Engineering Specificity to Mitigate On-Target Off-Tumor Toxicity

39 concepts32 papers1 proposal

The primary challenge in modern cancer immunotherapy is not just finding a way to kill cancer cells, but finding a way to kill only cancer cells. While Chimeric Antigen Receptor (CAR) T-cell therapy has revolutionized the treatment of liquid biopsies like leukemia, solid tumors remain a "formidable fortress" Zhang et al.. The central hurdle is the antigen specificity gap: most identified tumor antigens are overexpressed rather than unique to the malignancy. When a therapeutic agent hits the correct target (the antigen) but on the wrong cell (healthy tissue), the result is on-target, off-tumor toxicity.

The Problem of Shared Antigens

In an ideal scenario, a tumor would express a protein that exists nowhere else in the body. In reality, most targets are shared. As Zhang et al. note, the lack of absolute specificity means engineered immune cells often cannot distinguish between a malignant cell and a vital organ expressing the same marker at a lower level.

This overlap manifests in several dangerous ways across different cancers:

  • Hematopoietic Fratricide: In acute myeloid leukemia (AML), common targets like CD123 and CD33 are also present on healthy hematopoietic stem cells. Lu et al. explain that this overlap risks severe myelosuppression, effectively destroying the patient's ability to produce new blood cells. Similarly, targeting PD-L1 can lead to "fratricide," where CAR-T cells attack other activated therapeutic T cells Qin et al..
  • Vital Organ Damage: Proteins that drive aggressive growth, such as Podocalyxin, are often essential for the physiological function of the kidneys or vascular system Canals Hernaez et al..
  • Immune Dysregulation: Targeting markers like CD80 to suppress inflammation can inadvertently deplete B cells and dendritic cells, disrupting broader immune functions Yang et al..

Engineering Biological Logic Gates

To bridge the specificity gap, researchers are moving away from simple "see antigen, kill cell" mechanics toward sophisticated biological computing. Zhu et al. describe treating immune cells as programmable computers that use logic gates to improve discrimination.

Instead of recognizing one antigen, these cells can be programmed with an AND gate: they only activate if they detect two different antigens simultaneously. This significantly narrows the "off-tumor" window, as the probability of a healthy cell expressing both markers at high levels is much lower than expressing just one.

Temporal and Modular Control

Another strategy involves giving clinicians a "remote control" over the therapy. If toxicity begins to manifest, the therapy needs to be toggled off. Zhang et al. demonstrated this using a doxycycline-inducible (Tet-On) system. In this model, the CAR-T cells only express their targeting receptors when the patient is administered the antibiotic doxycycline. This allows for the anti-tumor activity to be tuned or halted in response to real-time safety data.

Similarly, Märkl et al. proposed a modular approach using Synthetic Agonistic Receptors (SARs) and bispecific antibodies. By separating the T-cell from its targeting mechanism, the therapy becomes controllable and reversible, providing a safety buffer against the high-density antigens found on healthy tissues.

Searching for Unique Epitopes

While engineering better "software" for immune cells is one path, others are searching for better "targets." This involves looking deeper than the protein level to the epitope or glycopeptide level.

For example, Zhou et al. highlight that the protein MUC1 is expressed on both healthy and cancerous cells, but its sugar coatings (glycosylation patterns) differ significantly in tumors. By targeting these specific "glycopeptide signatures," it may be possible to achieve the absolute specificity that has so far eluded standard antibody-based therapies. Identifying these "truly" tumor-specific antigens, such as the PRAME protein in adenoid cystic carcinoma Tang et al., remains a high priority for expanding the therapeutic window of adoptive cell therapies Du et al..

Key Foundational Work

Hurdles of CAR-T cell-based cancer immunotherapy directed against solid tumors (2016) — contributed 2 concept elements to this direction.
BAP31, a promising target for the immunotherapy of malignant melanomas (2015) — contributed 1 concept elements to this direction.
History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_on_target_off_tumor_toxicity_5a61c41f,
  title = {on-target off-tumor toxicity},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/on-target-off-tumor-toxicity/rdr_5a61c41fcb6f4dc1af4d66ad0426b8ac},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-21},
}

This exact version (version 1, frozen text):

@misc{lacuna_on_target_off_tumor_toxicity_5a61c41f_v1,
  title = {on-target off-tumor toxicity},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/on-target-off-tumor-toxicity/rdr_5a61c41fcb6f4dc1af4d66ad0426b8ac/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-21},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake