Detectable plasma severe acute respiratory syndrome coronavirus 2 spike antigen is associated with poor antibody response following third messenger RNA vaccination in kidney transplant recipients

Abstract

Background

Kidney transplant recipients (KTRs) generate lower antibody responses to messenger RNA (mRNA)‐based severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) vaccination, yet precise mechanisms for this poor response remain uncertain. One potential contributor is suboptimal spike antigen (sAg) translation and expression owing to transplant immunosuppression, which might lead to insufficient exposure to develop humoral and/or cellular immune responses.

Methods

Within a single‐arm clinical trial, 65 KTRs underwent ultrasensitive plasma sAg testing before, and 3 and 14 days after, the third mRNA vaccine doses. Anti‐SARS‐CoV‐2 spike antibodies (anti‐receptor binding domain [anti‐RBD]) were serially measured at 14 and 30 days post‐vaccination. Associations between sAg detection and clinical factors were assessed. Day 30 anti‐RBD titer was compared among those with versus without sAg expression using Wilcoxon rank sum testing.

Results

Overall, 16 (25%) KTRs were sAg positive (sAg+) after vaccination, peaking at day 3. Clinical and laboratory factors were broadly similar in sAg(+) versus sAg(‐) KTRs. sAg(+) status was significantly negatively associated with day 30 anti‐RBD response, with median (interquartile range) 10.8 (<0.4–338.3) U/mL if sAg(+) versus 709 (10.5–2309.5) U/mL if sAg(‐) (i.e., 66‐fold lower; p = .01).

Conclusion

Inadequate plasma sAg does not likely drive poor antibody responses in KTRs, rather sAg detection implies insufficient immune response to rapidly clear vaccine antigen from blood. Other downstream mechanisms such as sAg trafficking and presentation should be explored.

Inadequate plasma spike antigen does not drive poor severe acute respiratory syndrome coronavirus 2 messenger RNA vaccine antibody response in kidney transplant recipients. Rather, a detectable spike likely indicates insufficient immune response to vaccination to rapidly clear vaccine antigens from blood.

Abbreviations

COVID‐19

coronavirus disease 2019

DTT

dithiothreitol

IQR

interquartile range

KTR

kidney transplant recipient

mRNA

messenger RNA

MSD

Meso Scale Diagnostics

N

SARS‐CoV‐2 nucleocapsid

PCR

polymerase chain reaction

sAg

SARS‐CoV‐2 spike antigen

SARS‐CoV‐2

severe acute respiratory syndrome coronavirus 2

SOTR

solid organ transplant recipient

WHO

World Health Organization

1. INTRODUCTION

Messenger RNA (mRNA)‐based severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) vaccination induces strong anti‐spike antibody response in the general population and contributes to immunoprotection, whereas many solid organ transplant recipients (SOTRs) demonstrate attenuated responses and endure higher breakthrough infection rates. ¹ , ² Additional doses can increase immune responses, yet some SOTRs, particularly those with preceding poor antibody responses, remain unable to produce high‐titer antibodies. ³ , ⁴ Many clinical factors are negatively associated with antibody response, including proximity to transplant and use of anti‐metabolite immunosuppression, yet precise molecular mechanisms underlying these poor responses remain unclear. One unexplored possibility is whether post‐transplant immunosuppressive medications alter the ability of the spike mRNA sequence to be translated into spike protein, thus decreasing the amount of antigen to which the immune system is exposed and hampering the development of an adaptive response. High‐sensitivity assays are capable of detecting circulating spike antigen (sAg), specifically the S1 subunit, in healthy vaccinated populations, as well as in convalescent cohorts, ⁵ , ⁶ though these data are not available for immunocompromised populations. It is also uncertain whether the presence or absence of circulating sAg is associated with vaccine response in these high‐risk groups. Understanding whether sAg is produced following vaccination in immunocompromised populations and contributes to antibody response has important implications for SARS‐CoV‐2 and other infectious disease vaccine strategies. Namely, if standard mRNA vaccination does not lead to sufficient sAg expression, alternative approaches such as higher‐dose mRNA vaccines or a shift to other technologies such as protein‐based platforms may be preferable.

To study the relationship between sAg and antibody response after mRNA vaccination in immunocompromised individuals, we used an ultra‐sensitive assay to test plasma before and after third vaccine doses administered within a clinical trial of KTRs and measured associations with subsequent antibody response.

2. METHODS

2.1. Study population and sample acquisition

Study participants included kidney transplant recipients (KTRs) with poor response to two‐dose mRNA vaccination (<50 U/mL on Roche Elecsys anti‐SARS‐CoV‐2 S immunoassay) enrolled in the single‐arm, open‐label coronavirus disease 2019 (COVID‐19) Protection After Transplant Pilot trial funded by the National Institutes of Health (NCT04969263) and approved by the Johns Hopkins University IRB (IRB00288774). ⁷ Blood samples were obtained on the day of vaccination (day 0; pre third dose), and days 3, 14, and 30 days after vaccination. Plasma was isolated by Ficoll centrifugation and samples were stored at −80°C until utilized. All participants were tested with SARS‐CoV‐2 nasal swab polymerase chain reaction (PCR) at days 0 and 30 and with symptom screening by the study team.

2.2. SARS‐CoV‐2 antigen assays

Plasma from study participants at each time point was tested for spike (S1 subunit) and nucleocapsid (N) antigens as previously described. ⁶ , ⁸ Briefly, samples were thawed and spun to remove cellular debris and diluted with a solution of 10 mM dithiothreitol (DTT), protease inhibitors, and incubated at 37°C allowing the DTT to denature any antibodies bound to circulating antigens. Using an HD‐X Analyzer (Quanterix Corporation), 100 µL of diluted plasma was incubated with beads coated with anti‐SARS‐CoV‐2 antibodies. After washing, the beads were resuspended in the detector antibody solution, incubated, and washed. The beads were then resuspended in streptavidin conjugated β‐galactosidase solution, incubated, and washed before being resuspended in resorufin β‐D‐galactopyranoside and loaded into a microwell array for imaging. S1 antigen >9.6 pg/mL was considered positive and N‐antigen greater than 1.6 pg/mL was considered positive. ⁵ , ⁶ , ⁸

2.3. Anti‐SARS‐CoV‐2 receptor binding domain antibody measurement

Anti‐receptor binding domain (anti‐RBD) was measured using the semiquantitative Roche Elecsys anti‐SARS‐CoV‐2 S immunoassay on days 0, 14, and 30 as previously described ⁷ . Anti‐RBD in U/mL correlates ∼1:1 with World Health Organization (WHO) binding antibody units. Per the manufacturer, <0.8 U/mL was reported as negative (lower limit of quantification 0.4 U/mL). ⁵ , ⁶ , ⁸

2.4. Anti‐N antibody measurement

Anti‐N antibodies were measured using the qualitative Roche Elecsys Anti‐SARS‐CoV‐2 assay (cutoff of ≥1.0 deemed positive) and Meso Scale Diagnostics (MSD) electrochemiluminescent assay (Coronavirus Panel 3) according to the manufacturer's instructions. For the MSD assay, values greater than 12.3 WHO units were considered positive per the manufacturer.

2.5. Statistical analysis

Statistical analysis was conducted using Stata (version 15) and RStudio. Descriptive statistics are reported as counts (percentages) or medians (interquartile ranges [IQRs]), as appropriate. Continuous values were compared using the Wilcoxon rank‐sum test and Pearson's chi‐squared test for binary and categorical variables. A two‐sided p‐value of <.05 was considered significant. All figures were created in RStudio using ggplot2, dplyr, tidyverse, hrbrthemes, viridis, and scales packages.

3. RESULTS

3.1. Cohort characteristics

Of 81 participants, 16 were excluded due to prior confirmed clinical infection or receipt of monoclonal antibody (n = 3), baseline positive anti‐N antibody (n = 5), baseline positive plasma anti‐N antigen (n = 4), or missing timepoint data (n = 4) (Supplemental Figure 1). The included cohort was primarily male with a median (IQR) age 66 (56–73) years. Triple immunosuppression consisting of a calcineurin inhibitor, an antimetabolite, and steroids was common (74%) (Table 1). Among 65 included participants, 16 (25%) had S1 antigen detected at least once post‐vaccination (sAg(+)), and 49 (75%) had no detectable sAg at any post‐vaccination time point (sAg(‐)). There were no statistically significant differences in clinical, demographic, or transplant factors between these groups, though there were numerically fewer males, lower absolute lymphocyte counts, and less use of triple immunosuppression in the sAg(+) group (Table 1). Specifically, there was no association between the use of mycophenolate and sAg positivity (p = .72).

TABLE 1.

Baseline characteristics of participants by S1 antigen status by day 14 post‐intervention.^*

	S1 Antigen Status (sAg)
	sAg (+) a	sAg (−)
Characteristics	N = 16	N = 49	p‐Value
Male sex	8 (50)	35 (71)	0.12
Age (years), median (IQR)	63 (45–71)	67 (57–74)	0.23
Time since Transplant (years), median (IQR)	7 (3–10)	5 (2–9)	0.47
Absolute Lymphocyte Count (cells/mL), median (IQR)	0.78 (0.63–1.18)	1.09 (0.66–1.57)	0.14
Type of Vaccine Administrated
Moderna (mRNA‐1273)	6 (38)	13 (27)	0.40
Pfizer‐BioNTech (BNT162b2)	10 (62)	36 (73)
Total IgG (mg/dL), median (IQR)	801 (690–924)	833 (712–1057)	0.72
Immunosuppressants
Calcineurin Inhibitor (Tacrolimus/Cyclosporine)	16 (100)	48 (98)	0.56
Antimetabolites (MMF/MPA, Azathioprine)	14 (88)	40 (82)	0.59
Steroids	13 (81)	47 (96)	0.06
Triple Immunosuppression	11 (69)	37 (76)	0.06
Quantitative Anti‐RBD Antibody Levels (U/mL)
Day 0, median (IQR)	<0.4 (<0.4–0.6)	2.5 (<0.4–9.3)	0.01
Day 14, median (IQR)	<0.4 (<0.4–194.6)	823.5 (7.1–1970.5)	0.01
Day 30, median (IQR)	10.8 (<0.4–338.3)	709.5 (10.5–2309.5)	0.01
Negative Anti‐RBD levels (< 0.4 U/mL)
Day 0	12 (75)	19 (39)	0.01
Day 14	8 (50)	9 (18)	0.03
Day 30	9 (56)	9 (18)	<0.01
Duration between 2nd & 3rd vaccine (days), median (IQR)	159 (139–173)	168 (150–182)	0.13

^† Data are presented as median (IQR) for continuous measures, and n (%) for categorical measures.

^‡ S1 antigen greater than 9.6 at any time point was considered detectable (sAg(+)).

^§ Statistical tests were conducted using Chi‐square for binary/categorical variables & Wilcoxon rank‐sum for continuous variables.

3.2. Spike antigen results

For the sAg(+) participants, sAg levels were highest at day 3 post‐vaccination (median [IQR] = 61.2 pg/mL [41.5–172.4 pg/mL]) (Figure 1A) and became undetectable by day 14 in all but 3/16 (19%) participants. This included two participants with detectable sAg at day 0, which remained detectable through day 14. Neither participant reported recent or intercurrent infection with SARS‐CoV‐2 (and both remained N antigen and N antibody negative throughout). These participants received third doses of vaccine 34 and 159 days after the second doses, respectively.

FIGURE 1.

(A) Spike (S1) antigen over time after third doses of messenger RNA (mRNA) severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) vaccine in kidney transplant recipients. S1 antigen levels in pg/mL (y‐axis) are displayed immediately prior to (Day 0), three days after (Day 3), and 14 days after (Day 14) a third vaccine dose. Individuals’ antigen trajectories are represented by lines. A horizontal dashed line at 9.6 pg/mL represents the lower limit of detection of the assay. S1 antigen was undetectable in 49 (75%) of participants at all three time points. (B) Box and whisker plot of anti‐SARS‐CoV‐2 spike antibody titers (U/mL) thirty days after third vaccine doses compared between sAg(+) (detected at any time) (Left, Blue) and sAg(‐) (Right, Red). Individual antibody titers are represented by black dots overlayed on the boxplots. The difference in antibody titer between groups was compared using a Wilcoxon‐Mann‐Whitney test.

3.3. Association of sAg detection and anti‐spike antibody

Anti‐spike antibodies were significantly lower at day 30 post‐third dose in the sAg(+) group versus the sAg(‐) group (10.8 U/mL [IQR: <0.4–338.3] vs. 709.5 U/mL [IQR:10.5–2309.5], p = .01) (Figure 1B). Furthermore, the sAg(+) group had significantly lower anti‐spike titers on Day 0 and Day 14 (Table 1). sAg(+) status was also associated with a greater frequency of negative anti‐spike antibody post‐third vaccination (50% vs. 18% at day 14, p = .03) (Table 1). The two participants who were sAg(+) at baseline had antibody levels at day 30 of 6490 U/mL and <0.4 U/mL (negative). Inferences were similar on sensitivity analysis after the exclusion of the two participants with baseline sAg(+). In those with detectable day 30 anti‐spike Ab and detectable sAg (n = 7), there was no evident relationship with sAg level (spearman rank p = −.06; p = .64) (Supporting Information Table).

4. DISCUSSION

In this first study of plasma sAg in KTRs post SARS‐CoV‐2 mRNA vaccination we demonstrated that the presence of sAg was, paradoxically, associated with lower spike antibody response to third mRNA vaccine doses. These findings indicate that lack of translation and expression of sAg after mRNA vaccination is not likely the primary mechanism of ongoing poor vaccine response in this immunosuppressed group. Rather, sAg(‐) individuals showed much higher antibody responses to a third dose of vaccine, suggestive of boosting of low‐level preformed immune response against SARS‐CoV‐2 that likely also contributed to rapid antigen recognition and clearance (i.e., in < 3 days). This study and hypothesis are also consistent with a prior report that sAg is rarely detectable after a 2nd dose of vaccine in healthy individuals, who routinely develop high levels of antibodies. ⁵

Although this study was designed to observe the isolated association of third mRNA vaccination on plasma sAg generation, we did, interestingly, observe baseline high sAg(+) in two KTRs. These participants had no clinical or serological evidence of prior or incident COVID‐19, thus this finding might be due to uncleared sAg produced via primary vaccination, which has occasionally been detected out to 29 days in other populations. ⁵ Otherwise, it remains possible that persistent sAg(+) represents evidence of persistent subclinical SARS‐CoV‐2 infection, though a lack of positive plasma N antigen or nasal swabs PCR in either individual at any timepoint is less suggestive. ⁹ , ¹⁰ Alternatively, assay cross‐reactivity could lead to false positive sAg results, suggested to be ∼1% in other studies, ¹¹ though serial consistent false positive assays in these participants appear less likely. Regardless, in sensitivity analyses excluding these two individuals, the inference that sAg(+) is associated with worse antibody production remained robust.

This study was specifically designed to study KTRs with documented poor antibody response to the original two‐dose series, which may limit generalizability to other immunosuppressed groups. This phenotype, however, is common in KTRs among whom nearly 30% may remain anti‐spike seronegative following three mRNA vaccine doses. ¹² Furthermore, we limited our assessment to correlations between sAg detection and spike antibody production, and cannot comment upon the generation of other cellular immune responses, though cellular and humoral responses are often positively correlated. ⁷ Although we suspect pre‐formed adaptive responses are responsible for sAg clearance, we did not explicitly examine this mechanism or assess other factors such as activation of the innate immune system that may have contributed to clearance. Finally, we restricted the analysis to those without a history of prior COVID‐19 by clinical report or by anti‐N antibody, though operating characteristics may be imperfect in this heavily immunosuppressed cohort. We did, however, augment this with N antigen testing to surveil for subclinical infection peri‐vaccination.

This study clarifies that lack of plasma sAg exposure is not likely a primary mechanism of decreased humoral response to mRNA‐based SARS‐CoV‐2 vaccines in KTRs. This result, in part, supports why higher‐dose strategies have not improved antibody levels in some persistent poor responders. ¹³ Therefore, future efforts toward improving immunogenicity among transplant recipients should focus on alternative strategies such as novel vaccine platforms or modulation of immunosuppression.

CONFLICT OF INTEREST STATEMENT

Dr. Segev reports receiving consulting and/or speaking honoraria from Sanofi, Novartis, Veloxis, Mallinckrodt, Jazz Pharmaceuticals, CSL Behring, Thermo Fisher Scientific, Caredx, Transmedics, Kamada, MediGO, Regeneron, AstraZeneca, Takeda/Shire, Novavax, and Bridge to Life. Dr. Werbel has received consulting and/or speaking fees from CDC/IDSA COVID‐19 Real‐Time Learning Network, AstraZeneca, GlobalData, China Medical Tribune (CME), Medical Learning Institute (CME), and advisory board fees from AstraZeneca and Novavax. Dr. Karaba has received consulting fees from Hologic, Inc. and speaking fees from PRIME Education (CME). Dr. Walt has a financial interest in Quanterix Corporation, a company that develops an ultra‐sensitive digital immunoassay platform. He is an inventor of the Simoa technology, a founder of the company, and also serves on its Board of Directors. Dr. Walt's interests were reviewed and managed by Mass General Brigham and Harvard University in accordance with their conflict‐of‐interest policies. The remaining authors of this manuscript have no relevant financial disclosures or conflicts of interest related to this study.

Supporting information

Supporting information

Karaba AH, Swank Z, Hussain S, et al. Detectable plasma severe acute respiratory syndrome coronavirus 2 spike antigen is associated with poor antibody response following third messenger RNA vaccination in kidney transplant recipients. Transpl Infect Dis. 2024;26:00:e14281. 10.1111/tid.14281

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.