Characterization of Chimeric Antigen Receptor Modified T Cells Expressing scFv-IL-13Rα2 after Radiolabeling with 89Zirconium Oxine for PET Imaging

Background Chimeric antigen receptor (CAR) T cell therapy is an exciting cell-based cancer immunotherapy. Unfortunately, CAR-T cell therapy is associated with serious toxicities such as cytokine release syndrome (CRS) and neurotoxicity. The mechanism of these serious adverse events (SAEs) and how homing, distribution and retention of CAR-T cells contribute to toxicities is not fully understood. Methods To determine if radiolabelling of CAR-T cells could support positron emission tomography (PET)-based biodistribution studies, we labeled IL-13Rα2 targeting scFv-IL-13Rα2-CAR-T cells (CAR-T cells) with 89Zirconium-oxine (89Zr-oxine), and characterized and compared their product attributes with non-labeled CAR-T cells. The 89Zr-oxine labeling conditions were optimized for incubation time, temperature, and use of serum for labeling. In addition, product attributes of radiolabeled CAR-T cells were studied to assess their overall quality including cell viability, proliferation, phenotype markers of T-cell activation and exhaustion, cytolytic activity and release of interferon-γ upon co-culture with IL-13Rα2 expressing glioma cells. Results We observed that radiolabeling of CAR-T cells with 89Zr-oxine is quick, efficient, and radioactivity is retained in the cells for at least 8 days with minimal loss. Also, viability of radiolabeled CAR-T cells was similar to that of unlabeled cells as determined by TUNEL assay and caspase 3/7 enzyme activity assay. Moreover, there were no significant changes in T cell activation (CD24, CD44, CD69 and IFN-γ) or T cell exhaustion(PD-1, LAG-3 and TIM3) markers expression between radiolabeled and unlabeled CAR-T cells. In chemotaxis assays, migratory capability of radiolabeled CAR-T cells to IL-13Rα2Fc was similar to that of non-labeled cells. Conclusions Importantly, radiolabeling has minimal impact on biological product attributes including potency of CAR-T cells towards IL-13Rα2 positive tumor cells but not IL-13Rα2 negative cells as measured by cytolytic activity and release of IFN-γ. Thus, IL-13Rα2 targeting CAR-T cells radiolabeled with 89Zr-oxine retain critical product attributes and suggest 89Zr-oxine radiolabeling of CAR-T cells may facilitate biodistribution and tissue trafficking studies in vivo using PET.


Page 3/22
The approach of genetically engineering T cells to express chimeric antigen receptors (CARs) speci cally targeting and killing cancer cells, has become an important immunotherapeutic approach in the treatment of cancer [1][2][3][4][5][6][7][8][9][10]. CAR-T cells initiate T cell-mediated immune responses as they have target speci c antibodies or single chain variable fragment (scFv) of the antibody genetically engineered with the intracellular signaling domains of the T cell receptor and costimulatory molecules [11,12]. Recent advances in adoptive immunotherapy of cancer have led to the Food and Drug Administration (FDA) approval of CAR-cells targeting CD19 and B-cell maturation antigen (BCMA) for advanced B-cell malignancies or relapsed/refractory B-cell malignancies. These CAR-T therapies have shown complete response in a large cohort of patients [7,9,[13][14][15]. However, these therapies are also associated with serious adverse events (SAE) such as Cytokine release syndrome (CRS) and neurotoxicity. CRS is a common complication observed in large percentage of subjects and can range from mild to severe form.
CRS is often managed using tocilizumab (anti-IL-6R antibody), anakinra (anti-IL-1R antibody) or steroids but warrants a careful monitoring of the subjects under medical supervision and management.
Neurotoxicity, which can be lethal, is another adverse event seen in a subset of patients [16][17][18]. Although it is believed that the cytokines and factors released by CAR-T cells and dying tumor cells may be contributing partly to these effects [19][20][21][22][23][24], the pathogenicity of these serious adverse events is not completely understood. Therefore, an understanding of the biodistribution and persistence of CAR-T cells in various tissues and vital organs, and their relationship to observed toxicities could provide insights into CAR T-cell related toxicities.
Molecular imaging can noninvasively monitor binding, tra cking, biodistribution and persistence of CAR-T cells which could improve our understanding of the dynamics of mechanism(s) underlying CRS, neurological and other toxicities. Radiolabeled cells can be visualized in vivo with very high signal-tonoise ratios by using single photon emission computed tomography (SPECT) and positron emission tomography (PET). Since it is desired to have lower radiation exposure while still obtaining high sensitivity, resolution, speci city, and su cient duration to track the cells over multiple days, a long-lived positron-emitting radioisotope is needed. Zirconium-89 ( 89 Zr) is a cyclotron-produced PET isotope with a half-life of 3.27 days and is routinely used to radiolabel large molecules including cells [25]. In the present study, we have used 89 Zr-oxine ( 89 Zr-oxine complex) to develop a technology to radiolabel CAR-T cells.
We have optimized the labeling conditions so that labeled CAR-T cells can be used in tra cking and biodistribution studies in vivo in animal studies by assessing key biological attributes of labeled CAR-T cells to determine the effects of 89 Zr-radilabelling on the biological characteristics of these cells. at room temperature. Oxine in chloroform (1 mg/mL, 500 µL) and 89 ZrCl 4 were mixed in the presence of 10 µL of Tween-80 and pH of the reaction was adjusted to 7.0 to 7.3 with 1M Na 2 CO 3 . The reaction vial was vortexed to facilitate phase transfer and the two phases were allowed to separate (7-10 min). The chloroform phase was carefully extracted with a pipette and transferred to a separate vial. The pooled extract from multiple chloroform extractions was evaporated at 50 ºC under a stream of Argon. The residue containing 89 Zr (oxinate) 4 was re-constituted in 10% ethanol in PBS and sterile ltered. Puri ed and reconstituted 89 Zr (oxinate) 4 was analyzed on silica gel impregnated ITLC strips with ethyl acetate as mobile phase. The developed strips were cut and counted in an automated gamma counter. The radiochemical yields were 60 ± 19% (n = 8) with a radiochemical purity > 95%. The reference compound, Zr (oxinate) 4 was synthesized according to reported procedure [26] and data (MS and 1 H NMR) was consistent with the product formation.

Materials And Methods
Cell cultures and generation of scFv-IL-13Ra2-CAR-T cells.
Jurkat-T cells, A172, U87MG and T98G glioma cells lines were obtained from ATCC and grown as per the supplier's instructions. U251 glioma cell line was obtained from NCI and maintained in RPMI complete medium with 10% FBS. T98G and U87MG glioma cell lines were maintained in EMEM complete medium supplemented with 10% FBS. We have previously characterized these cell lines for IL-13Ra2 expression by RT-PCR for mRNA and ICC analyses for protein expression as described (Joshi et al, 2000). Human PBMCs were isolated from buffy coats of normal healthy blood donors who donated blood at the Division of Transfusion Medicine, NIH.
IL13Rα2-CAR T cells (termed CAR-T cell) were generated from CD4 and CD8 + ve T cells isolated from normal human blood donor buffy coat using Ficol-Paque plus density gradient technique (Cytiva Life sciences, Marlborough, MA), activated with Dynabeads human T cell activator CD3w/CD28 for T cell activation (Thermo Fisher Scienti c, Waltham, MA) and transduced with different multiplicity of infection (m.o.i) as described elsewhere (Joshi et al, unpublished observations). Brie y, a third-generation CAR construct consisting of single chain Fv (scFv) antibody sequence against IL-13Rα2 antibody (as ectodomain) and a CD28 transmembrane domain along with (CD3ζ) and (CD28 and or 41BB) endodomain sequences were placed into pCDH-MSCV-MCS-EF1-copGFP-T2A-Puro lentiviral vector (System Biosciences, Palo Alto, CA ) and packaged into 293T cells by co-transfecting with three helper plasmids (pRRE, pRev and pCMV-VSV-G) to produce self-inactivating (SIN) lentiviral vector expressing CAR as generated IL13Ra2-CAR pseudo-lentivector. Jurkat-T cells were used as a control in parallel to assess the e ciency of transduction using similar m.o.i, and expansion and nal manufacturing of CAR-T cells.
For CAR T cell expansion, the cells were activated with CD3/CD28 Dynabeads (Invitrogen, Carlsbad, CA) at a ratio of 3:1 (T cell to bead) and genetically modi ed via lentiviral transduction. T cells were maintained in culture at 0.6-1 X 10 6 cells/mL in T cell Transact supplemented with 50ng/ml IL-2 (Miltenyi Biotec, Waltham, MA).
Radiolabeling of CAR-T cells with 89 Zr-Oxine and retention of radioactivity. 89 Zr-oxine solution (1µCi-10 µCi) and 10 6 cells in phosphate-buffered saline were incubated at room temperature for 10-60 minutes at 1:20 or 1:40 volume ratios. We compared radiolabeling of CAR-T cells in serum free medium or in complete culture medium at 37°C, room temperature or 4°C. In an independent set of experiments, we incubated 1µCi of 89 Zr-oxine solution with varying number (1-10 X 10 6 ) of CAR-T cells to determine the number of CAR-T cells that can be radiolabeled in 1µCi of 89 Zr-oxine. The cells were washed with complete medium without fetal calf serum twice and with phosphate-buffered saline once.
For retention experiments, 89 Zr-Oxine labeled CAR-T cells (radiolabeled CAR T cells) were cultured in T cell medium and sampled at the indicated time points. Radioactivity within the cell pellet and supernatant was assayed in a gamma-counter (WizzardD2 Automatic Gamma counter, Perkin-Elmer, Waltham, MA) to determine retention of radioactivity. Cell-associated radioactivity was determined as the amount of radioactivity in the nal cell pellet at the speci ed time point.
E ux of 89 Zr-oxine from radiolabeled CAR-T cells.
In a parallel study to determine cellular e ux of radioactivity, we cultured 0.25 × 10 6 89 Zr-labeled CAR-T cells in each well of a six-well culture plate. The medium was replaced with fresh medium daily for 7 days, and radioactivity in the replaced medium was counted in a gamma counter. Three independent experiments were performed in quadruple runs and the results were expressed as mean ± SD.

Cell viability and proliferation of Radiolabeled CAR-T cells.
For all experiments, we examined the cell viability before labeling, after labeling, and at the indicated time points in culture by trypan blue exclusion technique. In an independent experiment, we incubated the cells with 1X PBS, which served as negative control. Each value is expressed as mean ± SD of three independent experiments performed in quadruplicate.
The effect of radiolabeling on CAR-T cellular proliferation was assessed by the CellTiter 96R AQuesous one solution (Promega, Madison, WI). Unlabeled and 89 Zr-labeled CAR-T cells (2,500 cells/well) were plated in quadruple wells of a 96-well culture plate and maintained at 37°C in a CO 2 incubator. Twenty microliters of MTS reagent was added in each well on day 3,5 and 7 and number of proliferating cells in each well was quanti ed at 490nm from absorbance of MTS formazon formed in presence of phenazine ethosulfate. Each value is expressed as mean ± SD of three independent experiments performed in quadruplicate.
We assessed the effect of radiolabeling on intracellular caspase 3/7 activity using earlytox Caspase-3/7 R100 assay kit following manufacturer's recommendations (Molecular Device, Sunnyvale, CA). Unlabeled cells served as control. A known number of unlabeled and 89 Zr-oxine labeled CAR-T cells (200,000/well) were plated in a 6-well culture plate and maintained at 37°C in a CO 2 incubator. Intracellular caspase 3/7 activity from radiolabeled CAR-T cells was determined on day 3 and 7. The endpoint uorescence was measured on a SpectraMax M5 plate reader (Molecular Device, San Jose, CA) and each value is expressed as arbitrary uorescence units.
The number of apoptotic nuclei were counted and quanti ed in quadruplicate on day 3 and 7. Brie y, 100,000 unlabeled and radiolabeled CAR-T cells were plated for 45 minutes in 4 well poly-L-lysine coated glass chambered slides and performed the assay as per manufacturer's instructions (Promega Corporation, Madison, WI). After a brief wash with 1X PBS, 4 elds in each well were counted for 500 cells/ eld by three investigators in a blinded manner (a total of 2000 cells/well) for the TUNEL positive apoptotic cells. The assay was performed in three independent experiments in quadruplicate and data were shown as mean ± SD. [25] Assessment of T cell stimulation and T cell exhaustion biomarker expression in 89 Zr-Oxine labeled CAR-T cells.
Radiolabeled CAR-T cells were evaluated for their expression of CD44, CD25, CD69, and intracellular IFN-γ as T cell activation markers after treating the cells with brefeldin A (Bio Legend, San Diego, CA) by indirect immuno uorescence assay (IFA, abCAM, Cambridge, MA). Radiolabeled CAR-T cells were also examined for PD-1, LAG-3 and TIM3 expression by IFA on day 7 for T cell exhaustion biomarker expression. Each value in both sets of experiments is expressed as mean ± SD of three independent readings performed in quadruplicate for scoring in a blinded fashion for % positive cells expressing ≥ 2 + immuno uorescence intensity.
Analysis of migration potential of 89 Zr-IL-13Rα2.
The migration potential of radiolabeled IL13Rα2-CAR T cells was assessed in 24-well ChemoTx plates with a 5-µm pore diameter (abCam, Cambridge, MA). In the lower chambers, 600 µL of unconditioned Dulbecco modi ed Eagle medium with 10, 50 and 1000ng/ml hu-IL-13Rα2Fc (R&D Systems, Minneapolis, MN ) and 300 µL of either U251-or T98G tumor cell culture conditioned medium mixed with 300 µL of complete medium were added. The upper chambers were loaded with 500,000 CAR T cells/200 µl. After 6 and 20 hours at 37 o C, residual cells were scraped off the polycarbonate lter, and the plate was centrifuged for 2 min at 400 x g. The lter was removed, and cells in the lower chamber were counted by trypan blue exclusion technique. Percentage migration was calculated as the number of cells in the lower chamber divided by the total number of cells plated per well. Each value is expressed as mean ± SD of three independent experiments performed in quadruplicate.
We next performed an in vitro assay to determine the cell killing activity of non-labeled and radiolabeled CAR-T cells by a robust homogeneous uorescence-based non-isotopic cytotoxicity assay. The IL-13Rα2 positive and IL-13Rα2 negative tumor cells are labeled by intracellular Calcein violet-acetoxymethyl ester that has good retention in target cells [27,28]. Release of Calcein Violet in the supernatants recovered at the end of 6 hour of co-culture of target : effector cells in the ratio of 1:10, 1:20, 1:30, 1:40 and 1:50 is measured quantitatively on a uorescent plate reader. The data are shown as mean ± SD of three independent experiments performed in quadruplicate involving co-cultures of radiolabeled and unlabeled CAR-T cells.
Non-labeled and radiolabeled IL13Rα2-CAR T (CAR-T) cells (100,000) were co-cultured for 20 hours with equal number of IL-13Rα2 positive tumor cells in a well of 96 well round bottom plate. The cultures were centrifuged at 3,500 x g for 10 minutes and supernatants were harvested for quantitative determination of IFN-γ secretion by ELISA assay (Bio legend, San Diego, CA). A plate-bound IL-13Rα2-Fc (R&D Systems, Minneapolis, MN; at 250, 500, 750 and 1,000ng/well) was included as positive controls in the assay. Each value is expressed as mean ± SD of three independent experiments performed in quadruplicate.

Statistical analysis:
The data were compared using unpaired Student's t-test analyses. P values less than 0.05, calculated by using GraphPad Prism software (Graph-Pad Software, La Jolla, Calif), were considered to indicate a signi cant difference. Two-way analysis of variance was used to compare labeling conditions (n = 4), the Wilcoxon test was used to obtain two-sided global P values for cell survival or proliferation (n = 4).
We observed an average yield of 60 ± 19% (mean ± SD, n = 8) of 89 Zr-oxine conjugate with radiochemical purity > 95% as investigated with chloroform extraction analyses. The 89 Zr-(oxinate) 3 was sterile ltered and used for subsequent experiments of cell labeling without further puri cation.

Radiolabeling of CAR-T cells with 89 Zr-oxine.
To determine the optimal labeling conditions, we compared cell labeling at 37°C, room temperature, and 4°C by using Jurkat and CAR-T cells in different medium. The highest radioactivity incorporation was achieved when the cells were labeled at room temperature or 4°C in phosphate-buffered saline, serum free RPMI or T cell medium (≤ 0.35 µCi/10 6 cells) compared to cells labeled at 37°C (≤ [0.24 µCi/10 6 cells] (Fig. 1A). Maximum radiolabeling of Jurkat and CAR-T cells occurred in rst 15 minutes (Fig. 1B). Use of complete cell media at room temperature or 4°C decreased the labeling e ciency to about 60-76% of that in phosphate-buffered saline (P ≤ .05). The labeling at 37°C was low under all buffer conditions (Fig. 1C). We also observed that 1.0 µCi of 89 Zr-oxine could be used to e ciently radiolabel 5x10 6 cells ( Fig. 1D). Increasing the number cells beyond 5X10 6 cells did not increase cell uptake of radioactivity in CAR-Jurkat or CAR-T cell under optimal radiolabeling conditions. Effect of radiolabeling with 89 Zr-oxine on viability and proliferation of CAR-T cells.
Next, we determined if radiolabeling of CAR-Jurkat and CAR-T cells with 89 Zr-oxine could exert any effect on cell viability and interfere with biological functions of radiolabeled CAR-T cells. As shown in Fig. 2A, we evaluated cell viability of CAR-Jurkat and CAR-T cells at different time points for 7 days after maintaining them in complete growth medium. No signi cant difference was observed in viable cell numbers between radiolabeled CAR-Jurkat or radiolabeled CAR-T cells and respective non-radiolabeled cells for 7 days post-labeling. We also examined for effects on metabolic and cell proliferation activity. A comparative study of cell proliferation between unlabeled and radiolabeled CAR-T cells showed that radiolabeled CAR-Jurkat or CAR-T cells showed no signi cant difference in proliferation rate from respective unlabeled cells when evaluated by the MTS cell proliferation assay (Fig. 2B). 89 Zr labeled cells retained at least one third of radioactivity on day 8 (≥ 0.35µCi/10 6 cells) from initial value of on day 1 of culture (1.16 µCi/10 6 cells) (Fig. 2C). The observed loss in cell associated activity on day 8 was found to be within predicted values after 2.5 half-life of isotope decay. This was further corroborated by our observations of radioactivity measured from cell free supernatants of radiolabeled CAR-T cells on a daily basis, which revealed no signi cant e ux of radioactivity for a period of 7 days. This con rmed that majority of the radioactivity was retained as cell associated radioactivity during the course of experiment. These data support the use of radiolabeled CAR-T cells for further in vivo evaluation. 89 Zr-oxine radiolabeling did not increase apoptosis in CAR-T cells.
Unlabeled and 89 Zr-oxine radiolabeled CAR-Jurkat and CAR-T cells were analyzed for apoptotic nuclei and Caspase-3/7 activities at day 3 and 7 of culture. As shown in Fig. 3A, no signi cant change in number of apoptotic nuclei was observed between labeled and non-labeled CAR-Jurkat or CAR-T cells on 3 and 7 time points. Also, 89 Zr-oxine radiolabeling did not induce signi cant increase in Caspase-3/7 activity in radiolabeled cells compared to unlabeled cells at day 3 or day 7 (Fig. 3B), which further corroborated the apoptosis assay results. 89 Zr-oxine labeling did not alter CAR-T cell phenotype or function.
As anticipated, we observed that stimulated CAR-T cells showed upregulated expression of CD25, CD44 and CD69 cell surface markers (Fig. 4A-C). As shown in Fig. 4D, stimulation of T cell with anti-CD3/CD28 antibody coated magnetic beads caused expression of intracellular IFN-γ. Radiolabeling of CAR-T cells with 89 Zr-oxine did not change the expression of intracellular IFN-γ. These results suggest that the TCR stimulation of CAR-T cells was not affected by radiolabeling.
To determine the impact of radiolabeling on T cell exhaustion, we examined the expression of three exhaustion markers on CAR-T cells. As shown in Fig. 5, 89 Zr-oxine radiolabeling of CAR-T cells did not show any signi cant change in T cell exhaustion markers (PD-1, LAG-3 and TIM3) compared to unlabeled CAR-T cells, suggesting that the radiolabeled CAR-T cells maintained the phenotype.
Effect of 89 Zr-oxine labeling on cell invasive ability of CAR-T cells.
Next, we examined the effect of 89 Zr-oxine radiolabeling on cell invasion of CAR-T cells using a Boyden chamber assay to measure the response to IL-13Rα2Fc chimeric protein or conditioned medium obtained from IL-13Rα2 positive and IL-13Rα2 negative human glioma cell lines. As shown in Fig. 6, unlabeled and radiolabeled CAR-T cells invaded equally and migrated to human IL-13Rα2Fc in a concentration dependent manner at 6 and 20 hr time points. Similarly, unlabeled and radiolabeled CAR-T cells invaded equally and migrated to conditioned medium from IL-13Rα2 positive glioma cells but not to condition medium from IL-13Rα2 negative glioma cells. Interestingly, there were a greater but non-signi cant number of CAR-T cells migrated to the lower chamber at 20 hour time point. In contrast, no signi cant difference in CAR-T cells migration was observed between unlabeled and radiolabeled. A marginal increase in CAR-T cell migration to bottom chamber between 6 hours and 20 hour time points suggests that viable CAR-T cells continued growing during additional 14 hours but non-signi cant in unlabeled and radiolabeled cell populations.
Radiolabeled CAR-T maintained their potency.
Next, we examined the potency of radiolabeled CAR-T cells against IL-13Rα2 positive U251 and U87 MG, and IL-13Rα2 negative T98G malignant glioma target cells by a cell killing assay. Radiolabeled CAR-T cells mediated the cell killing of U251 and U87MG cells in effector cell number dependent manner similar to unlabeled CAR-T cells. In contrast, no cell killing was observed with IL-13Rα2 negative T98G glioma cells with non-labeled and radiolabeled CAR-T cells (Fig. 7A).
We also determined the potency of CAR-T cells by INF-γ release assay when CAR-T effector cells were cocultured with IL-13Rα2 positive A172, U251 and U87MG cells and IL-13Rα2 negative T98G cells. CAR-T cells were co-cultured with equal numbers of target cells for 20 hrs. and IFN-γ measured in the supernatant by ELISA. As shown in Fig. 7B, the labeled and non-labeled CAR-T cells produced large and equal amount of IFN-γ in the supernatant when cultured with IL-13Rα2 positive glioma cell lines. In contrast, both labeled and non-labeled CAR-T cells secreted basal and minimal amounts of IFN-γ when cultured with IL-13Rα2 negative tumor cell line. Taken together, these data demonstrate that 89 Zrradiolabled IL-13Rα2 targeted CAR-T cells exhibit similar potency to that of non-labeled cells.

Discussion
The primary objective of our present study is to provide proof-of-principle results by developing a highly sensitive and stable imaging tool to evaluate the biodistribution and tra cking of CAR-T cells in vivo in animal studies. Towards that goal, we have studied several crucial biological attributes of unlabeled and interference in these attributes of CAR-T cells related to their functions in vivo. Maintaining these biological attributes after radiolabeling of CAR-T cells is of paramount importance to show that these cells may provide an integrated readout that is re ective of biodistribution and tra cking of functional and healthy CAR-T cells in vivo. The ability to trace radiolabeled CAR-T cells is of high importance for understanding the mechanism of action of e cacy and safety and may provide insight into particularly serious adverse events such as CRS and vital organ toxicities including neurotoxicity associated with this class of therapies.
The synthesis of 89 Zr-oxine complex was accomplished with simple steps of mixing, which resulted in more than 60% of 89 Zr being converted to the complex allowing us to add the resulting solution to the cell suspension for radiolabeling. Our optimization data for radiolabeling of CAR-T cells showed that the maximum radiolabeling occurred in 15 minutes of incubation at room temperature in serum free RPMI or T cell culture medium. A xed amount of 1 µCi of 89 Zr-oxine could optimally radiolabel ~ 4.5 X 10 6 CAR-T cells demonstrating a linear cell associated retention of 89 Zr-oxine with a labeling e ciency in the range of 30%. Since maximum labeling occurred at 4°C and room temperature, we conclude that 89 Zr-oxine complex does not require active cellular incorporation. Our data also con rm previously publish results that 89 Zr-oxine labeling does not depend on active cellular incorporation in vitro in the CAR-T cells [29]. As the cells are labeled and allowed to grow in cell growth medium for next 8 days, they divide after labeling and the speci c activity (activity per cell) is reduced to the predicted values over a period of 8 day incubation. These observations corroborate data reported by other investigators [25,29]. Our results also suggest that the cell growth medium has fetal bovine serum as one of the key constituents, it could reduce the labeling e ciency, maybe due to the interference of serum proteins and lipids that could bind the 89 Zr-oxine complex via transient noncovalent bonding, such as hydrogen bonds, π effect, and hydrophobic bonds. Once labeled, 89 Zr stably remained associated with live cells.
We optimized our labeling conditions such that CAR-T cells derived from Jurkat cell line and from human peripheral blood derived lymphocytes maintained their several of their crucial and functional biological attributes. These attributes were assessed by a matrix of cell-surface and functional biological assays and included CAR-T cell viability, cell proliferation, retention of radionuclide, chemotaxis and intracellular IFN-gamma expression. Radiolabeling of CAR-T cells with 89 Zr-oxime did not affect any of these biological attributes. Both radiolabeled CAR-Jurkat and CAR-T cells retained ≥ 0.5µCi/10 6 cell for the next eight days. This amount of activity is su cient to detect cells by powerful PET cameras after I.V. infusion in mice. Interestingly, 89 Zr-oxine labeled CAR-T cells with radioactivity burden did not lose their invasive attributes in Boyden chamber chemotaxis assays, providing additional evidence that the radiolabeling of CAR-T cells with 89 Zr-oxine did not affect biological functions.
Because interference with cellular and biological functions by radiolabeling with a metal like Zr may cause an inconsistency between the localization of radiolabeled cells detected by imaging techniques and real-time tra cking in vivo, we focused on investigating additional key biological attributes of radiolabeled CAR-T cells. Our in vitro data showed that 89 Zr-oxine used at the optimized labeling conditions did not induce any increase in apoptotic nuclei and caspase 3/7 activities in post-labeled cells. Similarly, we neither observed any signi cant change in T cell activation phenotype markers expression ( CD28, CD44 and CD69) nor intracellular IFN-γ levels in post labeled CAR-T cells. Furthermore, the expression of T cell exhaustion markers (PD-1, LAG-3 and TIM3) was unchanged between unlabeled and radiolabeled CAR-T cells. These data indicate that 89 Zr-oxine radiolabeling has no detrimental effects on cellular functions of CAR-T cells.
Furthermore, 89 Zr-oxine radiolabeling did not alter the most important feature, potency, of CAR-T cells as assessed by two independent assays, cytotoxicity and IFN-γ release assays. Co-culture of IL-13Rα2 positive and IL-13Rα2 negative glioma cells with varying number of unlabeled and radiolabeled CAR-T cells showed similar cytotoxic activity and released similar amount of IFN-γ. Thus, 89 Zr-oxine radiolabeling does not affect the potency of CAR-T cells.
Our strategy of radiolabeling of CAR-T cells with radio isotope 89 Zr offer unique advantages over other radiolabeling techniques. Various strategies have been employed in the past to label cells with imaging isotopes for non-invasive in vivo cell tracking for cell-based therapeutic product imaging. More common among these are, 18 F-FDG (for PET) [30,31] and 111 In-oxine (for SPECT) [32]based imaging analyses.
Though 18 F-FDG is useful for evaluating the delivery of cells and early fate of cells (approximately rst few hours), it is not appropriate for in vivo cell tracking after 24 h post-infusion because of its short halflife and poor retention in the cells. Inability of 18 F-FDG to allow cell tracking beyond 24 hours restricts its usefulness in cell-based therapy. For cell therapy based translational studies including adoptive T cell therapy, early engraftment period of rst couple of weeks post-cell therapy product infusion is the most critical time period for their biodistribution and tra cking in vivo [33,34]. Therefore, bio-imaging technologies should be robust over this time frame to allow evaluation of various interventions assessing adverse events and other toxicities. The capability to monitor genetically engineered or modi ed cells in vivo beyond 24 hours is also of high importance for evaluation of biodistribution and tra cking the cellular products using radiolabeled leukocytes. Our results are in agreement with other published data, which reported that 89 Zr-oxine synthesized with different technologies to radiolabel either CTL, NK, DC, bone marrow and stem cells have no interfering effects on their cell viability, cytotoxic and functional activity [25,[35][36][37][38][39], which may help designing more robust and effective biodistribution and cell tra cking studies using PET/CT or PET/MRI technology.

Conclusions
In conclusion, the 89 Zr-oxine labeling technique of CAR-T cells developed is simple, robust and stable allowing CAR-T cells to maintain their critical biological attributes and functions. Collectively, our in vitro data demonstrate that radiolabeled CAR-T cells exhibit similar proliferation, migratory capacity, activation status, potency and speci city to IL-13Rα2 as that of non-labeled cells.
We believe this success will allow us to better use PET-based non-invasive imaging of cell tra cking and biodistribution of CAR-T cells in vivo. To better realize the full potential of our technique, additional studies are planned to assess the tra cking and biodistribution of human CAR-T cells in vivo in animal models of human cancers that are relevant to cell tra cking and biodistribution. Figure 1 Radiolabeling of CAR-T cells with 89 Zr-oxine (A) 1 X 10 6 cells were radiolabeled with 89 Zr-oxine in PBS, RPMI, RPMI with 10% FBS, T cell medium or T cell medium with IL-2. (B) Optimal time for radiolabeling of CAR-Jurkat and CAR-T cells (C) Effect of temperature on radiolabeling of CAR-Jurkat and CAR-T cells.

Figures
(D) 89 Zr-oxine radiolabeled CAR-Jurkat and CAR-T cells retained cell associated radiolabel in a cell number dependent manner. Three independent experiments were performed in triplicate and results are expressed as mean ± SD.

Figure 2
Page 18/22 Effect of 89 Zr-oxine on cell viability and proliferation of CAR-Jurkat and CAR-T cells. (A) Cell viability of radiolabeled and unlabeled CAR-Jurkat and CAR-T cells was determined at different time points by trypan blue exclusion technique (B) Cell proliferation of radiolabeled and unlabeled CAR-Jurkat and CAR-T cells was determined at different time points by MTS cell proliferation assay. (C) Cell associated radio activities were measured after spinning 1 X 10 6 CAR Jurkat and CAR-T cells to determine retention of radioactivity at different time points. Three independent experiments were performed in triplicate and results were expressed as mean ± SD.   Radiolabeling with 89 Zr-oxine did alter expression of T cell exhaustion markers in CAR-T cells. IFA analysis of 89 Zr-oxine labeled and unlabeled CAR-T cells was performed for T cell exhaustion markers and no signi cant changes in PD-1, LAG-3 and TIM3 expression were noted. Representative data from three independent experiments is shown.