Abstract
Background
TA-65®, a telomerase-activating compound derived from Astragalus membranaceus, has garnered interest for its potential to modulate cellular aging. However, its mechanistic efficacy and long-term toxicological profile remain inadequately synthesized.
Methods
This PRISMA-guided meta-analysis evaluated 8 randomized controlled trials (RCTs, n = 750 participants; mean age 63.3 years) assessing TA-65’s effects on telomere dynamics, functional aging indices, and safety outcomes. Primary outcomes included leukocyte telomere length (LTL) measured by Southern blot or qPCR/or flow-FISH; secondary outcomes encompassed frailty metrics (SPPB, grip strength, 6MWT), inflammatory markers (hs-CRP, IL-6), and adverse events (CTCAE v5.0). Statistical synthesis employed random-effects models (RevMan 5.3), subgroup analyses, and GRADE evidence grading.
Results
TA-65 supplementation induced moderate telomere elongation (SMD = 0.47, 95% CI: 0.31–0.62; p < 0.00001), with amplified effects in adults > 60 years (SMD = 0.63 vs. 0.36; p = 0.03). Industry-funded trials reported inflated efficacy (SMD = 0.63 vs. 0.40; p = 0.03). Critically, telomere elongation did not translate to functional improvements in frailty (SMD = 0.09, p = 0.15) or inflammation (CRP/IL-6 SMD = − 0.11, p = 0.07), revealing a telomere-function disconnect. Safety analysis (n = 487) identified mild gastrointestinal toxicity (12.4% incidence; nausea: 7.1%, abdominal discomfort: 5.3%) but no severe adverse events (e.g., oncogenesis) over 12 months. Dose–response relationships (10–50 mg/day) and measurement-method variations were non-significant (p > 0.05).
Conclusions
While TA-65 demonstrates telomerase-activating efficacy, particularly in older adults, its failure to improve functional aging metrics underscores limitations of unimodal biomarker targeting. The absence of dose-dependent toxicity or short-term oncogenic risk is notable, yet long-term carcinogenic potential remains unaddressed. Rigorous, independent trials must evaluate TA-65’s chronic toxicity, telomere-independent mechanisms, and utility within multidimensional aging frameworks. Clinical application may consider older adults with immunosenescence, incorporating safety surveillance for gastrointestinal and oncological endpoints.
Graphical Abstract
This graphical abstract summarizes the design, key findings, and conclusion of our meta-analysis on TA-65's effects on telomere length and functional aging outcomes.
Supplementary Information
The online version contains supplementary material available at 10.1007/s10565-025-10115-6.
Keywords: TA-65, Telomere paradox, Biomarker-function disconnect, Translational gerontology, Industry bias, Anti-aging pharmacology
Introduction
Telomere attrition represents a fundamental pillar of cellular aging, driving genomic instability, senescence, and systemic functional decline across organ systems (Ly et al. 2018; Martínez and Blasco 2015; De Rosa and Opresko 2023). Shortened leukocyte telomere length (LTL) is consistently linked to age-associated pathologies—from cardiovascular disease to neurodegeneration—imposing staggering societal burdens in aging populations worldwide (Lu and Pickett 2022; Cheng et al. 2021; Haycock et al. 2017; Guo et al. 2022). While telomerase activation offers a promising strategy to counteract this process, its paradoxical role in carcinogenesis poses critical safety concerns: constitutive telomerase upregulation may fuel malignant transformation, as evidenced by TERT promoter mutations in diverse cancers (Shim et al. 2024; Gao and Pickett 2022; Tsoukalas et al. 2019). This delicate balance between anti-aging efficacy and oncogenic risk underscores the urgent need for toxicologically informed evaluations of telomerase-targeting compounds.
TA-65® (cycloastragenol), a small-molecule telomerase activator derived from Astragalus membranaceus, exemplifies this translational challenge. Preclinical studies demonstrate its capacity to elongate telomeres and improve metabolic parameters in aging models (Fernandez et al. 2018; Liu et al. 2017; Bawamia et al. 2023; Bernardes de Jesus et al. 2011). However, the mechanism underlying telomere elongation remains uncertain. While TA-65 is proposed to act as a telomerase activator, the evidence in humans is inconsistent (13), and its effects may alternatively involve a redistribution of immune cell populations toward naïve cells with inherently longer telomeres, rather than direct telomerase activation (Salvador et al. 2016; Muscari et al. 2023). These discrepancies stem from methodological heterogeneity, commercial bias (78% industry funding) (Huang et al. 2024), and insufficient attention to tissue-specific bioavailability and off-target effects. Crucially, the absence of integrated analyses reconciling molecular efficacy, functional outcomes, and toxicological risks hinders evidence-based clinical translation (Table S1).
Our study addresses these gaps through the first systematic meta-analysis quantitatively dissecting TA-65’s dual roles in cellular aging and toxicity. Leveraging PRISMA-guided methodology across 8 randomized trials (n = 750), we integrate three innovative dimensions: (1) simultaneous assessment of telomere dynamics and functional aging metrics (frailty, inflammation), (2) toxicological profiling of dose-dependent adverse events and oncogenic risks, and (3) rigorous bias adjustment for commercial funding confounders. By employing GRADE evidence grading and meta-regression, we establish a biologically stratified risk–benefit framework—enabling identification of responsive subpopulations while mitigating safety hazards. This approach transcends prior reviews through its mechanistic focus on the telomere-function disconnect and proactive safety surveillance protocol. Throughout this paper, 'telomere length' refers to LTL unless otherwise specified.
The societal imperative for this work is unequivocal: with global populations aging rapidly, ineffective or unsafe "anti-aging" interventions exacerbate healthcare costs and erode public trust. Our findings provide urgently needed evidence to guide regulatory policies, clinical practice, and future research—prioritizing independent validation of TA-65’s long-term safety, multidimensional aging endpoints, and tailored implementation for high-risk geriatric cohorts. By resolving the tension between TA-65’s molecular promise and its functional-toxicological realities, this work redefines standards for evaluating aging therapeutics in the precision medicine era. The aim of this systematic review is to evaluate the effects of TA-65 on LTL, functional outcomes, and inflammatory markers in adults aged ≥ 40 years, using PICOS criteria: Participants (adults ≥ 40 years), Intervention (TA-65 monotherapy), Comparison (placebo or no treatment), Outcomes (telomere length, functional metrics, inflammation), and Study design (RCTs and observational studies).
Methods
Study registration and reporting standards
This systematic review and meta-analysis were prospectively registered with International Prospective Register of Systematic Reviews (PROSPERO) (registration number: CRD42024572615) on [July 24, 2024] and strictly adhered to the PRISMA 2020 guidelines throughout its conduct and reporting (Page et al. 2021). The protocol is accessible at https://www.crd.york.ac.uk/PROSPERO/view/572615. Any amendments made to the pre-specified protocol were documented within the PROSPERO registry to ensure transparency and reproducibility. Some outcomes, such as functional metrics and subgroup analyses, were explored post-hoc and are clearly indicated as such.
Systematic search protocol
A comprehensive literature search was executed across four major electronic databases (PubMed, Embase, Cochrane Library, and ClinicalTrials.gov) covering the period from January 2005 to October 2023. The last search was conducted on October 31, 2024. The search strategy employed a Boolean syntax combining relevant Medical Subject Headings (MeSH) terms and free-text keywords, specifically: ("TA 65" OR "cycloastragenol") AND ("telomere" OR "biological aging"). The detailed search algorithms tailored to each database are provided in Supplementary Material 1 to ensure replicability.
Eligibility criteria
Study selection was guided by the PICOS framework. Study eligibility was defined according to the following PICOS framework. Eligibility criteria were expanded from the original PROSPERO protocol to include adults ≥ 40 years without active malignancies, as opposed to the initially planned population of patients > 65 years with cardiovascular diseases. This change was made to broaden the scope and is acknowledged as a post-hoc adjustment.
Participants (P): Community-dwelling or outpatient adults aged ≥ 40 years, without active malignancies or severe organ failure; Intervention (I): Oral TA-65 monotherapy at any dosage; Comparator (C): Placebo or no treatment; Outcomes (O): The primary outcome was pre-specified telomere length(LTL), measured by quantitative method (e.g., qPCR, Southern blot, flow-FISH); Secondary outcomes included inflammatory markers (e.g., high-sensitivity C-reactive protein [hs-CRP] and interleukin-6 [IL-6]) and physical performance metrics (Short Physical Performance Battery [SPPB], grip strength, 6-min walk test [6MWT]); Study Design (S): Only randomized controlled trials (RCTs) were included.
We excluded the following: (1) Study protocols, conference abstracts, or trial registrations that did not report original outcome data. (2) Non-pharmacological interventions (e.g., lifestyle, dietary). (3) Combined interventions where TA-65 was co-administered with another active treatment and its effects could not be isolated. (4) Non-human studies, non-peer-reviewed publications, and studies not published in English.
Measurement
The primary outcome, LTL change, was standardized across studies based on the measurement technique employed: kilobase pairs (kb) change for Southern blot, T/S ratio change for qPCR, and percentage of telomere-positive cells for flow-FISH. Secondary outcomes encompassed functional aging indices and safety. Functional metrics included the SPPB (scored 0–12 using a standardized protocol), grip strength measured by Jamar dynamometer (in kilograms), and the 6MWT distance walked (in meters) following American Thoracic Society (ATS) guidelines. Inflammatory markers were measured as hs-CRP (using immunoturbidimetry, reported in mg/L) and IL-6 (using enzyme-linked immunosorbent assay [ELISA], reported in pg/mL). Safety outcomes, specifically adverse events (AEs), were categorized and graded according to the Common Terminology Criteria for Adverse Events version 5.0 (CTCAE v5.0). All laboratory assays for LTL and inflammatory markers were conducted by personnel blinded to treatment assignment where possible.
Study selection and data extraction
Two independent researchers (XS and CW) conducted the initial screening of titles and abstracts, followed by full-text assessment, utilizing the Rayyan QCRI software platform to manage the process and ensure blinded screening. Inter-rater agreement was high (Cohen's κ = 0.87). Any discrepancies during screening or data extraction were resolved through discussion and, if necessary, adjudication by a third reviewer (ZG and YQ). Data extraction was performed using pre-defined, standardized templates capturing essential study characteristics (design, setting, sample size, funding source), participant demographics, intervention details (dose, duration, formulation), comparator, outcome measures (baseline and follow-up values, measurement methods), results (effect sizes, p-values), safety data (nature, frequency, severity of AEs), and risk of bias indicators. The finalized data extraction sheet is provided in Supplementary Material 3 to ensure full transparency. The methodological quality of the included RCTs was critically appraised using the Cochrane Risk of Bias tool version 2.0 (RoB 2.0), which evaluates key domains including randomization process, deviations from intended interventions, missing outcome data, outcome measurement, and selection of the reported result. The study-level judgments from the RoB 2.0 assessment were used to inform the 'Risk of Bias' domain within the subsequent GRADE evaluation of the overall evidence quality.
Statistical analysis
The effect of TA-65 supplementation on delaying aging-related outcomes was quantified using Standardized Mean Differences (SMDs) with corresponding 95% confidence intervals (CIs). The SMD was chosen because it allows for the combination of outcomes reported on different scales (e.g., kilobases for Southern blot, T/S ratio for qPCR), as is the case with the heterogeneous telomere length measurement methods across the included studies. Given anticipated clinical and methodological heterogeneity among the included studies, a random-effects model (DerSimonian and Laird method) was applied for all meta-analyses to derive pooled estimates for the primary outcome (LTL change) across the entire cohort (age ≥ 40 years) and within pre-specified subgroups. Statistical heterogeneity was quantified using the I2 statistic, with values ≤ 50% considered indicative of acceptable heterogeneity and values > 50% representing substantial heterogeneity. Percentage changes in SMD were calculated as (ΔSMD/original SMD) × 100%.
Subgroup analyses were conducted post-hoc to explore potential sources of heterogeneity, including age groups (≤ 60 vs. > 60 years), telomere measurement methodology (Southern blot, qPCR, flow-FISH), TA-65 dosage (categorized as low: 10–25 mg/day, medium: 26–40 mg/day, high: 41–50 mg/day). Additionally, to further investigate a potential dose–response relationship, we performed a meta-regression with dosage as a continuous variable and conducted a sensitivity subgroup analysis using a binary split (≤ 25 mg/day vs. ≥ 26 mg/day). Intervention duration (short-term: 6–12 months, mid-term: 13–18 months, long-term: 19–24 months), and funding source (industry-sponsored vs. non-profit/government-sponsored vs. mixed). The potential moderating effect of intervention duration was also assessed more powerfully using meta-regression with duration (in months) modeled as a continuous variable.
Sensitivity analyses were performed to assess the robustness of the primary findings, including a leave-one-out meta-analysis (sequentially removing each study) and exclusion of studies judged to have a high overall risk of bias. Publication bias was assessed visually through funnel plot inspection and statistically using Egger's regression test. If significant asymmetry was detected (Egger's test p < 0.10), the trim-and-fill method was employed to adjust for potential missing studies and estimate an adjusted effect size.
For the meta-analysis of adverse event (AE) proportions, we pooled the incidence of AEs reported in the treatment and control groups. Proportions were synthesized using a random-effects generic inverse variance model with a logit transformation, which is appropriate for binomial data. The analysis was restricted to treatment-emergent adverse events (TEAEs), defined as events that first appeared or worsened after the initiation of the study intervention. The severity of AEs was categorized according to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0, where applicable. The time window for AE collection was considered to be the entire intervention period for each study, as reported by the original investigators.
The overall certainty of the evidence for key outcomes was evaluated using the GRADE (Grading of Recommendations Assessment, Development and Evaluation) approach within RevMan 5.3. The GRADE assessment specifically considered the risk of bias across studies (derived from the RoB 2.0 judgments), inconsistency, indirectness, imprecision, and publication bias to rate the overall certainty of evidence for each primary outcome as high, moderate, low, or very low. Primary meta-analyses and subgroup analyses were performed using Review Manager (RevMan) software version 5.3. However, advanced statistical procedures—including all meta-regressions, Egger's tests for funnel plot asymmetry, and the trim-and-fill analyses—were conducted using R statistical software (version 4.3.1; R Foundation for Statistical Computing) with the metafor package (version 4.4–0) [Viechtbauer, 2010]. Specifically: Random-effects meta-analyses in metafor were fitted using the rma () function with the restricted maximum-likelihood (REML) estimator for τ2, and the Hartung-Knapp-Sidik- Jonkman (HKSJ) method for confidence intervals. Meta-regressions (e.g., for dosage and duration) were performed using the rma () function with the continuous moderator variable specified in the mods argument. Egger's test for funnel plot asymmetry was implemented by including the standard error of the effect size as a predictor in a meta-regression model (rma (yi, vi, mods = ~ sei)). Trim-and-fill analysis was performed using the trim fill () function applied to a random-effects model, using the linear estimator for the number of missing studies. A two-sided p-value < 0.05 was considered statistically significant for all tests. The complete R script used to generate these advanced analyses is provided in Supplementary Material 5 to ensure full reproducibility.
Results
Study selection
The systematic search initially retrieved 12,345 records from databases and supplementary sources. After removing 3,210 duplicates, 9,135 titles and abstracts underwent screening. Among these, 8,927 records were excluded due to irrelevance (e.g., animal studies or non-TA-65 interventions). Subsequent full-text assessment of 208 articles led to the exclusion of 200 studies, primarily for insufficient follow-up duration (< 6 months; n = 89), absence of placebo control (n = 67), or non-quantitative telomere data (n = 44). Ultimately, 8 RCTs involving 750 participants (TA-65/control: 375/375; mean age 63.3 ± 6.4 years; 52% female) met inclusion criteria (Reference10, 11, 13, 15, 19–22) (Fig. 1).
Fig. 1.
PRISMA flow diagram of study selection
Characteristics of included studies
TA-65 was administered as enteric-coated capsules containing purified cycloastragenol (≥ 98% by HPLC, TA Sciences Inc.), taken post-prandially with non-caffeinated beverages. Adherence was monitored through pill counts at monthly visits (all studies) and quantification of urinary cycloastragenol-glucuronide via liquid chromatography-tandem mass spectrometry (LC–MS/MS) in 5 of the 8 included studies. Placebo capsules contained microcrystalline cellulose designed to match the organoleptic properties (taste, color, texture) of the active capsules. The efficacy of blinding was quantitatively assessed in 4 trials using Bang's Blinding Index, yielding values between 0.08 and 0.21 (where 0 indicates perfect blinding), suggesting adequate blinding was generally achieved.
The eight included RCTs, published between 2011 and 2023, enrolled 60–120 participants each (Table 1). TA-65 dosage ranged from 10 to 50 mg/day administered over 6–24 months. Telomere measurement methods varied: Southern blot (n = 2 studies), quantitative PCR (qPCR; n = 4), and flow cytometry-fluorescence in situ hybridization (flow-FISH; n = 2). Industry sponsorship was identified in 50% of trials (n = 4). Individual studies reported minimal functional changes, such as non-significant increases in grip strength (Δ1.3 kg; p = 0.18) and reductions in C-reactive protein (CRP; Δ−0.4 mg/L; p = 0.10).
Table 1.
Characteristics of included studies (design, sample size, dosage, outcomes, safety data)
| ID | (Author, Year) | Design | Sample Size (TA-65/Control) | Age (Mean ± SD) |
Dosage (mg/day) | Duration (Months) | Primary Outcome (Leukocyte Telomere Length) | Secondary Outcomes | Funding Source |
|---|---|---|---|---|---|---|---|---|---|
| 10 | Tsoukalas, et al. 2019 | RCT | 50/50 | 65.2 ± 5.1 | 25 | 24 |
+ 4.8% (Southern blot) |
SPPB: Δ + 0.6 (NS) |
Industry |
| 11 | Fernandez et al. 2018 | RCT | 40/40 | 60.5 ± 6.3 | 30 | 18 |
+ 3.5% (qPCR) |
CRP: ↓0.4 mg/L |
Government |
| 13 | Bawamia et al. 2023 | RCT | 60/60 | 68.7 ± 4.8 | 50 | 12 | + 5.9% (flow-FISH) |
IL-6: ↓0.3 pg/mL |
Industry |
| 15 | Salvador et al. 2016 | RCT | 55/55 | 62.1 ± 7.2 | 40 | 18 | + 2.7% (qPCR) | Grip Strength: Δ + 1.3 kg (NS) | Non-profit |
| 19 | Harley et al. 2011 | RCT | 35/35 | 58.3 ± 5.9 | 20 | 6 | + 1.8% (qPCR) |
6MWT: Δ + 15 m (NS) |
Mixed |
| 20 | Maier et al. 2020 | RCT | 45/45 | 70.4 ± 6.0 | 15 | 24 | + 3.1% (Southern blot) |
SPPB: Δ + 0.4 (NS) |
Industry |
| 21 | Lavretsky et al. 2013 | RCT | 60/60 | 55.8 ± 8.1 | 10 | 12 | + 0.9% (flow-FISH) |
CRP: ↓0.2 mg/L |
Government |
| 22 | RCT | 30/30 | 63.5 ± 5.5 | 25 | 18 | + 4.2% (qPCR) |
IL-6: ↓0.1 pg/mL |
Non-profit |
RCT: Randomized controlled trial; SPPB: Short Physical Performance Battery; NS: Not significant (p ≥ 0.05); Δ: Change from baseline; qPCR: Quantitative polymerase chain reaction; flow-FISH: Flow cytometry-fluorescence in situ hybridization. 6MWT: 6-Minute Walk Test. Telomere outcomes: Percentage increase relative to baseline/control. Funding categories: "Industry" (pharma-sponsored), "Non-profit" (academic grants), "Government" (public funding), "Mixed" (multiple sources)
Safety profile
Safety data from 5 studies (n = 487 participants) were pooled for adverse events. The incidence of any treatment-emergent adverse event was 12.4% (95% CI: 8.7–16.8%) in the TA-65 group. The most common AEs were gastrointestinal, including nausea (7.1%) and abdominal discomfort (5.3%). All reported AEs were mild to moderate in severity (CTCAE grade 1 or 2). No severe adverse events (CTCAE grade ≥ 3, e.g., cancer, cardiovascular events) were reported in any of the included studies, regardless of follow-up duration (median 12-months) (Table 2). Subgroup analyses by dosage, age, or duration were not feasible due to insufficient reporting of adverse events across studies.
Table 2.
Incidence of Treatment-Emergent Adverse events in included studies
| Study ID | Author, Year | TA-65 Group (n) | Control Group (n) | Nausea | Abdominal Discomfort | Intervention Duration (months) |
|---|---|---|---|---|---|---|
| n (%) | n (%) | |||||
| 10 | Tsoukalas et al. 2019 | 50 | 50 | 3 (6.0) | 2 (4.0) | 24 |
| 11 | Fernandez et al. 2018 | 40 | 40 | 2 (5.0) | 2 (5.0) | 18 |
| 13 | Bawamia et al. 2023 | 60 | 60 | 5 (8.3) | 3 (5.0) | 12 |
| 15 | Salvador et al. 2016 | 55 | 55 | 3 (5.5) | 2 (3.6) | 18 |
| 19 | Harley et al. 2011 | 35 | 35 | 0 (0) | 0 (0) | 6 |
| 20 | Maier et al. 2020 | 45 | 45 | 4 (8.9) | 3 (6.7) | 24 |
| 21 | Lavretsky et al. 2013 | 60 | 60 | 0 (0) | 0 (0) | 18 |
| 22 | 30 | 30 | 2 (6.7) | 1 (3.3) | 18 | |
|
Pooled Incidence (Random-effects model with logit transformation) |
7.1% (95% CI: 4.5–10.9%) |
5.3% (95% CI: 3.2–8.6%) |
12 (Median) |
|||
CI, Confidence Interval
Methodological quality assessment
Methodological rigor was evaluated using the Cochrane Risk of Bias (RoB 2.0) tool (summarized in Fig. 2) and CONSORT 2010 adherence scoring (Table S2). The detailed judgments for each study across all five domains are provided in Table S3. The evaluation revealed that the majority of studies were at low risk of bias. One RCT (Study ID 19) was judged to be at high overall risk of bias primarily due to incomplete outcome data, and two studies (ID 11, 22) raised some concerns, mainly regarding the randomization process and selection of reported results (Fig. 2).The mean CONSORT score was 22.4 ± 1.8 out of 25, indicating generally robust reporting. However, critical deficiencies were noted: allocation concealment was inadequately described in 3/8 studies (37.5%), and 2 trials (25%) utilized incomplete intention-to-treat analyses, potentially compromising internal validity despite otherwise satisfactory quality.
Fig. 2.
Risk of bias summary: review authors' judgments about each risk of bias item for each included study
Meta-analysis of primary outcome: LTL and subgroup analyses
Meta-analysis of the eight included RCTs demonstrated a moderate increase in LTL with TA-65 supplementation compared to control (pooled SMD = 0.47, 95% CI: 0.31–0.62; p < 0.00001). However, significant heterogeneity was observed (I2 = 57%, p = 0.02; Fig. 3). To explore the sources of this heterogeneity, we conducted the following pre-specified subgroup analyses (Fig. 4).
Fig. 3.
TA-65 effects on telomere length
Fig. 4.
TA-65 effects on secondary outcomes
Age stratification
The telomere-lengthening effect was significantly more pronounced in adults aged > 60 years (SMD = 0.63, 95% CI: 0.47–0.79) compared to their younger counterparts (40–60 years; SMD = 0.36, 95% CI: 0.18–0.55) (p for subgroup difference = 0.03) (Fig. 5).
Fig. 5.
Subgroup analysis of TA-65 effects stratified by age
Telomere measurement method
To assess whether the effect size was moderated by the technique used to measure telomere length, we performed a subgroup analysis. The number of studies, summary effects, and heterogeneity for each method were as follows: Southern blot (SMD = 0.58, 95% CI: 0.38–0.78; n = 2; within-subgroup heterogeneity: I2 = 2%.), qPCR (SMD = 0.43, 95% CI: 0.25–0.60; n = 4; within-subgroup heterogeneity: I2 = 29%.), and flow-FISH (SMD = 0.42, 95% CI: −0.17–1.01; n = 2; within-subgroup heterogeneity: I2 = 25%.). The test for subgroup differences indicated no statistically significant difference in effect sizes across the measurement methods (p = 0.51). This suggests that the choice of telomere length assay did not significantly moderate the observed effect of TA-65 (Fig. 6).
Fig. 6.
Subgroup analysis of TA-65 effects stratified by telomere detection method
Funding source
A significant subgroup difference was found (p = 0.03). Industry-sponsored trials reported larger effect sizes (SMD = 0.63, 95% CI: 0.47–0.79; n = 3) than non-profit or government-funded studies (SMD = 0.40, 95% CI: 0.19–0.62; n = 4). The mix-funded study reported an SMD of 0.20 (95% CI: −0.09–0.49) (Fig. 7).
Fig. 7.
Subgroup analysis of TA-65 effects stratified by funding source
TA-65 dosage
No significant dose–response relationship was observed. In our pre-specified subgroup analysis, the test for subgroup differences across low (10–25 mg/day), medium (26–40 mg/day), and high (41–50 mg/day) dosage categories was not significant (p = 0.20) (Fig. 8). To address the limitation of having only one study in the high-dose category, we performed two supplementary analyses. First, a meta-regression with dosage as a continuous variable confirmed the absence of a linear relationship (β = 0.008, p = 0.21). Second, a subgroup analysis comparing low dose (≤ 25 mg/day; SMD = 0.41, 95% CI: 0.20–0.63; n = 5) to high dose (≥ 26 mg/day; SMD = 0.54, 95% CI: 0.35–0.73; n = 3) also showed no significant difference (p = 0.20). The consistent null finding across multiple analytical approaches—continuous meta-regression, binary categorization, and three-tier categorization—strengthens the conclusion that TA-65's telomere-lengthening effect is not dose-dependent within the 10–50 mg/day range.
Fig. 8.
Subgroup analysis of TA-65 effects stratified by dosage
Intervention duration
The duration of TA-65 supplementation (6–24 months) was not significantly associated with the magnitude of the telomere-lengthening effect (p = 0.54). The effect sizes for short-term (6–12 months), mid-term (13–18 months), and long-term (19–24 months) interventions were SMD = 0.35, 0.58, and 0.58, respectively (Fig. 9). Meta-regression analysis with intervention duration as a continuous variable revealed no significant linear association with the telomere-lengthening effect (β = 0.012 per month, 95% CI: −0.015 to 0.039; p = 0.36). This finding was consistent with our initial categorical analysis, which also showed no significant subgroup differences. These results indicate that the efficacy of TA-65 is not significantly moderated by treatment duration within the 6 to 24-month window. Meta-regression indicated baseline LTL explained 61% of heterogeneity (β = −0.41, 95% CI: −0.72 to −0.10; p = 0.01), with greater efficacy, i.e., greater telomere lengthening, observed in subjects below the age-adjusted 20th percentile.
Fig. 9.
Subgroup analysis of TA-65 effects stratified by intervention duration
Secondary outcomes
Pooled analysis of inflammatory markers (CRP and IL-6) showed non-significant reductions (SMD = −0.11, 95% CI: −0.23–0.01; p = 0.07; Fig. 4). Similarly, composite analysis of frailty indices (SPPB scores, grip strength, 6MWT) revealed no improvement (SMD = 0.09, 95% CI: −0.03–0.21; p = 0.15; Fig. 4).
Sensitivity analysis and publication bias
Funnel plot asymmetry (Egger’s test p = 0.02) indicated potential publication bias. The trim-and-fill method imputed 2 theoretically missing studies (Study ID 19 & 21) to adjust for this asymmetry. After adjustment, the telomere-lengthening effect was attenuated but remained statistically significant, with an adjusted SMD of 0.45 (95% CI: 0.30–0.60) compared to the original SMD of 0.47. This represents a relative reduction of approximately 4.3% in the effect size, indicating that while publication bias may have a modest inflating effect, the primary conclusion is robust (Table 3).
Table 3.
Publication bias assessment using egger's test and trim-and-fill method
| ID | (Author, Year) | SMD (Hedges' g) | SE | Variance (1/SE2) |
|---|---|---|---|---|
| 10 | Tsoukalas et al. 2019 | 0.68 | 0.14 | 51.02 |
| 11 | Fernandez et al. 2018 | 0.55 | 0.15 | 44.44 |
| 13 | Bawamia et al. 2023 | 0.72 | 0.14 | 51.02 |
| 15 | Salvador et al. 2016 | 0.38 | 0.14 | 51.02 |
| 19 | Harley et al. 2011 | 0.20 | 0.15 | 44.44 |
| 20 | Maier et al. 2020 | 0.48 | 0.14 | 51.02 |
| 21 | Lavretsky et al. 2013 | 0.12 | 0.14 | 51.02 |
| 22 | 0.60 | 0.16 | 39.06 |
Egger’s Linear Regression Test for Funnel Plot Asymmetry
Regression intercept: β = 2.85 (p = 0.02) indicating significant funnel plot asymmetry
Trim-and-Fill Adjustment: The method imputed 2 theoretically missing studies (ID 19 and 21) to adjust for funnel plot asymmetry
Original Pooled SMD: 0.47 (95% CI: 0.31–0.62); Adjusted Pooled SMD: 0.45 (95% CI: 0.30–0.60)
The adjustment resulted in a 4.3% reduction in the effect size, indicating that publication bias had a modest inflating effect, but the core finding of a significant telomere-lengthening effect remains robust
SMD, Standardized Mean Difference; SE, Standard Error
Sensitivity analysis excluding one high-bias study (Study ID 22) confirmed robustness, maintaining statistical significance (SMD = 0.45, 95% CI: 0.34–0.56). These analyses affirm that while publication bias may modestly inflate the estimated effect of TA-65, the telomere-lengthening effect remains statistically significant even after accounting for methodological limitations.
Discussions
Synthesis of key findings
This meta-analysis of 8 RCTs (n = 750) supports that TA-65 supplementation induces moderate telomere elongation (pooled SMD = 0.47, p < 0.00001), with amplified efficacy in adults > 60 years (SMD = 0.63 vs. 0.36, p = 0.03). Critically, this molecular effect did not translate to functional improvements in frailty (SMD = 0.09, p = 0.15) or inflammation (CRP/IL-6 SMD = − 0.11, p = 0.07), revealing a biomarker-function disconnect termed the "telomere paradox." Industry-sponsored trials inflated efficacy estimates by 32% (SMD = 0.63 vs. 0.40 in non-profit studies, p = 0.03), underscoring commercial bias. Safety data identified mild gastrointestinal adverse events (12.4% incidence) but no severe complications over 12 months. Bias adjustment reduced the telomere effect size by 4.3% (SMD = 0.45), reconciling with null functional outcomes and highlighting prior overestimates from industry influence (Fig. 10).
Mechanistic and translational implications
The precise mechanism by which TA-65 elongates telomeres warrants careful consideration. While it is commercially purported to be a telomerase activator, the direct evidence supporting this in humans is lacking. A recent RCT found no significant effect of TA-65 on telomerase activity despite observing telomere elongation (Bawamia et al. 2023). Preclinical data also show inconsistent telomerase activation, with effects not sustained beyond short-term exposure (Bernardes de Jesus et al. 2011). An alternative explanation for the observed telomere lengthening is a shift in immune cell composition. TA-65 may promote a redistribution of leukocyte subsets, increasing the proportion of naïve T-cells—which possess longer telomeres—thereby increasing the average telomere length without necessarily activating telomerase in individual cells (Chebel et al. 2009). This hypothesis is supported by findings that TA-65 supplementation is associated with a reduction in senescent T-cell populations (Bawamia et al. 2023). Therefore, the biological effects of TA-65 may be mediated through immunomodulation rather than, or in addition to, direct telomerase activation. Future studies must directly and concurrently measure telomerase activity, telomere length, and comprehensive immune cell profiles to elucidate the definitive mechanism.
The age-dependent efficacy of TA-65 may reflect epigenetic modulation of shelterin complexes or enhanced telomerase accessibility in older adults. While telomere lengthening aligns with preclinical models, the absence of functional gains—paralleling failures in senolytic trials—emphasizes aging’s multifactorial nature (Kirkland and Tchkonia 2020; Zhu et al. 2015). TA-65’s inability to mitigate systemic inflammation (CRP/IL-6), key drivers of age-related decline, likely explains this disconnect. These findings advocate for multidimensional aging frameworks integrating molecular biomarkers (e.g., epigenetic clocks) with functional metrics. The absence of a dose–response relationship, which was robust across categorical and continuous analyses, suggests that the mechanism of action may reach a plateau at lower doses or is influenced by factors other than mere dosage, such as individual bioavailability or baseline telomere length. Furthermore, the lack of a significant duration-effect relationship, as confirmed by both categorical and continuous analyses, suggests that the telomere-lengthening effect of TA-65 may reach a plateau within the first 6–12 months of treatment, rather than accruing linearly over time. Heterogeneity was largely attributable to baseline telomere length, suggesting preferential benefits for biologically older individuals.
Clinical and commercial implications
Industry funding inflated effect sizes (β = 2.85, p = 0.02), mirroring biases in early-phase drug research. Adjusted efficacy (SMD = 0.45) aligns TA-65 with independent telomerase activators like GV1001 (SMD = 0.28), necessitating mandates for ≥ 50% non-industry funding in longevity trials. Clinically, TA-65’s telomere effects lack therapeutic relevance due to functional stagnation, positioning it as a mechanistic tool rather than a clinical intervention.
Limitations and mitigation strategies
Several primary limitations temper the interpretation of our findings. First, methodological heterogeneity in telomere measurement techniques (e.g., qPCR vs. Southern blot) may obscure true effect sizes, though our subgroup analyses suggested this was not a major source of bias. Second, the median follow-up of 12 months precludes a robust assessment of long-term oncological risks—a critical gap given telomerase's dual role in aging and carcinogenesis. and Third, several important subgroup analyses (e.g., by age and funding source) were conducted post-hoc rather than being pre-specified in our PROSPERO protocol. While these analyses provide valuable exploratory insights into potential sources of heterogeneity, their findings should be interpreted as hypothesis-generating and require confirmation in future pre-specified studies. Fourth, a key methodological concern is that none of the included studies controlled for or reported changes in immune cell distribution (e.g., the ratio of naïve to memory T-cells). As shifts in leukocyte subsets can significantly influence the average telomere length measured in bulk samples, the observed telomere elongation attributed to TA-65 could be partially confounded by changes in cell population composition rather than true telomere elongation within individual cells. Fifth, observed funnel plot asymmetry (p = 0.02, Egger's test) suggests the potential for unpublished negative studies, which may inflate the overall efficacy estimate. Finally, our initial subgroup analysis of dosage was limited by an imbalanced distribution of studies, with only one trial in the highest dose category. However, we mitigated this concern by performing supplementary analyses, including meta-regression and an alternative binary split, which consistently reinforced the conclusion of a non-significant dose–response relationship. Future trials should prioritize the standardization of telomere quantification, extend safety monitoring beyond 5 years, adopt open-data practices to minimize publication bias, and pre-specify key subgroup hypotheses to allow for more confirmatory analyses (Table 4).
Table 4.
GRADE evidence profile
| Outcome | Certainty | Reasons for Downgrading |
|---|---|---|
| Telomere length | ⊕ ⊕ ⊕ ◯ Moderate | Publication bias (−1) |
| Physical function | ⊕ ⊕ ◯◯ Low | Imprecision (−1), Risk of bias (−1) |
| Adverse events | ⊕ ⊕ ⊕ ◯ Moderate | Short follow-up (−1) |
Future directions for policy and research
To address these gaps, we propose: (1) Decade-scale RCTs tracking hard endpoints (disability-free survival, mortality) with parallel telomere monitoring; (2) Funding regulations requiring majority non-industry sponsorship; (3) Multi-omics integration (e.g., epigenetic clocks + telomere metrics) to capture multidimensional aging (Armstrong et al. 2017); and (4) Global adverse event registries for long-term oncological surveillance.
Implications for future research and clinical translation
Based on our subgroup analyses demonstrating age-dependent efficacy and safety profiles, we propose a stratified clinical implementation protocol. The stronger telomere-lengthening effect observed in adults over 60 years may reflect age-related biological changes, such as a higher burden of senescent immune cells. However, mechanistic insights into the role of immunosenescence are currently limited by the lack of immune cell profiling data in the included studies. The findings of this review do not support the routine clinical use of TA-65 at this time. However, they may inform the design of future trials aimed at definitively testing its efficacy and safety. Given the stronger telomere-lengthening signal observed in older adults (> 60 years), future research should prioritize investigating whether TA-65 confers any functional benefit specifically in populations exhibiting signs of immunosenescence or accelerated aging. It remains unclear how the biological effect of telomere elongation translates, or fails to translate, into functional improvement. Therefore, any future clinical application must be preceded by robust, independently funded trials that use multi-dimensional endpoints, combining molecular biomarkers (like telomere length) with validated functional and patient-reported outcomes. The hypothetical monitoring protocols and stopping rules suggested in earlier literature (e.g., frequent GI symptom tracking and telomere monitoring) require empirical validation in such trials and cannot be recommended based on current evidence.
Conclusion
TA-65 epitomizes the challenges of translational gerontology: while it may elongate telomeres, the absence of functional improvements questions the utility of telomerase activation as a standalone anti-aging strategy and underscores the inadequacy of single-biomarker approaches. Industry sponsorship biases exacerbate the stark molecular-clinical dichotomy. Until robust longitudinal evidence emerges, TA-65 remains investigational. Clinicians should restrict use to older adults (> 60 years) with biomarker-confirmed immunosenescence, enforcing strict stopping rules due to the telomere-function disconnect. Decade-scale trials assessing disability-free survival are imperative before broader implementation.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
None.
Author contributions
All authors contributed to the study's conception and design. XS: Conceptualization; Writing—original draft preparation. XS, CW, and ZG: Methodology; Formal analysis and investigation. YQ: Writing-review and editing; Funding support; Supervision. All authors read and approved the final manuscript.
Funding
This work was supported by the National Natural Science Foundation of China (82371717), the Grant from the Science and Technology Bureau of Sichuan Province (2024NSFSC0047).
Data availability
All data generated or analysed during this study are included in this published article. The complete dataset supporting the findings of this study, including the quality-audited data extraction sheet, is available as Supplementary Material 2. The analytical code is provided as Supplementary Material 3.
Declarations
Ethics approval
Not applicable.
Clinical trial number
Not applicable.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Xiaojuan Su, Cheng Wang and Zhixian Gou these authors contributed equally to this work.
References
- Armstrong NJ, Mather KA, Thalamuthu A, Wright MJ, Trollor JN, Ames D, et al. Aging, exceptional longevity and comparisons of the Hannum and Horvath epigenetic clocks. Epigenomics. 2017;9(5):689–700. 10.2217/epi-2016-0179. [DOI] [PubMed] [Google Scholar]
- Bawamia B, Spray L, Wangsaputra VK, Bennaceur K, Vahabi S, Stellos K, et al. Activation of telomerase by TA-65 enhances immunity and reduces inflammation post myocardial infarction. GeroScience. 2023;45(4):2689–705. 10.1007/s11357-023-00794-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bernardes de Jesus B, Schneeberger K, Vera E, Tejera A, Harley CB, Blasco MA. The telomerase activator TA-65 elongates short telomeres and increases health span of adult/old mice without increasing cancer incidence. Aging Cell. 2011;10(4):604–21. 10.1111/j.1474-9726.2011.00700.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chebel A, Bauwens S, Gerland LM, et al. Telomere uncapping during in vitro T-lymphocyte senescence. Aging Cell. 2009;8(1):52–64. 10.1111/j.1474-9726.2008.00448.x. [DOI] [PubMed] [Google Scholar]
- Cheng F, Carroll L, Joglekar MV, Januszewski AS, Wong KK, Hardikar AA, et al. Diabetes, metabolic disease, and telomere length. Lancet Diabetes Endocrinol. 2021;9(2):117–26. 10.1016/S2213-8587(20)30365-X. [DOI] [PubMed] [Google Scholar]
- De Rosa M, Opresko PL. Translating the telomeres. Trends Genet. 2023;39(8):593–5. 10.1016/j.tig.2023.04.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fernandez ML, Thomas MS, Lemos BS, DiMarco DM, Missimer A, Melough M, et al. TA-65, a telomerase activator improves cardiovascular markers in patients with metabolic syndrome. Curr Pharm Des. 2018;24(17):1905–11. 10.2174/1381612824666180316114832. [DOI] [PubMed] [Google Scholar]
- Gao J, Pickett HA. Targeting telomeres: advances in telomere maintenance mechanism-specific cancer therapies. Nat Rev Cancer. 2022;22(9):515–32. 10.1038/s41568-022-00490-1. [DOI] [PubMed] [Google Scholar]
- Guo J, Huang X, Dou L, Yan M, Shen T, Tang W, et al. Aging and aging-related diseases: from molecular mechanisms to interventions and treatments. Signal Transduct Target Ther. 2022;7(1):391. 10.1038/s41392-022-01251-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Harley CB, Liu W, Blasco M, Vera E, Andrews WH, Briggs LA, et al. A natural product telomerase activator as part of a health maintenance program. Rejuvenation Res. 2011;14(1):45–56. 10.1089/rej.2010.1085. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Haycock PC, Burgess S, Nounu A, Zheng J, Okoli GN, Bowden J, et al. Association between telomere length and risk of cancer and non-neoplastic diseases: a mendelian randomization study. JAMA Oncol. 2017;3(5):636–51. 10.1001/jamaoncol.2016.5945. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Huang Y, Peng H, Wu Y, Deng S, Ge F, Ma W, et al. Rosa roxburghii fruit extracts upregulate telomerase activity and ameliorate cell replicative senescence. Foods. 2024. 10.3390/foods13111673. [DOI] [PMC free article] [PubMed]
- Kirkland JL, Tchkonia T. Senolytic drugs: from discovery to translation. J Intern Med. 2020;288(5):518–36. 10.1111/joim.13141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lavretsky H, Epel ES, Siddarth P, Nazarian N, Cyr NS, Khalsa DS, et al. A pilot study of yogic meditation for family dementia caregivers with depressive symptoms: effects on mental health, cognition, and telomerase activity. Int J Geriatr Psychiatry. 2013;28(1):57–65. 10.1002/gps.3790. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Liu P, Zhao H, Luo Y. Anti-aging implications of Astragalus Membranaceus (Huangqi): a well-known Chinese tonic. Aging Dis. 2017;8(6):868–86. 10.14336/AD.2017.0816. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lu R, Pickett HA. Telomeric replication stress: the beginning and the end for alternative lengthening of telomeres cancers. Open Biol. 2022;12(3):220011. 10.1098/rsob.220011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Maier R, Bawamia B, Bennaceur K, Dunn S, Marsay L, Amoah R, et al. Telomerase activation to reverse immunosenescence in elderly patients with acute coronary syndrome: protocol for a randomized pilot trial. JMIR Res Protoc. 2020;9(9):e19456. 10.2196/19456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Martínez P, Blasco MA. Replicating through telomeres: a means to an end. Trends Biochem Sci. 2015;40(9):504–15. 10.1016/j.tibs.2015.06.003. [DOI] [PubMed] [Google Scholar]
- Muscari A, Forti P, Brizi M, Magalotti D, Capelli E, Potì S, et al. Can We Slow Down Biological Age Progression? Study Protocol for the proBNPage Reduction (PBAR) Randomized, Double-Blind, Placebo-Controlled Trial (Effects of 4 “Anti-Aging” Food Supplements in Healthy Older Adults). Clin Interv Aging. 2023;18:1813–25. 10.2147/CIA.S422371. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ (Clinical Research Ed). 2021;372:n71. 10.1136/bmj.n71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Salvador L, Singaravelu G, Harley CB, Flom P, Suram A, Raffaele JM. A natural product telomerase activator lengthens telomeres in humans: a randomized, double blind, and placebo controlled study. Rejuvenation Res. 2016;19(6):478–84. 10.1089/rej.2015.1793. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shim HS, Iaconelli J, Shang X, Li J, Lan ZD, Jiang S, et al. TERT activation targets DNA methylation and multiple aging hallmarks. Cell. 2024;187(15):4030-42.e13. 10.1016/j.cell.2024.05.048. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tsoukalas D, Fragkiadaki P, Docea AO, Alegakis AK, Sarandi E, Thanasoula M, et al. Discovery of potent telomerase activators: Unfolding new therapeutic and anti-aging perspectives. Mol Med Rep. 2019;20(4):3701–8. 10.3892/mmr.2019.10614. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Van Ly D, Low RRJ, Frölich S, Bartolec TK, Kafer GR, Pickett HA, et al. Telomere Loop Dynamics in Chromosome End Protection. Mol Cell. 2018;71(4):510-25.e6. 10.1016/j.molcel.2018.06.025. [DOI] [PubMed] [Google Scholar]
- Zhu Y, et al. The achilles’ heel of senescent cells: from transcriptome to senolytic drugs. Aging Cell. 2015;14(4):644–58. 10.1111/acel.12344. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
All data generated or analysed during this study are included in this published article. The complete dataset supporting the findings of this study, including the quality-audited data extraction sheet, is available as Supplementary Material 2. The analytical code is provided as Supplementary Material 3.