By David Tuller, DrPH

From the start, one of my strategies for this PACE-debunking project was to draw in outside experts–people with no ax to grind and no pre-conceived notions about the trial and its methodology–and encourage them to scrutinize the matter. It was telling that many well-regarded scientists and researchers were willing to make scathing public comments about PACE and the related claims being made by the CBT/GET ideological brigades. When Bruce Levin, a professor of biostatistics at Columbia, calls something “the height of clinical trial amateurism,” people should pay attention and stop maligning PACE critics as being irrational, anti-science or vexatious.

That’s why I’m delighted that Simon Wessely himself, in a recent Twitter exchange, invited Mike Godwin, an American attorney and social commentator, to review the PACE trial controversy. After scrutinizing the published record, including the PACE research and the special issue of the Journal of Health Psychology dedicated to the issue, Godwin pronounced PACE to be “so profoundly flawed that it cannot be trusted.” In response, Sir Simon attempted to re-direct the narrative, reiterating his own positive beliefs about the trial. What he fails to understand is that, despite his knighthood and his widely hailed courage in “standing up for science,” his beliefs and opinions are irrelevant here. In this case, the facts are what count.

And when outside observers assess the facts about PACE, they see what the critics see–an uninterpretable mess. Mike Godwin is known for the so-called “Godwin’s Law of Nazi Analogies,” a trenchant observation that he described this way in a 1994 article in Wired: “As an online discussion grows longer, the probability of a comparison involving Nazis or Hitler approaches one.” Almost 25 years later, this observation has retained its relevance and social currency.

In any event, the embroilment of Godwin in the PACE debate can be traced to Professor Michael Sharpe. He has been engaged for weeks in a relentless but ultimately incoherent pro-PACE tweeting campaign that has once again exposed his inability to defend the study’s methodological flaws. Recently, in response to something someone tweeted in one of these tweet-chains inspired by yet another Sharpe head-scratcher, a PACE critic referenced “the banality of evil”—the brilliant but exhausted Hannah Arendt phrase that emerged from the philosopher’s coverage of the Eichmann trial in Jerusalem. At that point who should jump into the tweet-fray but Sir Simon himself.

(I can’t see Sir Simon’s tweets. He blocked me a year or two ago when I tweeted that some drivel and nonsense he’d tweeted was in fact “drivel and nonsense,” or it might have been “nonsense and drivel.” Screen-shots of his recent tweets were posted on a patient forum.)

Apparently triggered into action by the Holocaust allusion, Sir Simon tweeted out to Godwin with an oblique reference the Nazi-analogy “law” that bears his name. As it turns out, the two became chums decades ago when the young Simon attended an American high school as an exchange student. Who knew? Alas, bonds forged back then in chem lab or drama club were not enough to prevent Godwin from offering a blunt appraisal of the £5,000,000 garbage heap that Sir Simon has called “a thing of beauty.” (Godwin assured me that the two men remain friends despite their disagreement about PACE.)

Sir Simon has defended PACE up and down the queen’s realms, and he appears convinced of his own correctness. As he is fond of reminding people, he wrote the book on clinical trials so he knows something about them. (Can someone check whether Sir Simon’s book covers whether investigators can weaken outcome thresholds after data collection in ways that dramatically boost reported results?) But he appears blind to how others view the study. Did Sir Simon think his school buddy would end up bolstering his side of the argument? If so, he miscalculated.

Godwin did more than declare the trial to be “profoundly flawed.” In his comments, he touched on one of the main negative consequences of the PACE debacle—the fact that the trial and/or its CBT/GET recommendations have influenced how disability insurers and government social welfare agencies make their determinations. Here’s what Godwin, who also watched some of the recent Scottish parliamentary hearings on the issue, tweeted about that aspect of the debate: “It seems clear to me that the PACE trial will continue to be used to deny patients benefits. That sucks.”

Since Sir Simon invited Godwin into the conversation in the first place, the quick smack-down of PACE likely stung. But he managed to present a brave front in the face of the humiliation. In a series of tweets, he wrote back: “You won’t be surprised mike to find that I don’t agree with this…I continue to believe they did a good job with the trial and that it was carried out to a high standard the data is sound even if the results are relatively modest.”

Sir Simon even tweeted to Godwin the “ocean-liner PACE defense” essay posted by The National Elf website in late 2015, shortly after Virology Blog published my 15,000-word investigation. Despite his apparent fondness for this essay, the ocean-liner metaphor is a disaster for his argument, as Northwestern law professor Steven Lubet has noted. It is not just because of the inevitable Titanic association, although that’s a problem. But his account of how the navigators made mid-trip adjustments so the ocean-liner called PACE could reach its pre-determined destination raises further questions. To extend the metaphor, the goal of a clinical trial is to reach whatever destination you reach by following the course you committed to at the outset, not to reach a pre-determined destination by changing your course mid-way through—which is of course what happened in PACE.

Poor Sir Simon. He’s a smart guy but appears to remain mired in misconceptions. He doesn’t grasp that his defense of the PACE trial is intellectually and ethically bankrupt. Now he has unwittingly solicited some unvarnished comments about the trial from a childhood friend. Perhaps Godwin’s honest assessment will finally help Sir Simon accept with grace and humility what experts around the world have recognized: PACE is a piece of crap.

(Post-script: I wrote most of the above on Saturday. Afterwards, Professor Sharpe rejoined the conversation. I responded to his absurd comments in several tweets, which only reinforced my conviction that he is unable or unwilling to understand the legitimate concerns about PACE. In such circumstances, efforts at dialogue and communication seem pointless. On top of that, Professor Sharpe has now blocked me on Twitter, like his colleague Sir Simon.)

This morning I sent the following e-mail to key NICE executives involved in the effort to overhaul the clinical guidance for ME/CFS. I cc’d several stakeholders in the process as well.


Dear Sir Andrew and Professor Baker (and others)—

In an e-mail to Sir Andrew last year, I posed some questions concerning the involvement of NICE with the Improving Access to Psychological Therapies program being rolled out across England by the National Health Service. According to the NICE website, the organization is working with IAPT to “assess digitally enabled therapies for anxiety, depression and medically unexplained symptoms which offer the potential to expand these services further.”

Here is part of that e-mail from last fall:

Chronic fatigue syndrome is included in the list of thirteen conditions identified by “the NICE expert IAPT panel” as appropriate targets for interventions developed through this program. The IAPT site indicates that these interventions need to be consistent with NICE guidance. The site then refers readers to CG53 [the 2007 NICE guidance for CFS/ME] but makes no mention that this guidance is undergoing a full update.

Why is chronic fatigue syndrome still listed under the IAPT program? Does the Guidance Executive agree that it should be removed, since it is not a psychological or psychiatric disorder and the use of CBT and GET is very much under question? If it is not removed from under the auspices of IAPT, does that mean NICE intends to encourage the development of digital methods of delivering CBT and GET to people with CFS/ME, even as the guidance itself undergoes a full update?

Despite positive developments in the NICE process since then, these questions remain. IAPT’s interest in digitally delivered therapies raises particular concern among ME/CFS patients and advocates because it suggests the possible or even likely adoption or endorsement of some version of FITNET-NHS. That is the Bristol University trial investigating online CBT for children, based on the purported success of earlier Dutch research. Critics of the PACE trial, including me, have documented serious methodological weaknesses in both the Dutch study and the U.K. version.

Last year, before NICE announced its decision to pursue a full update of the guidance for this illness, FITNET-NHS was being cited as a study of interest to be considered in a future review. But given the identified deficiencies of this research into online CBT, it has no place being considered during the development of health policy or clinical guidelines for ME/CFS.

As the NICE process for producing the new ME/CFS guidance moves forward, so does IAPT’s implementation of its expansion strategy. IAPT’s recent publications contain no indication, as far as I can tell, that anyone from NICE has communicated with anyone from IAPT about the paradigm shift occurring in perceptions about and treatment of ME/CFS. Those involved with IAPT seem not to have noticed yet that the CBT/GET approach is rapidly losing credibility worldwide.

Does NICE have any current plans to communicate with IAPT about ME/CFS, and specifically about IAPT’s ongoing roll-out of CBT and GET services for these patients? Or will such communications need to wait until the new guidance is released in 2020? Does NICE at least plan in the interim to suspend its cooperation with the NHS on developing digitally enabled or non-digitally enabled IAPT services related to ME/CFS? Since a new NICE guidance is in the works, developing any ME/CFS services for IAPT based on what will soon be an outdated document would seem to be a questionable way to invest limited public resources.

Thanks for your attention to this matter.


David Tuller, DrPH
Senior Fellow in Public Health and Journalism
Center for Global Public Health
School of Public Health
University of California, Berkeley

Vincent, Kathy and Rich travel to ASM Microbe 2018 in Atlanta where they speak with Stacy Horner and Ken Stapleford about their careers and their research.

Click arrow to play
Download TWiV 498 (42 MB .mp3, 68 min)
Subscribe (free): iTunesRSSemail

Become a patron of TWiV!

Show notes at

IFN signaling

Type I IFN receptor binding and signal transduction.

By Gertrud U. Rey

Zika virus (ZIKV) infection causes microcephaly in newborns and is causally associated with Guillian–Barré syndrome in adults. To date, there are no drugs available to prevent or treat ZIKV infection. ZIKV vaccine research is challenging because adult immunocompetent mice are resistant to ZIKV infection and disease.

The primary immunologic response to ZIKV infection is the production of type I interferon, which binds to the interferon alpha/beta 1 receptor (IFNAR1) and activates signal transducer and activator of transcription (STAT) complexes (illustrated). Activation of this pathway leads to the production of hundreds of antiviral and immunomodulatory interferon-stimulated genes and a general antiviral state. In humans, ZIKV evades the innate immune response by degrading STAT2, thereby antagonizing type I interferon (click here for a review of how viruses interfere with interferon). However, in mice ZIKV is incapable of binding STAT2 and triggering its downstream transcriptional effects, possibly explaining the resistance of immunocompetent mice to ZIKV. Consequently, most mouse studies of ZIKV pathogenesis have been done in mice with genetic or acquired deficiencies of interferon signaling. Unfortunately, such models don’t mimic human disease sufficiently to make them suitable for evaluating vaccines and therapeutics.

Using a two-step approach, a group at Washington University School of Medicine has developed a new system to study ZIKV infection in mice (link to paper). First, they generated a mouse-adapted ZIKV strain by passaging the African strain ZIKV-Dak (parental strain) in mice that have an intact type I interferon response but lack mature B and T cell responses. Second, they developed a fully immunocompetent mouse model of ZIKV infection by introducing human STAT2 into the mouse Stat2 locus.

After passaging the parental virus strain four times, the authors obtained ZIKV-Dak-MA (MA indicates mouse-adapted), a strain containing three mutations. The first was a silent mutation in the nucleotide sequence coding for E protein, while the other two were non-synonymous mutations of lysine to arginine, and glycine to arginine at positions 399 and 18 of the NS3 and NS4B proteins, respectively. To test the infectivity of this mutated viral strain, wild type mice were first treated with an anti-Ifnar1 monoclonal antibody and then infected. Compared to the parental strain, ZIKV-Dak-MA was more lethal, with a higher level of replication in the mouse brains.

This higher infectivity in brains may be due to an increased ability of the mutant virus to cross the blood-brain barrier or because it can replicate better in neurons. To test these possibilities, the authors infected mice intracranially with mutant or parental virus in the absence of anti-Ifnar1 antibody treatment. Intracranial infection would bypass the blood-brain barrier, and omitting anti-Ifnar1 antibody would lead to cessation of viral replication before systemic dissemination. Although neither virus caused a lethal infection, mutant viral RNA accumulated to greater levels in multiple regions of the brain, compared to parental virus RNA. In contrast, intracranial infection of mice deficient in IFNAR1 receptor with either strain resulted in minimal differences in viral RNA produced by the two strains. This result suggests that the difference in replication between the two viruses is likely mostly due to different abilities to evade type I interferon responses.

To determine which of the two non-synonymous mutations was responsible for enhanced infection and disease, the authors engineered viruses with the non-synonymous mutations in either NS3, NS4B, or both. Mice were then treated with anti-Ifnar1 monoclonal antibody and challenged with either of the mutant viruses. Viruses with mutations in either NS4B or both NS3 and NS4B were about 95% lethal in mice, while the virus with the NS3 mutation only caused about 5% mortality. The authors obtained the same result in mouse and human neural stem cells, suggesting that the NS4B mutation is necessary and sufficient for enhanced virulence.

Single-cell RNA sequencing analysis of neural stem cells infected with the parental or the NS4B mutant virus revealed that the NS4B mutant is less effective at inducing interferon-stimulated genes and type I interferon responses compared to the parental virus. To determine whether ZIKV modulates type I interferon signaling responses, the authors pre-treated murine neural stem cells with either anti-Ifnar1 antibody or exogenous IFN-b and then infected the cells with parental or NS4B mutant virus. Viral yields produced by the NS4B mutant were higher in the absence, and lower in the presence of anti-Ifnar1 antibody. However, pre-treatment of cells with IFN-b caused the two viruses to replicate to similar levels, suggesting that the NS4B mutation likely results in diminished levels of IFN-b. The fact that the NS4B mutant was still sensitive to exogenous interferon suggests that it may be unable to bind to and degrade mouse STAT2.

In the second major part of this study, the authors generated a fully immunocompetent mouse model of ZIKV infection by introducing human STAT2 into the mouse Stat2 locus. Infection of these human STAT2 knock-in mice with ZIKV-Dak-MA led to 30% mortality, with higher viral levels observed in the spleen and brain. In contrast, none of the mice inoculated with parental virus died. In addition, pregnant ZIKV-Dak-MA recipients passed the virus to their babies. Higher levels of ZIKV-Dak-MA were detected in the placenta and fetal heads of STAT2 knock-in mice. The authors concluded that combining the adapted ZIKV strain with human STAT2 knock-in mice produces a model system that mimics the human mechanism of ZIKV pathogenesis.

The lack of a suitable immunocompetent mouse model for ZIKV infection has been an impediment to research progress in this field. The model system presented here has drawbacks, such as the need to use a specific viral mutant, and a disparity of phenotype observed in peripheral organs and the brain following subcutaneous inoculation with either mutant and parental strains. However, it presents an additional resource for determining mechanisms of pathogenesis, defining correlates of innate and adaptive immune protection, and for developing drugs.

By David Tuller, DrPH

Ten years ago, the National Health Service began rolling out across England a program called Improving Access to Psychological Therapies, or IAPT. This program arose out of the notion that many people were suffering from untreated depression, anxiety and other psychiatric disorders. In parallel with that, research suggested that treating these ailments would not only be good for public health but would in turn help restrain overall medical costs. The most common but not the only psychotherapeutic intervention offered by IAPT services is cognitive behavior therapy.

[click to continue…]

By David Tuller, DrPH

This morning I sent three more e-mails alerting interested parties to my concerns about two BMJ studies of children with ME/CFS. When it comes to research, kids are already a vulnerable population, and those with a stigmatizing illness even more so. That’s why it is both surprising and troubling that BMJ appears to have little interest in addressing the ethical and/or methodological violations that Virology Blog has documented in these pediatric studies.

[click to continue…]

TWiV 497: Europic 2018

At Europic 2018, a meeting on picornaviruses in the Netherlands, Vincent speaks with Sasha Gorbalenya, Jim Hogle, Ann Palmenberg and Frank van Kuppeveld about their careers and their research.

Click arrow to play
Download TWiV 497 (36 MB .mp3, 60 min)
Subscribe (free): iTunesRSSemail

Become a patron of TWiV!

Show notes at

N-terminal myristoylation

N-terminal myristoylation. An amide bond links myristate to an N-terminal glycine in the myristoylation site consensus sequence.

The common cold is an infection of the upper respiratory tract that may be caused by many different viruses, but most frequently by rhinoviruses. A compound that inhibits a cell enzyme and blocks rhinovirus replication has the potential to be developed into an antiviral drug (link to paper).

[click to continue…]

By David Tuller, DrPH

I have sent the following e-mail to Edward Sykes, the head of mental health and neuroscience at the Science Media Centre. The e-mail concerns the Lightning Process study published last year in Archives of Disease in Childhood. The SMC promoted the findings, which received widespread media coverage.

[click to continue…]

By David Tuller, DrPH

I am not sure what is going on at BMJ and why editors there seem incapable of acknowledging their flawed decision-making when it comes to two papers that should never have been accepted for publication. One violated BMJ’s policy that all trials must be properly prospectively registered, with no participants recruited beforehand. The other violated official U.K. guidelines by exempting itself from ethical review on false grounds.

[click to continue…]