Immunocompromised patients, including those who received an organ transplant, were excluded from coronavirus disease 2019 (COVID-19) vaccine clinical trials.1 , 2 Transplant patients are at high risk of severe disease3 but may not manifest adequate antispike antibody responses after COVID-19 vaccination, raising concern for persisting risk.4 , 5 The objective of this study was to evaluate the effectiveness of COVID-19 vaccination for the prevention of COVID-19 and COVID-19–related death.
Liver transplant recipients (aged > 18 years) in active care from December 15, 2020 to September 12, 2021 were identified from the Veterans Outcomes and Costs Associated with Liver Disease cohort using International Classification of Diseases codes.6 , 7 Patients were considered fully vaccinated 14 days after the second dose of either the Pfizer BNT162b2 mRNA or the Moderna 1273 mRNA. Patients who were unvaccinated, partially vaccinated (received only 1 dose of the vaccine or had fewer than 14 days of follow-up after the second dose) with an mRNA vaccine, or received the Janssen Ad26.SARS CoV2 vaccine were considered as control subjects. Outcomes evaluated included COVID-19 (documented by positive COVID-19 polymerase chain reaction), symptomatic COVID-19, and COVID-19–related death, assessed from the date of full vaccination (14 days after the second dose of an mRNA vaccine) for vaccinated patients and from the first availability of COVID-19 vaccines within the VA system (December 18, 2020) for unvaccinated patients.
Propensity scores were derived through the logistic regression adjusted by the covariates of age group, sex, current smoking, kidney transplantation, comorbidities including chronic obstructive pulmonary disease and hypertension, primary immunosuppression, and antimetabolite therapy in patients who received vaccinations vs unvaccinated control subjects. Inverse propensity weights were calculated and used as weighted score into the unadjusted and adjusted hazard regression.
Descriptive variables were compared using Wilcoxon and χ2 tests as appropriate. The hazard ratios of COVID-19 infection, symptomatic COVID-19, or COVID-19–related death associated with vaccination were estimated using Cox proportional hazard regression models, controlling for a priori determined variables known to be correlated with surviva,l and probability of being infected including age, body mass index, race, tobacco use, prior receipt of a kidney transplant, and time after transplantation.3 Event-free cases and control subjects were censored at the end of the study period. Patients who received only 1 dose of an mRNA vaccine or the Ad26.SARS CoV2 vaccine were included as control subjects and censored on the date of receipt of the first dose. Vaccine efficacy was calculated as 1 minus risk ratio, where risk ratio is the ratio of risk after full vaccination among vaccinated to risk after the matched vaccine date among unvaccinated.
Statistical significance was defined using 2-sided tests with α = 0.05. Statistical analysis was performed using SAS 4.9 (SAS Inc, Cary, NC). The study was approved by the institutional review boards at all participating VA medical centers.
Of 1924 eligible liver transplant recipients, 1133 (58.9%) were fully vaccinated with 2 doses of an mRNA vaccine. Patients who were vaccinated were more likely to be men, white, have chronic obstructive pulmonary disease, be hypertensive, have received a kidney transplant, have received a calcineurin inhibitor plus antimetabolite, less likely to be active smokers, and were fewer years after transplantation (Supplementary Table 1).
The outcomes of COVID-19 were observed in 24 of 1133 vaccinated and 43 of 791 control subjects, symptomatic COVID-19 in 14 vaccinated and 26 control subjects, and COVID-19–related death in 2 vaccinated and 11 control subjects. On multivariable analysis, full vaccination with a COVID-19 vaccine was associated with a significant reduction in COVID-19 (adjusted hazard ratio, 0.36; 95% confidence interval, 0.26–0.51; P < .0001), symptomatic COVID-19 (adjusted hazard ratio, 0.42; 95% confidence interval, 0.27–0.65; P < .0001), and COVID-19–related death (adjusted hazard ratio, 0.13; 95% confidence interval, 0.04–0.37; P = .0002) (Supplementary Table 2 and Figure 1 ).
Figure 1.
Cumulative incidence of (A) COVID-19 infection, (B) symptomatic COVID-19, or (C) COVID-19–related death in mRNA vaccinated and control liver transplant participants.
Of the 2 postvaccination COVID-19 deaths, the first was an 80-year-old man who was 9 years post-transplant on tacrolimus, with compensated graft cirrhosis, nonischemic cardiomyopathy, atrial fibrillation, and chronic kidney disease who developed COVID-19 18 days after the second dose of mRNA-1273 vaccine. This patient required mechanical ventilation, pressors, and dialysis and died from respiratory failure. The second patient was an 82-year-old man with coronary artery disease, diabetes mellitus, and chronic kidney disease who was 8 years post-transplant on single-agent tacrolimus and developed COVID-19 pneumonia 10 days after the second dose of the mRNA-1273 vaccine. He required noninvasive ventilation, refused intubation, and died of respiratory failure.
Recent studies report concerning data that only 54% of solid organ transplant recipients, 43% on antimetabolite therapy and 46% of kidney transplant recipients, mount antispike antibodies after COVID-19 mRNA vaccine.4 , 5 However, our data show that receipt of full vaccination with 2 doses of an mRNA vaccine was associated with a 64%% decrease in COVID-19 infection, 58% decrease in symptomatic COVID-19, and 87% decrease in COVID-19–related death in liver transplant recipients.
Protection from the vaccination may be of lower magnitude than that observed in healthy individuals, yet it exceeds the level of protection predicted by antibody titers.4 , 5 In addition to humoral immunity, adaptive T-cell–mediated responses may also confer vaccine protection and require further study. Similar observations have been reported with influenza in liver transplant recipients in whom, despite low antibody titers, vaccination has been associated with significant reduction in hospitalizations and death.8
Limitations of the present study include a retrospective design with potential for residual confounding, a male-predominant cohort, small numbers of outcomes, lack of measurement of antispike antibodies or T-cell response to COVID-19, and lack of systematic testing for severe acute respiratory syndrome coronavirus 2. Control subjects could have received the vaccine outside the VA, resulting in misclassification bias toward the null. Individuals in either group may have tested positive for COVID-19 outside the VA system, possibly more likely to have occurred in unvaccinated individuals and again biasing toward the null. The study was completed before the recommendation for a third dose of the mRNA BNT162b2 vaccine. Strengths of the present study include a large sample size for a geographically diverse, national transplant population. We were able to complete this study at a point where we had adequate numbers of unvaccinated control subjects. In summary, our findings show that full vaccination with 2 doses of a COVID-19 mRNA vaccine is associated with a decrease in COVID-19 infection and death in liver transplant recipients.
Acknowledgments
CRediT Authorship Contributions
Binu V. John, MD, MPH (Conceptualization: Lead; Funding acquisition: Lead; Investigation: Lead; Methodology: Equal; Supervision: Equal; Writing – original draft: Lead). Yangyang Deng, MS (Formal analysis: Lead; Investigation: Equal; Methodology: Equal; Software: Equal; Writing – review & editing: Supporting). Nidah S. Khakoo, MD (Data curation: Lead; Investigation: Equal; Validation: Equal; Writing – review & editing: Supporting). Tamar H. Taddei, MD (Funding acquisition: Supporting; Project administration: Equal; Resources: Equal; Writing – review & editing: Equal). David E. Kaplan, MD (Funding acquisition: Supporting; Methodology: Equal; Project administration: Equal; Writing – review & editing: Equal). Bassam Dahman, PhD (Data curation: Lead; Formal analysis: Lead; Methodology: Lead; Resources: Equal; Software: Lead; Supervision: Equal; Writing – review & editing: Equal).
Footnotes
Conflicts of interest This author discloses the following: Binu V. John receives institutional research funding from Exelixis, Exact Sciences, Glycotest Inc, Glaxo Smithkline, and Viking therapeutics and advises Dova. The remaining authors disclose no conflicts.
Funding Services supporting the analysis and interpretation of the data of this research project were generated by the VCU Massey Cancer Center Biostatistics Shared Resource, supported, in part, with funding from NIH-NCI Cancer Center Support Grant P30 CA016059. The VCU Massey Cancer Center Biostatistics Shared Resource or the NIH-NCI had no role in the design of the current study. We acknowledge data and support from the VA COVID-19 shared data resource.
Disclaimer The authors prepared this work in their personal capacity. The opinions expressed in this article are the author's own and do not reflect the view of the Department of Veterans Affairs or the United States government.
Note: To access the supplementary material accompanying this article, visit the online version of Gastroenterology at www.gastrojournal.org and at https://doi.org/10.1053/j.gastro.2021.11.001.
Supplementary Material
Supplementary Table 1.
Baseline Characteristics of Study Participants
| Variables | Vaccine |
Control |
P |
|---|---|---|---|
| (n = 1133) | (n = 791) | ||
| Patient who received the first dose | … | ||
| Pfizer BNT162b2 mRNA vaccine | 539 (47.6) | … | |
| Moderna 1273 mRNA vaccine | 596 (52.4) | … | |
| Sex | <.0001 | ||
| Male | 1106 (97.6) | 736 (93.1) | |
| Female | 27 (2.4) | 55 (6.9) | |
| Median age, y (IQR) | 68.9 (7.9) | 68.6 (8.7) | .4629 |
| Age group | .0184 | ||
| <50 y | 30 (2.6) | 37 (4.7) | |
| 50–59.9 y | 110 (9.7) | 76 (9.6) | |
| 60–69.9 y | 516 (45.5) | 341 (43.1) | |
| 70–84.9 y | 477 (42.1) | 333 (42.1) | |
| >85 y | 0 (0.0) | 4 (0.5) | |
| White | 755 (66.6) | 479 (60.6) | .0062 |
| Median body mass index (IQR) | 30.2 (5.5) | 30.2 (4.8) | .3851 |
| Current smoker | 227 (20.0) | 180 (22.8) | .1505 |
| Chronic obstructive pulmonary disease | 319 (28.2) | 137 (17.3) | <.0001 |
| Hypertension | 917 (80.9) | 562 (71.1) | <.0001 |
| Kidney transplantation | 223 (19.7) | 117 (14.8) | .0057 |
| Tacrolimus | 917 (80.9) | 454 (57.4) | <.0001 |
| Cyclosporine | 78 (6.9) | 47 (5.9) | .4092 |
| Calcineurin inhibitor–based primary immunosuppression | 1055 (93.1) | 568 (71.8) | <.0001 |
| Mycophenolate | 610 (53.8) | 331 (41.8) | <.0001 |
| Azathioprine | 86 (7.6) | 36 (4.5) | .0071 |
| Antimetabolites | 696 (61.4) | 367 (46.4) | <.0001 |
| Median year from transplantation (IQR) | 5.6 (5.0) | 8.4 (7.5) | <.0001 |
Values are n (%) unless otherwise defined. IQR, interquartile range.
Supplementary Table 2.
Inverse Probability of Treatment Weighting Multivariable Hazard Ratios for the Risk of COVID-19, Symptomatic COVID-19, and COVID-19–related Death, Comparing Liver Transplant Recipients Who Are Fully Vaccinated for COVID-19 With an mRNA Vaccine vs Control Subjects
| Variable | COVID-19 |
Symptomatic COVID-19 |
COVID-19 related death |
|||
|---|---|---|---|---|---|---|
| aHR (95% CI) | P | aHR (95% CI) | P | aHR (95% CI) | P | |
| Number of patients | 1924 | 1924 | 1924 | |||
| Control | 791 | 791 | 791 | |||
| Vaccine | 1133 | 1133 | 1133 | |||
| Number of events | 67 | 40 | 13 | |||
| Control | 43 | 26 | 11 | |||
| Vaccine | 24 | 14 | 2 | |||
| Group | ||||||
| Control | REF | REF | REF | |||
| Vaccine | 0.36 (0.26–0.51) | <.0001 | 0.42 (0.27–0.65) | <.0001 | 0.13 (0.04–0.37) | .0002 |
| Age | 1.03 (0.97–1.04) | .7228 | 1.03 (0.98–1.03) | .8007 | 1.16 (1.05–1.28) | .0026 |
| Body mass index | 1.03 (1.01–1.06) | .0016 | 1.06 (1.03–1.09) | <.0001 | 1.09 (1.04–1.14) | .0001 |
| Race | ||||||
| Nonwhite | REF | REF | REF | |||
| White | 1.29 (0.93–1.78) | .1309 | 1.11 (0.74–1.67) | .6146 | 1.56 (0.63–3.90) | .3404 |
| Tobacco use | ||||||
| Current nonsmoker | REF | REF | REF | |||
| Current Smoker | 1.20 (0.84–1.71) | .3154 | 1.48 (0.96–2.28) | .0780 | 2.54 (1.21–5.31) | .01353 |
| Kidney transplantation | ||||||
| No | REF | REF | REF | |||
| Yes | 1.11 (0.75–1.64) | .6085 | 1.03 (0.61–1.74) | .9079 | 1.32 (0.59–2.92) | .5010 |
| Year from transplantation | 0.94 (0.90–0.97) | .0007 | 0.96 (0.91–1.01) | .0847 | 0.94 (0.83–1.05) | .2778 |
References
- 1.Polack F.P., et al. N Engl J Med. 2020;383:2603–2615. doi: 10.1056/NEJMoa2034577. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.John B.V., et al. JAMA Intern Med. 2021;181:1306–1314. doi: 10.1001/jamainternmed.2021.4325. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Webb G.J., et al. Lancet Gastroenterol Hepatol. 2020;5:1008–1016. doi: 10.1016/S2468-1253(20)30271-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Boyarsky B.J., et al. JAMA. 2021;325:2204–2206. doi: 10.1001/jama.2021.7489. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Boyarsky B.J., et al. JAMA. 2021;325:1784–1786. doi: 10.1001/jama.2021.4385. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.John B.V., et al. Hepatology. 2021;74:879–891. doi: 10.1002/hep.31776. [DOI] [PubMed] [Google Scholar]
- 7.John B.V., et al. Am J Gastroenterol. 2021;116:1913–1923. doi: 10.14309/ajg.0000000000001280. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kumar D., et al. Clin Infect Dis. 2018;67:1322–1329. doi: 10.1093/cid/ciy294. [DOI] [PubMed] [Google Scholar]