T-Cell Receptor Stimulation Enhances the Expansion and Function Of

Published OnlineFirst September 26, 2019; DOI: 10.1158/1078-0432.CCR-18-3199

Clinical Trials: Immunotherapy Clinical Cancer Research T-Cell Receptor Stimulation Enhances the Expansion and Function of CD19 Chimeric Antigen Receptor–Expressing T Cells Natalia Lapteva1,2, Margaret Gilbert1, Iulia Diaconu1,LisaA.Rollins1, Mina Al-Sabbagh1, Swati Naik1,3,4,RobertA.Krance1,3,4, Tamara Tripic1, Manasa Hiregange1, Darshana Raghavan1,OlgaDakhova1, Rayne H. Rouce1,3,4,HaoLiu1,5, Bilal Omer1,3,4, Barbara Savoldo1,3, Gianpietro Dotti1,2,6, Conrad Russel Cruz1, Keli Sharpe1, Melissa Gates1, Aaron Orozco1, April Durett1, Elizabeth Pacheco1, Adrian P. Gee1,3,CarlosA.Ramos1,6,7, Helen E. Heslop1,3,4,6,7, Malcolm K. Brenner1,3,4,6,7, and Cliona M. Rooney1,2,3,4,8,9

Abstract

Purpose: Current protocols for CD19 chimeric antigen cells (CD19.CAR-VST) without prior cytoreductive chemo- receptor–expressing T cells (CD19.CAR-T cells) require reci- therapy into 8 patients after allogeneic stem cell transplant. pients to tolerate preinfusion cytoreductive chemotherapy, Results: Absent virus reactivation, we saw no CD19. and the presence of sufficient target antigen on normal or CAR-VST expansion. In contrast, in patients with viral reacti- malignant B cells. vation, up to 30,000-fold expansion of CD19.CAR-VSTs was þ Patients and Methods: We investigated whether additional observed, with depletion of CD19 B cells. Five patients stimulation of CD19.CAR-T cells through their native recep- remain in remission at 42–60þ months. tors can substitute for cytoreductive chemotherapy, inducing Conclusion: Dual T-cell receptor and CAR stimulation can expansion and functional persistence of CD19.CAR-T even in thus potentiate effector cell expansion and CAR-target cell patients in remission of B-cell acute lymphocytic leukemia. We killing, even when infusing low numbers of effector cells infused a low dose of CD19.CAR-modified virus-specificT without cytoreduction.

Introduction lymphodepleting chemotherapy. Optimal CAR-T–cell expansion and persistence currently requires administration of toxic cytor- Chimeric antigen receptor–expressing T cells directed to the eductive chemotherapy, which may be undesirable or impractical CD19 antigen (CD19.CAR-T) have proved remarkably effective for some patients (4). Even when cytoreductive therapy is an in treatment of pre-B acute lymphocytic leukemia (ALL) and acceptable option, engraftment of CAR-T cells may be insufficient other B-cell malignancies (1–3). In principle, signaling through for sustained antitumor activity if there are limiting numbers of CD19 and other CARs could be reinforced by concomitant or normal and malignant target cells expressing the CAR-target sequential signaling through the native T-cell receptor (TCR). antigen (5). We therefore examined whether combined TCR and Demonstration of such an effect would have two major benefits, CAR stimulation can obviate preinfusion cytoreduction, and by inducing CAR-T–cell expansion and reducing the toxicity of whether it can also expand functional CAR-T cells even when these cells are infused in small numbers and/or their cognate 1 Center for Cell and Gene Therapy, Baylor College of Medicine, Houston Meth- antigen is present at insufficient levels for their desired expansion odist Hospital, Texas Children's Hospital, Houston, Texas. 2Division of Immu- nology, Department of Pathology, Baylor College of Medicine, Houston, Texas. and persistence. fi 3Division of Hematology and Oncology, Department of Pediatrics, Baylor To assess the putative bene ts of dual TCR/CAR stimulation, we College of Medicine, Houston, Texas. 4Texas Children's Hospital, Houston, Texas. chose patients who received allogeneic hematopoietic stem cell 5Division of Biostatistics, Dan L. Duncan Cancer Center, Baylor College of transplantation (HSCT) as treatment for high risk pre-B-cell 6 Medicine, Houston, Texas. Department of Medicine, Baylor College of Medicine, leukemia. These patients are at risk both for leukemic relapse 7 8 Houston, Texas. Houston Methodist Hospital, Houston, Texas. Program of and for severe viral infections, so that a T-cell product with both Translational Biology and Molecular Medicine, Baylor College of Medicine, Houston, Texas. 9Department of Molecular Virology and Microbiology of Baylor antileukemic and antiviral activity could provide a safe means to fi College of Medicine, Houston, Texas. protect patients from both. Adoptively transferred virus-speci cT cells (VST) proliferate extensively after infusion into recipients of Note: Supplementary data for this article are available at Clinical Cancer – Research Online (http://clincancerres.aacrjournals.org/). T-cell depleted allogeneic HSCT with viral reactivation, and then return to the long-term memory population, where they retain the Corresponding Author: Cliona M. Rooney, Baylor College of Medicine, One ability to reexpand in response to virus reactivation (6). We Baylor Plaza, Houston, TX 77030. Phone: 832-824-4693; Fax: 832-825-4732; fi E-mail: [email protected] reasoned that if donor VSTs were modi ed with leukemia- specific CD19.CAR, they also should expand in the presence of Clin Cancer Res 2019;XX:XX–XX viral infection or reactivation and protect patients from both viral doi: 10.1158/1078-0432.CCR-18-3199 infections and leukemic relapse, thereby achieving dual ends 2019 American Association for Cancer Research. through a single means.

www.aacrjournals.org OF1

Lapteva et al.

cell manufacture at median of 11 days prior to transplant (range Translational Relevance 1–31 days prior to HSCT). It is widely accepted that cytotoxic lymphodepletion prior to adoptive T-cell immunotherapy is essential for the expan- Adenoviral and retroviral vectors sion and antitumor efficacy of adoptively transferred chimeric Both vectors were produced by the Vector Production Facility of antigen receptor–modified T cells (CAR-T cell). However, not the Center for Cell and Gene Therapy, Baylor College of Medicine. all patients, in particular hematopoietic stem cell transplant The CD19.CAR scFv domain, FMC-63, targeting the CD19 antigen (HSCT) recipients, can tolerate this treatment. Here we show was provided by Heddy Zola (Child Health Research Institute, that viruses can induce exponential expansion of CD19 Women's and Children's Hospital, Adelaide, South Australia, chimeric antigen receptor expressing virus-specific T cells Australia; ref. 7). The CAR.CD19-28z vector was generated as without cytokine release syndrome or neurotoxicity in HSCT described previously (8–10). Briefly, a spacer region derived from recipients at high risk for relapse of B-cell acute lymphoblastic the human IgG1-CH2CH3 domain was cloned in-frame between leukemia. This work provides proof of the concept that CAR-T the scFv and the signaling domains and cloned into the SFG cells can be expanded via their T-cell receptor (TCR) without retroviral backbone. Clinical grade packaging cell lines were lymphodepletion and supports investigation of viral vaccines generated with the use of PG13 cells (gibbon ape leukemia virus or oncolytic viruses to expand T cells bispecific for cognate viral pseudotyping packaging cell line; CRL-10686, ATCC). antigens via their TCR and for tumor antigens via their CAR. To generate antigen-presenting cells (APC) for VST generation, we used a previously described adenoviral vector, Ad5f35-pp65 containing the immunodominant pp65 protein of CMV (11, 12), to transduce EBV-transformed B lymphoblastoid cell lines (LCL). Ad5f35-pp65–transduced LCLs presented de novo–expressed We therefore tested the efficacy of CD19.CAR-VSTs in pp65, the hexon and penton proteins from the Ad vector and þ eight HSCT recipients in remission from high-risk, CD19 B-cell endogenous EBV proteins (13). ALL (B-ALL). Patients received donor T cells specific for cytomegalovirus (CMV), Epstein–Barr virus (EBV), and adenovirus Generation of EBV-transformed B LCLs (multivirus-specific T cells) modified with a second generation LCLs were generated in our Good Manufacturing Practices CD19.CAR. CD19.CAR-VSTs could be detected by PCR for a (GMP) facility by infection of peripheral blood mononuclear median of 182 weeks (range 8 weeks to 5 years), but only in the cells (PBMC) from each stem cell donor, with a clinical grade EBV presence of viral reactivation was there a substantive expansion produced by a clinical grade B95-8 producer cell line, in the of CD19.CAR-VSTs and associated B-cell aplasia, despite the presence of cyclosporin A (1 mg/mL; Sandoz) as described pre- presence of significant numbers of normal B cells in all patients viously (14). Autologous LCLs were irradiated and used as APCs at the time of infusion. Hence, concomitant signaling through the after transduction with Ad5f35-pp65. TCR and CAR can enhance the proliferation and the function of CAR-VSTs. Generation of multivirus-specific T cells VSTs specific for CMV, EBV, and adenovirus were expanded from 40–60 mL of peripheral blood from the stem cell donor. Patients and Methods PBMCs were purified using Lymphoprep (Axel-Shield PoC As). To Patients generate APCs for the first stimulation of VSTs, donor LCLs were The study was conducted in accordance with the U.S. Common transduced with Ad5f35-pp65 at 10,000 virus particles per cell, Rule following approval by an Institutional Review Board of irradiated, and cocultured with autologous PBMCs at a 1:40 ratio Baylor College of Medicine the Recombinant DNA Advisory of LCL:PBMCs in T-cell medium comprising 45% advanced Committee and the FDA. Written consent was obtained from all RPMI1640 (Hyclone), 45% Click medium (Irvine Scientific), study subjects. 2 mmol/L/L GlutaMAX TM-I (Invitrogen), and 10% FBS All patients were in remission from B-ALL and had no (HyClone, Thermo Fisher Scientific) in presence of 10 ng/mL IL7 detectable disease at the time of infusion (MRD-negative). The and 100 ng/mL IL15 (R&D Systems). In this protocol we modified subject's stem cell donor was also the donor for the CAR-VST the manufacturing strategy from that reported previously (15) product that was generated from up to 100 mL of peripheral to ensure early transduction of central memory CMV, EBV, blood (Fig. 1A). Patients were infused at least 1 month from and adenovirus-specific T lymphocytes (VST; ref. 16). Three lymphodepleting antibodies and off steroids for treatment of days after culture initiation, VSTs were transduced with the GVHD at the time of infusion. They were allowed to remain on g-retroviral vector expressing CD19.CAR-28z. Transduced VSTs other GVHD prophylaxis. Patients were eligible to receive a (CD19.CAR-VSTs) were cultured for 8–10 days with IL7 and IL15, single dose of 1 107 CD19.CAR-VSTs/m2 after day 30 post and if sufficient cells were obtained, they were cryopreserved for HSCT and were subsequently monitored for toxicity and detec- infusion. Otherwise, CD19.CAR-VSTs were restimulated with tion of transduced VSTs, virus-specific immunity, and B-cell autologous Ad-pp65-LCLs for another 8 days. Ten products were immune reconstitution. manufactured and eight were infused. For these 10 cell lines, þ Seven of 8 patients received myeloablative conditioning ther- viability by 7-AAD staining was 92% 4%, CD3 88% 10%, þ þ þ þ þ apy and 1 received reduced intensity conditioning therapy. Four CD3 CD4 15% 8%, CD3 CD8 63% 20%, CD56 CD3neg þ þ patients received transplants from matched related donors, 2 [natural killer (NK) cells] 11% 10%, CD4 CD45RAnegCD62L þ þ from matched unrelated donors, and 2 from mismatched unre- 10.7% 4.2%, and CD8 CD45RAnegCD62L 30.1% 21%. The þ lated donors. Two of the 8 patients in this cohort had received a average CD19.CAR frequency within the CD19.CAR-VSTs was þ prior myeloablative allogeneic HSCT. Donors were procured for 63% 20%. Of note, the mean fluorescent intensity of CAR VSTs

OF2 Clin Cancer Res; 2019 Clinical Cancer Research

TCR Stimulation for Enhanced CAR-T–Cell Function

A B-ALL paents in remission aer allo-HSCT

Day −8 to −10 Day 0 GVHD Follow-up T-cell Condioning HSCT prophylaxis qPCR, ﬂow cytometry, infusion and IFNγ ELIspot 7 1 × 10 3−4 hours, 1, 2, 4, 6, 8, 12, Donor 2 CD19CAR-VST cells/m 24 weeks aer infusion procurement manufacturing Day 74-158 then yearly for 15 years Day −1 to −31 B C 105

4 g DNA 10 μ 103

102

101 CD19.CAR 1 CD19CAR-copy number/ CD19CAR-copy CD3 0 1 2 4 8 26 52 104 182 Weeks aer infusion

Figure 1. Outline of study design and CD19.CAR-VST cell expansion. A, Stem cell donor blood, obtained between 1 and 31 days before HSCT, was used to manufacture CD19.CAR-VSTs. Cells were infused into patients with B-ALL in remission of HSCT between day 74 and 158 after HSCT. Immune cell reconstitution was monitored by flow cytometry, and virus specificity was analyzed by INFg ELISpot at the indicated times after infusion. The presence of the CD19.CAR transgene was analyzed using qPCR assay. B, Expansion of CD19.CAR-VSTs in patients with EBV virus load. Relatively low levels of expansion of CD19.CAR-VSTs were detected by qPCR in 5 patients without detectable EBV, CMV, or adenovirus (black dashed lines, individual patients; black solid line, mean values). CD19.CAR-VSTs expanded in 3 patients with reactivated EBV (red dashed lines, each patient; red solid line, mean values). C, No significant expansion of CD19.CAR-VSTs in the blood of representative patient with no detectable viral reactivation/infections. did not differ from that of similarly transduced T cells stimulated tohaemagglutinin-activated T-cell blasts in chromium release with CD3 and CD28 antibodies (Supplementary Fig. S1). All T cell assays was a product release criterion. lines generated for infusion were generated in the GMP facility To test NK-cell cytotoxicity, NK cells from patient blood were and tested negative for Mycoplasma and bacteria. expanded using a previously reported protocol (17). Target cells were chromium51-labeled donor-derived CD19.CAR-VSTs, GD2. Cytotoxicity assays CAR-transduced VST (control), nontransduced VSTs (NT), or Cytotoxic activity of CD19.CAR-VSTs was measured using a K562 cells. standard 6-hour Cr51 (Perkin Elmer) release assay. Percent- specific lysis was calculated as specific lysis [(experimental IFNg enzyme-linked immunospot assay release spontaneous release)/(maximum release spontane- IFNg enzyme-linked immunospot (ELISpot) assays were used ous release)] 100. To assess the CAR function, we tested the to test the frequency of T cells specific for each virus, both in the þ ability of each line to kill CD19 Raji Burkitt lymphoma cells, infusion product, and in patients' PBMCs preinfusion and þ þ while donor-derived LCLs provided additional CD19 and EBV on weeks 1, 2, 4, 8, 12, and months 6 and 12 after infusion. autologous target cells (Supplementary Fig. S2A). Raji cells were CD19.CAR-VSTs (1 105/well; Supplementary Fig. S2B) or authenticated in July 2017 using short tandem repeat profiles PBMCs (3 105/well) were plated either with 1 105 LCLs or method by MD Anderson Cancer Center Characterized Cell Line with 1 mg/mL of overlapping peptide libraries (pepmixes) span- core (Houston, TX) and Mycoplasma tested using MycoAlert ning EBNA-1, LMP1, LMP2, BARF-1 of EBV, pp65 of CMV, and (Lonza) on a monthly laboratory schedule of Mycoplasma testing. hexon and penton of adenovirus (JPT Peptide Technologies). Target tumor cells used in the assays were cultured for no longer Data are presented with background (no stimulation) subtracted. than 6 months. Less than 10% lysis of recipient or donor phy- In some assays, HLA-negative K562cs (18) cells expressing

www.aacrjournals.org Clin Cancer Res; 2019 OF3

Lapteva et al.

costimulatory molecules were added at 1 105 cells/well. The Genotyping of Fcg receptors assay was developed as described previously (6, 19). Genotyping of FcgRIIA, FcgRIIIA, and FcgRIIIB was conducted on genomic DNA extracted from patient PBMCs (QIAamp DNA Transgene qPCR Blood Kit, Qiagen) by PCR using Platinum Taq DNA Polymerase To measure the frequency of circulating, transduced CD19. (Invitrogen). Primers and PCR conditions adapted from previ- CAR-VSTs, the CD19.CAR transgene copy number was analyzed ously published methods (21–23). in patient PBMC using qPCR as described before (15). Briefly, genomic DNA was extracted from patient PBMCs and analyzed Statistical analysis with primers from Applied Biosystems (Thermo Fisher Scientific). Data are presented as mean SD and were analyzed using Forward primer 50-CCACCCGCAAGCATTACC-30, reverse primer paired two-tailed Student t test to determine statistical signifi- 50-CGCTCCTGCTGAACTTCACTCT-30 and probe FAM- CACCAC- cance (P < 0.05) of differences when comparing two treatment GCGACTTCGCAGCCT-30 TAMRA. Primers were used at a final groups. The association of CD19.CAR-VST expansion and EBV concentration of 500 nmol/L each, and the probe was used at reactivation was analyzed using Fisher exact test. Data were a final concentration of 120 nmol/L. qPCR was performed analyzed using GraphPad Prism version 5.0 Software (GraphPad with TaqMan Universal Master Mix (Applied Biosystems) on Software Inc). 7900H using default TaqMan program: 50C, 2 minutes; 95C, 10 minutes, (95C, 15 seconds; 60C, 1 minute) 40 cycles. Results Coculture of CD19.CAR-VSTs with B cells Patient characteristics, virus reactivation, and expansion of B cells were purified by negative selection using a human B Cell CD19.CAR-VSTs Isolation Kit (Stemcell Technologies). CD19.CAR-modified Patient demographics are shown in Table 1. Eight pediatric EBV-specific T cells (EBVST) were cocultured with B cells at a allogeneic HSCT recipients in remission from B-ALL received a 1:1 ratio, with or without overlapping peptide libraries spanning single dose of 1 107 CD19.CAR-VSTs/m2 (Fig. 1A). The median the protein sequences (pepmixes) of EBNA-1, LMP-1, LMP-2, and patient age at enrollment was 8 years (range 3–20 years) and BARF-1. Cells were stained with anti-CD3 and goat anti-human median post-HSCT day of infusion was 112 days (range 74– IgG (HþL) AF647 antibodies and acquired for 100 seconds using 158 days). There were no infusion-related toxicities. At the time Gallios Flow Cytometer (Beckman Coulter Inc.) on day 0 and 7 of of infusion, absolute peripheral blood lymphocyte counts were þ each stimulation. 946 569/mL, and average CD19 B-cell counts were 273 258 cells/mL; hence patients were not severely lymphopenic (Table 1). Flow cytometry All patients were negative for minimal residual disease at the time CD19.CAR expression levels in VSTs and PBMCs was measured of treatment. by flow cytometry using goat anti-human-IgG (HþL)-AF647 We measured the frequency of CD19.CAR-VSTs in peripheral antibody (Jackson ImmunoResearch) and analyzed using a blood before and after infusion by stimulating PBMCs with viral FACSCalibur flow cytometer. VSTs were phenotyped using antigens in IFNg ELISpot assays, and by determining the copy fluorochrome-conjugated CD3, CD4, CD8, CD16, CD19, number of transgene-positive/expressing cells by PCR and flow CD45RA, CD45RO, CD62L, CCR7, CD27, and CD28 mAbs (BD cytometry, respectively. We determined the in vivo antiviral Biosciences) and acquired using BD FACSCanto-II (3-laser) activity of these cells by measuring changes in viral load by PCR. Cytometer and analyzed with FACSDiva Software v6.1.3. EBV DNA was detected in circulating PBMCs in only 3 patients, CD19.CAR expression was not analyzed in follow-up samples while no patients reactivated CMV or experienced detectable with low transgene copy numbers by qPCR. adenoviral infection. Therefore, in 5 patients, the only apparent source of CD19.CAR-VST stimulation was the normal B cells Detection of EBV and adenovirus DNA in PBMCs by qPCR (CAR-mediated stimulation), while in 3 patients EBV antigens DNA was isolated from purified PBMCs using Blood DNA were also available for concomitant CAR- and TCR-mediated Isolation Kit (Qiagen). A total of 500 ng of DNA was analyzed stimulation. by qPCR as described previously (20). Adenovirus and CMV DNA In the 5 patients without detectable EBV, CMV, or adenovirus were measured by Viracor IBT Laboratories. reactivation or infection, there was a relatively low level of

Table 1. Patient characteristics at the time of T-cell infusion Disease status Days after Lymphocytes/mL CD19þ B cell EBV copies/mg Tacrolimus level at time infusion Subject no. Age (y) before HSCT HSCT blood at infusion # at infusion PBMC DNA CNI in ng/mL (if applicable) 1 8 CR2 104 316 142 3,445 Weaning 4.3 2 3 CR1 158 681 176a ND No N/A 3 14 CR2 89 598 453 149 Weaning 3.8 4 20 CR3 83 1,404 357 255 Weaning 2.2 5 4 CR2 74 1,851 1,038 ND Weaning <2 6 8 CR3 120 878 109 ND No N/A 7 9 CR3 129 1,487 737 ND No N/A 8 7 CR1 139 350 112 ND No N/A Mean SD 9 6 N/A 112 29 946 569 273 258 481 1,201 NOTE: Boldface indicates mean SD. Abbreviations: CNI, calcineurin inhibitors; CR, complete remission #; N/A, not applicable; ND, not detected (limit of detection is 10 copies). aAnalyzed on day 6 after infusion.

OF4 Clin Cancer Res; 2019 Clinical Cancer Research

TCR Stimulation for Enhanced CAR-T–Cell Function

A C + 31,000 copies CD19CAR triVSTs 105 6,000 105 CD19+ B Cells 150

CD19.CAR EBV DNA (copies/ g DNA g DNA 4 4 #CD19+ Cells/ m m m 10 10 EBV 103 103 102 2 10 1 L 2,000 m 10 50 g) 10 100

-1 CD19CAR-copy number/ CD19CAR-copy 1 number/ CD19CAR-copy 10 0 4h 1 2 6 12 50 0246913243852 Weeks Weeks

B Preinfusion 2 weeks D Preinfusion 68% 32% Week 3 Month 6

PBMC 300

5 Month 9 CD3 Month 12 Month 24 200 Month 30 CD19.CAR CD19.CAR

17% 47% 100

M C

CD3 0

Spot-forming cells per 3 ¥ 10 ne 16% 0% CL o L Adpx l EBVpx CMVpx LCL a CD19 CD19

Figure 2. CD19.CAR-VST expansion associated with EBV reactivation in patient 1. A, Expansion of CD19.CAR-VSTs and control of reactivated EBV. B, CD19.CAR-VSTs were detectable by ﬂow cytometry in patient's peripheral blood (top). CD19þ normal B cells were depleted from blood on week 2 after infusion (bottom). C, Normal B þ þ cells were completely depleted from the blood of patient from week 2 until week 38. CD19 B-cell number remains low (<100 CD19 cells/mL) at 18 months after infusion. D, Reactivity in PBMCs to three viruses (EBV, CMV, and adenoviruses) was measured in PBMCs by IFNg ELISpot. EBV responses were analyzed with LCLs and overlapping peptide libraries to EBNA1, LMP1/2, BZLF1, BARF-1, and EBNA3a, 3b, and 3c (EBVpx). Responses to CMV were analyzed using overlapping pepmixes to pp65 (CMV px); adenovirus responses were analyzed using overlapping peptide libraries for hexon and penton (Ad px). LCLs alone were plated (timepoints 9, 12, 24, and 30 months) as controls and did not produce IFNg (0 SFC/well).

þ expansion of VSTs [as measured by IFNg ELISpot assays or of CAR cells have persisted for over 60 months, at which time the CD19.CAR-VSTs by PCR (Fig. 1B) or flow cytometry (Fig. 1C)]. transgene was detected at 0.8 copies/mg DNA. In this patient, Transgene expression in PBMCs was 20 copies/mg PBMC DNA CD19.CAR-VSTs could be detected by flow cytometry at week 2 þ for most of the timepoints in all of these patients, except for post-infusion (Fig. 2B) with 32% of CD3 T cells expressing the patients 2 and 6, whose transgene levels increased at week 2 to CAR. This striking CD19.CAR-VST cell expansion coincided with 77.8 and 56.4 copies/mg PBMC DNA, respectively. complete B-cell aplasia, until month 6, when B cells became Three patients (1, 3, and 4) had measurable peripheral blood detectable at low levels (18/mL; Fig. 2C); B lymphopenia persisted EBV load at the time of infusion (Table 1), and this was associated for 17 months post-infusion. Coincident with increasing trans- with a correspondingly greater expansion of transgene-positive gene expression, we detected an increase in the frequency of EBV- T cells in PBMC than in patients without a detectable virus load specific T cells that remain detectable for at least 2 years post- (P ¼ 0.02; Fig. 1B). infusion (Fig. 2D). Patient 1 experienced no cytokine release Patient 1 had the highest virus load with 3,445 EBV gen- syndrome, neurotoxicity, or other CD19.CAR-VST–associated omes/mg PBMC DNA at the time of infusion, peaking at 5,204 toxicities even when CD19.CAR-VSTs comprised 32% of periph- þ copies/mg by 4 days post-infusion, and decreasing to undetectable eral blood CD3 T cells (Fig. 2B). This patient remains in remis- levels by day 18 (Fig. 2A). This patient also showed the greatest sion from B-ALL at 60þ months. expansion of CD19.CAR-VSTs, increasing from 7.6 transgene Patient 3 had a low level of EBV reactivation (149 copies/mg copies/mg at 3 hours post-infusion to 30,810 copies by week 2. DNA) and concomitant, but transient, CD19.CAR-VST

www.aacrjournals.org Clin Cancer Res; 2019 OF5

Lapteva et al.

Figure 3. Expansion of CD19.CAR-VSTs upon EBV reactivation in patients 3 and 4. A and B, qPCR data for the CD19.CAR and EBV in patients 3 and 4, respectively. C, CD19.CAR-VSTs were detected by ﬂow cytometry PBMCs from patient 4 on day 10 after infusion, and CD19þ B cells were depleted coincident with CD19.CAR-VST expansion. D, Transient depletion of CD19þ B cells and expansion of CD19.CAR-VSTs in patient 4 during the study period.

expansion, with the highest peak of 215 copies/mg PBMC DNA at Patient 4 had an EBV load of 255 copies/mg peripheral blood week 4, dropping to 6.2 copies/mg DNA at month 6 (Fig. 3A). This DNA at the time of infusion. CD19.CAR-VSTs increased from 5 patient had EBV-specific T-cell reactivity throughout the study. copies at the 3-hour timepoint to 4,426 transgene copies/mg DNA Adenovirus-specific T cells were undetectable at the time of on day 11 post-infusion when they were detectable by flow infusion, but increased in frequency at week 4 post-infusion and cytometry (Fig. 3B and C). Although no increase in the frequency were still detected 3 months after T-cell infusion, even though of T cells reactive with EBV pepmixes representing EBV type 2 adenovirus was not detected in the patient's blood (Supplemen- latency antigens (LMP1, LMP2, EBNA1, and BARF1) was detected þ tary Fig. S3A). Patient 3 did not develop CD19 normal B-cell by IFNg ELISpot (Supplementary Fig. S3B), there was an increase aplasia, relapsed at 15 months after T-cell infusion, and died of in the response to the autologous EBV-LCL that expresses multiple disease. EBV latency and lytic cycle antigens. The high EBV virus load

Table 2. Clinical outcomes, CD19.CAR-VST cell expansion, and B-cell depletion Highest CD19.CAR copies/mg Clinical outcome Reported MRD status after Patient # DNA (time after infusion) (time after infusion)a CD19.CAR-VST therapy 1 30,810 (week 2) Disease-free (60 moþ) Negative (55 mo A.I.) 2 77.8 (week 2) Relapsed 41 mo after T-cell infusion Detected MRD (41 mo A.I.) 3 215 (week 4) DOD 13 mo after T-cell infusion Detected MRD (3 mo A.I.) 4 4,426 (week 1.5) Disease-free (42 moþ) Negative (21þ mo A.I.) 5 5.17 (month 30) Disease-free (48 moþ) Negative (42þ mo A.I.) 6 56.4 (week 2) Disease-free (48 moþ) Negative (47þ mo A.I.) 7 12.7 (week 1) DOD 11 mo after T-cell infusion Detected MRD (2 mo A.I.) 8 14 (week 2) Disease-free (42 moþ) Negative (19þ mo A.I.) Abbreviations: A.I., after T-cell infusion; DOD, died of disease; mo, month; MRD, minimal residual disease. aReported at the time of article submission.

OF6 Clin Cancer Res; 2019 Clinical Cancer Research

TCR Stimulation for Enhanced CAR-T–Cell Function

decreased to undetectable levels by week 2 after infusion (Fig. 3B). this failure, first by determining whether CD19.CAR-VSTs could þ þ Normal CD19 B cells were depleted in this patient for 2 months proliferate in vitro in response to normal CD19 B cells or EBV- þ (Fig. 3D) and the patient remains in remission for more than antigen–expressing CD19 B cells. CD19.CAR-VSTs cocultured þ 42 months after CD19.CAR-VST infusion with transgene detect- with purified autologous CD19 B cells from three different able at 12.9 copies/mg DNA. donors could recognize and kill normal B cells (Fig. 4A), but Of note, the higher frequency of CD19.CAR-VSTs in patients 1 proliferated poorly (total fold expansion 13 10-fold) in and 4 was accompanied by normal B-cell aplasia (See Figs. 2B and response to serial stimulation with autologous B cells. However C and 3D). proliferation was enhanced when the autologous B cells were Overall, 5 of the 8 patients remain disease free, 2 patients pulsed with EBV pepmixes (246 115-fold; Fig. 4B and C), died of disease (patients 3 and 7), and patient 2 relapsed supporting the in vivo indication that stimulation through the TCR at 41 months post T-cell infusion and was transferred to a was more effective at inducing T-cell proliferation than stimula- commercial CD19.CAR-T–cell therapy (Table 2). There was no tion through the CAR. increase in the frequency of CD19.CAR-VSTs at the time of We next determined whether lack of CD19.CAR-VST expansion relapse as measure by PCR for the transgene in peripheral following CAR stimulation alone could be explained by the Fc blood. binding domain of the IgG1-derived CH2-CH3 spacer of the CAR. Hudecek and colleagues suggested that an IgG4-derived Analysis of CD19.CAR-VSTs ex vivo CH2-CH3 spacer element in CARs can bind to the Fc receptor CD19.CAR-VSTs showed little expansion in 5 of the 8 patients, on myeloid cells, leading to activation-induced cell death (AICD) despite the high numbers of circulating B cells (Supplementary in vivo (24). To test whether binding of our IgG1 CH2.CH3 Table S1), suggesting that while CD19.CAR-VSTs could expand domain by monocytes could induce AICD or even direct killing rapidly in response to TCR stimulation, they expanded poorly in of CAR-VSTs, we cocultured CD19.CAR-VSTs and NT-VSTs with þ response to CAR stimulation. We investigated the mechanism for CD14 -purified monocytes for 3 days. Both CD19.CAR-VSTs and

Figure 4. CD19.CAR-VSTs were cytotoxic to autologous normal B cells and expanded in vitro. A, B cells were puriﬁed by positive selection with CD19 microbeads and cocultured at 1:1 ratio with CD19.CAR-VSTs or NT-VSTs for 3 days. CD19.CAR-VSTs eliminated B cells (CD20þ) on day 1 of culture. B, CD19.CARþ EBVST cell expansion in response to multiple stimulations with B cells pulsed with EBV pepmixes or B cells alone. C, Expansion of CD19.CARþCD3þ EBVSTs on day 7 of coculture with B cells alone (13 10-fold cumulative expansion) or B cells pulsed with EBV pepmixes (246 115-fold). Representative results from one donor after the third stimulation are shown.

www.aacrjournals.org Clin Cancer Res; 2019 OF7

Lapteva et al.

Figure 5. VSTs are not eliminated by monocytes in cocultures. A, CD19.CAR-VSTs or NT-VSTs were cocultured with autologous CD14þ monocytes for 3 days with or without EBV pepmixes (px) spanning EBNA1, LMP1, and LMP2 antigens. Viability of CD3þ T cells was assessed on day 3 by flow cytometry with 7ADD (n ¼ 3). B, Fold of expansion of CD19.CAR-VSTs and NT-VSTs 3 days after coculture with monocytes with or without EBV pepmix stimulation. Cells were harvested and analyzed by flow cytometry with Dynabeads to obtain absolute cell numbers (n ¼ 3). No statistically significant difference for CAR.VSTs pulsed with EBV pepmixes was observed at early days (day 3) after stimulation. C, Monocytes are eliminated on day 1 of coculture with CD19.CAR-VSTs and nontransduced cells.

NT-VSTs showed high viability after 3 days of coculture with RIIA: H/H131, high affinity binding; H/R131, intermediate; monocytes in the presence or absence of EBV pepmixes (Fig. 5A), R/R131, and low affinity binding to IgG; Polymorphism RIIA: and retained their ability to respond to EBV pepmixes (Fig. 5B). V/V158, high affinity; V/F158, intermediate; and F/F158, low Monocytes became undetectable after 3 days of coculture with affinity binding; and polymorphism RIIIB: NA1/NA1, high affin- either CD19.CAR VSTs or NT-VSTs (Fig. 5C) both in the presence ity; NA1/NA2, intermediate; and NA2/NA2, low affinity binding and absence EBV pepmixes. Because NK cells also express FcRs, we to IgG. We found no association between high and low affinity expanded NK cells from patients 1 and 5 as effector cells and FcR binding and expansion of CD19.CAR-VSTs (Supplementary added autologous CD19.CAR-VSTs as targets. We found no Table S2). significant death of CD19.CAR-VSTs induced by either patient's NK cells (Supplementary Fig. S4). Finally, we excluded the possibility that the affinity of patient Discussion FcRs for immunoglobulin constant regions determined the in vivo We infused a low dose of second-generation CD19.CAR-VSTs expansion and persistence of the CH2-CH3 containing CD19. into 8 patients who had received HSCT for high-risk B-ALL in the CAR-VSTs, rather than the presence of TCR stimulation from viral absence of cytoreduction. Although all patients had circulating þ reactivation. We analyzed known polymorphic regions of CD32 normal CD19 B cells at the time of infusion, CD19.CAR-VST and CD16 FcRs that might mediate binding and killing (21–23). expansion and normal B-cell elimination occurred only in the

OF8 Clin Cancer Res; 2019 Clinical Cancer Research

TCR Stimulation for Enhanced CAR-T–Cell Function

presence of concomitant TCR stimulation by viral antigens. The The substantial in vivo expansion and persistence of CD19. degree of CD19.CAR-VST expansion and subsequent B-cell apla- CAR-VSTs in the presence of virus reactivation suggested sia correlated with EBV load in peripheral blood, suggesting that that the CH2.CH3 domain of IgG1 does not lead to lung in the absence of the lymphoproliferative milieu resulting from entrapment or activation-induced cell death as has been cytoreductive chemotherapy, in patients with low/no disease observed in animal models using CARs containing CH2.CH3 burden, CAR stimulation alone can be insufficient for effector domains from IgG4 (24). Supporting this suggestion, in vitro cell expansion and that this deficiency can be overcome by the coculture of our IgG1 domain-containing CD19.CAR-VSTs addition of TCR stimulation. with FcR-positive monocytes or NK cells did not result in In most clinical studies, significant expansion and function T-cell death, nor did the affinity of the CD32 and CD16 FcRs of CAR-T cells, requires extensive cytoreduction with cytoxan in these patients alter the performance of the CAR-VST, and fludarabine. Among other effects, these drugs produce whose behavior was unaffected by whether the patients lymphodepletion, which liberates homeostatic cytokines to expressed the high- or low-affinity polymorphisms of these support CAR-T–cell expansion. In this study HSCT recipients receptors. received no further cytoreductive therapy before receiving a In this study, CD19.CAR-VSTs expanded to a greater degree in low dose (107/m2) of CD19.CAR-VSTs between 74 and the presence of virus reactivation than observed with our earlier 158 days posttransplant. Although their immune reconstitution study using VSTs transduced with the same CD19.CAR (ref. 15; was incomplete at this time, T and B lymphocytes circulated with Supplementary Fig. S6). These related studies used different normal frequencies (946 569 lymphocytes/mL; Table 1) versus approaches to manufacture CD19.CAR-VST. In the earlier study, less than 100 lymphocytes/mL after cytoxan and fludara- VSTs were transduced on day 19 of culture, 3 days after the third bine (25, 26). Hence, the presence of near-normal numbers of stimulation when T cells demonstrated an effector memory þ circulating CD19 B cells was insufficient to induce substantial phenotype. In this study, we transduced VSTs 3 days after the CD19.CAR-VST expansion. first stimulation, when T cells with a central memory phenotype In patients with EBV reactivation, CD19.CAR-VST expansion and long-term memory potential were transduced (16). When we þ and corresponding elimination of normal B cells occurred analyzed CAR T cells isolated from patient 4, 3 weeks without cytokine release syndrome (CRS), even when the after infusion we found the proportion of central memory T cells þ CD19.CAR-VST cells comprised 32% of all peripheral CD3 T was higher than in the infused T-cell line (54% vs. 4.3% þ þ þ þ cells (30,000 transgene copies/mg DNA in patient 1). CRS has CD45RO CCR7 of CD19.CAR CD3 T cells in the blood been associated with lymphodepletion and high tumor burden, and in the line, respectively), showing that like unmodified þ neither of which applied to our patients. In several clinical trials VSTs, CAR VSTs were able to reenter the memory compartment we have observed massive expansion of virus-specific T-cells in (Supplementary Fig. S7). Furthermore, the early-transduced HSCT recipients with virus infection or reactivation, including CD19.CAR-VSTs showed greater expansion in the presence or several with bulky EBV-driven lymphoma, and none of these absence of virus reactivation than in previous study (Supplemen- patients experienced severe CRS or neurotoxicity. It is possible tary Fig. S6; ref. 15). Therefore, we propose that early that VSTs that are derived from the memory compartment are less transduction, allowing infusion of CD19.CAR-VSTs with a central reactive with myeloid cells than na€ve cells that are expanded memory phenotype was likely a major contributor to the polyclonally in most clinical trials of CAR-T cells. This virus greater expansion, persistence, and function of CD19.CAR-VSTs specificity likely also explains why none of our patients experi- in this study. enced GVHD, even during CD19.CAR-VST expansion, and Extensive CD19.CAR-VST expansion was observed only in despite HLA mismatches in 2 patients. This has been a consistent patients who experienced virus reactivation and B-cell depletion finding in patients receiving virus-specific T cells from allogeneic occurred only in these patients. The only virus detected during the donors (27), even when derived from partially HLA-mismatched post-infusion period was EBV, so we cannot determine whether third-party donors HLA-matched at only one or two alleles (28), the same CD19.CAR-VST expansion would have occurred in and is likely a result of the greatly reduced TCR repertoire of T cells association with CMV or adenovirus reactivation. While we did specific activated with viral antigens rather than polyclonally observe an increase in the frequency of adenovirus-specific T cells activated with CD3 and CD28 antibodies." without detectable adenovirus in blood, it is possible that the Although our results suggest that stimulation through the TCR patient had a subclinical adenovirus infection, with virus presence can promote the activity of CD19.CAR-VSTs, repeated stimulation in other sites such as in stool or urine. Lack of CD19.CAR-VST via the CAR may not always prove beneficial for TCR expres- expansion in the absence of virus reactivation suggests that sion (29, 30). As illustrated in Supplementary Fig. S5, repeated signaling through the CAR on its own was not sufficient to induce stimulation in vitro via the CAR can lead to downregulation of the substantial CAR-T–cell expansion or function in vivo, despite the TCR. This may explain the lack of apparent expansion of VSTs in CD28-derived costimulatory endodomain and the presence of þ patient 4. At the time of infusion, this patient had a measurable normal CD19 B cells. In contrast, stimulation via the TCR EBV load that responded rapidly to CD19.CAR-VSTs, which achieved this goal, supporting our initial hypothesis that in vivo expanded equally rapidly as measured by PCR and eliminated TCR stimulation could enhance the function of the CAR in T cells. normal B cells. However, there was no apparent increase in the These attributable benefits of dual TCR/CAR stimulation may frequency of circulating T cells for EBV, CMV, or adenovirus as result from more physiologic signals transmitted by the TCR or measured by IFNg ELISpot assay. In contrast, in patients 1 and 3 because EBV reactivation is associated with inflammation pro- who had a high virus load, CD19.CAR-VST expansion correlated duced by EBV-derived pathogen-associated molecular patterns with both a decrease in virus load and an increase in the frequency that stimulate pattern recognition receptors of the innate immune of EBVSTs. This paradox is difficult to explain and is the subject of system producing a more conducive microenvironment for T-cell further studies. activation.

www.aacrjournals.org Clin Cancer Res; 2019 OF9

Lapteva et al.

In conclusion, we have demonstrated that VSTs modified with a interests in Tessa Therapeutics, Allogene, and BluebirdBio. She is also an CD19.CAR containing a long IgG1 spacer and a CD28 endodo- advisory board member/unpaid consultant for CellGenix. No potential con- fl main can expand, persist long term, and produce TCR-mediated icts of interest were disclosed by the other authors. þ antiviral effects through their TCRs and concomitant anti-CD19 Authors' Contributions B-cell activity through their CARs. These benefits are observed Conception and design: H. Liu, G. Dotti, H.E. Heslop, M.K. Brenner, even in patients who are unsuited to potent cytoreductive C.M. Rooney drugs and were not associated with toxicities. The use of CD19. Development of methodology: N. Lapteva, I. Diaconu, O. Dakhova, CAR-VSTs, perhaps with a viral vaccine, may be of particular value B. Savoldo, G. Dotti, A.P. Gee, C.M. Rooney post-remission induction for patients with low disease burden Acquisition of data (provided animals, acquired and managed patients, who are at high risk of relapse and it would be of interest to provided facilities, etc.): N. Lapteva, M. Gilbert, L.A. Rollins, S. Naik, discover whether the enhancement attributable to dual receptor R.A. Krance, T. Tripic, M. Hiregange, R.H. Rouce, B. Omer, A. Durett, E. Pacheco, C.A. Ramos, A. Orozco, K. Sharpe, M. Gates, D. Raghavan stimulation can be extended to other diseases such as breast Analysis and interpretation of data (e.g., statistical analysis, biostatistics, cancer, in which patients have dormant disseminated tumor cells computational analysis): N. Lapteva, I. Diaconu, M. Al-Sabbagh, H. Liu, that may on their own be insufficient to produce stimulation A. Durett, M.K. Brenner, C.M. Rooney, M. Gates through a CAR alone. Writing, review, and/or revision of the manuscript: N. Lapteva, M. Gilbert, I. Diaconu, L.A. Rollins, S. Naik, R.A. Krance, R.H. Rouce, B. Omer, G. Dotti, Disclosure of Potential Conflicts of Interest C.R. Cruz, C.A. Ramos, H.E. Heslop, M.K. Brenner, C.M. Rooney Administrative, technical, or material support (i.e., reporting or organizing I. Diaconu is an employee/paid consultant for Bluebirdbio and ElevateBio. data, constructing databases): M. Gilbert, M. Al-Sabbagh, A. Durett, E. Pacheco, R.H. Rouce reports receiving commercial research grants from TESSA Pharma- C.M. Rooney ceuticals, reports receiving other commercial research support and speakers Study supervision: E. Pacheco, C.A. Ramos bureau honoraria from Novartis and Gilead/Kite Pharma. G. Dotti reports Other (cell product manufacturing): A.P. Gee receiving other commercial research support from Cell Medica. C.R. Cruz holds ownership interest (including patents) in and is an advisory board member/ unpaid consultant for Mana Therapeutics. C.A. Ramos is an employee/paid Acknowledgments consultant for Novartis and Celgene and reports receiving commercial research This work was supported by grants from V-Foundation, Alex's Lemonade grants from Tessa Therapeutics. H.E. Heslop is an employee/paid consultant for Stand Foundation, NIH-NCI P50 CA126752 SPORE in lymphoma and Leu- Gilead Biosciences, Novartis, Cytosen, Tessa Therapeutics, and Marker Thera- kemia and lymphoma Society SCOR 7001. peutics, reports receiving commercial research grants from Cell Medica and Tessa Therapeutics, and holds ownership interest (including patents) in AlloVir The costs of publication of this article were defrayed in part by the payment of advertisement and Marker Therapeutics. M.K. Brenner is an employee/paid consultant for and page charges. This article must therefore be hereby marked in holds ownership interest (including patents) in Tessa Therapeutics, Marker accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Therapeutics, Allovir, Allogene, and BluebirdBio. C.M. Rooney is an employee/ paid consultant for Tessa Therapeutics, holds ownership interest (including Received December 24, 2018; revised April 29, 2019; accepted September 17, patents) in Allovir and Marker Therapeutics, while spouse has ownership 2019; published first October 1, 2019.

References 1. Schuster SJ, Svoboda J, Chong EA, Nasta SD, Mato AR, Anak O,€ et al. enhances chimeric T-cell resistance to T regulatory cells. Leukemia 2006; Chimeric antigen receptor T cells in refractory B-Cell Lymphomas. N Engl J 20:1819–28. Med 2017;377:2545–54. 10. Roessig C, Scherer SP, Baer A, Vormoor J, Rooney CM, Brenner MK, et al. 2. Kochenderfer JN, Dudley ME, Carpenter RO, Kassim SH, Rose JJ, Telford Targeting CD19 with genetically modified EBV-specific human T lympho- WG, et al. Donor-derived CD19-targeted T cells cause regression of malig- cytes. Ann Hematol 2002;81:S42–3. nancy persisting after allogeneic hematopoietic stem cell transplantation. 11. Sili U, Huls MH, Davis AR, Gottschalk S, Brenner MK, Heslop HE, et al. Blood 2013;122:4129–39. Large-scale expansion of dendritic cell-primed polyclonal human cytotoxic 3. Park JH, Riviere I, Gonen M, Wang X, Senechal B, Curran KJ, et al. Long-term T-lymphocyte lines using lymphoblastoid cell lines for adoptive immu- follow-up of CD19 CAR therapy in acute lymphoblastic leukemia. N Engl J notherapy. J Immunother 2003;26:241–56. Med 2018;378:449–59. 12. Leen AM, Myers GD, Sili U, Huls MH, Weiss H, Leung KS, et al. 4. Fried S, Avigdor A, Bielorai B, Meir A, Besser MJ, Schachter J, et al. Early and Monoculture-derived T lymphocytes specific for multiple viruses expand late hematologic toxicity following CD19 CAR-T cells. Bone Marrow and produce clinically relevant effects in immunocompromised indivi- Transplant 2019;54:1643–50. duals. Nat Med 2006;12:1160–6. 5. Turtle CJ, Hanafi LA, Berger C, Gooley TA, Cherian S, Hudecek M, et al. 13. Leen A, Ratnayake M, Foster A, Heym K, Ahmed N, Rooney CM, et al. CD19 CAR-T cells of defined CD4þ:CD8þ composition in adult B cell ALL Contact-activated monocytes: efficient antigen presenting cells for the patients. J Clin Invest 2016;126:2123–38. stimulation of antigen-specific T cells. J Immunother 2007;30: 6. Leen AM, Christin A, Myers GD, Liu H, Cruz CR, Hanley PJ, et al. Cytotoxic 96–107. T lymphocyte therapy with donor T cells prevents and treats adenovirus 14. Smith CA, Ng CY, Heslop HE, Holladay MS, Richardson S, Turner EV, et al. and Epstein-Barr virus infections after haploidentical and matched unre- Production of genetically modified Epstein-Barr virus-specific cytotoxic T lated stem cell transplantation. Blood 2009;114:4283–92. cells for adoptive transfer to patients at high risk of EBV-associated 7. Nicholson IC, Lenton KA, Little DJ, Decorso T, Lee FT, Scott AM, et al. lymphoproliferative disease. J Hematother 1995;4:73–9. Construction and characterisation of a functional CD19 specific single 15. Cruz CR, Micklethwaite KP, Savoldo B, Ramos CA, Lam S, Ku S, et al. chain Fv fragment for immunotherapy of B lineage leukaemia and lym- Infusion of donor-derived CD19-redirected virus-specific T cells for B-cell phoma. Mol Immunol 1997;34:1157–65. malignancies relapsed after allogeneic stem cell transplant: a phase 1 study. 8. Vera J, Savoldo B, Vigouroux S, Biagi E, Pule M, Rossig C, et al. T Blood 2013;122:2965–73. lymphocytes redirected against the kappa light chain of human immuno- 16. Sun J, Huye LE, Lapteva N, Mamonkin M, Hiregange M, Ballard B, et al. globulin efficiently kill mature B lymphocyte-derived malignant cells. Early transduction produces highly functional chimeric antigen receptor- Blood 2006;108:3890–7. modified virus-specific T-cells with central memory markers: a Production 9. Loskog A, Giandomenico V, Rossig C, Pule M, Dotti G, Brenner MK. Assistant for Cell Therapy (PACT) translational application. J Immunother Addition of the CD28 signaling domain to chimeric T-cell receptors Cancer 2015;3:5.

OF10 Clin Cancer Res; 2019 Clinical Cancer Research

TCR Stimulation for Enhanced CAR-T–Cell Function

17. Lapteva N, Durett AG, Sun J, Rollins LA, Huye LL, Fang J, et al. Large-scale 24. Hudecek M, Sommermeyer D, Kosasih PL, Silva-Benedict A, Liu L, Rader C, ex vivo expansion and characterization of natural killer cells for clinical et al. The nonsignaling extracellular spacer domain of chimeric antigen applications. Cytotherapy 2012;14:1131–43. receptors is decisive for in vivo antitumor activity. Cancer Immunol Res 18. Suhoski MM, Golovina TN, Aqui NA, Tai VC, Varela-Rohena A, 2015;3:125–35. Milone MC, et al. Engineering artificial antigen-presenting cells to 25. Dudley ME, Yang JC, Sherry R, Hughes MS, Royal R, Kammula U, et al. express a diverse array of co-stimulatory molecules. Mol Ther 2007; Adoptive cell therapy for patients with metastatic melanoma: evaluation of 15:981–8. intensive myeloablative chemoradiation preparative regimens. J Clin 19. Gerdemann U, Katari UL, Papadopoulou A, Keirnan JM, Craddock JA, Oncol 2008;26:5233–9. Liu H, et al. Safety and clinical efficacy of rapidly-generated trivirus- 26. Heczey A, Louis CU, Savoldo B, Dakhova O, Durett A, Grilley B, et al. CAR T directed T cells as treatment for adenovirus, EBV, and CMV infections cells administered in combination with lymphodepletion and PD-1 inhi- after allogeneic hematopoietic stem cell transplant. Mol Ther 2013;21: bition to patients with neuroblastoma. Mol Ther 2017;25:2214–24. 2113–21. 27. Cruz CR, Hanley PJ, Liu H, Torrano V, Lin YF, Arce JA, et al. Adverse events 20. Wagner HJ, Cheng YC, Huls MH, Gee AP, Kuehnle I, Krance RA, et al. following infusion of T cells for adoptive immunotherapy: a 10-year Prompt versus preemptive intervention for EBV lymphoproliferative dis- experience. Cytotherapy 2010;12:743–9. ease. Blood 2004;103:3979–81. 28. Tzannou I, Papadopoulou A, Naik S, Leung K, Martinez CA, Ramos CA, 21. Yuan FF, Tanner J, Chan PK, Biffin S, Dyer WB, Geczy AF, et al. Influence of et al. Off-the-shelf virus-specific T cells to treat BK virus, human herpesvirus FcgammaRIIA and MBL polymorphisms on severe acute respiratory syn- 6, cytomegalovirus, epstein-barr virus, and adenovirus infections after drome. Tissue Antigens 2005;66:291–6. allogeneic hematopoietic stem-cell transplantation. J Clin Oncol 2017; 22. van Schie RC, Wilson ME. Saliva: a convenient source of DNA for 35:3547–57. analysis of bi-allelic polymorphisms of Fc gamma receptor IIA (CD32) 29. Yang Y, Kohler ME, Chien CD, Sauter CT, Jacoby E, Yan C, et al. TCR and Fc gamma receptor IIIB (CD16). J Immunol Methods 1997;208: engagement negatively affects CD8 but not CD4 CAR T cell expansion and 91–101. leukemic clearance. Sci Transl Med 2017;9:pii: eaag1209. 23. Calemma R, Ottaiano A, Trotta AM, Nasti G, Romano C, Napolitano M, 30. Omer B, Castillo PA, Tashiro H, Shum T, Huynh MTA, Cardenas M, et al. et al. Fc gamma receptor IIIa polymorphisms in advanced colorectal cancer Chimeric antigen receptor signaling domains differentially regulate pro- patients correlated with response to anti-EGFR antibodies and clinical liferation and native T cell receptor function in virus-specific T cells. outcome. J Transl Med 2012;10:232. Front Med 2018;5:343.

www.aacrjournals.org Clin Cancer Res; 2019 OF11

T-Cell Receptor Stimulation Enhances the Expansion and Function of CD19 Chimeric Antigen Receptor−Expressing T Cells

Natalia Lapteva, Margaret Gilbert, Iulia Diaconu, et al.

Clin Cancer Res Published OnlineFirst September 26, 2019.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-18-3199

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2019/09/26/1078-0432.CCR-18-3199.DC1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/early/2019/11/08/1078-0432.CCR-18-3199. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.