Daily Respiratory Research Analysis
Three impactful respiratory studies span basic mechanisms, therapeutics, and prevention. A mechanistic study reveals how SMARCA4-BRD4 chromatin remodeling controls innate immunity and epithelial plasticity during RSV injury. A phase 2 RCT shows a novel influenza polymerase inhibitor accelerates viral clearance, and a machine learning model in people with HIV predicts incident TB risk better than TST/IGRA, enabling targeted prevention.
Summary
Three impactful respiratory studies span basic mechanisms, therapeutics, and prevention. A mechanistic study reveals how SMARCA4-BRD4 chromatin remodeling controls innate immunity and epithelial plasticity during RSV injury. A phase 2 RCT shows a novel influenza polymerase inhibitor accelerates viral clearance, and a machine learning model in people with HIV predicts incident TB risk better than TST/IGRA, enabling targeted prevention.
Research Themes
- Host epigenetic regulation of antiviral responses (RSV)
- Antiviral therapeutics and virologic endpoints (influenza)
- AI-driven risk stratification for respiratory infections (TB in HIV)
Selected Articles
1. SMARCA4 regulates inducible BRD4 genomic redistribution coupling intrinsic immunity and plasticity in epithelial injury-repair.
Using CUT&RUN in RSV-infected basal epithelial cells, the authors show that SMARCA4 orchestrates BRD4 redistribution from mesenchymal gene bodies to open chromatin and super-enhancers that regulate cytokines, adhesion, antiviral programs, and immune lncRNAs. SMARCA4 knockdown reduces BRD4 occupancy and nucleosome-free region boundaries, implicating SWI/SNF ATPases in maintaining enhancer openness and coupling lncRNA expression to intrinsic antiviral immunity and epithelial plasticity.
Impact: This work uncovers a chromatin-level mechanism linking innate immunity, lncRNA regulation, and epithelial state transitions during RSV injury, revealing potential epigenetic targets for therapy.
Clinical Implications: While preclinical, targeting SMARCA4-BRD4 enhancer dynamics or downstream immune lncRNAs could modulate epithelial repair and antiviral responses in severe RSV infection and related airway injury.
Key Findings
- RSV replication repositions 2339 BRD4 peaks to open chromatin upstream of inducible cytokine, adhesion, and antiviral genes.
- RSV redistributes BRD4 into super-enhancers that regulate immune response–associated lncRNAs; SMARCA4 knockdown reduces BRD4 occupancy on 739 peaks.
- SMARCA4 maintains nucleosome-free region boundaries at super-enhancers and controls lncRNAs important for IRF1 autoregulation, coupling intrinsic immunity with epithelial plasticity.
Methodological Strengths
- Genome-wide CUT&RUN mapping of BRD4 under RSV infection with and without SMARCA4 knockdown
- Integration of enhancer architecture (super-enhancers, nucleosome-free regions) with lncRNA regulation and innate immune programs
Limitations
- Findings are primarily from in vitro basal epithelial models; in vivo validation is limited.
- Functional rescue and therapeutic modulation of specific lncRNAs or enhancers were not tested.
Future Directions: Validate enhancer–lncRNA–immune circuitry in vivo, define druggable nodes (e.g., BRD4/SMARCA4 interfaces), and test whether modulating these pathways improves outcomes in RSV or other epithelial injury.
Coordinated expression of differentiation and innate pathways is essential for successful mucosal injury-repair. Previously, we discovered that the core SWI/SNF complex ATPase, SWI/SNF-related, matrix associated, actin dependent regulator of chromatin, subfamily A, member 4 (SMARCA4)/Brg1, maintains tumor protein 63 + basal progenitor cells in an epithelial-committed state. In response to viral injury, SMARCA4 complexes BRD4 to activate innate inflammation and promote mesenchymal transition/plasticity. To investigate how innate inflammation couples with plasticity, Cleavage Under Targets and Release Using Nuclease of BRD4 binding was applied to wild type and SMARCA4 knockdown (KD) in mock- or respiratory syncytial virus (RSV)-infected basal cells. In mock-infected cells, BRD4 binds 4017 high-confidence peaks within gene bodies controlling mesenchymal transition pathways. By contrast, RSV replication repositions 2339 BRD4 peaks to open chromatin regions upstream of the genes controlling inducible cytokine, cell adherence, and antiviral programs. Also, we note RSV redistributes BRD4 into super enhancers regulating immune response-associated long noncoding (lnc)RNAs. In SMARCA4 KD cells, BRD4 distribution is reduced on 739 peaks after RSV infection. The boundaries of nucleosome-free regions are reduced by SMARCA4 KD, suggesting its role in maintaining open chromatin of super enhancers. Specifically, SMARCA4-BRD4 enhancer controls lncRNAs important in interferon response factor 1 autoregulation. These data indicate how SWI/SNF ATPases couple BRD4 to lncRNA expression controlling cell state and intrinsic immunity in epithelial injury-repair.
2. Efficacy and Safety of WXSH0208 Tablets in Treatment of Acute Uncomplicated Influenza Infection in Adults: A Multicenter Randomized, Double-Blind, Placebo-Controlled Phase 2 Trial.
In a multicenter double-blind phase 2 RCT (n=209 ITT-infected), WXSH0208 shortened time to RT-qPCR negativity by ~48–49 hours versus 95.6 hours on placebo, across dosing regimens, with similar symptom trajectories and acceptable safety. These results validate influenza polymerase targeting in uncomplicated influenza and warrant phase 3 clinical outcome studies.
Impact: Demonstrates robust virologic efficacy of a novel influenza polymerase inhibitor in a randomized, placebo-controlled setting, informing the pipeline for next-generation antivirals.
Clinical Implications: Although symptom relief was not accelerated, accelerated viral clearance with acceptable safety supports further trials to assess clinical endpoints (symptom duration, complications, transmission) and positioning relative to neuraminidase inhibitors and baloxavir.
Key Findings
- Median time to RT-qPCR negativity: ~48–49 h with WXSH0208 vs 95.6 h with placebo (P<.001) across dosing arms.
- No significant difference in time to symptom alleviation among groups.
- Treatment-emergent adverse events were mostly mild/moderate and comparable or lower than placebo (48.3–51.7% vs 58.3%).
Methodological Strengths
- Randomized, double-blind, placebo-controlled, multicenter phase 2 design with ITT-infected analysis
- Predefined virologic primary endpoint (time to RT-qPCR negativity) with consistent effects across doses
Limitations
- Clinical symptom endpoints were not improved; study not powered for complications or transmission outcomes.
- Short follow-up focused on early virologic kinetics; resistance emergence was not reported.
Future Directions: Proceed to phase 3 trials emphasizing clinical endpoints, resistance monitoring, high-risk subgroups, and head-to-head comparisons versus standard antivirals.
BACKGROUND: WXSH0208 is a selective inhibitor of influenza RNA polymerase subunit, demonstrating antiviral activity in preclinical studies against influenza A and B virus infections. The purpose of this study was to investigate the efficacy and safety of WXSH0208 in adult outpatients with uncomplicated influenza. METHODS: We conducted a multicenter phase 2 trial based on a randomized, double-blind, placebo-controlled design at 23 research centers in China from November 2023 to March 2024. Participants were randomized 1:1:1:1 to receive one of the following treatments within 48 hours of symptom onset: WXSH0208 10 mg once daily for 5 days, 20 mg once daily for 5 days, 30 mg once daily for 3 days, or placebo. The primary outcome was the time to negative detection of viral load by reverse transcriptase quantitative polymerase chain reaction in the intention-to-treat infected population. RESULTS: Of 240 randomized patients, 209 were included in the intention-to-treat infected analysis. The median time to negative detection of viral load was 49.3 hours in the WXSH0208 10 mg group, 48.0 hours in the 20 mg group, and 48.2 hours in the 30 mg group, as compared with 95.6 hours in the placebo group (P < .001). Time to alleviation of influenza symptoms was comparable among all groups. Treatment-emergent adverse events were reported in 48.3% to 51.7% of WXSH0208 recipients and 58.3% of placebo recipients, with most being mild or moderate in severity. CONCLUSIONS: WXSH0208 showed no evident safety concerns and was superior to placebo in reducing viral load in adult outpatients with uncomplicated influenza. Clinical Trials Registration. CTR20233250 (www.chinadrugtrials.org.cn).
3. Machine Learning-based Prediction of Active Tuberculosis in People With HIV Using Clinical Data.
Using routinely collected enrollment data, a random forest model predicted incident active TB among people with HIV with AUC 0.83 internally and 0.67 in external validation after adjustment, outperforming TST/IGRA via a lower number needed to diagnose (1.96 vs 4). This approach requires no new testing and could target preventive TB therapy efficiently.
Impact: Provides an externally validated, low-cost, scalable risk prediction tool for TB progression in PWH, a critical gap for preventive therapy deployment.
Clinical Implications: Clinicians and programs could prioritize IPT/TPT for PWH at highest modeled risk without additional testing, potentially reducing incident TB and optimizing resource allocation.
Key Findings
- Random forest model trained on SHCS predicted incident TB with AUC 0.83; adjusted parsimonious models yielded AUC 0.72 (internal) and 0.67 (external).
- Outperformed TST/IGRA by a lower number needed to diagnose high-risk individuals (1.96 vs 4).
- Model uses routinely collected enrollment data, implying minimal incremental cost and no added data burden.
Methodological Strengths
- External validation across independent national cohort (Austria) with matched controls
- Comparison against standard of care tests (TST/IGRA) using pragmatic metrics (number needed to diagnose)
Limitations
- Relatively small number of TB events limits model complexity and may affect generalizability.
- Performance attenuated after adjustment and in external validation; prospective impact evaluation is needed.
Future Directions: Prospective implementation trials to test preventive therapy targeting by model risk; retraining/transfer learning in high-burden, diverse settings; transparent model sharing and calibration.
BACKGROUND: Coinfections of Mycobacterium tuberculosis (MTB) and human immunodeficiency virus (HIV) impose a substantial global health burden. Patients with MTB infection face a heightened risk of progression to incident active TB, which preventive therapy can mitigate. Current testing methods often fail to identify individuals who subsequently develop incident active TB. METHODS: We developed random forest models to predict incident active TB using patients' medical data at HIV-1 diagnosis. Training our model involved using clinical data routinely collected at enrollment from the Swiss HIV Cohort Study (SHCS). This dataset encompassed 55 people with HIV (PWH) who developed incident active TB 6 months after enrollment and 1432 matched PWH without TB enrolled between 2000 and 2023. External validation used data from the Austrian HIV Cohort Study, comprising 43 people with incident active TB and 1005 people without TB. RESULTS: We predicted incident active TB with an area under the receiver operating characteristic curve of 0.83 (95% CI: .8-.86) in the SHCS. After adjusting for ethnicity and the region of origin and refitting the model with fewer parameters, we obtained comparable receiver operating characteristic curve values of 0.72 (SHCS) and 0.67 (Austrian HIV Cohort Study). Our model outperformed the standard of care (tuberculin skin test and interferon-gamma release assay) in identifying high-risk patients, demonstrated by a lower number needed to diagnose (1.96 vs 4). CONCLUSIONS: Models based on machine learning offer considerable promise for improving care for PWH, requiring no additional data collection and incurring minimal additional costs while enhancing the identification of PWH that could benefit from preventive TB treatment.