A Plea for the Use of Ivermectin Against COVID-19

To: Dr Julian Elliott
Executive Director
National Covid Clinical Evidence Taskforce
Level 4, 553 St Kilda Rd
Melbourne, Vic. 3004

Re: Call for an Urgent Review of the NCCET Recommendation regarding the use of ivermectin in the management of COVID-19 within 14 days

I refer to the current recommendation by the National Covid Clinical Evidence Taskforce (NCCET) regarding the use of the drug ivermectin for the management of COVID-19.

The NCCET serves an important role in reviewing and recommending treatment for COVID-19 to peak health professional bodies across Australia. The current recommendation (Communique Ed. 48 – 5.8.21) regarding the use of the drug ivermectin is as follows:

The available research evidence does not yet provide reasonable certainty to recommend for or against the use of ivermectin and therefore the Taskforce recommends ivermectin not be used outside of randomised trials. The certainty of the current evidence base varies from low to very low depending on which outcome is being measured, as a result of serious risk of bias and serious imprecision in the 18 included studies.

In addition to uncertainty around benefits for patients with COVID-19, there are common side effects and harms associated with ivermectin, including diarrhoea, nausea and dizziness.

Given this uncertainty of benefit, and concerns of harms, we recommend that ivermectin only be provided in research trials, where there is the potential to generate further evidence on the effectiveness, or otherwise, of ivermectin …

This is a high priority recommendation and will be updated as soon as new evidence becomes available.

Ivermectin has been the subject of more than sixty clinical trials, including more than thirty randomised controlled trials, and used successfully in national COVID-19 mass treatment campaigns in India, Mexico and several other countries to reduce the number of cases and prevent serious complications of the disease leading to hospitalisation and death.

Despite this, and in the absence of NCCET members’ personal experience in treating COVID-19 patients with ivermectin, the NCCET has selected in an arbitrary and imprecise manner a small number of published clinical trials (eighteen) upon which to base its current negative recommendation for ivermectin use. NCCET has failed to apply sophisticated, defined and detailed meta-analysis techniques as employed in widely discussed published reviews on ivermectin (see references attached). When lives are at risk, the highest standards of evaluation are required.

The emphasis on minor and generally uneventful “harms associated with ivermectin, including diarrhoea, nausea and dizziness” contained in the above NCCET statement demonstrates a total lack of therapeutic perspective in relation to the much more serious side effects of other drugs used to treat COVID-19, including many over-the-counter non-prescription drugs, and the dire consequences of a lack of effective therapeutic management of COVID-19 individuals.

The NCCET has sought to respond to critics of its recommendation on ivermectin in the Communique of August 5 by justifying its limited consideration of the ivermectin literature by posing, and then, answering its own question in the following way:

But hasn’t ivermectin been shown to be effective as an early COVID-19 treatment in randomised controlled trials overseas?

Despite some early suggestions that ivermectin may provide both prophylactic and therapeutic benefit, the available research evidence does not yet provide reasonable certainty to recommend for or against the use of ivermectin. More robust, well-designed randomised controlled trials are needed to demonstrate whether or not ivermectin is effective.

Some widely discussed meta-analyses of ivermectin studies (e.g. The British Ivermectin Research Development (BIRD) Group meta analysis) have significant weaknesses, for example they include a large trial which has been discredited and retracted (Elgazzar et al.). Even in these reviews, when patient populations are separated by severity and comparisons to active treatments removed, no meaningful effect is found.

Given the national importance of the NCCET advice on ivermectin, I invited internationally recognised and experienced literature review specialists Tess Lawrie MBBCh PhD and Edmund Fordham PhD FInstP of Evidence Based Medicine Consultancy Ltd (UK) and EbMCsquared, a community interest company located in Bath, England, to comment on the above NCCET interpretations of the literature. Their expert analysis is attached, titled, “Commentary on NCCET Statement on Ivermectin”, dated August 7.

The analysis reveals and details (with references) serious flaws in the selective NCCET interpretation of the “cherry picked” literature. It ignores the broad sweep of clinical evidence from other randomised controlled clinical trials, observational trials and national treatment programs and demands (in the NCCET’s own words) as a matter of high priority to review this recommendation in the national interest.

In addition, related to the current NCCET recommendation is the statement by the Therapeutic Goods Administration (August 18):

There is currently insufficient evidence to support the safe and effective use of ivermectin, doxycycline and zinc (either separately, or in combination) for the prevention or treatment of COVID-19. More robust, well-designed clinical trials are needed before they could be considered an appropriate treatment option.

In reality, there is insufficient evidence not to support the use of ivermectin while new and expensive drugs are being expedited through the regulatory process and given provisional approval with far less clinical trial, efficacy and safety data supporting their use.

Australia is in the grip of a pandemic of enormous consequences. Every possible useful therapeutic approach is needed in this crisis. Ivermectin, especially in combination with zinc and doxycycline, has been shown to be effective in relation to COVID-19 management. Other new antiviral medications have been recently approved by the TGA with relatively minimal safety and efficacy data by comparison to ivermectin.

Ivermectin has been in use for more than three decades. Four billion doses have been administered, it is on the World Health Organisation List of Essential Drugs, and it is one of the world’s most useful and well tolerated drugs available. Its breakthrough discovery is attributed to Satoshi Omura and William Campbell, who were awarded the Nobel Prize in Medicine in 2015, reflecting the magnitude of their achievement and the importance of ivermectin to medicine.

The current approach to symptomatic COVID-19 individuals is largely to do nothing and simply observe until they get better or worse, perhaps much worse, and need to go to hospital. The do-nothing approach places enormous strain on our health care system. Evidence for this “do nothing, watch and observe” approach is lacking. Ivermectin offers a potentially effective, low-cost, safe and rational approach to the management of such individuals with little or no disadvantage. The NCCET recommendation on ivermectin is considered by many experts to be misinformation and is viewed as contributing to needless hospitalisation. But for this recommendation, many COVID-19 infected individuals could be receiving early effective treatment.

Greg Hunt, Minister for Health and Aged Care, has written regarding ivermectin in a reply to Senator Malcolm Roberts (July 27): “It remains open for doctors to prescribe existing medicines ‘off-label’ based on their own clinical judgement.” Indeed, this has always been the case previously.

Given the evidence available, doctors should be able to prescribe ivermectin as monotherapy or in combination without stigma or hindrance by a restrictive recommendation from the NCCET or the TGA. The NCCET and the TGA should re-examine the accumulating international experience with ivermectin from all sources supporting its safe and effective use and should support and encourage ongoing efforts by many to clarify the important role of ivermectin in the management of COVID-19.

I request the NCCET review and issue revised recommendations for the use of ivermectin within fourteen days in the light of the submitted information as a matter of urgent priority and national interest.

Regards,

Phillip M. Altman

BPharm(Hons), MSc, PhD

Clinical Trials and Regulatory Affairs Consultant

References

Bryant, A, Lawrie, TA, Dowswell, T, Fordham, EJ, Mitchell, S, hill, SR and Tham, TC. Ivermectin for Prevention and Treatment of COVID-19 infection: A Systematic Review, Meta-analysis, and Trial Sequential Analysis to Inform Clinical Guidelines. American Journal of Therapeutics 28, e434-e460 (2021).

Kory, P, Meduri, U, Varon, J, Iglesias, J and Marik, PE. Review of the Emerging Evidence Demonstrating the Efficacy of Invermectin in the Prophylaxis and Treatment of COVID-19. American Journal of Therapeutics 28, e299-e318 (2021).

**********

Dr Tess Lawrie MBBCh PhD

Dr Edmund Fordham PhD FInstP EurIng

Commentary on NCCET Statement on Ivermectin in COVID-19

We have considered the extracts quoted below from the current NCCET statement regarding the use of ivermectin in COVID-19. Our responses and commentary to these statements follow.

The current recommendation regarding ivermectin is as follows:

Despite some early suggestions that ivermectin may provide both prophylactic and therapeutic benefit, the available research evidence does not yet provide reasonable certainty to recommend for or against the use of ivermectin.

And a specific critique asserts:

Some widely discussed meta-analyses of ivermectin studies (e.g. The British Ivermectin Research Development (BiRD) Group meta analysis) have significant weaknesses, for example they include a large trial which has been discredited and retracted (Elgazzar et  al.). Even in these reviews, when patient populations are separated by severity and comparisons to active treatments removed, no meaningful effect is found.

Overall assertion

The available research evidence from (i) randomised controlled trials, (ii) observational trials, (iii) clinical success of multiple unrelated clinicians in many parts of the world, (iv) the phenomenology of whole-country effects with both temporal correlation to introduction of ivermectin, and the contrasting experimental control of states or other administrative divisions with differing public health policies, all point overwhelmingly to the efficacy of ivermectin in both the prevention and management of COVID-19 [1].

The phrase “reasonable certainty” is undefined and vague, and no declaration as to what level of certainty would be regarded as “reasonable” is given. It is not a “level of certainty” recognised in formal meta-analysis.

The formal review of Bryant et al [2] found “moderate certainty” evidence which is normally considered more than sufficient for regulatory approval of existing drugs in a new indication. For example, corticosteroids have become a standard of care for inflammatory stage COVID-19 on the basis of a single randomised controlled trial (RCT) of dexamethasone [3], on what is generally considered as “moderate certainty” evidence. The review of Bryant et al [2] found “moderate certainty” evidence over twenty-four RCTs, not just one.

The prophylaxis trials were assessed as “low certainty” but report quantitative results in prophylaxis fully consistent with much larger observational trials, some very large [4].

“Low” certainty evidence in the past has been sufficient for the inclusion of ivermectin on the WHO Essential Medicines (Children) List in the indication of scabies [5] where measures of effect were in fact inferior to the previously recommended drugs.

On the basis of prior decisions in COVID-19, and for ivermectin in an anti-parasitic indication, the continued hesitancy of regulatory authorities worldwide with respect to ivermectin in COVID-19 is completely anomalous.

“Reasonable” is not recognised in formal meta-analysis according to PRISMA guidelines [6], which recognise very low, low, moderate and high certainty, typically from appraisals of Risk of Bias in contributing studies. There is always a measure of subjectivity in such appraisals but allocation of grades and conclusions of “levels of certainty” follow strict rules.

“High” certainty evidence is rare, confined to strong effects in very large clinical trials or meta-analyses pooling several such large studies. “Moderate” certainty evidence is generally considered extremely powerful, and more than sufficient for regulatory approval of existing medicines in new indications. “Low” certainty evidence has led to prior regulatory approvals to meet clear clinical needs. We address subsequent critiques of [2] below.

Much of the evidence was summarised as early as November 2020 by Kory et al and now published in their narrative review in the American Journal of Therapeutics [1] (May- June issue).

The formal systematic review and meta-analysis by Bryant et al [2] (July-August issue of same journal) was an exercise in support of the narrative review of Kory et al [1], but restricted by deliberate choice to RCTs only, as conventionally considered the highest quality of medical evidence.

For example, the review protocol excluded by policy notable studies such as the ICON study [7] demonstrating strong advantage in overall mortality in a large propensity-matched retrospective study, with obvious confounders addressed, simply because the patient allocation was not randomised. The most pronounced benefits were seen in severe disease.

Similarly in prophylaxis the very large trial of Behera et al [4] with well over 3000 participants was excluded for the same reasons, though delivering quantitative measures of Risk Reduction (for infection) very close to the meta-analysis of the RCTs.

Including high-quality observational trials was found to lead to results just as reliable as RCTs in the synthesis of Anglemyer [15]. Adding the many known observational trials to the meta-analysis of Bryant et al [2] is likely only to strengthen the findings further.

In any serious scientific appraisal, the evidence presented by these non-randomised trials cannot be dismissed as of no account, just because they lacked certain formal constraints, being part of the experience of hard-working clinicians in stressed circumstances.

(To pre-empt widespread misunderstanding, what is called “the BiRD group” or more accurately the British Ivermectin Recommendation Development panel (not “Research”) was an ad hoc panel of clinicians, researchers and other stakeholders, with international representation, convened for an “Evidence to Decision” framework event on February 20, 2021, to hear the evidence summarised in an earlier version of reference [2]. The BiRD panel published its recommendation quite separately from Bryant et al [2]. The authors of Bryant et al [2] comprise: two members of the steering group (who did not vote), four ordinary members of the BiRD panel (consumer representative, health economist and two active clinicians), and one professional systematic reviewer who did not take part in the BiRD panel but contributed extensively to the research. Hence the authors of Bryant et al [2] are not congruent with the membership of the BiRD panel, a much larger group, and include one major contributor who remains uninvolved with BiRD.)

Subsequent critiques of [2]

Some widely discussed meta-analyses of ivermectin studies (such as The British Ivermectin Research Development (BiRD) Group meta-analysis) have significant weaknesses, for example they include a large trial which has been discredited and retracted (Elgazzar et al). Even in these reviews, when patient populations are separated by severity and comparisons to active treatments removed, no meaningful effect is found.

These claims are categorically false, though regularly asserted by those with an agenda driven independently of the actual evidence.

1. The claim of “significant weakness” in [2] is confined entirely to the inclusion of the disputed trial of Elgazzar [8]. The review of [2] was exhaustive of all RCTs found at the review closure and the first anywhere to follow strict PRISMA guidelines [6]. At the time of publication of [2], there was no reason to doubt the veracity of Elgazzar [8]; indeed it would have been a protocol violation to exclude it.

It is untrue to state that the study has been “retracted”. Professor Elgazzar has retracted nothing, asserts defamation and has intimated legal action. The server ResearchGate has withdrawn the preprint in response to a complaint, without giving Elgazzar the right of reply. Whether or not the study is “discredited” remains to be determined.

Notwithstanding these uncertainties, a “Letter to the Editor” of the American Journal of Therapeutics [9] concerning the Elgazzar dispute has been accepted for publication and should appear shortly. We show explicitly the consequences of deleting the disputed trial in the leading mortality outcome, and in prophylaxis (Elgazzar [8] contributed arms to both outcomes). Whilst the quantitative result inevitably changes, the mortality outcome remains clear, demonstrating a 49 per cent reduction in favour of ivermectin (aRR=0.51, 95% CI 0.27 – 0.95).

Similarly the prophylaxis outcome remains in quantitative effect virtually unchanged, and in fact slightly improved in that the point estimate for reduction in COVID-19 infection increases from 86 per cent to 87 per cent (aRR=0.13, 95% CI 0.08 – 0.21), with similarly tight 95 per cent confidence intervals again fully consistent with the larger observational trials of ivermectin prophylaxis.

2. The assertion, “when patient populations are separated by severity and comparisons to active treatments removed, no meaningful effect is found”, lacks any logic.
3. Removing comparison to active treatments would be a pointless exercise. The pragmatic and pre-specified inclusion of “active” treatment comparators is a strength, not a weakness, of Bryant et al [2] and would lead to under-estimation of the effect of ivermectin, not over-estimation. In other words, Bryant et al [2] is conservative by design, against the effect of ivermectin. The fact that consistent positive effects are observed makes the results more convincing, not less.
4. Separation by severity has been dealt with explicitly by Neil and Fenton [10] who apply a Bayesian meta-analysis to the full set of trials in Bryant et al [2], with an explicit separation of disease severity between “severe” and “mild-moderate”. The study of Niaee [11] was excluded because disease severity was not distinguished. A “leave one out” sensitivity analysis is performed systematically on the entire data set, including the disputed trial of Elgazzar [8]. Again the conclusions remain robust to the removal of particular studies. For some studies with known heterogeneity the results are actually improved.
5. Neil and Fenton [10] find for severe disease a 90.7 per cent posterior probability that the risk ratio favours ivermectin, and for mild-moderate COVID-19 there is an 84.1 per cent probability the risk ratio favours ivermectin. They conclude that the results support the conclusions of Bryant et al [2] over other claims such as that of Roman et al [12]. The removal of Elgazzar [8] (Niaee [11] already excluded) provides the worst reduction in evidence but still results in a Bayesian posterior probability of effective risk reduction of 77 per cent.
6. Other meta-analyses have been accepted for publication [12], in spite of demonstrated reporting errors available at pre-print stage, with very similar titles to [2] but asserting the opposite conclusions. Roman et al [12] make a limited selection (1173 patients over ten trials compared to 3406 patients over twenty-four trials in [2]) of the trials reviewed in [2]. The assertions in [12] commit the elementary fallacy of supposing that lack of statistically significant evidence (in their highly selective survey) is the same thing as a positive demonstration of no benefit. These claims of Roman et al [12] were dismissed by Neil and Fenton [13], an earlier version of [10].
7. Similar assertions have been made by propagandists in news media [14] but are simply untrue, as demonstrated explicitly in [9].
8. The context where essentially all studies are referenced to placebo (or non-pharmaceutical precautions) is prophylaxis. As previously mentioned, the prophylaxis effect reported in [2] is actually slightly improved by the removal of Elgazzar [8], and consistent with large non-randomised trials of ivermectin prophylaxis. There is no question of categorising by severity in the prophylaxis context and virtually all studies are referenced against no active comparators. The reduction in infection risk by 87 per cent cannot be said to constitute “no meaningful effect”. It is a very strong effect, achieved with ivermectin alone (or in one trial, combined with topical iota-carageenan nasal sprays).

Moreover, there has been no credible challenge to the prophylaxis results. It is not credible that ivermectin should achieve a prophylactic effect (by whatever mechanism) and fail to achieve a therapeutic effect, at least in the initial (viremic) phase of the illness.

References

1. Kory, P., Meduri, G.U., Varon, J., Iglesias, J., and Marik, P.E. (2021). Review of the emerging evidence demonstrating the efficacy of Ivermectin in the prophylaxis and treatment of Covid-19. American Journal of Therapeutics, 28(3), e299-e318 DOI: 10.1097/MJT.0000000000001377
2. Bryant, A., Lawrie, T. A., Dowswell, T., Fordham, E. J., Mitchell, S., Hill, S. R. and Tham, T. C. (2021). Ivermectin for Prevention and Treatment of COVID-19 Infection: A Systematic Review, Meta-analysis, and Trial Sequential Analysis to Inform Clinical Guidelines. American Journal of Therapeutics, 28, e434–e460. doi: 10.1097/mjt.0000000000001402
3. Horby, P., Lim, W. S., Emberson, J., Mafham, M., Bell, J., Linsell, L., et al. (2020). Effect of Dexamethasone in Hospitalized Patients with COVID-19: Preliminary Report. New England Journal of Medicine, doi: 10.1056/NEJMoa2021436
4. Behera, P., Patro, B. K., Padhy, B. M., Mohapatra, P. R., Bal, S. K., Chandanshive, P. D., et al. (2021). Prophylactic role of ivermectin in SARS-CoV-2 infection among healthcare workers. Research Square preprint. doi: 10.21203/rs.3.rs-208785/v1
5. Cantey, P. (2018), WHO Expert Committee on the Selection and Use of Essential Medicines: Application for inclusion of ivermectin on the WHO Model List of Essential dicines (EML) and Model List of Essential Medicines for Children (EMLc) in the indication of Scabies. WHO Expert Committee Application. https://www.who.int/selection_medicines/committees/expert/22/applications/s6.6_ivermectin.pdf
6. Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D. et al (2021). The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ, 372. doi: 10.1136/bmj.n71 Accessed 22 July 2021.
7. Cepelowicz-Rajter, J., Sherman, M. S., Fatteh, N., Vogel, F., Sacks, J. and Rajter, J.-J. (2020). Use of Ivermectin is Associated with Lower Mortality in Hospitalized Patients with COVID-19 (ICON study). Chest, 159(1), 85-92. DOI: 10.1016/j.chest.2020.10.009
8. Elgazzar, A., Hany, B., Youssef, S. A., Hafez, M., Moussa, H. and Eltaweel, A. (2020). Efficacy and Safety of Ivermectin for Treatment and Prophylaxis of COVID-19 Pandemic. Research Square preprint doi: 10.21203/rs.3.rs-100956/v2 Accessed 22 July 2021.
9. Bryant, A., Lawrie, T. A. and Fordham, E. J. (2021). Letter to the Editor, American Journal of Therapeutics, accepted, to appear (August 2021).
10. Neil, M. and Fenton, N. E. (2021). Bayesian Meta Analysis of Ivermectin Effectiveness in Treating Covid-19 (with sensitivity analysis to account for possibly flawed studies). Research Gate preprint. doi: 10.13140/RG.2.2.19713.58723 Accessed 10 August 2021.
11. Niaee, M. S., Namdar, P., Allami, A., Zolghadr, L., Javadi, A., Karampour, A., … Gheibi, N. (2021). Ivermectin as an adjunct treatment for hospitalized adult COVID-19 patients: A randomized multi-center clinical trial. Asian Pacific Journal of Tropical Medicine, 14(6), 266. https://doi.org/10.4103/1995-7645.318304
12. Roman, Y. M., Burela, P. A., Pasupuleti, V., Piscoya, A., Vidal, J. E. and Hernandez, A. V. (2021). Ivermectin for the treatment of COVID-19: A systematic review and meta-analysis of randomized controlled trials. Clinical Infectious Diseases, doi: 10.1093/cid/ciab591
13. Neil, M. and Fenton, N. E. (2021). Bayesian Meta Analysis of Ivermectin Effectiveness in Treating Covid-19 Disease. Research Gate preprint doi: 10.13140/RG.2.2.31800.88323 12 July Accessed 22 July 2021
14. Davey, M. (2021). Huge study supporting ivermectin as Covid treatment withdrawn over ethical concerns. Guardian, 15 July. https://www.theguardian.com/science/2021/jul/16/huge-study-supporting-ivermectin- as-covid-treatment-withdrawn-over-ethical-concerns Accessed 22 July 2021.
15. Anglemyer, A., Horvath, H. and Bero, L. (2014). Healthcare outcomes assessed with observational study designs compared with those assessed in randomized trials. Cochrane Database of Systematic Reviews, DOI: 10.1002/14651858.MR000034.pub2