The numbers reveal: The vaccines work

Vaccination efforts against the coronavirus began in Israel two months ago, and it is a good time to look into the effects of vaccination on morbidity rate in Israel and how – if indeed – vaccine effectiveness is affected by new variants of the virus. The only vaccines administered in Israel for now are mRNA vaccines, which will be the focus of this article.

Positive effects of the vaccine in Israel

As the vaccination effort began in Israel, health authorities – The Israel Ministry of Health and some of the state-mandated healthcare providers - started collecting morbidity data on the vaccinated population. Studies of vaccine effectiveness in the real world have great importance not only to Israel, but to the rest of the world. Israel has a few advantages when studying the effect of vaccines on morbidity rate. Firstly, every citizen in Israel has a digital medical file that contains their medical history, including medical examinations and vaccinations. With these medical files, it is possible to monitor the incidence of COVID-19 quickly and easily and compare that data against records of vaccination. Secondly, the vaccination operation progressed quickly, initially at least, so a large share of the high-risk population was vaccinated within a short time. Finally, sadly, the pandemic was spreading at a fast rate at the time that vaccination efforts began, and despite the decline in incidence in the past weeks, there is still a high chance of exposure and contraction. Nonetheless, when compared with a controlled experiment, where a vaccinated group is compared with a group that was treated with a placebo (a saline solution or a vaccine against a different disease), examination of the effects of the vaccine on the real world is complex, and researchers in Israel tried different approaches to do so. Although the findings obtained in these studies are not identical, they are similar enough to point at a clear conclusion: the vaccines reduce the general morbidity rate and – critically – the rate of severe morbidity, at a high level of effectiveness, resembling the one declared by Pfizer and Moderna in their clinical trials.

The most extensive study, that also produced the most encouraging results, was performed by the Clalit Research Institute, led by Prof. Ran Balicer in cooperation with experts from Harvard University. The researchers compared 600,000 vaccinated subjects (a week since the second dose) with an identical number of subjects that were not vaccinated, when each of the vaccinated subject was matched with a subject in the unvaccinated group with a similar demographic profile (age, medical background and risk factors, neighborhood of residence, ethnicity etc.). For example, a 76-year-old vaccinated Arab man, with a history of four flu vaccines in the last five years and two COVID-19-related risk factors, was matched with an unvaccinated Arab man, aged 76 to 77, who lives in the same neighborhood, has an identical set of risk factors and was the recipient of 3-4 flu vaccines in the last five years. Comparing research subjects in this way eliminates many possible confounding factors and helps researchers estimate how effective the vaccine is. The New England Journal of Medicine study found that the vaccine's efficiency rate with respect to preventing infection was 46% three weeks after the first dose and 92% the second week after the second dose. The researcher also discovered that the vaccine effectively prevents the appearance of symptoms (94%), hospitalizations (87%) and serious illness (92%). Furthermore, the researchers found that the vaccine prevents mortality from COVID-19 with 72% effectiveness three weeks after the first dose. Due to the short time frame, not enough data was collected to measure the efficiency of preventing mortality two weeks after the second dose.

Another healthcare provider, Maccabi Health Services, also issued a press release that stated a 95% efficacy rate for the vaccines. According to the Maccabi study, only 608 out of 602,000 vaccinated subjects that (0.1%) were infected with COVID-19 once a week has passed since their second dose, and only 7 of them experienced severe symptoms. That figure is smaller compared with 3.9% that were infected with corona amongst the roughly half a million subjects in the unvaccinated group. In another study by Maccabi, in cooperation with the KI institute, the researchers followed 200,000 vaccinated subjects at the age of 60 and over, for a few weeks starting at the first dose of the vaccine. In a short document, they present a 67% decline in the confirmed cases rate and a 78% decline in the hospitalization rate, when comparing subjects in the second week since the first dose with subjects in the second week since their second dose.

Prof. Dvir Aran, an immunologist and computational biology expert from the Technion, compared vaccinated patients with the general population and calculated that the effectiveness of the vaccines in Israel as of February 22nd is a 73-85% reduction of infections (the confirmed cases rate). Moreover, he found an 89-97% reduction in hospitalizations and severe morbidity. This study has not been peer reviewed yet either.

Additionally, journalist Nadav Eyal tweeted a joint article by Pfizer and the Israeli Ministry of Health. According to this document, five metrics were studied regarding the effectiveness of the vaccine one week after the second dose and all of them presented the high effectiveness of the vaccine: 89.4% in preventing infections, 93.7% in preventing symptomatic disease, 93.3% in preventing hospitalization, 93.9% in preventing severe cases and 92.9% in prevention of mortality. It should be noted that the article has not been officially published online and the publication by Eyal lacked graphs and tables, so these results should be taken with a grain of salt. 

One of the smaller healthcare providers, Kupat Holim Meuhedet, issued a press release on March 8th detailing findings from its study into COVID-19 vaccine effectiveness among 102,000 recipients. The study compared morbidity rates during the five weeks after receiving the second dose. According to the study, the vaccine is 79.4% effective the first week after the second dose, and it rises to 89.4% the following week. The researchers followed up on the recipients of the vaccine and found that effectiveness goes up to 96% two weeks after the second dose. In other words, among 459 patients who were diagnosed with COVID-19 after the second dose, 55% were diagnosed the first week, 27% were diagnosed the second week and only 5-9% were diagnosed in the each of the subsequent three weeks. This study led Meuhedet researchers to recommend that the Israel Ministry of Health delay the distribution of Green Passes from 7 days after the second dose to 14 days after it. They reasoned that this could confer needed protection to vaccinated people who were not yet not fully protected and reduce the chance that they pass on the disease.

Prof. Yariv Wine from Tel Aviv University investigated whether vaccinated persons express COVID-19 antibodies in their breast milk. Working with Dr. Michal Rosenberg-Friedman and physicians at the Lis Maternity and Women's Hospital, the levels of antibodies were monitored in the blood and breast milk of 10 breastfeeding women who had received the Pfizer vaccine. The study, uploaded to medRxiv but not yet peer-reviewed, found that the breast milk of vaccinated women contained COVID-19 antibodies. The levels of milk antibodies rise along with the levels of antibodies in their blood, and reach their maximal level two weeks after receiving the second dose. This study shows that vaccination of lactating women provides protection not only to them but also to the children they nurse and care for.

Studies in Israel demonstrated that vaccines reduce the general morbidity rate and the severe cases rate in particular. Rows of vials of messenger RNA vaccines against COVID-19 | Shutterstock, CROCOTHERY

Vaccine reduces morbidity rate, possibly infections as well

Prof. Eran Segal, a computational biology researcher at the department of computer science at the Weizmann institute, Dr. Uri Shalit, an artificial intelligence researcher at the Technion, and their group members, followed morbidity rates in different age groups since the beginning of the vaccination drive, while taking into account the effects of the lockdown. At the time of the publication, which is not peer reviewed as of yet, the researchers found a decline of 49% in morbidity, a decline of 36% in hospitalization and a decline of 29% in severe cases rate amongst cases at the age of 60 and over on Feb. 6th compared with three weeks prior. However, at younger age groups, where vaccination rates are lower at the initial stage of the vaccine drive, moderate declines were observed: a 18% decline in morbidity rate, and a 10.5% decline in hospitalization rate. The effect of the vaccines is greater in cities where vaccination rates are higher. 

Prof. Khitam Muhsen, an epidemiologist from the Tel-Aviv University, and her group members, studied the effect of the first vaccine administration on its own through a comparison of morbidity rate%ages in the vaccinated population and the general population. In the pre-print, Prof. Muhsen found that the vaccine reaches a 51% effectiveness rate in preventing infections 13 days following the first dose.

Prof. Doron Gazit and his group followed the spreading dynamics of morbidity and the actual severe cases rate compared with a model that presents a scenario where no vaccines are administered, as well as models that present different efficacy rates for the vaccines. The results of their study, which has not been peer reviewed yet, demonstrate that the vaccination effort slows down the spreading of the pandemic, and they estimate that the vaccines are at least 50% effective.  

Another important and encouraging observation was published in studies that have not been peer reviewed by two research groups in Israel. The group of Prof. Roy Kishony, a computational biology researcher at the Technion, in cooperation with Maccabi, found that the viral load - meaning the amount of viruses in the upper airways - is at least 4 times lower in those vaccinated and diagnosed in 12-28 days of the first dose of the vaccine, compared with those vaccinated and diagnosed within 12 days since the first dose. Dr. Yaniv Erlich and other researchers at the MyHeritage coronavirus test laboratory, together with Prof. Dvir Aran of the Technion and other researchers from the virology laboratory at Sheba Medical Center, compared the viral load in about 16,000 positive coronavirus samples during the months of December-January. The researchers found that starting mid-January - but not earlier - a 1.6 to 20 fold decline in the viral load was observed in samples of people at the age of 60 and over, but not in samples from younger subjects. These findings correlate to a situation where those at the age of 60 and over were the first to be vaccinated, so the effect of the vaccines is first exhibited in this group, but only following a sufficient time since the first dose. Both studies support additional studies, which demonstrate that vaccines are capable of not only reducing severe cases, but the overall morbidity as well, since the lower the viral load, the lower the chance of spreading the disease. 

An initial study published by researchers at the Mayo Clinic in the U.S. compared about 30,000 public health workers in several states that were vaccinated with roughly 30,000 public health workers that were not, and found that mRNA vaccines had an 88% effectiveness in preventing infections.

Due to a shortage in vaccine doses, authorities in the United Kingdom decided to postpone the second dose of the vaccine for most recipients. Preliminary studies, that have not yet been peer reviewed, suggest that as little as one dose can be effective. A study on the population of Scotland found that the Pfizer vaccine was 85% effective in preventing hospitalization and severe morbidity beginning five weeks after the first dose. The equivalent figure for the Astra-Zeneca vaccine was 94%. A study conducted among hospital staff in England found that the Pfizer vaccine was  72% effective in protecting against infection or illness three weeks after the first dose (the same finding as the Clalit study), and 86% effective a week after the second dose. Another study from England examined the effects of vaccination on people aged 70 and up, showing at least 50% effectiveness in preventing infections and 60-70% effectiveness in preventing hospitalizations and mortality approximately a month after the first dose.

All the studies described here are preliminary, and it is likely that the estimates will shift according to newly obtained data in the coming weeks and months. Nonetheless, all the studies conducted so far, which were performed by independent researchers in various countries that employed different approaches and methods, point to the same trend: vaccines are at least 50% effective in preventing illness and infecting others and over 90% effective after the second dose.

Different studies point towards the same trend: Vaccines are at least 50% effective in preventing infection and over 80% effective in preventing disease | Shutterstock, Toor Kan

No unusual or severe side effects

At this time, about 181 million vaccine doses have been administered around the world (about half of those were mRNA vaccines), and there are no reports of mortality cases nor extreme reactions to the vaccines, such as: Antibody-dependent enhancement (ADE) or harm to pregnant women.

The coronavirus vaccine follow-up committee of in Israel published a presentation summarizing the reported side effects following vaccination, as of Jan. 31st. There were no surprises in the report - the same side effects were observed amongst the vaccinated population as in Pfizer’s trial and at similar rates, disregarding a lower rate of reports regarding mild side effects. It is assumed that it is a case of under-reporting, meaning, that vaccinated people with mild side effects that do not require medical assistance do not report them. Forty-eight vaccinated people were hospitalized soon after receiving the vaccine, but in all cases studied, it was found that hospitalization was related to underlying conditions rather than to the vaccine. Six vaccinated subjects experienced a severe allergic reaction. No vaccinated person has died because of the vaccine.

 In England, data from the “Yellow Card” system, used to report side effects of the vaccine, describe a similar situation, as of Jan 31st. Up to this date, 6.6 million people have been vaccinated with the first dose of Pfizer’s vaccine, about 3 million people were vaccinated with AstraZeneca’s vaccine and about half a million people received both vaccination doses of either vaccine. A small number of severe allergic reactions was reported there as well: approximately 1-2 cases per 100,000 doses. The Journal of the American Medical Association (JAMA) reported 66 cases of severe allergic reaction out of 18 million mRNA vaccine units administered in the U.S. as of Jan. 18th. That is why people at risk of allergic reaction were instructed to avoid the vaccine, especially if there is known sensitivity to an ingredient in the vaccine; and of course, there is a longstanding instruction to wait 15 minutes at the vaccination facility following the injection (about 80% of allergic reaction cases in the U.S. occurred within 15 minutes of injection).

While Pfizer and Moderna’s corona vaccines are the first approved mRNA vaccines, the technology has existed for a few decades, and mRNA vaccines against different viruses or cancer are at different stages of clinical trials for over a decade. Data from these subjects, together with tens of thousands tested in Pfizer and Moderna’s trials, a few of which received the vaccine as early as April 2020, present no expected long term side effects.

About 181 million doses of the vaccine administered across the world, no mortality nor unusual reactions to the vaccines reported | Picture of a man injected with a vaccine | Dr P. Marazzi / Science Photo Library

Vaccinating people who have recovered from corona

Several studies from abroad and from Israel, published online but not peer reviewed as of yet, have demonstrated that administration of a single vaccine dose to people who have recovered from COVID-19 increased their antibody levels to 10 to 1,000 times higher than the antibody levels in recovered patients that were not vaccinated, or the levels people that have not contracted the virus and received a single dose. The studies also point at generation of type T and type B memory cells. According to these studies, the level of antibodies of recovered patients that were given a single dose was similar to that of those who have not contracted the virus and were treated with two doses, or higher.  According to the Israeli study, there is no difference in immune response between different ethnicities (Jews, Arabs and Druse).

The U.S. Center for Disease Control and Prevention (CDC) issued updated guidelines for mRNA vaccines. According to these guidelines, there is no prevention to vaccinate recovered patients for COVID-19, but people who are currently infected or should not be vaccinated, nor should those who were recently exposed to a confirmed patient. Regarding people who have received the full vaccine (two weeks following the second dose), the CDC states that there is no need for isolation following exposure to confirmed cases, but only if no symptoms appear, based on the assumption that those that were vaccinated and infected but do not experience symptoms, their chances of infecting others are reduced. In Israel recovered patients are not vaccinated as of yet, in order to use the limited amount of existing vaccinations on those who are completely unprotected. Nonetheless, according to these studies, it is possible to vaccinate recovered patients with one dose, thus saving on vaccine doses.

According to studies, antibody levels in recovered patients that were vaccinated with a single dose is similar to that of those who have not contracted the virus and were treated with both doses, or higher. Illustration of a virus next to an mRNA molecule | Shutterstock, Mike Mareen

mRNA vaccines are effective against different strains of the virus. Other vaccines are less effective

One of the more common vaccine-related debates in the past weeks regarded their effectiveness against the different variants of the virus. As in all living creatures, the novel coronavirus evolves. As the virus infects more people, it is more likely to develop new abilities that will enhance its ability to infect - as in the British variant - and maybe to avoid antibodies, especially neutralizing antibodies, made against the original strain. Neutralizing antibodies are antibodies that attach to the virus’s spike protein in a manner that prevents its attachment to receptors on human cells.

Indeed, Prof. Gideon Schreiber and his group members at the Weizmann Institute found the same mutations that promote concern amongst scientists regarding the British, South African and Brazilian variants using “in vitro evolution” experiments. They have also discovered another mutation in the same region of attachment to the receptor, which has not been found in the general population. This mutation could improve the virus’s ability to bind to the receptor and interfere with the ability of antibodies to neutralize the virus. A similar “in vitro evolution” study of the virus, conducted by researchers from Italy and U.S., also demonstrated that mutations like the one found in the South African variant could hamper the effectiveness of neutralizing antibodies. 

Concerns regarding reduction in vaccine effectiveness were found to be valid in the cases of the vaccines by Novavax, Johnson & Johnson and AstraZeneca (Oxford). Novavax developed a protein-based vaccine, while Johnson & Johnson's and AstraZeneca’s vaccines were based on an adenovirus vector. All three companies conducted their experiments in England and South Africa, while their vaccines were found to be effective against the British variant (about 90% effectiveness), in the study arm conducted in South Africa they were found to be relatively less effective (Johnson & Johnson: 64% effectiveness; Novavax: 60%, AstraZeneca: 22%, with a drop to 10% when focusing on people infected with the South African variant only). Thus, South African officials have decided not to administer the million vaccine doses purchased from AstraZeneca. By contrast, mRNA vaccines demonstrate good effectiveness against all variants.

Producers of mRNA vaccines published studies about the effectiveness of neutralizing antibodies found in the serum (the liquid component of blood) derived from vaccinated people one (Moderna) or two weeks to a month (Pfizer) following the second dose. The Moderna study, published in the New England Journal of Medicine, investigated effectiveness against the British, South African and several European variants, including the variant found in minks. BioNTech-Pfizer published two studies in the New England Journal of Medicine and in Nature Medicine, in which the company examined the effectiveness of the serum of vaccinated people against a diverse selection of variants, including the Brazilian variant. The three studies found that the vaccine by both companies produces neutralizing antibodies in a similar effectiveness against all variants, except for the South African variant, for which vaccine effectiveness was three times lower than average for the two vaccines. Despite the reduced effectiveness, however, the level of neutralizing antibodies was still high enough to confer protection.

Several studies, most of which have not been peer reviewed yet, examined the effectiveness of the antibodies produced in response to vaccines against the British variant, and demonstrated with tests conducted on cell cultures that the blood serum derived from vaccinated people is able to neutralize the British variant in an effectiveness similar to that demonstrated against the original strain, or slightly lower - but not in a manner that prevents the vaccine from working against it. A testament to that is the success of the vaccination drive in Israel - as the main variant still spreading through Israel is the British one.    

Some of these studies, along with other studies, examined whether the antibodies that were produced in response to vaccines were effective against the South African and Brazilian variants. Several studies, which have not yet been peer reviewed, as well as a study appearing in Cell, showed that serum derived from vaccinated people is able to neutralize both variants in a lesser manner (numbers indicate a 5 to 12 times lower effectiveness). However, the Cell paper showed that the Pfizer vaccine was 3.6 times more effective than the Astra-Zeneca vaccine, a finding suggesting that mRNA vaccines might be more effective against the South African variant than other vaccines. Another study, that is yet to be peer reviewed, investigated the effectiveness of a single vaccine dose to people who have contracted COVID-19. The results presented a 1,000 -fold increase in the level of neutralizing antibodies effective against the original strain of the coronavirus, as well as the South African strain - and even against the first SARS virus - in a cell culture test. 

Finally, researchers from Oxford University, England investigated the effectiveness of neutralizing antibodies and the activity of type T immune cells. In a paper that was not peer reviewed yet, the researchers found, that following two vaccine doses, the neutralizing antibody activity towards the South African variant was 3-5 times lower than the one found against the original strain, and only two of 25 samples were completely unable to neutralize it. By contrast, samples from people who have recovered from the disease, or from people who were administered with a single dose of the vaccine, were completely unable to neutralize this variant. Hence, it is important to receive both doses of the vaccine. The researchers found that samples from people who were vaccinated with both doses contained antibodies that also identified the spike protein of SARS, MERS and the 4 coronavirus strains that cause the common cold, so it is possible that the vaccine will also confer some protection against these viruses as well. The most important findings were regarding type T cells: the researchers found that the T- cell response in those who were vaccinated with two doses was similar in all four variants (the original, the British, the Brazilian and the South African), because the T cells recognize areas of the spike protein that are identical in all four variants. From this the researchers conclude that even if there is a reduction in the activity of neutralizing antibodies, activity of T cells will aid in vanquishing this disease. This conclusion is corroborated by another study that looked into the recognition ability of type T cells from recoverees from COVID-19 (unvaccinated). Here too, the study showed that T cells recognize regions throughout the spike protein, suggesting that the T-cell response is unaffected by point mutations. 

Summary

All the studies conducted in Israel and around the world so far show that mRNA vaccines are safe and effective in reducing infection and morbidity rates of the original strain as well as its different variants. If you have access to the vaccine and have not done so yet, go get vaccinated without delay!

Last update: March 9th, 2021