A pharmaceutical research scandal revealing that clinical trials of Alzheimer’s drugs may have contained fudged and faked data has cast doubt on the companies and clinics linked to the trials, and a pall over the future of similar research, reports Science.
In 2023, American drug developer T3D Therapeutics glimpsed a breakthrough after a decade of research on a novel compound to treat Alzheimer’s disease. A clinical trial for its drug, code-named T3D-959, initially seemed to support the company’s belief that the pill stimulates the brain’s ability to process sugar and could reverse cognitive decline in the neurodegenerative condition.
But before it trumpeted the good news, the small company took a closer look at detailed data for each participant. It found “a nightmare scenario”, according to a July 2025 legal complaint filed by T3D: the results were “medically impossible”.
Some Alzheimer’s patients in the placebo group were reported as improving, even though the disease inexorably erodes cognitive abilities. Many trial participants did not even have the memory-robbing condition, the company claimed, and there was no sign of T3D-959 in blood samples from others who purportedly received it.
After excluding the data it deemed unreliable, the company said too few patients remained to confidently affirm any possible benefits for its drug. The trial, underwritten by $35m in private and federal funding, was a write-off, and the company hopes to raise funds for a new trial.
In its complaint, T3D charged Clinilabs, the contract research organisation (CRO) to which it had paid $14.5m to manage the trial, with covering up “all encompassing” fraud by five of the South Florida clinics that had recruited many of the subjects, administered the drug, and reported results.
The trial sites are all for-profit companies with small offices – or, in one case, a mobile clinic plastered with patient recruitment ads – that specialise in clinical testing rather than routine treatments.
T3D alleged they had manipulated data, boosted revenue by enrolling many candidates who did not have Alzheimer’s, and collected payments for screening out others who obviously should never have been considered.
Clinilabs and the five trial sites have denied the allegations. The CRO instead blamed T3D for a poor trial design and said the complaint was replete with “demonstrable misrepresentations”.
The dispute illuminates the potential perils of a booming industry: CROs like Clinilabs that run drug trials for many pharma companies, enlisting local clinics to find and test participants.
They have become a huge business – estimated at more than $85bn in 2024. The Miami area, with more than 6m residents, is a major player, in part because of a racially diverse, relatively elderly population that appeals to drug developers.
Trial sites in Miami alone have been involved in more than 11 000 studies, including 1 300 that are currently recruiting volunteers, according to the federal website ClinicalTrials.gov.
The problem in South Florida extends far beyond T3D’s drug, a Science investigation shows.
Other drug trials have also come to grief there, and a raft of records from the US Food and Drug Administration (FDA) show similar lapses at three of the same five sites during studies for other companies.
When such problems occur, they “can put patients in jeopardy and inhibit trust in research”, said Holly Fernandez Lynch, a University of Pennsylvania Law Professor who specialises in medical ethics and health policy.
Alzheimer’s drug development can be particularly susceptible, according to experts in clinical testing. For people seeking the fee paid to trial participants, the symptoms are easy to fake. Volunteers who actually are cognitively impaired might not understand or recognise violations of procedures.
And because cognitive tests have subjective aspects, the data are easy to manipulate.
A troubled trial industry
The lawsuit by T3D Therapeutics accuses five South Florida testing sites of falsification of data and other fraud in tests of its experimental compound T3D-959.
Science found that the sites worked on scores of other AD trials, including both approved and failed drugs.
T3D officials say its lead candidate drug looked promising in the lab. T3D-959 activates a class of molecules in the cell nucleus known as PPAR-delta/gamma receptors, which affect glucose and fat metabolism and may reduce inflammation, a factor in Alzheimer’s. Studies in lab animals modelling the brain disease showed the compound was safe and potentially effective.
T3D’s preclinical work was strong enough for FDA to approve human testing of the drug, leading to the now-disputed trial in 2021.
When it ended in March 2023, the overall findings looked promising. But when T3D reviewed the “unblinded” results – data matched to each participant – they seemed too good to be true.
Science took a closer look at the five Miami-area sites accused by T3D: IMIC Medical Research, the CCM Clinical Research Group, the Miami Dade Medical Research Institute, the Premier Clinical Research Institute, and Health Care Family Rehabilitation.
The five firms have worked on 59 other Alzheimer’s trials involving 46 drugs, including antibodies approved to treat the disease despite concerns about their safety and efficacy, and Novo Nordisk’s glucagonlike peptide-1 inhibitor semaglutide, for which the results have been discouraging.
FDA inspectors found all but IMIC made apparent errors in one or more trials. Because of redactions in their reports it was unclear in most cases which drugs had been tested, but three of the reports – including one for the semaglutide trial – noted possible enrolment of participants who did not meet study requirements.
According to records provided by Redica Systems, a company that analyses FDA reports, the agency has issued warnings or pushed for changes at some of the locations.
Suzanne Hendrix of Pentara Research, a consultant for pharma companies, said her separate, proprietary analysis on behalf of T3D validated its allegations. Hendrix declined to provide her work to Science, but according to slides she presented at a conference, several of the trial sites enrolled participants who probably didn’t have Alzheimer’s in studies for T3D, BioVie, and another company.
Hendrix told Science she has also found cases in which a site submitted the same blood sample for several participants.
She also contends that doctors involved in problematic trials often move between local clinics, and sites accused of improper practices sometimes close down, then set up shop in the same area under new names.
Claims denied
In its legal reply to T3D’s claims, Clinilabs, which specialises in managing tests of drugs and devices for the central nervous system, called for the case to be dismissed. The allegations are “a complete sham”, it said.
The five accused trial sites filed court motions to dismiss T3D’s case as vague, legally flawed, and groundless. Only IMIC provided a response to questions from Science, calling T3D’s claims “incorrect, unfounded, and baseless”.
Annovis Bio and BioVie drew on many testing locations from the same Florida area, although not the sites used by T3D. They say their trials of candidate Alzheimer’s treatments also failed to show hoped-for benefits because of similar problems.
BioVie has abandoned its Alzheimer’s programme, at least temporarily.
Last month, Bristol Myers Squibb announced that “irregularities due to clinical trial execution” at some study sites prompted the exclusion of data for its drug Cobenfy, which was being tested as a treatment for Alzheimer’s-linked psychosis.
The company said it would have to enrol new patients.
Do says he confirmed through various means that at least four of the Bristol Myers problem sites matched South Florida locations implicated in BioVie’s trial. Bristol Myers declined to provide further details.
Collusion between dishonest trial volunteers and the managers of trial sites may explain some of the apparent problems, which may extend beyond South Florida. Hendrix says her review of the scholarly literature suggests about 4% of trial volunteers at local clinics around the country are “professional patients” who fabricate or exaggerate symptoms to be paid for participation.
A 2018 review in Contemporary Clinical Trials argues that many more people pretend to participate honestly in drug studies but instead engage in “pill dumping” – surreptitiously discarding rather than ingesting the experimental treatment.
Annovis CEO Maria Maccecchini says the suspect clinical trial locations in the study of buntanetap offered cash payments and generous spreads of food and beverages to entice volunteers. Many people, she adds, pretended to suffer from cognitive impairment to join the trial, for example by intentionally drawing a clock incorrectly.
Maccecchini adds that the participants whose data had to be tossed out included many grandmothers who saw a chance to make thousands of dollars a year by simultaneously joining multiple trials for which they did not qualify.
Both BioVie and T3D say they have referred their findings about the trial sites to the FDA for investigation. The US Department of Health and Human Services, which oversees the FDA, declined to comment.
Besides compromising drug development, poor practices at some Miami-area clinical trial sites may have put subjects at risk. For example, FDA repeatedly cited Americo Padilla, a paediatric psychiatrist at New Med Research in Miami Gardens, for improperly enrolling children for a trial testing rimegepant, a drug from Pfizer.
Marketed as Nurtec ODT, it was approved to treat migraine headaches in adults, and Pfizer hoped to win approval for children. According to FDA documents, Padilla, who also worked at the Miami Dade Medical Research Institute, one of the sites in T3D’s lawsuit, failed to obtain proper informed consent from parents and could not provide FDA inspectors with lab results needed to monitor side effects.
Padilla did not reply to requests for comment and New Med says he is no longer affiliated with the trial site.
Pfizer would not comment about Padilla’s work, but told Science it tracks anomalies in trial site data and terminated New Med as a site for the rimegepant study. Trial records show the termination occurred in 2024, shortly after FDA found Padilla’s violations.
The plethora of drug trials in South Florida suggests many more may be affected. A clinical study of Eli Lilly’s antibody drug for Alzheimers, donanemab (now sold as Kisunla), at Miami-area trial sites had data problems similar to T3D’s claims, according to Hendrix.
But the trial was so large – more than 1700 subjects dispersed among 274 trial sites globally – that violations at a handful of locations probably did not compromise its overall data, she says.
Lilly declined to comment about the alleged lapses. Greely notes that even in a big trial, problematic data for a small number of participants can skew results if a drug shows only borderline effectiveness.
See more from MedicalBrief archives:
Science, not corporate zeal, must drive search for Alzheimer’s drug
Biogen halts sales, development of controversial Alzheimer’s drug
Concern over FDA’s fast-track approval of Alzheimer’s drugs
Alzheimer’s Society calls for ‘failed’ dementia drug trials to be revisited