Interleukin-15-armored ALPPL2 CAR T cells demonstrate robust preclinical efficacy for solid tumors

Chimeric antigen receptor (CAR) T therapy has achieved significant success in hematologic malignancies. Its clinical efficacy in solid tumors remains limited due to several challenges, including the limited number of tumor-specific antigens and insufficient CAR T cell persistence. Armored CAR T cells, engineered to express cytokines, enzymes, or switch receptors, are one of the most active areas of CAR T research for solid tumors. They are developed to provide survival signals, block inhibitory effects, and enable the engineered T cells to function more effectively in the hostile tumor microenvironment.¹ Unlike conventional CAR T cells, armored CAR T cells have a modified functional state and, in many cases, can alter the surrounding tumor microenvironment, enhancing antitumor activity.

In our recent publication in Cancer Immunology Research, we developed a highly specific CAR composed exclusively of human sequences targeting alkaline phosphatase, placental-like 2 (ALPPL2). T cells expressing this receptor exhibited increased NK-like effector differentiation and sustained proliferative capacity when coexpressing interleukin-15 (IL-15), showing strong preclinical antitumor activity and laying the groundwork for future clinical trials.²

ALPPL2 was identified as a tumor-specific target, expressed in several solid tumors, including pancreatic ductal adenocarcinoma (PDAC), gastric cancer, ovarian cancer, mesothelioma, and yolk sac tumors (YST). However, its expression is restricted to only the placenta among healthy human tissues, rendering it an attractive immunotherapeutic target for cancer treatment.³ ALPPL2 was known to regulate naive pluripotency by interacting with IGF2BP1 to stabilize mRNA levels of TFCP2L1 and STAT3.⁴ However, its biological functions in tumors remain poorly elucidated. To target ALPPL2, earlier studies have developed therapeutic modalities like antibodies, aptamers, and antibody-drug conjugates (ADCs).^3,4,5,6 Here, we first reported IL-15 armored ALPPL2-targeted CAR T cells. In our study, we screened a panel of ALPPL2 antibodies by panning a human antibody-based phage-displayed scFv library, thereby reducing the potential risk of immunogenicity. Even though another highly homologous alkaline phosphatase family member, placental-type alkaline phosphatase (ALPP), and ALPPL2 differ by only 11 amino acids, the optimal antibody AP2 1-1 binds only to ALPPL2, not ALPP, indicating high specificity and minimal off-target toxicity. Furthermore, the AP2 1-1 scFv showed strong binding affinity for ALPPL2, with a measured K_D of 0.28 nM and an EC₅₀ of 0.13 nmol/L.

IL-15 is a vital cytokine that drives the proliferation, survival, and cytotoxicity of CD8⁺ T cells. Among the various CAR T-arming approaches, IL-15 has emerged as one of the most promising cytokines. Building on AP2 1-1, we developed the conventional second-generation CAR (ABBz) and a version with IL-15-armoring (15.ABBz). Both demonstrated notable cytotoxicity toward ALPPL2-positive tumor cells in vitro. Compared to ABBz, coexpression of IL-15 enriched for the CD8⁺ T cell subset and showed reduced effector cytokine production. After repeated exposure to tumor cells, 15.ABBz expressing T cells showed further enrichment of the CD8⁺ and a reduction in CD4⁺ T cell subset. There were no notable differences in exhaustion marker expression, effector or memory T cell subset composition between ABBz and 15.ABBz-expressing T cells. Importantly, compared to ABBz, 15.ABBz T cells produced lower amounts of effector cytokines and demonstrated greater cytotoxicity against ALPPL2-positive tumor cells.

In both in vitro repeat tumor cell exposures and in vivo experiments, 15.ABBz T cells demonstrated superior expansion and longer persistence. Finally, we also used two xenograft mouse models—one with PDAC and another with YST—to test the antitumor efficacy of ABBz and 15.ABBz T cells. As expected, 15.ABBz T cells induced superior and durable tumor regressions and provided a significant survival advantage in both models.

Overall, we developed a highly specific ALPPL2-targeting, fully human CAR that was co-expressed in T cells with IL-15, resulting in reduced T cell apoptosis, promoting NK-like effector differentiation with greater expansion capacity, and enhancing antitumor efficacy (Figure 1). Given that ALPPL2 is strictly expressed in multiple solid tumors, we believe these IL-15-armored CAR T cells could be a promising cancer therapy justifying further clinical development. These results are consistent with previous findings from our group on IL-15-armored glypican-3 (GPC3) CAR T cells.⁷ The outcomes of GPC3 CAR T cell clinical trials are expected to offer valuable insights to support the clinical development of ALPPL2 CAR T cell therapy. In patients, compared to GPC3 CAR T cells, IL-15-armored GPC3 CAR T cells exhibited significantly higher T cell expansion and antitumor activity. However, IL-15 coexpression also increased the risk of cytokine release syndrome (CRS) and required activation of inducible caspase-9 in 3/12 patients.⁸ Therefore, the potential systemic side effects caused by increased CAR T cell potency should be carefully evaluated in future clinical applications of IL-15-armored ALPPL2 CAR T cells.

Figure 1.

Schematic overview of IL-15-armored ALPPL2 CAR T cells, which demonstrate enhanced expansion and persistence, reduced apoptosis, and increased cytotoxicity

This figure was created using BioRender.

During the preparation of this manuscript, China’s National Medical Products Administration (NMPA) approved Satri-cel (satricabtagene autoleucel) for treating Claudin18.2-positive and HER2-negative advanced gastric/gastroesophageal junction adenocarcinoma in patients who have failed at least two prior lines of therapy, mainly supported by the phase 2 clinical study.⁹ This marks the world’s first approved CAR T cell therapy for solid tumors, making it arguably the most important translational milestone in developing solid tumor CAR T treatments. It also offers a valuable translational roadmap for our IL-15-armored ALPPL2 CAR T cells. Key lessons from this experience include modulating toxicity associated with higher-potency CAR T cells and selecting appropriate patients and diseases to maximize therapeutic benefit.

Acknowledgments

This manuscript was partially supported by the funding from the Cancer Prevention and Research Institute of Texas (RP190561 and RP250465 to Z.A. and N.Z.) and The Welch Foundation (AU-0042-20030616 to Z.A.).

Declaration of interests

J.L., N.Z., Z.A., and A.H. hold a patent for ALPPL2-targeting CAR T cells and related methods; Z.A. is a member of the Scientific Advisory Boards at ImmuneOnc Therapeutics and CrossBridge Bio, owns shares in these companies as well as Incendia Therapeutics, and receives funding from Merck Research Laboratories for his lab. A.H. worked as a consultant for Waypoint Bio and Kite BioPharma.