Comparative Effectiveness and Antibody Responses to Moderna and Pfizer-BioNTech COVID-19 Vaccines among Hospitalized Veterans — Five Veterans Affairs Medical Centers, United States, February 1–September 30, 2021

The mRNA COVID-19 vaccines (Moderna and Pfizer-BioNTech) provide strong protection against severe COVID-19, including hospitalization, for at least several months after receipt of the second dose (1,2). However, studies examining immune responses and differences in protection against COVID-19–associated hospitalization in real-world settings, including by vaccine product, are limited. To understand how vaccine effectiveness (VE) might change with time, CDC and collaborators assessed the comparative effectiveness of Moderna and Pfizer-BioNTech vaccines in preventing COVID-19–associated hospitalization at two periods (14–119 days and ≥120 days) after receipt of the second vaccine dose among 1,896 U.S. veterans at five Veterans Affairs medical centers (VAMCs) during February 1–September 30, 2021. Among 234 U.S. veterans fully vaccinated with an mRNA COVID-19 vaccine and without evidence of current or prior SARS-CoV-2 infection, serum antibody levels (anti-spike immunoglobulin G [IgG] and anti-receptor binding domain [RBD] IgG) to SARS-CoV-2 were also compared. Adjusted VE 14–119 days following second Moderna vaccine dose was 89.6% (95% CI = 80.1%–94.5%) and after the second Pfizer-BioNTech dose was 86.0% (95% CI = 77.6%–91.3%); at ≥120 days VE was 86.1% (95% CI = 77.7%–91.3%) for Moderna and 75.1% (95% CI = 64.6%–82.4%) for Pfizer-BioNTech. Antibody levels were significantly higher among Moderna recipients than Pfizer-BioNTech recipients across all age groups and periods since vaccination; however, antibody levels among recipients of both products declined between 14–119 days and ≥120 days. These findings from a cohort of older, hospitalized veterans with high prevalences of underlying conditions suggest the importance of booster doses to help maintain long-term protection against severe COVID-19.^†

During February 1–September 30, 2021, adults aged ≥18 years hospitalized at five VAMCs (Atlanta, Georgia; the New York City borough of the Bronx; Houston, Texas; Los Angeles, California; and Palo Alto, California) were screened for inclusion in this test-negative case-control assessment (1,3). Patients with COVID-19–like illness^§ who received a positive SARS-CoV-2 nucleic acid amplification test result were included as case-patients and those with COVID-19–like illness and negative SARS-CoV-2 test results were included as controls^¶ (4).

Data on demographic characteristics, clinical history, and COVID-19 vaccination history were abstracted from electronic health records.** Full vaccination was defined as receipt of 2 doses of an mRNA COVID-19 vaccine (Moderna or Pfizer-BioNTech) ≥14 days before the SARS-CoV-2 test. Participants who received only 1 dose of an mRNA COVID-19 vaccine, 2 mRNA doses with receipt of the second dose <14 days before the SARS-CoV-2 test, mixed mRNA vaccine products, 3 vaccine doses, or the Janssen (Johnson & Johnson) COVID-19 vaccine were excluded from the analysis.^††

Available residual clinical serum specimens were collected from fully vaccinated hospitalized control patients at all sites and tested at CDC. Specimens were tested using the V-PLEX SARS-CoV-2 panel 2 kit (Meso Scale Diagnostics)^§§ to measure binding IgG levels against three SARS-CoV-2 antigens: the spike protein (anti-spike), the receptor-binding domain of the spike protein (anti-RBD), and the nucleocapsid protein (anti-nucleocapsid). Levels were reported in international binding antibody units (BAU) per milliliter (mL). Control participants with anti-nucleocapsid antibodies (>11.8 BAU/mL), suggesting a prior SARS-CoV-2 infection, were excluded from the final analysis.

VE to prevent COVID-19–associated hospitalization (calculated as 1 – adjusted odds ratio [aOR] × 100)^¶¶ was estimated using multivariable logistic regression to compare the odds of full vaccination between case-patients and controls. Models were adjusted for VAMC site, admission date, and age (with the use of cubic splines), sex, and race/ethnicity.*** VE between subgroups was compared using 95% CIs. In the antibody analysis, pairwise comparisons of median anti-spike IgG and anti-RBD IgG levels using the Wilcoxon rank-sum test and p-values were calculated among participants by age category, vaccine product received, and time since vaccination (14–119 days and ≥120 days after the second vaccine dose). Because vaccines might not elicit a strong immune response^††† in some persons with immunocompromising conditions,^§§§ differences including and excluding this group were examined. Analyses were conducted using SAS (version 9.4; SAS Institute). For all analyses, statistical significance was set at p<0.05. Protocols were reviewed and approved by the VAMC Research and Development Committee at each site. The activity was also reviewed by CDC and conducted consistent with applicable federal law and CDC policy.^¶¶¶

During February 1–September 30, 2021, a total of 2,329 hospitalized U.S. veterans with COVID-19–like illness met inclusion criteria. After excluding 433 persons with missing data or ineligible vaccination status,**** 755 case-patients and 1,141 controls were included in the analysis. Among these 1,896 patients, 1,758 (92.7%) were male, the median age was 67 years (IQR = 59–75 years), 942 (49.7%) were Black, and 162 (8.5%) were Hispanic (Table 1). Effectiveness of the Moderna vaccine was 89.6% (95% CI = 80.1%–94.5%) 14–119 days after the second vaccine dose and 86.1% (95% CI = 77.7%–91.3%) at ≥120 days (Table 2). Effectiveness of the Pfizer-BioNTech vaccine was 86.0% (95% CI = 77.6%–91.3%) at 14–119 days and 75.1% (95% CI = 64.6%–82.4%) at ≥120 days.

Antibody testing was performed on sera available from 259 of 638 (40.6%) fully vaccinated controls. No significant differences in age, sex, or vaccine product received were observed between fully vaccinated controls with and without available sera (Supplementary Table 1, https://stacks.cdc.gov/view/cdc/112103). After excluding 25 (9.7%) control specimens with anti-nucleocapsid antibodies, the analysis included 90 (38.5%) controls fully vaccinated with the Moderna vaccine (median age = 72 years; median interval from second dose to serum collection = 75 days; 24 [26.7%] with an immunocompromising condition) and 144 (61.5%) who were fully vaccinated with the Pfizer-BioNTech vaccine (median age = 73 years; median interval from second dose to serum collection = 102 days; 38 [26.4%] with an immunocompromising condition). Among fully vaccinated Moderna controls, anti-spike IgG levels were higher among persons with sera collected 14–119 days after the second vaccine dose (median = 759 BAU/mL; IQR = 348–2,086 BAU/mL) compared with ≥120 days (median = 266 BAU/mL; IQR = 133–441 BAU/mL) (p = 0.002) (Figure). Anti-spike IgG levels were also higher among fully vaccinated Pfizer-BioNTech controls at 14–119 days after receipt of dose 2 (median = 187 BAU/mL; IQR = 50–493 BAU/mL) than at ≥120 days (median = 62 BAU/mL; IQR = 25–141 BAU/mL) (p = 0.001). At 14–119 days after the second dose, anti-spike IgG levels were higher among controls fully vaccinated with the Moderna vaccine compared with those who received the Pfizer-BioNTech vaccine among persons aged 18–64 years (median = 612 versus 340; p = 0.018) and ≥65 years (median = 792 versus 152; p<0.001). At ≥120 days, anti-spike IgG levels were also higher among controls fully vaccinated with the Moderna vaccine compared with the Pfizer-BioNTech vaccine among persons aged 18–64 years (median = 267 versus 106; p = 0.006) and ≥65 years (median = 266 versus 57; p = 0.003). Relative differences in anti-RBD IgG levels across groups were similar to differences in anti-spike IgG levels (Supplementary Table 2, https://stacks.cdc.gov/view/cdc/112104), and differences in anti-SARS-CoV-2 antibody levels were similar across groups with immunocompromised persons included or excluded from the analysis.

Acknowledgments

Daoling Bi, Cristina Cardemil, Aaron Curns, Fiona Havers, Jefferson Jones, Lindsay Kim, L. Clifford McDonald.

Corresponding author: Kristina L. Bajema, [email protected].

¹CDC COVID-19 Response Team; ²Karna, LLC, Atlanta, Georgia; ³Michael E. DeBakey Veterans Affairs Medical Center, Houston, Texas; ⁴Department of Medicine, Baylor College of Medicine, Houston, Texas; ⁵Atlanta VA Medical Center, Atlanta, Georgia; ⁶Department of Medicine, Emory University School of Medicine, Atlanta, Georgia; ⁷Department of Global Health, Rollins School of Public Health, Emory University, Atlanta, Georgia; ⁸Veterans Affairs Greater Los Angeles Healthcare System, Los Angeles, California; ⁹Department of Medicine, David Geffen School of Medicine at UCLA, Los Angeles, California; ¹⁰Veterans Affairs Palo Alto Health Care System, Palo Alto, California; ¹¹Public Health Program Office, Department of Veterans Affairs, Washington, DC; ¹²Department of Medicine, Stanford University, Stanford, California; ¹³Department of Medicine, Icahn School of Medicine at Mount Sinai, New York, New York; ¹⁴James J. Peters Veterans Affairs Medical Center, Bronx, New York, New York; ¹⁵General Dynamics Information Technology, Falls Church, Virginia.

References

1. Bajema KL, Dahl RM, Prill MM, et al.; SUPERNOVA COVID-19; Surveillance Group; Surveillance Platform for Enteric and Respiratory Infectious Organisms at the VA (SUPERNOVA) COVID-19 Surveillance Group. Effectiveness of COVID-19 mRNA vaccines against COVID-19–associated hospitalization—five Veterans Affairs Medical Centers, United States, February 1–August 6, 2021. MMWR Morb Mortal Wkly Rep 2021;70:1294–9. https://doi.org/10.15585/mmwr.mm7037e3 PMID:34529636
2. Tenforde MW, Self WH, Naioti EA, et al.; IVY Network Investigators; IVY Network. Sustained effectiveness of Pfizer-BioNTech and Moderna vaccines against COVID-19–associated hospitalizations among adults—United States, March–July 2021. MMWR Morb Mortal Wkly Rep 2021;70:1156–62. https://doi.org/10.15585/mmwr.mm7034e2 PMID:34437524
3. Meites E, Bajema KL, Kambhampati A, et al. Adapting the Surveillance Platform for Enteric and Respiratory Infectious Organisms at United States Veterans Affairs Medical Centers (SUPERNOVA) for COVID-19 among hospitalized adults: surveillance protocol. Front Public Health 2021;9:739076. https://doi.org/10.3389/fpubh.2021.739076 PMID:34778173
4. Jackson ML, Nelson JC. The test-negative design for estimating influenza vaccine effectiveness. Vaccine 2013;31:2165–8. https://doi.org/10.1016/j.vaccine.2013.02.053 PMID:23499601
5. Khoury DS, Cromer D, Reynaldi A, et al. Neutralizing antibody levels are highly predictive of immune protection from symptomatic SARS-CoV-2 infection. Nat Med 2021;27:1205–11. https://doi.org/10.1038/s41591-021-01377-8 PMID:34002089
6. Gilbert PB, Montefiori DC, McDermott A, et al. Immune correlates analysis of the mRNA-1273 COVID-19 vaccine efficacy trial. medRxiv [Preprint posted online August 15, 2021]. https://www.medrxiv.org/content/10.1101/2021.08.09.21261290v4
7. Self WH, Tenforde MW, Rhoads JP, et al.; IVY Network. Comparative effectiveness of Moderna, Pfizer-BioNTech, and Janssen (Johnson & Johnson) vaccines in preventing COVID-19 hospitalizations among adults without immunocompromising conditions—United States, March–August 2021. MMWR Morb Mortal Wkly Rep 2021;70:1337–43. https://doi.org/10.15585/mmwr.mm7038e1 PMID:34555004
8. Steensels D, Pierlet N, Penders J, Mesotten D, Heylen L. Comparison of SARS-CoV-2 antibody response following vaccination with BNT162b2 and mRNA-1272. JAMA 2021;326:1533–5. https://doi.org/10.1001/jama.2021.15125 PMID:34459863
9. Laing ED, Weiss CD, Samuels EC, et al. Durability of antibody responses and frequency of clinical and sub-clinical SARS-CoV-2 infection six months after BNT162b2 COVID-19 vaccination in healthcare workers. medRxiv [Preprint posted online October 18, 2021]. https://www.medrxiv.org/content/10.1101/2021.10.16.21265087v1
10. Doria-Rose N, Suthar MS, Makowski M, et al.; mRNA-1273 Study Group. Antibody persistence through 6 months after the second dose of mRNA-1273 vaccine for COVID-19. N Engl J Med 2021;384:2259–61. https://doi.org/10.1056/NEJMc2103916 PMID:33822494

Abbreviations: COPD = chronic obstructive pulmonary disease; VAMC = Veterans Affairs medical center.
* Case-patients were defined as patients with COVID-19–like illness (i.e., presence of fever, new or worsened cough or shortness of breath, loss of taste or smell, oxygen saturation on room air <94%, requirement for noninvasive ventilation or endotracheal intubation with mechanical ventilation, or chest radiograph or computed tomography pulmonary findings consistent with pneumonia) who tested positive for SARS-CoV-2 by nucleic acid amplification test performed within 14 days before admission or during the first 72 hours of hospitalization. Controls were defined as patients with COVID-19–like illness and negative SARS-CoV-2 test results during the same period.
^† Included non-Hispanic American Indian and Alaska Native, non-Hispanic Asian and Other Pacific Islander, non-Hispanic multiple races, or non-Hispanic Other race.
^§ Included residence before admission to VAMC and non-VAMC nursing facilities as well as other VAMC long-term housing (e.g., domiciliary).
^¶ The Bronx is a borough in New York City.
** COVID-19 vaccination status includes unvaccinated, defined as no receipt of any SARS-CoV-2 vaccine, and fully vaccinated, defined as receipt of both doses of an mRNA (Pfizer-BioNTech or Moderna) ≥14 days before the first SARS-CoV-2 test performed within 14 days before admission or during the first 72 hours of hospitalization.
^†† Included coronary artery disease, myocardial infarction, peripheral vascular disease, carotid artery stenosis.
^§§ Body mass index ≥30 kg/m².
^¶¶ Included HIV/AIDS, malignancy, history of solid organ or stem cell transplant, or immunosuppressive therapy (systemic steroids, chemotherapy, or other immunosuppressive therapy within 1 month of SARS-CoV-2 test).
*** Tobacco use was defined as smoking of cigarettes, cigars, or pipes. Current use of tobacco was defined as use within the previous 12 months of hospitalization, whereas former use occurred >12 months before hospitalization.

Abbreviation: VE = vaccine effectiveness.
*All nonstratified models adjusted for study site, time (admission date), age, sex, and race/ethnicity. Stratified models exclude adjustment for stratification variable.
^† Includes unvaccinated, defined as no receipt of any SARS-CoV-2 vaccine, and fully vaccinated, defined as receipt of both doses of an mRNA (Pfizer-BioNTech or Moderna) ≥14 days before the first SARS-CoV-2 test performed within 14 days before admission or during the first 72 hours of hospitalization.
^§ The five Veterans Affairs medical centers are located in Atlanta, Georgia; the New York City borough of the Bronx; Houston, Texas; Los Angeles, California; and Palo Alto, California.
^¶ Among fully vaccinated, time since second dose of COVID-19 mRNA vaccine.

FIGURE. Serum anti-spike and anti-receptor binding domain immunoglobulin G levels* after full vaccination among hospitalized veterans without current or previous SARS-CoV-2 infection^† — five Veterans Affairs medical centers,^§ United States, February 1–September 30, 2021^⁋

Abbreviations: BAU = binding antibody units; IgG = immunoglobulin G; RBD = receptor binding domain.

* Anti-spike and anti-RBD IgG levels were measured in sera of hospitalized veterans collected at or within 2 days of hospital admission. In these box and whisker plots, the central horizontal line of each box plot represents the median, with the box denoting the IQR and the whiskers representing 1.5 x IQR.

^† Excluded 25 controls with anti-nucleocapsid antibodies (>11.8 BAU/mL), suggesting a previous SARS-CoV-2 infection.

^§ The five Veterans Affairs medical centers are located in Atlanta, Georgia; the New York City borough of the Bronx; Houston, Texas; Los Angeles, California; and Palo Alto, California.

^¶ Serum specimens collected during March 22–August 31, 2021.