CTLA-4 blockade shifts the B cell repertoire toward autoimmunity

Elif Çakan, Meng Wang, Yile Dai, Adrien Mirouse, Clarence Rachel Villanueva-Pachas, Delphine Bouis, Joshua M. Boeckers, Ruchi Gera, Sally Yraita, L.m. de la Escalera Clapp, Ana Luisa Perdigoto, Fabien Delmotte, Christopher Massad, Antonietta Bacchiocchi, Aaron M. Ring, Yuval Kluger, Harriet M. Kluger, Kevan C. Herold, Eric MeffreJournal of Clinical Investigation 2025 · DOI

TL;DR

Checkpoint inhibitors like anti-CTLA-4 and anti-PD-1 are revolutionizing cancer treatment but can cause autoimmune side effects. This study reveals that blocking CTLA-4, but not PD-1, leads to a transient increase in autoreactive B cells, potentially explaining these side effects and also contributing to anti-tumor responses.

What Problem Does This Paper Address?

Cancer immunotherapies that target immune checkpoints, specifically CTLA-4 and PD-1, have shown remarkable success in treating various cancers. However, a significant challenge is the emergence of immune-related adverse events (irAEs), which are essentially autoimmune reactions. While both CTLA-4 and PD-1 are crucial for maintaining immune tolerance, it's been unclear how their blockade specifically impacts the B cell repertoire, particularly concerning autoreactive B cells that could drive these irAEs. This paper aims to dissect the distinct roles of CTLA-4 and PD-1 in regulating B cell tolerance, focusing on how their inhibition affects the development and selection of autoreactive B cells.

Methods

The researchers employed a multi-pronged approach:

  • B Cell Repertoire Analysis: They isolated B cells from cancer patients before and after treatment with checkpoint inhibitors (CPIs). Recombinant antibodies were generated from single B cells and tested for reactivity against self-antigens (like dsDNA, insulin, LPS, and HEp-2 cells) using ELISA. This allowed them to quantify the frequency of autoreactive B cells.
  • Humanized Mouse Models: To further investigate the mechanisms, they used humanized mice (NOD-scid IL2Rγnull mice engrafted with human hematopoietic stem cells and thymic tissue). These mice were treated with anti-CTLA-4, anti-PD-1, or isotype control antibodies. B cells from these mice were then analyzed for autoreactivity.
  • Single-Cell RNA Sequencing (scRNA-seq): scRNA-seq was used to analyze the transcriptomic profiles of B cells and CD4+ T cells (specifically regulatory T cells, or Tregs) from patients at different time points during treatment. This helped identify changes in cell populations and gene expression patterns.
  • Rapid Extracellular Antigen Profiling (REAP): This high-throughput yeast-based technology was used to characterize the specific antigens recognized by autoreactive B cell-derived antibodies.

Datasets

The study utilized:

  • Patient Samples: Peripheral blood mononuclear cells (PBMCs) from cancer patients undergoing treatment with anti-CTLA-4 and/or anti-PD-1 checkpoint inhibitors.
  • Humanized Mouse Models: Spleen and blood samples from humanized mice treated with CPIs.
  • Fetal Tissues: Human fetal tissues (bone marrow, spleen, liver, thymus) were used to generate the humanized mouse models.

No new benchmarks were introduced; the study focused on analyzing existing biological systems and patient data.

Key Insights and Ideas

  • CTLA-4, Not PD-1, is the Gatekeeper for Autoreactive B Cells: The most striking finding is that CTLA-4 blockade, but not PD-1 blockade, leads to a significant increase in autoreactive mature naive B cells. This suggests CTLA-4 plays a critical, previously underappreciated role in eliminating these potentially harmful B cells.
  • Central Tolerance Remains Intact: The study confirmed that neither CTLA-4 nor PD-1 blockade disrupts central B cell tolerance in the bone marrow, as evidenced by the normal frequencies of autoreactive B cells in transitional B cell populations.
  • Peripheral Tolerance is the Target: The impact of CTLA-4 blockade is specifically on the peripheral tolerance checkpoint, which normally filters out autoreactive B cells that escape central tolerance.
  • Autoreactive B Cells Recognize Tissue-Specific Antigens: REAP analysis revealed that some of the autoreactive B cells generated after CTLA-4 blockade recognize antigens found in tissues commonly affected by irAEs (e.g., skin, heart, thyroid, brain).
  • Tregs are Affected by Combination Therapy: scRNA-seq showed an expansion and altered transcriptional profile of regulatory T cells (Tregs) in patients treated with combination therapy, suggesting a complex interplay between T cells and B cells during treatment.
  • Broader Repertoire Implications: The findings suggest that CTLA-4 blockade broadens the peripheral B cell repertoire, which could be beneficial for anti-tumor responses by including B cells that recognize tumor antigens, but also detrimental by allowing autoreactive clones to emerge.

Key Results

  • Anti-CTLA-4 and anti-PD-1 combination therapy significantly increased the frequency of HEp-2-reactive (54% vs. 17%) and polyreactive (37.3% vs. 8.5%) mature naive B cells in cancer patients.
  • These increases were transient, returning to baseline levels when anti-CTLA-4 treatment was stopped.
  • Anti-PD-1 monotherapy did not alter autoreactive B cell frequencies in either patients or humanized mice.
  • Anti-CTLA-4 injections in humanized mice, but not anti-PD-1, led to a significant accumulation of autoreactive mature naive B cells.
  • Specific autoreactive clones identified after CTLA-4 blockade recognized antigens like dermatopontin (DPT) and lipocalin 1-like protein 1 (LCN1P1), which are expressed in irAE-targeted organs.
  • scRNA-seq identified an expansion of a specific Treg cluster (cluster 2) in patients treated with combination therapy, characterized by the upregulation of genes like CXCR3 and FOXP3.

Limitations and Future Work

  • Transient Nature of Autoreactivity: The study highlights that the emergence of autoreactive B cells is transient, which is a key observation but also means long-term consequences require further investigation.
  • Limited Patient Numbers for Specific Analyses: While the study used multiple patients, some analyses, particularly scRNA-seq, were performed on a limited number of individuals.
  • Potential for Anti-Tumor Role: The paper acknowledges that the broadened repertoire might also include B cells that recognize tumor antigens, suggesting a potential benefit for anti-tumor immunity. This duality warrants further exploration.
  • Timing of Anti-B Cell Therapy: The authors suggest that anti-B cell therapy could potentially prevent irAEs, but this might compromise anti-tumor responses. Determining the optimal timing for such interventions is a critical area for future research.
  • Investigating Tumor-Infiltrating B Cells: Future studies are needed to examine the reactivity of B cells within tumors and identify the specific antigens they recognize, which could provide deeper insights into their role in anti-tumor immunity.

Summary updated

Abstract

Checkpoint inhibitors targeting CTLA-4 and PD-1 revolutionized the treatment of cancer patients, but their use is limited by the emergence of immune-related adverse events (irAEs). We assessed autoreactive B cell frequencies in the blood of cancer patients before and after treatment with checkpoint inhibitors by testing the reactivity of recombinant antibodies cloned from single B cells. We found that anti-PD-1 and anti-CTLA-4 combination therapy induced the emergence of autoreactive mature naive B cells, whereas central B cell tolerance remained functional. In contrast, anti-PD-1 alone did not alter autoreactive B cell counterselection. Anti-CTLA-4 injections in humanized mice also resulted in the production of autoreactive B cells, whereas anti-PD-1 did not. We conclude that CTLA-4 but not PD-1 is required for the removal of developing autoreactive mature naive B cells and that CTLA-4 blockade broadens the peripheral B cell repertoire, which likely contains clones that promote not only irAEs but also antitumor responses.
BibTeX
@article{cakan2025ctla,
  author = {Elif Çakan and Meng Wang and Yile Dai and Adrien Mirouse and Clarence Rachel Villanueva-Pachas and Delphine Bouis and Joshua M. Boeckers and Ruchi Gera and Sally Yraita and L.M. de la Escalera Clapp and Ana Luisa Perdigoto and Fabien Delmotte and Christopher Massad and Antonietta Bacchiocchi and Aaron M. Ring and Yuval Kluger and Harriet M. Kluger and Kevan C. Herold and Eric Meffre},
  title = {{CTLA-4 blockade shifts the B cell repertoire toward autoimmunity}},
  journal = {Journal of Clinical Investigation},
  year = {2025},
  doi = {10.1172/jci189074},
}
Figure