Intratumoral CTLA-4 Blockade: Decoupling Local Immune Remodeling From Systemic Toxicity

Abstract

In the randomized phase 1b NIVIPIT trial, Tselikas et al. showed that low-dose intratumoral ipilimumab combined with intravenous nivolumab markedly reduced severe treatment-related adverse events while maintaining considerable antitumor activity [1]. These findings support intratumoral CTLA-4 blockade as a strategy to dissociate local immune activation from systemic toxicity and provide a conceptual framework for spatially restricted immunotherapy.

Increasing evidence suggests that the efficacy of anti-CTLA-4 therapy depends not only on checkpoint blockade itself but also on FcγR-mediated immune remodeling within the tumor microenvironment [2]. Unlike anti-PD-1 antibodies, whose efficacy largely depends on receptor occupancy and reinvigoration of exhausted T cells, ipilimumab is an Fc-competent IgG1 antibody capable of engaging Fcγ receptors and mediating depletion of intratumoral regulatory T (Treg) cells [3]. Therefore, intratumoral delivery may preserve the key effector mechanisms of anti-CTLA-4 therapy at the tumor site while reducing systemic exposure and peripheral immune activation. However, because murine and human FcγR systems differ, the contribution of individual FcγR subtypes requires validation in human tumors.

The NIVIPIT trial supports this concept by showing that substantial antitumor activity can be achieved in injected lesions despite markedly reduced circulating ipilimumab levels. This finding further suggests that CTLA-4 blockade is highly dependent on spatial immune context. Its efficacy may be shaped by localized interactions among activated Treg cells, FcγR-expressing macrophages, antigen-presenting cells, and tumor-reactive T cells [4]. In this setting, the injected lesion is not merely a drug delivery site, but may function as an immune-remodeling niche whose cellular composition determines both local tumor regression and the extent of systemic immune mobilization (Figure 1).

However, the study also highlights an important conceptual distinction between local immune remodeling and systemic immune priming. Although injected lesions responded strongly, responses in uninjected lesions were less pronounced, and systemic pharmacodynamic activation appeared attenuated compared with intravenous ipilimumab. Peripheral immune activation and inflammatory cytokine induction were also weaker in the intratumoral treatment arm. These findings suggest that intratumoral CTLA-4 blockade may effectively remodel the injected tumor microenvironment but may not fully reproduce the systemic immune amplification induced by conventional intravenous CTLA-4 blockade. Therefore, the apparent safety advantage of intratumoral therapy may partly reflect the same biological constraint that limits its ability to drive broad systemic antitumor immunity.

This distinction may have important translational implications. Systemic anti-CTLA-4 therapy likely exerts at least two partially separable functions: local depletion of CTLA-4-expressing Treg cells and broader systemic expansion or reinvigoration of antitumor immunity. Intratumoral administration appears highly effective at the former while restraining the latter. This may explain why local CTLA-4 blockade is particularly attractive for accessible lesions, oligometastatic disease, or neoadjuvant settings in which regional immune remodeling may be sufficient to induce meaningful clinical responses. In contrast, patients with disseminated metastatic disease may still require a certain degree of systemic immune amplification to achieve durable control across distant tumor sites. Thus, local CTLA-4 blockade should not be viewed simply as a lower-toxicity substitute for conventional therapy, but rather as a biologically distinct immunotherapeutic strategy with different spatial and systemic properties.

Another notable finding of the study is that durable clinical benefit was associated with increased baseline infiltration of activated Treg cells and FcγR-positive macrophages. This observation challenges the oversimplified view that immunosuppressive populations uniformly predict resistance to immunotherapy. Instead, activated Treg cells may simultaneously represent immune suppression and therapeutic vulnerability. In tumors enriched with FcγR-expressing myeloid cells, CTLA-4-positive Treg cells may provide a permissive niche for Fc-dependent depletion by anti-CTLA-4 antibodies. In this context, anti-CTLA-4 therapy may function less as a conventional checkpoint inhibitor and more as a spatial immune-remodeling therapy that depends on coordinated interactions between Treg cells, macrophages, and Fcγ receptor signaling within the tumor microenvironment. These findings further suggest that biomarkers reflecting FcγR-dependent immune remodeling may be more informative than simple measurements of immune infiltration when predicting response to CTLA-4-targeted therapy.

One additional issue is how lesion selection should be incorporated into future trial design. If the injected lesion acts as a local immune-remodeling site, its cellular composition, anatomical location, size, vascularity, and accessibility may influence both local efficacy and the extent of systemic immune mobilization. Therefore, future studies should not only evaluate patient-level biomarkers but also characterize lesion-level immune states before treatment, including the abundance of CTLA-4-expressing Treg cells and FcγR-expressing myeloid cells. This may help identify which lesions are most suitable for intratumoral CTLA-4 blockade and which patients may require additional systemic or innate immune stimulation.

Several limitations should temper the interpretation of the NIVIPIT study. The phase 1b sample size was limited, all patients received intravenous nivolumab, and the trial was not powered for formal efficacy comparisons with conventional intravenous ipilimumab. Moreover, responses differed between injected and uninjected lesions, while the reported biomarker associations remain exploratory. It therefore remains uncertain whether reduced systemic toxicity can be achieved without compromising long-term disease control, particularly in patients with extensive or noninjectable disease. Future randomized studies should separately evaluate injected-lesion response, distant-lesion response, survival outcomes, and immune changes in paired biopsies, while prospectively validating Treg- and FcγR-related biomarkers.

These insights provide important directions for the development of next-generation CTLA-4-targeted therapies. Compared with other intratumoral immunotherapies, including PD-1/PD-L1 blockade, oncolytic viruses, cytokines, and innate agonists, local CTLA-4 blockade is distinguished by its potential to combine checkpoint antagonism with FcγR-dependent depletion of CTLA-4-high Treg cells. However, direct comparative evidence remains limited. Future strategies may need to balance regional immune remodeling with controlled systemic immune priming. Fc-engineered anti-CTLA-4 antibodies, tumor-restricted biologics, masked antibodies, and intratumoral immunostimulatory approaches may help improve this balance. In particular, combining local CTLA-4 blockade with innate immune activation, such as STING or Toll-like receptor agonists, may enhance systemic T-cell priming while preserving the favorable safety profile of intratumoral delivery [5]. Local innate immune activation may compensate for insufficient peripheral immune amplification and promote broader antitumor immunity across uninjected lesions. In parallel, image-guided delivery strategies and biomarker-defined lesion selection may further optimize the spatial precision of intratumoral immunotherapy.

Overall, the NIVIPIT study refines our understanding of anti-CTLA-4 biology by highlighting the spatial requirements for effective FcγR-dependent immune remodeling. More importantly, it suggests that the next generation of CTLA-4-directed therapies may depend not simply on reducing systemic exposure, but on strategically integrating local immune activation with sufficient systemic antitumor priming to achieve durable disease control across metastatic sites.

Luyan Chen: conceptualization, investigation, visualization, writing – original draft. Mengke Niu: investigation, visualization, writing – original draft. Haidong Cui: investigation, validation, writing – original draft. Zhijun Dai: conceptualization, supervision, project administration, funding acquisition, writing – review and editing. Ming Yi: conceptualization, supervision, project administration, funding acquisition, writing – review and editing. All authors have read and approved the final manuscript.

The authors have nothing to report.

The authors declare no conflicts of interest.

No new datasets were generated or analyzed in this commentary.