The centerpiece of the field remains the chimeric antigen receptor T cell, or CAR-T cell. In this approach, T cells are extracted from a patient’s blood and fitted with an artificial receptor that combines an antibody-like recognition domain with intracellular signaling machinery, allowing the cells to latch onto a chosen surface antigen on tumor cells and kill them without the usual constraints of natural antigen presentation. The clinical results in B-cell malignancies, where CD19 serves as a near-universal target, have been remarkable enough to earn regulatory approval in multiple countries. Yet the review emphasizes that this success has exposed the technology’s limits: patients can relapse when tumor cells stop displaying the target antigen, a phenomenon known as antigen escape, and the engineered cells themselves can exhaust, lose persistence, or trigger dangerous immune reactions.

To counter antigen escape, researchers are developing multi-targeted receptor designs. Dual-targeting constructs that recognize two antigens simultaneously, such as the CD19 and CD22 combination tested in B-cell acute lymphoblastic leukemia, are designed to prevent tumor cells from simply shedding one molecular flag and surviving. The review also highlights logic-gated receptor architectures, in which a cell must display a specific combination of antigens before the therapeutic T cell fires, reducing the risk of attacking healthy tissue that carries the target on its surface. These designs borrow concepts from synthetic biology, treating the engineered cell as a programmable unit whose activation rules can be written into its genome.

Gene editing has become the second pillar of next-generation engineering. Early CAR-T products were made with viral vectors that inserted the receptor gene at random locations in the genome, raising long-standing safety questions about insertional mutagenesis. The review describes how CRISPR-based systems, adenine base editors, and prime editing now allow precise, non-viral integration of therapeutic sequences into defined genomic safe harbors, or even targeted rewriting of endogenous genes so that tumor-restricted payloads are delivered only where they are needed. Quadruple base-edited allogeneic CAR T cells have been reported to outperform cells engineered with conventional CRISPR nucleases, and base-edited CAR7 T cells have shown promise in relapsed T-cell acute lymphoblastic leukemia, a setting in which the patient’s own T cells cannot easily be used because the tumor and the therapy would attack one another.

Allogeneic, or off-the-shelf, products represent one of the most consequential shifts in the field. Autologous manufacturing, in which each patient’s cells are individually engineered, is slow, expensive, and sometimes impossible in patients whose disease or prior chemotherapy has depleted their lymphocytes. Donor-derived or stem-cell-derived products could in principle be manufactured in large batches and infused immediately, but they carry the risk of graft-versus-host disease, in which the transferred cells attack the recipient’s healthy tissues. The review notes that natural killer cells have emerged as a particularly attractive platform here, because they naturally destroy stressed and malignant cells, carry a lower incidence of graft-versus-host disease, and can be generated as readily available allogeneic therapeutic products. CAR-NK cells, engineered to express tumor-targeting receptors while retaining their innate killing machinery, are now being tested across hematologic malignancies and solid tumors alike.

An even newer branch of the family tree is the CAR-NKT cell, derived from natural killer T cells that combine features of both innate and adaptive immunity. Recent preclinical and early clinical work summarized in the review describes allogeneic CD70-directed CAR-NKT cells for metastatic renal cell carcinoma, mesothelin-specific CAR-NKT cells generated from cord-blood hematopoietic stem and progenitor cells for triple-negative breast cancer, and CD33-directed CAR-NKT cells for bone marrow-resident myeloid malignancies. A phase 1 trial of allogeneic iPSC-derived iNKT cells in recurrent head and neck cancer has also been reported, pointing toward induced pluripotent stem cells as a renewable, standardized starting material for entire product lines of immune cells. In parallel, large-scale protocols for generating induced natural killer cells and CAR-iNK cells from CD34-positive hematopoietic stem and progenitor cells suggest that manufacturing capacity, rather than biology, may become the field’s decisive bottleneck.

Solid tumors remain the hardest problem, and the review devotes substantial attention to why. Unlike the bloodstream, where CAR-T cells can circulate freely and encounter their targets, solid tumors are wrapped in a physically dense and chemically hostile microenvironment. Abnormal vasculature impedes cell trafficking, immunosuppressive metabolites such as kynurenine dampen T-cell function, and regulatory immune cells actively suppress the attack. Engineering responses include nanobody-based receptors that recognize microenvironment markers rather than tumor cells themselves, kynureninase-modified CAR-T cells that metabolize the immunosuppressive amino acid kynurenine, and manganese-oxide nanoenzyme-armed CAR-NK cells designed to alleviate the immunosuppressive milieu. Tumor-infiltrating lymphocytes, harvested from surgically removed tumors and expanded ex vivo, offer a complementary strategy because they arrive pre-selected for tumor reactivity, and they have shown durable benefit particularly in melanoma. T-cell receptor engineered T cells, meanwhile, can target intracellular antigens presented on HLA molecules, a class of targets fundamentally invisible to CARs, with a recent phase 1 trial of autologous T cells directed at the PRAME antigen in HLA-A*02-positive patients with advanced solid tumors.

Safety engineering has matured alongside efficacy engineering. Cytokine release syndrome and neurotoxicity remain the signature toxicities of potent CAR-T products, and the review discusses management strategies ranging from early corticosteroid use, which has been associated with reduced risk of high-grade cytokine release syndrome without compromising treatment outcomes, to inducible caspase 9 suicide switches that can be triggered to eliminate the infused cells if toxicity escalates. A more sobering concern is the emerging evidence on second primary malignancies after CAR-T therapy, analyzed through global pharmacovigilance databases and a meta-analysis of more than 5,500 lymphoma and myeloma patients, alongside reports of CAR-positive T-cell lymphoma. These findings have intensified interest in site-specific transgene integration into genomic safe harbors and in rigorous long-term surveillance of treated patients, framing durability and safety as inseparable design goals rather than competing priorities.

Manufacturing, the review argues, is where many of the field’s promises will ultimately be won or lost. Conventional autologous production relies on viral transduction and open-handling steps that demand centralized facilities, weeks of turnaround time, and costs that can exceed hundreds of thousands of dollars per patient. The response has been threefold: automated, closed-system bioreactors that reduce contamination risk and operator variability; decentralized and point-of-care manufacturing that brings production closer to the patient, with regulatory pathways now being delineated in the United States and abroad; and non-viral engineering methods such as Sleeping Beauty transposition and orthogonal CRISPR systems that cut cost and simplify logistics. Academic consortia are working to harmonize quality-control testing across centers, and cost-effectiveness studies are exploring how standardized processes could widen access beyond wealthy health systems, including in resource-limited regions where the current price of cell therapy is simply prohibitive.

The review closes by sketching a future in which adoptive cell therapy becomes a genuinely modular component of precision cancer immunotherapy rather than a standalone intervention. Combination regimens pairing engineered cells with checkpoint inhibitors, oncolytic viruses that can expand dual-specific CAR T cells in vivo, and biomarker-driven patient selection are expected to define the next wave of clinical trials. Preclinical tools are evolving in step, with tumor organoid-immune co-culture systems and humanized mouse models enabling researchers to test candidate cell products against patient-specific tumor tissue before infusion. Single-cell T-cell receptor sequencing of diagnostic biopsies is feeding high-throughput discovery pipelines that can convert a patient’s own tumor into a source of therapeutic receptors. None of these advances removes the fundamental challenges of antigen heterogeneity, immune resistance, toxicity, and scale, but together, the authors conclude, they indicate that engineered living cells are steadily advancing from a specialized treatment for a handful of blood cancers toward a central pillar of precision oncology.

Subject of Research: Engineering innovations, challenges, and manufacturing perspectives in adoptive cell therapy for cancer

Article Title: Emerging frontiers in adoptive cell therapies: engineering innovations, current challenges, and manufacturing perspectives

Article References: Hosseini Nezhad, M. S., Zeighami Gol, S., Gharavi, P., & Javid, H. (2026). Emerging frontiers in adoptive cell therapies: engineering innovations, current challenges, and manufacturing perspectives. Molecular Biology Reports, 53(1), Article 1554. https://doi.org/10.1007/s11033-026-12720-y

Image Credits: AI Generated

DOI: 10.1007/s11033-026-12720-y

Keywords: adoptive cell therapy, CAR-T cells, CAR-NK cells, tumor-infiltrating lymphocytes, TCR-engineered T cells, gene editing, allogeneic cell therapy, cytokine release syndrome, solid tumors, cell therapy manufacturing, cancer immunotherapy, iPSC-derived immune cells