Can Stem Cell Therapy Improve Frailty?
From CRATUS and Laromestrocel to Zhu 2024 and Nguyen 2026 — what the MSC frailty trials actually show
Frailty is not simply a condition of being old. It is a clinical syndrome in which muscle strength, gait capacity, endurance, balance, resilience, and the ability to regulate inflammation all decline together — making a person vulnerable to even minor physiological stressors. The goal of frailty treatment is not cosmetic anti-aging, but restoring functional capacity and physiological reserve for everyday life.
This is precisely why clinical trials of mesenchymal stem cells (MSC) for frailty have attracted growing attention. The rationale for MSC treatment rests not on a simple idea of cells replacing damaged tissue, but on their paracrine effects: anti-inflammatory signaling, immune modulation, vascular function improvement, and reshaping of the local cellular environment — all of which can address the biological mechanisms underlying frailty.
The Biology of Frailty and the Rationale for MSC Therapy
The pathophysiology of frailty involves four overlapping mechanisms: chronic low-grade systemic inflammation (inflammaging), mitochondrial dysfunction, exhaustion of endogenous stem cells, and immune senescence (immunosenescence). The depletion of skeletal muscle satellite cells and their reduced regenerative capacity is a central driver of muscle loss and declining mobility.
Using a patient’s autologous stem cells faces a fundamental limitation: cells already exposed to years of chronic inflammation have severely diminished replicative lifespan and stemness. By contrast, allogeneic MSCs from young, healthy donors (allo-hMSCs) offer several advantages:
- Low expression of MHC class II, co-stimulatory molecules, and MHC class I — reducing the risk of immune rejection
- Strong homing capacity to sites of tissue damage and inflammation
- Secretion of TGF-β, IL-10, and exosomes that promote M2 anti-inflammatory macrophage polarization and suppress TNF-α, IL-1β, and IL-6
- TGF-β signaling suppresses harmful CD4+/CD8+ T-cell proliferation and Th1 activation while increasing regulatory T cells (Treg)
Bone Marrow-Derived MSC Trials
CRATUS Phase I (NCT02065245)
The CRATUS program, led by the University of Miami and Longeveron, began with an open-label, dose-escalation Phase I study in 15 patients aged 60 and older with frailty. A single intravenous infusion of allogeneic bone marrow-derived MSC (hBM-MSC) was given at three dose levels.
The cell suspension was delivered at 2 mL/min through a dedicated infusion bag, with the bag agitated every 15 minutes to maintain uniform dispersion. Through one month post-infusion, no treatment-emergent serious adverse events (TE-SAE) were observed. One patient in the 20M group showed a mild donor-specific anti-HLA antibody response (cPRA); no T-cell-mediated rejection was detected.
For efficacy, six-minute walk distance (6MWD) improved significantly at 3 and 6 months (p=0.001), and serum TNF-α declined markedly at 6 months (p<0.0001). Notably, the 100M dose group showed greater improvements in FEV1, SF-36, and EQ-5D than the 200M group, challenging the assumption that higher cell doses always produce greater benefit.
CRATUS Phase II
Phase II enrolled 30 patients (mean age 75.5 years) in a randomized, double-blind, placebo-controlled design with 1:1:1 allocation to 100M, 200M, or placebo groups.
No TE-SAE were observed at one month. Functionally, only the 100M group achieved statistically significant improvements in 6MWD, SPPB, and FEV1 (p=0.01), along with quality of life gains (p=0.03). Immunologically, serum TNF-α fell significantly in the 100M group, and the pro-inflammatory TNF-α content within B cells declined across both cell-treated groups, suggesting systemic immune rejuvenation beyond simple dose-response linearity.
Laromestrocel (Lomecel-B) Phase 2b (Ruiz 2026, Cell Stem Cell)
This is the strongest evidence to date for MSC treatment of frailty. Conducted across 12 US centers, 148 ambulatory patients aged 70–85 with mild to moderate frailty were randomized to five arms: 25M, 50M, 100M, or 200M of Laromestrocel (a processed hBM-MSC product), or placebo — all as a single intravenous infusion.
At 9 months, the 200M group showed a 6MWD increase of 63.4 m over placebo (95% CI: 17.1–109.6 m, p=0.0077). A dose-response relationship was confirmed at 6 months. The most clinically meaningful finding: 30.8% of treated patients reversed their frailty classification on the Clinical Frailty Scale (CFS), recovering to a “Well” category, compared to 14.8% in the placebo group.
The functional gains were not isolated to a single metric. The increase in 6MWD correlated with patient-reported PROMIS Physical Function scores. The research team proposed soluble TIE2 (sTIE2) in blood as a potential treatment biomarker.
The sTIE2 biomarker and vascular protection mechanism: TIE2 is a receptor on vascular endothelial cells that, when bound by Angiopoietin, activates the PI3K/AKT survival pathway and suppresses NF-κB-driven inflammation. In the frailty environment, MMP14 becomes overactivated and cleaves TIE2 from the endothelial surface, releasing it as a soluble, inactive fragment (sTIE2) into the bloodstream. This leads to microvascular damage and progressive skeletal muscle dysfunction. TIMP2 and TIMP3 secreted by Laromestrocel inhibit MMP14, blocking TIE2 cleavage and restoring endothelial homeostasis. Synergy with VEGF signaling is proposed to rebuild the microvascular network within skeletal muscle, improving ATP production capacity and ultimately extending walking distance.
Bone Marrow-Derived MSC Clinical Trial Summary
| Trial | N | Age | Dose | Key Outcomes | Notes |
|---|---|---|---|---|---|
| CRATUS Ph I (NCT02065245) | 15 | ≥60 | 20M / 100M / 200M | 6MWD↑ (p=0.001), TNF-α↓ (p<0.0001) | 100M best for FEV1·QOL; 1 cPRA event |
| CRATUS Ph II (NCT02065245) | 30 | 75.5 (mean) | 100M / 200M / Placebo | Only 100M: 6MWD·SPPB·FEV1 significant (p=0.01) | Double-blind RCT; non-linear dose response confirmed |
| Laromestrocel Ph 2b (NCT03169231) | 148 | 70–85 | 25M / 50M / 100M / 200M / Placebo | 200M: +63.4 m 6MWD (p=0.0077); 30.8% CFS reversal | Multi-center confirmatory; sTIE2 biomarker identified |
Umbilical Cord-Derived MSC Trials
HUC-MSCs (human umbilical cord-derived MSCs) have distinct practical advantages over bone marrow-derived MSCs. They are obtained non-invasively from post-natal discarded tissue, carry the biological potency of a young donor, and express very low or absent MHC class I and II, minimizing allogeneic rejection risk.
Zhu 2024 (NCT04314011)
A randomized, double-blind, placebo-controlled trial conducted at Shanghai East Hospital between 2020 and 2022, enrolling 30 patients aged 60–80 with frailty (Fried phenotype score 1–4). HUC-MSC at 1×10⁶ cells/kg were infused intravenously twice, one month apart (Day 1 and Day 30), with 6-month follow-up.
- SF-36 Physical Component Summary (PCS): Improvement began within one week and was sustained and statistically significant vs. placebo through 6 months (p=0.042)
- EQ-VAS (patient-reported health): Significant advantage at 2 months, widening at 6 months (p=0.002)
- TUG (Timed Up-and-Go): Consistent improvement throughout the observation period (p<0.05)
- Grip strength: Significantly greater than placebo at 6 months (p=0.002)
- 4-meter gait speed: Average 2.05 seconds faster than placebo (p=0.21; below the significance threshold but showed a recovery signal)
- Serum TNF-α and IL-17: Both declined at 6 months (p=0.034, p=0.033)
No significant safety differences were observed compared to placebo.
Nguyen 2026 (NCT04919135) — Vinmec Research, Vietnam
A Phase 1/2 randomized controlled trial by the Vinmec Research Institute of Stem Cell and Gene Technology, enrolling 44 patients aged 60–85 (CFS 3–6 or Fried ≥3). HUC-MSC at 1.5×10⁶ cells/kg were infused intravenously twice, at 3-month intervals (Day 1 and Month 3).
A key design feature: both treatment and control arms received identical standard nutritional supplements (amino acid preparation, calcium, joint support compound), allowing the investigators to isolate the cell-specific effect from nutritional improvement.
The secondary efficacy analysis focused on mechanistic biomarkers:
- Seahorse XF analysis: Measurement of oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) in CD3+ T cells extracted from patient blood, to assess whether mitochondrial metabolism was restored
- qPCR: Direct quantification of CDKN2A (p16) — a senescence-promoting transcription factor — in CD3+ T cells, as an indicator of cellular senescence reversal
This trial is the outcome report of the study design published in J Gerontology Series A (2022), with results published in EBioMedicine (2026).
Umbilical Cord-Derived MSC Clinical Trial Summary
| Trial | N | Age | Dose · Schedule | Key Efficacy Results | Design Feature |
|---|---|---|---|---|---|
| Zhu 2024 (NCT04314011) | 30 | 60–80 | 1×10⁶/kg ×2 (Day 1, 30) | SF-36 PCS↑ (p=0.042), Grip↑ (p=0.002), TUG↑, TNF-α·IL-17↓ | 6-month follow-up; saline placebo control |
| Nguyen 2026 (NCT04919135) | 44 | 60–85 | 1.5×10⁶/kg ×2 (Day 1, Month 3) | Mitochondrial metabolic recovery, CDKN2A↓ | Matched nutritional supplement in both arms; Seahorse XF + qPCR mechanistic analysis |
Stepwise Trial Design: Success and Failure
Taiwan UMC119-06-05 (Meridigen/Meribank) — A Model of Phased Risk Management
The Taiwanese umbilical cord MSC product UMC119-06-05 followed a stepwise regulatory path:
Phase I (NCT04914403, n=6): Small-scale safety screening in patients aged 60–85 with mild to moderate frailty (CFS 4–6). Exclusion criteria strictly filtered NYHA class III/IV heart failure, GOLD III/IV COPD, malignancy within 5 years, dementia (MMSE <24), and uncontrolled diabetes or renal failure — all conditions that could mask or confound safety events.
Phase I/II (NCT06501066): Following safety confirmation, a randomized, single-blind expansion trial comparing a single IV infusion of 100M cells vs. placebo in pre-frail and frail participants. The composite safety endpoint includes death, non-fatal pulmonary embolism, stroke, unexplained dyspnea hospitalization, and major lab abnormalities — tracked over 12 months.
VA MESCAFY (NCT05284604) — The Cost of Ignoring Real-World Patient Populations
Led by Dennis T. Villareal, MD, with support from the US Department of Veterans Affairs and Baylor College of Medicine, MESCAFY was designed to enroll 36 veterans aged 65–85 with frailty (PPT score 18–31, CFS 5–6, 6MWD 200–400 m) and deliver repeat infusions of BM-MSC (100M on Day 1 and Day 90).
The trial was withdrawn on January 31, 2023, without enrolling a single patient.
The reason was straightforward: the combination of narrow 6MWD eligibility criteria, strict comorbidity exclusions calibrated to a complex veteran population, and the logistical demands of repeat high-cost cell infusions produced a patient pool that was effectively impossible to fill. This is a sharp reminder that even scientifically sound protocols will fail if they do not account for the real-world characteristics of the target patient population.
Trial Design Comparison
| UMC119-06-05 (Taiwan) | MESCAFY (VA, USA) | |
|---|---|---|
| Sponsor | Meridigen / Meribank | US Department of Veterans Affairs |
| Target N | Ph I: 6 → Ph I/II: multi-center | 36 |
| Cell source | UC-MSC | BM-MSC |
| Infusion schedule | Single (Ph I/II) | 2× (Day 1, Day 90) |
| Status | Active | Withdrawn |
| Key lesson | Stepwise screening satisfies regulatory requirements | Ignoring real patient population demographics leads to recruitment failure |
Safety Concern: MSC Infusion and Hypercoagulability
The most significant safety risk associated with intravenous MSC infusion is hypercoagulability and the associated risk of thromboembolism. Findings from the Vinmec group, drawn from animal models and clinical cohorts (NCT05292625, NCT04919135), provide important guidance.
Mechanism: HUC-MSCs and adipose-derived MSCs express high levels of Tissue Factor (TF) on their cell surface membranes — the initiating protein of the extrinsic blood coagulation cascade. Bone marrow-derived MSCs express TF at comparatively lower levels. When TF-expressing MSCs enter the venous bloodstream, they bind Factor VII and platelets, rapidly accelerating microthrombus formation.
Clinical observation: In patients receiving HUC-MSC intravenously, serum D-dimer showed a transient but sharp spike immediately after infusion. This pattern was absent in patients who received the same cells intrathecally (into the spinal canal), confirming the route-dependence of the coagulation response.
Animal toxicology: IV bolus injection of UC-MSC or adipose-derived MSC at ≥50×10⁶ cells/kg caused fatal multi-organ thrombosis (portal vein thrombosis, pulmonary embolism) in mouse and rabbit models. At human-equivalent clinical doses (~0.5×10⁶ cells/kg), only localized vascular irritation around the infusion site was observed, resolving within 96 hours.
Practical safety measures:
- Prophylactic low-molecular-weight heparin or aspirin at least 24 hours before infusion
- D-dimer monitoring for 72 hours after infusion
- Recognize that ex vivo TF expression data do not fully predict in vivo coagulation risk; patient-specific factors including hypoxic priming and systemic inflammation level must be considered
Limits of the Current Evidence
It is premature — and misleading — to interpret the available trials as proof that stem cell therapy has reversed biological aging.
First, study participants are older adults with frailty or pre-frailty, not healthy adults. Second, primary outcomes are functional metrics — 6MWD, SPPB, TUG, grip strength, quality of life, inflammatory biomarkers — not biological age reversal or lifespan extension. Third, the following remain unestablished:
- Large-scale Phase 3 randomized controlled trials
- Hard clinical outcomes: falls, hospitalizations, disability onset, long-term care entry, mortality reduction
- Optimal dosing intervals and dose for repeat administration
- Long-term safety (beyond 12 months)
- Cost-effectiveness and criteria for patient selection in real clinical settings
The largest study to date — Laromestrocel Phase 2b (N=148) — is still modest relative to confirmatory Phase 3 trial standards.
Conclusion: A Measured Assessment
Taken together, the current evidence shows that MSC treatment — from both bone marrow and umbilical cord sources — consistently produces signals in the direction of functional recovery in frailty. Safety profiles have been reported as relatively favorable. Improvements in gait capacity, physical function, quality of life, and inflammatory biomarkers have been observed in a consistent direction across multiple trials.
What has shifted in recent research is the central question itself: not “does it make patients younger?” but rather “do they walk better, feel less fatigued, show improved inflammation and vascular function, and report better quality of life?” — measurable functional endpoints that matter for daily independence.
Future trials must move beyond simple cell infusion with subjective improvement measures. Physical performance, gait speed, muscle strength, balance, patient-reported outcomes, inflammatory and vascular biomarkers, and long-term clinical events must all be tracked systematically. Frailty is a multidimensional syndrome; its treatment cannot be judged by a single blood test or questionnaire alone.
For stem cell therapy to earn a place in genuine frailty care, the field must demonstrate that functional gains translate into sustained independence and reduced healthcare utilization over time. Rigorous clinical evidence — not commercial enthusiasm — will determine whether this promise is real.
This article synthesizes findings from Golpanian S et al. (CRATUS Phase I, J Gerontol A 2017), Tompkins BA et al. (CRATUS Phase II, J Gerontol A 2017), Ruiz JG et al. (Laromestrocel Phase 2b, Cell Stem Cell 2026), Zhu Y et al. (HUC-MSC frailty RCT, 2024), Nguyen LT et al. (Vinmec UC-MSC RCT, EBioMedicine 2026), and Vinmec safety analyses of MSC-associated hypercoagulability.
MSC trials using bone marrow and umbilical cord sources have shown consistent signals of improvement in gait speed, grip strength, quality of life, and inflammatory markers. The Laromestrocel Phase 2b trial (N=148) is the strongest evidence to date, with 6MWD +63.4 m and 30.8% reversal of frailty status. However, large-scale Phase 3 trials, long-term safety, and hard clinical outcomes have not yet been established.
Golpanian S, et al.; Tompkins BA, et al.; Ruiz JG, et al.; Zhu Y, et al.; Nguyen LT, et al.."CRATUS Phase I/II; Laromestrocel Phase 2b (Cell Stem Cell 2026); HUC-MSC frailty RCT; Vinmec UC-MSC RCT (EBioMedicine 2026)." J Gerontol A Biol Sci Med Sci; Cell Stem Cell; EBioMedicine, 2026