covid: vaccine effectiveness in cirrhosis

 A recent VA study assessed mRNA vaccine effectiveness in veterans with cirrhosis (see covid mRNA vaccine in cirrhosis jamaintmed2021 in dropbox, or doi:10.1001/jamainternmed.2021.4325). 

 

Details: 

-- mRNA vaccines were administered to 20,037 veterans (51% Moderna, 49% Pfizer), from Dec 18,2020 to March 17, 2021 

    -- 99.7% of those receiving a first dose who had a follow-up of least 42 days received a second dose 

-- median age 69, 97% male, 45% age 60-70yo/43% 70-85yo, 61% white/23% Black/8% Latinx, 42% alcohol-associated cirrhosis, 77% overweight or obese, 52% diabetes, 39% current smoker/32% former 

-- propensity matching for the control group (also 20,037 veterans as controls) to achieve similarity regarding age, sex, race or ethnicity, duration of follow-up, comorbidities, alcohol-associated liver disease, and severity of liver disease estimated by the Child-Turcotte-Pugh score 

-- primary outcome: Covid-19 infections, documented by a positive PCR, 28 days after the dose of either mRNA vaccine vs controls

-- secondary outcomes: Covid-19 infections, documented by positive PCR, 7 days after the second dose of either vaccine; hospitalization, or death from Covid 28 days after the first dose of either vaccine or 7 days after the second dose of either vaccine. vs controls 

 

Results: 

-- 28 days after the first vaccine dose: 

    -- 6 vs 17 Covid infections: 64.8% reduction in Covid infections (10.9%-86.1%), p=0.03

    -- 100% protection against hospitalization or death (numbers too low to be significant) 

        -- no Covid-related hospitalizations in anyone receiving the vaccine vs 3 in unvaccinated ones

        -- no Covid-related deaths in anyone receiving the vaccine vs 2 in unvaccinated ones 

    -- no difference in Covid infections between vaccine recipients and controls on days 0-7, 7-14, 14-21, and 21-28 days after the first dose (ie, benefit only after 28 days)

-- 7 days after the second dose: 3 patients in the vaccine group and 14 in the control group developed Covid-19 infections 

    -- 78.6% reduction in Covid infections (25.5%-93.8%), p=0.0154; 100% reduction in Covid-related hospitalization or deaths (0 vs 2 hospitalizations, 0 vs 1 control died) neither statistically significant  

-- overall findings were similar for Pfizer and Moderna

 

-- decompensated cirrhosis: 3142 patients 

   -- no difference in the first 28 days after the first shot, after 28 days 1 patient in the vaccine group and 2 in the control group had a Covid infection, no patient in the vaccine group vs 1 patient the control group was hospitalized and died. Overall protection 50.3% (4.9%-100%), p=0.568, not significant

-- compensated cirrhosis: 16,895 patients 

    -- after the first 28 days, 5 patients in the vaccine group vs 15 patients in the control group developed Covid infection, 66.8% reduction (8.6%-87.9%), P=0.0328 

    -- no cases of hospitalization or death with vaccine vs 2 hospitalizations and 1 death in the control group 

-- no difference in subgroup analysis of women, though the numbers of women and infection were small 

 

Commentary: 

-- this study did find that US veterans given the mRNA vaccine had a decrease in Covid-19 infections and especially severe infections despite underlying cirrhosis. And this was an important study in that patients with cirrhosis had been systematically excluded from the original mRNA vaccine studies 

    -- guidelines for immunocompromised persons were based on expert opinion only

-- though it seems that the vaccine was less effective in those with cirrhosis, the confidence intervals overlapped those for the general population in the original studies. Likely that this finding of "similar" effectiveness with cirrhosis is attributable to the low numbers of covid infections, but cannot say for sure based on this study

-- a prior blog noted the likely utility of a third dose of mRNA vaccine as assessed by antibody titers in those who were immunocompromised (specifically with solid organ transplants: see http://gmodestmedblogs.blogspot.com/2021/07/covid-3rd-vaccine-dose-for.html ). This VA study, in contrast, assessed clinical outcomes. It would have been useful if this VA study also had assessed antibody responses in order to ascertain the relationship between those antibody responses and clinical outcomes. For example, it would be useful to know if those veterans who did have clinical Covid infections in fact did not have detectable antibody responses at the time, since from the prior blog about half of those without antibody responses after two vaccines did have them after the third shot

-- chronic liver disease is associated with vaccine hyporesponsiveness, including to hepatitis B, pneumococcal, and influenza vaccines, and this is especially true in those with decompensated cirrhosis  

-- this article broadens the relationship between immunocompromised individuals and vaccine effectiveness, since it includes a very large number of people who had cirrhosis creating their immunocompromise. And it invites more studies dealing with other common immunocompromising conditions, such as poorly control diabetes or HIV or chronic steroid use 

-- another finding in the study was that the effectiveness of the vaccine was only really evident at least 28 days after the first dose. This brings up a few issues: 

    -- it is likely that immunocompromise does blunt the immunologic response, and this might take more than the initial 12 to 14 days to achieve vaccine efficacy found in the initial trials of both the Pfizer and Moderna vaccines 

    -- studies on the Delta variant of SARS-CoV-2 also have found that it takes longer for the vaccine to be effective, typically weeks after the second dose (see http://gmodestmedblogs.blogspot.com/2021/07/covid-moderna-j-work-ag-delta-virus-and.html). this VA study was done in a time period prior to the US invasion of the Delta variant, though would still be good to know the genetics of the virus in the infected veterans to see if any played a part in either the effectiveness of the vaccine or the time until effectiveness, to disentangle the relative contributions of immunocompromise and SARS-CoV-2 variants 

-- they were relatively few clinical events, making some of the outcomes difficult to interpret, such as in those with decompensated cirrhosis (also too few people in this category, only 3142 versus 16,895 with compensated cirrhosis). And, as noted above, lots of overlapping confidence intervals limiting our ability to have definitive effectiveness rates

-- propensity matched equivalency of controls to cirrhotic patients is an attempt to equalize the two groups according to some of the known baseline factors associated with severe Covid. They used the "nearest neighbor" approach, which has its own potential biases, see http://gmodestmedblogs.blogspot.com/2020/03/tramadol-fo-oa-inc-mortalityprobs-with.html for a poignant example 

 

Limitations: 

--as a retrospective observational study, we cannot determine causality, only association. There might well be some unmeasured residual confounding 

--this was a specific population of veterans, limited by their diversity (not a lot of women) and perhaps some other differences as compared to nonveterans. And even then, these were a select group of veterans who were connected to the VA system on a regular basis, likely different from the larger group of veterans 

-- the number of events was quite small. Important findings of decreased clinical events after vaccination did reach statistical significance, though many did not because of very small numbers 

 

So, a few points: 

-- though there were very few clinical events in patients with cirrhosis, there was a statistically significant decrease in those who were 28 days after the first vaccine, likely increasing by seven days after the second dose. this reinforces the importance of getting the second dose of vaccine

-- and, despite the poverty of data for people who are immunocompromised , this finding with cirrhosis is likely generalizable to other immunocompromising conditions

    -- though there is the imperative to have more specific data to determine who those people are. per the recent blog: 

        -- those with lupus (and what degree? lupus nephritis?), those on certain meds (steroids/how long and what dose, methotrexate, etc), those with poorly controlled diabetes (and at what level of A1c), those with HIV and CD4 <200 (or different target), those with renal failure (and what level? it seems by the study above that there is (not surprisingly) a gradient of vaccine efffectiveness, with apparently better effectiveness in those with less severe immunocompromise (compensated vs uncompensated cirrhosis)

-- there was no significant difference between the two mRNA vaccines 

-- so the good news is that the vaccine seems to work in some cases of immunocompromise (likely solid organ transplants, cirrhosis). now we need more granular data to determine the optimum approach to different subsets of people...

geoff

 

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

to get access to all of the blogs (2 options):

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

 

or: go to https://www.bucommunitymedicine.org/ , a website from the Community Medicine section at Boston Medical Center.  This site does have a very searchable and accessible list of my blogs (though there have been a few that did not upload over the last year or two). but overall it is much easier to view blogs and displays more at a time.

 

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list