Hello, all. My name is Gavin Harris, and I will briefly be talking to you all today on COVID-19 vaccination and the immune system. I am an infectious diseases and critical care physician at Emory University who specializes in disaster preparedness. I have no conflicts of interest in this talk, but I do receive funding from NETEC, the CDC, FEMA, and the NIH. I want to start by first giving a brief perspective on the state of vaccinations throughout the world as of Sunday, the 27th of February, 2022. This data is courtesy of the Johns Hopkins University Coronavirus Resource Center, publicly sourced, and it lays bare how far we certainly have come with COVID-19 vaccinations, with the world hitting a global milestone of 10 billion administered doses in little more than a year in January, the largest vaccination program in history. And while this means that 64.3% of the global population have received at least one dose, less than 20% of the people across the entire African continent have received even one dose of a COVID-19 vaccine. In fact, vaccine inequity in the world has never been higher, as this graph of COVID-19 vaccine doses per 100 people shows by income group. If we are to combat this pandemic, this is a massive problem that continues to pose immense challenges. Until viral circulation and disease is reduced in every corner of the world, the risk of variant circulation and vaccine-resistant strains will continue. That being said, there are currently 24 studies in 11 countries that have been reported in pre-print and published literature and reports, including ongoing and planned studies. And 284 vaccine effectiveness studies in 31 countries published, pre-print, planned, or ongoing, mostly taking place in the Americas or Europe. I mean, there is a true wealth of information emerging almost daily. Given this scale, it would not be possible to discuss all trial data in the next 20 minutes, so I will instead pick out the most recent, relevant, and up-to-date studies to provide an update of what we know and how it relates to the patient populations we treat in intensive care units. This talk will focus on briefly detailing immune responses to vaccination, and given the limited time of this talk, I will focus mainly on discussing vaccination immunity duration and how it relates to special populations, for these are the patients that we are now mainly seeing in our intensive care units, immunocompromised populations and those reinfected, pregnant women, etc. We'll also talk about the efficacy of booster vaccinations and against variants of concern and how that affects transmission and breakthrough infections. We will also talk about specific major safety issues that have been found with vaccination and finish with some forthcoming data from vaccine studies, what we can expect from trial data over the next several months. And before we launch into the main discussion points, I want to address age-old terminology issues that continue to pervade medical research, policy, and conversations. Vaccine immunogenicity, efficacy, and effectiveness are not interchangeable. Vaccine immunogenicity refers to the ability of a vaccine to induce an immune response, whether antibody or cell-mediated. Efficacy is the extent to which a vaccine provides a benefit under ideal conditions, measured in Phase III randomized controlled trials, for instance. And then there is effectiveness. Effectiveness is the extent to which a vaccine provides benefit under real-world conditions. I will be discussing studies measuring all three of these endpoints, and I just want you to be aware that they do not mean the same things. First let's talk about the immune response to vaccination. As a precursor, one of the first studies to evaluate vaccine effectiveness was published in the New England Journal of Medicine in July 2021 from the United Kingdom, where data on vaccination is easily available in the National Vaccine Registry. Evaluating symptomatic cases in a test-negative case-control study, they compared vaccination status in patients reporting COVID-19 symptoms who either tested positive or negative. Then, in a secondary analysis, the relative incidence of infections in both variants was estimated and found a two-dose vaccination strategy of either an mRNA vaccine, in this case the Pfizer or BioNTech, or an adenoviral vector vaccine, the AstraZeneca vaccine. They found they both improved vaccine effectiveness markedly. After two doses, effectiveness rose again against each variant. BioNTech was 94% versus the Alpha variant and 88% versus the Delta variant, whereas with the AstraZeneca vaccine, 75% versus the Alpha variant and 67% versus the Delta variant. And in a final analysis just published on the 9th of February for comparison for the Johnson & Johnson adenoviral vaccine, researchers found that a single dose provided 52.9% protection versus moderate to severe COVID-19. Of note, this was a study looking at vaccine efficacy versus the first occurrence of PCR-confirmed moderate to severe COVID-19 with an onset of 14 days post-administration and at least 28 days after administration in a per-protocol population. This study was otherwise known as the Phase III Ensemble Trial. And on the left side of the screen are the pictorial representations of cumulative incidence of moderate to severe COVID-19 on top and on the bottom vaccine efficacy. Vaccine efficacy persisted through approximately six to seven months post-vaccine administration and six to seven months with a modest decline. This could be due to more neutralizing variant emergence. On the right side, we see severe to critical COVID-19 vaccine efficacy, the Kaplan-Meier curve again on top. Total efficacy in this case was 74.6%, with no evidence of waning for six to seven months post-administration, as you can see on the bottom right corner. So to briefly summarize, this trial began on the 21st of September, 2020, and data cutoff for the final analysis was the 9th of July, 2021. The level of protection was proportionate to the severity of disease and the nature of the viral mutation. Real-world data has shown varying degrees of efficacy of adenovirus-26, and this could really be important in closing the equity gap. We will certainly talk about adverse events later in the talk, and for most purposes, I will discuss the mRNA vaccines for the remainder, of which the most studies have been currently done in immunocompromised populations. So for that, I'm going to begin on the 21st of January this year with an article published in the Journal of the American Medical Association, or JAMA, that looked to assess mRNA COVID-19 vaccine performance, either Pfizer or Moderna, versus symptomatic infection from either Omicron or Delta variants. It was a test-negative case-control analysis of an adult population, and you can see the dates on the screen. Of almost 5,000 COVID-19 testing sites throughout 49 U.S. states, looking at three doses of an mRNA vaccine versus unvaccinated individuals, and measuring outcomes of the association between symptomatic infection and vaccination, with a secondary outcome looking at median cycle thresholds. Here is a table of their results, looking at the odds ratio of a patient coming to test with a confirmed symptomatic infection by either variant, as determined by genomic sequencing. As you can see, odds ratios were lower for those who received three doses of mRNA vaccines and had confirmed variant infection with Delta, and certainly lower with three versus two doses of a vaccine than that of unvaccinated persons. But an adjusted odds ratio of less than two indicates that relatively fewer test-positive cases had prior receipt of three doses, and with values closer to zero representing a stronger magnitude of association. These odds ratios correspond to an effectiveness of those with three doses versus unvaccinated of 67.3% for Omicron, 93.5% for Delta, and for three doses versus two doses, there was a corresponding effectiveness of 66.3% for Omicron and 84.5% for Delta. Again, with higher odds ratios for Omicron, it suggests that booster dosing is less protective against that variant, Omicron, than Delta, correlating with immune evasion findings of in vitro data. And importantly, from a public health perspective, this implies that higher booster rates may be needed to achieve the benefits seen during the Delta predominant period. Now, just published last week in Nature and highlighted by the University of Minnesota's Center for Infectious Diseases Research and Policy, an impressive study was published by Kaiser Permanente in Southern California, looking at three doses of the Moderna mRNA vaccine versus the Omicron and Delta variants in hospitalizations. This, again, was a test negative case control study with almost 27,000 COVID-19 cases caused by either variant in December 2021. Sixteen percent of those were Delta and 84% of those cases were the Omicron variant. For as we all know, by the end of 2021, the genomic epidemiology showed that Omicron became the most dominant variant, which you can see here by the small red line that is outleading all the other lines in this graph. So what did they find? Well, first, in keeping with our terminology, we're looking at vaccine effectiveness. So this is a real-world study. They found that a two-dose vaccine effectiveness versus the Omicron variant was 44% effective at 14 to 90 days but fell quite quickly. Three doses versus the Delta, by comparison, did actually extremely well, having a vaccine effectiveness of above 85% even after 60 days. And three doses of vaccine effectiveness versus the Omicron variant were shown to be effective at 71% in the initial period and 47.4% after 60 days. This led them to the conclusion that a three-dose vaccine effectiveness regimen versus hospitalizations with Delta or Omicron, actually, had above 99% vaccine effectiveness across the entire study population, with only four patients being hospitalized. And all of those patients were above 60 years old and all had chronic diseases or were immunocompromised. But how long this protection holds is something under tremendous examination, and the past month has seen a flurry of reports and studies published. So I will highlight just a few. The first is published last week in The Lancet, a systematic review and metaregression looking at preprint and peer-reviewed published articles from the 17th of June 2021 to the 2nd of December 2021, the era of the Delta variant. This study aimed to estimate vaccine efficacy or effectiveness one to six months after full vaccination. What they found was that both versus severe disease remained high up to six months for the vaccines that were studied, greater than 80% for the two mRNA vaccines. But by six months, there was definitely a drop in efficacy or effectiveness for severe disease by a mean of 9.5 to 10 percentage points. By comparison, there was a six-month drop in efficacy or effectiveness for rates of infection, 21 percentage points, and symptomatic disease, 25 to 32 percentage points, significantly increased. Now, there are three possibilities for these findings. One could be a lower vaccine effectiveness against the emergence of a new variant, truly waning immunity, which we will discuss, and the other is bias, as the authors noted. They were varying populations and study sizes for all the studies that they looked at. And indeed, there was a Centers for Disease Control, Morbidity, and Mortality report published on the 18th of February of this year that looked at such encounters. They found that there were 241,204 emergency department or urgent care encounters and over 93,000 hospitalizations from a period ranging from August to January of this year. It found that vaccine effectiveness versus COVID-19 associated visits and hospitalizations was higher after a third dose than the second, but it did wane with time since vaccination. Additionally, emergency encounters and urgent care encounters were 87 percent and 91 percent related to vaccine effectiveness during the two months after the third dose, but then decreased over time by the fourth month after that third dose. And interestingly, hospitalization vaccine effectiveness exceeded both of those during the same time period. So this suggests that antibody-induced responses does not actually present the entire picture. In January of this year, two studies were published that looked at cellular immunity in nature. The first, by Liu et al., looked at whether CD8 T cell responses contributed to protection versus severe disease. What they found was that cellular immunity was induced by current vaccines and highly conserved to the SARS-CoV-2 Omicron spike protein. And individuals who received either the Johnson & Johnson, so an adenoviral vaccine, or an mRNA vaccine, in this case Pfizer, showed durable spike-specific CD8 and CD4 T cell responses in several lineages versus both the Delta and Omicron variants. So despite waning antibody and neutralizing responses, this is extremely relevant for the Omicron variant, which evades the antibody response at a much greater rate. And on the same day, out of South Africa, a research team there demonstrated that 70 to 80% of CD4 and CD8 Omicron-specific spike T cell responses were actually maintained across all vaccinated groups, with a magnitude similar to Beta and Delta observations. And they also found that the majority of T cell responses cross-recognized the variant. Of course, this needs to be correlated with clinical real-world findings of decreased hospitalizations. But for immunocompromised patients, there is a greater challenge because of these fundamental issues that are crucial to understanding the differing immune responses in immunocompromised patients. For them, you can have specific or isolated defects in the immune system, such as those seen with HIV and a decrease in CD4 T cell lymphocytes. Or you can also have systemic issues, as can be seen with patients who receive bone marrow transplantation, leading to severe global immunodeficiency as a result of that treatment, or even those with solid organ transplantation who are on immunosuppression. These concepts are important as to why immunosuppressed patients do not appear to mount a standard immune response, and is important to keep in mind, especially with patients on anti-CD20 therapy, such as the monoclonal antibodies, the most notorious of which being rituximab. This drug has such a profound direct effect on B cells, and we currently recommend that patients be boosted as long after the completion of a vaccine series as possible. Now, two recent studies helped to shed some light on this issue. One, in preprint, in cancer cell, examined antibody responses following a third mRNA vaccination in patients with cancer in Denmark, looking at almost 540 patients, as you can see some with solid malignancies and others with hematologic malignancies. And they evaluated the anti-spike IgG titers at 36 days, 3 months, 6 months following two vaccinations, and then at 39 days and 3 months following a third vaccination. Now, for the immunocompromised population studied, about a third had lymphoma, a little over a third had chronic lymphocytic leukemia, or CLL, and almost a third had multiple myeloma. Many of the solid cancers were receiving active treatments, and among those with hematologic malignancies, 15% were receiving chemotherapy. Importantly, 12% were receiving anti-CD20 therapy, and 16% were the recipients of stem cell transplantation. As you can see, seropositivity for anti-spike IgG titers were higher in solid organ transplantation, or solid organ malignancies, versus those with hematologic malignancies. And improvement was most pronounced in patients, not surprisingly, with those solid malignancies. Less than 1% were seronegative 3 months post a third vaccine. And this suggests a fourth boosting doses for those undergoing treatment with anti-CD20 regimens may be warranted. Those with hematologic malignancies is certainly warranted. And then the question arises, is cell-mediated immunity conserved in those patients treated with anti-CD20 therapies who received an mRNA vaccination? And to look at this, a brief report was published on the 25th of February in JAMA Neurology looking at this question. It was a prospective study looking at outcomes of the quantification of CD4 and CD8 T-cells specific for those spike proteins of the vaccine strain and Delta Omicron variants, comparing a frequency before and after the third vaccination dose. What they found was that it advocated for booster dosing. Robust T-cell responses were generated from both variants, but this was a very small sample size. And we have all seen these patients in the ICUs with serious complications. So it really remains to be seen whether a fourth dose will alter these complications. Certainly an area of ongoing research. This is the data from that study from the Aquarolizumab-treated individuals, with each dot representing one patient and the lines connecting pre-booster and post-booster vaccination frequencies of spike-specific activation-induced marker assay positive CD8 T-cells specific for the vaccine strain and the Delta Omicron variants. The main thing I want you to get from this graph is that you can see the dramatic increase in those areas of positivity of the CD8 T-cell positivity following the booster administration for the vaccine strain and the Delta and Omicron variants. So just to summarize this, the observational data from various countries following national rollouts of mRNA vaccination supports trial findings that greater than 90% vaccine effectiveness is certainly possible in preventing COVID-19-related hospitalizations, ICU admissions, and death or severe disease. As discussed, we've also found that effectiveness significantly wanes over time and is certainly decreased against certain variants. And the efficacy data is consistent with immunogenicity studies that demonstrate robust binding and neutralizing antibody responses with variability by age, immunocompromised status, and over time from vaccination. All right, moving quickly to immune determinants of major safety signals. First, we have myocarditis and pericarditis, mainly in males and young adults. This has been reported more frequently than expected following the receipt of mRNA vaccinations. A similar pattern has not been reported following the receipt of the Johnson & Johnson vaccines. So among over 192 million people who had received an mRNA vaccine over the period seen in the slide, there were 1991 reports of myocarditis, and 1626 of those met the actual case definition of myocarditis. And this was a study that examined publicly available VAERS data in adults greater than 12 years of age in the United States. When broken down by age ranges, as you can see, from the 12 to 16 years old male ages, 70.7 cases per million doses administered. For the next age group, 105 cases per million doses. And for the next age group, 52 to 56.3 cases per million for Pfizer and Moderna, respectively. Now males comprised 82% of the cases of myocarditis that were seen. For females in the similar age group, there was a much less incidence. Most of these cases importantly were mild, the onset within the first week after vaccine receipt, and highest after the second vaccination, again in adolescent males and young adults. Of additional great concern has been vaccine-induced thrombotic thrombocytopenia, which has not been identified with the mRNA vaccines with the same signal. Now this, if you look at the left side of the screen, again has been associated with the adenoviral vaccines. And it's associated with autoantibodies to the platelet factor 4 antigen complex, similar to those found in patients with autoimmune heparin-induced thrombocytopenia, or HIT. Now thrombosis has often been found to occur at unusual sites, including the cerebral sinuses and the mesenteric vesicles, vessels, excuse me, sometimes with multiple sites involved. Most of the initially reported events occurred within the first two weeks of receipt of an initial vaccine dose, and in women or females under 60 years of age. But it should be known that a clear causal relation between either vaccines and thromboembolic disorders overall has not yet been identified. Another potential immune signal has been noted, again with adenoviral vector vaccines, but not with the same signal intensity with the mRNA vaccines, this with Guillain-Barre syndrome, or GBS. And it should also be known that GBS has been reported in the setting of COVID-19 infection as well. In the United States, as of July 24th last year, there were 132 reports from adenoviral vaccination, leading to a rate of 9.8 cases per million doses. And in Europe, with a similar time period, there were 227 cases noted. Lastly, another immune determinant of a possible major safety signal is that of sudden sensorineural hearing loss. Two studies were published in the Journal of the American Medical Association in the last week of February exploring this topic, one finding no link and the other showing a marginally higher incidence among Pfizer vaccine recipients. This study out of Israel looked at over 2 million patients who received the first dose of the mRNA vaccine, in which they found 91 cases of hearing loss reported. And with a similar amount of patients receiving the second dose, less numbers, but still a pretty elevated number of cases reported. This suggests a link, but overall the effect size is extremely small, so it's unclear whether the signal presents over the greater number in the general population. But unlike side effects from COVID-19 infection, which can certainly include hearing loss and other neurologic difficulties, sudden sensorineural hearing loss can be reversed or reduced if diagnosed and treated early. Lastly, just a brief mention on forthcoming data expected in vaccine trials. There are three major areas of COVID-19 vaccination that we are all eagerly awaiting data on. The first is a pan-coronavirus vaccine. In the United States, multiple institutions from the public, private, and military sectors are working on this. Additionally, the mixing and matching of variant vaccines to address antigenic shifts or drifts, similar to the influenza vaccine production. And of course, pediatric vaccination, which is beyond the scope of our lecture, but especially in younger children from infants to those currently eligible for vaccination. This is a huge area of continued viral circulation and reservoir, and given the essential need to keep schools open, is aggressively being explored. To expand on this briefly and to underscore the urgency of vaccinating these age groups, on the 28th of February, 2022, data was posted on a preprint server from the New York State Department of Health. So this study has not yet undergone peer review. It estimated the Pfizer-BioNTech vaccine effectiveness versus cases and hospitalizations among children ages 5 to 11 and 12 to 17 from December 2021 and January 2022. They looked at outcomes of new laboratory-confirmed cases or hospitalizations and compared by vaccination status and estimated vaccine effectiveness. Preliminary data shows that the Pfizer-BioNTech vaccine works substantially less well at preventing infection and hospitalizations in the younger age groups and shows a rapid and substantial decline in protection. To look at it directly, they noted vaccine effectiveness versus cases declined from 66% to 51% in the 12 to 17 age group and 68% to 12% in the 5 to 11 age group. And vaccine effectiveness versus hospitalizations declined 85% to 73% in the 12 to 17 age group and 100% effectiveness to 48% in the 5 to 11 age group, again over the period of the 13th of December to the 30th of January. The peer review process will certainly take some time as this is preliminary data, but it is not encouraging, especially as many areas are lifting public health mitigation restrictions. More adjusting and investigation is ongoing as this is one of the most vulnerable populations on the planet. So unfortunately, on that sour note, I will have to leave you, but I just wanted to give a special thanks to some of my colleagues and mentors at Emory University, Anish Mehta and Nadine Rafael from the Emory University School of Medicine. And thank you everyone for your attention.

COVID-19 vaccines

immune system

vaccine distribution

vaccine efficacy

booster vaccinations

safety signals