Deep Dive: CTLA-4 Inhibitors: One Target, Two Antibody Designs

Ipilimumab is a human IgG1 antibody. Tremelimumab is IgG2. That single digit is not a footnote. IgG1 binds activating Fc-gamma receptors far more readily than IgG2, and in preclinical models that difference is what lets ipilimumab-type antibodies deplete CTLA-4-high regulatory T cells inside a tumor, on top of simply blocking the checkpoint. Cancer Cell Tremelimumab’s IgG2 backbone was chosen specifically to blunt that mechanism.

In first-line non-small cell lung cancer, nivolumab paired with ipilimumab went on to earn approval. Durvalumab paired with tremelimumab was also tested in first-line NSCLC, in the phase 3 MYSTIC trial, which compared durvalumab alone and durvalumab plus tremelimumab against chemotherapy; the combination missed its primary endpoints, with no statistically significant overall survival or progression-free survival benefit over chemotherapy. JAMA Oncol Reviewing that divergence, two field leaders writing in Annals of Oncology called the reason for it explicitly unresolved, and named three tangled candidates: the IgG1/IgG2 difference itself, the dosing schedule, or something biologically distinct about pairing CTLA-4 blockade with an anti-PD-1 antibody rather than an anti-PD-L1 antibody. Ann Oncol Tremelimumab’s dosing was later redesigned twice, first into the single-priming-dose STRIDE regimen that succeeded in hepatocellular carcinoma, and separately into a different capped, up-to-five-dose regimen that succeeded in non-small cell lung cancer’s POSEIDON trial; neither redesign used the dosing MYSTIC had used. Because tremelimumab itself produced better results once its own dosing changed, MYSTIC’s failure cannot be attributed to isotype alone: the dosing schedule used in that specific trial is a live alternative explanation, tested and revised within tremelimumab’s own development program.

MYSTIC was not the first time tremelimumab’s underlying design bet came up short. Years earlier, in tremelimumab’s first phase 3 test, monotherapy tremelimumab failed to beat chemotherapy in previously untreated melanoma, around the same era ipilimumab monotherapy succeeded in that same disease. Section 1 covers both episodes and one hypothesis, still unproven, for why the pattern keeps recurring.

That unresolved question is also a clue to a larger pattern. CTLA-4 monotherapy never matched the tolerability or tumor-type breadth that PD-1 and PD-L1 monotherapy achieved on their own, so industry investment shifted toward PD-(L)1 as the backbone treatment, leaving CTLA-4 in a supporting role whose correct isotype, dose, and partner drug are difficult to predict in advance, exactly as MYSTIC demonstrated. A randomized trial directly comparing two competing anti-CTLA-4 antibodies at matched doses would help answer that, but is not the kind of trial industry has been eager to sponsor. Ann Oncol Rather than adding more antibodies to compete head-to-head with ipilimumab and tremelimumab, the current wave of CTLA-4 development is aimed at re-engineering the mechanism itself, a story picked up in Section 10.

The rest of this review returns to that isotype-and-development-program context rather than setting it aside after this introduction. Section 4 identifies hepatocellular carcinoma as the one tumor type where clinicians choose between an ipilimumab-based and a tremelimumab-based regimen, but that choice is a matter of label and sequencing, not a verdict on which antibody performs better, since the two have never been tested against each other directly. Section 10 returns to the isotype question directly, in the context of why so few new anti-CTLA-4 antibodies have reached the clinic.