A 68Ga–BCMA Ectodomain Tracer Enables PET/MR Longitudinal Tracking of Clinical-Grade BCMA/CD19 CAR T Cells

A 68Ga–BCMA Ectodomain Tracer Enables PET/MR Longitudinal Tracking of Clinical-Grade BCMA/CD19 CAR T Cells | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article A 68 Ga–BCMA Ectodomain Tracer Enables PET/MR Longitudinal Tracking of Clinical-Grade BCMA/CD19 CAR T Cells Min Yang, Wenyao Zhou, Xinyu Wang, Rui Hou, Junjie Yan, Donghui Pan, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9063774/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract CAR T cell therapies targeting CD19 and B cell maturation antigen (BCMA) induce profound responses in B cell malignancies, yet relapse highlight the need for non-invasive, quantitative tools to track cell kinetics. Here we develop an engineering-free, antigen-based PET/MR strategy to track tandem scFv BCMA/CD19 CAR-T product using a 68 Ga-labelled minimal BCMA ectodomain probe (BED). [ 68 Ga]Ga-NOTA-BED retains nanomolar affinity and high specificity for BCMA-scFv-containing cells, detects as few as ~ 2×10 4 cells without compromising effector functions. In mouse models, the probe enables quantitative discrimination of CAR-positive clusters, revealing a linear relationship between PET signal and cell number. Longitudinal PET/MR in lymphoma and myeloma xenografts visualizes heterogeneous CAR T expansion and trafficking patterns that align with distinct response phenotypes under varying antigen burden. This antigen-derived, human-sequence probe provides a repeatable, low-burden framework for kinetic phenotyping of dual-target CAR T therapies without additional cell engineering, and is positioned for clinical translation (NCT:07280793) as an imaging companion to guide patient-specific monitoring and trial design. Biological sciences/Biological techniques/Imaging/Positron-emission tomography Health sciences/Health care/Medical imaging/Radionuclide imaging Figures Figure 1 Figure 3 Figure 5 Figure 6 Introduction Chimeric antigen receptor (CAR) T cell therapies directed against CD19 and B cell maturation antigen (BCMA) have transformed treatment paradigms for B cell malignancies and multiple myeloma, delivering high response rates in heavily pretreated populations and establishing living cell products as a major therapeutic class 1 , 2 , 3 . However, heterogeneous responses and relapse remain frequent and are often linked to antigen heterogeneity/escape and insufficient in vivo expansion or persistence, limiting durable disease control. In BCMA‑directed settings, tumor‑intrinsic resistance mechanisms such as biallelic loss of BCMA have been documented after initial response, underscoring the vulnerability of single‑antigen targeting and the need for strategies that both broaden antigen coverage and enable mechanistic monitoring of treatment failure 4 . Dual‑target CAR designs have therefore gained momentum as a rational countermeasure to antigen escape 5 , 6 . In relapsed/refractory multiple myeloma, a bispecific BCMA/CD19 CAR (“BC19”) demonstrated feasibility, tolerability, and encouraging clinical activity in a phase I/II trial, supporting the translational rationale for BCMA/CD19 co‑targeting 7 . Beyond oncology, a homologous BCMA/CD19 CAR platform has been explored in refractory myasthenia gravis, enabling deep immune profiling of therapeutic responses and disease flares and further highlighting the clinical relevance of this CAR lineage across indications 8 . These advances sharpen a central unmet need: noninvasive, quantitative tools that can map where CAR T cells traffic and how they expand over time in the same subject, rather than inferring kinetics from sparse peripheral blood sampling or endpoint biopsies 9 , 10 . Positron emission tomography (PET) offers a whole‑body, quantitative readout that can, in principle, report on cell therapy biodistribution, tumor infiltration, expansion, persistence, and on‑target/off‑tumor risks, thereby enabling iterative optimization of dosing and product design 9 , 11 , 12 . Yet, repeatable PET tracking of living, proliferating CAR T cells remains challenging. Direct ex vivo radiolabeling enables immediate tracking with low background but rapidly loses interpretability after homing because signal is diluted by cell division and confounded by label loss or redistribution upon cell death 13 , 14 , 15 , 16 , 17 . Indirect reporter gene imaging avoids dilution by embedding a stable imaging tag, but it requires additional genetic engineering and can introduce operational complexity, regulatory hurdles, immunogenicity concerns, and reporter‑dependent physiologic background 18 , 19 , 20 , 21 , 22 , 23 , 24 . These constraints help explain why robust, longitudinal CAR T PET remains uncommon despite compelling proof‑of‑concept demonstrations in both preclinical models and patients. An emerging alternative is antigen‑based CAR‑PET, which uses the CAR’s target antigen as the imaging probe—aiming to preserve the timing flexibility of injectable tracers while avoiding any modification of the therapeutic cells 25 . In a recent study, the soluble CD19 ectodomain (~32 kDa) was proposed as an “ideal” probe to image CD19 CAR T cells and explicitly emphasized compatibility with currently FDA‑approved CAR T products without changing established clinical workflows 25 . However, translating antigen‑based CAR‑PET into a high‑frequency longitudinal imaging paradigm requires careful optimization of probe size, clearance, binding, and functional non‑interference—particularly when the goal is to capture expansion dynamics over weeks. Here, we extend the antigen‑based CAR‑PET paradigm to a clinically used tandem scFv BCMA/CD19 CAR T platform by developing an engineering‑free PET/MR tracer derived from a minimal BCMA extracellular domain fragment (BED) with a molecular weight of 5,396 Da. This small antigen probe is radiolabeled with 68 Ga (t 1/2 ≈ 67.7 min) to bind the anti‑BCMA scFv module on the CAR surface, thereby enabling rapid post‑scan clearance and low‑burden repeat imaging. Leveraging this fast‑clearing probe with PET/MR, we perform five scans over 22 days to quantify CAR T trafficking and proliferation kinetics during treatment of BCMA+ U266 and CD19+ Raji xenografts. This framework enables serial, whole‑body mapping of dual‑target CAR T expansion and response heterogeneity while avoiding the key limitations of dilution‑prone direct labeling and the added genetic modification required for reporter gene strategies. Results Design and Synthesis of a BCMA CAR Targeted Probe for Monitoring BC19 CAR T Cells BCMA is a defining surface marker of malignant plasma cells and a central therapeutic target in multiple myeloma 2 . As tumor heterogeneity and antigen escape increasingly motivate dual‑targeted or multispecific CAR platforms, imaging tools that remain compatible with diverse CAR architectures are needed for spatiotemporal monitoring in vivo. To meet this need, we developed an antigen‑based imaging probe derived from the core extracellular region of human BCMA (BCMA extracellular domain, BED; amino acid 5–54; calculated MW 5.4 kDa). BED contains the key epitope recognized by anti‑BCMA scFvs, providing direct compatibility not only with monospecific BCMA CAR T cells but also with multispecific constructs that include a BCMA‑binding arm, such as tandem scFv BCMA/CD19 CAR T cells (BC19 CAR T) (Fig. 1 ). This design bypasses additional genetic modification of CAR T cells and is intended as a generalizable imaging strategy for combinatorial CAR formats that retain an anti‑BCMA recognition module. Structural modeling (PyMOL) suggested a binding mode in which BED engages the anti‑BCMA scFv within the dual‑scFv (tandem) BC19 CAR configuration (Fig. 2 A). Given its compact size, BED is predicted to impose minimal steric hindrance and to preserve access to the spatially distinct anti‑CD19 scFv arm, thereby reducing the likelihood of functional interference in multispecific CAR settings (Fig. 2 A). The BED probe was produced in an Escherichia coli expression system using an N-terminal 6×His-SUMO fusion strategy (Fig. 2 B). The His tag enabled Ni‑NTA affinity purification, and the SUMO moiety enhanced solubility and folding. Following site‑specific cleavage by Ulp1 and polishing by size‑exclusion chromatography, BED was obtained at > 98% purity with a yield of ~ 5 mg/L. SDS‑PAGE and time‑of‑flight mass spectrometry confirmed the expected molecular mass and integrity of the final product (Fig. S1A, Fig. S2A), supporting its suitability for subsequent radiochemistry and in vivo studies. Effector cell validation and physicochemical characterization of BED and NOTA‑BED We first validated CAR expression on engineered T cells by flow cytometry. CD19 CAR T cells showed 93.6% positivity by anti‑FMC63 staining, whereas BCMA CAR T cells showed 77.1% positivity by BCMA antigen binding. Bispecific BC19 CAR T cells exhibited the expected dual‑targeting phenotype, with 68.5% of cells double‑positive for BCMA and CD19 binding; mock‑transduced T cells showed negligible binding (Fig. 2 C). These data establish a robust cellular platform for evaluating BED‑based imaging. Bio‑layer interferometry (BLI) demonstrated high‑affinity binding of unmodified BED to the anti‑BCMA scFv (K D = 0.89 nM; Fig. 2 D). For PET probe construction, a NOTA‑BED conjugate was synthesized and purified. SDS‑PAGE showed a single band with a slight upward shift compared to BED, consistent with successful chelator conjugation and high purity (Fig. S1A). TOF‑MS confirmed a molecular mass of 5672.52 Da, matching the theoretical value for single NOTA attachment (5672 Da; Fig. S2B). Importantly, BLI confirmed that NOTA‑BED retained nanomolar binding affinity (K D = 1.11 nM; Fig. 2 E; Table.1), indicating that conjugation preserved the functional epitope. [ 68 Ga]Ga‑NOTA‑BED enables sensitive and specific detection of CAR T cells in vitro Radiochemical quality control showed that purified [ 68 Ga]Ga‑NOTA‑BED achieved radiochemical purity (RCP) > 98% and radiochemical yield (RCY) > 50%, with a molar activity of ~ 17.5 MBq/nmol, supporting sensitive in vitro assays and in vivo imaging. Time‑dependent uptake assays demonstrated rapid and specific binding of [ 68 Ga]Ga‑NOTA‑BED to BC19 CAR T cells (5.45 ± 0.46% and 5.06 ± 0.51% added dose per 5×10 5 cells at 1 h and 2 h, respectively; Fig. 3 A). A 1000‑fold molar excess of unlabeled probe reduced uptake to near background (0.24 ± 0.08% and 0.17 ± 0.09%), confirming specificity. Mock‑T controls showed negligible uptake (< 0.1%). To exclude nonspecific interactions with shared CAR scaffold elements, uptake was compared across CAR phenotypes (Fig. 3 B). Accumulation was strictly dependent on the presence of an anti‑BCMA scFv: BCMA CAR T cells showed the highest uptake (13.54 ± 3.14% at 1 h), whereas BC19 CAR T cells showed lower but substantial uptake (6.53 ± 1.54%), consistent with reduced BCMA scFv availability and/or steric effects in the tandem architecture. In contrast, CD19 CAR T cells and mock‑T cells remained at background levels. These results indicate that [ 68 Ga]Ga‑NOTA‑BED recognizes the anti‑BCMA antigen‑binding fragment with minimal cross‑reactivity to common signaling/scaffold components. Because antigen engagement in vivo may reduce availability of the CAR’s anti‑BCMA binding site, we modeled competitive conditions using tumor–effector co‑cultures (Fig. 3 C). In the presence of BCMA‑negative Raji cells, probe uptake by BC19 CAR T cells was minimally affected (82.63 ± 2.61% of control even at E:T = 5:1). In contrast, co‑culture with BCMA‑high U266 cells reduced uptake to 66.53 ± 0.94% (E:T = 1:1) and 57.01 ± 2.21% (E:T = 5:1), yet absolute uptake remained measurable (3.59 ± 0.14% at E:T = 5:1). These data suggest that high antigen burden may attenuate early imaging by occupying the CAR’s anti‑BCMA scFv, but [ 68 Ga]Ga‑NOTA‑BED retains sufficient sensitivity to track CAR T presence and dynamics under competitive conditions. [ 68 Ga]Ga‑NOTA‑BED enables quantitative and highly specific CAR T tracking in vivo To validate in vivo specificity, we established a bilateral subcutaneous cell‑cluster model in BALB/c mice by inoculating mock‑T and distinct CAR T subtypes into opposite flanks. PET imaging revealed selective probe accumulation at sites expressing an anti‑BCMA scFv (Fig. 3 D). BCMA CAR T sites showed the highest uptake (1.05 ± 0.18%ID/g at 1 h), followed by BC19 CAR T sites (0.90 ± 0.04%ID/g), consistent with in vitro binding. CD19 CAR T and mock‑T sites were indistinguishable from background (0.15 ± 0.04 and 0.11 ± 0.03%ID/g; Fig. 3 E), demonstrating robust in vivo discrimination of target versus non‑target populations. We next assessed sensitivity by PET/MR imaging of mice receiving serial dilutions of CAR T cells. PET signal decreased stepwise with decreasing cell number, yet discrete foci remained visible above background at 2 × 10 4 cells (Fig. 3 F). At this dose, uptake (0.18 ± 0.02%ID/g) exceeded muscle background (0.04 ± 0.01%ID/g; Fig. 3 G). Moreover, PET signal correlated linearly with injected CAR T cell number (R 2 = 0.856; Fig. 3 H), supporting quantitative use of [ 68 Ga]Ga‑NOTA‑BED as a surrogate readout of CAR T cell abundance in vivo . BED probe binding does not impair BC19 CAR T functionality We evaluated whether BED binding interferes with BC19 CAR T cytotoxicity using luciferase-based killing assays against Raji-Luc (CD19+) and U266-Luc (BCMA+) targets. BC19 CAR T cells showed strong E:T-dependent killing compared with mock-T controls (Fig. 4 A), achieving maximal lysis of 96.92 ± 0.67% for Raji-Luc and 99.36 ± 0.39% for U266-Luc (Fig. 4 B,C). In interference experiments performed at a fixed E:T ratio (2:1), addition of BED up to 100 nM—well above the estimated peak plasma concentration required for imaging—did not alter killing (Fig. 4 D), with lysis remaining 89.87 ± 5.85% (Raji-Luc) and 98.89 ± 1.08% (U266-Luc) (Fig. 4 E,F; P > 0.05). To assess effects on CAR activation, cytokine secretion was quantified. TNF-α and IFN-γ levels in co-culture supernatants were unchanged in the presence of up to 100 nM BED (Fig. 4 G–J; P > 0.05), indicating that probe binding does not measurably disrupt antigen recognition, downstream signaling, or effector function. Together with the probe’s hydrophilicity and expected rapid renal clearance in vivo, these findings support a favorable biological safety profile for repeated imaging applications. Longitudinal PET/MR visualizes CAR T dynamics and therapeutic response heterogeneity in Raji lymphoma We performed longitudinal PET/MR and IVIS monitoring using [ 68 Ga]Ga‑NOTA‑BED in a Raji tumor model to map spatiotemporal CAR T kinetics during therapy (Fig. 5 A). Although BC19 CAR T–treated mice ultimately achieved tumor control, early response dynamics were heterogeneous. Co‑registration of PET/MR‑derived CAR T distribution with IVIS‑derived tumor burden enabled classification into three response phenotypes (Fig. 5 B; Fig. S3). Fast responders (e.g., R2) showed rapid tumor clearance by day 8, and PET signals remained near baseline thereafter (0.19–0.36%ID/g; Fig. 5 C), consistent with rapid antigen loss and limited sustained expansion. Intermediate responders (e.g., R1) displayed a classical rise‑and‑fall immune kinetic: tumor burden increased to day 8 then declined, while PET signal increased and peaked around day 17 before contracting as tumor was eliminated (Fig. 5 D). Slow responders (e.g., R5) showed prolonged coexistence of tumor growth and CAR T accumulation, with progressive PET signal increase reaching a maximum at day 22 (Fig. 5 E). PBS controls exhibited uncontrolled tumor growth and no specific tracer accumulation (Fig. 5 B, 5 F). Quantitatively, excluding non‑proliferative fast responders, treated mice reached an average peak uptake of 1.29 ± 0.48%ID/g (Fig. 5 G) and a peak tumor‑to‑muscle ratio of 15.29 ± 2.29 (Fig. 5 H), both significantly higher than PBS controls (P < 0.01). Individual uptake curves further highlighted marked inter‑animal heterogeneity in both peak magnitude and timing (Fig. 5 I). IVIS confirmed potent antitumor activity, with tumor flux decreasing to 3.43×10 7 p/s by day 22 (approximately 1/220 of PBS peak 7.55×10 9 p/s at day 18; Fig. 5 J). Treated mice maintained stable body weight (Fig. 5 K), showed no overt organ pathology (Fig. S4), and achieved 100% survival versus a median survival of 18 days in PBS controls (P < 0.001; Fig. 5 L). Terminal immunohistochemistry further supported specificity: residual microscopic lesions in treated mice remained densely infiltrated by CD3 + T cells despite low CD19 signal (Fig. 5 M), consistent with effective CAR T homing and sustained tumor‑site localization. Longitudinal PET/MR tracking under high antigen burden in U266 multiple myeloma To evaluate probe performance under conditions of high antigen burden and potential CAR occupancy, we established a composite U266 myeloma model. In contrast to the uniform remission observed in the Raji setting, BC19 CAR T treatment produced heterogeneous outcomes that separated into Cure and Progress subgroups (Fig. S5). To better capture early and delayed kinetics in this model, we added an additional early scan (day 4) and shifted the day 8 timepoint to day 11 (Fig. 6 A). PET/MR imaging mapped BC19 CAR T distribution longitudinally (Fig. 6 B). In the Cure subgroup, tracer uptake rose promptly as tumor burden declined, peaking around day 14 (~ 1.44%ID/g) and returning toward baseline following tumor eradication and antigen loss (Fig. 6 C). In the Progress subgroup, kinetics were delayed and variable: some animals showed a gradual PET rise without tumor control (Fig. S5), whereas others showed transient PET increases followed by declines despite continued tumor growth by IVIS (Fig. 6 D). This PET–IVIS mismatch is consistent with the combined effects of high tumor load, CAR occupancy/competition, and/or functional impairment within the myeloma microenvironment. Controls (mock‑T and PBS) displayed exponential tumor growth by IVIS without specific PET signal in tumor regions (Fig. 6 E,F). Across treated animals, peak uptake averaged 1.19 ± 0.28%ID/g (Fig. 6 G) with a peak tumor‑to‑muscle ratio of 11.33 ± 2.94 (Fig. 6 H), significantly exceeding controls (P < 0.01), and individual curves delineated distinct kinetic patterns (Fig. 6 I). Body weights remained stable in the BC19 group but declined in controls due to progressive disease (Fig. 6 K). No major organ pathology was observed on H&E (Fig. S6), and the BC19 group maintained 100% survival over 23 days (P < 0.05; Fig. 6 L). Terminal immunohistochemistry confirmed tracking specificity: Progress tumors retained strong BCMA expression and showed prominent intratumoral CD3 + infiltration, whereas control tumors lacked CD3 staining (Fig. 6 M). [ 68 Ga]Ga‑NOTA‑BED exhibits favorable translational safety. To assess the translational viability of the [ 68 Ga]Ga-NOTA-BCMA probe, we performed comprehensive human radiation dosimetry and chemical toxicity evaluations. The probe exhibited favorable in vivo safety and biocompatibility. Biodistribution studies and OLINDA/EXM estimates indicated predominant renal clearance, with the kidneys acting as the dose-limiting organ (8.71 × 10 − 1 mSv/MBq). Notably, the total-body effective dose (0.0128 mSv/MBq) was lower than that of [ 18 F]F-FDG (0.019 mSv/MBq), supporting its safe application in longitudinal repetitive imaging (Table. 2). Additionally, a single intravenous administration at a 10-fold higher dose induced no abnormal vital signs, and histological analysis of major organs showed no structural damage (Fig. S8), verifying no notable acute toxicity. Discussion We developed an engineering‑free, antigen‑based PET/MR strategy for longitudinal tracking of a clinically used BCMA/CD19 tandem scFv CAR‑T product. A minimal BCMA ectodomain fragment radiolabelled with 68 Ga enabled five PET/MR scans over 22 days, resolving heterogeneous CAR‑T kinetics in vivo while preserving cytotoxicity and cytokine release. The work extends antigen‑probe CAR‑PET 25 toward a repeatable monitoring approach compatible with ongoing BCMA/CD19 CAR‑T programs. This study establishes a compact probe with preserved binding and practical radiochemistry. NOTA‑BED retained nanomolar affinity for the anti‑BCMA scFv, and [ 68 Ga]Ga‑NOTA‑BED was produced with RCP > 98%, RCY > 50%, and molar activity ~ 17.5 MBq/nmol. In vivo, the tracer enabled sensitive and quantitative detection: PET signal remained discernible above background at ~ 2 × 10 4 CAR‑T cells (in vivo LOD), and uptake scaled with cell number (R 2 = 0.856), supporting its use as a quantitative surrogate under the tested conditions. Critically, because antigen‑binding tracers could in principle interfere with CAR signaling, we directly tested functional compatibility: BED up to 100 nM did not impair BC19 CAR‑T cytotoxicity against both Raji‑Luc and U266‑Luc, and TNF‑α/IFN‑γ release remained unchanged, indicating that tracer binding is functionally tolerated at concentrations exceeding those anticipated for typical PET microdose exposures 25 , 26 . The main conceptual advance is the feasibility of repeatable longitudinal imaging without additional cell engineering, achieved by aligning probe size and radionuclide physics with serial studies. Antigen‑probe CAR‑PET was articulated as a strategy to image clinical CAR products without adding reporter genes 14 . Here, probe miniaturization to 5.4 kDa—smaller than clinically evaluated HER2-binding Affibody scaffolds (6.5-7 kDa) 27 , 28 —facilitates rapid post-scan clearance, reduces carry-over background and permits repeated administration. Pairing this human-sequence antigen fragment with 68 Ga (t 1/2 ≈ 67.7 min), as in established Affibody PET workflows 29 , further supports short-interval “inject-and-scan” imaging, albeit with an intrinsically brief imaging window per injection 30 . Together, these choices enabled five scans within 22 days, a time span that captures the clinically relevant expansion 31 , 32 –contraction phase of CAR‑T therapy and is not readily accessible with many existing PET cell‑tracking workflows. Mechanistically, our data are consistent with scFv‑dependent surface labeling rather than nonspecific T‑cell retention. Rapid, competitively blockable uptake in BCMA‑scFv–positive CAR‑T populations, and background‑level uptake in CD19 CAR‑T and mock T cells, indicate that PET signal reflects the presence and accessibility of the BCMA‑binding arm. Nevertheless, competitive co‑culture experiments also foreground a key interpretive constraint of antigen‑based CAR‑PET: antigen occupancy under high tumour burden can reduce tracer binding by engaging the same scFv epitope used for imaging. This feature is not merely a technical caveat; it implies that PET signal is a composite of cell number, CAR surface density, and CAR accessibility (that is, the fraction of unoccupied binding sites), and may therefore diverge from tumour burden metrics in high‑antigen settings. Longitudinal PET/MR revealed distinct kinetic phenotypes across tumour contexts, highlighting why repeated imaging is informative beyond single time points. In the CD19 + Raji model, five scans delineated fast, intermediate, and slow responder trajectories: fast responders displayed rapid tumour clearance with near‑baseline PET thereafter; intermediate responders showed a rise‑and‑fall PET trajectory aligned with regression; and slow responders exhibited delayed, progressive PET accumulation peaking late. In the BCMA‑high U266 model, outcomes diverged into cure versus progress subgroups: cures exhibited early tumour‑region PET increases coincident with tumour decline, whereas progressors showed delayed or discordant PET trajectories (including transient rises followed by declines despite continued tumour growth). These patterns are consistent with at least two non‑exclusive mechanisms: competitive antigen occupancy reducing tracer access in high‑burden tumours, and/or functional impairment of CAR‑T cells in the myeloma microenvironment, emphasizing that PET‑derived CAR‑T “signal” is informative but not synonymous with efficacy. The translational relevance is sharpened by the maturation of clinically used BCMA/CD19 CAR‑T programs. A phase I/II study reported feasibility and activity of bispecific BCMA/CD19 CAR‑T therapy in relapsed/refractory multiple myeloma 7 , and a homologous BCMA/CD19 lineage has been explored in refractory myasthenia gravis with deep immune profiling 8 . For these programmes, imaging methods that require additional genetic modifications can be difficult to retrofit. Reporter‑gene PET has clinical proof‑of‑concept for tracking engineered T cells 18 , but it adds genetic payload, manufacturing complexity, and regulatory considerations. Direct ex vivo radiolabelling can map early biodistribution, but suffers from label dilution with proliferation and signal confounding from dead‑cell label redistribution, limiting interpretability across the weeks‑long kinetic arc 13 . In contrast, antigen‑based BED imaging retains the practical advantage of not altering the therapeutic cells while enabling serial snapshots across time, aligning with the “engineering‑free” premise emphasized in antigen‑probe CAR‑PET 25 and extending it to a high‑frequency, multi‑scan workflow. Several limitations should be addressed explicitly before human translation. First, small‑animal xenografts incompletely recapitulate human tumour architecture, antigen heterogeneity, vascular permeability, and immune context, and do not reliably predict immune responses to repeat dosing. That said, BED is derived from a human BCMA extracellular domain and is essentially congruent with soluble BCMA apart from the chelator conjugation, so the intrinsic risk of immunogenicity is expected to be comparatively low; nevertheless, the NOTA (and any linker) constitutes a non‑native chemical moiety that could act as a hapten and warrants empiric evaluation of anti‑drug antibodies and their impact on clearance, background, and safety under repeat administration. Second, renal clearance is expected to dominate for a ~ 5.4‑kDa protein, which may concentrate activity in kidney/bladder and make kidney‑limited dosimetry a key constraint under repeat imaging; formal dosimetry and mitigation strategies (hydration/voiding protocols, renal dose modelling) are therefore required. Third, 68 Ga’s short half‑life (≈ 67.7 min) limits imaging windows per injection and may reduce sensitivity if human pharmacokinetics are slower than in mice; longer‑lived isotopes (e.g. 18 F, t 1/2 ≈109.7 min) may be needed depending on clinical workflows. Fourth, quantitative comparability across models and scanners is nontrivial: %ID/g (mouse) and SUV (human) differ in scaling; moreover, partial‑volume effects and ROI definition can bias lesion quantification, particularly for small or regressing tumours. Fifth, although RCY > 50% and RCP > 98% are strong starting points, chelator choice (NOTA versus alternatives) and radiochemistry conditions may be further optimized to improve kit‑readiness and increase molar activity, thereby minimizing the risk of receptor occupancy by cold mass. Finally, response subgrouping (fast/intermediate/slow; cure/progress) emerges from limited cohorts; larger studies are needed to improve statistical power, define predictive thresholds (for example, time‑to‑peak or peak amplitude), and test reproducibility. Where injected mass/activity, scan start time, and cohort sizes were not specified, we assume typical PET microdose conditions and ~ 1 h post‑injection imaging for 68 Ga 33 ; these parameters should be explicitly optimized and reported in translational studies. Future work can follow a pragmatic, translation‑forward path. Because diagnostic PET radiopharmaceuticals are administered at microdose‑level masses and are used at the low end of the dose–response curve, dose‑related pharmacologic or toxic effects from the nonradioactive mass are generally unlikely, and regulatory guidances explicitly encourage tailored, risk‑based nonclinical packages to facilitate timely early‑phase studies 34 , 35 , 36 . In this context, an early first‑in‑human imaging sub‑study embedded within ongoing BCMA/CD19 CAR‑T programmes 7 is feasible to establish human pharmacokinetics, lesion targeting, and whole‑body/kidney dosimetry, while repeat‑dose considerations (including immunogenicity and renal dose) are addressed in parallel through focused nonclinical work. In parallel, radiochemistry and formulation should be optimized toward higher molar activity and kit‑readiness 37 , 38 , and the generalizability of BED imaging should be tested across clinically relevant CAR architectures (tandem versus bicistronic), ideally integrated with longitudinal blood‑based measures 39 , 40 to link imaging kinetics to expansion, persistence, and outcome. In summary, we developed an engineering-free, antigen-based PET/MR strategy to longitudinally track a clinically used tandem scFv BCMA/CD19 CAR-T platform using a 68 Ga-labeled BCMA ectodomain probe. The compact, human-derived BED tracer retains ~ 1 nM binding, clears rapidly after each scan, and supports repeat imaging with a short-lived radionuclide and minimal procedural burden. [ 68 Ga]Ga-NOTA-BED achieved high specificity and sensitivity (≈ 2×10 4 cells) without measurably impairing CAR T function. Five PET/MR scans over 22 days resolved heterogeneous CAR T trafficking and expansion kinetics that aligned with response in Raji lymphoma and revealed distinct dynamics under high antigen burden in U266 myeloma. These data support clinical translation of an on-demand imaging companion for patient-specific interpretation of CAR T expansion and lesion engagement; notably, a prospective study (ClinicalTrials.gov Identifier: NCT07280793, “CAR-T Cell Efficacy With Molecular Imaging in Multiple Myeloma”) has been registered based on this study and will be posted publicly on ClinicalTrials.gov. Methods Cell culture and cell lines Human Burkitt’s lymphoma Raji cells and human multiple myeloma U266 cells were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). To generate reporter cell lines stably expressing green fluorescent protein (GFP) and firefly luciferase (Luc), Raji and U266 cells were transduced with a lentiviral vector (pASLenti-pA-Luc2-CMV-EF1-EGFP-P2A-Puro-WPRE; OBiO Technology, China). Transduced cells were selected with puromycin (2 µg/ml; Beyotime, Catalog No. ST551) for 14 days, resulting in stable polyclonal populations co-expressing GFP and luciferase (hereafter referred to as Raji-Luc-GFP and U266-Luc-GFP). All tumor cell lines were maintained in RPMI 1640 medium (Adamas-life, Catalog No. C8016) supplemented with 10% fetal bovine serum (FBS; Adamas-life, Catalog No. C8010) and 1% penicillin–streptomycin (Beyotime, Catalog No. C0222). Primary human T cells were isolated from donor peripheral blood mononuclear cells (PBMCs) to produce untransduced control T cells (Mock-T) and lentivirally transduced CD19 CAR T, BCMA CAR T, and bispecific BC19 CAR T cells. All PBMCs were donated by healthy donors under protocols approved by the Medical Ethics Committee of the Affiliated Hospital of Xuzhou Medical University (ethics approval no. XYFY2020-KL062-01). For in vitro expansion and maintenance, T cells were cultured in X-VIVO™ 15 medium (Lonza, Catalog No. 04-418Q) supplemented with 10% FBS and freshly prepared recombinant human cytokines, including IL-2 (300 U/ml; Beyotime, Catalog No. P5115), IL-7 (5 ng/ml; Yeasen, Catalog No. 90188ES10), and IL-15 (5 ng/ml; Yeasen, Catalog No. 90113ES10). All cells were cultured in a humidified incubator at 37°C with 5% CO 2 . Flow cytometric analysis of CAR expression Surface CAR expression and transduction efficiency were evaluated by multicolor flow cytometry. CD19 CAR T, BCMA CAR T, bispecific BC19 CAR T, and untransduced Mock-T cells in logarithmic growth phase were harvested. Approximately 5 × 10 5 cells per condition were washed with ice-cold FACS buffer (PBS + 1% BSA; Beyotime, Catalog No. ST023) and blocked with human Fc receptor reagent (5 µL; Beyotime, Catalog No. C1752S) for 10 min at 4°C. Cells were stained with APC-conjugated anti-FMC63 antibody (1:50; ACROBiosystems, Catalog No. FM3-AY54A1) and FITC-labeled human BCMA protein (3 µg/mL; ACROBiosystems, Catalog No. BCA-HF254) for 60 min at 4°C in the dark; single-stained controls were included for compensation. After washing, 7-AAD (5 µL; Proteintech, Catalog No. PD00101) was added 5 min before acquisition. Data were collected on a BD FACSCelesta cytometer (BD Biosciences, USA) and analyzed with FlowJo v10.8 (BD Biosciences). At least 10,000 live cells per sample were recorded, and CAR expression was quantified as the percentage of APC + and/or FITC+ cells among live cells, with Mock-T cells used as gating controls. Preparation and quality control of BED The core extracellular domain of BCMA (UniProtKB Q02223, residues A5 to A54) was selected as the scaffold for probe construction and is hereafter referred to as BED. A 6×His-SUMO dual tag was introduced at the N-terminus, and the codon-optimized gene was cloned into the pET-28a(+) vector (XbaI/XhoI sites), yielding the recombinant plasmid pET28a-His-SUMO-BCMA(5–54). The plasmid was transformed into Escherichia coli BL21(DE3), and protein expression was induced with 0.1 mM isopropyl-β-D-thiogalactopyranoside (IPTG) at 16°C for 16 hours. Cells were harvested and lysed, and the His-SUMO-BCMA fusion protein was purified by Ni-NTA affinity chromatography. The SUMO tag was subsequently removed by Ulp1 protease digestion, followed by a second Ni-NTA purification step to remove the cleaved tag; the flow-through fraction containing BED was collected. The protein was further purified by size-exclusion chromatography (Superdex 75 pg column, ÄKTA pure system; GE Healthcare), concentrated by ultrafiltration, buffer-exchanged, and lyophilized to obtain BED powder. Protein purity, aggregation state, and molecular mass were systematically assessed by SDS-PAGE, SEC-HPLC, and time-of-flight mass spectrometry (TOF-MS), respectively. Conjugation and purification of NOTA-BED Lyophilized BED was dissolved in anhydrous dimethyl sulfoxide (DMSO), and NOTA-NHS ester solution (50 mg/ml) was added dropwise at a molar ratio of 1:8 (protein:chelator). N,N-diisopropylethylamine (DIPEA) was added to adjust the pH to 9.0, and the reaction was carried out at 40°C for 8 hours with continuous stirring. The reaction mixture was purified by reversed-phase high-performance liquid chromatography (RP-HPLC) using a preparative C18 column (PrepPurite Gold, 5 µm, 10 × 250 mm; Wepure) at a flow rate of 5.0 ml/min with a linear gradient of solvent A (0.1% TFA in water) and solvent B (0.1% TFA in acetonitrile). Elution was monitored at 280 nm. The target fraction was collected and analyzed by TOF-MS to confirm successful NOTA conjugation and determine the exact molecular mass. BLI-based affinity measurement Binding kinetics between BED probes and CAR molecules were evaluated using bio-layer interferometry (BLI) on an Octet R2 system (Sartorius). His-tagged CAR proteins (30 µg/ml) were immobilized onto Ni-NTA biosensors and equilibrated in PBST buffer (PBS containing 0.05% Tween-20). Serial dilutions of BED and NOTA-BED (0.78 to 50 nM) were prepared and applied for association (300 s) and dissociation (300 s) measurements. Data were analyzed using Octet Analysis Studio software (Sartorius) with a 1:1 global fitting model to derive the association rate constant ( k on ), dissociation rate constant ( k off ), and equilibrium dissociation constant (K D = k off / k on ). Radiolabeling with [ 68 Ga]Ga and quality control Freshly eluted [ 68 Ga]GaCl 3 solution (500 µL in 0.05 M HCl) was buffered with 1.25 M sodium acetate to adjust the pH to 4.0 to 4.5. NOTA-BED (50 µg) was added and incubated at 60°C for 15 min. The reaction mixture was purified using a C18 solid-phase extraction cartridge (Waters). After sequential conditioning with ethanol and water, the sample was loaded, washed with water to remove free [ 68 Ga]Ga, and eluted with 300 µL of ethanol containing 10 mM HCl to obtain [ 68 Ga]Ga-NOTA-BED. Radiochemical purity (RCP) was determined by radio-HPLC under the same chromatographic conditions as the non-radioactive probe and calculated as the percentage of radioactivity corresponding to the main peak. Radiochemical yield (RCY) and molar activity were calculated based on starting activity and BED amount. Time-dependent cellular uptake and blocking assays To evaluate the binding specificity and kinetic properties of the probe, BC19 CAR T cells and Mock-T cells in the logarithmic growth phase were harvested, washed with PBS, and resuspended in serum-free X-VIVO™ 15 medium at a density of 5 × 10 5 cells per tube. Approximately 37 kBq (1 µCi) of [ 68 Ga]Ga-NOTA-BED was added to each tube, followed by incubation at 37°C for 60 or 120 min.For competitive blocking, a 1000-fold molar excess of unlabeled NOTA-BED (50 µg) was added 1 hour prior to radiotracer incubation. After incubation, the cells were centrifuged and washed three times with ice-cold PBS containing 1% BSA to remove unbound and nonspecifically associated tracer. Cell pellets were resuspended in 500 µL of PBS, and radioactivity was measured using a gamma counter (Perkin Elmer, American). Cellular uptake was expressed as the percentage of the added dose (% AD) per 5 × 10 5 cells. All experiments were performed in triplicate. Uptake specificity among different CAR T subtypes To assess the selectivity of the probe for the BCMA CAR scFv, Mock-T cells (negative control), CD19 CAR T cells (nontarget control), BCMA CAR T cells (single-target positive control), and BC19 CAR T cells (dual-target experimental group) were incubated with [ 68 Ga]Ga-NOTA-BED (5 × 10 5 cells per tube) under the same conditions for 60 min. Cellular uptake was calculated as % AD/5 × 10 5 cells as described above, and values among the different groups were compared to determine probe specificity toward BCMA CAR-expressing T cells. Tumor cell competition assay To simulate the competitive inhibition of probe uptake by tumor-associated BCMA antigens in the tumor microenvironment, a co-incubation model was established using CAR T cells and tumor target cells. BC19 CAR T cells (effector cells, E) were mixed with Raji (CD19⁺) or U266 (BCMA⁺) target cells (T) at effector-to-target (E:T) ratios of 1:1 and 1:5. BC19 CAR T cells incubated alone served as the baseline positive control (defined as 100% uptake), and those pre-blocked with excess unlabeled NOTA-BED served as the specificity control. Cell mixtures were pre-incubated at 37°C for 1 hour, followed by the addition of 37 kBq (1 µCi) of [ 68 Ga]Ga-NOTA-BED and further incubation for 1 hour. After incubation, the cells were centrifuged and washed with ice-cold PBS, and the radioactivity of the cell pellets was measured. Relative uptake was calculated as counts per minute (CPM) normalized to the baseline positive control group and expressed as a percentage. Targeting specificity assessment To evaluate the in vivo targeting selectivity of the BED probe, healthy BALB/c mice received subcutaneous injections of Mock-T cells or CAR T cells with different antigen specificities (3 × 10 6 cells in 100 µL sterile saline) into the bilateral axillae. Immediately after cell implantation, 3.7–5.5 MBq (100–150 µCi) of [ 68 Ga]Ga-NOTA-BED was administered via the tail vein. Static micro-positron emission tomography (micro-PET) imaging was performed 1 hour post-injection. Detection sensitivity and limit of detection To determine the minimum detectable CAR T cell number, in vivo , a cell quantity gradient was established. BC19 CAR T cells (5 × 10 3 to 5 × 10 6 cells in 100 µL sterile saline) were injected subcutaneously into BALB/c mice (n = 3), simulating target tissues of varying cell quantities. One hour post-probe injection, mice were imaged using a Bruker 9.4T small-animal PET/magnetic resonance (PET/MR) system (Bruker BioSpec 94/30, Germany) under 1.0–1.5% isoflurane anesthesia. The lower limit of detection (LOD) was defined as the lowest cell quantity yielding a discernible signal distinguishable from the background. PET/MR images reconstruction PET images were reconstructed using a three-dimensional ordered-subset expectation maximization (3D OSEM) algorithm with attenuation and scatter correction, with anatomical reference provided by a T2-weighted TurboRARE sequence (TR/TE = 2400/33 ms; resolution = 0.15 × 0.15 × 1.0 mm 3 ). Image analysis was performed using PMOD software (Version 4.4; Switzerland). Regions of interest (ROIs) were manually delineated on fused PET/MR images to quantify radioactivity concentrations (%ID/g). Bioluminescence-based cytotoxicity assay To systematically evaluate whether the NOTA-BED probe affects the cytotoxic function of CAR T cells, a co-culture system of effector and target cells was established. Raji-Luc-GFP (CD19+) and U266-Luc-GFP (BCMA+) cells were used as target cells, while BC19 CAR T cells and Mock-T cells served as effectors. Initially, various effector-to-target (E:T) ratios (4:1, 2:1, 1:1, and 1:2; n = 3) were tested to determine the optimal killing conditions. Subsequently, an interference assay was performed at a fixed E:T ratio of 2:1 by introducing graded concentrations of NOTA-BED (0, 0.1, 1, 10, and 100 nM; n = 4). All assays were conducted in black 96-well plates, in which 1 × 10 4 target cells per well were co-incubated with effector cells and probes in a final volume of 200 µL for 48 hours at 37°C. For bioluminescence detection, a solution containing 0.3 mg/ml D-luciferin potassium salt (Ambeed, Catalog No.A162108) and 2% Triton X-100 was added to each well. After incubation for 5 min in the dark, bioluminescence images were acquired using an IVIS imaging system (PerkinElmer, USA). Total photon flux (photons/s) was quantified using Living Image software (PerkinElmer). Specific lysis was calculated as: $$\:\text{S}\text{p}\text{e}\text{c}\text{i}\text{f}\text{i}\text{c}\:\text{l}\text{y}\text{s}\text{i}\text{s}\:\left(\text{\%}\right)=\:\left[1\:-\frac{\left(\text{S}\text{a}\text{m}\text{p}\text{l}\text{e}\:\text{f}\text{l}\text{u}\text{x}\:-\:\text{B}\text{a}\text{c}\text{k}\text{g}\text{r}\text{o}\text{u}\text{n}\text{d}\:\text{f}\text{l}\text{u}\text{x}\right)}{\left(\text{M}\text{e}\text{a}\text{n}\:\text{t}\text{a}\text{r}\text{g}\text{e}\text{t}\:\text{c}\text{o}\text{n}\text{t}\text{r}\text{o}\text{l}\:\text{f}\text{l}\text{u}\text{x}\:-\:\text{B}\text{a}\text{c}\text{k}\text{g}\text{r}\text{o}\text{u}\text{n}\text{d}\:\text{f}\text{l}\text{u}\text{x}\right)}\right]\text{*}\:100\text{\%}$$ 1 CAR-specific lysis was calculated by normalizing CAR T sample flux to Mock-T control flux: $$\:\text{C}\text{A}\text{R}-\text{s}\text{p}\text{e}\text{c}\text{i}\text{f}\text{i}\text{c}\:\text{l}\text{y}\text{s}\text{i}\text{s}\:\left(\text{\%}\right)=\:\left[1\:-\frac{\left(\text{C}\text{A}\text{R}\:\text{T}\:\text{s}\text{a}\text{m}\text{p}\text{l}\text{e}\:\text{f}\text{l}\text{u}\text{x}\right)}{\left(\text{M}\text{e}\text{a}\text{n}\:\text{M}\text{o}\text{c}\text{k}-\text{T}\:\text{c}\text{o}\text{n}\text{t}\text{r}\text{o}\text{l}\:\text{f}\text{l}\text{u}\text{x}\right)}\right]\text{*}\:100\text{\%}$$ 2 Cytokine release analysis To further assess the immune activation and effector function of CAR T cells under probe intervention, parallel co-culture assays corresponding to the interference experiment (fixed E:T ratio of 2:1) were performed under identical conditions. After 48 hours of co-incubation, cell culture supernatants were collected and centrifuged at 350 g for 5 min to remove residual cells and debris. Supernatants were diluted 1:10 with the provided sample diluent, and cytokine levels were measured using human interferon-γ (IFN-γ) and tumor necrosis factor-α (TNF-α) enzyme-linked immunosorbent assay (ELISA) kits (Elabscience; Catalog Nos. Catalog No. E-EL-H0108 for IFN-γ and Catalog No. E-EL-H0109 for TNF-α) according to the manufacturer’s instructions. Absorbance was measured at 450 nm using a microplate reader (BioTeK, American). Absolute cytokine concentrations were determined by fitting sample absorbance values to standard curves generated using a four-parameter logistic regression model. Establishment of tumor-bearing models and therapeutic protocols All animal experiments were conducted in strict accordance with the national regulations for animal welfare and were approved by the Institutional Animal Care and Use Committee (IACUC) of the Jiangsu Institute of Nuclear Medicine (Approval No. JSINM-2025-097). To simulate localized tumor burden and systemic dissemination, composite subcutaneous-systemic models were established in 4–6-week-old female NCG mice. For the Raji lymphoma model, mice received a subcutaneous injection of 1 × 10 6 Raji-Luc cells on Day − 10, followed by a tail vein injection of 2 × 10 5 Raji-Luc cells on Day − 2. On Day 0, mice were screened via bioluminescence imaging; those with a total flux of approximately 1 × 10 7 p/s in the subcutaneous tumor were randomly assigned to the CAR-T group (5 × 10 6 BC19 CAR-Tcells, i.v), or the PBS group (equivalent volume PBS, i.v). The U266 myeloma model was established similarly, with subcutaneous injection (3.5 × 10 6 U266-Luc cells) on Day − 7 and intravenous injection (8 × 10 5 U266 -Luc cells) on Day − 2. On Day 0, mice with a subcutaneous flux of approximately 2 × 10 8 p/s were randomized into CAR-T (5 × 10 6 BC19 CAR-T cells, i.v), Mock-T (5 × 10 6 Mock-T cells, i.v), or PBS groups (equivalent volume PBS, i.v). Multimodal longitudinal imaging and therapeutic evaluation Tumor burden and CAR T cell trafficking were longitudinally monitored using synchronized bioluminescence imaging and PET/MR imaging. Bioluminescence imaging was performed at baseline (Day 0) and all subsequent follow-up points to assess tumor progression. Following CAR T cell infusion, serial PET/MR imaging was conducted in tandem with Bioluminescence imaging on Days 1, 8, 12, 17, and 22 for the Raji model, and on Days 4, 11, 14, 18, and 21 for the U266 model. For PET/MR imaging, mice received approximately 3.7 MBq of the radiotracer via tail vein injection. After a 1 hour uptake period, 10-min static PET/MRI acquisitions were performed using a Bruker 9.4T small-animal PET/MR system (Bruker BioSpec 94/30, Germany). Image analysis was conducted using PMOD software (version 4.4, Switzerland), with radioactivity uptake quantified as the percentage of injected dose per gram of tissue (%ID/g) based on volumes of interest (VOIs) delineated on fused PET/MR images. Bioluminescence imaging was performed 5–10 min after intraperitoneal injection of D-luciferin (30 mg/mL, 100 µL), and total photon flux (photons/s) was quantified using Living Image software (PerkinElmer, American). Body weight monitoring and survival analysis To assess the therapeutic efficacy of CAR T cell treatment and evaluate potential systemic toxicity associated with repeated PET/MR imaging, body weight and clinical status were recorded every 3–4 days following CAR T cell infusion. The humane endpoints were defined as a body weight loss exceeding 20% of the baseline or manifestation of severe cachexia. Animals that did not reach humane endpoint criteria were euthanized on Day 23, which was designated as the experimental endpoint. Survival time was recorded for each animal and utilized for subsequent survival analysis. Histopathological and immunohistochemical analysis At the experimental endpoint, mice were euthanized for tissue harvesting. Major organs, including the heart, liver, spleen, lungs, and kidneys, as well as tumor tissues, were collected, fixed in 4% paraformaldehyde for 24 hours, and paraffin-embedded. Serial sections (4 µm) were prepared for histological and immunohistochemical (IHC) analyses. For histopathological evaluation, sections were stained with hematoxylin and eosin (H&E) to assess tissue morphology and potential treatment-related toxicity. For IHC analysis, tumor sections were incubated with anti-human CD3 (1:500, clone CAL54; Abcam, cat. no. ab237707) to identify infiltrating CAR T cells, and with either anti-CD19 (5 µg/mL; R&D Systems, cat. no. MAB48671) or anti-BCMA (5 µg/mL; R&D Systems, cat. no. MAB10762) antibodies to visualize tumor-associated antigens. Slides were digitally scanned, and representative images were acquired. CAR T cell distribution and infiltration patterns across different treatment groups were qualitatively assessed using ImageJ software. Statistical Analysis All quantitative data are expressed as mean ± standard deviation (SD). Statistical analyses and graphical representations were performed using GraphPad Prism (version 8.0). For comparisons between two independent groups, unpaired two-tailed Student’s t-tests were applied. For comparisons involving more than two groups, one-way analysis of variance (ANOVA) was used; Welch’s ANOVA was applied when variance homogeneity was not satisfied, followed by Tukey’s multiple comparisons post hoc test. Longitudinal datasets, including tumor volume and body weight over time, were analyzed using two-way repeated-measures ANOVA with Tukey’s post hoc correction. For survival analysis, an event was defined as death, body weight loss exceeding 20% of baseline, or tumor volume reaching 1500 mm 3 . Survival curves were generated using the Kaplan–Meier method and compared using the log-rank (Mantel–Cox) test. All statistical tests were two-tailed, and a P value 0.05), *P < 0.05, ** P < 0.01, *** P < 0.001, and **** P < 0.0001. Declarations Author contributions W.Z., X.W., X.Z. and M.Y. conceived the study and designed the experiments. W.Z. and X.W. conducted the experiments and collected and analysed the data. R.H., J.Y., D.P., C.C., Y.X., L.W. and M.S. contributed experimental or analysis tools. W.Z., X.W., X.Z. and M.Y. wrote the manuscript. All authors carefully reviewed and approved the manuscript. Acknowledgements This work was supported by the National Natural Science Foundation of China (grant No. 32371434, 12575361), and Wuxi Taihu Light Science and Technology Project, grant No. K20253011, K20251002. Data availability The authors declare that the main data supporting the findings of this study are available within the Article and its Supplementary Information. The corresponding author will make raw data and step-by-step protocols available upon request. References June CH, O’Connor RS, Kawalekar OU, Ghassemi S (2018) Milone MC. CAR T cell immunotherapy for human cancer. Science 359:1361–1365 Munshi NC et al (2021) Idecabtagene vicleucel in relapsed and refractory multiple myeloma. N Engl J Med 384:705–716 Westin JR et al (2023) Survival with axicabtagene ciloleucel in large B-cell lymphoma. N Engl J Med 389:148–157 Samur MK et al (2021) Biallelic loss of BCMA as a resistance mechanism to CAR T cell therapy in a patient with multiple myeloma. Nat Commun 12:868 Simon S, Riddell SR (2020) Dual targeting with CAR T cells to limit antigen escape in multiple myeloma. Blood Cancer Discov 1:130–133 Hirabayashi K et al (2021) Dual-targeting CAR-T cells with optimal co-stimulation and metabolic fitness enhance antitumor activity and prevent escape in solid tumors. Nat Cancer 2:904–918 Shi M et al (2024) Bispecific CAR T cell therapy targeting BCMA and CD19 in relapsed/refractory multiple myeloma: a phase I/II trial. Nat Commun 15:3371 Huang X et al (2026) BCMA/CD19 CAR T cell therapy for refractory myasthenia gravis: Proteomic signatures and single-cell transcriptomics of disease flares. Sci Adv 12:eaeb6424 Volpe A, Pham J, Sellmyer MA, Ponomarev V (2025) Practical considerations for clinical translation of PET imaging of adoptive cell therapies. npj Imaging 3:59 Shishido SN et al (2024) Liquid biopsy approach to monitor the efficacy and response to CAR-T cell therapy. J Immunother Cancer 12:e007329 Volpe A et al (2020) Spatiotemporal PET imaging reveals differences in CAR-T tumor retention in triple-negative breast cancer models. Mol Ther 28:2271–2285 Krebs S, Ponomarev V, Slovin S, Schöder H (2019) Imaging of CAR T-cells in cancer patients: paving the way to treatment monitoring and outcome prediction. J Nucl Med 60:879–881 Weist MR et al (2018) PET of adoptively transferred chimeric antigen receptor T cells with 89Zr-oxine. J Nucl Med 59:1531–1537 Kim S-Y et al (2024) Direct and indirect chimeric antigen receptor T-Cell imaging with PET/MRI in a tumor xenograft model. Radiology 310:e231406 Wang X-y et al (2021) Feasibility study of 68Ga-labeled CAR T cells for in vivo tracking using micro-positron emission tomography imaging. Acta Pharmacol Sin 42:824–831 Kurebayashi Y, Choyke PL, Sato N (2021) Imaging of cell-based therapy using 89Zr-oxine ex vivo cell labeling for positron emission tomography. Nanotheranostics 5:27 Leland P, Kumar D, Nimmagadda S, Bauer SR, Puri RK, Joshi BH (2023) Characterization of chimeric antigen receptor modified T cells expressing scFv-IL-13Rα2 after radiolabeling with 89Zirconium oxine for PET imaging. J Transl Med 21:367 Keu KV et al (2017) Reporter gene imaging of targeted T cell immunotherapy in recurrent glioma. Sci Transl Med 9:eaag2196 Morath V et al (2025) PET-based tracking of CAR T cells and viral gene transfer using a cell surface reporter that binds to lanthanide complexes. Nat Biomed Eng, 1–21 Sellmyer MA et al (2020) Imaging CAR T cell trafficking with eDHFR as a PET reporter gene. Mol Ther 28:42–51 Song X et al (2024) Noninvasive longitudinal PET/CT imaging of CAR T cells using PSMA reporter gene. Eur J Nucl Med Mol Imaging 51:965–977 Minn I et al (2019) Imaging CAR T cell therapy with PSMA-targeted positron emission tomography. Sci Adv 5:eaaw5096 Zhang N et al (2025) Development of an in situ CAR-T cell protocol through optical and PSMA-targeted PET imaging. P Natl Acad Sci USA 122:e2504950122 Murty S et al (2020) PET reporter gene imaging and ganciclovir-mediated ablation of chimeric antigen receptor T cells in solid tumors. Cancer Res 80:4731–4740 Fröse J et al (2024) Development of an antigen-based approach to noninvasively image CAR T cells in real time and as a predictive tool. Sci Adv 10:eadn3816 Bergström M, Grahnen A, Långström B (2003) Positron emission tomography microdosing: a new concept with application in tracer and early clinical drug development. Eur J Clin Pharmacol 59:357–366 Zou S et al (2025) Non-invasive assessment of HER2 expression in patients with urothelial carcinoma using [68Ga] Ga-HER2 affibody PET/CT imaging: preliminary clinical findings. Eur J Nucl Med Mol Imaging 52:2782–2791 Xu Y et al (2019) PET imaging of a 68Ga labeled modified HER2 affibody in breast cancers: from xenografts to patients. Br J Radiol 92:20190425 Zhou N et al (2021) Impact of 68Ga-NOTA-MAL-MZHER2 PET imaging in advanced gastric cancer patients and therapeutic response monitoring. Eur J Nucl Med Mol Imaging 48:161–175 Rodnick ME et al (2022) Synthesis of 68Ga-radiopharmaceuticals using both generator-derived and cyclotron-produced 68Ga as exemplified by [68Ga] Ga-PSMA-11 for prostate cancer PET imaging. Nat Protoc 17:980–1003 Wei J et al (2020) The model of cytokine release syndrome in CAR T-cell treatment for B-cell non-Hodgkin lymphoma. Signal Transduct Target Ther 5:134 Ghorashian S et al (2019) Enhanced CAR T cell expansion and prolonged persistence in pediatric patients with ALL treated with a low-affinity CD19 CAR. Nat Med 25:1408–1414 Maecke HR, Hofmann M (2005) Haberkorn U. 68Ga-labeled peptides in tumor imaging. J Nucl Med 46:172S–178S Schwarz SW, Oyama R (2015) The role of exploratory investigational new drugs for translating radiopharmaceuticals into first-in-human studies. J Nucl Med 56:497–500 Decristoforo C, Lyashchenko SK (2019) Recommendations for conducting clinical trials with radiopharmaceuticals. In: (ed^ (ed) Nuclear Medicine Textbook: Methodology and Clinical Applications. Springer Korde A et al (2022) Practical considerations for navigating the regulatory landscape of non-clinical studies for clinical translation of radiopharmaceuticals. EJNMMI Radiopharm Chem 7:18 Satpati D (2021) Recent breakthrough in 68Ga-radiopharmaceuticals cold kits for convenient PET radiopharmacy. Bioconjug Chem 32:430–447 Gillings N et al (2021) Guideline on current good radiopharmacy practice (cGRPP) for the small-scale preparation of radiopharmaceuticals. EJNMMI Radiopharm Chem 6:8 Liu L et al (2022) Computational model of CAR T-cell immunotherapy dissects and predicts leukemia patient responses at remission, resistance, and relapse. J Immunother Cancer 10:e005360 Kirouac DC, Zmurchok C, Deyati A, Sicherman J, Bond C, Zandstra PW (2023) Deconvolution of clinical variance in CAR-T cell pharmacology and response. Nat Biotechnol 41:1606–1617 Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryMaterial.docx Dataset 1 Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9063774","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":632075739,"identity":"8147bdb9-2753-4c7b-8f6c-bdba4d3f8083","order_by":0,"name":"Min Yang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyElEQVRIiWNgGAWjYNACAwY5xgbGBtK0GJOqhYEhkXj1BsfPHn51o+BOenP/4bZHNxhs8uUdCGk5k5dmnWPwLLdxRmK7cQ5DmuXGA4S0HMgxM84xOAzUwtgmncNw2MCQkBMNzr8Ba0ln7D9IrJYbOcaPgVoSGBsSIVrkCehgkLzxxowZqMUQ6BegFoM0AwNCWvjO5xh/zvlzWN6w//gz6ZwKGwN5Qg5TOMDAJgFiQLxgAAoQAlqAZjJ/ADOQREbBKBgFo2AUoAAAbGhDkKC1BSUAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-6976-8526","institution":"Jiangsu Institute of Nuclear Medicine","correspondingAuthor":true,"prefix":"","firstName":"Min","middleName":"","lastName":"Yang","suffix":""},{"id":632075740,"identity":"8cee23cd-4d41-4ee3-bb6a-c1604b4d3ea9","order_by":1,"name":"Wenyao Zhou","email":"","orcid":"","institution":"Jiangsu Institute of Nuclear Medicine","correspondingAuthor":false,"prefix":"","firstName":"Wenyao","middleName":"","lastName":"Zhou","suffix":""},{"id":632075741,"identity":"219673e1-fa48-464b-a7de-bf24bc7db7e7","order_by":2,"name":"Xinyu Wang","email":"","orcid":"https://orcid.org/0000-0002-9167-2077","institution":"Jiangsu Institute of Nuclear Medicine","correspondingAuthor":false,"prefix":"","firstName":"Xinyu","middleName":"","lastName":"Wang","suffix":""},{"id":632075742,"identity":"2e37ee3a-8eec-4d21-8c38-64c25e556a71","order_by":3,"name":"Rui Hou","email":"","orcid":"","institution":"Xuzhou Medical University","correspondingAuthor":false,"prefix":"","firstName":"Rui","middleName":"","lastName":"Hou","suffix":""},{"id":632075743,"identity":"6552554d-23d4-4b31-a1d3-7f9d3dd4704d","order_by":4,"name":"Junjie Yan","email":"","orcid":"https://orcid.org/0000-0001-8016-2277","institution":"Jiangsu Institute of Nuclear Medicine","correspondingAuthor":false,"prefix":"","firstName":"Junjie","middleName":"","lastName":"Yan","suffix":""},{"id":632075744,"identity":"35ae252f-e8ab-41a7-b687-d68d0462110c","order_by":5,"name":"Donghui Pan","email":"","orcid":"","institution":"Jiangsu Institute of Nuclear Medicine","correspondingAuthor":false,"prefix":"","firstName":"Donghui","middleName":"","lastName":"Pan","suffix":""},{"id":632075745,"identity":"4e54754f-203a-404e-b8a4-f8cff3a4246e","order_by":6,"name":"Chongyang Chen","email":"","orcid":"","institution":"Jiangsu Institute of Nuclear Medicine","correspondingAuthor":false,"prefix":"","firstName":"Chongyang","middleName":"","lastName":"Chen","suffix":""},{"id":632075746,"identity":"440c48a5-ae74-44dd-b086-aac859589016","order_by":7,"name":"Yuping Xu","email":"","orcid":"","institution":"Jiangsu Institute of Nuclear Medicine","correspondingAuthor":false,"prefix":"","firstName":"Yuping","middleName":"","lastName":"Xu","suffix":""},{"id":632075747,"identity":"9755e166-c5bb-4fc5-b09c-8a930e6a4586","order_by":8,"name":"Lizhen Wang","email":"","orcid":"","institution":"Jiangsu Institute of Nuclear Medicine","correspondingAuthor":false,"prefix":"","firstName":"Lizhen","middleName":"","lastName":"Wang","suffix":""},{"id":632075748,"identity":"2916b739-c6fb-4b6c-8f63-308e561b4d2f","order_by":9,"name":"Ming Shi","email":"","orcid":"","institution":"Xuzhou Medical University","correspondingAuthor":false,"prefix":"","firstName":"Ming","middleName":"","lastName":"Shi","suffix":""},{"id":632075749,"identity":"277e4b69-d1e2-4f8a-b462-7290b1153892","order_by":10,"name":"Xueyan Zhou","email":"","orcid":"","institution":"Xuzhou Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xueyan","middleName":"","lastName":"Zhou","suffix":""}],"badges":[],"createdAt":"2026-03-08 11:10:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9063774/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9063774/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109332864,"identity":"4f405107-190a-491d-b1cc-5975874c0de7","added_by":"auto","created_at":"2026-05-15 16:22:04","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":6818376,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eConcept of BCMA antigen-based [\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e68\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003eGa]Ga-NOTA-BED probe for longitudinal PET/MR monitoring of BC19 CAR-T cell kinetics.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDual-targeted BC19 (CD19/BCMA) CAR-T cells were engineered with a tandem scFv construct to recognize CD19+ lymphoma and BCMA+ multiple myeloma. To track their in vivo behavior, we developed a standardized PET imaging strategy using [\u003csup\u003e68\u003c/sup\u003eGa]Ga-NOTA-BED, a recombinant probe derived from the extracellular domain of human BCMA (BED). The probe binds to the anti-BCMA scFv within the CAR construct, enabling imaging under antigen-noncompetitive conditions in CD19+ tumors and antigen-competitive conditions in BCMA+ tumors. Following infusion of BC19 CAR-T cells into tumor-bearing mice, longitudinal [\u003csup\u003e68\u003c/sup\u003eGa]Ga-NOTA-BED PET/MR imaging enables noninvasive visualization and quantification of CAR T-cell kinetics over time.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-9063774/v1/f11494f95412fcfcfde18e94.png"},{"id":109332858,"identity":"bb993871-85b7-49ba-9dcd-5cbb57d12331","added_by":"auto","created_at":"2026-05-15 16:22:03","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":28326798,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003e[\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e68\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003eGa]Ga-NOTA-BED enables specific and sensitive CAR-T cell detection in vitro and quantitative tracking in vivo.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Time-dependent cell uptake and competitive blocking. Radioactivity accumulation in BC19 CAR T and Mock-T cells following incubation with [\u003csup\u003e68\u003c/sup\u003eGa]Ga-NOTA-BED for 1 or 2 h. Competitive blocking was performed by the addition of a 1000-fold molar excess of unlabeled NOTA-BED. (B) Tracer uptake across different CAR phenotypes. [\u003csup\u003e68\u003c/sup\u003eGa]Ga-NOTA-BED binding in T cells engineered with BCMA CAR, BC19 CAR, CD19 CAR, or Mock-T constructs. (C) In vitro competition assay in tumor co-culture models. Probe uptake by BC19 CAR-T cells in the presence of BCMA-negative Raji cells or BCMA-high U266 cells at varying effector-to-target (E:T) ratios. (D–E) In vivo targeting specificity in a dual-axillary cell-cluster model. (D) Representative coronal and axial PET images of healthy BALB/c mice acquired 1 h post-injection of [\u003csup\u003e68\u003c/sup\u003eGa]Ga-NOTA-BED. Mice were subcutaneously inoculated with 5 × 10\u003csup\u003e6\u003c/sup\u003e engineered T cells (Mock-T, CD19 CAR T, BCMA CAR T, or BC19 CAR T) into the bilateral axillary regions immediately prior to probe administration. (E) Quantitative ROI analysis of radioactivity uptake in the indicated cell-inoculation sites and muscle background.(F) In vivo limit of detection (LOD). Representative PET/MR fusion images of mice inoculated with serial dilutions of BC19 CAR T cells ranging from 5 × 10\u003csup\u003e6\u003c/sup\u003e to 5 × 10\u003csup\u003e3\u003c/sup\u003e. (G–H) Quantitative imaging analysis. (G) Tracer uptake at the injection sites plotted against the number of implanted CAR-T cells. (H) Pearson correlation analysis between PET signal intensity (%ID/g) and the log-transformed number of injected BC19 CAR-T cells. Data are presented as mean ± SD from at least three independent experiments (n = 3 per group). ns, not significant (\u003cem\u003eP\u003c/em\u003e \u0026gt; 0.05); *\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.05, **\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.01, ***\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.001, ****\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.0001 by one-way or two-way ANOVA; Pearson correlation was used for linear correlation analysis.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-9063774/v1/94e924c195b72cc0c041a3b1.png"},{"id":109332865,"identity":"9cd27f3e-e383-4876-8513-736c36b42430","added_by":"auto","created_at":"2026-05-15 16:22:04","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":3144095,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLongitudinal PET/MR tracking of CAR T-cell dynamics under non-antigen-competitive conditions in Raji lymphoma models.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Experimental design for longitudinal [\u003csup\u003e68\u003c/sup\u003eGa]Ga-NOTA-BED PET/MR and bioluminescence imaging (IVIS) in Raji tumor–bearing mice. (B) Spatiotemporal visualization of CAR T-cell distribution and tumor burden. Representative longitudinal IVIS, PET/MR, and maximum intensity projection (MIP) images of BC19 CAR T–treated mice (categorized by response kinetics: rapid R2, intermediate R1, and slow R5) and PBS controls (P3). White dashed lines delineate tumor regions. Red “×” symbols indicate animal mortality due to tumor progression or humane endpoint criteria. (C–F) Individual kinetic profiles. Temporal changes in tumor burden (Total flux, blue curves) and intratumoral CAR T-cell signal (tracer uptake, %ID/g, pink curves) for representative mice in the rapid (C), intermediate (D), slow (E), and PBS (F) groups. (G–H) Quantitative assessment of peak tracer accumulation. Comparison of peak radioactivity uptake (%ID/g\u003csub\u003epeak\u003c/sub\u003e, G) and peak tumor-to-muscle ratios (T/M\u003csub\u003epeak\u003c/sub\u003e, H) between treatment and control groups. (I) Individual longitudinal uptake curves. Longitudinal tracer uptake (%ID/g) over time for individual mice in the treatment (pink) and PBS (blue) groups. (J) Quantitative bioluminescence analysis. Longitudinal monitoring of subcutaneous tumor total flux. (K) Body weight monitoring in BC19 CAR T and PBS groups during the treatment period. (L) Kaplan–Meier survival analysis comparing BC19 CAR T and PBS groups. (M) Endpoint immunohistochemical (IHC) analysis. Representative IHC images of tumor or residual tissues stained for CD19 (target antigen) and CD3 (T-cell infiltration) in BC19 CAR T and PBS groups. Red arrows indicate infiltrating CD3-positive cells. Scale bars: 100 μm (main images) and 50 μm (zoom-in insets). Data are presented as mean ± SD (n = 5 per group). *P \u0026lt; 0.05, **P \u0026lt; 0.01, ***P \u0026lt; 0.001, ****P \u0026lt; 0.0001 by one-way or two-way ANOVA, or by log-rank (Mantel–Cox) test for survival analysis.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-9063774/v1/969fea0d991f1b08da248833.png"},{"id":109405451,"identity":"a9e7fff1-4d6d-4ee4-b2a6-be6b62eed072","added_by":"auto","created_at":"2026-05-17 13:18:14","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":3091230,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLongitudinal PET/MR tracking of CAR-T cell dynamics under antigen-competitive conditions in U266 multiple myeloma models.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Experimental design for serial [\u003csup\u003e68\u003c/sup\u003eGa]Ga-NOTA-BED PET/MR and IVIS bioluminescence imaging in U266 tumor-bearing mice. (B) Spatiotemporal visualization of CAR-T cell distribution and tumor burden. Representative longitudinal IVIS (left) and PET/MR (right) images of mice in the BC19 CAR-T treated group (stratified into Cure and Progress subgroups), MOCK-T group, and PBS group from day 0 to 21. White dashed lines delineate tumor regions. Red “×” symbols indicate animal mortality due to tumor progression or humane endpoint criteria. (C-F) Individual kinetic profiles. Dynamic changes in tumor burden (total flux, blue curves) and intratumoral tracer uptake (%ID/g mean, red curves) for representative mice: Cure subgroup (U3, C), Progress subgroup (U5, D), MOCK-T group (U-M1, E), and PBS group (U-P1, F). (G-H) Quantitative assessment of peak tracer accumulation. Comparison of peak tumor uptake (%ID/g\u003csub\u003epeak\u003c/sub\u003e, G) and peak tumor-to-muscle ratios (T/M\u003csub\u003epeak\u003c/sub\u003e, H) across the indicated experimental groups. (I) Individual longitudinal uptake profiles. Longitudinal tracer uptake (%ID/g) over time for individual mice in the Cure (pink, n = 4), Progress (purple, n = 2), MOCK-T (blue, n = 3), and PBS (gray, n = 3) cohorts. (J) Quantitative bioluminescence analysis. Longitudinal monitoring of subcutaneous tumor total flux in BC19 subgroups and control cohorts. (K) Body weight monitoring in BC19-treated (n = 6), MOCK-T (n = 3), and PBS (n = 3) mice during the study period. (L) Kaplan-Meier survival analysis comparing in BC19-treated (n = 6), MOCK-T (n = 3), and PBS (n = 3) mice during the study period. (M) Histological and immunohistochemical (IHC) analysis. Representative H\u0026amp;E, BCMA (target antigen), and CD3 (T-cell infiltration) staining of tumor or residual tissues. Red arrows indicate infiltrating CD3+ cells. Scale bars, 100 μm (main images) and 50 μm (zoom-in insets). Data are presented as mean ± SD. *P \u0026lt; 0.05, **P \u0026lt; 0.01, ***P \u0026lt; 0.001, ****P \u0026lt; 0.0001 by one-way or two-way ANOVA; survival differences were assessed by log-rank (Mantel-Cox) test.\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-9063774/v1/b917265554b3eb68498c714d.png"},{"id":109405275,"identity":"b465ee8a-e3e0-4103-b6b3-a04ca6781a8e","added_by":"auto","created_at":"2026-05-17 13:15:32","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":3045680,"visible":true,"origin":"","legend":"Dataset 1","description":"","filename":"SupplementaryMaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-9063774/v1/32134f9f7b54a0bc2d1be829.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"\u003cp\u003eA \u003csup\u003e68\u003c/sup\u003eGa–BCMA Ectodomain Tracer Enables PET/MR Longitudinal Tracking of Clinical-Grade BCMA/CD19 CAR T Cells\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eChimeric antigen receptor (CAR) T cell therapies directed against CD19 and B cell maturation antigen (BCMA) have transformed treatment paradigms for B cell malignancies and multiple myeloma, delivering high response rates in heavily pretreated populations and establishing living cell products as a major therapeutic class\u003csup\u003e1\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e2\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e3\u003c/sup\u003e. However, heterogeneous responses and relapse remain frequent and are often linked to antigen heterogeneity/escape and insufficient in vivo expansion or persistence, limiting durable disease control. In BCMA‑directed settings, tumor‑intrinsic resistance mechanisms such as biallelic loss of BCMA have been documented after initial response, underscoring the vulnerability of single‑antigen targeting and the need for strategies that both broaden antigen coverage and enable mechanistic monitoring of treatment failure\u003csup\u003e4\u003c/sup\u003e. Dual‑target CAR designs have therefore gained momentum as a rational countermeasure to antigen escape\u003csup\u003e5\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e6\u003c/sup\u003e. In relapsed/refractory multiple myeloma, a bispecific BCMA/CD19 CAR (“BC19”) demonstrated feasibility, tolerability, and encouraging clinical activity in a phase I/II trial, supporting the translational rationale for BCMA/CD19 co‑targeting\u003csup\u003e7\u003c/sup\u003e. Beyond oncology, a homologous BCMA/CD19 CAR platform has been explored in refractory myasthenia gravis, enabling deep immune profiling of therapeutic responses and disease flares and further highlighting the clinical relevance of this CAR lineage across indications\u003csup\u003e8\u003c/sup\u003e. These advances sharpen a central unmet need: noninvasive, quantitative tools that can map where CAR T cells traffic and how they expand over time in the same subject, rather than inferring kinetics from sparse peripheral blood sampling or endpoint biopsies\u003csup\u003e9\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e10\u003c/sup\u003e. \u003c/p\u003e\n\u003cp\u003ePositron emission tomography (PET) offers a whole‑body, quantitative readout that can, in principle, report on cell therapy biodistribution, tumor infiltration, expansion, persistence, and on‑target/off‑tumor risks, thereby enabling iterative optimization of dosing and product design\u003csup\u003e9\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e11\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e12\u003c/sup\u003e. Yet, repeatable PET tracking of living, proliferating CAR T cells remains challenging. Direct ex vivo radiolabeling enables immediate tracking with low background but rapidly loses interpretability after homing because signal is diluted by cell division and confounded by label loss or redistribution upon cell death\u003csup\u003e13\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e14\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e15\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e16\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e17\u003c/sup\u003e. Indirect reporter gene imaging avoids dilution by embedding a stable imaging tag, but it requires additional genetic engineering and can introduce operational complexity, regulatory hurdles, immunogenicity concerns, and reporter‑dependent physiologic background\u003csup\u003e18\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e19\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e20\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e21\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e22\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e23\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e24\u003c/sup\u003e. These constraints help explain why robust, longitudinal CAR T PET remains uncommon despite compelling proof‑of‑concept demonstrations in both preclinical models and patients.\u003c/p\u003e\n\u003cp\u003eAn emerging alternative is antigen‑based CAR‑PET, which uses the CAR’s target antigen as the imaging probe—aiming to preserve the timing flexibility of injectable tracers while avoiding any modification of the therapeutic cells\u003csup\u003e25\u003c/sup\u003e. In a recent study, the soluble CD19 ectodomain (~32 kDa) was proposed as an “ideal” probe to image CD19 CAR T cells and explicitly emphasized compatibility with currently FDA‑approved CAR T products without changing established clinical workflows\u003csup\u003e25\u003c/sup\u003e. However, translating antigen‑based CAR‑PET into a high‑frequency longitudinal imaging paradigm requires careful optimization of probe size, clearance, binding, and functional non‑interference—particularly when the goal is to capture expansion dynamics over weeks.\u003c/p\u003e\n\u003cp\u003eHere, we extend the antigen‑based CAR‑PET paradigm to a clinically used tandem scFv BCMA/CD19 CAR T platform by developing an engineering‑free PET/MR tracer derived from a minimal BCMA extracellular domain fragment (BED) with a molecular weight of 5,396 Da. This small antigen probe is radiolabeled with \u003csup\u003e68\u003c/sup\u003eGa (t\u003csub\u003e1/2\u003c/sub\u003e ≈ 67.7 min) to bind the anti‑BCMA scFv module on the CAR surface, thereby enabling rapid post‑scan clearance and low‑burden repeat imaging. Leveraging this fast‑clearing probe with PET/MR, we perform five scans over 22 days to quantify CAR T trafficking and proliferation kinetics during treatment of BCMA+ U266 and CD19+ Raji xenografts. This framework enables serial, whole‑body mapping of dual‑target CAR T expansion and response heterogeneity while avoiding the key limitations of dilution‑prone direct labeling and the added genetic modification required for reporter gene strategies.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003eDesign and Synthesis of a BCMA CAR Targeted Probe for Monitoring BC19 CAR T Cells\u003c/h2\u003e \u003cp\u003eBCMA is a defining surface marker of malignant plasma cells and a central therapeutic target in multiple myeloma\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. As tumor heterogeneity and antigen escape increasingly motivate dual‑targeted or multispecific CAR platforms, imaging tools that remain compatible with diverse CAR architectures are needed for spatiotemporal monitoring in vivo. To meet this need, we developed an antigen‑based imaging probe derived from the core extracellular region of human BCMA (BCMA extracellular domain, BED; amino acid 5\u0026ndash;54; calculated MW 5.4 kDa). BED contains the key epitope recognized by anti‑BCMA scFvs, providing direct compatibility not only with monospecific BCMA CAR T cells but also with multispecific constructs that include a BCMA‑binding arm, such as tandem scFv BCMA/CD19 CAR T cells (BC19 CAR T) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). This design bypasses additional genetic modification of CAR T cells and is intended as a generalizable imaging strategy for combinatorial CAR formats that retain an anti‑BCMA recognition module.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eStructural modeling (PyMOL) suggested a binding mode in which BED engages the anti‑BCMA scFv within the dual‑scFv (tandem) BC19 CAR configuration (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). Given its compact size, BED is predicted to impose minimal steric hindrance and to preserve access to the spatially distinct anti‑CD19 scFv arm, thereby reducing the likelihood of functional interference in multispecific CAR settings (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe BED probe was produced in an \u003cem\u003eEscherichia coli\u003c/em\u003e expression system using an N-terminal 6\u0026times;His-SUMO fusion strategy (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). The His tag enabled Ni‑NTA affinity purification, and the SUMO moiety enhanced solubility and folding. Following site‑specific cleavage by Ulp1 and polishing by size‑exclusion chromatography, BED was obtained at \u0026gt;\u0026thinsp;98% purity with a yield of ~\u0026thinsp;5 mg/L. SDS‑PAGE and time‑of‑flight mass spectrometry confirmed the expected molecular mass and integrity of the final product (Fig. S1A, Fig. S2A), supporting its suitability for subsequent radiochemistry and in vivo studies.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eEffector cell validation and physicochemical characterization of BED and NOTA‑BED\u003c/h2\u003e \u003cp\u003eWe first validated CAR expression on engineered T cells by flow cytometry. CD19 CAR T cells showed 93.6% positivity by anti‑FMC63 staining, whereas BCMA CAR T cells showed 77.1% positivity by BCMA antigen binding. Bispecific BC19 CAR T cells exhibited the expected dual‑targeting phenotype, with 68.5% of cells double‑positive for BCMA and CD19 binding; mock‑transduced T cells showed negligible binding (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). These data establish a robust cellular platform for evaluating BED‑based imaging.\u003c/p\u003e \u003cp\u003eBio‑layer interferometry (BLI) demonstrated high‑affinity binding of unmodified BED to the anti‑BCMA scFv (K\u003csub\u003eD\u003c/sub\u003e = 0.89 nM; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). For PET probe construction, a NOTA‑BED conjugate was synthesized and purified. SDS‑PAGE showed a single band with a slight upward shift compared to BED, consistent with successful chelator conjugation and high purity (Fig. S1A). TOF‑MS confirmed a molecular mass of 5672.52 Da, matching the theoretical value for single NOTA attachment (5672 Da; Fig. S2B). Importantly, BLI confirmed that NOTA‑BED retained nanomolar binding affinity (K\u003csub\u003eD\u003c/sub\u003e = 1.11 nM; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE; Table.1), indicating that conjugation preserved the functional epitope.\u003c/p\u003e \u003cp\u003e \u003cb\u003e[\u003c/b\u003e \u003csup\u003e \u003cb\u003e68\u003c/b\u003e \u003c/sup\u003e \u003cb\u003eGa]Ga‑NOTA‑BED enables sensitive and specific detection of CAR T cells\u003c/b\u003e \u003cb\u003ein vitro\u003c/b\u003e\u003c/p\u003e \u003cp\u003eRadiochemical quality control showed that purified [\u003csup\u003e68\u003c/sup\u003eGa]Ga‑NOTA‑BED achieved radiochemical purity (RCP)\u0026thinsp;\u0026gt;\u0026thinsp;98% and radiochemical yield (RCY)\u0026thinsp;\u0026gt;\u0026thinsp;50%, with a molar activity of ~\u0026thinsp;17.5 MBq/nmol, supporting sensitive in vitro assays and in vivo imaging. Time‑dependent uptake assays demonstrated rapid and specific binding of [\u003csup\u003e68\u003c/sup\u003eGa]Ga‑NOTA‑BED to BC19 CAR T cells (5.45\u0026thinsp;\u0026plusmn;\u0026thinsp;0.46% and 5.06\u0026thinsp;\u0026plusmn;\u0026thinsp;0.51% added dose per 5\u0026times;10\u003csup\u003e5\u003c/sup\u003e cells at 1 h and 2 h, respectively; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). A 1000‑fold molar excess of unlabeled probe reduced uptake to near background (0.24\u0026thinsp;\u0026plusmn;\u0026thinsp;0.08% and 0.17\u0026thinsp;\u0026plusmn;\u0026thinsp;0.09%), confirming specificity. Mock‑T controls showed negligible uptake (\u0026lt;\u0026thinsp;0.1%).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo exclude nonspecific interactions with shared CAR scaffold elements, uptake was compared across CAR phenotypes (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). Accumulation was strictly dependent on the presence of an anti‑BCMA scFv: BCMA CAR T cells showed the highest uptake (13.54\u0026thinsp;\u0026plusmn;\u0026thinsp;3.14% at 1 h), whereas BC19 CAR T cells showed lower but substantial uptake (6.53\u0026thinsp;\u0026plusmn;\u0026thinsp;1.54%), consistent with reduced BCMA scFv availability and/or steric effects in the tandem architecture. In contrast, CD19 CAR T cells and mock‑T cells remained at background levels. These results indicate that [\u003csup\u003e68\u003c/sup\u003eGa]Ga‑NOTA‑BED recognizes the anti‑BCMA antigen‑binding fragment with minimal cross‑reactivity to common signaling/scaffold components.\u003c/p\u003e \u003cp\u003eBecause antigen engagement in vivo may reduce availability of the CAR\u0026rsquo;s anti‑BCMA binding site, we modeled competitive conditions using tumor\u0026ndash;effector co‑cultures (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). In the presence of BCMA‑negative Raji cells, probe uptake by BC19 CAR T cells was minimally affected (82.63\u0026thinsp;\u0026plusmn;\u0026thinsp;2.61% of control even at E:T\u0026thinsp;=\u0026thinsp;5:1). In contrast, co‑culture with BCMA‑high U266 cells reduced uptake to 66.53\u0026thinsp;\u0026plusmn;\u0026thinsp;0.94% (E:T\u0026thinsp;=\u0026thinsp;1:1) and 57.01\u0026thinsp;\u0026plusmn;\u0026thinsp;2.21% (E:T\u0026thinsp;=\u0026thinsp;5:1), yet absolute uptake remained measurable (3.59\u0026thinsp;\u0026plusmn;\u0026thinsp;0.14% at E:T\u0026thinsp;=\u0026thinsp;5:1). These data suggest that high antigen burden may attenuate early imaging by occupying the CAR\u0026rsquo;s anti‑BCMA scFv, but [\u003csup\u003e68\u003c/sup\u003eGa]Ga‑NOTA‑BED retains sufficient sensitivity to track CAR T presence and dynamics under competitive conditions.\u003c/p\u003e \u003cp\u003e \u003cb\u003e[\u003c/b\u003e \u003csup\u003e \u003cb\u003e68\u003c/b\u003e \u003c/sup\u003e \u003cb\u003eGa]Ga‑NOTA‑BED enables quantitative and highly specific CAR T tracking\u003c/b\u003e \u003cb\u003ein vivo\u003c/b\u003e\u003c/p\u003e \u003cp\u003eTo validate \u003cem\u003ein vivo\u003c/em\u003e specificity, we established a bilateral subcutaneous cell‑cluster model in BALB/c mice by inoculating mock‑T and distinct CAR T subtypes into opposite flanks. PET imaging revealed selective probe accumulation at sites expressing an anti‑BCMA scFv (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). BCMA CAR T sites showed the highest uptake (1.05\u0026thinsp;\u0026plusmn;\u0026thinsp;0.18%ID/g at 1 h), followed by BC19 CAR T sites (0.90\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04%ID/g), consistent with in vitro binding. CD19 CAR T and mock‑T sites were indistinguishable from background (0.15\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04 and 0.11\u0026thinsp;\u0026plusmn;\u0026thinsp;0.03%ID/g; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE), demonstrating robust \u003cem\u003ein vivo\u003c/em\u003e discrimination of target versus non‑target populations.\u003c/p\u003e \u003cp\u003eWe next assessed sensitivity by PET/MR imaging of mice receiving serial dilutions of CAR T cells. PET signal decreased stepwise with decreasing cell number, yet discrete foci remained visible above background at 2 \u0026times; 10\u003csup\u003e4\u003c/sup\u003e cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF). At this dose, uptake (0.18\u0026thinsp;\u0026plusmn;\u0026thinsp;0.02%ID/g) exceeded muscle background (0.04\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01%ID/g; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG). Moreover, PET signal correlated linearly with injected CAR T cell number (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.856; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH), supporting quantitative use of [\u003csup\u003e68\u003c/sup\u003eGa]Ga‑NOTA‑BED as a surrogate readout of CAR T cell abundance \u003cem\u003ein vivo\u003c/em\u003e.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eBED probe binding does not impair BC19 CAR T functionality\u003c/h3\u003e\n\u003cp\u003eWe evaluated whether BED binding interferes with BC19 CAR T cytotoxicity using luciferase-based killing assays against Raji-Luc (CD19+) and U266-Luc (BCMA+) targets. BC19 CAR T cells showed strong E:T-dependent killing compared with mock-T controls (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA), achieving maximal lysis of 96.92\u0026thinsp;\u0026plusmn;\u0026thinsp;0.67% for Raji-Luc and 99.36\u0026thinsp;\u0026plusmn;\u0026thinsp;0.39% for U266-Luc (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB,C). In interference experiments performed at a fixed E:T ratio (2:1), addition of BED up to 100 nM\u0026mdash;well above the estimated peak plasma concentration required for imaging\u0026mdash;did not alter killing (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD), with lysis remaining 89.87\u0026thinsp;\u0026plusmn;\u0026thinsp;5.85% (Raji-Luc) and 98.89\u0026thinsp;\u0026plusmn;\u0026thinsp;1.08% (U266-Luc) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE,F; P\u0026thinsp;\u0026gt;\u0026thinsp;0.05).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo assess effects on CAR activation, cytokine secretion was quantified. TNF-α and IFN-γ levels in co-culture supernatants were unchanged in the presence of up to 100 nM BED (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG\u0026ndash;J; P\u0026thinsp;\u0026gt;\u0026thinsp;0.05), indicating that probe binding does not measurably disrupt antigen recognition, downstream signaling, or effector function. Together with the probe\u0026rsquo;s hydrophilicity and expected rapid renal clearance in vivo, these findings support a favorable biological safety profile for repeated imaging applications.\u003c/p\u003e\n\u003ch3\u003eLongitudinal PET/MR visualizes CAR T dynamics and therapeutic response heterogeneity in Raji lymphoma\u003c/h3\u003e\n\u003cp\u003eWe performed longitudinal PET/MR and IVIS monitoring using [\u003csup\u003e68\u003c/sup\u003eGa]Ga‑NOTA‑BED in a Raji tumor model to map spatiotemporal CAR T kinetics during therapy (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). Although BC19 CAR T\u0026ndash;treated mice ultimately achieved tumor control, early response dynamics were heterogeneous. Co‑registration of PET/MR‑derived CAR T distribution with IVIS‑derived tumor burden enabled classification into three response phenotypes (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB; Fig. S3).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFast responders (e.g., R2) showed rapid tumor clearance by day 8, and PET signals remained near baseline thereafter (0.19\u0026ndash;0.36%ID/g; Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC), consistent with rapid antigen loss and limited sustained expansion. Intermediate responders (e.g., R1) displayed a classical rise‑and‑fall immune kinetic: tumor burden increased to day 8 then declined, while PET signal increased and peaked around day 17 before contracting as tumor was eliminated (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD). Slow responders (e.g., R5) showed prolonged coexistence of tumor growth and CAR T accumulation, with progressive PET signal increase reaching a maximum at day 22 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE). PBS controls exhibited uncontrolled tumor growth and no specific tracer accumulation (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB, \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eF).\u003c/p\u003e \u003cp\u003eQuantitatively, excluding non‑proliferative fast responders, treated mice reached an average peak uptake of 1.29\u0026thinsp;\u0026plusmn;\u0026thinsp;0.48%ID/g (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eG) and a peak tumor‑to‑muscle ratio of 15.29\u0026thinsp;\u0026plusmn;\u0026thinsp;2.29 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eH), both significantly higher than PBS controls (P\u0026thinsp;\u0026lt;\u0026thinsp;0.01). Individual uptake curves further highlighted marked inter‑animal heterogeneity in both peak magnitude and timing (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eI). IVIS confirmed potent antitumor activity, with tumor flux decreasing to 3.43\u0026times;10\u003csup\u003e7\u003c/sup\u003e p/s by day 22 (approximately 1/220 of PBS peak 7.55\u0026times;10\u003csup\u003e9\u003c/sup\u003e p/s at day 18; Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eJ). Treated mice maintained stable body weight (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eK), showed no overt organ pathology (Fig. S4), and achieved 100% survival versus a median survival of 18 days in PBS controls (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eL). Terminal immunohistochemistry further supported specificity: residual microscopic lesions in treated mice remained densely infiltrated by CD3\u0026thinsp;+\u0026thinsp;T cells despite low CD19 signal (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eM), consistent with effective CAR T homing and sustained tumor‑site localization.\u003c/p\u003e\n\u003ch3\u003eLongitudinal PET/MR tracking under high antigen burden in U266 multiple myeloma\u003c/h3\u003e\n\u003cp\u003eTo evaluate probe performance under conditions of high antigen burden and potential CAR occupancy, we established a composite U266 myeloma model. In contrast to the uniform remission observed in the Raji setting, BC19 CAR T treatment produced heterogeneous outcomes that separated into Cure and Progress subgroups (Fig. S5).\u003c/p\u003e \u003cp\u003eTo better capture early and delayed kinetics in this model, we added an additional early scan (day 4) and shifted the day 8 timepoint to day 11 (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). PET/MR imaging mapped BC19 CAR T distribution longitudinally (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). In the Cure subgroup, tracer uptake rose promptly as tumor burden declined, peaking around day 14 (~\u0026thinsp;1.44%ID/g) and returning toward baseline following tumor eradication and antigen loss (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC). In the Progress subgroup, kinetics were delayed and variable: some animals showed a gradual PET rise without tumor control (Fig. S5), whereas others showed transient PET increases followed by declines despite continued tumor growth by IVIS (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD). This PET\u0026ndash;IVIS mismatch is consistent with the combined effects of high tumor load, CAR occupancy/competition, and/or functional impairment within the myeloma microenvironment.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eControls (mock‑T and PBS) displayed exponential tumor growth by IVIS without specific PET signal in tumor regions (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE,F). Across treated animals, peak uptake averaged 1.19\u0026thinsp;\u0026plusmn;\u0026thinsp;0.28%ID/g (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eG) with a peak tumor‑to‑muscle ratio of 11.33\u0026thinsp;\u0026plusmn;\u0026thinsp;2.94 (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eH), significantly exceeding controls (P\u0026thinsp;\u0026lt;\u0026thinsp;0.01), and individual curves delineated distinct kinetic patterns (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eI). Body weights remained stable in the BC19 group but declined in controls due to progressive disease (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eK). No major organ pathology was observed on H\u0026amp;E (Fig. S6), and the BC19 group maintained 100% survival over 23 days (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05; Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eL). Terminal immunohistochemistry confirmed tracking specificity: Progress tumors retained strong BCMA expression and showed prominent intratumoral CD3\u0026thinsp;+\u0026thinsp;infiltration, whereas control tumors lacked CD3 staining (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eM).\u003c/p\u003e \u003cp\u003e \u003cb\u003e[\u003c/b\u003e \u003csup\u003e \u003cb\u003e68\u003c/b\u003e \u003c/sup\u003e \u003cb\u003eGa]Ga‑NOTA‑BED exhibits favorable translational safety.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo assess the translational viability of the [\u003csup\u003e68\u003c/sup\u003eGa]Ga-NOTA-BCMA probe, we performed comprehensive human radiation dosimetry and chemical toxicity evaluations. The probe exhibited favorable in vivo safety and biocompatibility. Biodistribution studies and OLINDA/EXM estimates indicated predominant renal clearance, with the kidneys acting as the dose-limiting organ (8.71 \u0026times; 10\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e mSv/MBq). Notably, the total-body effective dose (0.0128 mSv/MBq) was lower than that of [\u003csup\u003e18\u003c/sup\u003eF]F-FDG (0.019 mSv/MBq), supporting its safe application in longitudinal repetitive imaging (Table. 2). Additionally, a single intravenous administration at a 10-fold higher dose induced no abnormal vital signs, and histological analysis of major organs showed no structural damage (Fig. S8), verifying no notable acute toxicity.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eWe developed an engineering‑free, antigen‑based PET/MR strategy for longitudinal tracking of a clinically used BCMA/CD19 tandem scFv CAR‑T product. A minimal BCMA ectodomain fragment radiolabelled with \u003csup\u003e68\u003c/sup\u003eGa enabled five PET/MR scans over 22 days, resolving heterogeneous CAR‑T kinetics in vivo while preserving cytotoxicity and cytokine release. The work extends antigen‑probe CAR‑PET\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e toward a repeatable monitoring approach compatible with ongoing BCMA/CD19 CAR‑T programs.\u003c/p\u003e \u003cp\u003eThis study establishes a compact probe with preserved binding and practical radiochemistry. NOTA‑BED retained nanomolar affinity for the anti‑BCMA scFv, and [\u003csup\u003e68\u003c/sup\u003eGa]Ga‑NOTA‑BED was produced with RCP\u0026thinsp;\u0026gt;\u0026thinsp;98%, RCY\u0026thinsp;\u0026gt;\u0026thinsp;50%, and molar activity\u0026thinsp;~\u0026thinsp;17.5 MBq/nmol. In vivo, the tracer enabled sensitive and quantitative detection: PET signal remained discernible above background at ~\u0026thinsp;2 \u0026times; 10\u003csup\u003e4\u003c/sup\u003e CAR‑T cells (in vivo LOD), and uptake scaled with cell number (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.856), supporting its use as a quantitative surrogate under the tested conditions. Critically, because antigen‑binding tracers could in principle interfere with CAR signaling, we directly tested functional compatibility: BED up to 100 nM did not impair BC19 CAR‑T cytotoxicity against both Raji‑Luc and U266‑Luc, and TNF‑α/IFN‑γ release remained unchanged, indicating that tracer binding is functionally tolerated at concentrations exceeding those anticipated for typical PET microdose exposures\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe main conceptual advance is the feasibility of repeatable longitudinal imaging without additional cell engineering, achieved by aligning probe size and radionuclide physics with serial studies. Antigen‑probe CAR‑PET was articulated as a strategy to image clinical CAR products without adding reporter genes\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. Here, probe miniaturization to 5.4 kDa\u0026mdash;smaller than clinically evaluated HER2-binding Affibody scaffolds (6.5-7 kDa)\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e\u0026mdash;facilitates rapid post-scan clearance, reduces carry-over background and permits repeated administration. Pairing this human-sequence antigen fragment with \u003csup\u003e68\u003c/sup\u003eGa (t\u003csub\u003e1/2\u003c/sub\u003e \u0026asymp; 67.7 min), as in established Affibody PET workflows\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e, further supports short-interval \u0026ldquo;inject-and-scan\u0026rdquo; imaging, albeit with an intrinsically brief imaging window per injection\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. Together, these choices enabled five scans within 22 days, a time span that captures the clinically relevant expansion\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e\u0026ndash;contraction phase of CAR‑T therapy and is not readily accessible with many existing PET cell‑tracking workflows.\u003c/p\u003e \u003cp\u003eMechanistically, our data are consistent with scFv‑dependent surface labeling rather than nonspecific T‑cell retention. Rapid, competitively blockable uptake in BCMA‑scFv\u0026ndash;positive CAR‑T populations, and background‑level uptake in CD19 CAR‑T and mock T cells, indicate that PET signal reflects the presence and accessibility of the BCMA‑binding arm. Nevertheless, competitive co‑culture experiments also foreground a key interpretive constraint of antigen‑based CAR‑PET: antigen occupancy under high tumour burden can reduce tracer binding by engaging the same scFv epitope used for imaging. This feature is not merely a technical caveat; it implies that PET signal is a composite of cell number, CAR surface density, and CAR accessibility (that is, the fraction of unoccupied binding sites), and may therefore diverge from tumour burden metrics in high‑antigen settings.\u003c/p\u003e \u003cp\u003eLongitudinal PET/MR revealed distinct kinetic phenotypes across tumour contexts, highlighting why repeated imaging is informative beyond single time points. In the CD19\u0026thinsp;+\u0026thinsp;Raji model, five scans delineated fast, intermediate, and slow responder trajectories: fast responders displayed rapid tumour clearance with near‑baseline PET thereafter; intermediate responders showed a rise‑and‑fall PET trajectory aligned with regression; and slow responders exhibited delayed, progressive PET accumulation peaking late. In the BCMA‑high U266 model, outcomes diverged into cure versus progress subgroups: cures exhibited early tumour‑region PET increases coincident with tumour decline, whereas progressors showed delayed or discordant PET trajectories (including transient rises followed by declines despite continued tumour growth). These patterns are consistent with at least two non‑exclusive mechanisms: competitive antigen occupancy reducing tracer access in high‑burden tumours, and/or functional impairment of CAR‑T cells in the myeloma microenvironment, emphasizing that PET‑derived CAR‑T \u0026ldquo;signal\u0026rdquo; is informative but not synonymous with efficacy.\u003c/p\u003e \u003cp\u003eThe translational relevance is sharpened by the maturation of clinically used BCMA/CD19 CAR‑T programs. A phase I/II study reported feasibility and activity of bispecific BCMA/CD19 CAR‑T therapy in relapsed/refractory multiple myeloma\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e, and a homologous BCMA/CD19 lineage has been explored in refractory myasthenia gravis with deep immune profiling\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. For these programmes, imaging methods that require additional genetic modifications can be difficult to retrofit. Reporter‑gene PET has clinical proof‑of‑concept for tracking engineered T cells\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e, but it adds genetic payload, manufacturing complexity, and regulatory considerations. Direct ex vivo radiolabelling can map early biodistribution, but suffers from label dilution with proliferation and signal confounding from dead‑cell label redistribution, limiting interpretability across the weeks‑long kinetic arc\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. In contrast, antigen‑based BED imaging retains the practical advantage of not altering the therapeutic cells while enabling serial snapshots across time, aligning with the \u0026ldquo;engineering‑free\u0026rdquo; premise emphasized in antigen‑probe CAR‑PET\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e and extending it to a high‑frequency, multi‑scan workflow.\u003c/p\u003e \u003cp\u003eSeveral limitations should be addressed explicitly before human translation. First, small‑animal xenografts incompletely recapitulate human tumour architecture, antigen heterogeneity, vascular permeability, and immune context, and do not reliably predict immune responses to repeat dosing. That said, BED is derived from a human BCMA extracellular domain and is essentially congruent with soluble BCMA apart from the chelator conjugation, so the intrinsic risk of immunogenicity is expected to be comparatively low; nevertheless, the NOTA (and any linker) constitutes a non‑native chemical moiety that could act as a hapten and warrants empiric evaluation of anti‑drug antibodies and their impact on clearance, background, and safety under repeat administration. Second, renal clearance is expected to dominate for a\u0026thinsp;~\u0026thinsp;5.4‑kDa protein, which may concentrate activity in kidney/bladder and make kidney‑limited dosimetry a key constraint under repeat imaging; formal dosimetry and mitigation strategies (hydration/voiding protocols, renal dose modelling) are therefore required. Third, \u003csup\u003e68\u003c/sup\u003eGa\u0026rsquo;s short half‑life (\u0026asymp;\u0026thinsp;67.7 min) limits imaging windows per injection and may reduce sensitivity if human pharmacokinetics are slower than in mice; longer‑lived isotopes (e.g. \u003csup\u003e18\u003c/sup\u003eF, t\u003csub\u003e1/2\u003c/sub\u003e\u0026asymp;109.7 min) may be needed depending on clinical workflows. Fourth, quantitative comparability across models and scanners is nontrivial: %ID/g (mouse) and SUV (human) differ in scaling; moreover, partial‑volume effects and ROI definition can bias lesion quantification, particularly for small or regressing tumours. Fifth, although RCY\u0026thinsp;\u0026gt;\u0026thinsp;50% and RCP\u0026thinsp;\u0026gt;\u0026thinsp;98% are strong starting points, chelator choice (NOTA versus alternatives) and radiochemistry conditions may be further optimized to improve kit‑readiness and increase molar activity, thereby minimizing the risk of receptor occupancy by cold mass. Finally, response subgrouping (fast/intermediate/slow; cure/progress) emerges from limited cohorts; larger studies are needed to improve statistical power, define predictive thresholds (for example, time‑to‑peak or peak amplitude), and test reproducibility. Where injected mass/activity, scan start time, and cohort sizes were not specified, we assume typical PET microdose conditions and ~\u0026thinsp;1 h post‑injection imaging for \u003csup\u003e68\u003c/sup\u003eGa\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e; these parameters should be explicitly optimized and reported in translational studies.\u003c/p\u003e \u003cp\u003eFuture work can follow a pragmatic, translation‑forward path. Because diagnostic PET radiopharmaceuticals are administered at microdose‑level masses and are used at the low end of the dose\u0026ndash;response curve, dose‑related pharmacologic or toxic effects from the nonradioactive mass are generally unlikely, and regulatory guidances explicitly encourage tailored, risk‑based nonclinical packages to facilitate timely early‑phase studies\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. In this context, an early first‑in‑human imaging sub‑study embedded within ongoing BCMA/CD19 CAR‑T programmes\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e is feasible to establish human pharmacokinetics, lesion targeting, and whole‑body/kidney dosimetry, while repeat‑dose considerations (including immunogenicity and renal dose) are addressed in parallel through focused nonclinical work. In parallel, radiochemistry and formulation should be optimized toward higher molar activity and kit‑readiness\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e, and the generalizability of BED imaging should be tested across clinically relevant CAR architectures (tandem versus bicistronic), ideally integrated with longitudinal blood‑based measures\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e to link imaging kinetics to expansion, persistence, and outcome.\u003c/p\u003e \u003cp\u003eIn summary, we developed an engineering-free, antigen-based PET/MR strategy to longitudinally track a clinically used tandem scFv BCMA/CD19 CAR-T platform using a \u003csup\u003e68\u003c/sup\u003eGa-labeled BCMA ectodomain probe. The compact, human-derived BED tracer retains\u0026thinsp;~\u0026thinsp;1 nM binding, clears rapidly after each scan, and supports repeat imaging with a short-lived radionuclide and minimal procedural burden. [\u003csup\u003e68\u003c/sup\u003eGa]Ga-NOTA-BED achieved high specificity and sensitivity (\u0026asymp;\u0026thinsp;2\u0026times;10\u003csup\u003e4\u003c/sup\u003e cells) without measurably impairing CAR T function. Five PET/MR scans over 22 days resolved heterogeneous CAR T trafficking and expansion kinetics that aligned with response in Raji lymphoma and revealed distinct dynamics under high antigen burden in U266 myeloma. These data support clinical translation of an on-demand imaging companion for patient-specific interpretation of CAR T expansion and lesion engagement; notably, a prospective study (ClinicalTrials.gov Identifier: NCT07280793, \u0026ldquo;CAR-T Cell Efficacy With Molecular Imaging in Multiple Myeloma\u0026rdquo;) has been registered based on this study and will be posted publicly on ClinicalTrials.gov.\u003c/p\u003e "},{"header":"Methods","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003eCell culture and cell lines\u003c/h2\u003e \u003cp\u003eHuman Burkitt\u0026rsquo;s lymphoma Raji cells and human multiple myeloma U266 cells were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). To generate reporter cell lines stably expressing green fluorescent protein (GFP) and firefly luciferase (Luc), Raji and U266 cells were transduced with a lentiviral vector (pASLenti-pA-Luc2-CMV-EF1-EGFP-P2A-Puro-WPRE; OBiO Technology, China). Transduced cells were selected with puromycin (2 \u0026micro;g/ml; Beyotime, Catalog No. ST551) for 14 days, resulting in stable polyclonal populations co-expressing GFP and luciferase (hereafter referred to as Raji-Luc-GFP and U266-Luc-GFP). All tumor cell lines were maintained in RPMI 1640 medium (Adamas-life, Catalog No. C8016) supplemented with 10% fetal bovine serum (FBS; Adamas-life, Catalog No. C8010) and 1% penicillin\u0026ndash;streptomycin (Beyotime, Catalog No. C0222).\u003c/p\u003e \u003cp\u003ePrimary human T cells were isolated from donor peripheral blood mononuclear cells (PBMCs) to produce untransduced control T cells (Mock-T) and lentivirally transduced CD19 CAR T, BCMA CAR T, and bispecific BC19 CAR T cells. All PBMCs were donated by healthy donors under protocols approved by the Medical Ethics Committee of the Affiliated Hospital of Xuzhou Medical University (ethics approval no. XYFY2020-KL062-01). For \u003cem\u003ein vitro\u003c/em\u003e expansion and maintenance, T cells were cultured in X-VIVO\u0026trade; 15 medium (Lonza, Catalog No. 04-418Q) supplemented with 10% FBS and freshly prepared recombinant human cytokines, including IL-2 (300 U/ml; Beyotime, Catalog No. P5115), IL-7 (5 ng/ml; Yeasen, Catalog No. 90188ES10), and IL-15 (5 ng/ml; Yeasen, Catalog No. 90113ES10). All cells were cultured in a humidified incubator at 37\u0026deg;C with 5% CO\u003csub\u003e2\u003c/sub\u003e.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e\n\u003ch3\u003eFlow cytometric analysis of CAR expression\u003c/h3\u003e\n\u003cp\u003eSurface CAR expression and transduction efficiency were evaluated by multicolor flow cytometry. CD19 CAR T, BCMA CAR T, bispecific BC19 CAR T, and untransduced Mock-T cells in logarithmic growth phase were harvested. Approximately 5 \u0026times; 10\u003csup\u003e5\u003c/sup\u003e cells per condition were washed with ice-cold FACS buffer (PBS\u0026thinsp;+\u0026thinsp;1% BSA; Beyotime, Catalog No. ST023) and blocked with human Fc receptor reagent (5 \u0026micro;L; Beyotime, Catalog No. C1752S) for 10 min at 4\u0026deg;C. Cells were stained with APC-conjugated anti-FMC63 antibody (1:50; ACROBiosystems, Catalog No. FM3-AY54A1) and FITC-labeled human BCMA protein (3 \u0026micro;g/mL; ACROBiosystems, Catalog No. BCA-HF254) for 60 min at 4\u0026deg;C in the dark; single-stained controls were included for compensation. After washing, 7-AAD (5 \u0026micro;L; Proteintech, Catalog No. PD00101) was added 5 min before acquisition. Data were collected on a BD FACSCelesta cytometer (BD Biosciences, USA) and analyzed with FlowJo v10.8 (BD Biosciences). At least 10,000 live cells per sample were recorded, and CAR expression was quantified as the percentage of APC\u0026thinsp;+\u0026thinsp;and/or FITC+ cells among live cells, with Mock-T cells used as gating controls.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003ePreparation and quality control of BED\u003c/h2\u003e \u003cp\u003eThe core extracellular domain of BCMA (UniProtKB Q02223, residues A5 to A54) was selected as the scaffold for probe construction and is hereafter referred to as BED. A 6\u0026times;His-SUMO dual tag was introduced at the N-terminus, and the codon-optimized gene was cloned into the pET-28a(+) vector (XbaI/XhoI sites), yielding the recombinant plasmid pET28a-His-SUMO-BCMA(5\u0026ndash;54). The plasmid was transformed into \u003cem\u003eEscherichia coli\u003c/em\u003e BL21(DE3), and protein expression was induced with 0.1 mM isopropyl-β-D-thiogalactopyranoside (IPTG) at 16\u0026deg;C for 16 hours. Cells were harvested and lysed, and the His-SUMO-BCMA fusion protein was purified by Ni-NTA affinity chromatography. The SUMO tag was subsequently removed by Ulp1 protease digestion, followed by a second Ni-NTA purification step to remove the cleaved tag; the flow-through fraction containing BED was collected. The protein was further purified by size-exclusion chromatography (Superdex 75 pg column, \u0026Auml;KTA pure system; GE Healthcare), concentrated by ultrafiltration, buffer-exchanged, and lyophilized to obtain BED powder. Protein purity, aggregation state, and molecular mass were systematically assessed by SDS-PAGE, SEC-HPLC, and time-of-flight mass spectrometry (TOF-MS), respectively.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eConjugation and purification of NOTA-BED\u003c/h2\u003e \u003cp\u003eLyophilized BED was dissolved in anhydrous dimethyl sulfoxide (DMSO), and NOTA-NHS ester solution (50 mg/ml) was added dropwise at a molar ratio of 1:8 (protein:chelator). N,N-diisopropylethylamine (DIPEA) was added to adjust the pH to 9.0, and the reaction was carried out at 40\u0026deg;C for 8 hours with continuous stirring. The reaction mixture was purified by reversed-phase high-performance liquid chromatography (RP-HPLC) using a preparative C18 column (PrepPurite Gold, 5 \u0026micro;m, 10 \u0026times; 250 mm; Wepure) at a flow rate of 5.0 ml/min with a linear gradient of solvent A (0.1% TFA in water) and solvent B (0.1% TFA in acetonitrile). Elution was monitored at 280 nm. The target fraction was collected and analyzed by TOF-MS to confirm successful NOTA conjugation and determine the exact molecular mass.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eBLI-based affinity measurement\u003c/h2\u003e \u003cp\u003eBinding kinetics between BED probes and CAR molecules were evaluated using bio-layer interferometry (BLI) on an Octet R2 system (Sartorius). His-tagged CAR proteins (30 \u0026micro;g/ml) were immobilized onto Ni-NTA biosensors and equilibrated in PBST buffer (PBS containing 0.05% Tween-20). Serial dilutions of BED and NOTA-BED (0.78 to 50 nM) were prepared and applied for association (300 s) and dissociation (300 s) measurements. Data were analyzed using Octet Analysis Studio software (Sartorius) with a 1:1 global fitting model to derive the association rate constant (\u003cem\u003ek\u003c/em\u003e\u003csub\u003eon\u003c/sub\u003e), dissociation rate constant (\u003cem\u003ek\u003c/em\u003e\u003csub\u003eoff\u003c/sub\u003e), and equilibrium dissociation constant (K\u003csub\u003eD\u003c/sub\u003e = \u003cem\u003ek\u003c/em\u003e\u003csub\u003eoff\u003c/sub\u003e / \u003cem\u003ek\u003c/em\u003e\u003csub\u003eon\u003c/sub\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eRadiolabeling with [\u003csup\u003e68\u003c/sup\u003eGa]Ga and quality control\u003c/h2\u003e \u003cp\u003eFreshly eluted [\u003csup\u003e68\u003c/sup\u003eGa]GaCl\u003csub\u003e3\u003c/sub\u003e solution (500 \u0026micro;L in 0.05 M HCl) was buffered with 1.25 M sodium acetate to adjust the pH to 4.0 to 4.5. NOTA-BED (50 \u0026micro;g) was added and incubated at 60\u0026deg;C for 15 min. The reaction mixture was purified using a C18 solid-phase extraction cartridge (Waters). After sequential conditioning with ethanol and water, the sample was loaded, washed with water to remove free [\u003csup\u003e68\u003c/sup\u003eGa]Ga, and eluted with 300 \u0026micro;L of ethanol containing 10 mM HCl to obtain [\u003csup\u003e68\u003c/sup\u003eGa]Ga-NOTA-BED. Radiochemical purity (RCP) was determined by radio-HPLC under the same chromatographic conditions as the non-radioactive probe and calculated as the percentage of radioactivity corresponding to the main peak. Radiochemical yield (RCY) and molar activity were calculated based on starting activity and BED amount.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eTime-dependent cellular uptake and blocking assays\u003c/h2\u003e \u003cp\u003eTo evaluate the binding specificity and kinetic properties of the probe, BC19 CAR T cells and Mock-T cells in the logarithmic growth phase were harvested, washed with PBS, and resuspended in serum-free X-VIVO\u0026trade; 15 medium at a density of 5 \u0026times; 10\u003csup\u003e5\u003c/sup\u003e cells per tube. Approximately 37 kBq (1 \u0026micro;Ci) of [\u003csup\u003e68\u003c/sup\u003eGa]Ga-NOTA-BED was added to each tube, followed by incubation at 37\u0026deg;C for 60 or 120 min.For competitive blocking, a 1000-fold molar excess of unlabeled NOTA-BED (50 \u0026micro;g) was added 1 hour prior to radiotracer incubation. After incubation, the cells were centrifuged and washed three times with ice-cold PBS containing 1% BSA to remove unbound and nonspecifically associated tracer. Cell pellets were resuspended in 500 \u0026micro;L of PBS, and radioactivity was measured using a gamma counter (Perkin Elmer, American). Cellular uptake was expressed as the percentage of the added dose (% AD) per 5 \u0026times; 10\u003csup\u003e5\u003c/sup\u003e cells. All experiments were performed in triplicate.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eUptake specificity among different CAR T subtypes\u003c/h2\u003e \u003cp\u003eTo assess the selectivity of the probe for the BCMA CAR scFv, Mock-T cells (negative control), CD19 CAR T cells (nontarget control), BCMA CAR T cells (single-target positive control), and BC19 CAR T cells (dual-target experimental group) were incubated with [\u003csup\u003e68\u003c/sup\u003eGa]Ga-NOTA-BED (5 \u0026times; 10\u003csup\u003e5\u003c/sup\u003e cells per tube) under the same conditions for 60 min. Cellular uptake was calculated as % AD/5 \u0026times; 10\u003csup\u003e5\u003c/sup\u003e cells as described above, and values among the different groups were compared to determine probe specificity toward BCMA CAR-expressing T cells.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eTumor cell competition assay\u003c/h2\u003e \u003cp\u003eTo simulate the competitive inhibition of probe uptake by tumor-associated BCMA antigens in the tumor microenvironment, a co-incubation model was established using CAR T cells and tumor target cells. BC19 CAR T cells (effector cells, E) were mixed with Raji (CD19⁺) or U266 (BCMA⁺) target cells (T) at effector-to-target (E:T) ratios of 1:1 and 1:5. BC19 CAR T cells incubated alone served as the baseline positive control (defined as 100% uptake), and those pre-blocked with excess unlabeled NOTA-BED served as the specificity control. Cell mixtures were pre-incubated at 37\u0026deg;C for 1 hour, followed by the addition of 37 kBq (1 \u0026micro;Ci) of [\u003csup\u003e68\u003c/sup\u003eGa]Ga-NOTA-BED and further incubation for 1 hour. After incubation, the cells were centrifuged and washed with ice-cold PBS, and the radioactivity of the cell pellets was measured. Relative uptake was calculated as counts per minute (CPM) normalized to the baseline positive control group and expressed as a percentage.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eTargeting specificity assessment\u003c/h2\u003e \u003cp\u003eTo evaluate the in vivo targeting selectivity of the BED probe, healthy BALB/c mice received subcutaneous injections of Mock-T cells or CAR T cells with different antigen specificities (3 \u0026times; 10\u003csup\u003e6\u003c/sup\u003e cells in 100 \u0026micro;L sterile saline) into the bilateral axillae. Immediately after cell implantation, 3.7\u0026ndash;5.5 MBq (100\u0026ndash;150 \u0026micro;Ci) of [\u003csup\u003e68\u003c/sup\u003eGa]Ga-NOTA-BED was administered via the tail vein. Static micro-positron emission tomography (micro-PET) imaging was performed 1 hour post-injection.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eDetection sensitivity and limit of detection\u003c/h2\u003e \u003cp\u003eTo determine the minimum detectable CAR T cell number, \u003cem\u003ein vivo\u003c/em\u003e, a cell quantity gradient was established. BC19 CAR T cells (5 \u0026times; 10\u003csup\u003e3\u003c/sup\u003e to 5 \u0026times; 10\u003csup\u003e6\u003c/sup\u003e cells in 100 \u0026micro;L sterile saline) were injected subcutaneously into BALB/c mice (n\u0026thinsp;=\u0026thinsp;3), simulating target tissues of varying cell quantities. One hour post-probe injection, mice were imaged using a Bruker 9.4T small-animal PET/magnetic resonance (PET/MR) system (Bruker BioSpec 94/30, Germany) under 1.0\u0026ndash;1.5% isoflurane anesthesia. The lower limit of detection (LOD) was defined as the lowest cell quantity yielding a discernible signal distinguishable from the background.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003ePET/MR images reconstruction\u003c/h2\u003e \u003cp\u003ePET images were reconstructed using a three-dimensional ordered-subset expectation maximization (3D OSEM) algorithm with attenuation and scatter correction, with anatomical reference provided by a T2-weighted TurboRARE sequence (TR/TE\u0026thinsp;=\u0026thinsp;2400/33 ms; resolution\u0026thinsp;=\u0026thinsp;0.15 \u0026times; 0.15 \u0026times; 1.0 mm\u003csup\u003e3\u003c/sup\u003e). Image analysis was performed using PMOD software (Version 4.4; Switzerland). Regions of interest (ROIs) were manually delineated on fused PET/MR images to quantify radioactivity concentrations (%ID/g).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eBioluminescence-based cytotoxicity assay\u003c/h2\u003e \u003cp\u003eTo systematically evaluate whether the NOTA-BED probe affects the cytotoxic function of CAR T cells, a co-culture system of effector and target cells was established. Raji-Luc-GFP (CD19+) and U266-Luc-GFP (BCMA+) cells were used as target cells, while BC19 CAR T cells and Mock-T cells served as effectors. Initially, various effector-to-target (E:T) ratios (4:1, 2:1, 1:1, and 1:2; n\u0026thinsp;=\u0026thinsp;3) were tested to determine the optimal killing conditions. Subsequently, an interference assay was performed at a fixed E:T ratio of 2:1 by introducing graded concentrations of NOTA-BED (0, 0.1, 1, 10, and 100 nM; n\u0026thinsp;=\u0026thinsp;4). All assays were conducted in black 96-well plates, in which 1 \u0026times; 10\u003csup\u003e4\u003c/sup\u003e target cells per well were co-incubated with effector cells and probes in a final volume of 200 \u0026micro;L for 48 hours at 37\u0026deg;C. For bioluminescence detection, a solution containing 0.3 mg/ml D-luciferin potassium salt (Ambeed, Catalog No.A162108) and 2% Triton X-100 was added to each well. After incubation for 5 min in the dark, bioluminescence images were acquired using an IVIS imaging system (PerkinElmer, USA). Total photon flux (photons/s) was quantified using Living Image software (PerkinElmer).\u003c/p\u003e \u003cp\u003eSpecific lysis was calculated as:\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$\\:\\text{S}\\text{p}\\text{e}\\text{c}\\text{i}\\text{f}\\text{i}\\text{c}\\:\\text{l}\\text{y}\\text{s}\\text{i}\\text{s}\\:\\left(\\text{\\%}\\right)=\\:\\left[1\\:-\\frac{\\left(\\text{S}\\text{a}\\text{m}\\text{p}\\text{l}\\text{e}\\:\\text{f}\\text{l}\\text{u}\\text{x}\\:-\\:\\text{B}\\text{a}\\text{c}\\text{k}\\text{g}\\text{r}\\text{o}\\text{u}\\text{n}\\text{d}\\:\\text{f}\\text{l}\\text{u}\\text{x}\\right)}{\\left(\\text{M}\\text{e}\\text{a}\\text{n}\\:\\text{t}\\text{a}\\text{r}\\text{g}\\text{e}\\text{t}\\:\\text{c}\\text{o}\\text{n}\\text{t}\\text{r}\\text{o}\\text{l}\\:\\text{f}\\text{l}\\text{u}\\text{x}\\:-\\:\\text{B}\\text{a}\\text{c}\\text{k}\\text{g}\\text{r}\\text{o}\\text{u}\\text{n}\\text{d}\\:\\text{f}\\text{l}\\text{u}\\text{x}\\right)}\\right]\\text{*}\\:100\\text{\\%}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eCAR-specific lysis was calculated by normalizing CAR T sample flux to Mock-T control flux:\u003cdiv id=\"Equ2\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ2\" name=\"EquationSource\"\u003e\n$$\\:\\text{C}\\text{A}\\text{R}-\\text{s}\\text{p}\\text{e}\\text{c}\\text{i}\\text{f}\\text{i}\\text{c}\\:\\text{l}\\text{y}\\text{s}\\text{i}\\text{s}\\:\\left(\\text{\\%}\\right)=\\:\\left[1\\:-\\frac{\\left(\\text{C}\\text{A}\\text{R}\\:\\text{T}\\:\\text{s}\\text{a}\\text{m}\\text{p}\\text{l}\\text{e}\\:\\text{f}\\text{l}\\text{u}\\text{x}\\right)}{\\left(\\text{M}\\text{e}\\text{a}\\text{n}\\:\\text{M}\\text{o}\\text{c}\\text{k}-\\text{T}\\:\\text{c}\\text{o}\\text{n}\\text{t}\\text{r}\\text{o}\\text{l}\\:\\text{f}\\text{l}\\text{u}\\text{x}\\right)}\\right]\\text{*}\\:100\\text{\\%}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e2\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eCytokine release analysis\u003c/h2\u003e \u003cp\u003eTo further assess the immune activation and effector function of CAR T cells under probe intervention, parallel co-culture assays corresponding to the interference experiment (fixed E:T ratio of 2:1) were performed under identical conditions. After 48 hours of co-incubation, cell culture supernatants were collected and centrifuged at 350 g for 5 min to remove residual cells and debris.\u003c/p\u003e \u003cp\u003eSupernatants were diluted 1:10 with the provided sample diluent, and cytokine levels were measured using human interferon-γ (IFN-γ) and tumor necrosis factor-α (TNF-α) enzyme-linked immunosorbent assay (ELISA) kits (Elabscience; Catalog Nos. Catalog No. E-EL-H0108 for IFN-γ and Catalog No. E-EL-H0109 for TNF-α) according to the manufacturer\u0026rsquo;s instructions. Absorbance was measured at 450 nm using a microplate reader (BioTeK, American). Absolute cytokine concentrations were determined by fitting sample absorbance values to standard curves generated using a four-parameter logistic regression model.\u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003eEstablishment of tumor-bearing models and therapeutic protocols\u003c/h2\u003e \u003cp\u003e All animal experiments were conducted in strict accordance with the national regulations for animal welfare and were approved by the Institutional Animal Care and Use Committee (IACUC) of the Jiangsu Institute of Nuclear Medicine (Approval No. JSINM-2025-097).\u003c/p\u003e \u003cp\u003eTo simulate localized tumor burden and systemic dissemination, composite subcutaneous-systemic models were established in 4\u0026ndash;6-week-old female NCG mice. For the Raji lymphoma model, mice received a subcutaneous injection of 1 \u0026times; 10\u003csup\u003e6\u003c/sup\u003e Raji-Luc cells on Day\u0026thinsp;\u0026minus;\u0026thinsp;10, followed by a tail vein injection of 2 \u0026times; 10\u003csup\u003e5\u003c/sup\u003e Raji-Luc cells on Day\u0026thinsp;\u0026minus;\u0026thinsp;2. On Day 0, mice were screened via bioluminescence imaging; those with a total flux of approximately 1 \u0026times; 10\u003csup\u003e7\u003c/sup\u003ep/s in the subcutaneous tumor were randomly assigned to the CAR-T group (5 \u0026times; 10\u003csup\u003e6\u003c/sup\u003e BC19 CAR-Tcells, i.v), or the PBS group (equivalent volume PBS, i.v). The U266 myeloma model was established similarly, with subcutaneous injection (3.5 \u0026times; 10\u003csup\u003e6\u003c/sup\u003e U266-Luc cells) on Day\u0026thinsp;\u0026minus;\u0026thinsp;7 and intravenous injection (8 \u0026times; 10\u003csup\u003e5\u003c/sup\u003e U266 -Luc cells) on Day\u0026thinsp;\u0026minus;\u0026thinsp;2. On Day 0, mice with a subcutaneous flux of approximately 2 \u0026times; 10\u003csup\u003e8\u003c/sup\u003ep/s were randomized into CAR-T (5 \u0026times; 10\u003csup\u003e6\u003c/sup\u003e BC19 CAR-T cells, i.v), Mock-T (5 \u0026times; 10\u003csup\u003e6\u003c/sup\u003e Mock-T cells, i.v), or PBS groups (equivalent volume PBS, i.v).\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003eMultimodal longitudinal imaging and therapeutic evaluation\u003c/h2\u003e \u003cp\u003eTumor burden and CAR T cell trafficking were longitudinally monitored using synchronized bioluminescence imaging and PET/MR imaging. Bioluminescence imaging was performed at baseline (Day 0) and all subsequent follow-up points to assess tumor progression. Following CAR T cell infusion, serial PET/MR imaging was conducted in tandem with Bioluminescence imaging on Days 1, 8, 12, 17, and 22 for the Raji model, and on Days 4, 11, 14, 18, and 21 for the U266 model. For PET/MR imaging, mice received approximately 3.7 MBq of the radiotracer via tail vein injection. After a 1 hour uptake period, 10-min static PET/MRI acquisitions were performed using a Bruker 9.4T small-animal PET/MR system (Bruker BioSpec 94/30, Germany). Image analysis was conducted using PMOD software (version 4.4, Switzerland), with radioactivity uptake quantified as the percentage of injected dose per gram of tissue (%ID/g) based on volumes of interest (VOIs) delineated on fused PET/MR images. Bioluminescence imaging was performed 5\u0026ndash;10 min after intraperitoneal injection of D-luciferin (30 mg/mL, 100 \u0026micro;L), and total photon flux (photons/s) was quantified using Living Image software (PerkinElmer, American).\u003c/p\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003eBody weight monitoring and survival analysis\u003c/h2\u003e \u003cp\u003eTo assess the therapeutic efficacy of CAR T cell treatment and evaluate potential systemic toxicity associated with repeated PET/MR imaging, body weight and clinical status were recorded every 3\u0026ndash;4 days following CAR T cell infusion. The humane endpoints were defined as a body weight loss exceeding 20% of the baseline or manifestation of severe cachexia. Animals that did not reach humane endpoint criteria were euthanized on Day 23, which was designated as the experimental endpoint. Survival time was recorded for each animal and utilized for subsequent survival analysis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section3\"\u003e \u003ch2\u003eHistopathological and immunohistochemical analysis\u003c/h2\u003e \u003cp\u003eAt the experimental endpoint, mice were euthanized for tissue harvesting. Major organs, including the heart, liver, spleen, lungs, and kidneys, as well as tumor tissues, were collected, fixed in 4% paraformaldehyde for 24 hours, and paraffin-embedded. Serial sections (4 \u0026micro;m) were prepared for histological and immunohistochemical (IHC) analyses. For histopathological evaluation, sections were stained with hematoxylin and eosin (H\u0026amp;E) to assess tissue morphology and potential treatment-related toxicity. For IHC analysis, tumor sections were incubated with anti-human CD3 (1:500, clone CAL54; Abcam, cat. no. ab237707) to identify infiltrating CAR T cells, and with either anti-CD19 (5 \u0026micro;g/mL; R\u0026amp;D Systems, cat. no. MAB48671) or anti-BCMA (5 \u0026micro;g/mL; R\u0026amp;D Systems, cat. no. MAB10762) antibodies to visualize tumor-associated antigens. Slides were digitally scanned, and representative images were acquired. CAR T cell distribution and infiltration patterns across different treatment groups were qualitatively assessed using ImageJ software.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cp\u003eAll quantitative data are expressed as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation (SD). Statistical analyses and graphical representations were performed using GraphPad Prism (version 8.0). For comparisons between two independent groups, unpaired two-tailed Student\u0026rsquo;s t-tests were applied. For comparisons involving more than two groups, one-way analysis of variance (ANOVA) was used; Welch\u0026rsquo;s ANOVA was applied when variance homogeneity was not satisfied, followed by Tukey\u0026rsquo;s multiple comparisons post hoc test. Longitudinal datasets, including tumor volume and body weight over time, were analyzed using two-way repeated-measures ANOVA with Tukey\u0026rsquo;s post hoc correction.\u003c/p\u003e \u003cp\u003eFor survival analysis, an event was defined as death, body weight loss exceeding 20% of baseline, or tumor volume reaching 1500 mm\u003csup\u003e3\u003c/sup\u003e. Survival curves were generated using the Kaplan\u0026ndash;Meier method and compared using the log-rank (Mantel\u0026ndash;Cox) test. All statistical tests were two-tailed, and a \u003cem\u003eP\u003c/em\u003e value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant. Levels of statistical significance are indicated in the figures as follows: ns (not significant, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05), *P\u0026thinsp;\u0026lt;\u0026thinsp;0.05, **\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01, ***\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, and ****\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.0001.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor contributions\u003c/h2\u003e \u003cp\u003eW.Z., X.W., X.Z. and M.Y. conceived the study and designed the experiments. W.Z. and X.W. conducted the experiments and collected and analysed the data. R.H., J.Y., D.P., C.C., Y.X., L.W. and M.S. contributed experimental or analysis tools. W.Z., X.W., X.Z. and M.Y. wrote the manuscript. All authors carefully reviewed and approved the manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eThis work was supported by the National Natural Science Foundation of China (grant No. 32371434, 12575361), and Wuxi Taihu Light Science and Technology Project, grant No. K20253011, K20251002.\u003c/p\u003e\u003ch2\u003eData availability\u003c/h2\u003e \u003cp\u003eThe authors declare that the main data supporting the findings of this study are available within the Article and its Supplementary Information. The corresponding author will make raw data and step-by-step protocols available upon request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eJune CH, O\u0026rsquo;Connor RS, Kawalekar OU, Ghassemi S (2018) Milone MC. CAR T cell immunotherapy for human cancer. Science 359:1361\u0026ndash;1365\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMunshi NC et al (2021) Idecabtagene vicleucel in relapsed and refractory multiple myeloma. N Engl J Med 384:705\u0026ndash;716\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWestin JR et al (2023) Survival with axicabtagene ciloleucel in large B-cell lymphoma. N Engl J Med 389:148\u0026ndash;157\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSamur MK et al (2021) Biallelic loss of BCMA as a resistance mechanism to CAR T cell therapy in a patient with multiple myeloma. Nat Commun 12:868\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSimon S, Riddell SR (2020) Dual targeting with CAR T cells to limit antigen escape in multiple myeloma. Blood Cancer Discov 1:130\u0026ndash;133\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHirabayashi K et al (2021) Dual-targeting CAR-T cells with optimal co-stimulation and metabolic fitness enhance antitumor activity and prevent escape in solid tumors. Nat Cancer 2:904\u0026ndash;918\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShi M et al (2024) Bispecific CAR T cell therapy targeting BCMA and CD19 in relapsed/refractory multiple myeloma: a phase I/II trial. Nat Commun 15:3371\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang X et al (2026) BCMA/CD19 CAR T cell therapy for refractory myasthenia gravis: Proteomic signatures and single-cell transcriptomics of disease flares. Sci Adv 12:eaeb6424\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVolpe A, Pham J, Sellmyer MA, Ponomarev V (2025) Practical considerations for clinical translation of PET imaging of adoptive cell therapies. npj Imaging 3:59\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShishido SN et al (2024) Liquid biopsy approach to monitor the efficacy and response to CAR-T cell therapy. J Immunother Cancer 12:e007329\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVolpe A et al (2020) Spatiotemporal PET imaging reveals differences in CAR-T tumor retention in triple-negative breast cancer models. Mol Ther 28:2271\u0026ndash;2285\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrebs S, Ponomarev V, Slovin S, Sch\u0026ouml;der H (2019) Imaging of CAR T-cells in cancer patients: paving the way to treatment monitoring and outcome prediction. J Nucl Med 60:879\u0026ndash;881\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeist MR et al (2018) PET of adoptively transferred chimeric antigen receptor T cells with 89Zr-oxine. J Nucl Med 59:1531\u0026ndash;1537\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim S-Y et al (2024) Direct and indirect chimeric antigen receptor T-Cell imaging with PET/MRI in a tumor xenograft model. Radiology 310:e231406\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang X-y et al (2021) Feasibility study of 68Ga-labeled CAR T cells for in vivo tracking using micro-positron emission tomography imaging. Acta Pharmacol Sin 42:824\u0026ndash;831\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKurebayashi Y, Choyke PL, Sato N (2021) Imaging of cell-based therapy using 89Zr-oxine ex vivo cell labeling for positron emission tomography. Nanotheranostics 5:27\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeland P, Kumar D, Nimmagadda S, Bauer SR, Puri RK, Joshi BH (2023) Characterization of chimeric antigen receptor modified T cells expressing scFv-IL-13Rα2 after radiolabeling with 89Zirconium oxine for PET imaging. J Transl Med 21:367\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKeu KV et al (2017) Reporter gene imaging of targeted T cell immunotherapy in recurrent glioma. Sci Transl Med 9:eaag2196\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorath V et al (2025) PET-based tracking of CAR T cells and viral gene transfer using a cell surface reporter that binds to lanthanide complexes. Nat Biomed Eng, 1\u0026ndash;21\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSellmyer MA et al (2020) Imaging CAR T cell trafficking with eDHFR as a PET reporter gene. Mol Ther 28:42\u0026ndash;51\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSong X et al (2024) Noninvasive longitudinal PET/CT imaging of CAR T cells using PSMA reporter gene. Eur J Nucl Med Mol Imaging 51:965\u0026ndash;977\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMinn I et al (2019) Imaging CAR T cell therapy with PSMA-targeted positron emission tomography. Sci Adv 5:eaaw5096\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang N et al (2025) Development of an in situ CAR-T cell protocol through optical and PSMA-targeted PET imaging. P Natl Acad Sci USA 122:e2504950122\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMurty S et al (2020) PET reporter gene imaging and ganciclovir-mediated ablation of chimeric antigen receptor T cells in solid tumors. Cancer Res 80:4731\u0026ndash;4740\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFr\u0026ouml;se J et al (2024) Development of an antigen-based approach to noninvasively image CAR T cells in real time and as a predictive tool. Sci Adv 10:eadn3816\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBergstr\u0026ouml;m M, Grahnen A, L\u0026aring;ngstr\u0026ouml;m B (2003) Positron emission tomography microdosing: a new concept with application in tracer and early clinical drug development. Eur J Clin Pharmacol 59:357\u0026ndash;366\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZou S et al (2025) Non-invasive assessment of HER2 expression in patients with urothelial carcinoma using [68Ga] Ga-HER2 affibody PET/CT imaging: preliminary clinical findings. Eur J Nucl Med Mol Imaging 52:2782\u0026ndash;2791\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu Y et al (2019) PET imaging of a 68Ga labeled modified HER2 affibody in breast cancers: from xenografts to patients. Br J Radiol 92:20190425\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou N et al (2021) Impact of 68Ga-NOTA-MAL-MZHER2 PET imaging in advanced gastric cancer patients and therapeutic response monitoring. Eur J Nucl Med Mol Imaging 48:161\u0026ndash;175\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRodnick ME et al (2022) Synthesis of 68Ga-radiopharmaceuticals using both generator-derived and cyclotron-produced 68Ga as exemplified by [68Ga] Ga-PSMA-11 for prostate cancer PET imaging. Nat Protoc 17:980\u0026ndash;1003\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWei J et al (2020) The model of cytokine release syndrome in CAR T-cell treatment for B-cell non-Hodgkin lymphoma. Signal Transduct Target Ther 5:134\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGhorashian S et al (2019) Enhanced CAR T cell expansion and prolonged persistence in pediatric patients with ALL treated with a low-affinity CD19 CAR. Nat Med 25:1408\u0026ndash;1414\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaecke HR, Hofmann M (2005) Haberkorn U. 68Ga-labeled peptides in tumor imaging. J Nucl Med 46:172S\u0026ndash;178S\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchwarz SW, Oyama R (2015) The role of exploratory investigational new drugs for translating radiopharmaceuticals into first-in-human studies. J Nucl Med 56:497\u0026ndash;500\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDecristoforo C, Lyashchenko SK (2019) Recommendations for conducting clinical trials with radiopharmaceuticals. In: (ed^ (ed) Nuclear Medicine Textbook: Methodology and Clinical Applications. Springer\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKorde A et al (2022) Practical considerations for navigating the regulatory landscape of non-clinical studies for clinical translation of radiopharmaceuticals. EJNMMI Radiopharm Chem 7:18\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSatpati D (2021) Recent breakthrough in 68Ga-radiopharmaceuticals cold kits for convenient PET radiopharmacy. Bioconjug Chem 32:430\u0026ndash;447\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGillings N et al (2021) Guideline on current good radiopharmacy practice (cGRPP) for the small-scale preparation of radiopharmaceuticals. EJNMMI Radiopharm Chem 6:8\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu L et al (2022) Computational model of CAR T-cell immunotherapy dissects and predicts leukemia patient responses at remission, resistance, and relapse. J Immunother Cancer 10:e005360\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKirouac DC, Zmurchok C, Deyati A, Sicherman J, Bond C, Zandstra PW (2023) Deconvolution of clinical variance in CAR-T cell pharmacology and response. Nat Biotechnol 41:1606\u0026ndash;1617\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":false,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9063774/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9063774/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eCAR T cell therapies targeting CD19 and B cell maturation antigen (BCMA) induce profound responses in B cell malignancies, yet relapse highlight the need for non-invasive, quantitative tools to track cell kinetics. Here we develop an engineering-free, antigen-based PET/MR strategy to track tandem scFv BCMA/CD19 CAR-T product using a \u003csup\u003e68\u003c/sup\u003eGa-labelled minimal BCMA ectodomain probe (BED). [\u003csup\u003e68\u003c/sup\u003eGa]Ga-NOTA-BED retains nanomolar affinity and high specificity for BCMA-scFv-containing cells, detects as few as ~ 2×10\u003csup\u003e4\u003c/sup\u003e cells without compromising effector functions. In mouse models, the probe enables quantitative discrimination of CAR-positive clusters, revealing a linear relationship between PET signal and cell number. Longitudinal PET/MR in lymphoma and myeloma xenografts visualizes heterogeneous CAR T expansion and trafficking patterns that align with distinct response phenotypes under varying antigen burden. This antigen-derived, human-sequence probe provides a repeatable, low-burden framework for kinetic phenotyping of dual-target CAR T therapies without additional cell engineering, and is positioned for clinical translation (NCT:07280793) as an imaging companion to guide patient-specific monitoring and trial design.\u003c/p\u003e","manuscriptTitle":"A 68Ga–BCMA Ectodomain Tracer Enables PET/MR Longitudinal Tracking of Clinical-Grade BCMA/CD19 CAR T Cells","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-15 16:21:58","doi":"10.21203/rs.3.rs-9063774/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"aa94335b-b5ba-4824-b9ee-885d799037a1","owner":[],"postedDate":"May 15th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-04-30T01:28:25+00:00","index":3,"fulltext":"This content is not available."}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":68194811,"name":"Biological sciences/Biological techniques/Imaging/Positron-emission tomography"},{"id":68194812,"name":"Health sciences/Health care/Medical imaging/Radionuclide imaging"}],"tags":[],"updatedAt":"2026-05-15T16:21:58+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-15 16:21:58","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9063774","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9063774","identity":"rs-9063774","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}