Olfactory Receptor OR2H1 Is an Effective Target for CAR T Cells in Human Epithelial Tumors

Malignant solid tumors represent a massive global health issue with growing incidence, high mortality rates, and exorbitant health care costs (1). Given many solid tumors are immunogenic (2–4), immunotherapies remain a promising approach. The field of immunotherapy has exhibited remarkable progress among epithelial tumors in recent years, demonstrating dramatic clinical responses in malignancies such as head and neck, urothelial, and lung cancers (5–7).

Chimeric antigen receptor (CAR) T cells are genetically engineered T cells that combine the tumor antigen recognition domain from an mAb with T-cell intracellular signaling domains to allow specific T-cell–mediated killing of tumor cells (8). Although CAR T cells are changing the management of hematologic malignancies (9–11), their effectiveness in solid tumors has been largely unsuccessful thus far. Tumor heterogeneity and strong immunosuppressive networks at solid tumor beds have been identified as barriers to using CAR T cells to treat solid tumors (12, 13). The main obstacle to translate the success of CAR T-cell therapy into solid tumors, however, remains the paucity of targets that are expressed on the tumor cell surface and not on normal tissues, which is crucial to mitigating potentially life-threatening on-target, off-tumor toxicity (14). Identifying accessible tumor antigens that are not expressed in vital organs would allow genetic engineering of the infusion product to prevent or ameliorate the inhibitory effect of the metabolic restrictions, immunosuppressive networks, and intrinsic cell stress pathways that render effector T cells dysfunctional at tumor beds.

Olfactory receptors (ORs) are G-protein–coupled receptors (GPCRs) with extracellular domains that interact with odorant molecules in the nose to initiate a neuronal response leading to the perception of smell (15). The transmembrane domain, along with the increased frequency of alteration in solid tumors of multiple histologies and very low or absent expression in normal tissues make these particularly interesting targets for CARs. Accordingly, we performed an exhaustive screening of more than 400 members of the family of olfactory receptors, focusing on those that have a limited pattern of expression in healthy tissues but are expressed, at least at the mRNA level, in multiple tumors. Among them, OR2H1 was the molecule with the most tumor-specific expression. Here, we report that the olfactory receptor OR2H1 is a safe and effective target for CAR T cells in a broad range of human carcinomas and restricted expression limited to testis among human tissues.

Human ovarian carcinoma tissues were procured under protocols approved by the Committee for the Protection of Human Subjects at Dartmouth-Hitchcock Medical Center (Lebanon, NH; No. 17702) and under a protocol approved by the Institutional Review Board at Christiana Care Health System (Newark, DE; No. 32214) and H. Lee Moffitt Cancer Center (Tampa, FL; MCCNo. 18974). Lung cancer tumors were obtained through a fresh tumor tissue collection protocol at H. Lee Moffitt Cancer Center (MCCNo. 20023). Breast cancer specimens were obtained under a protocol approved by the Institutional Review Board (IRB) of the University of Pennsylvania (Philadelphia, PA; No. 805139) and the IRB of The Wistar Institute (Philadelphia, PA; No. 21204259). Written informed consent was obtained from all subjects.

mRNA sequencing data files related to epithelial tumors of multiple histologies were downloaded from the The Cancer Genome Atlas (TCGA) data portal (2016). Downloaded files included RNA sequencing by Expectation-Maximization data, and values were expressed as transcripts per million mapped reads.

Tissue microarrays (TMAs) were obtained from US BioMax, Inc. (Ov401sur, GA802a, LC953; Derwood, MD) containing 16 cases of high-grade serous ovarian cancer (HGSOC), four endometrioid ovarian adenocarcinomas, two mucinous ovarian adenocarcinoma, four adult granulosa cell tumors, 14 cases of ovarian adenocarcinoma, 28 squamous cell carcinomas of the lung, 26 lung adenocarcinomas, two lung adenosquamous carcinomas, two bronchioalveolar lung carcinomas, two large-cell lung carcinomas, two small-cell lung carcinomas, 48 cases of extrahepatic biliary duct adenocarcinoma, and 27 intrahepatic biliary duct adenocarcinomas. An additional TMA was obtained from Novus Biologicals (NBP2–30232) containing 58 normal human tissues.

Human HGSOC cell line OVCAR3 (NCI-DTP catalog No. OVCAR-3, RRID:CVCL_0465) was obtained originally from ATCC and cultured in our lab in RPMI-1640 (Sigma-Aldrich) media supplemented with 10% FBS, penicillin (100 IU/mL), streptomycin (100 μg/mL), L-glutamine (2 mmol/L), and sodium pyruvate (0.5 mmol/L; Thermo Fisher Scientific) without changes in expression of OR2H1, proliferation rates, or morphology. Human non–small cell lung carcinoma (NSCLC) cell line H2009 (ATCC catalog No. CRL-5911, RRID:CVCL_1514) was procured from the ATCC and passaged fewer than 10 times in our lab. Cell lines tested negative for Mycoplasma contamination before injection in mice each time. No further cell line authentication was performed.

To quantify OR2H1 expression, we used a panel of normal tissue RNA from Clontech and RNA extracted from human ovarian, lung, and breast cancer tissues from our established tumor bank. Total mRNA were reverse transcribed using the SuperScript IV First-Strand Synthesis System (Invitrogen catalog No. 18090010). Quantification of human OR2H1 was performed on the 7900HT Fast Real-Time PCR system (Thermo Fisher Scientific) using the SYBR Select Master Mix (Applied Biosciences catalog No. 4472897) and primers (forward: 5′-TCACTCAGTACAGCTCCCATGC-3′; and reverse: 5′-TTCAGTTCTTGCAATTAAGTCAGACTCT-3′). Expression was normalized to levels of the endogenous reference control gene GAPDH using primers (forward: 5′-CCTGCACCACCAACTGCTTA-3′; and reverse: 5′-AGTGATGGCATGGACTGTGGT-3′).

Frozen human tumors and cells lines were cultured and lysed in RIPA buffer (Thermo Fisher Scientific) with a protease-phosphatase inhibitor cocktail (Sigma-Aldrich), followed by protein extraction and denaturation. Proteins were quantified by bicinchoninic acid assay (Thermo Fisher Scientific). Western blotting was performed using the NuPage system. Membranes were blotted with anti-OR2H1 (Invitrogen PA5–113405, RRID:AB_2868138), followed by anti-rabbit IgG-HRP (Abcam catalog No. ab6759, RRID:AB_955434), and reblotted with anti–β-actin horseradish peroxidase (HRP; Cell Signaling Technology catalog No. 5125, RRID:AB_1903890). Images were captured using BioRad ChemiDoc imaging system.

An OR2H1 antibody corresponding to the CAR single-chain variable fragment (scFv) was generated by cloning the variable heavy and variable light chains of the scFv into pcDNA3.4 and expressed in Chinese hamster ovary cells, followed by purification of the IgG1 (outsourced to Genscript).

To initially optimize the staining, immunocytochemistry was performed by first placing 10⁵ OVCAR3 (NCI-DTP catalog No. OVCAR-3, RRID:CVCL_0465), H2009 wild-type (WT; ATCC catalog No. CRL-5911, RRID:CVCL_1514), or H2009-OR2H1 knockout (KO) cells onto a coverslip within 6-well plates overnight. The cells were then fixed with chilled methanol, washed with PBS, and blocked with peroxidase block and protein blocking solution (BioVision K405–50) followed by incubation at 4˚C overnight with OR2H1 IgG containing the same scFv to be used in the OR2H1 CAR at various ratios (1:25, 1:50, 1:100, and 1:200). After washing, the cells were then stained with goat anti-human IgG conjugated to HRP (Abcam ab6858, RRID:AB_955433) secondary antibody for 2 hours at room temperature, followed by completion of IHC procedure according to the manufacturer’s instructions (BioVision K405–50). The slides were developed using the DAB chromogen and hematoxylin. Negative control slides were included that used the secondary antibody only. Coverslips were mounted on microscope slides with ProLong gold antifade reagent (Invitrogen P36934) and imaged.

HGSOC slides from our tumor bank and TMAs were then deparaffinized and subjected to antigen retrieval, followed by IHC staining as above according to manufacturer instructions (BioVision K405–50) with OR2H1 IgG. Slides were mounted as above and imaged on the BOND RX (Leica Biosystems).

In order to estimate the number of copies of OR2H1 in target cells, we conjugated OR2H1 IgG to PE using the PE/R-Phycoerythrein Conjugation Kit- Lightning-Link (Abcam 102918) according to the manufacturer's instructions. We stained H2009-WT and H2009-OR2H1 KO cells with the OR2H1 IgG-PE antibody and measured the median fluorescence index on the BD FACS Canto flow cytometer compared with anti-human IgG isotype control conjugated to PE. We then ran a BD Quantibrite PE tube (BD Biosciences catalog No. 340495, RRID:AB_2868736) using the same instrument settings and used the resulting standard curve to estimate the number of OR2H1-IgG antibodies bound per cell according to the manufacturer's instructions.

To determine the sequence of an antibody fragment recognizing the extracellular domain of OR2H1, ELISA-based screening and validation of a phage display library of 270 healthy human donors’ samples was performed against peptide Ag (HQQIDDFLCEV-Cys), followed by Phage DNA extraction and antibody sequencing (outsourced to Proteogenix).

We designed the CAR construct using the olfactory receptor signal peptide, followed by the OR2H1 scFv, linked by a glycine/serine spacer, followed by CD8α hinge and transmembrane domains, and intracellular fragments of 4–1BB and CD3ζ (Supplemental Figure 1). We ordered the constructs from Genscript flanked by EcoRI and NotI and cloned into the pBMN-I-GFP (RRID:Addgene_1736) retroviral vector.

We generated retrovirus by transfecting 293GP cells (RRID:CVCL_E072) with either pBMN-I-GFP (RRID:Addgene_1736) or pBMNI-GFP-OR2H1 CAR, similar to previously described methods (16). Briefly, we plated the cells in a 10-cm culture dish. When the cells reached 70% to 80% confluency, we transfected them with a mix of our DNA and the RD114 envelope vector (catalog No. 17576, RRID:Addgene_17576) using Lipofectamine 3000 (Invitrogen catalog No. L3000015). We collected the supernatant containing the retroviral particles at 48, 72, and 96 hours after transfection.

To prepare T cells for transduction, we performed red blood cells lysis on healthy human apheresis samples, followed by activation of T cells in AIM-V, 5% FBS, 1% penicillin/streptomycin, 1% L-glutamine, 300 U/mL IL2 (PeproTech catalog No. 200–02–50), and 0.05 μg/mL purified anti-human CD3 (OKT3) antibody (Tonbo Biosciences catalog No. 70–0037, RRID:AB_2621474) at a concentration of 10⁶ cells/mL. We performed three spin infections at 48, 72, and 96 hours on Retronectin-coated plates (Takara catalog No. T100B). We removed and added media containing 300 U/mL IL2 (PeproTech catalog No. 200–02–50) to maintain a concentration of 10⁶ cells/mL. Transduction efficiency and T-cell populations were measured with a BD LSRII flow cytometer.

We plated 10⁵ target cells per well in a flat-bottom 96-well plate in 100-μL culture media. The following day, we transfected the target cells with a luciferase-expressing vector [Promega pGL4.51(luc2/CMV/Neo); RRID:Addgene_132962] using the JetPrime system (Polyplus catalog No. 101000015) for tumor cells and the JetOptimus system (Polyplus catalog No. 101000051) for primary healthy cells. The media was removed and replaced with fresh culture media 6 to 8 hours after transfection. The appropriate number of either OR2H1 CAR or mock-transduced T cells were added the following day and cocultured for 8 hours in the case of tumor cells and 24 hours for primary cultures (adipocytes, hepatocytes, and neurons). Cytotoxicity was measured via Luciferase Assay (Promega catalog No. E1501). Cytotoxicity was calculated as (maximum viability control – individual well)/(maximum viability control − maximum death control) × 100 as a percentage.

In parallel, either H2009 (ATCC catalog No. CRL-5911, RRID:CVCL_1514) or OVCAR3 (NCI-DTP catalog No. OVCAR-3, RRID:CVCL_0465) cells were cocultured with either OR2H1 CAR or mock-transduced T cells at a ratio of 1:5. After coculture of 8 hours, the supernatant was harvested and used for IFNγ detection using ELISA MAX Deluxe Set Human IFNγ (Biolegend 430101), and absorbance was measured at 450 nm with a Gen5 Microplate reader and Image Software (Biotek).

NOD/SCID gamma (NSG) mice, originally from Jackson labs, were maintained by the H. Lee Moffitt Cancer Center animal facility. Animal experiments were approved by the Institutional Animal Care and Use Committee at the University of South Florida (Tampa, FL).

Five million H2009 (ATCC catalog No. CRL-5911, RRID:CVCL_1514) or OVCAR3 (NCI-DTP catalog No. OVCAR-3, RRID:CVCL_0465) were implanted subcutaneously in the right flank of NSG mice, followed by retro-orbital injection of either 5 × 10⁶ OR2H1 CAR or mock-transduced T cells on day 5 or days 9 and 14. Flank tumors were measured with calipers and tumor volumes calculated as length × width × height.

CRISPR RNA (crRNA) targeting OR2H1 (GACAGCCACGTATCGGTCAA; IDT) was reconstituted in 100-μmol/L Nuclease-Free Duplex Buffer (IDT). The crRNA was then mixed with Alt-R CRISPR-Cas9 tracrRNA, ATTO 550 (IDT catalog No. 1075927) in a sterile PCR tube. crRNA:tracrRNA duplexes were annealed by heating at 95°C for 5 minutes, then slowly cooled to room temperature. Nine microliters of crRNA-tracrRNA duplexes was mixed with 180 pmol of TrueCut Cas9 protein (Thermo Fisher Scientific, A36498), followed by incubation at room temperature for 10 minutes to form Cas9 ribonucleoproteins. To ablate OR2H1 from H2009 tumor cells, 2 × 10⁶ cells were resuspended in 100-μL buffer R (Neon Transfection System; Thermo Fisher Scientific), and 15 μL of the Cas9 RNPs were added to the resuspended cells. Electroporation was performed at 1,230 V, 30 milliseconds, and two pulses. H2009 cells (ATCC catalog No. CRL-5911, RRID:CVCL_1514) were cultured [RPMI-1640, 10% FBS, penicillin (100 IU mL−1), streptomycin (100 μg mL−1), l-glutamine (2 mmol/L), sodium pyruvate (0.5 mmol/L); Thermo Fisher Scientific], and OR2H1 ablation was confirmed via Western blotting.

We assessed a potential mechanism by which OR2H1 enhances glucose metabolism using the fluorescent glucose analog 2-(N-(7-Nitrobenz-2-oxa-1,3-diazol-4-yl)Amino)-2-Deoxyglucose (2-NBDG; Invitrogen N13195). We diluted 2-NBDG in glucose-free R10 at a concentration of 200 μmol/L and incubated with H2009-WT (ATCC catalog No. CRL-5911, RRID:CVCL_1514) and H2009-OR2H1 KO cells for 30 minutes at 37°C and 5% CO₂. We then measured the signal in the FITC channel via the BD LSRII flow cytometer to evaluate glucose uptake in the wild-type cells compared with the KO.

RNAscope 2.5 LSx Red ISH was used to evaluate OR2H1 RNA expression in ACD prequalified human mutinormal TMA and human multitumor (TMA) using the target probe and images were collected. This was outsourced to ACD, a biotechne brand.

All experiments were repeated at least twice with similar results. Differences between the means of experimental groups were calculated using a two-tailed unpaired Student t test or two-way ANOVA with multiple comparisons. Error bars represent the standard error of the mean. All statistical analyses were done using GraphPad Prism 9.0 (GraphPad Prism, RRID:SCR_002798). P < 0.05 was considered statistically significant.

The data generated in this study are available within the article and its Supplementary Data Files.

To understand the relevance of ORs as immunotherapeutic targets in solid tumors, we first examined the TCGA dataset and cBioPortal (17) to determine the pattern of expression of all members of the family in a variety of malignancies and healthy human tissues. We found that OR2H1 is overexpressed in a variety of solid epithelial tumors, ranging from ∼4% of colon cancer to 69% of intrahepatic cholangiocarcinoma (Fig. 1A). Common tumors with frequent expression included prostate cancer (>38%) and serous endometrial carcinoma (∼27%), implying potential applicability to a multitude of patients. Further supporting the therapeutic potential of OR2H1 CAR T cells, we confirmed the pattern of expression in normal human tissues via real-time QPCR is limited to the testis (Fig. 1B), in agreement with the GTEX Portal and The Human Protein Atlas (Supplementary Fig. S1A). RNAscope was also performed to further assess the expression pattern of OR2H1. Using testis and ileum as positive and negative controls, respectively, we identified expression in multiple tumor types including pancreatic and ovarian carcinomas (Fig. 1C).

We additionally evaluated OR2H1 expression in a variety of epithelial tumors via immunohistochemistry. For that purpose, we used a recombinant IgG1 generated with the variable heavy/light chain sequences of a human OR2H1-reactive scFv, identified through the ELISA-based screening of a phage display library including samples from 270 healthy human donors against the HQQIDDFLCEV antigen, contained in the predicted extracellular domain of OR2H1 (Fig. 2A). Using this human antibody, we found specific OR2H1 signal in 20% of ovarian carcinomas (8/40), 13% of lung carcinomas (8/60) and 59% of cholangiocarcinomas (43/73). Importantly, no expression of OR2H1 was found in 58 healthy human tissues (Fig. 2B).

Widespread expression of OR2H1 was confirmed by quantitative real-time PCR analysis of eight NSCLC, 17 breast cancers, 53 high-grade serous ovarian cancers (HGSOCs), and 29 non–high-grade serous ovarian cancers from our tumor bank (Fig. 3A–D). OR2H1 expression in normal ovary adjacent to OR2H1-expressing ovarian carcinoma was confirmed to be negative via RT-qPCR (Fig. 3E). Expression of OR2H1 was confirmed in H2009 and OVCAR3 cells lines (Fig. 3A and C), with an estimated 18,843 copies of OR2H1 surface protein per H2009 cell (Supplementary Fig. S1B), which is above the threshold target antigen density required to induce CAR T-cell cell responses and cytokine production (18). Protein expression was additionally confirmed in a variety of histological subtypes via Western blotting, with good correlation between RNA and protein expression (Fig. 3F). Therefore, OR2H1 is a potential target in a multitude of solid, epithelial tumors likely with a range of admissible toxicities, particularly in the context of terminal cancer.

To target OR2H1 by redirecting T cells, we generated a human CAR targeted with the same OR2H1-reactive variable heavy/light sequences used to generate our recombinant IgG1 [CDR3 sequences TRGPLL (VH) and AAWDDSVRGPV (VL)], linked by a glycine/serine spacer, in frame with CD8 hinge and transmembrane domains, along with the intracellular domains of costimulatory 4–1BB and CD3ζ (Fig. 4A). T cells were retrovirally transduced with a green fluorescent protein (GFP)-containing vector and demonstrated transduction efficiencies of greater than 90% in all experiments (Fig. 4B). Importantly, there was no cytotoxic killing by OR2H1 CAR T cells of healthy primary human adipocytes, hepatocytes, or neurons (Fig. 4C; Supplementary Fig. S1C). In contrast, OR2H1 CAR T cells showed strong cytotoxic activity against OR2H1⁺ lung cancer H2009 cells, with nearly 90% specific killing in a dose-dependent manner, compared with mock-transduced T cells (Fig. 4D). Equally important, the cytotoxic effect of OR2H1 CAR T cells was lost following CRISPR/Cas9-mediated ablation of OR2H1 in H2009 tumor cells (Fig. 4D), supporting specific OR2H1 killing and no off-target effects. Accordingly, a single injection of OR2H1 CAR T cells in established (day 5) H2009 tumor-bearing mice abrogated tumor growth, while mock-transduced T cells were unable to prevent accelerated malignant progression (Fig. 4E–G). When tumors were allowed to progress for 9 days, two injections of either OR2H1 CAR or mock-transduced T cells on days 9 and 14 produced results comparable with a single injection at earlier temporal points (Fig. 4H–J). Tumors treated with OR2H1 CAR T cells not only showed reduced tumor volumes and weights but also significant central necrosis compared with those treated with mock-transduced T cells (Fig. 4K), with specific CAR T-cell IFNγ production (Fig. 4L).

To demonstrate the therapeutic potential of OR2H1 CAR T cells in tumors of various histologies with a relatively low level of OR2H1 expression, we repeated these experiments using OVCAR3 ovarian cancer cells, which have OR2H1 expression below the median of positive HGSOCs (Fig. 3C). OR2H1 CAR T cells again effectively killed OR2H1^low OVCAR3 cells in vitro in a dose-dependent manner, while mock-transduced T cells induced negligible effects (Fig. 5A). Compared with mock-transduced T cells, a single injection of OR2H1 CAR T cells was again able to delay the progression of OVCAR3 tumors already established in immunodeficient mice (Fig. 5B–D), with significant histological alterations (Fig. 5E), in addition to reduced tumor volume and weight and IFNγ production (Fig. 5F). We additionally demonstrate applicability of OR2H1 CAR T cells to cisplatin-resistant ovarian cancer using A2780-derived CP70 cells (Supplementary Fig. S1D). As expected, cytotoxic killing relative to OVCAR3 correlates with levels of mRNA expression (Supplementary Fig. S1E). These data underscore the potential of treating carcinomas of different histologic origins and variable levels of OR2H1 expression, using OR2H1-directed CAR T cells.

Another important barrier for the implementation of CAR T cells against solid tumors is rapid immunoediting, resulting in antigen loss in response to treatment (19). To elucidate the possible cost of losing OR2H1 in terms of malignant progression, we ablated OR2H1 in H2009 tumor cells using a CRISPR/Cas9 system. As shown in Fig. 4C, OR2H1 deletion completely eliminated the cytotoxic effect of OR2H1 CAR T cells in vitro, further supporting specific on-target killing. As expected, abrogated malignant progression upon OR2H1 CAR T-cell administration completely disappeared in the absence of OR2H1 expression in vivo (Fig. 6A–C). Most importantly, OR2H1-ablated tumors grew significantly more slowly in vivo, compared with the isogenic cell line (Fig. 6D and E). This suggests that even if OR2H1 expression was lost in patients with cancer under immune pressure, this could still result in a significant therapeutic benefit. Accordingly, OR2H1⁺ H2009 (WT) cells showed enhanced uptake of the fluorescent glucose analog 2-NBDG, compared with OR2H1-ablated isogenic cells, suggestive of OR2H1 involvement in glucose metabolism (Fig. 6G–H).

We showed that the olfactory receptor OR2H1 is an effective immunotherapeutic target for CAR T cells in a variety of solid tumors with variable levels of OR2H1, mediating significant abrogation of malignant growth, likely with a limited range of admissible toxicities. Accordingly, T cells redirected against OR2H1 delayed growth of NSCLC (H2009) and HGSOC (OVCAR3) both in vitro and in vivo. This work additionally shows the applicability of this therapy to a wide variety of patients, given the expression of OR2H1 in a subset of solid tumors across multiple histologies, including HGSOC, lung carcinoma, cholangiocarcinoma, breast carcinoma, and ovarian cancers of multiple other histologies (e.g., clear cell, endometrioid, mucinous, granulosa cell, and low-grade serous). Importantly, we did not detect expression of OR2H1 in any healthy tissue, with the exception of testis. Given chemical castration is routine in the treatment of prostate cancer, loss of androgen signaling in response to treatment with OR2H1 CAR T cells would likely be an acceptable outcome, and even a therapeutic outcome in the context of prostate cancer. Theoretically, OR2H1 could be also expressed in the nasal epithelium to some degree, although this tissue was not included in the panel of normal tissues studied. However, any toxicities related to olfaction as a result of targeting the nasal epithelium may be acceptable in the context of terminal cancer. Accordingly, OR2H1 CAR T cells did not induce any cytotoxic activity in primary cultures of normal adipocytes, hepatocytes, or neurons. Therefore, OR2H1, and possibly other olfactory receptors with limited patterns of expression in other tumors, represent promising targets for translation of the success of CAR T cells from hematologic malignancies to solid tumors.

Other studies have suggested ORs as potential therapeutic targets in various malignancies, such as OR51E2 in prostate cancer (20–23), OR51B4 in colorectal cancer (24), OR10H1 in bladder cancer (25), OR2B6 (26) and OR2W3 (27) in breast cancer, and OR2C3 in melanoma (28). Although overexpression of multiple ORs has been shown in various epithelial malignancies (25–29), reports of targeting ORs are limited and typically describe ORs as tumor antigens recognized by CD8 T cells, which can be used as targets for anticancer vaccines (29). To our knowledge, this is the first report of targeting an olfactory receptor for the treatment of malignancy.

In addition to the specificity of our OR2H1 CAR T cells shown by loss of cytotoxic activity with OR2H1 ablation, we demonstrated delayed malignant progression associated with OR2H1 ablation. Other ectopic ORs (ORs outside the nasal epithelium) are regulators of cellular glucose metabolism (30, 31) and glucose stimulated insulin secretion in pancreatic β cells (32). Similarly, we found enhanced 2-NBDG uptake in H2009-WT cells compared with H2009-OR2H1 ablated isogenic cells, indicating OR2H1 involvement in glucose metabolism. Impaired tumor growth following OR2H1 ablation further supports therapeutic benefit of targeting this receptor.

Overall, our study characterizes the expression pattern of OR2H1 in epithelial tumors of multiple histologies and demonstrates the therapeutic efficacy of targeting OR2H1-expressing epithelial tumors with CAR T cells redirected against OR2H1.

A.L. Martin reports grants from H. Lee Moffitt Cancer Center during the conduct of the study and has a patent for “Olfactory receptors for use as targets for antigen binding molecules to detect and treat cancer,” pending to H. Lee Moffitt Cancer Center. C.M. Harro reports grants from H. Lee Moffitt Cancer Center during the conduct of the study. B.A. Perez reports grants and personal fees from Bristol Myers Squibb; personal fees from AstraZeneca; and personal fees from G1 Therapeutics outside the submitted work. R.M. Wenham reports grants and personal fees from Merck and Ovation Diagnostics; personal fees from Genentech GSK/Tesaro, Clovis, AstraZeneca, AbbVie, Legend Biotech, Regeneron, Seagen, Sonnet Biotherapeutics, Shattuck Labs, Novocure, and Eisai; and personal fees from Immunogen outside the submitted work. J.R. Conejo-Garcia reports grants from H. Lee Moffitt Cancer Center Foundation during the conduct of the study; personal fees from Alloy Therapeutics; grants, personal fees, and other support from Anixa Biosciences; and other support from Compass Therapeutics outside the submitted work; in addition, J.R. Conejo-Garcia has a patent for “Compass Therapeutics,” pending to The Wistar Institute and a patent for Anixa Bioscience issued, licensed, and with royalties paid from The Wistar Institute. No disclosures were reported by the other authors.

A.L. Martin: Data curation, formal analysis, validation, investigation, visualization, methodology, writing–original draft. C.M. Anadon: Data curation, investigation, visualization, methodology, writing–review and editing. S. Biswas: Investigation, visualization, methodology, writing–review and editing. J.A. Mine: Investigation, writing–review and editing. K.F. Handley: Investigation, visualization, writing–review and editing. K.K. Payne: Investigation, visualization, methodology, writing–review and editing. G. Mandal: Investigation, methodology, writing–review and editing. R.A. Chaurio: Investigation, methodology, writing–review and editing. J.J. Powers: Investigation, methodology, writing–review and editing. K.B. Sprenger: Investigation, visualization, methodology, writing–review and editing. K.E. Rigolizzo: Investigation, writing–review and editing. P. Innamarato: Investigation, writing–review and editing. C.M. Harro: Investigation, writing–review and editing. S. Mehta: Investigation, writing–review and editing. B.A. Perez: Investigation, writing–review and editing. R.M. Wenham: Resources, data curation, supervision, funding acquisition, writing–review and editing. J.R. Conejo-Garcia: Conceptualization, resources, data curation, supervision, funding acquisition, visualization, methodology, project administration, writing–review and editing.

Support for Shared Resources was provided by Cancer Center Support Grant (CCSG) CA076292 to H. Lee Moffitt Cancer Center. This study was supported by R01CA157664, R01CA124515, R01CA178687, and R01CA211913 to J.R. Conejo-Garcia. A.L. Martin was supported by Moffitt Foundation. S. Biswas was supported by 1K99CA266947–01. S. Biswas and K.F. Handley were supported by Junior Scientist Research Partnership Award, H. Lee Moffitt Cancer Center. K.K. Payne was supported by T32CA009140 and The American Cancer Society Postdoctoral Fellowship.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.