Rentosertib Shifts Six Proteomic Aging Clocks in Phase 2a IPF Substudy, but Cannot Separate Anti-Aging From Anti-Fibrotic Effects
核心洞察
A nested analysis of 42 idiopathic pulmonary fibrosis (搜索) patients found rentosertib (搜索)-treated groups showed lower predicted biological age across six proteomic clocks, while placebo participants showed minimal change or slight increases.
The 30 mg twice-daily regimen produced the most consistent cross-clock signal, with 21 of 54 treatment-versus-placebo comparisons reaching a false-discovery-rate-adjusted q value below 0.10.
Rentosertib (搜索) significantly altered trajectories of 326 proteins versus two with placebo, with modulated proteins enriched 1.74-fold for age-associated proteins in UK Biobank data.
A nested proteomic substudy of a Phase 2a trial has found that rentosertib (搜索), an experimental TNIK (搜索) inhibitor developed by Insilico Medicine (搜索) for idiopathic pulmonary fibrosis (搜索) (IPF), was associated with reductions in predicted biological age across six independent protein-based aging clocks. The analysis, published September 7 in Nature Biotechnology, examined serum samples from 42 patients and reported that treated groups generally showed declines in predicted biological age from baseline, whereas placebo participants showed minimal changes or slight increases.
The findings are explicitly exploratory. The authors state that proteomic clocks "cannot deconvolute aging- and disease-specific effects," meaning a drug that eases lung scarring could make blood proteins appear younger simply because patients are less ill. The study measured surrogate molecular signals, not lifespan, healthspan, or any clinical aging outcome.
Trial Design and Proteomic Cohort
The parent trial was a randomized, double-blind, placebo-controlled Phase 2a study conducted at 22 sites in China from July 19, 2023, through June 11, 2024, and registered as NCT05938920. Seventy-one adults older than 40 years with confirmed, stable IPF were assigned in a 1:1:1:1 ratio to oral rentosertib (搜索) 30 mg once daily, 30 mg twice daily, 60 mg once daily, or placebo for 12 weeks, with 18 patients in each active arm and 17 in the placebo group. All participants continued standard-of-care medications.
The aging analysis was a nested substudy rather than the full randomized cohort. Of 55 participants who completed the trial, 43 consented to the proteomic assessment; one was excluded because week-12 measurements were unavailable, leaving 42 Asian participants with a mean age of 67.1 years. Blood was collected at baseline and at weeks 2, 4, and 12 and analyzed on the Olink Explore 3072 platform, yielding 2,841 proteins after quality-control exclusions.
Six published proteomic aging clocks were applied using standardized implementations: ProtAge, ipfP3GPT, OrganAge chronological, OrganAge mortality, PAOPAC, and PAC. Four were trained primarily to estimate chronological age, while two were trained to predict mortality risk. The clocks were developed by several research groups; according to Insilico, their developers include teams at Harvard, Oxford, Peking University, and Insilico itself. One widely cited model, ProtAge, was built from UK Biobank data and first described in a Nature Medicine study.
Clock Signals Concentrated at Week 4
Across 54 treatment-versus-placebo comparisons covering three active regimens, six clocks, and three post-baseline time points, 21 reached a false-discovery-rate-adjusted q value below 0.10. The signal was concentrated at week 4, when 11 of 18 comparisons showed significantly lower changes in predicted biological age for treated patients. The 30 mg twice-daily regimen produced nine significant comparisons, compared with seven for 60 mg once daily and five for 30 mg once daily.
On the four chronological-age clocks, the 60 mg once-daily group read approximately 2.71 to 3.46 years younger at week 4 than at baseline, a change that was statistically significant versus placebo. The mortality-trained clocks did not show significant changes in that arm. The twice-daily 30 mg regimen was detected by both clock classes and showed reductions across five of six clocks at week 4, producing the broadest agreement across clock types. Organ-specific mortality-based clocks also detected reductions in predicted age for the artery, brain, pancreas, stomach, and immune system in selected treatment groups, though the authors note these models are less established and harder to interpret.
By week 12, the number of significant comparisons had fallen and clock estimates appeared to plateau. Removal of six participants with higher-grade adverse events did not materially alter the findings, and body mass index was not clearly associated with changes in biological age.
Proteomic Shifts and Pathway Signals
Rentosertib (搜索) significantly altered the trajectories of 326 proteins, compared with only two in the placebo group. The 30 mg twice-daily regimen produced the broadest response, including 142 uniquely affected proteins. Changes included reductions in fibrosis- and extracellular matrix-associated proteins and alterations in proteins involved in metabolism and stress resistance. Only 5–9% of proteomic shifts in the 30 mg twice-daily and 60 mg once-daily groups were transient; most responses were classified as sustained or delayed, indicating that underlying proteomic effects persisted through week 12 despite the aging-clock plateau.
Comparison with 55,319 older adults in the UK Biobank showed that rentosertib (搜索)-modulated proteins were enriched 1.74-fold for age-associated proteins. The 30 mg twice-daily regimen showed a pattern opposite to that of normal aging trajectories, whereas the 60 mg once-daily regimen did not. Gene set enrichment analysis indicated generally opposing senescence-associated patterns between placebo and treated groups, and growth-factor signaling pathways involving receptor tyrosine kinases, mitogen-activated protein kinases, RAS, and PI3K–Akt were downregulated across treated groups.
LTBP2 (搜索), described in the paper as a master regulator of fibrosis in the TGF-beta (搜索) pathway, was the only important feature shared by all six clocks. Other influential proteins were associated with inflammation, extracellular-matrix remodeling, fibrosis, and cellular senescence.
Disconnect Between Lung Function and Aging Signal
The 60 mg once-daily group produced the greatest improvement in forced vital capacity (FVC) in the original trial but showed less consistent aging-clock responses than the 30 mg twice-daily group. At 12 weeks, the 60 mg once-daily group had a mean FVC change of 98.4 milliliters, compared with a decline of 20.3 milliliters in the placebo group. The trial's primary endpoint was safety: the proportion of participants experiencing at least one treatment-emergent adverse event was 72.2% in the 30 mg once-daily group, 83.3% in the 30 mg twice-daily group, 83.3% in the 60 mg once-daily group, and 70.6% with placebo. The most common events leading to discontinuation were associated with liver toxicity or diarrhea.
The authors offered the dose-response mismatch as indirect evidence against the explanation that younger-looking blood proteins simply reflect less severe disease, noting that changes in lung capacity explained little of the change in clock readings. They acknowledge that direct testing in people without IPF is still required.
Limitations and Regulatory Path
The authors list substantial limitations: the small sample size, short 12-week follow-up, the computational nature of much of the analysis, the lack of complementary omics data, and the inability to fully separate anti-fibrotic from anti-aging effects. The proteomic cohort included only participants who completed the trial and agreed to additional sampling, so the clock analysis cannot automatically be generalized to all 71 randomized participants. Most proteomic clocks have had little validation in clinical settings, and participants continued standard lung medications that could influence protein levels. The reported q-value threshold of 0.10 is more permissive than the 0.05 threshold commonly used for confirmatory claims.
The authors propose a stepwise path for aging research: collect aging biomarkers as exploratory measures in disease trials, test promising drugs in older adults without the target disease, and pursue formal biomarker qualification or composite clinical endpoints with regulators. They note that under current FDA frameworks, such studies would not be run as trials for an aging indication, and any clinical claim would require conventional clinical outcomes or a qualified biomarker.
Insilico Medicine (搜索) sponsored the Phase 2a trial and supplied rentosertib (搜索). The conflict-of-interest statement identifies founder and CEO Alex Zhavoronkov as first author and other authors as Insilico employees; the remaining authors reported no competing interests. The authors reported no specific funding for the proteomic analysis.
Rentosertib (搜索) remains an experimental treatment for IPF and is not approved for any indication, including aging. Insilico announced on July 7, 2026, that the drug had entered Phase 3 development for IPF in China. The Nature Biotechnology authors said that confirming any broader aging effect would require studies outside IPF, including trials in other age-related diseases or in generally healthy older adults.