🇪🇸 Leer este artículo en español

When Checkpoints Fall Short

Checkpoint inhibitors changed oncology forever. But not for everyone: there are patients with primary resistance, tumors that relapse, and toxicities that limit dosing. And outside of oncology, the underlying question remains open: if immunotherapy can awaken the immune system against cancer, why not use it against cells causing autoimmune diseases?

A review published in 2026 in Biochemical Pharmacology (Abdelaziz and Shokr) makes precisely that journey: what comes after checkpoint inhibitors. And the map has two protagonists: T cell engagers (TCEs) and antibody-drug conjugates (ADCs).

The Two Guided Missiles

T cell engagers (TCEs). They are bispecific antibodies that bind on one side a cytotoxic T cell and on the other the target cell, forcing contact and cell death without depending on MHC. That is: they do not need the tumor to “present” the antigen in the classical way; the drug acts as a bridge. We already use them in oncology — blinatumomab (CD19×CD3) in acute lymphoblastic leukemia or teclistamab (BCMA×CD3) in multiple myeloma — but their potential goes beyond that.

Antibody-drug conjugates (ADCs). An antibody recognizing a surface antigen + a cytotoxic payload released after receptor internalization. The classic example in our practice: anti-HER2 conjugates. The idea is simple: use the antibody as a selective delivery vehicle.

What This Review Brings

1. The CD47/SIRPα axis as a case study. Agents like evorpacept and BYON4228, combined with rituximab, show objective response rates around 50% in non-Hodgkin lymphoma, and next-generation designs reduce the hematologic toxicity that limited earlier ones.

2. From oncology to autoimmunity. The most striking findings come from compassionate use: the CD19×CD3 TCE (blinatumomab) in refractory anti-synthetase syndrome and the BCMA×CD3 (teclistamab) in systemic sclerosis achieved rapid clinical improvement — but with grade 3 cytokine release syndrome (CRS) in 40% and 100% of patients, respectively, and no neurotoxicity (PMID: 42476268).

3. Engineering to tame toxicity. In TCEs with attenuated CD3 affinity, grade 1-2 CRS appears in less than 20% of patients. The lesson: cytotoxic effect can be maintained while reducing the fire.

4. The adverse side of ADCs. Not everything works: an anti-TNF ADC with a glucocorticoid receptor modulator did not outperform adalimumab in a phase 2b trial. Safety profile is the Achilles’ heel: TCEs pay in CRS; ADCs, in off-target payload release toxicity.

🧐 Critical Reading

• It’s a narrative review by two authors, not a meta-analysis. It serves as a conceptual map, not a clinical practice guideline. The evidence level is descriptive.
• Autoimmunity data are compassionate use, with small series. The “100% grade 3 CRS” with teclistamab in systemic sclerosis must be read as a warning: the efficacy/safety balance in autoimmune diseases is not the same as in oncology, where we tolerate more toxicity for tumor control.
• Promising numbers (50% ORR in lymphoma) come from early-stage combinations; confirmation in controlled trials is lacking.

💡 Why It Matters to Clinicians

Because the same mechanism that eliminates tumor B cells can eliminate autoreactive clones, and the drugs we already handle in oncology (blinatumomab, teclistamab) are the spearhead of that translation. Knowing the mechanism and expected toxicity (CRS, and its management) is no longer just a trial topic: it’s becoming cross-cutting clinical competence.

Reflection

The boundary between oncology and immunology is blurring. The challenge for the next decade will not only be “killing the bad cell,” but doing so with precision — without depending on MHC, with less CRS, with payloads that don’t escape. Guided missiles are already in the air; the question is how long they’ll take to land in daily practice.

---

Reference: Abdelaziz AM, Shokr MM. Beyond immune checkpoint blockade: T cell engagers and antibody-drug conjugates in cancer and autoimmune disease. Biochemical Pharmacology. 2026. DOI: 10.1016/j.bcp.2026.118276 · PMID: 42476268 · https://doi.org/10.1016/j.bcp.2026.118276

— This analysis was generated by ANGIE (Always Next to Guide, Inspire and Empower), an artificial intelligence system with SOUL profiles, designed by Dr. Javier Pumares Pérez.

Connect on LinkedIn · ppj.es