Mel Gibson & Dr. Neil Riordan on the Joe Rogan Experience #1066, fact-checked

What the evidence shows: Riordan claimed that exactly 800 human clinical trials have ever been conducted with mesenchymal stem cells (MSCs) and that his book references all of them. Clinical trial registries such as ClinicalTrials.gov are continuously updated as new studies launch and are registered, so any fixed count of MSC trials becomes stale quickly, and framing a single figure as the complete universe of trials overstates certainty. Registry-based academic analyses of MSC trials, including a comprehensive review of MSC trials for a single indication (osteoarthritis), describe large and steadily growing trial counts rather than a static total, and note that registries like ClinicalTrials.gov, the EU registry, and China's registry each capture only part of the global picture. This indicates the true number of MSC human trials was likely already higher than 800 by 2018, when this interview aired, and has continued to grow since, meaning the 'exactly 800, and that's all of them' framing understated the field's scope even at the time it was said.

What the evidence shows: Riordan's figure of roughly $2.5 billion per approved drug traces to a widely cited 2016 Tufts Center for the Study of Drug Development estimate (about $2.6 billion), which was funded by pharmaceutical industry sponsors, is not publicly reproducible, and includes an estimated cost of capital (opportunity cost) alongside assumptions about high clinical failure rates that roughly double the reported cash outlay. An independent peer-reviewed study published in JAMA in 2020, using only publicly available data on FDA-approved drugs from 2009-2018, estimated a median capitalized R&D cost of about $985 million per approved drug (mean approximately $1.3 billion), with substantial variation by therapeutic area. The $2.5 billion figure is not fabricated, but it sits at the high end of a contested range built on a non-transparent, industry-funded methodology, while independently verifiable data point to costs well below that figure for many approved drugs.

What the evidence shows: Riordan is likely referring to a 2017 Spanish randomized controlled trial (Noriega et al., published in Transplantation) of intradiscal allogeneic mesenchymal stromal cell (MSC) injection for degenerative disc disease, the best-known European trial of this kind circulating at the time of the podcast. That study does not report that ~50% of discs became 'normal' on MRI: in a cohort of only 24 patients, disc degeneration graded by the Pfirrmann scale improved in the MSC group and worsened in controls, while clinical improvement (pain/disability) was seen in a subset described as 40% of the cohort, not a full return of discs to normal appearance. A larger, more rigorous European follow-up trial (RESPINE, a 2024 French multicentre, double-blind, placebo-controlled RCT with 114 patients) tested the same allogeneic MSC approach and found no significant difference between MSC injection and sham placebo at 12 months on either clinical response rates (74% vs 69%, p=0.77) or secondary MRI-based measures. Taken together, current evidence shows small early-phase signals of benefit in some patients but no robust confirmation that intradiscal stem cell injection restores discs to a 'normal' MRI appearance in half of cases, and the larger, better-controlled trial found no advantage over placebo. The claim overstates both the strength and the specificity (MRI 'normalization') of the supporting data.

What the evidence shows: A peer-reviewed, controlled study matching the '172 patients' detail does exist: Wang et al. (2013, Stem Cells and Development, PMID 23941289) enrolled 172 rheumatoid arthritis patients with inadequate response to standard treatment, randomizing 136 to umbilical cord MSC infusion (plus DMARDs) and 36 to a medium/placebo control (plus DMARDs), with follow-up up to 8 months. That study reported serum TNF-alpha and IL-6 both decreased significantly (P<0.05) after treatment and no comparable improvement in controls, so Riordan's cytokine claim and his description of near-universal clinical improvement are substantially grounded in real, controlled, published data, not fabricated. However, several details in his retelling do not match the paper: the published dose was 4x10^7 (40 million) cells per infusion, not the roughly 120 million cells Riordan cites (he appears to be describing his own Panama clinic's typical dosing rather than the dose actually used in this Chinese-hospital trial, which he was not involved in and does not credit by name). The paper also does not report a precise 'eight and a half months' duration or a 30-patient retreatment subgroup with a further 50% cytokine drop, details that could not be independently verified in the available record and may come from unpublished follow-up data from Riordan's own clinic. Riordan founded and directs the Stem Cell Institute in Panama, which sells umbilical cord MSC infusions for rheumatoid arthritis and other conditions, giving him a financial incentive to present these results as more definitive, more applicable to his own clinic's dosing, and more directly attributable to his own work than the underlying study supports. No umbilical cord MSC product is FDA-approved for rheumatoid arthritis.

What the evidence shows: Riordan's company did complete and publish a small stem cell trial in multiple sclerosis: a 2018 paper in the Journal of Translational Medicine describes 20 subjects who received intravenous umbilical-cord mesenchymal stem cell infusions, with statistically significant improvements reported on some disability and quality-of-life measures at one month. However, the study was explicitly designed and labeled by its own authors as a safety and feasibility study, not an efficacy trial: it was open-label, uncontrolled, had no placebo or comparator arm, and enrolled only 20 patients, a sample size too small to draw reliable conclusions about a chronic, relapsing-remitting disease with substantial natural fluctuation. The paper's own conclusion states only that "potential therapeutic benefits should be further investigated," not that they were established. The study, and a subsequent correction issued in 2021, drew published concerns from other researchers about cost, transparency, and research integrity, to which Riordan and a co-author formally responded. Riordan is a co-founder and officer of the Stem Cell Institute/MediStem Panama, the same for-profit entity that performed the infusions and sells this treatment to paying patients, meaning the trial was not independently conducted. As of the podcast, the described "statistically significant" findings had already been published years earlier (2018, corrected 2021) contrary to the "submitted for publication" framing, and remain unreplicated by an independent, controlled trial.

What the evidence shows: The University at Buffalo hamster heart-failure study Riordan is describing (Shabbir et al., 2009, and related work from the same lab) did find that bone marrow mesenchymal stem cells (MSCs) improved cardiac function in TO-2 cardiomyopathic hamsters despite few or no cells persisting in the heart, consistent with a secreted-factor (trophic/paracrine) mechanism rather than cells directly repopulating heart tissue. However, the cells in that study were not injected intravenously: the researchers delivered them intramuscularly into the hamstrings, designing the protocol that way specifically because systemic IV infusion of MSCs is known from other cited work to cause the cells to become trapped in the lungs. The paper does not report an IV arm at all. So the underlying finding Riordan cites, few or no cells reaching the heart yet heart failure improving, is accurately described, but the intravenous route he attributes to it is not what this study tested; the study used intramuscular injection instead. More broadly, this and related animal studies are preclinical: MSC-based heart failure treatment remains investigational, with human trials showing modest, mixed effects rather than an established therapy.

What the evidence shows: Riordan claimed that in stem-cell trials for heart failure, a biomarker (likely BNP, brain natriuretic peptide) fell and ejection fraction rose in every single treated patient. The best available synthesis of controlled evidence, a 2016 Cochrane systematic review and meta-analysis of 38 randomized controlled trials (1,907 participants) of autologous bone-marrow-derived stem/progenitor cells in chronic ischemic heart disease and congestive heart failure, found only low-quality evidence of a modest average improvement in left ventricular ejection fraction, with wide confidence intervals. In the long-term MRI-measured subgroup, the mean difference in ejection fraction was -1.60 (95% CI -8.70 to 5.50, n=25), not statistically distinguishable from zero. The Cochrane authors explicitly cautioned that findings should be interpreted with caution given low event rates and imprecision, which is inconsistent with a uniform 100% response across patients. Universal, exceptionless improvement in both a natriuretic peptide biomarker and ejection fraction has not been demonstrated in peer-reviewed controlled trials; individual patient variability in response is the norm in this literature, not the exception. Ejection fraction is a standard, well-defined measure of heart pumping efficiency, so the claim's description of the metric is accurate, but the assertion of a universal (100%) responder rate across trials is not supported by the controlled-trial evidence base.

What the evidence shows: Riordan describes a single, undocumented patient case (an ejection fraction of 11%, versus a normal range of roughly 55-70%, that per his own account rose to about 42%) as evidence that stem cell infusion can reverse severe heart failure. Controlled clinical research on stem cell therapy for heart failure exists but does not support this kind of dramatic single-anecdote recovery: a 2025 phase 3 randomized controlled trial (PREVENT-TAHA8, published in The BMJ) testing intracoronary mesenchymal stem cell infusion in 396 post-heart-attack patients with reduced ejection fraction found a modest average improvement in ejection fraction (about 6 percentage points more than controls at six months) and a reduced rate of heart-failure hospitalization, but no significant effect on overall mortality. No controlled trial has demonstrated a rapid, near-total recovery of the kind implied by Riordan's anecdote. The American Heart Association's mainstream guidance on heart failure notes that stem cell approaches remain an active research area rather than an established treatment. The specific case Riordan describes has not been published or independently verified, so its diagnosis, treatment, and outcome cannot be confirmed.

What the evidence shows: Riordan described animal data showing mesenchymal stem cell injections can reverse ovarian failure, and cited one patient with an autoimmune-related fertility problem who conceived after treatment. Current peer-reviewed literature confirms extensive preclinical (rodent) evidence that mesenchymal stem cells or their derivatives can restore ovarian function and reduce granulosa cell apoptosis in chemically induced ovarian failure models. However, translation to humans remains at an early stage: a 2026 meta-epidemiological analysis found stem cell trials for premature ovarian failure are predominantly early-phase, single-center, and small-sample, with pregnancy-outcome data still sparse and unstandardized. No large randomized controlled trial has established that stem cell therapy reliably restores fertility or reverses ovarian failure in humans. A single patient's pregnancy following treatment is an anecdote that cannot be attributed to the intervention with any certainty, since spontaneous conception, misdiagnosis of ovarian failure, or other concurrent treatments cannot be ruled out without a controlled study. The claim accurately reflects the state of animal research but overstates its bearing on the individual human case presented as supporting evidence.

What the evidence shows: Riordan's claim about CIRM's funding running low is broadly accurate for the era: California's Proposition 71 provided roughly $3 billion starting in 2004, and by the late 2010s the agency had nearly exhausted that money, leading to what CIRM's own annual report calls "a long period of uncertainty" that ended only when voters approved a further $5.5 billion under Proposition 14 in November 2020. However, the claim that CIRM pivoted almost entirely away from embryonic stem cell research toward adult and umbilical cord cell studies is not supported. CIRM's own program pages show it has continued to fund human embryonic stem cell-derived clinical trials, including an active cardiomyocyte therapy for chronic ischemic heart disease, alongside adult stem cell, induced pluripotent stem cell, and gene therapy programs. As of its most recent public reporting, CIRM has funded over 100 clinical trials and more than 237 stem cell and gene therapy programs across multiple cell types, not a program narrowed down to adult or cord blood stem cells alone. The claim's framing that embryonic research was set aside is therefore misleading, even though the underlying point about the original funding nearly running out is factually grounded.

What the evidence shows: Riordan, founder of the for-profit Stem Cell Institute in Panama, cited an internal "cohort analysis" claiming 100% of spinal cord injury patients treated within one year of injury regained some neurologic function, versus 82% treated one to two years post-injury. No peer-reviewed publication reporting these specific figures could be located; the claim appears to originate from unpublished clinic data rather than an independently reviewed study. The largest independent meta-analysis of stem cell therapy for spinal cord injury (62 trials, 2,439 patients) found that stem cells improved ASIA impairment scale grade by at least one level in only about 49% of patients, well below the 82-100% Riordan reports, and concluded current evidence is not strong enough to support clinical translation of stem cell therapy for spinal cord injury. Independent natural-history data also show that a meaningful fraction of incomplete spinal cord injuries recover some neurologic function spontaneously in the first year without any stem cell intervention, which complicates attributing improvement solely to treatment in an uncontrolled cohort. Because Riordan's figures come from his own clinic, are not published in a form that can be independently verified, and lack a control group, the claim is unsupported by the current independent evidence base.

What the evidence shows: Riordan stated that neurosurgeons involved in designing his spinal cord injury trial hold that patients recover 98-99% of their eventual neurologic function within six months of injury, using this as the basis for a six-month endpoint. The natural history literature on spinal cord injury (SCI) recovery, most notably the International Campaign for Cures of Spinal Cord Injury Paralysis (ICCP) panel's 2007 analysis, describes spontaneous recovery as highly heterogeneous rather than following one fixed percentage: recovery in motor-complete injuries (AIS A/B) is limited and largely confined to the zone of partial preservation, while recovery in motor-incomplete injuries (AIS C/D) is substantially larger and more variable. The same analysis found the majority of spontaneous recovery occurs within the first three months post-injury, with a smaller amount continuing for up to 18 months or longer, which is broadly consistent with recovery slowing markedly by six months but does not support a uniform "98 to 99%" figure applicable to all patients regardless of injury completeness or level. Because outcomes differ so much by baseline AIS grade, injury level, and cause, a single blanket recovery percentage at a fixed time point is an oversimplification of the underlying evidence, even though the general direction (recovery plateaus within roughly six months to a year) is broadly supported. Status: misleading/oversimplified rather than fabricated.