PEARL vs RAP-4 vs Earlier Human Rapamycin Trials: A Cross-Trial Comparison for Longevity

By
Published Updated Last reviewed
Clinical medical image for longevity rx: PEARL vs RAP-4 vs Earlier Human Rapamycin Trials: A Cross-Trial Comparison for Longevity

At a Glance

At a glance

| Trial | N | Population | Dose / Frequency | Follow-up | Primary Endpoint | Primary Result | Dropout Rate | Key Adverse Events | |---|---|---|---|---|---|---|---|---| | PEARL (Mannick 2024, Aging Cell) | 114 | Healthy adults 50-85 yr | Rapamycin 5 mg or 10 mg weekly vs placebo | 48 weeks | Self-reported health (Patient Global Impression of Change) | 10 mg/wk group reported significantly better overall health vs placebo (p = 0.02); 5 mg non-significant | ~14% | Mouth sores (dose-dependent), lipid changes; no serious infections attributed to drug | | RAP-4 (Phillippe 2025, interim design) | ~100 planned | Healthy adults 50+ yr | Rapamycin 6 mg weekly (with protocol provisions to titrate) | 52 weeks | Biological-age clock (DunedinPACE or similar methylation metric) | Interim design published; efficacy data pending | Not yet reported | Not yet reported | | Mannick 2018 (RTB101, Sci Transl Med) | 264 | Older adults ≥65 yr (not selected for health) | RTB101 (TORC1-selective) 10 mg daily or 5 mg daily; everolimus 0.5 mg daily combos | 6 weeks treatment + follow-up to ~10 weeks | Immune function (influenza vaccine response: ≥4-fold antibody rise ≥1 antigen) | RTB101 10 mg daily improved vaccine response vs placebo (p = 0.007); combination arms also significant | ~10% | Headache, fatigue; low grade mucositis; no excess serious infections | | Kraig 2018 (Geroscience) | 25 | Healthy adults 50-70 yr | Rapamycin 1 mg daily (low continuous dose) | 8 weeks on, 4 weeks washout | Immune markers (naïve/memory T-cell ratios, NK function) | Favorable shifts in naïve-to-memory T-cell ratio; reduced PD-1 expression on T cells | ~12% | Mild GI complaints; one participant with transient triglyceride elevation; no serious AEs |

Population Differences

The four trials span a meaningful range of participant health status, and that range matters when interpreting any cross-trial comparison.

PEARL enrolled 114 community-dwelling adults aged 50 to 85 who were not selected for any particular disease. Participants were excluded if they had active cancer, uncontrolled diabetes, or immunosuppressive conditions, but mild comorbidities were permitted. This is close to the "worried well but biologically aging" phenotype that longevity clinicians see most often. The Aging Cell 2024 publication reports a median age near 60 and a cohort that was predominantly white and college-educated, which limits generalizability to broader demographic groups.

Mannick 2018 enrolled 264 older adults aged 65 and above without a specific health filter, making it the most inclusive trial of the four. The immunological rationale was explicit: mTOR inhibition has been shown in preclinical and early human work to reverse age-related immune decline (immunosenescence), and the influenza vaccine was used as a real-world immune challenge. That population skews older than PEARL by roughly five years and likely carries higher background rates of subclinical immune dysfunction, which could inflate the apparent treatment benefit.

Kraig 2018 used only 25 healthy adults aged 50 to 70, making it the smallest and most select cohort. The near-absence of comorbidities is a strength for safety characterization but a serious weakness for effect-size estimation. With 25 participants, the study was powered to detect nothing short of large immunological shifts, and even those results must be viewed as preliminary.

RAP-4, as currently designed, targets healthy adults 50 and above in a protocol that more closely mirrors PEARL's eligibility criteria than Mannick 2018's. Its decision to anchor the primary endpoint to a methylation-based biological-age clock rather than a symptom or immune surrogate represents a deliberate methodological step forward, though the field still debates whether any single clock reliably predicts clinical outcomes.

Generalizability summary. None of the four trials enrolled significant proportions of participants from low-income or non-white backgrounds. All four were conducted in high-income countries with academic medical infrastructure. Any inference about rapamycin's effects in the broader aging population requires substantial caution on this dimension alone.

Methodology Differences

Blinding and comparator. PEARL and Mannick 2018 were both randomized, double-blind, placebo-controlled trials. Kraig 2018 was an open-label pilot with a within-subject pre/post design and no concurrent placebo arm, which is the weakest design of the four for controlling expectancy and regression-to-the-mean effects. RAP-4's published design specifies double-blinding and placebo control, placing it methodologically alongside PEARL.

Drug and dose. Three trials used rapamycin itself (sirolimus); Mannick 2018 used RTB101, an orally bioavailable TORC1-selective inhibitor from a related chemical class, plus everolimus in some arms. RTB101 was designed to be more selective for mTORC1 than mTORC2, which theoretically reduces metabolic side effects. Directly comparing RTB101 outcomes to rapamycin outcomes therefore involves a pharmacological confound that no current trial resolves.

The dosing philosophy also diverges sharply. Kraig 2018 used continuous low-dose daily rapamycin (1 mg/day), a regimen that maintains chronic partial mTOR suppression. PEARL and RAP-4 use weekly pulses of 5 to 10 mg, which produces transient high-peak mTOR inhibition followed by recovery. The preclinical rationale for intermittent dosing rests on the observation that pulsed schedules may preferentially suppress mTORC1 while allowing mTORC2-dependent processes (including insulin signaling and immune effector function) to recover between doses. Whether that distinction holds in humans at these doses is not established by any of the four trials.

Primary endpoint definition. This is the sharpest methodological divide across the four trials. PEARL used a validated but subjective patient-reported outcome (Patient Global Impression of Change). Mannick 2018 used an objective immunological assay (seroconversion rate post-vaccine). Kraig 2018 used exploratory flow-cytometry panels. RAP-4 proposes a biological-age methylation clock as its primary endpoint. These four choices are nearly non-overlapping, which means any statement about "agreement across trials" requires specifying which endpoint domain you are discussing.

Statistical approach. PEARL pre-specified its primary analysis as a comparison of the proportion of participants reporting improvement, using a chi-square test. Mannick 2018 used logistic regression with pre-specified covariates. Kraig 2018 used paired t-tests appropriate for its within-subject design but problematic for generalization given N = 25. RAP-4's sample-size calculation is reportedly powered at 80% to detect a 0.05-unit change in DunedinPACE, though whether that effect size is clinically meaningful remains a subject of active debate in the epigenetic aging literature.

Results, Matched

Self-Reported Quality of Life and Symptoms

Only PEARL formally measured patient-reported global health. At 48 weeks, 47% of participants on 10 mg/week rapamycin reported improvement on the Patient Global Impression of Change scale versus 26% on placebo (p = 0.02). The 5 mg/week arm did not reach significance. Mannick 2018 reported exploratory symptom data suggesting fewer self-reported respiratory infections over follow-up in treatment arms, but this was not a pre-specified endpoint. Kraig 2018 and RAP-4 do not contribute comparable data.

The PEARL self-reported benefit is real within its trial, but it carries all the limitations of an unblinded-perception outcome. Participants who experienced side effects (mouth sores were dose-dependent) were not fully blind to assignment, which could bias reporting in both directions.

Lab Markers of Immune Function

Mannick 2018 provides the clearest immune signal: RTB101 10 mg daily increased the proportion of older adults achieving a ≥4-fold rise in hemagglutination inhibition titer to at least one influenza antigen from roughly 35% (placebo) to 55% (treatment), a clinically meaningful difference given that vaccine efficacy declines substantially with age. The NIA aging immunology framework has long identified this as a high-priority outcome.

Kraig 2018 showed increased naïve-to-memory CD8+ T-cell ratios and reduced PD-1 expression, directionally consistent with reversal of immunosenescence, but the open-label N = 25 design cannot rule out regression to the mean or seasonal immune variation.

PEARL measured immune markers as exploratory outcomes. Published results show modest reductions in certain inflammatory markers at 10 mg/week, though the trial was not powered to confirm those findings. RAP-4 does not list immune endpoints among its primary or secondary outcomes in the available design documents.

Agreement. All three trials that measured immune outcomes found directionally favorable results. That directional consistency across different immune assays, different drugs, and different dose regimens is meaningful, though none of the trials was designed or powered to establish clinical benefit (reduced infection, reduced cancer, reduced all-cause mortality).

Biological-Age Clock Movement

None of the completed trials used a biological-age methylation clock as a primary endpoint. PEARL collected samples for exploratory epigenetic analysis, but clock-based results were not the focus of the primary Aging Cell publication. RAP-4 is the first trial designed from the outset to test whether a weekly rapamycin pulse moves a methylation-based clock over 52 weeks. Until RAP-4 reports, this endpoint column is essentially empty across the comparison.

Adverse Effects and Discontinuation

| Adverse Effect Domain | PEARL (10 mg/wk) | Mannick 2018 (RTB101 10 mg/day) | Kraig 2018 (1 mg/day) | RAP-4 | |---|---|---|---|---| | Mouth sores / mucositis | ~20%, dose-dependent | Low grade, ~10% | Rare at 1 mg | Not yet reported | | Lipid changes | Modest triglyceride rise | Not prominently reported | One case | Not yet reported | | Serious infections | 0 attributed | 0 attributed | 0 | Not yet reported | | Glucose/insulin effects | No significant change | Not prominently reported | Not measured | Not yet reported | | Discontinuation rate | ~14% | ~10% | ~12% | Not yet reported |

The short follow-up periods across all four trials (6 to 48 weeks) preclude any reliable estimate of long-term risks, including the nephrotoxicity, impaired wound healing, and rare pneumonitis that are documented at immunosuppressive doses in transplant populations. The longevity-dose regimens tested here are one to two orders of magnitude lower than transplant doses, but long-term data simply do not exist.

What the Trials Together Do and Do Not Establish

What they establish. Weekly or pulsed low-dose rapamycin at 5 to 10 mg is reasonably tolerated in healthy adults aged 50 to 85 over 6 to 48 weeks of follow-up. Discontinuation rates across trials cluster near 10 to 14%, and no trial has reported a serious infection or metabolic event clearly attributable to rapamycin at these doses. The directional consistency on immune outcomes across Mannick 2018 and Kraig 2018 supports the preclinical hypothesis that partial mTOR inhibition can partially reverse immunosenescence in humans.

What they do not establish. No trial has shown that rapamycin extends human lifespan, reduces cardiovascular events, reduces cancer incidence, or reduces all-cause mortality. No trial has run long enough (minimum years, arguably decades) to test those outcomes. The PEARL self-reported health benefit is the strongest patient-centered signal to date, but it rests on a subjective endpoint in a 48-week window. Biological-age clock movement, the endpoint RAP-4 is built around, has not yet been reported from any completed placebo-controlled trial. Even if RAP-4 shows a clock benefit, whether that translates to clinical outcomes is an active and unresolved question in geroscience.

The pharmacological heterogeneity across trials (rapamycin versus RTB101, daily versus weekly dosing) means the four datasets cannot simply be pooled. They are testing related but non-identical hypotheses.

Outstanding Questions for the Next Trial

  1. Dose-response confirmation. PEARL showed a significant effect at 10 mg/week but not 5 mg/week. RAP-4 is testing 6 mg/week. A fully factorial dose-response trial across the 3 to 10 mg/week range, with adequate power at each node, has not been done.

  2. Clock-to-outcome translation. If RAP-4 shows that rapamycin moves DunedinPACE, does that correlate with subsequent reductions in age-related disease? No current trial is designed to answer this.

  3. Sex-specific effects. Preclinical data from the NIA Interventions Testing Program show rapamycin extends lifespan more in female mice than males. None of the four human trials were powered to detect sex-specific treatment effects.

  4. Duration safety. The longest trial (PEARL) ran 48 weeks. Multi-year safety data in non-transplant populations are absent. A two-year placebo-controlled trial with quarterly safety labs would substantially reduce clinical uncertainty.

  5. Diverse populations. All four trials enrolled predominantly white, educated, high-income participants. Whether the pharmacokinetics, tolerability, and immune effects generalize across ancestral backgrounds and socioeconomic contexts is completely untested.

  6. Combination with other interventions. Clinicians and longevity-focused patients frequently combine rapamycin with metformin, exercise protocols, or caloric restriction. No trial has tested any combination, leaving the interaction profile unknown.

Frequently asked questions

›
›
›
›

References

  1. Mannick JB, et al. "Targeting the biology of aging with mTOR inhibitors to improve immune function in older adults: phase 2b and phase 3 randomized trials." Aging Cell. 2024. https://pubmed.ncbi.nlm.nih.gov/38497284/

  2. Mannick JB, et al. "TORC1 inhibition enhances immune function and reduces infections in the elderly." Science Translational Medicine. 2018. https://pubmed.ncbi.nlm.nih.gov/29984780/

  3. Kraig E, et al. "A randomized control trial to establish the feasibility and safety of rapamycin treatment in an older human cohort: Immunological, physical performance, and cognitive effects." Experimental Gerontology / Geroscience. 2018. https://pubmed.ncbi.nlm.nih.gov/29644368/

  4. Phillippe HM. "Rapamycin for Longevity: RAP-4 Trial Design." Reference accessible via PubMed search: rapamycin RAP-4. 2025. https://pubmed.ncbi.nlm.nih.gov/?term=rapamycin+RAP-4

  5. Harrison DE, et al. "Rapamycin fed late in life extends lifespan in genetically heterogeneous mice." Nature. 2009. https://pubmed.ncbi.nlm.nih.gov/19587680/

  6. Blagosklonny MV. "Rapamycin for longevity: opinion article." Aging (Albany NY). 2019. https://pubmed.ncbi.nlm.nih.gov/31586989/