24 juni 2026
Keywords
● PD-1 Blockade
● Drug Holiday
● Exhausted T cells
● Tpex Adaptive Resistance
Main Findings
In this preprint, the authors investigate how the timing and duration of PD-1 pathway blockade affect exhausted CD8 T cell (Tex) reinvigoration during chronic infection and cancer. PD-1 blockade can transiently reinvigorate Tex cells by targeting progenitor exhausted T cells (Tpex), which proliferate and differentiate into downstream effector-like Tex intermediate (Tex-Int) and terminal Tex (Tex-Term) states. However, previous work has shown that this reinvigoration often occurs as a single early burst, even when PD-1 blockade is continued. The authors therefore ask whether continuous long-term blockade is sufficient to sustain Tex reinvigoration, or whether intermittent treatment could preserve Tpex function and enable renewed responsiveness to PD-1 blockade.
Using chronic LCMV infection with trackable antigen-specific CD8 T cells, the authors compare three treatment schedules: short-term PD-1 blockade, continuous long-term PD-1 blockade, and an intermittent “drug holiday” regimen in which treatment is paused for eight weeks before re-blockade. Continuous long-term blockade provides little additional benefit compared with short-term treatment. Both regimens induce an initial burst of Tex expansion, but this response contracts over time, and continuous blockade does not sustain improved proliferation, effector function or systemic viral control. In contrast, the drug holiday regimen enables a second wave of Tex reinvigoration upon retreatment, with increased Tex expansion, improved effector function and enhanced viral control.
The authors then examine how these schedules affect Tex subset composition. Continuous PD-1 blockade reduces the effector-like Tex-Int population and is associated with a reduced Tpex pool, whereas the drug holiday regimen preserves the ability of Tpex cells to generate Tex-Int progeny. Single-cell RNA sequencing supports this conclusion, showing that long-term blockade leads to loss of Tex-Int cells, enrichment of Tex-Term cells and reduced preservation of the Tpex compartment. More detailed analysis of Tpex cells shows that long-term blockade enriches transcriptional states associated with quiescence, terminal differentiation potential and adaptive resistance.
The authors next identify CD22 as an inhibitory receptor expressed by a subset of Tpex cells. CD22+ Tpex cells show higher expression of transcription factors associated with stemness, higher TCF-1 expression and stress-response genes, while also showing features of quiescence and restrained differentiation. In adoptive transfer experiments, CD22+ Tpex cells respond robustly to short-term PD-1 blockade and show strong self-renewal capacity. However, when CD22+ Tpex cells are derived from mice that had undergone prolonged continuous PD-1 blockade, their ability to expand and generate downstream Tex-Int progeny is impaired. These findings suggest that continuous PD-1 blockade induces a Tpex-intrinsic adaptive resistance state, in which CD22 marks a responsive but restrained progenitor population.
Finally, the authors test whether targeting CD22 can overcome this adaptive resistance. In mice receiving long-term PD-1 blockade, addition of anti-CD22 improves circulating Tex expansion and viral control. In treatment-naive chronically infected mice, combined PD-1 and CD22 blockade further improved viral control compared with PD-1 blockade alone. The authors then engineer a PD-1–CD22 bispecific antibody, which enhances Tex numerical expansion, Tpex proliferation, effector cytokine production, degranulation and viral control compared with single-target blockade. The authors also extend this approach to an AT3-OVA tumour model, where PD-1–CD22 bispecific treatment reduces tumour growth and increases intratumoural CD8 T cells.
Together, these findings suggest that intermittent PD-1 blockade can preserve Tpex responsiveness, and that CD22 co-targeting may overcome Tpex adaptive resistance during prolonged therapy.
Limitations
1. The reduced benefit of long-term versus short-term PD-1 blockade is replicated in a tumour model (MCA1956), supporting the generality of that finding. However, the core drug-holiday experiment which includes withdrawal followed by retreatment, and the resulting second wave of Tex reinvigoration is demonstrated only in the chronic LCMV model. Whether the same washout-retreat benefit would hold in a progressive, heterogeneous tumour remains untested.
2. CD22 is also highly expressed by B cells, so the cellular mechanism of CD22 blockade is not completely resolved. The authors perform B cell depletion experiments and observe that some effects of combined PD-1 and CD22 targeting are preserved, but they also acknowledge that B cell depletion alters viral load and lymphoid architecture, complicating interpretation. Additional T cell-specific genetic approaches would strengthen the claim that CD22 acts directly within Tpex cells.
3. It also remains unclear whether CD22 is the only resistance associated inhibitory receptor induced by prolonged PD-1 blockade, or whether it is one component of a broader adaptive resistance program. The scRNA-seq data suggest that long-term blockade alters multiple Tpex transcriptional states and stress response pathways. Further mechanistic work will be needed to determine whether CD22 blockade directly releases Tpex from inhibitory signalling.
4. The study would be strengthened by validation in human samples, particularly longitudinal samples from patients receiving continuous PD-1 blockade, treatment interruption and retreatment. Such data would help determine whether human Tpex cells also acquire CD22-associated adaptive resistance.
Significance/Novelty
This study identifies adaptive resistance within the Tpex compartment as a previously underappreciated limitation of prolonged PD-1 blockade. While PD-1 blockade is known to transiently reinvigorate exhausted CD8 T cells, this work shows that continuous treatment can weaken Tpex function and limit the generation of more functional Tex-Int progeny. The authors also uncover CD22 as a previously unrecognized inhibitory pathway associated with restrained CD22+ Tpex cells. Finally, by developing a PD-1–CD22 bispecific antibody, they translate these mechanistic findings into a potential therapeutic strategy for restoring Tex reinvigoration.
Preprint rating
· Scientific quality=4
· Novelty=5
· Significance=5
Please award 1-5 stars for the preprint (won’t be public, only use for internal assessment). 5 stars being an outstanding preprint, 1 being a poor preprint.
Credit
Reviewed by Chih-Wei Chu as part of a cross-institutional journal club between the Icahn School of Medicine at Mount Sinai, the University of Oxford, the Karolinska Institute, the University of Toronto, and the UT MD Anderson James P. Allison Institute.
The author declares no conflict of interests in relation to their involvement in the review.