Guest contribution by Leo Goldstein
- The use of remdesivir for COVID-19 has been approved by the FDA based on a single RCT conducted by NIAID with the participation of Gilead Sciences, the exclusive maker of remdesivir. A final report from this study was released on October 8, five months after the drug was approved.
- The final report shows that at least 35% of patients were treated with hydroxychloroquine, likely with azithromycin. The data in the final report suggest that hydroxychloroquine, rather than remdesivir, was the main factor driving the patients in this study.
- Nothing in the study supports the hypothesis that remdesivir is an effective antiviral for SARS-COV-2.
- The study's own figures show a connection between RDV and increased mortality in the most severe patients. It is also possible to conclude that RDV is net harmful to most hospital patients.
- The trial was conducted and reported with several shortcomings including:
- The study was not double-blind, but was reported as such
- The pre-registered protocol was changed several times during the course of the experiment
- The primary result was changed in the middle of the study, apparently because the researchers noticed a lack of effectiveness of their drug when trying
- The outcome measures were subjective and not reliable
- The study had a conflict of interest, compounded by the design that gave NIAID and Gilead an impact on the hospitals and doctors who treated patients
- At least three of the study's authors and authors have recently reported no grants and / or personal fees received from Gilead.
This study, also known as the Adaptive COVID-19 Treatment Study (ACTT-1), was registered as NCT04280705 and published by the National Institute for Allergies and Infectious Diseases (NIAID) under the direction of Dr. Anthony Fauci performed. The study began as a "multicenter, adaptive, randomized, blinded, controlled trial of the safety and effectiveness of investigational therapeutics for the treatment of COVID-19 in hospitalized adults," but quickly became a phase 3 study for remdesivir. It was reported under the title "Remdesivir for the Treatment of Covid-19 – Preliminary Report" (Beigel – Lane PR, 2020). The lead author is H. Clifford Lane, co-chair of the NIH COVID-19 Treatment Guidelines Panel, and Associate Director of NIAID.
This study was the only study to support the emergency approval of remdesivir (RDV) for the treatment of COVID-19. The updated version "Remdesivir for the treatment of Covid-19 – final report" (Beigel – Lane FR, 2020) (also "the paper") was published on October 8th and is discussed here.
The assumption that remdesivir is effective against SARS-COV-2 is likely based on the similarity of its in vitro effects to that of chloroquine, without taking into account that chloroquine and hydroxychloroquine build up in the lungs while RDV does not ( Goldstein, 2020). .
Unless otherwise stated, table and figure numbers refer to the supplementary appendix (Beigel – Lane FR, 2020), which was published on October 11 from https://www.nejm.org/doi/suppl/10.1056/NEJMoa2007764/suppl_file/ nejmoa2007764_appendix.pdf was downloaded. 2020.
Deadly methodological flaws
Each of the various study errors described below will void the results of this study.
Not double blind
The abstract of the paper falsely claims that the study was double-blind. In fact, the study wasn't double-blind. The doctors at the European site and at "some" other sites were not blind and knew what they were administering.
Too many websites
Contrary to common sense and best methodological practice (Kraemer, 2000), the study was carried out at 60 locations plus 13 sub-locations. Such a multi-site design is a potentially misleading “centralized multi-center collaborative RCT” (Kraemer & Robinson, 2005).
Note that fewer than 600 Remdesivir courses were spread across 73 sites, with an average of 9 RDV courses per site – a small sample enough to satisfy appetites but not enough for hospitals to draw their own conclusions about efficiency can pull.
Taking into account the deliberate unblinding of the treating physicians and the limited supply of medicines, the sites could have been given incentives to compete for the best results from RDV and could be rewarded with a prioritized supply of medicines in the future. RDV was considered a miracle cure, and access to it was priceless.
Arbitrarily removing websites from the final statistics
The May 6 update removed two websites from the registry:
Rocky Mountain Regional Veteran Affairs Medical Center – Infectious Disease Department, Aurora, Colorado, USA, 80045
University of Florida Health – Shands Hospital – Department of Infectious Diseases and Global Medicine, Gainesville, Florida, USA, 32610
These two sites began enrolling patients between April 2nd and April 15th, as indicated by the April 16th enrollment status. Registration at all locations ended on April 19th. Neither location was mentioned on Beigel Lane (both versions, including attachments), and their removal was not explained. This raises the suspicion that these spots were removed due to undesirable results.
Simultaneous use of other treatments
Administration of HCQ before and / or after RDV treatment was allowed. Co-administration of RDV with HCQ was also allowed in some locations (which included a written policy on the use of HCQ for COVID-19) but prohibited in other locations.
~ 35% of patients in both the remdesivir and placebo groups also received hydroxychloroquine (Table S3). ~ 80% of the patients received antibiotics but the paper does not specify which patients. 93% of patients were recruited on March 22 or later after President Trump tweeted about HCQ & azithromycin. In late March – early April, HCQ + AZ was the standard of care in some countries (Sermo, April 15). New York State required that patients be hospitalized to receive HCQ-based treatment. Even before the president's tweet, azithromycin was widely known to have some effect against the coronavirus, and therefore his purchases at the expense of other antibiotics had risen sharply (Vaduganathan et al., 2020).
The paper does not provide information on the use of zinc and vitamin C, both of which have been widely used in COVID-19 treatments (Sermo 2020, April 9).
Significant changes to the registered log
The protocol was changed many times during the study. The primary result was changed from “Percentage of Subjects Giving Each Severity Rating on an 8-point Grading Scale” to “Time to Recovery” on April 8th.
The relevant part of the order scale is:
8 – death (accordingly removed from the scales and counted separately)
7 – hospitalization, invasive mechanical ventilation or extracorporeal membrane oxygenation (ECMO)
6 – Requires hospitalization, non-invasive ventilation, or use of high flow oxygen equipment
5 – Admitted to the hospital, requiring additional oxygen
4 – in the hospital, without supplemental oxygen, but with ongoing medical care
"Recovery" is defined as a score below 4 that is normally discharged from hospital.
Incorrect result measurement
There is a problem with the outcome measurements on this scale. The results reflect subjective decisions made by the doctor, e.g. B. “Receive invasive mechanical ventilation”, “Require non-invasive ventilation” (here is the equivalent of receiving) and the like. These values cannot be measured objectively, such as blood oxygen content or heart rate. And the doctor is likely not blind and motivated to rate RDV as superior to a placebo. Most of the secondary outcomes were also defined by the ratings on this scale.
Lack of basic information
The paper did not report baseline statistics for patients. Instead, subjective ratings were given on the same grading scale. Selected comorbidities have also been reported.
Unexplained data incongruence between the preliminary report and the final report
In terms of data completion, the figures in the final report differ significantly from the figures in the preliminary report.
No benefits from RDV have been shown
The average mortality in the RDV arm was not much lower than in the placebo group until the first measurement date – 15 days after the start of treatment. After that, death rates were reversed. At the end of the study (day 29), the reported all-cause mortality in the RDV group was 11% versus 15% in the placebo arm (according to Table S12). This difference is due to the different initial conditions and data manipulations by the authors. Tables S13 and S15 support this conclusion.
Table S15 shows the differences in scores between the remdesivir and placebo arms on selected days. Immediately after randomization, a small difference develops in favor of RDV, which remains almost the same for some time. This suggests that patients do not improve because of RDV, but that the initial selection or evaluation of patients is manipulated to be positive for RDV. After a change of only 0.1 points over eight days of treatment (RDV treatment is carried out for up to 10 days), the difference jumps by 0.3 points (from day 11 to day 15) within four days without treatment – just in time to measure the result a day fifteen!
This table excludes deaths, layoffs and withdrawals.
Table 15 shows the results according to the objective national early warning value. Accordingly, the symptoms of the RDV arm at the start of treatment are better than those of the placebo arm: 5.7: 6.1 (the higher the value, the worse the symptoms). Similar to the previous table, the difference increased in the first few days, mainly due to a worsening of the placebo arm, and was 5.6: 6.6. Then the difference begins to decrease and on the 11th day it becomes 6.3: 6.6 (worse for RDV than at the beginning of the study), then increases until the 15th day. This table excludes deaths, layoffs and withdrawals.
It is possible that when administered early, RDV may have measurable antiviral effects against SARS-COV-2 in humans, but this was not demonstrated in this study.
RDV increased deaths in the sickest patients
Even the manipulated data cannot hide the damage RDV has done to the sickest patients. Unfortunately, remdesivir was approved and recommended for the treatment of the sickest patients before the study was completed.
The following is the reported mortality up to day 29 (from Table S5):
Score 7 21% 19%
Score 6 20% 20%
The real odds are even worse for RDV. The following picture is graphic E from Figure S5. It is a Kaplan-Meier survival estimate in the Score 7 group (the heaviest group) per day.
The trick is easy to see. RDV patients seem to have a slightly better chance of survival up to day 15, the first reported day of measurement. After day 15, their chances drop sharply and become much worse than the placebo arm. The difference decreases again on the 26th day shortly before the second measurement day.
More worryingly, this graph is very different from the graph published in the preliminary report (revised May 22nd and revised May 26th), which shows even worse performance of the RDV arm.
This graph shows not only the sharp reversal in mortality rates after day 15 in favor of the placebo arm, but also a sudden narrowing of the gap on day 26, three days before the final reporting day, day 29. No explanation for these data after the study changes was given.
Hydroxychloroquine + azithromycin
Table S8 is intended to analyze the patients who have not taken hydroxychloroquine. Just 46% (486 out of 1062, both arms) of patients recovered by day 29 without the use of hydroxychloroquine. This percentage is much more significant than any remdesivir percentage.
This table shows the mean time to recovery (MTTR) organized by the time the RDV or placebo was administered after symptoms appeared. Patients were randomized to the first quartile (“Q-I”) and treated for 6 days or less after onset of symptoms; the fourth quartile (“Q-IV”) was randomized to be treated 13 days or later. Only patients who have recovered by day 29 are included. The remainder (~ 30%) are those who died, stopped, or did not recover by day 29.
The Q-III (10 to ≤ 12 days) had the shortest recovery time – 7 days. If RDV were an effective antiviral for SARS-COV-2, the Q-I would have the fastest recovery as it was recognized early. The recovery would be significantly longer in Q-III when the viral stage is over in most patients. However, the Q-III has the shortest MTTR for RDV. This contradiction alone belies the hypothesis that RDV is a potent antiviral agent.
Another thing noteworthy is that the MTTR for the placebo group in Q-I is 24 days, which is 1.5-2x longer than for other quartiles. Placebo would be expected to have similar times to recovery or a monotonous change.
Both contradictions are resolved by accepting that many patients other than RDV have received effective antiviral treatment.
- Most of the patients randomized to Q-I did not receive this other antiviral treatment and therefore had the longest average recovery time.
- Patients randomized to Q-II and Q-III were more likely to receive the other antiviral treatment early, after symptoms appeared, and before treatment with remdesivir. Therefore, both RDV and placebo arms recovered the fastest in Q-II and Q-III.
- Q-III's RDV arm had the shortest MTTR of just 7 days, meaning that most patients in this group did not receive the full 10-day course of toxic RDV. This effect, resulting from receiving less remdesivir, created the gap between him and the other groups.
- Q-IV was less likely to receive the other antiviral treatment or to be late and have a longer MTTR
- In each quartile, the RDV arm reported results than the “placebo” arm based on patient selection or some other manipulation
Note that someone who is convinced that HCQ has no effect may interpret this statistic as evidence that HCQ interferes with the effects of RDV.
HCQ and RDV
Gilead falsely and dishonestly claimed that HCQ interfered with the antiviral effects of RDV, and the FDA slavishly repeated those claims. It is explained above how the statistics of trials using both HCQ and RDV can be misinterpreted in this way.
In addition, Gilead presented the results of a laboratory test in a cell culture in which chloroquine phosphate interfered with the conversion of RDV into what Gilead called "Remdesivir" Triphosphate"(GS-441524 triphosphate or GS-443902). This result is not related to HCQ, which is taken as hydroxychloroquine sulfate.
WHO solidarity process
The WHO has just published a preprint of the preliminary report (WHO, 2020) from the solidarity study of four antiviral drugs against COVID-19. It confirmed the ineffectiveness of RDV given as recommended by Gilead. This is despite the fact that Gilead is well connected with and makes significant contributions to the WHO.
The solidarity report has also claimed HCQ's lack of effectiveness, but HCQ was administered in toxic doses using the same regimen as in the RECOVERY study – 2,400 mg in the first 24 hours (6x recommended dose), 800 mg for the next 9 Hours days (2x recommended dose). These doses are 4-5x higher than the recommended doses for COVID-19 (and even for malaria).
Gilead Disclosed and Undisclosed Conflicts of Interest
- Thomas F. Patterson, M.D. – Gilead Adviser (Personal Fees) in 2019
- William R. Short, M.D., M.P.H., – Adviser to Gilead (personal fees) (reported in 2019 and 2018)
- Norio Ohmagari, M.D., Ph.D., – One of the researchers in another Gilead-sponsored study of RDV for COVID-19 (Grein et al.) That completed in early March. This counts as a research grant.
The following grants and personal fees received by Gilead over the past 36 months have been published by the following report authors:
- Anu Osinusi – Gilead Sciences, Inc. Employee
- Thomas Benfield – Grants and Personal Fees
- Gerd Fätkenheuer – grants and personal fees
- Roger Paredes – Grants and Personal Fees
- Anne Luetkemeyer – Grant (s)
- Sarah Pett – Grant
- Giota Touloumi – Grant
Inexplicable late start of treatment
It is surprising that a drug intended to act as an antiviral agent was administered far too late. Only 25% of patients started RDV treatment within 6 days of symptoms appearing. 75% of patients started too late for antiviral treatment between 7 and 34 days after symptoms appeared.
With a few exceptions, the aforementioned NIAID / Gilead researchers who conducted this study appear to have most of their experience with HIV and attempts to use RDV in hepatitis C or Ebola instead of acute respiratory infections. Hence, some decision makers could not be fully aware of the two-tier dynamics of COVID-19. The viral phase is followed by the immune response phase, and the overreaction of the immune response is more dangerous than the virus. The immune response reduces viral load and influence, so that these phases overlap. Their overlap is often classified as a separate phase (Siddiqi – Mehra, 2020, Fig. 1).
It should be noted that early administration of RDV cannot be routine as RDV is administered by infusion (not injection) and therefore hospitalization is required. RDV is not safe and its dangers are unknown. This justifies its use only if the diagnosis is certain and the patient is sick enough. After all, you can't charge $ 3,000 for cough and fever treatments.
RDV & COVID-19 Severity
In COVID-19, the phase and severity of the disease are routinely confused. Patients in the early (viral) phase are routinely classified as mild. Patients with a severe late-stage immune overreaction (immune response) are routinely classified as serious.
Antiviral drugs are useless after the virus phase. Hydroxychloroquine is useful in all phases of COVID-19 as it has antiviral effects against both SARS-COV-2 and immunomodulator.
Even if RDV were an effective antiviral agent, its only effect in the immune response phase is toxicity (Zampino et al., 2020). It is expected to be especially dangerous for heavy patients. Even the manipulated data show higher mortality rates in the RDV arm of the most severe subgroup.
Suspected differentiated care
At the start of the study, the placebo arm was slightly worse than the RDV arm: 30% versus 24% of the patients with invasive mechanical ventilation (IVM, also called intubation) or ECMO according to Table S1.
Table S3 shows that the percentage of placebo versus RDV patients who received ECMO or intubation during the treatment process was 45.5%: 34.6% (as treated versus the intent-to-treat population; slightly smaller) . This is a 52% increase in placebo versus 44% in the RDV group, which cannot be explained by the properties of the RDV in this study.
Most of the increase in ECMO patients was due to intubation. Intubation is a controversial procedure in COVID-19 treatment. It is believed to have been overused during the pandemic and sometimes caused preventable deaths. The percentage of patients intubated (“need for invasive ventilation”) is part of the measured results. Given the lack of positive effects from RDV and the incentives of the sites, the disproportionate intubation of placebo patients raises the suspicion that some of them were unnecessarily intubated to improve outcomes for remdesivir and Gilead. Some of the patients may have died from it.
- According to Reuters, many hospitals have concluded that RDV is not what it was promised.
- This study was published in NEJM on October 8th, the same day as another newspaper reporting old news about the tragicomic RECOVERY process and a screeching anti-Trump editorial “Dying in a Leadership Vacuum”. This is not an accident, it is a pattern. The PR version of this paper was published on May 22nd – the same day the infamous (Mehra et al., 2020) was published in The Lancet. Notably, Mehra was associated with Brigham and Women & # 39; s Hospital, which was contracted by Gilead for an RDV for a COVID-19 clinical trial, possibly another.
- It is strange that research into a potential treatment for pandemic diseases began in Phase 3, which is required for FDA approval of a novel drug. Phase 3 studies do not allow the flexibility required for research, especially in emergencies. It appears that the pandemic was used as an opportunity to shorten the approval of the drug, which under normal circumstances could not be tested.
- Contrary to the principles of scientific archiving, the DOI number and the URL (doi: 10.1056 / NEJMoa2007764, https://www.nejm.org/doi/full/10.1056/NEJMoa2007764) of the preliminary report are used again in the final report. despite massive changes in the text and in the appendices.
- The preliminary report claimed that the study outcome measure was changed on April 2, 2020 "without knowledge of the outcome data from the study and before intermediate data was available." This subsentence is removed in the final report, essentially admitting that it was changed with knowledge of the result data.
- The unusually large number of sites and their selection by Gilead could indicate an attempt to co-opt influential institutions. (Roussel & Raoult, 2020) found a near perfect correlation between the amounts received from Gilead and public opposition to hydroxychloroquine in France.
- If someone wanted to test the hypothesis that HCQ and HCQ + AZ are effective for COVID-19 treatment, re-analyzing the raw data from this study would work.
- This analysis does not include reporting of adverse events from the study
No competing interest
The author does not declare a competing interest.
No funding was provided for this work.
All relevant ethical guidelines were followed.
John H. Beigel, Kay M. Tomashek, MPH, Lori E. Dodd, Aneesh K. Mehta, Barry S. Zingman, Andre C. Kalil, MPH, Elizabeth Hohmann, Helen Y. Chu, MPH, Annie Luetkemeyer, Susan Kline, MPH, Diego Lopez de Castilla, Robert W. Finberg, Kerry Dierberg, MPH, Victor Tapson MD, Lanny Hsieh, Thomas F. Patterson, Roger Paredes, Daniel A. Sweeney, William R. Short, MPH, Giota Touloumi, David Chien Lye , MB, BS, Dr. med. Norio Ohmagari, Dr. med. Myoung-don Oh, Dr. med. Guillermo M. Ruiz-Palacios, Dr. med. Thomas Benfield, Dr. med. Gerd Fätkenheuer, Dr. med. Mark G. Kortepeter, Dr. Robert L. Atmar, C. Buddy Creech, MPH, Jens Lundgren, Abdel G. Babiker, Sarah Pett, James D. Neaton, Timothy H. Burgess, MD, MPH, Tyler Bonnett, MS, Michelle Green, MPH, MBA, Mat Makowski, Ph.D., Anu Osinusi, MD, MPH, Seema Nayak, MD, and H. Clifford Lane, MD for ACTT-1 Study Group Members, Remdesivir for the Treatment of Covid-19 – Final Report, NEJM, 2020 , https://www.nejm.org/doi/full/10.1056 / NEJMoa2007764
Beigel, …, Lane for ACTT-1 Study Group Members, Remdesivir for the Treatment of Covid-19 – Preliminary Report, NEJM, 2020, available at https://www.nejm.org/doi/suppl/10.1056/NEJMoa2007764 / supplement_file / nejmoa2007764_preliminary-report.pdf (without attachments)
Goldstein, Leo; Remdesivir has insignificant antiviral effects against SARS-COV-2, but dangerous adverse events, defyccc.com, 2020, https://defyccc.com/wp-content/uploads/RDV-ineffective-in-COVID-19-Final -Draft .pdf
Kraemer, Helena Chmura, Ph.D., Pitfalls of Multi-Site Randomized Clinical Trials on Efficacy and Effectiveness, Schizophrenia Bulletin, 2000, https://doi.org/10.1093/oxfordjournals.schbul.a033474
Kraemer, Helena Chmura; Robinson Thomas N., Are Certain Multicenter Randomized Clinical Trial Structures Misleading For Clinical And Policy Decisions ?, Contemporary Clinical Trials, 2005, https://doi.org/10.1016/j.cct.2005.05.002
Mandeep R. Mehra, Sapan S. Desai, Frank Ruschitzka, Amit N. Patel, RETRACTED: Hydroxychloroquine or chloroquine with or without macrolide for the treatment of COVID-19: a multinational registration analysis, The Lancet, 2020, https://doi.org / 10.1016 / S0140-6736 (20) 31180-6
Y. Roussel, Y .; Raoult, D .; Influence of Conflicts of Interest on Public Positions in the COVID-19 Era, Gilead Sciences Case, New Microbes and New Infections, 2020, https://doi.org/10.1016/j.nmni.2020.100710
Sermo, Zinc and Vitamins C and D, Recommended by Global Doctors for Treatment and Building Resistance to COVID-19, April 8, 2020, https://www.sermo.com/press-releases/sermo-reports-zinc- and-vitamins -c-and-d-recommended-by-global-doctors-for-treatment-and-build-up-resistance-to-Covid-19 /
Sermo, results from the third week: 17% more COVID practitioners worldwide who used hydroxychloroquine (33% -50%) and azithromycin (41% -58%), April 15, 2020, https://www.sermo.com / Press Releases / Sermon Reports-Week-3-Results-Worldwide-17-Point-Increase-in-Covid-Treaters-Who-Have-used-hydroxychloroquine-33-50-and-azithromycin-41-58 / /
Siddiqi, Hasan K .; Mehra, Mandeep R .; COVID-19 Disease in Native and Immunocompromised States: A Clinical Therapeutic Staging Proposal, The Journal of Heart and Lung Transplantation, March 20, 2020, https://doi.org/10.1016/j.healun.2020.03.012
Vaduganathan M, van Meijgaard J, Mehra MR, Joseph J, O & # 39; Donnell CJ, Warraich HJ. Prescription fill patterns for commonly used medications during the U.S. COVID-19 pandemic. JAMA. 2020. doi: 10.1001 / jama.2020.9184
WHO Solidarity Process Consortium, Hongchao Pan, Richard Peto, Quarraisha Abdool Karim, Marissa Alejandria, Ana Maria Henao Restrepo, Cesar Hernandez Garcia, Marie Paule Kieny, Reza Malekzadeh, Srinivas Murthy, Marie-Pierre Preziosi, Srinath Reddy, Vasey Roses, Mirta Roses Siyath John-Arne Rottingen, Soumya Swaminathan; Repurposed Antiviral Drugs For COVID-19; Interim results of the WHO SOLIDARITY study (form), medRxiv, https://doi.org/10.1101/2020.10.15.20209817
Zampino R., Mele F., Florio LL, et al. Liver damage in COVID-19 patients treated with remdesivir. Hepatol Int. Published online July 28, 2020. doi: 10.1007 / s12072-020-10077-3