Background: Transcriptomic Replenishment Blind Spot of Single Immune Checkpoint Inhibitor (ICI) Monotherapy and Immune Evasion Data Bottleneck in Solid Tumor R&D

Existing anti-PD-1/PD-L1 monoclonal antibody-based immune checkpoint inhibitor therapies have been limited to a unidirectional protein-level intervention that blocks ligand-receptor binding on tumor cell surfaces. The critical blind spot of this static standard guideline lies in its inability to computationally control in silico the transcriptomic replenishment feedback loops operating within the tumor microenvironment (TME). In other words, even if an antibody temporarily blocks the PD-L1 protein on the cell surface, the tumor cell's genetic machinery continuously resynthesizes and re-expresses PD-L1 mRNA, restoring the immune evasion pathway within hours. In real-world clinical data, the objective response rate (ORR) of anti-PD-1 monotherapy is only about 20–25% in non-small cell lung cancer (NSCLC) and 10–20% in triple-negative breast cancer (TNBC), which is attributed to the pharmacodynamic limitation of failing to suppress the transcriptional replenishment of immune checkpoint proteins. The superposition of cellular dissociation-induced structural collapse noise, IFN-γ-induced adaptive resistance, and TME-mediated regulatory T cell (Treg)/myeloid-derived suppressor cell (MDSC) immune suppression fluxes has repeatedly caused the conventional single-modality ICI pipeline to fail to maintain an effective anti-tumor immune-active concentration. This data bottleneck has become a structural cause of increased Phase III failure rates and delayed IND approval timelines across solid tumor immuno-oncology R&D.

Discovery: Activation of Self-Assembled DNA Hydrogel Dual-Action Modality and Demonstration of Transcriptome-Proteome Dual-Resolution Simultaneous Immune Checkpoint Repression Tensor Synchronization

In this study, we developed a smart DNA hydrogel based on nucleic acid self-assembly nanoarchitecture, designed as a dual-action therapeutic modality that simultaneously activates the blockade of immune checkpoint proteins on the cell surface (protein level) and the gene silencing of their corresponding mRNA transcripts (nucleic acid level) in a single macromolecular platform. Specifically, a CpG oligonucleotide adjuvant is integrated into the gel matrix backbone to activate the innate immune Toll-like receptor (TLR9) pathway, while antisense oligonucleotides (ASOs) or siRNA payloads are loaded into the 3D network of the hydrogel to degrade the mRNA of PD-L1 (CD274) and CTLA-4-related genes at the post-transcriptional level. This dual-interference matrix proactively neutralizes the tumor cell's genetic machinery that resynthesizes surface immune checkpoint proteins, disruptively suppressing the rate of immune evasion pathway reactivation compared to conventional antibody monotherapy. In silico molecular dynamics simulations were used to computationally pre-calculate the DNA hydrogel-target mRNA binding free energy (ΔG) and gel degradation rate constant, optimizing the payload release kinetics. In vivo mouse solid tumor models showed that the topological variation curve of downstream transcriptomic networks—particularly the IFN-γ/JAK-STAT adaptive resistance axis PD-L1 re-expression pathway—was significantly suppressed. In multiple independent experiments with batch effect removal pipelines, tumor volume reduction, increased CD8+ cytotoxic T lymphocyte (CTL) infiltration, and decreased Treg ratio were statistically reproducible, demonstrating molecular integrity.

Establishment of a PD-L1/CTLA-4 Dual Transcriptional Repression Axis Tuning and Reversible Anti-Tumor Immune Homeostasis Precision Stratification Model

Based on omics matrix analysis, a model was established to precisely stratify predictive molecular phenotypes for the DNA hydrogel dual-action therapy. Tumor mutational burden (TMB), PD-L1 expression H-score, microsatellite instability (MSI) status, and the ratio of immune cell composition in the TME are integrated into a multivariate tensor to stratify the probability of dual-action therapy response in patient groups. In particular, the gene silencing effect of the DNA hydrogel is maximized in patients with high IFN-γ signature/high PD-L1 transcript replenishment, while in immune desert phenotypes, the CpG adjuvant axis activates innate immune priming, overcoming the TME immune activation threshold through a dual-complementary mechanism. By up-clamping (slow release via cross-linking density adjustment) and down-clamping (rapid release via enzymatic reactive cross-linking) the rate-limiting step of hydrogel degradation-payload release kinetics, a responsive backbone was created that reversibly and autonomously regulates the effective immune checkpoint inhibition concentration within the homeostatic range under tumor microenvironment stress conditions such as pH decrease and MMP overexpression. This functions as a prototype for adaptive drug delivery governance that responds to the spatiotemporal heterogeneity of tumors.

Prospects: Establishment of a Programmable Nucleic Acid Nano-Medicine Standard and Activation of Next-Generation IND Digital Governance

The DNA hydrogel dual-action platform is a declarative turning point that completely resets immuno-oncology R&D governance from a static, post-hoc symptomatic system to a proactive, AI-driven multidimensional tensor-based programmable infrastructure. Following the successful commercialization of Alnylam Pharmaceuticals' ONPATTRO (2018, exceeding $500 million in annual revenue) and the validation of the Moderna/BioNTech mRNA platform, the nucleic acid therapeutics market is projected to grow to approximately $25 billion by 2030 (Grand View Research, 2024), and DNA hydrogel-based dual-action immune checkpoint modulators will be positioned as the next-generation modality in this pipeline. In the context of global multinational pharmaceutical companies conducting large-scale pivotal clinical trials for immune checkpoint combination therapies—Bristol-Myers Squibb's Nivolumab+Ipilimumab (CheckMate series), Merck's Pembrolizumab combination expansion, Roche/Genentech's Atezolizumab+Tiragolumab (TIGIT axis)—by linking the efficiency coefficient of transcriptional-level immune checkpoint inhibition of this platform as a correction variable, the computational zeroing of batch-to-batch PD-L1 re-expression variance and the implementation of a closed-loop system that provides real-time feedback on genetic gradient correction coefficients in high-throughput screening are possible. Furthermore, by standardizing the linkage between the companion diagnostic (CDx) panel—Foundation Medicine's FoundationOne CDx, Guardant Health's Guardant360 CDx—and the molecular phenotype stratification model of this platform, a digital governance master asset can be established that disruptively shortens the FDA/EMA IND approval evaluation framework.

The presentation of preclinical data on nucleic acid nanostructure-based immune checkpoint modulation at AACR 2026 and ASCO 2026 is accelerating, and this study will establish itself as a reference architecture that leads the establishment of a programmable nucleic acid nano-medicine standard in this field.