Vong, Judy

TPS2698 Background: Chimeric antigen receptor (CAR) T-cell therapies for treating solid tumors are challenging due to a lack of tumor-specific targets that discriminate cancer from normal cells. Previous studies using CEA T-cell receptors and T-cell engagers have resulted in dose-limiting, on-target, off-tumor toxicities (1,2). A2B530 is an autologous logic-gated, CEA-targeted Tmod CAR T-cell therapy that addresses the challenges of on-target, off-tumor toxicity by combining a CAR-activating receptor with a blocking receptor to discriminate tumor from normal cells (3). The activator recognizes CEA on the surface of both tumor and normal cells while specificity for tumor cells is provided by a blocker that binds HLA-A*02. In patients with both germline HLA-A*02 and tumor-associated HLA-A*02 LOH, the blocker prevents on-target, off-tumor toxicity on normal cells owing to retained HLA-A*02 expression (4). HLA-A*02 LOH can be detected using next-generation sequencing (Tempus AI, Inc.). With this definitive discriminator target, A2B530 can potentially provide a therapeutic window to treat patients with CEA-expressing solid tumors exhibiting HLA LOH. Methods: EVEREST-1 (NCT05736731) is a seamless, phase 1/2, open-label, nonrandomized study to evaluate the safety and efficacy of A2B530 in adult patients. Patients are enrolled through BASECAMP-1 (NCT04981119), a master prescreening study that identifies patients with HLA LOH at any time in the course of their disease. BASECAMP-1–eligible patients undergo leukapheresis and, when clinically appropriate, their banked T cells are manufactured for the EVEREST-1 study. The key inclusion criteria include histologically confirmed recurrent unresectable, locally advanced, or metastatic cancers associated with CEA expression: non-small cell lung, colorectal, or pancreatic cancers. Patients should have received ≥1 line of prior therapy such as checkpoint inhibitor, molecular targeted, or chemotherapy. The primary objective of phase 1 is to evaluate the safety and tolerability of A2B530 and determine the recommended phase 2 dose (RP2D). The dose-escalation portion is based on the Bayesian optimal interval design. The dose-expansion phase will confirm RP2D and collect biomarker data to further characterize A2B530. Phase 2 will assess overall response rate per RECIST v1.1. As of January 29, 2024, 8 patients have been enrolled on EVEREST-1. A2B530 was successfully manufactured for all patients, and all patients have received A2B530 infusion, with the first patient dosed in May 2023. Dose escalation is ongoing. 1. Parkhurst, et al. Mol Ther. 2022. 2. Tabernero, et al. J Clin Oncol. 2017. 3. Hamburger, et al. Mol Immunol. 2020. 4. Sandberg, et al. Sci Transl Med. 2022. Clinical trial information: NCT05736731 .

209 Background: Metastatic colorectal (CRC), pancreatic (PANC), and gastroesophageal cancers are the leading causes of GI cancer–related mortality (5-y survival: 15%, 3%, and 5%-6%, respectively) (ACS 2022). HLA LOH is a recurrent mechanism of immune escape observed in 15%-20% of GI cancers (Hecht R., ASCO GI 2022). The Tmod platform is a logic-gated chimeric antigen receptor (CAR) T-cell modular system, comprising a carcinoembryonic antigen (CEA)- or mesothelin (MSLN)-targeting CAR activator and a separate HLA-A*02-targeting blocker receptor. Both in vitro/in vivo, Tmod CAR T therapy kills cells with HLA-A*02 LOH (tumor) without harming cells with retained HLA-A*02 expression (normal). However, HLA-A*02 LOH can only be therapeutically exploited if patients are identifiable through a feasible and timely clinical workflow. Methods: We established a biobanking protocol (BASECAMP-1, NCT04981119) to determine whether HLA-A*02 LOH patients can be prospectively identified. Patients with CRC, PANC, or non-small cell lung cancer (NSCLC), and a high risk for incurable relapse, were screened first using a standard HLA assay. Heterozygous HLA-A*02 positive tumor samples were then assessed for LOH using a bioinformatic algorithm applied via the Tempus xT platform. Results: As of Sep 1, 2022, 83 patients were consented at 4 institutions. HLA status was obtained from 70 patients and 28 were identified as HLA-A*02:01 heterozygous (40%; expected frequency based on USA NMDP data, 27.6%). LOH results were available for 16 patients; 4 LOH-positive patients were identified (25%, 2 PANC, 2 NSCLC). The LOH assay sensitivity declines below a tumor purity of 40% (Hecht R., ASCO GI 2022). Six patients had a tumor purity of 20% (all with PANC, a tumor known for high stromal content), limiting possible LOH detection. The impact of tumor purity on LOH sensitivity was highlighted in a patient with a low initial sample tumor purity (30%) that resulted in a 41% probability of HLA-A*02:01 LOH (below positive threshold). A second sample with a higher tumor purity (70%), obtained from formalin-fixed, paraffin-embedded sections, resulted in a 92% probability of HLA-A*02:01 LOH (positive). Conclusions: BASECAMP-1 prospective identification of HLA-A*02 LOH is feasible in the real-world setting. The frequencies of the HLA-A*02 allele and of HLA-A*02 LOH in this cohort mirrored expected population frequencies. LOH results can be obtained within a clinically feasible workflow and timeframe, although samples with a < 40% tumor purity have a reduced sensitivity for LOH detection, an issue recurrently observed in patients with PANC. The BASECAMP-1 strategy enables prospective identification of appropriate patients for future therapeutic clinical trials using Tmod CEA and MSLN logic-gated CAR T cells. Clinical trial information: NCT04981119 .

190 Background: Metastatic colorectal (CRC), pancreatic (PANC), and gastroesophageal (GE) cancers are the leading causes of GI cancer–related mortality (5-yr survival rate, 14%, 3% and ̃5-6%, respectively). T-cell immunotherapy targeting GI-associated tumor antigens has been attempted, but efficacy has been constrained by on-target off-tumor toxicity, limiting the therapeutic window. The Tmod (TM) platform is an AND-NOT logic-gated CAR T modular system, versions of which have a CEA- or MSLN-targeting CAR activator and a separate HLA-A*02-targeting blocker receptor to protect normal cells. Tmod CAR T exploits HLA LOH, common in GI malignancies (10-33% in primary solid tumors [TCGA]) and can kill tumor cells without harming healthy cells in vitro and in vivo. However, the prevalence of HLA LOH across GI tumors is unknown in the real-world setting. We utilized the Tempus xT oncology NGS database of patients with multiple GI tumors. From a standard-of-care NGS assay, GI cancer patients can be readily identified for HLA LOH and future treatment with Tmod CAR T therapy. Methods: The occurrence of HLA LOH in GI tumors of 1439 patients was assessed using paired germline and somatic DNA sequencing using a research assay [6]. CRC, PANC and GE patients with ≥ stage 3 were then extracted, and rates of HLA LOH were identified (ie, whether loss occurred across high-frequency HLA-A alleles). In addition, mutations in KRAS and BRAF, as well as MSI status were stratified to determine any association with HLA-A LOH. Results: HLA-A LOH was detected in 830 (17.3%) of all solid tumor records, and a similar proportion when all GI cancer records were analyzed (17.0%). For GI subtypes, these values ranged from 13.5% to 23.1% (Table). No high-frequency HLA-A allele (A*01, A*02, A*03, A*11) was more likely to be lost. Clinical biomarkers ( KRAS, BRAF and MSI status) were not associated with HLA-LOH. Conclusions: The frequency of HLA LOH among advanced solid tumor cancers in this dataset is 17.3%, with a range of 13.5-23% between CRC, PANC and GE. The HLA LOH frequency observed in these GI tumors is consistent with that in primary tumors from TCGA, which also used germline-matched and tumor samples. Clinical biomarkers were not associated with HLA LOH. Tempus NGS was able to identify HLA LOH, which can be used for Tmod CAR T therapy to an enhanced therapeutic window. Identification of these patients in BASECAMP-1 (NCT04981119) will enable novel Tmod CAR T therapy. [Table: see text]