 1.0 mg/mL concentration in 1×PBS at 4ºC.Synthesis of Tz-TCO-AF488 (Scheme 1d).Tz-TCO-AF488 was synthesized according to our previously published protocol.[11,22,23] Brie y, Trastuzumab (500 µL of 10 mg/mL in 1×PBS) was treated with TCO-NHS ester (30 moles equiv. in 10-20 µL of dry DMSO) for 1 h with gentle stirring.Samples were puri ed by ultracentrifugation followed by SEC.The degree of functionalization was determined based on the change of molecular weight measured by MALDI-TOF.The resulting Tz(TCO) 6 (500 µL of 10 mg/mL in 1×PBS) was treated with NHS-AlexaFluor 488 (10 moles equiv.dissolved in 10 µL of anhydrous DMSO) and stirred for 1 h in the dark.The degree of labeling (DOL) of the uorophore was maintained at 2-4.The products were puri ed by ultra ltration followed by size-exclusion column chromatography.
Tz-TCO-DFO and Tz-TCO-DFO[ 89 Zr] were synthesized and characterized according to our published protocol.[23] Radiolabeling of Tz-TCO-DFO with 89 Zr was performed using a previously reported procedure.[23] 100 µL of 1.0 M oxalic acid containing 3.0 mCi of 89 Zr was mixed with 50 µL of 2.0 M Na 2 CO 3 and incubated for 3 min at room temperature.Then, 200 µL of 0.5 M HEPES (pH ~ 7) containing 1 mg of Tz-TCO-DFO was added.The resulting mixture was incubated for 1 h at room temperature with gentle rocking.The radiolabeling e ciency was measured using ITLC (Pall Life Science) and 20 mM citric radiochemical purity and 2.8 ± 0.2 mCi/mg speci c activity.For further studies, the speci c activity of Tz-TCO-DFO[ 89 Zr] was adjusted to 37 mBq (1.0 mCi)/mg using unlabeled Tz-TCO-DFO.
Characterization of uSPIO constructs.Core sizes of uSPIO constructs are characterized using Hitachi 7600 TEM.At least 150 uSPIO particles in one continuous TEM ROI were measured for an average core size using ImageJ.Hydrodynamic diameters (HD) of uSPIO constructs were measured by Malvern Zetasizer Nano ZS90 in pH = 7.4 1×PBS solution, with 3 measurements and 10 runs per measurement for each construct.Intensity distribution was used for HD analysis.
In Vitro binding.BT-474 cells were grown in 46-X medium supported by 10% Fetal Bovine Serum, 100 units/mL penicillin, and 100 units/mL Penicillin-Streptomycin (Pen-Strep).Cells were seeded into 4-well chamber slides at Passage 3, at a seeding density of 50,000 cells/chamber.After 2 days of cell growth, the medium was removed from all chambers and 0.25 mL of 35 Tz-TCO-AF488 in medium was added to chambers 1, 2, and 3 of each chamber slide, 0.25 mL of medium was added to chamber 4 of each slide as controls.Chambers were incubated on ice for 30 min.After incubation, all chambers were washed with sterile 1×PBS.Then 0.25mL of 35 uSPIO-Tt-Rhod in medium was added to chambers 1 and 2 of each chamber slide, 0.25mL of 35 uSPIO-Rhod in medium was added to chamber 3 of each chamber slide.All chamber slides were incubated in the incubator maintained at 37°C and 5% CO 2 .Chamber slides were incubated for 45 min and 4 hours, respectively (Supporting Information S3).
After the incubation period, medium was removed, and slides were washed with sterile 1×PBS.Cells were xed with 4% Paraformaldehyde (PFA) solution for 15 min in the dark at room temperature, aspirated and washed with 1×PBS.2.5 Hoescht 33342 solution was used for nuclei staining in the dark at room temperature for 8 min, aspirated and washed with 1×PBS.The chamber dividers were removed, and slides were mounted with cover slips using mounting agent with photobleaching inhibitor.Mounted slides were allowed to solidify overnight before confocal microscopy.
Confocal microscopy.Cells were imaged using a Zeiss LSM700 Single-point, laser scanning confocal microscope equipped with 405, 488, and 639 nm lasers used to excite blue nuclei, green Trastuzumab and red uSPIO, respectively.A 20 nm gap between the excitation and detection wavelengths was used for uorescence detection in all channels.A dual-line scanning method was used, and alignment was enforced before each scanning.The pinhole size of 1 AU (Airy Unit) was set for each color channel under the same magni cation.Confocal images were taken using a 43x oil immersion objective.Image processing was done with ImageJ.Individual channels were used for individual analysis and then merged for display.In vivoTumor and Mouse models.BT-474 breast cancer cells (ATCC) were grown in 46-X media supplemented with 10% Fetal bovine serum (Sigma-Aldrich), 100 units/mL of penicillin and streptomycin for inoculation of 6-8 weeks old athymic Nu/Nu female mice (Charles River Laboratories).Estrogen pellet 17β-estradiol (0.72mg, 90-day release, Innovative Research of America) was implanted in the subdermal area on the back of each mouse 24 h prior to inoculation.BT-474 cells.(2×10 6 cells in 50% of matrigel/media) were inoculated into the 2nd mammary fat-pad of the mouse.Mice were used for imaging when tumors reached 200-300 mm 3 .All animal procedures were performed in accordance with the United States Public Health Service Policy on the Humane Care and Use of Laboratory Animals and approved by the Johns Hopkins University Animal Care and Use Committee.
Two step, pretargeted delivery and imaging.PET-MR imaging was performed to visualize biodistribution of both the pretargeting component and the delivery component in mice bearing BT-474 tumors (Fig. 1).
Mice were randomly split into 2 groups: pretargeted and non-pretargeted groups (n = 3).Pretargeted mice were intravenously injected with 13.875 MBq (0.375 mCi) of Tz-TCO-DFO[ 89 Zr] containing 0.2 mg of Tz-TOC-DFO.Non-pretargeted group received same volume of saline.PET-MR images were collected for each mouse 24 h post Tz-TCO-DFO [ 89 Zr] injection.These images served as a baseline MR.After these images were obtained, 280 µg uSPIO-Tt was administered to all mice of by the tail vein injection.Prior to the injection, uSPIO-Tt was sterilized by ltration through 0.22 µm PVDF lters.PET-MR images were recorded for each mouse at 1, 4 and 24 hours post uSPIO-Tt injection.
The eld of view (FOV) of axial T 1 MR images were chosen to cover the entire chest and abdominal region of the animal using 20 slices with slice thickness of 1.5 mm.T 1 maps were acquired using RARE fast-spin echo sequence, with TE set at 13 ms and varying TR at 650, 1000, 1500, 2000, 4000 and 8000 ms, and in plane resolution of 0.25 mm.T 1 maps were reconstructed with IDL and/or Python software developed in our lab.T 2 -contrast MR images were used for co-registration with PET scans.The FOV of coronal T 2 MR images covered the entire body as well.T 2 contrast images were acquired with RARE fast spin-echo sequence, with T E = 32 ms and T R = 3200 ms.
PET images were reconstructed with an on-board Bruker ParaVision software and processed with PMOD software (version 4.004) to fuse PET and MR coronal images.The standardized uptake value (SUV, %ID/cc) for the pretargeting component Tz-TCO-[DFO] 89 Zr in relevant tissues and tumors were analyzed with PMOD software, taken account of radioactive decay.PET-MR fused images and 3D PET-MR images were processed with ParaVision and PMOD software.
Ex vivo biodistribution.uSPIO uptake was measured from quantitative T 1 analysis performed with the IDL programs and ImageJ image analysis software.uSPIO uptake was analyzed in the tumors, spleen, liver, and kidney.These organs were selected due to high iron concentrations detected by ICP-MS analysis in our previous study of untargeted uSPIO.[24] ImageJ was used to select ROIs for different organs in each slice.For each ROI, the surface area and the mean T 1 values are recorded.After processing of all slices, an average T 1 was calculated by weighing individual mean T 1 with corresponding surface areas.All mice and all time points from each treatment group were analyzed.

Statistical analysis
Statistical analysis done with JMP Pro 17 Software; P values of < 0.05 were considered signi cant.

Results
Nanomaterial Synthesis and Characterization.uSPIO-based nanoparticles were conjugated with rhodamine, tetrazine and tetrazine + rhodamine.Their core sizes were nm, nm, and nm, respectively, with good uniformity (Supporting Information S1.1, S1.2).These sizes also match the unconjugated uSPIO size reported in our previous publication.[24] The hydrodynamic diameters (HD) of these uSPIO conjugates were nm, and nm, respectively.All constructs have narrow HD distributions and are slightly larger than the unconjugated uSPIO, due to the conjugation of uorophores and/ or click-reactive group (Supporting Information S2).
In vitro Confocal Imaging.Processed images (Fig. 2) show that in non-pretargeted BT-474 cells treated with uSPIO-Tt-Rhod only, internalization is very low, suggesting minimal endocytosis of free uSPIO-Tt-Rhod.In both click-treated and none click-treated pretargeted groups, trastuzumab labeling was successful, as proved by the localization of green uorescence on the cell surface.In the group with nonclick-trastuzumab labeling, uSPIO was minimally internalized even after 4 h incubation, suggesting that click chemistry-driven ligation between components is necessary for uSPIO-Tt-Rhod membrane transport and endocytosis.In the cells preincubated with Tz-TCO-AF488, internalization of uSPIO-Tt-Rhod was observed starting from 45 min post uSPIO incubation.After 4 h incubation, more areas with higher signal intensity from rhodamine suggest that uSPIO-Tt-Rhod internalization increased with longer incubation time.
Individual and merged channels were examined in pretargeted groups (Figures S4.1 and S4.2). Figure S4.1 shows the channels for the 45 min incubation time point.Green channel (Tz labeling) signal readings are mostly from the cell surface, while some regions have initiated the caveolin-mediated endocytosis mediated by binding of Tz-TCO-AF488to HER2.In the click-reactive group, uSPIO-Tt-Rhod (red channel) were mainly located on the cell surface and some internalized due to click reaction between TCO and tetrazine, which has a very fast reaction kinetics.The yellow color seen on the merged image, mainly located near cell surface, suggested colocalization and reaction between the pretargeting component (Tz-TCO-AF488) and the delivery component (uSPIO-Tt-Rhod).In the none click-reactive group, uSPIO was not effectively internalized.At 4 h incubation time point (Figure S4.2), signi cantly more Tz-TCO-AF488 pretargeting component was internalized compared to 45 min incubation time.In the none click-reactive group, uSPIO-Tt-Rhod did not internalize even after 4 h of incubation, suggesting limited e ciency of non-speci c internalization In vivo Study.Quantitative PET image analysis shows the highest radioactivity accumulation in the tumor region after 24 h of 89 Zr injection (22%ID/cc in tumor), while less accumulations were detected in the liver and spleen (13 and 11%ID/cc, respectively, Figs.3a).Tz-TCO-DFO[ 89 Zr] retained in the tumor for at least 48 h (24 h before and 24 h after uSPIO injection), taking into account the radioactive decay (Fig. 3b).3D fused PET-MR images produced using PMOD software are shown in Fig. 3c&d.
At 1 h post uSPIO-Tt injection, T 1 map of the entire tumor ROI in the pretargeted group shows an overall percent decrease in T 1 with some regions demonstrating T 1 decrease over 40% (from 2810 ms to 2080 ms, Fig. 4, yellow arrow and Supporting Information S5), yet in the none pretargeted group tumor T 1 drop of only was detected at 1 h post-contrast.At 4 h post-injection, T 1 in the pretargeted group further decreased by of the baseline value, with some regions having T 1 reduced below 2000 ms.In contrast the T 1 in the non-pretargeted group recovered to 95% of its original value at 4 h time point.At 24 h post-injection, T 1 shortening is still persistent in the pretargeted group with an averaged reduction, while in the non-pretargeted group, T 1 has fully recovered.
Ex vivo Biodistribution Analysis.Representative T 1 maps and corresponding original gray-scale MR image are shown in Fig. 5. Organs' ROIs are highlighted with white dashed lines in the gure.The ROIs are depicted simply for representing the organs, the actual regions for T 1 analysis were chosen carefully to avoid any false registration.Figure 6 shows quantitative results obtained from the analysis.In the pretargeted group (round-shape marker), the tumor demonstrated 4-fold higher T 1 drop compared to the non-pretargeted group (diamond-shape marker) at 1 h (p < 0.01) as well as at 4 h post injection (p < 0.005) (Fig. 6a).At 24 h post injection, the difference between the two treatment groups was almost 10-fold (p < 0.0001).When cross comparing all major organs (Fig. 6b), non-pretargeted group experienced higher retention of uSPIO in all 3 major organs (liver, kidneys, and spleen) within 4 h post-injection than pretargeted group (p < 0.05, p < 0.01, p = 0.005), suggesting a non-speci c uptake of nanoparticles.It is also worth noting that the tumor demonstrated the strongest T 1 reduction compared to three major organs in the pretargeted group at 4 h post-injection, suggesting a target-speci c accumulation of the nanoparticles by tumor in the pretargeting delivery approach.At 24 h the most signi cant T 1 drop was detected in the liver and spleen in the non-pretargeted group.This agrees well with the ICP-MS analysis for uSPIO uptake in non-pretargeted mice shown in our previous study.[24] Discussions Image guidance enables visualization of administrated components and hence becomes key to translational theranostics.[25,26] In the pretargeting strategy, both components are conjugated with image probes utilizing orthogonal molecular imaging platforms.Combining PET imaging with MRI 24% ± 3.8% 6% ± 0.5% 25% ± 2.8% 11% ± 0.3% maximizes the information from the high sensitivity of a radioisotope and excellent spatial resolution and the enhanced soft tissue contrast of MRI.[27] An integrated bimodal imaging system such as PET-MRI can provide simultaneous image guidance of both pretargeting and delivery components.T 1 contrastenhancing MR reporters are favored for diagnostic imaging.Based on our previously developed sub-5nm hydrophilic uSPIO with enhanced T 1 relaxivity and ample modi cation sites as well as our 7 Tesla PET-MRI system, we developed PET-MR image-guided pretargeting strategy featuring the rapid click-chemistry driven two-step delivery system.[24] The in vitro confocal imaging showed that uSPIO-Tt-Rhod only internalizes when pretargeting components are present.This con rms the hypothesis that pretargeting is necessary and su cient for uSPIO-Tt-Rhod internalization.No cell death was seen in any of the treatment groups incubated with reagents and/or nanocarriers used during the labeling, as con rmed by comparing with cell morphology in the control group.At 4 h of incubation, more regions showed internalization in the click-reactive group compared to none click-reactive group, suggesting pretargeting might be a driving force for the enhanced ligand internalization.uSPIO-Tt-Rhod internalized and co-localized with Tz-TCO-AF488 only in the clickreactive group, suggesting the binding through TCO-tetrazine click chemistry reaction is e cient and stable.
Here we have assumed that the red signal from rhodamine corresponded to uSPIO, and the green signal corresponded to Alexa 488 labeled trastuzumab, and no free dissociated rhodamine or Alexa 488 was present at the time of imaging.This is a valid assumption for 2 reasons: 1) a total of 4 washing steps for uSPIO-Rhod and 6 steps for Tz-A488 have been performed throughout the binding study, this was also suggested by the clean background from all confocal images; 2) the untargeted uSPIO-Rhod group showed very low intensity and highly localized red signal, suggesting that the red uorescence signal in the pretargeted group originated from the uSPIO-Tt-Rhod bound to Tz-TCO-AF488.
The in vivo coregistered PET-MR image veri ed the targeting e ciency and precision of the trastuzumab pretargeting.Tumor uptake data are in good agreement with our previous study, as well as the confocal study discussed earlier.[23] Internalized uSPIO starts disassembling after 2 h post injection and are mostly disintegrated at 24 h time point due to the presence of proteolytic enzymes and the acidic environment of late endosomes and lysosomes.In addition, compartmentalization of uSPIO in the intracellular organelles, such as late endosomes and lysosomes, can cause reduced T 1 effects because of slow water exchange between cellular compartments and signal decay in T 2 weighted images due to increased T 2 susceptibility effects of concentrated and clustered nanoparticles.
[28] Therefore, the actual amount of uSPIO accumulated in the tumor region can be higher than what was extrapolated from T 1weighted MR images.Nonetheless, our results still provide solid proof that the pretargeting strategy can enhance in vivo localization and have increased tumor retention of nanoparticles.
uSPIO-Tt showed a signi cant uptake in liver and spleen, most likely due to the presence of Kupffer cell macrophages that rapidly recognize, engulf, and retain nanoparticles in these organs.Of special interest is the uSPIO uptake in kidneys at 24 h post-injection.While organs in the non-pretargeted group are predicted to have higher non-speci c accumulation, kidneys do not permanently retain nanoparticles and can eventually secrete them out even though the particle size is above the theoretical renal secretion threshold (14 nm vs 5 nm).However, this does align well with the ICP-MS analysis that showed minimal iron uptake in the kidney at 24 h.On the contrary, in the pretargeted group, the kidney uptake of uSPIO is more signi cant than other organs.However, mouse HER2 is not cross-reactive with trastuzumab, and a plausible explanation is that a trastuzumab-uSPIO nanoconjugates in circulation were clogging the renal lter, causing ine ciency in renal secretion.Bar graphs of uSPIO distribution for pretargeted vs nonpretargeted delivery show higher tumor uptake in the pretargeted group and overall higher non-speci c uptake in the non-pretargeted group (Fig. 6c, 6d).These results strongly suggest the selective and prolonged retention of nanocarriers by the pretargeting delivery approach.

Conclusions
We demonstrated that the pretargeting delivery strategy with Tz and uSPIO based nanocarrier increased site-speci c accumulation of the carriers, prolonged the delivery component retention, and reduced offtarget non-speci c uptake.PET-MRI simultaneous dual-modality imaging enabled by the radiolabeled Tz-TCO-DFO[ 89 Zr] and magnetic uSPIO-Tt with biocompatible surface coating with complimentary clickreactive groups allowed noninvasive monitoring of the delivery and biodistribution of both components.PET-MRI provides excellent sensitivity, spatial resolution and renders re ned soft tissue details that not only enables accurate location of the site of action, but also allows quantitative analysis for pretargeting system.This study provided solid foundation for the use of ultrasmall superparamagnetic iron-oxide core nanoparticles as drug carriers for image-guided pretargeted delivery approach and its potential application in molecular imaging and theranostics.Percent decrease of T 1 in organs from quantitative T 1 analysis.a) Tumor T 1 drop in both pretargeted and non-pretargeted groups; b) Major organs T 1 drop in both groups.Pretargeting and non-pretargeting groups are distinguished by the marker shape.In both gures, round-shape marker stands for the pretargeted group, while diamond-shape marker stands for the non-pretargeted group.c) Biodistribution of uSPIO in the pretargeted group from T 1 analysis.d) Biodistribution from T 1 analysis in the nonpretargeted group.Error bars were calculated from results for 3 different animals in both groups.e)-g) Statistical Analysis between pretargeted and non-pretargeted groups, at all time points.

Figure 3 a
Figure 3

Figure 5 See
Figure 5

Figure 6
Figure 6 pathways, such as pinocytosis or clathrin-mediated endocytosis in the delivery component internalization.This further con rms the effectiveness of the pretargeting strategy in enhanced in vitro internalization of nanocarriers.