17 January 2021 Professor Robert Clancy
Historically, pandemics generate suspicion, speculation and emotion, before logic and empirical decisions determine optimal management. The current COVID-19 pandemic is no exception. Twelve months on, there is an emerging consensus supporting an integration of a four-pillar plan: public health strategies; vaccination; early pre-hospital treatment; and hospital treatment. This position replaces an early confusion, with supporting data appearing on a near daily basis.
Public health strategies are well understood and highly effective, forming the bedrock for disease control, while hospital management is a work in progress but with progressive improvement in outcomes. Typical data for high risk subjects (over 50 years of age with one or more co-morbidities) in the US is currently 18-20 per cent hospitalisation, with mortality around 1 per cent. Less attention has been given to ongoing symptoms, with about 80 per cent of hospitalised patients having profound fatigue and/or breathlessness 3-4 months after discharge. Many are unable to return to full time work six months after the infection is controlled.
The area of intense disagreement is community management combining prevention by vaccine and reduction of hospital admission, using pre-hospital treatment. There is a global expectation that vaccines will dramatically change the current face of COVID-19 while there is broad-based denial that any of the available (unpatented) drugs beneficially alter the natural history of infection. Expectation of a vaccine nirvana alongside therapeutic nihilism are both incorrect, although each is promoted with a vigour rooted in socio-political conviction (and supported by the Pharma industry).
The conclusion, based on logic and data, is that vaccines and early treatment strategies are both necessary for optimal disease control. As a result, a community plan has been formulated, aimed at keeping patients out of hospital. Experienced physicians have developed protocols based on evidence, with sequenced multi-drug regimens that support over 80 per cent reductions in admissions to hospital and death. Implementation of this approach would effectively end the US, UK, Canada, and EU hospitalisation crises.
The objective of this brief review is to argue in support of these conclusions, based on an untangling of the pathobiology of COVID-19 over the last 12 months; review of the available data on the three vaccines used in the Western world; and current data supporting significant benefit of pre-hospital drug treatment.
The Vaccine.
The idea that a vaccine would induce sterilising immunity, and therefore prevent community spread by creating herd immunity, has become the dominant political and medical story. Political, economic and social planning has been based on a sense of certainty that this will happen. This was never a likely outcome, as such success was asking more of the immune apparatus responsible for containment of a respiratory virus than had been observed. The first principal of vaccinology has always been that a vaccine is unlikely to give better protection than does the disease itself. First-cousin corona viruses causing recurrent airways infections over many years manifest clinically as “common colds”. Recurrent COVID-19 infections have been documented during the first year of the pandemic. Mucosal immunological memory for corona viruses is predictably poor.
The best model for understanding mucosal compartment virus infection and likely vaccine performance is influenza. The 80-year history of influenza vaccine development has resulted in an annual vaccine-induced protection of 20-60 per cent, with much lower response rates in those over 65. This older age group is the high-risk target for COVID-19. This can be understood in terms of immune senescence, which is significant within the mucosal immune system in those over 65. The saving evidence for benefit in older subjects is Kristin Nichol’s study 20 years ago in healthy older subjects that showed a reduction of 20 per cent in those vaccinated in the “all death rate” increase that was associated with influenza. Interestingly, given current concern about thrombotic events in late-stage COVID-19 disease, this included those dying from cardiac and cerebral thrombotic events.
Influenza and COVID-19 are both RNA viruses lacking “error edit” seen in more sophisticated DNA-based organisms, with spontaneous mutations causing “antigen drift” that can change both biological characteristics and response to vaccines. Over 800 mutations have been described to date for COVID-19 virus. Two strains with characteristic mutations are of current concern due to their greater infectivity (a UK and a South African strain). Given that short-term protection and rapid “antigen drift” appear to be likely outcomes, the focus must be on scanning for variant mutations.This process needs to link to vaccine production much as is standard practise with influenza vaccines.
The future for adenovirus vectors must be doubtful, given that antibody is generated against the vector. The mRNA vaccines have their own challenges, including their need for “super-cold chain” storage and many unanswered questions with respect to longterm safety. Early human studies with rabies, CMV and influenza mRNA vaccines were aborted at Phase One level due to less than encouraging results. Reversion to traditional inactivated and “split virus” antigen vaccines (as made by Chinese vaccine producers), or similar, may become more appropriate to a predictable provision of seasonal vaccines that cover the dominant circulating antigens. In that mode a vaccine that deserves close watching is the Novavax recombinant spike protein nanoparticle vaccine with its Matrix-MI adjuvant. It is a “21st century subunit antigen vaccine” based on proven technology, with excellent safety and immunogenicity results in a Phase One study. Current clinical trials are of great importance.
The objective of any COVID-19 vaccine is to limit virus replication within the mucosal compartment of the airways. This requires specific activation of the mucosal immune system, which differs from systemic immunity, geared to protect the internal spaces of the body. Blood antibody levels characterise the systemic immune response. These antibodies are very effective at neutralising a virus that passes through the blood stream in its normal course of infection, such as the measles or mumps virus. These vaccines readily induce sterilising immunity.
The influenza and COVID-19 viruses are different. They enter the bloodstream only in late disease, when the inflammatory response to the virus has become the main cause of tissue damage, and when immune containment has become irrelevant. Limitation of virus expansion by the local mucosal immune response is subject to different rules: protection is mediated mainly by traffic of Th17 cells from lymphoid aggregates within the wall of the gut. These cells are activated by virus from the infected airways swallowed into the gut. These specific Th17 cells then “home” to the airways’ mucosa where they recruit cells and toxic molecules, called cytokines, that destroy the virus. Disease is determined by the balance between the viral load and the protective immune response. If the response is defective, as seen in older subjects, the viral load increases, causing a massive catch-up inflammatory response responsible for non- specific damage to lungs and other tissues. At the same time this response induces a hypercoagulable state (caused by the same cells and cytokines that, in health, target the pathogen). This second phase of COVID-19 disease often requires admission to hospital.
A carpet of T reg cells within the mucosa exists to limit mucosal inflammation as a protective mechanism for the lungs. This avoids destructive inflammation that would otherwise occur in response to the sea of microbes that bathe the mucosal surfaces (although it is clearly overwhelmed in COVID-19 disease).
The relevance of this immune machinery to Covid19 vaccines can be summarised:
# the systemic and mucosal immune compartments communicate poorly, with minimal mixing. Some mixing occurs in regions such as the nasal cavity and the alveolar space as demonstrated by injected pneumococcal vaccines where IgG antibody from blood can inhibit pneumocooci in the nose and the gas exchange apparatus of the lungs, with protection aginst infection in those regions. This may explain the high level of “PCR-ve Covid19 infections” in the mRNA trials, as discussed below.
# mucosal immune responses to the virus are transient.
# Immune senescence at a mucosal level is marked.
These basic observations well explain the limitations of injected influenza vaccines: blood antibody levels poorly correlate with virus protection; vaccine-induced protection is partial and of short duration; older subjects are poorly protected.
Also, this is the model that makes scientific sense for injected COVID-19 vaccines, irrespective of the carriers used. The extraordinary claims of over 90 per cent protection by the pharma companies for both mRNA vaccines (Moderna, and Pfizer) and the adeno vector vaccine (Oxford/AstraZeneca) come as a surprise, given knowledge of mucosal immunology. The Chinese “classic” inactivated and split unit vaccines, in early data suggest somewhat less (mid 50 per cent) protection. However, all these quoted “protection levels” are meaningless at present, for reasons to be discussed, and cries to choose this vaccine or another based on the scanty and largely company released data are at best, premature. They disrupt confidence in carefully constructed national programmes critical to vaccine distribution. In my view it is most likely that there will be little difference in protection levels between vaccines when real life assessment is completed, though this will not be known for some time. Then decisions can be made with confidence in regard to safety, protection levels, cost, storage and capacity to distribute over vast areas, often with poor cold chain resilience. An important issue will be adopting a system that can adjust to antigen drift, perhaps imposing the need to produce each year a vaccine variant to cover contemporary strains within the community. Influenza and CSL have proven the value of local production in both reflecting local virus strains and controlling production of a vital vaccine.
Three vaccines have been released in the US and UK after limited observation and review due to the dramatic circumstances of the pandemic. Follow-up has been restricted to about two months, and limited data has been released by the companies. Combining post-release clinical data from the pharmaceutical companies and information from the FDA allows a limited assessment of the current status. In addition, important reviews have been made by Peter Doshi (on Pfizer and Moderna mRNA vaccines), and D. Harrison (on Oxford adenovirus vector vaccine). Their analyses included data made available (for the first time) from the FDA review. The outcome of these reviews suggest that little difference exists between the three vaccines. Current opinion can be summarised as follows:
# The previously quoted “90 per cent protection” was for symptomatic disease (without evidence of reduction in admission to hospital or death, which remains unclear in early post vaccine release data). Significant reaction at injection sites suggests biased reporting in mRNA vaccines due to its “unblinding effect”. The eight-week follow-up is short term and the duration of protection is not known. Decline in antibody titre and the known natural history of regular recurrences of coronavirus and influenza infections suggest protection against symptomatic disease will last months (if indeed antibody levels are a reliable index, which is unproven), maybe one year. All vaccines stimulate short term non-specific protection, and the contribution of such non-specific mechanisms remains to be clarified.
# The vaccines give little (at best 50 per cent) protection against infection (as opposed to “clinical symptoms”) over the eight weeks observation period, when asymptomatic infection was recorded (with carefully monitored subjects, following the Oxford vaccine). When FDA data (not provided by Pfizer) was analysed to include “suspected COVID-19 disease”, protection against infection dropped to 20 per cent (increasing only to 30 per cent when Week One was not included). Leakage of serum IgG antibody, as noted above, could well explain this important subset, making PCR tests negative, but not preventing infection in the tracheobronchial system, which is the focus of early disease. This would mean that lower protection for both vector and mRNA vaccines is real and the lower protection values discussed are real.
# Careful following up of the Oxford vaccine trial showed re-infection rates in those given the vaccine who had positive cultures at the start of the trial was the same as “first-up infections” in the placebo group.
These results are not surprising, but will have a major impact on planning strategies for COVID-19 management. They will force a re-think by health authorities and government, who may not now rely on any of the current vaccines to achieve the expected durable immunity. In principle, they are the results anticipated for an injected influenza vaccine, except that COVID-19 is a more infectious and more lethal disease. There is, therefore, an urgent need to determine the dynamics of duration and degree of protection, especially in those over 65.
The adenovirus vector cannot be relied on for second immunisations (as antibody to the simian virus carrier limits response). For the same reason, the vector vaccine has limited value in Sub-Saharan Africa, where exposure to simian viruses is common.
The safety status of mRNA viruses is far from being understood, with reports suggesting a range of hypersensitivity and autoimmune responses. Variation in antigen levels, their exposure duration (as there is no control of antigen dose) and range of immune responses (including tolerance or non-responsiveness) must be studied before any clear picture emerges, and this will take time. The decision in Australia to not panic and rush release of vaccines was a wise one.
All three vaccines are experimental, with limited experience in humans,and using little-tried delivery systems. The high viral mutation rate demands programmes be established now to monitor “antigen drift” and co-ordinate with vaccine manufacturers, much as is routine with influenza. “Antigen shift” (characterised by a greater change in viral structure) as seen with influenza due to its segregate nucleus will not occur with COVID-19, which has linear RNA. Experience with HIV (also a linear RNA virus) suggests hyper-infection can occur, and this needs to be monitored. Given the multiple issues with currently available genetic vaccines, consideration of more traditional inactivated and sub-unit antigen vaccines should be assessed, particularly as the experimental delivery systems being used have not been shown to offer any significant advantage. Progress of the Chinese vaccines needs to be closely followed; and of more relevance to Australia, Phase 2 and Phase 3 trials of the NovaVax recombinant spike protein nanoparticle vaccine will be of enormous importance.
To summarise current position with vaccines;
# Little protection against infection occurs, although protection against symptomatic disease is significant, but is likely to be far less than 90 per cent. It remains to be demonstrated whether this translates into protection against admission to hospital and mortality since this was not the case in two months of follow-up with the mRNA vaccines. The duration level of protection in high risk individuals over time need to be monitored. The likely outcome is that vaccines push the disease profile towards asymptomatic infection, rather than induce any discrete sterilising immune state. It is unhelpful and risky to attempt to choose between available vaccines until far more data becomes available.
# Re-infection in vaccinated subjects appears to occur at a similar rate as it does for community non-vaccinated controls.
# There is no realistic chance of herd immunity, given the high rate of asymptomatic infections in vaccinated individuals. This becomes more probable should the current intention of about 30 per cent of the population (US figures) to not be vaccinated irrespective of advice given, be accurate. In Australia, every encouragement should support over 90 per cent vaccination, with whatever of the available vaccines are available: dissention and argument over “false news” undermines this endeavour. In other words, though immunity with COVID-19 vaccination appears to be neither complete nor durable, any chance of approaching “herd immunity” depends on a near 100 per cent vaccination rate. Time will answer the critical questions, and vaccines still in trials may provide a better choice in the longer run. None of the “clever” delivery systems have yet proved to be an advantage over traditional (or 21st century variations) adjuvenated split vaccines (other than for those who own the patent).
# It can be predicted that endemic spread of the virus throughout the population will occur. The observations in the UK trial of high levels of asymptomatic infection, and “PCR negative” infections in the Pfizer study, focus attention on confirming the dynamics of asymptomatic infection post-vaccination, and the degree of transmission in the community, from that source. Data on these critical issues is limited.
# There is a potential danger that as vaccination levels increase, but remain short of comprehensive cover, the virus could spread with “hotspots” difficult to identify. Such spread could promote the emergence of resistant strains. The worst scenario would be an increase in mortality due to spread from unrecognised vaccinated subjects with asymptomatic infection to those without vaccination protection. On this, current data is scanty, indicating about 20 per cent of COVID-19 infections are asymptomatic, but that the infectivity of these is reduced, perhaps three- or fourfold. Similar data in the post-vaccine world will be of central importance.
Early Drug Treatment
It could be summarised that in a “post-truth world” those with COVID-19 disease are denied safe, effective treatment which, if given early, can reduce admission to hospital and death. The main purpose of the comments to follow is to show that the data has moved on, and that science-based decisions can and must be made if lives are to be saved.
Two drugs are effective: hydroxychloroquine (HCQ) and ivermectin (IVM), with most effective trials including nutraceutical, zinc and intracellular antibiotics.
These antivirals have been available as antimicrobial drugs for many years. Their antiviral activity is due to intracellular processes that inhibit virus assemblage – HCQ reduces acidity within cytoplasmic vesicles, and IVM blocks communication between the cytoplasm and the nucleus, while both have many sites of action that impair the inflammatory response.
The basic principal in treating viral infections is to treat early. This is well established with acyclovir treatment for shingles, herpes simplex infections, and neuraminidase inhibitors in influenza. The same principal applies to treatment of COVID-19. Treatment during the first “virus dominant” phase is directed at reducing the viral load within the mucosal compartment, while in the second phase treatment aims at inhibition of the damaging inflammatory response. Patients with significant second-phase disease are usually in hospital, and treated with organ support, anticoagulation and anti-inflammatory drugs.
Although obvious that different phases of the disease require different therapies, confusion over the different causes of each phase led to an incorrect assessment of drug efficacy early in the course of the pandemic. It also has influenced criticism from many with narrow or no clinical experience. Thus several failed randomised clinical trials (RCT) of HCQ in hospital patients have been used to dismiss the value of HCQ in early treatment, without recognising differences between the pathogenesis of the two phases. Every study of early treatment, has shown protection, confirmed in multiple meta analyses. Both drugs have a high-level safety records. Earlier concern about HCQ and a prolongation of the Q-T interval has been resolved as a very rare event.
The data base supporting the value of treatment of COVID-19 disease is so strong that it is hard to understand the current philosophy of “wait until you are sick enough, then go to hospital” (The question I put to naysayers is ‘Would you give, or not give, HCQ or IVM to your grandparent with early COVID-19 disease in an Australian aged care facility?’). Suggested reasons for unscientific denial include: ideological unmovable mindsets; a rapidly evolving pandemic where new data appears on a daily basis, making it hard to keep up with the data flow; failure to understand the value of non-RCT data sets which, from the scientific and ethical viewpoints, are appropriate to the circumstances of a pandemic; and a total focus on an anticipated COVID-free world following the release of vaccines. An important point was that in the early stages of the pandemic confusion abounded, many treatments were suggested, and data was poor. The early RCTs showing no benefit for HCQ were on hospitalised patients, which was the wrong group to study. Subsequent trials were mainly quality Observational Studies, which sat poorly with purists who neither understood the disease nor the “real life” circumstances of a pandemic and refused to move on with the data. Denial was reinforced by bureaucrats, media and the pharmaceutical industry (which has no interest in cheap drugs without patents). The constant argument was “where are the RCTs?” (even after some were actually done!).
Nobody questions the gold star status of a well-done RCT. Thus far, there have been no large, high-quality, conclusive RCTs on any form of therapy or vaccine for COVID-19. What has not been recognised is that there are many other valuable assessment tools. Most drugs used in clinical practise have never been subjected to a RCT.
The RCT mantra selectively used against HCQ and IVM is cynical, given the experience with Remdesivir. This antiviral agent has been tried in the treatment of COVID-19 and has been shown in a RCT to reduce time in hospital by four days, but with no reduction in mortality. On this scanty evidence it was rushed through the regulatory process. Although three additional RCTs failed to confirm this slight benefit, it continues to be used at around A$4,000 a course, with many significant side effects.
Of direct relevance to early COVID-19 management is the continued reference to a RCT using HCQ in hospitalised patients with COVID-19 disease, as noted above, which of course is irrelevant to early treatment. It is noted that RCTs, at the unrealistic level some demand for HCQ and IVM, in the context of the pressures surrounding a pandemic, when no Pharma company will cover the expenses for cheap out-of-patent drugs, is a difficult ask. RCTs are essentially the tool of Big Pharma, with an average duration of four to five years, and with average costings of $1 million to 2million. However, recently RCTs have been done for both drugs in “real life” circumstances, adding to the many high quality observational studies with careful “propensity matching” of controls (to reduce bias). No one study irrespective of whether it is a RCT or an observational study (note the Remdesivir experience) is perfect – with both HCQ and IVM it is the sheer number of good observational and RCT studies consistently showing protection when used early in disease that makes an irrefutable case.
Review of clinical studies in early (pre-hospital) disease as at end of November 2020 illustrate the data:
# All 27 trials of HCQ showed protection (OR 0.37 (0.29-0.47)). Ten of these were RCT (OR 0.71 (0.54-0.95)) (the Odds Ratio , or OR of, say, 0.37, means “63 per cent protection”, and 0.71 would be “29 per cent protection”). The figures in ( ) are the 95 per cent confidence levels: if below 1.0, this is equivalent to {at least} a P value below 0.05)
# 26 of 32 prophylaxis studies using HCQ showed protection (OR for five post-exposure studies: 0.61 (0.4-0.74))
# IVM in eight studies, half of which were RCT, showed protection in early treatment studies (OR 0.28(0.13-0.59) P=0.004)
As this clinical data continues to accumulate, regions around the world are adopting therapy with HCQ or IVM with dramatic results, when compared to adjoining areas that have not adopted this therapy. This has been marked in regions in Brazil.
The most recent observational study, with good propensity matching, co-ordinated by Peter McCullough of Baylor University Medical Center in an early sequenced multidrug trial, combined HCQ with IVM in 869 high risk subjects (age over 50, with at least one co-morbidity), using the Cleveland Clinic COVID-19 hospitalisation calculator for controls. The early ambulatory treatment regimen was associated with estimated 87.6 per cent and 74.9 per cent reductions in hospitalisation and death respectively, (P below 0.0001).
A comprehensive and detailed review “Ivermectin Reduces the Risk of Death from COVID-19” (January 3:2021) by T Lawrie, from The Evidence-based Medicine Consultancy, confirmed the view of the US “Front Line COVID-19 Critical Care Alliance” that the evidence on IVM “demonstrates a strong signal of therapeutic efficacy” recommending its global adoption for prophylaxis and treatment of COVID-19. Seventeen treatment and prophylaxis studies were critically analysed (nine of which were RCTs). “Moderate Certainty Evidence” in RCTs showed IVM reduced death by an an average of 83 per cent (65-92. 95 per cent confidence limits). This Forest Plot included 1107 subjects, with a risk of death at 1.4 per cent versus 8.4 per cent in controls. In nine separately analysed observational trials similar data was found: reduction of deaths was 69 per cent (0.16-0.61), with risk of death 3.9 per cent vs 9.9 per cent. In the same analysis, four quality studies with “moderate certainty evidence” showed IVM prophylaxis among health workers and COVID-19 contacts reduced the risk of infection by 88 per cent (0.08-0.18) in 851 participants, with 4.3 per cent vs 34.5 per cent exposed, contracting COVID-19.
Conclusions
The incidence and mortality of COVID-19 in the US and Europe is of crushing proportions, with no end in sight. Australia has been protected through its island position and good quality public health, but this does not guarantee that conditions will not worsen. The ferocity of infection has been noted in both NSW and Victoria. The isolation and economic impact of lockdowns must have a limited horizon. Relaxing national borders is not being discussed. Planning on the basis that all this will change following the introduction of vaccines needs reassessment, as early review of trial data, while showing short-term protection from significant symptomatic disease, must be tempered by evidence that infection is little reduced when asymptomatic and “PCR-ve COVID-19” cases are counted.
Current vaccines remain experimental, as issues of safety and asymptomatic infection (and the infectivity of asymptomatic carriers and the implications of these observations for non- vaccinated individuals) are assessed, as must be the duration and level of protection in those vaccinated. This data is particularly needed for those most at risk. Uncertainties regarding the capacity of current vaccines to attain herd immunity due to continued asymptomatic infection dictate that additional measures to reduce the impact of the pandemic must be put in place.
Two drugs used early reduce admission into hospital and death, including in those considered high-risk subjects, and they go a significant way to filling this need: HCQ and IVM. Both can be used as prophylactic or therapeutic medications.
From uncertain beginnings, an impressive data base has more recently accumulated that strongly supports the use of HCQ and/or IVM. Their use in concert with vaccines can no longer be denied; indeed this is the only science-based option.
Governments must look at how best to manage vaccine production, as it is probable that “antigen drift” will demand vaccine adjustment on a regular basis. This may favour mRNA vaccines, given that vector vaccines are essentially “one-off” due to antibody generation against the vector with the first dose. Although traditional antigen vaccines (and their sophisticated replacements) used for influenza vaccines (and in the NovaVax and Chinese COVID-19 vaccines) may in the longer run prove superior and most adapted for the continued matching of “antigen drift”. With many issues to be resolved as the place of vaccines becomes clearer, it is important to reinforce the value of both vaccine and early drug treatment. The uncertainties as to where vaccines will travel reinforces the necessity to focus on early treatment regimens. While more effective drugs will be developed, currently HCQ and IVM fill a void and will save lives.
This is not a time to argue “vaccine or antiviral”, but one for an integration of early sequenced multidrug therapy and vaccination for COVID-19 naïve subjects, into the COVID-19 management plan to enable optimal disease management and community confidence in rebuilding our societies. Equally, it is not a time for those who should know better to publicly argue for one or other vaccine, when evidence for any such choice is well down the line, and the Australian community need to get behind a “vaccine for us all” programme, if even the slightest chance of herd immunity is to be achieved.
Robert Clancy is Foundation Professor Pathology, Medical School University Newcastle, Clinical Immunologist and (Previous) Head of the Newcastle Mucosal Immunology Group, with special interest in airways infection and vaccine development
Resource : here.
Comments