Marketed widely in the 1950s and 1960s, diethylstilbestrol (DES) (Eli Lilly), prescribed by the medical profession for prevention of miscarriage, led to extensive harm that would prove fatal for some and would span generations. Supplied to millions of pregnant women over 3 decades, DES became the first identified cause of “prenatal drug-induced cancer in humans”. The drug was recalled in 1971. The full intergenerational impact of these prescriptions is still not known.⁵

Thalidomide is one of the saddest chapters in pharmaceutical history and an example of how premature safety claims can have tragic consequences. Created as a sedative and marketed in Germany in 1957 by Chemie Grünenthal, thalidomide would soon be launched in the UK (Distillers, UK), and many other countries would follow. At this stage, the first thalidomide-affected baby had already been born in Germany, 25 December 1956, to a Chemie Grünenthal employee. By 1958, thalidomide was licensed and promoted in the UK as a “wonder drug” to treat headaches, insomnia, and nausea in pregnant women – advertisements emphasised safety, with catch phrases such as “non-toxic” and “no known toxicity”.

The first publication to link thalidomide and birth defects appeared in 1961 in The Lancet, as a letter from an Australian, William McBride.⁶ This same year the drug was formally withdrawn in Germany and in the UK, the Thalidomide Society was established in the UK, and efforts began to secure compensation for victims. In 1968, Chemie Grünenthal was brought to trial in Germany, charged with intent to commit bodily harm and involuntary manslaughter, but in 1970 this trial was ended prematurely by the German government, who stated that it was “not in the public interest”.⁷

Efforts have continued in the UK to secure compensation from the 1970s through to the present. It was only on 29 November 2023 that the Australian Prime Minister announced a “formal national apology to all Australians impacted by the Thalidomide Tragedy”, more than half a century on from the earliest harms.⁸

Through the diligent work of FDA scientist Frances Kelsey, who demanded further safety trials prior to market authorisation, thalidomide was never approved for release in the USA. She protected an entire nation.⁹

The Selective Serotonin Reuptake Inhibitor (SSRI) antidepressant, paroxetine, became a very successful commercial product for SmithKlineBeecham (SKB) (later GlaxoSmithKline, GSK). In the late 1990s, the company conducted two randomised, controlled trials in adolescents with depression (Study 329 & Study 377). Company documents, subpoenaed through litigation, reported that Study 377 “failed to demonstrate any separation from placebo” and consequently the company had “no plans to publish data from Study 377”. Study 329 showed “trends in efficacy” but the differentiation from placebo “was not statistically significant”.¹⁰ This Study 329 was ghost written and then published by Keller and 21 co-authors in 2001, with the conclusion that paroxetine was “generally well tolerated and effective” for adolescents with depression.¹¹ Although SKB/GSK decided not to present the studies’ data to the FDA for a label change to treat adolescent depression, they used the Keller et al. publication to promote off-label prescriptions for depressed teens. Later, independent researchers gained access to raw data from Study 329 and found increased suicidality and no significant efficacy.¹² Despite calls for retraction of the original Study 329 publication, the Journal of the American Academy of Child and Adolescent Psychiatry (JAACAP) has refused to do so.¹³

GSK suppressed negative data about their drug paroxetine and effects on depression and suicide. An internal GSK document advised staff to withhold data that indicated paroxetine had no beneficial effect in adolescents.¹⁴ In 2012, GSK pleaded guilty to fraud allegations and failure to report safety data, with payment of $3 billion in criminal fines, the largest fraud settlement in US history at the time.¹⁵

There have been further disputes over the increased suicidality caused by SSRIs in adolescents and young adults, with calls to remove the FDA Black Box label. However, both Study 329 data re-analysis¹³ and separate further data support continuation.¹⁶

This GSK paroxetine chapter is by no means an isolated case of hidden data. In 2015, Eli-Lilly and Takeda Pharmaceuticals were fined $2.4 billion USD for concealment of the carcinogenic effects of pioglitazone (Actos).¹⁷

Avandia (rosiglitazone) gained FDA approval for management of diabetes in May 1999 and was widely prescribed for control of blood glucose, until it was shown to increase risk of myocardial infarction by 43% and increase risk of death from cardiovascular causes by 64%.¹⁸ In May 2007, Steven Nissen of the Cleveland Clinic published controlled trial data that showed, in the rosiglitazone group, as compared with control, the odds ratio for myocardial infarction was 1.43 (95% confidence interval (CI), 1.03 to 1.98; p = 0.03), and the odds ratio for death from cardiovascular causes was 1.64 (95% CI, 0.98 to 2.74; p = 0.06).¹⁹

In July 2007, a panel of FDA advisers voted 22 to 1 against removal of Avandia from the marketplace. As late as 2009, GSK continued with promotion of Avandia as “safe and free from cardiovascular side effects”.²⁰ In contrast, by February 2010, a US senate finance committee was able to conclude that GSK had “full knowledge of the cardiac risks of Avandia in late 2004 or early 2005”. David Graham, FDA scientist, has estimated combined US heart attacks, strokes and deaths caused by Avandia to be in the order of 100,000 events.²¹ The drug was removed from the European market in September 2010, based on cardiovascular risks, and remains banned to this day.

Pursuit of surrogate end points can be dangerous, exemplified here with a focal target of blood glucose control, yet accompanied by significant adverse events.

While such corporate products and medical prescriptions as diethylstilbestrol, thalidomide, paroxetine and rosiglitazone are now infamous chapters in medical history, still greater events loom over more recent history, and we consider two of these, Merck’s Vioxx (rofecoxib) scandal, and the roll out of gene-based COVID-19 vaccines.

Developed by Merck, the cyclooxygenase-2 (COX 2) inhibitor Vioxx (rofecoxib) marketed as a non-steroidal anti-inflammatory drug (NSAID) for pain relief in 1999, obtained FDA approval (21 May 1999) based on equivalence to other NSAIDs in short term use. Efforts to explore long term value in rheumatoid arthritis further supported sales, with fewer gastrointestinal side effects when compared with typical NSAID naproxen.²²

In this VIGOR paper,²² Merck concealed adverse cardiovascular events in the Vioxx arm of the study that would prove to be a serious statistical signal. Just prior to publication, Merck informed the FDA of three adverse cardiovascular events, published on an FDA website, but The New England Journal of Medicine (NEJM) article was neither retracted nor corrected.

The full VIGOR data unmasked high rates of cardiovascular events with Vioxx (rofecoxib) compared to naproxen, with a relative risk of 2.38 (95% CI 1.39–4.00) for rofecoxib against naproxen over a 12-month study period.²³ The time lag between initial FDA approval and the appearance of this more complete VIGOR trial data in print was over 18 months.

Initial responses to this data from Merck included claims that naproxen had a protective effect against heart attacks and strokes, that was not possessed by Vioxx, and that the increased cardiovascular risks seen with Vioxx occurred only in people with known cardiovascular disease.²⁴ This was later found to be untrue, once data for healthy individuals who had suffered harm on Vioxx had been uncovered.

Merck tried to influence lead American physicians with support and finance for research, and they defamed, withdrew support, and tried to discredit or “neutralise” those who failed to promote use of Vioxx, a matter uncovered by the Federal Court in Melbourne, Australia.²⁵ In contrast, the Chair of the Study Data and Safety Board (SDSB) for the study, Michael Weinblatt, owned $72,000 in Merck stock and was on a $60,000 contract for 12 days’ work for the company.²⁶

Internal Merck emails are now known to have shown as early as 18 November 1999 (unblinded minutes), that an interim safety analysis of VIGOR showed excess deaths and cardiovascular adverse experiences – 79 cardiac events for rofecoxib compared with 41 for the control group on a traditional NSAID, naproxen.^26,27 Yet Merck made a press release on 22 May 2001, entitled “Merck Reconfirms Favourable Cardiovascular Safety of Vioxx”. Merck even created a “fake journal” with the medical publisher Elsevier: The Australasian Journal of Bone and Joint Medicine, with six issues between 2002 and 2005, that collated articles favourable to Merck’s drugs Vioxx and Fosamax.²⁸

The FDA appears to have been complicit with Merck in early suppression of the adverse event data of VIGOR. Eventually the FDA did instruct Merck (April 11^th, 2002) to include a precaution about cardiovascular risks in their package insert.²⁴ Dr David Graham, an FDA scientist in its Office of Drug Safety, revealed this interplay in his testimony to the US Senate (below).

Vioxx remained on the market until the completion of the APPROVE study in 2004. The intention was to promote use of Vioxx to treat polyps of the colon. But again, the drug demonstrated at least double the cardiovascular risk compared with placebo, this time in a patient population considered to be at low risk of cardiovascular disease.²⁹

Merck announced withdrawal of Vioxx on 30 September 2004, the largest prescription drug recall in history to date.

Over 20 million people in the US are believed to have taken the drug, of whom an estimated 88,000 to 139,000 suffered myocardial infarctions, with 30–40% fatality rate (testimony of Dr Graham to the US Senate).³⁰ His figures on estimated cardiac arrests were also published in The Lancet, despite opposition from the FDA.³¹ Dr Graham further testified to the Senate that conflicts of interest at the FDA had delayed the Vioxx recall.³² Discovery documents in litigation reveal corporate pharma may conceal data early, at any cost to achieve market growth.^33,34 Here the FDA appeared complicit and slow to withdraw the product.^24,35 Published in the NEJM, prominent cardiologist Eric Topol included strokes as well as myocardial infarctions to estimate 160,000 events per 10 million people prescribed Vioxx, and he noted a global cohort of up to 80 million had been prescribed Vioxx.²⁴

By August 2005, 13,000 class action lawsuits had been filed against Merck. By November 2007, Merck had created a settlement fund of $4.85 billion USD, the largest ever in US history at the time. Merck agreed to compensate victims in exchange for a no-fault agreement – specifically, no legal admission of fault. Yet payment of $4.85 billion USD in compensation to claimants could clearly be interpreted as an admission of fault.^25,26

When the Vioxx scandal broke, Merck had a capital market value of between $40 and $50 billion USD. Despite the greatest drug scandal in the world, enormous fines and atrocious damage to image, Merck has continued to grow in the last 2 decades and has increased its value six-fold to over $300 billion USD.

Initially marketed December 2020, as Emergency Use Authorisation (EUA) in the USA, and provisional authorisation in Australia and other nations, the gene-based COVID-19 vaccines of modified mRNA type, (Pfizer-BioNTech’s BNT162b2, Moderna’s mRNA-1273) and viral-vector-DNA type (AstraZeneca’s ChAdOx1-S, Janssen’s Ad26.COV2.S, Gamaleya’s Sputnik V) have constituted the majority of over 13 billion doses of all COVID-19 vaccines.^36–41 In contrast, COVID-19 vaccines that employ traditional well-tested inactivated virus or recombinant protein antigen-based technologies have been utilised mainly in a few non-Western nations (e.g., Bharat Biotech’s Covaxin, Sinovac’s CoronaVac, Cinnagen-Vaxine’s SpikoGen, Cuba’s Genetic Engineering and Biotechnology Centre’s Abdala).⁴²

Purposed for protection against transmission of the SARS-CoV-2 virus and reduced disease severity, official sales narratives included – “safe and effective”, and “millions of lives saved”. Indications of serious harm appeared from 2021 with record high adverse event reports to pharmacovigilance. These included suspected death reports as indicated by VAERS data⁴³ (Figure 1), peer-reviewed VAERS and EudraVigilance data,⁴⁴ excess mortality above expected from collation of official death statistics by Our World in Data⁴⁵ and insurance data for excess mortality and disability⁴⁶ correlated with COVID-19 vaccination. Montano (2022) compared COVID-19 vaccines (Janssen, Moderna, Pfizer-BioNTech) with influenza vaccines, and found extremely high elevated relative risk for serious and fatal adverse events across most organ systems [⁴⁴, in Table 3b]. Excess mortality is defined as mortality above normal background rates at ourworldindata.org which is under the jurisdiction of Oxford University, UK.