UBX0101
UBX0101, intra-articular MDM2 to p53 interaction inhibitor (Unity Biotechnology)
Written by Aaron CuhaReviewed Sep 2026
Also known as: UBX 0101
The first senolytic taken into a randomised phase 2 trial for a symptom endpoint. In 183 people with knee osteoarthritis, pain scores at 12 weeks were the same on every dose of UBX0101 as on placebo, according to the results Unity Biotechnology posted to ClinicalTrials.gov.
Overview
UBX0101 is a small molecule that blocks the interaction between MDM2 and p53, which stabilises p53 and pushes senescent cells into apoptosis. It was developed by Unity Biotechnology for injection directly into an osteoarthritic knee, and it is a different molecule from UBX1325 (foselutoclax), the BCL-xL inhibitor Unity took into eye disease.
The animal rationale is specific. A 2017 Nature Medicine paper using a transgenic mouse that allows senescent cells to be tracked and killed showed that senescent cells accumulate in cartilage and synovium after anterior cruciate ligament transection, and that clearing them attenuated post-traumatic osteoarthritis, reduced pain and increased cartilage development. Intra-articular injection of a senolytic molecule reproduced the effect in transgenic, non-transgenic and aged mice. A 2023 paper reported that UBX0101 reduced oxidative protein modification in aged arthritic mouse knees, with different responses in young and aged animals.
The human programme is documented on ClinicalTrials.gov and the phase 2 results are posted there. NCT04129944 randomised 183 patients with painful knee osteoarthritis to a single intra-articular injection of placebo or UBX0101 at 0.5, 2.0 or 4.0 mg, triple masked. The primary endpoint was change in the WOMAC pain subscale at 12 weeks. Mean pain scores fell in every arm, from about 2.1 at baseline to 1.12 on placebo and 1.16, 1.04 and 1.10 on the three UBX0101 doses. There is no separation between drug and placebo in those numbers. Serious adverse events were reported in 1 of 46 on placebo and 2, 1 and 3 in the ascending dose arms, and one death was recorded in the 4.0 mg arm; the posted record does not attribute a cause. A long-term follow-up study of 161 of these patients, NCT04349956, was terminated, and the registry gives the reason as inability to achieve primary or secondary study objectives.
The widely repeated statement that the UBX0101 phase 2 failed could not be resolved to a peer-reviewed publication. It is verifiable in the registry, which is where the numbers above come from. No senolytic has improved a clinical outcome in a randomised trial in humans, and UBX0101 is the clearest worked example.
Mechanism of action
Established in mice and in cells: inhibiting the MDM2 to p53 interaction prevents p53 degradation, and senescent cells, which depend on suppressing p53-driven apoptosis to survive, die selectively. Delivery is by direct injection into the joint, which was intended to confine the effect to the knee. In humans, whether the injection cleared senescent cells from the joint was not established by the trials, which measured pain and function rather than senescent cell burden.
Human evidence
Four registered trials in knee osteoarthritis, two phase 1 and one phase 2 with results posted to ClinicalTrials.gov, plus a terminated long-term follow-up. The phase 2 showed no separation from placebo on its primary pain endpoint. No peer-reviewed publication of any of these trials was located.
- NCT04129944, phase 2, 183 randomised, triple masked, single intra-articular injection. WOMAC pain at week 12: placebo 1.12, UBX0101 0.5 mg 1.16, 2.0 mg 1.04, 4.0 mg 1.10, from baselines of 2.05 to 2.20. Manufacturer funded and manufacturer reported.
- Secondary endpoints in the same trial, WOMAC function and average daily pain on an 11-point scale, also showed no separation at week 12 or at week 24.
- Serious adverse events by arm: 1 of 46 placebo, 2 of 45, 1 of 46 and 3 of 46. One death recorded in the 4.0 mg arm; the registry posting gives no cause.
- NCT04349956, long-term follow-up of 161 of those patients, TERMINATED, reason given as inability to achieve primary or secondary study objectives.
- NCT03513016 (78 participants) and NCT04229225 (35 participants), both phase 1 and both COMPLETED, have no posted results and no located publication.
What this does not tell you: A single injection into one knee tested for 12 weeks cannot rule out that repeated dosing, a different molecule, or a different tissue would behave differently. Neither can it tell you whether senescent cells were cleared, because the trial measured pain and function rather than senescent cell burden. What it does show is that clearing senescent cells in this way did not reduce knee pain more than placebo in 183 people.
Reading the research record
This programme is the most useful single data point in the senolytic field, and it is almost invisible in the peer-reviewed literature. Three completed trials and a terminated follow-up produced one posted results record and no publication that could be located in PubMed. The failure of the phase 2 is widely repeated in commentary; the verifiable source for it is the registry posting, and this page takes its numbers from there rather than from the commentary.
The placebo arm is the other lesson. Pain scores fell by roughly half on placebo, which is what intra-articular injection trials routinely show and precisely why an uncontrolled senolytic study reporting improvement proves very little. The open-label pilots that generate the public claims about senolytics improving human function had no arm like this one.
The evidence, charted
Fig. 1 · evidence composition
4of 6 citations (67%) are in people
Counted from the citation list on this page. The Human count is the same number shown in the badge at the top. Both understate any literature larger than the sources we cite.
Fig. 2 · evidence over time
Evidence spans 3 distinct years, 2017 to 2026, counted from the citation list on this page.
Fig. 3 · legal status at a glance
US
Not approved
UK
Approved
AU
Approved
CA
Not approved
Approved in 2 of 4, prescription route in 0, not approved in 2. A jurisdiction's classification is a regulatory fact, not a verdict on the science; see Legal status below for the full text and any notes.
Fig. 4 · dose response
No human dose response curve exists
We draw this figure where the data supports it. For this compound in humans it does not, so the panel stays empty rather than borrowing an animal curve and implying it transfers.
Key studies & citations
- Human2026
A Study to Assess the Safety and Efficacy of a Single Dose of UBX0101 in Patients With Osteoarthritis of the Knee
Unity Biotechnology, phase 2, randomised, triple masked, 183 participants (placebo 46, UBX0101 0.5 mg 45, 2.0 mg 46, 4.0 mg 46). Results posted December 2021. WOMAC pain subscale fell from about 2.1 at baseline to 1.12 on placebo versus 1.16, 1.04 and 1.10 on the three doses at week 12. Function and daily pain scores likewise showed no separation. Serious adverse events 1, 2, 1 and 3 by arm; one death recorded in the 4.0 mg arm, cause not given in the posting.
ClinicalTrials.gov - Human2026
Long-Term Follow-Up Study of Patients With Osteoarthritis of the Knee Treated With UBX0101
161 participants enrolled, status TERMINATED. The registry states the reason for stopping as inability to achieve primary or secondary study objectives. Verified live September 2026.
ClinicalTrials.gov - Human2026
A Safety and Tolerability Study of UBX0101 in Patients With Osteoarthritis of the Knee
Phase 1, 78 participants, status COMPLETED. No results are posted on the registry record and no peer-reviewed publication of this trial was located.
ClinicalTrials.gov - Human2026
A Study of Single and Repeat Dose Administration of UBX0101 in Patients With Osteoarthritis of the Knee
Phase 1 single and repeat dosing, 35 participants, status COMPLETED. No results posted and no publication located.
ClinicalTrials.gov - Animal2017
Local clearance of senescent cells attenuates the development of post-traumatic osteoarthritis and creates a pro-regenerative environment
Mouse. Using the p16-3MR transgenic model after anterior cruciate ligament transection, selective elimination of senescent cells attenuated post-traumatic osteoarthritis, reduced pain and increased cartilage development, reproduced by intra-articular injection of a senolytic molecule in transgenic, non-transgenic and aged mice. Cells from patients undergoing knee replacement were treated in culture, not in the patients.
Nature Medicine - Animal2023
Senolytic treatment reduces oxidative protein stress in an aging male murine model of post-traumatic osteoarthritis
Mouse. UBX0101 reduced protein oxidative modification in aged arthritic knee joints, with divergent responses between young and aged animals, suggesting the benefit of senolysis in the joint depends on the age of the tissue.
Aging Cell
Frequently asked questions
Did the UBX0101 osteoarthritis trial fail?
The posted results show no separation from placebo. In 183 randomised patients, WOMAC pain at 12 weeks was 1.12 on placebo and 1.16, 1.04 and 1.10 on the three UBX0101 doses. The long-term follow-up study was terminated, with the registry giving inability to achieve primary or secondary study objectives as the reason. No peer-reviewed publication of the trial was located, so the registry posting is the source.
Is UBX0101 the same as UBX1325?
No. UBX0101 blocks the MDM2 to p53 interaction and was injected into osteoarthritic knees. UBX1325, also called foselutoclax, is a BCL-xL inhibitor injected into the eye for retinal disease. Same company, different molecules and different programmes.
Why did the placebo group improve so much?
Injecting anything into a painful knee produces large symptomatic improvement in trials, which is why placebo-controlled designs matter here. Pain scores fell by about half in the placebo arm, roughly the same as in every drug arm.
Does this mean senolytics do not work?
It means one senolytic, given once into one joint, did not beat placebo on knee pain over 12 weeks in 183 people. It is the only randomised senolytic trial with a symptom endpoint that has reported. As of September 2026, no senolytic has improved a clinical outcome in a randomised trial in humans.