Abstract
Background
A severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) mRNA vaccine booster elicits sufficient antibody responses that protect against coronavirus disease 2019, whereas adverse reactions such as fever have been commonly reported. Associations between adverse reactions and antibody responses have not been fully characterized, nor has the influence of antipyretic use.
Methods
This is a prospective observational cohort study in Japan, following our prior investigation of BNT162b2 2-dose primary series. Spike-specific immunoglobulin G (IgG) titers were measured for SARS-CoV-2–naive hospital healthcare workers who received a BNT162b2 booster. The severity of solicited adverse reactions, including the highest body temperature, and self-medicated antipyretics were reported daily for 7 days following vaccination through a web-based self-reporting diary.
Results
The data of 281 healthcare workers were available. Multivariate analysis extracted fever after the booster dose (β = .305, P < .001) as being significantly correlated with the specific IgG titers. The analysis of 164 participants with data from the primary series showed that fever after the second dose was associated with the emergence of fever after the booster dose (relative risk, 3.97 [95% confidence interval, 2.48–6.35]); however, the IgG titers after the booster dose were not associated with the presence or degree of fever after the second dose. There were no significant differences in the IgG titers by the use, type, or dosage of antipyretic medication.
Conclusions
These results suggest an independent correlation between mRNA vaccine–induced specific IgG levels and post–booster vaccination fever, without any significant influence of fever after the primary series. Antipyretic medications for adverse reactions should not interfere with the elevation of specific IgG titers.
Keywords: antipyretic, antibody, reactogenicity, SARS-CoV-2, vaccine
Spike-specific IgG titers after a BNT162b2 booster were measured for healthcare workers. Adverse reactions and self-medicated antipyretics were reported. Post–booster vaccination fever was correlated with the specific IgG titers. Antipyretics used for adverse reactions did not suppress specific IgG induction.
Administration of a messenger RNA (mRNA) coronavirus disease 2019 (COVID-19) vaccine has shown high vaccine efficacy, substantially reducing the risk of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and the development of severe or critical disease [1–4]. These efficacy data are consistent with evidence from immunogenicity studies that show robust specific antibody responses to the mRNA COVID-19 vaccines [5, 6]. It is notable that the specific immunoglobulin G (IgG) and neutralization titers of vaccinees who received an mRNA COVID-19 vaccine exceeded those of recovered COVID-19 patients [7, 8]. On the other hand, the reactogenicity of mRNA COVID-19 vaccines is known to be relatively high, and fever is a common adverse reaction. Around 15% of vaccinees had fever after the second or third dose of an mRNA COVID-19 vaccine [1, 9, 10], while only 1%–2% did so after influenza or pneumococcal vaccination [11, 12]. The possible relation between adverse reactions, including fever, and the antibody responses to mRNA COVID-19 vaccines remains to be fully elucidated. Antipyretic or pain medications (antipyretics) are often used to mitigate the frequent adverse events. Public health authorities allow the use of antipyretics in response to adverse events [13, 14], but data for the possible influence of their use on the antibody responses to COVID-19 vaccines are insufficient.
We previously investigated the correlation of adverse reactions with the specific antibody responses to the 2-dose primary series of BNT162b2 vaccine (Pfizer/BioNTech). The influence of antipyretic use on antibody responses was also investigated. A positive correlation of degree of fever after the second dose and little interference from the antipyretic medications on the antibody titers were shown [15]. In the present study of the same cohort, we prospectively investigated the association of adverse reactions, which were evaluated using a standardized assessment tool, and the use of antipyretics with the specific antibody responses to a booster dose of BNT162b2 vaccine. In addition, using the data of the participants for whom information on both the second and booster doses was available, we also evaluated whether the specific antibody titers after the booster dose are affected by fever after the second dose.
METHODS
Participants
Eligible participants were healthcare workers who received three 30-µg doses of BNT162b2 at Fukuoka City Hospital in Japan. The primary 2-dose vaccine series with a 21-day interval was administered between March and June 2021, and the booster dose was given between December 2021 and January 2022. Included in the analysis were vaccinees who had serum sampling done ≥14 days after the booster dose and who completely responded to questionnaires about their background and solicited adverse reactions. The exclusion criteria were (1) previous laboratory-confirmed COVID-19 diagnosis, (2) positive results for antibodies targeting the viral nucleocapsid protein [IgG(N)], (3) the use of an antipyretic within 24 hours before the booster dose, and (4) receipt of immunosuppressive therapy.
Participant Consent Statement
All participants provided written informed consent before undergoing any of the study procedures. The study was approved by the ethics review board of Fukuoka City Hospital (approval number 228) and registered in the University Hospital Medical Information Network Clinical Trials Registry (registration number UMIN000046246).
Demographic Characteristics, Reactogenicity, and Antipyretic Medications
Participant background information was collected by a web-based questionnaire. Local and systemic adverse reactions were reported daily for 7 days after the booster dose through a web-based self-reporting diary. The solicited data were as follows: (1) local reactions (pain at the injection site, redness, and swelling) and (2) systemic events (fever, fatigue, headache, chills, vomiting, diarrhea, muscle pain, joint pain, and lymphadenopathy). Axillary body temperature was measured twice daily, morning and night, and whenever the participant felt feverish. The highest body temperature during the 7 days was used in the analysis. All solicited reactions except lymphadenopathy were recorded based on standardized assessment scales developed by the US Food and Drug Administration guidelines [16]. Lymphadenopathy was evaluated for its presence or absence. The use of an antipyretic was left up to the participant. Information on the self-medicated antipyretics, including name, dosage, timing, and reason for use, was collected daily with the solicited adverse reactions for the 7 days after the booster dose.
Previously collected data on the receptor-binding domain of the S1 subunit of the viral spike protein (S-RBD) IgG titers and adverse reactions to the second dose were used in the present study. The major differences in the method of data collection were that in the earlier studies the adverse reaction information was collected for 5 days, not 7, after each of the doses and that we had used an originally defined subjective scaling method, except for fever. The detailed methods are shown in our previous study [15].
Serological Testing
Serum samples were collected twice, before and after the booster dose. The interval between the booster dose and the sampling after vaccination was scheduled at approximately 1 month to match it with the interval between the second dose and the sampling [15]. The quantitative levels of IgG(S-RBD) and IgG(N) were measured using the SARS-CoV-2 IgG II assay and SARS-CoV-2 IgG assay, respectively (Abbott Laboratories, Abbott Park, Illinois). Signal-to-cutoff values of ≥1.4 AU/mL were applied for IgG(N) positivity [17].
Statistical Analysis
The IgG(S-RBD) titers were log-transformed for analysis. The median with interquartile range (IQR), geometric mean titer (GMT), fold change, 95% confidence interval (CI), and relative risk (RR) were calculated. Between-group differences were calculated with Student t test, analysis of variance, or post hoc Dunnett test in line with suitability. Correlation coefficients were calculated using Spearman correlation coefficient. Multivariate linear regression models with a stepwise selection procedure were established. Multicollinearity among variables was examined using variance inflation factors. The level of significance was set at <5%, 2-sided. All analyses were performed using the SAS software package, release 9.4 (SAS Institute, Cary, North Carolina).
RESULTS
Demographic Characteristics
Among 419 staff members who received the BNT162b2 booster, serum samples were collected from 346, and 316 satisfied the inclusion criteria. Of these, 13 were excluded due to a history of COVID-19 or IgG(N) ≥1.4 AU/mL, 20 due to prevaccination use of antipyretics, and 2 due to having received immunosuppressive therapy, leaving the data of 281 participants available for analysis. Demographic and background information are summarized in Table 1. The median age was 41 years (IQR, 33–50 years), 72.6% were female, all were Japanese, and 77.2% had no coexisting conditions. The median interval between the second and booster doses was 262 days (IQR, 260–264 days; range, 219–288 days). Serum samples before and after the booster dose were obtained approximately 8 months after the second dose (median, 247 days [IQR, 244–252 days]) and 1 month after the booster dose (median, 32 days [IQR, 29–33 days]), respectively.
Table 1.
Demographic Characteristics, Spike Receptor-Binding Domain Immunoglobulin G Titer, and Fold Change
| Characteristic | No. (%) | IgG(S-RBD) Titers Before Dose 3 | IgG(S-RBD) Titers After Dose 3 | Change in Titers After Dose 3 | |||
|---|---|---|---|---|---|---|---|
| GMT (95% CI), AU/mL | P Value | GMT (95% CI), AU/mL | P Value | Fold Change, Mean (95% CI) | P Value | ||
| All eligible participants | 281 | 573 (528–621) | 16 707 (15 403–18 122) | 29.4 (26.7–32.5) | |||
| Sex | |||||||
| Female | 204 (72.6) | 602 (546–665) | .047 | 16 110 (14 743–17 604) | .197 | 26.7 (24.4–29.3) | <.001 |
| Male | 77 (27.4) | 502 (436–577) | 18 395 (15 325–22 085) | 36.7 (31.1–43.3) | |||
| Age, y | |||||||
| Median (IQR) | 41 (33–50) | r = –0.215a | <.001 | r = 0.009a | .877 | r = 0.231a | <.001 |
| <40 | 127 (45.2) | 648 (582–723) | .014 | 16 970 (15 199–18 948) | .509 | 26.2 (23.4–29.3) | .006 |
| 40–54 | 118 (42.0) | 534 (472–603) | 15 937 (14 055–18 072) | 29.9 (26.2–34.0) | |||
| ≥55 | 36 (12.8) | 468 (345–634) | 18 453 (13 623–24 995) | 39.5 (30.7–50.7) | |||
| Smoking | |||||||
| Never | 234 (83.3) | 602 (549–659) | .030 | 17 249 (15 747–18 894) | .134 | 28.7 (26.2–31.4) | .473 |
| Ex-smoker | 26 (9.3) | 457 (385–543) | 15 627 (12 922–18 899) | 34.2 (27.8–42.1) | |||
| Current smoker | 21 (7.5) | 441 (321–605) | 12 713 (9214–17 540) | 28.8 (21.3–39.1) | |||
| Alcohol use | |||||||
| None | 113 (40.2) | 598 (523–684) | .493 | 17 781 (15 604–20 261) | .465 | 29.7 (26.4–33.4) | .698 |
| Sometimes | 136 (48.4) | 569 (510–634) | 16 048 (14 358–17 937) | 28.2 (25.0–31.8) | |||
| Almost every day | 32 (11.4) | 508 (378–682) | 15 907 (11 890–21 282) | 31.3 (23.0–42.5) | |||
| BMI, kg/m2 | |||||||
| Median (IQR) | 21.2 (19.7–23.3) | r = –0.032a | .599 | r = .091a | .130 | r = .137a | .022 |
| <18.5 | 26 (9.3) | 531 (373–758) | .417 | 15 055 (11 162–20 307) | .545 | 27.2 (20.7–35.7) | .079 |
| 18.5–24.9 | 220 (78.6) | 589 (539–645) | 16 666 (15 268–18 192) | 28.3 (25.8–31.0) | |||
| ≥25.0 | 34 (12.1) | 506 (408–682) | 18 332 (15 695–24 540) | 39.9 (32.2–44.5) | |||
| Job category | |||||||
| Nurse | 140 (49.8) | 575 (516–641) | .483 | 15 704 (14 270–17 283) | .488 | 27.3 (24.7–31.2) | .061 |
| Clerk | 40 (14.2) | 648 (507–828) | 15 731 (12 771–19 378) | 24.3 (18.8–31.3) | |||
| Doctor | 30 (10.7) | 467 (362–602) | 17 364 (12 091–24 936) | 37.2 (27.3–50.7) | |||
| Radiologist | 13 (4.6) | 605 (438–837) | 19 694 (12 239–31 690) | 32.5 (17.7–59.8) | |||
| Pharmacist | 12 (4.3) | 652 (389–1093) | 19 696 (14 270–17 283) | 30.2 (18.3–49.9) | |||
| Other | 46 (16.4) | 555 (443–694) | 18 953 (15 130–23 742) | 34.2 (28.4–41.1) | |||
| Comorbidities | |||||||
| Allergic rhinitis | |||||||
| Yes | 43 (15.3) | 537 (448–645) | .508 | 14 801 (11 989–18 277) | .213 | 27.6 (23.2–32.7) | .563 |
| No | 238 (84.7) | 580 (530–635) | 16 963 (15 513–18 548) | 29.4 (26.7–32.5) | |||
| Dislipidemia | |||||||
| Yes | 17 (6.1) | 656 (464–929) | .405 | 20 333 (12 526–33 007) | .381 | 31.0 (21.7–44.3) | .710 |
| No | 264 (94.0) | 568 (522–618) | 16 497 (15 205–17 898) | 29.0 (26.7–31.6) | |||
| Hypertension | |||||||
| Yes | 14 (5.0) | 455 (262–790) | .364 | 12 314 (6393–23 719) | .313 | 27.1 (18.0–40.7) | .683 |
| No | 267 (95.0) | 580 (535–629) | 16 975 (15 678–18 382) | 29.3 (26.9–31.8) | |||
| Asthma | |||||||
| Yes | 5 (1.8) | 798 (470–1352) | .281 | 21 154 (11 389–39 292) | .442 | 26.5 (15.4–45.8) | .759 |
| No | 276 (98.2) | 570 (525–618) | 16 634 (15 321–18 063) | 29.2 (26.9–31.7) | |||
| Diabetes | |||||||
| Yes | 3 (1.1) | 439 (96–2004) | .504 | 41 850 (14 723–118 987) | .021 | 95.3 (56.1–162.0) | .003 |
| No | 278 (98.9) | 575 (529–624) | 16 542 (15 251–17 943) | 28.8 (26.5–31.2) | |||
| Heart disease | |||||||
| Yes | 3 (1.1) | 539 (324–894) | .876 | 33 411 (3446–323 966) | .081 | 62.0 (6.1–632.8) | .058 |
| No | 278 (98.9) | 573 (528–622) | 16 581 (15 290–17 985) | 28.9 (26.7–31.4) | |||
| Malignancy | |||||||
| Yes | 3 (1.1) | 460 (45–4648) | .581 | 19 838 (1219–322 849) | .666 | 43.1 (3.7–497.3) | .327 |
| No | 278 (98.9) | 574 (529–623) | 16 676 (15 374–18 088) | 29.0 (26.8–31.5) | |||
| Chronic kidney disease | |||||||
| Yes | 1 (0.4) | 626 (NA) | NA | 32 300 (NA) | NA | 51.6 (NA) | NA |
| No | 280 (99.4) | 573 (528–621) | 16 669 (15 364–18 080) | 29.1 (26.8–31.6) | |||
Abbreviations: BMI, body mass index; CI, confidence interval; GMT, geometric mean titer; IgG(S-RBD), immunoglobulin G spike receptor-binding domain; IQR, interquartile range; NA, not available.
r values refer to the Spearman correlation coefficient.
IgG(S-RBD) Titer and Fold Change by Demographic Characteristics
The booster dose increased the GMT of IgG(S-RBD) 29.4-fold (95% CI, 26.7–32.5), from 573 AU/mL (95% CI, 528–621) to 16 707 AU/mL (95% CI, 15 403–18 122). The IgG(S-RBD) titers and fold changes according to demographic characteristics are shown in Table 1. In the analysis of the IgG(S-RBD) titers, sex and age showed no significant correlation (P = .197 and P = .645, respectively). In contrast, both were significantly correlated with the fold changes. The mean fold change in males was higher than that in females (36.7-fold [95% CI, 31.1–43.3] vs 26.7-fold [95% CI, 24.4–29.3]; P < .001). Age showed a positive correlation with the fold changes (r = 0.244, P < .001). Comorbidities other than diabetes were not correlated. Diabetes had statistically significant correlations with both the IgG(S-RBD) titers and fold changes, but it was not further analyzed due to the small number of participants with diabetes.
IgG(S-RBD) Titer and Fold Change by Adverse Reaction
The IgG(S-RBD) titers and fold changes after the booster dose by the solicited adverse reactions are shown in Table 2. None of the local reactions had a significant correlation with the IgG(S-RBD) titers. Among the systemic reactions, fever showed a positive correlation with the IgG(S-RBD) titers (r = 0.262, P < .001). Fatigue (P = .022), headache (P = .024), chills (P = .001), and lymphadenopathy (P < .001) were also positively correlated. Neither local nor systemic adverse reactions significantly affected the fold changes in IgG(S-RBD) titers.
Table 2.
Influence of Adverse Reaction Variables on Spike Receptor-Binding Domain Immunoglobulin G Titer
| Reaction | Variable | No. (%) | GMT (95% CI), AU/mL | P Value | Fold Change, Mean (95% CI) | P Value |
|---|---|---|---|---|---|---|
| Local reactions | ||||||
| Pain at injection site | No | 3 (1.1) | 14 818 (4032–54 450) | .763 | 23.9 (10.9–52.4) | .618 |
| Yes | 278 (98.9) | 16 730 (15 413–18 155) | 29.2 (26.9–31.7) | |||
| Redness | No | 185 (65.8) | 16 761 (15 198–18 484) | .915 | 28.9 (26.2–31.9) | .734 |
| Yes | 96 (34.2) | 16 604 (14 328–19 244) | 29.7 (25.7–34.4) | |||
| Swelling | No | 148 (52.7) | 16 707 (14 866–18 772) | .998 | 30.5 (27.4–33.8) | .269 |
| Yes | 133 (47.3) | 16 707 (14 907–18 728) | 27.8 (24.5–31.6) | |||
| Systemic reactions | ||||||
| Fever | Absolute value | 281 (100) | r = 0.262a | <.001 | r = 0.082a | .170 |
| <37.0°C | 114 (40.6) | 14 011 (12 431–15 793) | <.001 | 26.9 (23.7–30.5) | .100 | |
| 37.0°C–37.9°C | 97 (34.5) | 16 302 (14 174–18 750) | 28.8 (25.3–32.9) | |||
| ≥38.0°C | 64 (24.9) | 23 020 (19 670–26 941) | 33.7 (28.1–40.5) | |||
| Fatigue | No | 42 (15.0) | 13 329 (10 713–16 581) | .022 | 26.2 (20.8–32.9) | .278 |
| Yes | 239 (85.1) | 17 382 (15 933–18 967) | 29.7 (27.2–32.4) | |||
| Headache | No | 94 (33.5) | 14 652 (12 885–16 661) | .024 | 29.3 (25.6–33.5) | .947 |
| Yes | 187 (66.6) | 17 848 (16 099–19 783) | 29.1 (26.3–32.3) | |||
| Chills | No | 135 (48.0) | 14 534 (13 077–16 151) | .001 | 27.3 (24.3–30.7) | .126 |
| Yes | 146 (52.0) | 19 006 (16 862–21 419) | 31.0 (27.6–34.8) | |||
| Vomiting | No | 272 (96.8) | 16 512 (15 198–17 943) | .121 | 29.1 (26.7–31.6) | .721 |
| Yes | 9 (3.2) | 23 752 (16 021–35 221) | 31.6 (22.5–44.5) | |||
| Diarrhea | No | 246 (87.5) | 16 811 (15 385–18 365) | .693 | 29.3 (26.8–32.0) | .812 |
| Yes | 35 (12.5) | 15 999 (13 005–19 683) | 28.4 (23.6–34.2) | |||
| Muscle pain | No | 40 (14.2) | 17 434 (14 142–21 493) | .674 | 28.9 (22.2–37.7) | .944 |
| Yes | 241 (85.8) | 16 588 (15 181–18 126) | 29.2 (26.8–31.8) | |||
| Joint pain | No | 128 (45.6) | 16 044 (14 315–17 980) | .370 | 27.9 (24.6–31.6) | .335 |
| Yes | 153 (54.5) | 17 282 (15 396–19 404) | 30.2 (27.1–33.7) | |||
| Lymphadenopathy | No | 194 (69.0) | 15 332 (13 845–16 975) | <.001 | 29.1 (26.1–32.4) | .956 |
| Yes | 87 (31.0) | 20 234 (17 857–22 935) | 29.2 (26.2–32.7) |
Abbreviations: CI, confidence interval; GMT, geometric mean titer.
r values refer to the Spearman correlation coefficient.
Factors that showed a P value of <.2 in the univariate analyses, including sex, body mass index, smoking history, heart disease, fever, fatigue, headache, chills, vomiting, and lymphadenopathy, were incorporated in the multivariate analysis. Among these variables, only fever (standardized regression coefficient: β = .305 [95% CI, .193–.417]; P < .001) was extracted as being independently correlated with the IgG(S-RBD) titers (adjusted R2 = 0.090). A similar analysis for the fold change in IgG(S-RBD) titers extracted age (β = .248 [95% CI, .136–.360]; P < .001), male sex (β = .183 [95% CI, .071–.294]; P = .001), and fever (β = .136 [95% CI, .024–.248]; P = .018) as significant (adjusted R2 = 0.105). All variance inflation values were <5 in the linear regression models, indicating the absence of multicollinearity among the factors.
Influence of Antipyretic Medications on IgG(S-RBD) Titer and Fold Change
The analyses of the influence of antipyretics on the IgG(S-RBD) titers and fold changes are shown in Table 3. In total, 119 (42.4%) participants used antipyretics during the 7 days after the booster dose. None of the participants prophylactically used an antipyretic after the vaccination to prevent adverse events. The GMT of IgG(S-RBD) was comparable between the groups with and without antipyretic use (17 466 AU/mL [95% CI, 15 279–19 966] vs 16 170 AU/mL [95% CI, 14 601–17 906]; P = .357). Similarly, there was no significant difference in the mean fold change between the 2 groups (29.4-fold [95% CI, 26.1–33.2] vs 28.9-fold [95% CI, 25.9–32.4]; P = .893). No significant influence of antipyretic use was found when stratified by fever grade, classified into <37.0°C, 37.0°C–37.9°C, and ≥38.0°C. The most commonly used antipyretic combination was acetaminophen monotherapy (43/119 [36.1%]), followed by loxoprofen monotherapy (35/119 [29.4%]) and ibuprofen monotherapy (14/119 [11.8%]). The IgG(S-RBD) titers and fold changes of the group with nonsteroidal anti-inflammatory drugs (NSAIDs) such as loxoprofen and ibuprofen were comparable to those of acetaminophen. The IgG(S-RBD) titers by all combinations of antipyretics used are summarized in Supplementary Table 1.
Table 3.
Influence of Antipyretics on Spike Receptor-Binding Domain Immunoglobulin G Titer and Fold Change After a Booster Dose
| Characteristic | No. (%) | GMT (95% CI) | P Value | Fold Change, Mean (95% CI) |
P Value | |
|---|---|---|---|---|---|---|
| Use of antipyretic medications | ||||||
| All participants | No | 162 (57.7) | 16 170 (14 601–17 906) | .357 | 28.9 (25.9–32.4) | .893 |
| Yes | 119 (42.4) | 17 466 (15 279–19 966) | 29.4 (26.1–33.2) | |||
| By fever grade | ||||||
| <37.0°C | No | 85 (74.6) | 14 217 (12 491–16 181) | .681 | 27.8 (23.8–32.4) | .400 |
| Yes | 29 (25.4) | 13 425 (10 006–18 009) | 24.6 (19.9–30.3) | |||
| 37.0°C–37.9°C | No | 50 (51.6) | 16 489 (13 602–19 985) | .869 | 30.0 (24.7–36.4) | .556 |
| Yes | 47 (48.5) | 16 106 (13 041–19 893) | 27.7 (23.2–33.2) | |||
| ≥38.0°C | No | 27 (38.6) | 23 388 (17 914–30 542) | .874 | 31.0 (22.4–42.8) | .465 |
| Yes | 43 (61.4) | 22 793 (18 612–27 906) | 35.6 (28.3–44.6) | |||
| Type of antipyretic | ||||||
| Only acetaminophen | … | 43 (36.1) | 17 132 (13 886–21 136) | Reference | 26.3 (21.3–32.5) | Reference |
| Only loxoprofen | … | 35 (29.4) | 14 820 (11 167–19 669) | .744 | 25.4 (2.6–31.2) | .991 |
| Only ibuprofen | … | 14 (11.8) | 20 306 (13 383–30 811) | .815 | 41.2 (28.5–59.7) | .068 |
| Othera | … | 27 (22.7) | 20 613 (15 651–27 149) | .637 | 35.9 (28.3–45.6) | .135 |
| Total dosage of antipyretic during 7-days after vaccination | ||||||
| Acetaminophen, mg, median (IQR) | 900 (400–1200) | r = 0.072b | .648 | r = .247b | .111 | |
| Loxoprofen, mg, median (IQR) | 120 (60–180) | r = 0.097b | .580 | r = .223b | .198 | |
| Ibuprofen, mg, median (IQR) | 200 (150–400) | r = –0.346b | .226 | r = –.201b | .491 | |
| Timing of antipyretic use | ||||||
| Antipyretic use on the day of vaccination | No | 108 (90.8) | 17 151 (14 856–19 797) | .398 | 30.0 (26.4–34.1) | .308 |
| Yes | 11 (9.2) | 20 903 (14 378–30 395) | 24.3 (16.9–34.9) | |||
Abbreviations: CI, confidence interval; GMT, geometric mean titer; IQR, interquartile range.
Including other type of antipyretics, combination drugs, or combination of several antipyretics.
r values refer to the Spearman correlation coefficient.
Association Between the Presence or Absence of an Adverse Reaction After the Second Dose and the Emergence of the Same Reaction After the Booster Dose
Using the data of 164 participants for whom information on both the second and booster doses were available, the probability of a solicited adverse reaction after the booster dose was investigated by the presence or absence of the corresponding reaction after the second dose (Table 4). The RR of all solicited adverse reactions was >1.0. The presence of swelling at the injection site, fever of ≥38.0°C, headache, chills, muscle pain, and joint pain after the second dose showed a significant association with the emergence of the corresponding reaction after the booster dose. The RR was the highest for the fever, at 3.97 (95% CI, 2.48–6.35).
Table 4.
Probability of an Adverse Reaction After the Booster Dose According to the Presence or Absence of the Same Reaction After the Second Dose
| Adverse Reaction | Corresponding Reactions After Booster Dose | Incidence of a Corresponding Reaction After Booster Dose, % | RR (95% CI) | ||
|---|---|---|---|---|---|
| Present | Absent | ||||
| Local reactions after dose 2 | |||||
| Pain at injection site | Present | 154 | 1 | 99.4 | 1.12 (.89–1.41) |
| Absent | 8 | 1 | 88.9 | Reference | |
| Redness | Present | 14 | 16 | 46.7 | 1.56 (.98–2.48) |
| Absent | 40 | 94 | 29.9 | Reference | |
| Swelling | Present | 45 | 23 | 66.2 | 1.93 (1.39–2.66) |
| Absent | 33 | 63 | 34.4 | Reference | |
| Systemic reactions after dose 2 | |||||
| Fever ≥38.0°C | Present | 21 | 12 | 63.6 | 3.97 (2.48–6.35) |
| Absent | 21 | 110 | 16.0 | Reference | |
| Fatigue | Present | 129 | 14 | 90.2 | 1.35 (1.00–1.84) |
| Absent | 14 | 7 | 66.7 | Reference | |
| Headache | Present | 80 | 19 | 80.8 | 1.46 (1.15–1.85) |
| Absent | 36 | 29 | 55.4 | Reference | |
| Chills | Present | 52 | 18 | 74.3 | 2.00 (1.48–2.68) |
| Absent | 35 | 59 | 37.2 | Reference | |
| Diarrhea | Present | 3 | 17 | 15.0 | 1.14 (.37–3.50) |
| Absent | 19 | 125 | 13.2 | Reference | |
| Muscle pain | Present | 85 | 6 | 93.4 | 1.22 (1.06–1.40) |
| Absent | 56 | 17 | 76.7 | Reference | |
| Joint pain | Present | 59 | 21 | 73.8 | 1.77 (1.33–2.35) |
| Absent | 35 | 49 | 41.7 | Reference | |
Information on vomiting and lymphadenopathy after the second dose were not collected.
Abbreviations: CI, confidence interval; RR, relative risk.
IgG(S-RBD) Titer by the Presence of Fever After the Second and Booster Doses
The 164 participants were divided into 4 groups by the presence or absence of fever of ≥38.0°C after each dose. The IgG(S-RBD) titers of these 4 groups 1 month after the second dose, 8 months after the second dose, and 1 month after the booster dose are shown in Table 5. The GMTs of IgG(S-RBD) 1 month after the second dose were higher for the groups with fever after the second dose than for those without. Eight months after the second dose, the differences among the groups were subtle, ranging from 537 to 796 AU/mL, without significant differences. The GMTs of IgG(S-RBD) 1 month after the booster dose were higher in the groups with fever after the booster dose, and they were comparable between the groups with and without fever after the second dose. A multivariate analysis incorporating fever after the second dose was additionally done. Fever after the booster dose was extracted as being significantly correlated with the IgG(S-RBD) titers after the booster dose (β = .246 [95% CI, .097–.395]; P = .001), but fever after the second dose was not.
Table 5.
Spike Receptor-Binding Domain Immunoglobulin G Titers by the Presence or Absence of Fever After the Second and Booster Doses
| Fever ≥38.0°C After Vaccination (After Dose 2/Dose 3) | No. (%) | Female Sex, No. (%) | Age, y, Median (IQR) | 1 Month After Dose 2 | 8 Months After Dose 2 | 1 Month After Dose 3 | |||
|---|---|---|---|---|---|---|---|---|---|
| GMT (95% CI), AU/mL | P Value | GMT (95% CI), AU/mL | P Value | GMT (95% CI), AU/mL | P Value | ||||
| Absent/Absent | 110 (67.1) | 83 (75.5) | 42.0 (33.0–49.0) | 8281 (7305–9386) | Reference | 537 (474–607) | Reference | 15 438 (13 665–17 441) | Reference |
| Present/Absent | 12 (7.3) | 9 (75.0) | 33.5 (29.0–39.5) | 12 026 (8204–17 628) | .169 | 796 (552–1147) | 0.144 | 14 644 (10 630–20 174) | .990 |
| Absent/Present | 21 (12.8) | 12 (57.1) | 41.0 (46.0–51.0) | 9172 (6717–12 524) | .879 | 738 (515–1058) | 0.125 | 22 012 (16 555–29 270) | .060 |
| Present/Present | 21 (12.8) | 17 (81.0) | 40.0 (28.0–44.0) | 12 343 (9562–15 932) | .032 | 627 (481–817) | 0.682 | 22 479 (16 575–30 487) | .042 |
Abbreviations: CI, confidence interval; GMT, geometric mean titer; IQR, interquartile range.
DISCUSSION
A booster dose of an mRNA COVID-19 vaccine induces a higher antibody titer than is produced by the primary 2-dose series [18, 19]. Our previous study showed that the IgG(S-RBD) titers after the second dose of the primary series were relatively low for male participants and the elderly [15], as shown in other studies [20, 21]. In contrast, in the present study of the booster dose for the same cohort, male sex and age showed positive correlations with the fold change in IgG(S-RBD) titers, reaching comparable levels in the IgG(S-RBD) titers by sex and age. Other studies have also shown that the specific IgG titers after the BNT162b2 booster were comparable for sex and age [22, 23]. The mechanism for the difference in the immunogenicity of a BNT162b2 vaccine by sex and age between the second and booster doses is unclear, but the booster vaccination would drive antibody production, especially in the populations with relatively weak responses to the primary series.
The relation between the emergence of vaccine-related adverse reactions and antibody induction by the primary series of mRNA COVID-19 vaccines has been reported. Studies that used scores based on the sum of the presence or a severity scale of solicited adverse reactions showed no significant correlation of adverse reactions with the spike-specific IgG titers [24–26]. On the other hand, when each reaction was separately analyzed, the presence of fever after the second dose was shown to be correlated with high IgG titers [27, 28]. We previously showed that degree of fever after the second dose was correlated with the IgG(S-RBD) titers by a multivariate analysis [15]. In the present study, the degree of fever after the booster dose, which was defined by the highest body temperature after the vaccination, was again shown to have a significant correlation with the IgG(S-RBD) titers. Note that the clinical significance of the difference in the IgG(S-RBD) titers due to postvaccination fever (eg, the GMTs for the participants with fever of <37.0°C and ≥38.0°C were 14 011 AU/mL and 23 020 AU/mL, respectively) for protection against SARS-CoV-2 infection is difficult to evaluate. In general, the IgG(S-RBD) titers measured with the assay used in our study were shown to be correlated with the neutralizing antibody levels, a surrogate marker for protection, although the correlation at relatively high IgG(S-RBD) titers, as detected in our study, has been inconsistent across studies [29–31]. A positive correlation was reported not only for wild-type virus but also for variants including B.1.617.2 (Delta) and B.1.1.529 (Omicron) [32]. The mechanism of how fever after SARS-CoV-2 mRNA vaccination is linked to antibody production remains unclear. Elucidating the mechanism will lead to a better understand of the sufficient antibody induction mechanism of mRNA COVID-19 vaccines.
The possibility of a relation between postvaccination fever and the SARS-CoV-2–specific antibody titers has created concern that antipyretics, which can suppress fever, may have a negative influence on antibody responses to SARS-CoV-2 vaccination. To date, few studies have been done to determine the influence of antipyretics on the immunogenicity of COVID-19 vaccines. We previously showed that the use of antipyretics for fever or other adverse reactions did not interfere with the antibody responses to the primary BNT162b2 series [15]. In the present study for the BNT162b2 booster, no influence of antipyretic use on the IgG(S-RBD) titers was observed again. Although several in vitro laboratory studies have demonstrated that NSAIDs inhibit several pathways leading to antibody responses [33–35], the IgG(S-RBD) titers of the group with NSAIDs were comparable to those of the group with acetaminophen. The present study was not designed to evaluate the influence by the type of antipyretics, but it is suggested that neither acetaminophen nor NSAIDs would interfere with the elevation of IgG(S-RBD) titers. Antipyretics may also be used prophylactically. One study of a COVID-19 adenoviral vector vaccine reported that prophylactic acetaminophen use reduced many adverse reactions without interfering with antibody responses [36]. There are no studies on the influence of prophylactic use of antipyretics on antibody responses to mRNA COVID-19 vaccines. Taken together, our results indicate that the use of antipyretics for emerging adverse events, regardless of the type, acetaminophen or NSAIDs, would be helpful for alleviating adverse reactions, including fever, without interfering with antibody responses to SARS-CoV-2 mRNA vaccination.
It would be of great interest for physicians to determine whether the presence of an adverse reaction after the 2-dose primary series is useful to predict the emergence of the corresponding reaction after the booster dose. In the present study, the presence of swelling as a local reaction and several systemic reactions, including fever after the second dose, was associated with the emergence of the corresponding reactions after the booster dose. Fever after the second dose showed the highest RR of 3.97. Of interest, fever after the booster dose was independently correlated with the IgG(S-RBD) titers after the booster dose, irrespective of fever after the second dose. Thus, participants with fever after the second dose may be more likely to have fever after the booster dose, but fever after the second dose would not affect the antibody responses to the booster dose. To our knowledge, no investigations of the correlation between postvaccination fever and specific IgG levels have been done in the same cohort throughout the primary and booster vaccinations. These findings impress a potential linking of postvaccination fever with the antibody production induced by mRNA COVID-19 vaccines.
This study has some limitations that should be acknowledged. First, it is a single-center observational study with a relatively small sample size of 281 participants. Additionally, the population was relatively young and predominantly female. Second, the data collection methods for the solicited adverse reactions were slightly different between the second and booster doses. The data collection period for adverse reactions was 7 days after vaccination for the booster dose, whereas it was 5 days for the second dose. However, most common adverse reactions have been known to occur within a few days after vaccination [37]; thus, the difference would have little impact on our findings. Finally, the type, dosage, and timing of antipyretic usage were chosen by each participant and are thus arbitrary. Besides, this study has insufficient statistical power due to the relatively small sample size, which may have led to underestimation in the possible negative influence of antipyretic use on the antibody responses. Large-scale randomized controlled studies will be necessary to clarify the influence of antipyretics on the immunogenicity outcomes of COVID-19 vaccines, but we believe that our real-world data from healthcare workers with self-medicated antipyretics is informative.
In conclusion, a booster dose of BNT162b2 vaccine can restore waning immunity and significantly increase the antibody titers, especially in males and the elderly who had relatively low antibody responses to the primary 2-dose series. Post–booster vaccination fever could be independently correlated with mRNA vaccine–induced specific IgG levels, without any significant influence of fever after the primary series. Although the relatively small sample size in this study means that our results are inconclusive, they indicate that antipyretic medications would be helpful to alleviate suffering from adverse reactions, without suppressing the specific IgG responses induced by a BNT162b2 vaccine booster.
Supplementary Material
Contributor Information
Naoki Tani, Medicine and Biosystemic Science, Kyushu University Graduate School of Medical Sciences, Fukuoka, Japan.
Hideyuki Ikematsu, Ricerca Clinica Co, Fukuoka, Japan.
Takeyuki Goto, Medicine and Biosystemic Science, Kyushu University Graduate School of Medical Sciences, Fukuoka, Japan.
Kei Gondo, Clinical Laboratory, Fukuoka City Hospital, Fukuoka, Japan.
Takeru Inoue, Medicine and Biosystemic Science, Kyushu University Graduate School of Medical Sciences, Fukuoka, Japan.
Yuki Yanagihara, Pharmacy, Fukuoka City Hospital, Fukuoka, Japan.
Yasuo Kurata, Pharmacy, Fukuoka City Hospital, Fukuoka, Japan.
Ryo Oishi, Department of Infectious Diseases, Fukuoka City Hospital, Fukuoka, Japan.
Junya Minami, Department of Infectious Diseases, Fukuoka City Hospital, Fukuoka, Japan.
Kyoko Onozawa, Department of Infectious Diseases, Fukuoka City Hospital, Fukuoka, Japan.
Sukehisa Nagano, Department of Neurology, Fukuoka City Hospital, Fukuoka, Japan.
Hiroyuki Kuwano, Department of Surgery, Fukuoka City Hospital, Fukuoka, Japan.
Koichi Akashi, Medicine and Biosystemic Science, Kyushu University Graduate School of Medical Sciences, Fukuoka, Japan.
Nobuyuki Shimono, Center for the Study of Global Infection, Kyushu University Hospital, Fukuoka, Japan.
Yong Chong, Medicine and Biosystemic Science, Kyushu University Graduate School of Medical Sciences, Fukuoka, Japan.
Supplementary Data
Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
Notes
Acknowledgments. We thank all of the staff members of Fukuoka City Hospital for their support for sample collection and data reduction.
Financial support. This work was supported by our own resources.
References
- 1. Polack FP, Thomas SJ, Kitchin N, et al. . Safety and efficacy of the BNT162b2 mRNA Covid-19 vaccine. N Engl J Med 2020; 383:2603–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Baden LR, el Sahly HM, Essink B, et al. . Efficacy and safety of the mRNA-1273 SARS-CoV-2 vaccine. N Engl J Med 2020; 384:403–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Moreira EDJ, Kitchin N, Xu X, et al. . Safety and efficacy of a third dose of BNT162b2 Covid-19 vaccine. N Engl J Med 2022; 386:1910–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Bar-On YM, Goldberg Y, Mandel M, et al. . Protection against Covid-19 by BNT162b2 booster across age groups. N Engl J Med 2021; 385:2421–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Falsey AR, Frenck RW, Walsh EE, et al. . SARS-CoV-2 neutralization with BNT162b2 vaccine dose 3. N Engl J Med 2021; 385:1627–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Gruell H, Vanshylla K, Tober-Lau P, et al. . mRNA booster immunization elicits potent neutralizing serum activity against the SARS-CoV-2 Omicron variant. Nat Med 2022; 28:477–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Jalkanen P, Kolehmainen P, Häkkinen HK, et al. . COVID-19 mRNA vaccine induced antibody responses against three SARS-CoV-2 variants. Nat Commun 2021; 12:3991. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Yu Y, Esposito D, Kang Z, et al. . mRNA vaccine-induced antibodies more effective than natural immunity in neutralizing SARS-CoV-2 and its high affinity variants. Sci Rep 2022; 12:2628. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Paran Y, Saiag E, Spitzer A, et al. . Short-term safety of booster immunization with BNT162b2 mRNA COVID-19 vaccine in healthcare workers. Open Forum Infect Dis 2022; 9:ofab656. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Thomas SJ, Moreira ED, Kitchin N, et al. . Safety and efficacy of the BNT162b2 mRNA Covid-19 vaccine through 6 months. N Engl J Med 2021; 385:1761–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Pillsbury AJ, Glover C, Jacoby P, et al. . Active surveillance of 2017 seasonal influenza vaccine safety: an observational cohort study of individuals aged 6 months and older in Australia. BMJ Open 2018; 8:e023263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Merck . Pneumovax 23 (pneumococcal vaccine polyvalent) sterile, liquid vaccine for intramuscular or subcutaneous injection, prescribing information.http://www.merck.com/product/usa/pi_circulars/p/pneumovax_23/pneumovax_pi.pdf. Accessed 3 May 2022.
- 13. Centers for Disease Control and Prevention . Possible side effects after getting a COVID-19 vaccine.https://www.cdc.gov/coronavirus/2019-ncov/vaccines/expect/after.html. Accessed 3 May 2022.
- 14. National Health Service . Coronavirus (COVID-19) vaccines side effects and safety.https://www.nhs.uk/conditions/coronavirus-covid-19/coronavirus-vaccination/safety-and-side-effects/. Accessed 3 May 2022.
- 15. Tani N, Chong Y, Kurata Y, et al. . Relation of fever intensity and antipyretic use with specific antibody response after two doses of the BNT162b2 mRNA vaccine. Vaccine 2022; 40:2062–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. US Food and Drug Administration . Guidance for industry: toxicity grading scale for healthy adult and adolescent volunteers enrolled in preventive vaccine clinical trials.https://www.fda.gov/media/73679/download. Accessed 3 May 2022.
- 17. Madhusudhanan N, Lenin M, Ellen A, et al. . Clinical evaluation of the Abbott Alinity SARS-CoV-2 spike-specific quantitative IgG and IgM assays among infected, recovered, and vaccinated groups. J Clin Microbiol 2021; 59:e00388-21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Busà R, Sorrentino MC, Russelli G, et al. . Specific anti-SARS-CoV-2 humoral and cellular immune responses after booster dose of BNT162b2 Pfizer-BioNTech mRNA-based vaccine: integrated study of adaptive immune system components. Front Immunol 2022; 13:856657. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Romero-Ibarguengoitia ME, Rivera-Salinas D, Hernández-Ruíz YG, et al. . Effect of the third dose of BNT162b2 vaccine on quantitative SARS-CoV-2 spike 1–2 IgG antibody titers in healthcare personnel. PLoS One 2022; 17:e0263942. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Debes AK, Xiao S, Colantuoni E, et al. . Association of vaccine type and prior SARS-CoV-2 infection with symptoms and antibody measurements following vaccination among health care workers. JAMA Intern Med 2021; 181:1660–2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Müller L, Andrée M, Moskorz W, et al. . Age-dependent immune response to the BioNTech/Pfizer BNT162b2 COVID-19 vaccination. Clin Infect Dis 2021; 73:2065–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Eliakim-Raz N, Leibovici-Weisman Y, Stemmer A, et al. . Antibody titers before and after a third dose of the SARS-CoV-2 BNT162b2 vaccine in adults aged ≥60 years. JAMA 2021; 326:2203–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Lustig Y, Gonen T, Meltzer L, et al. . Superior immunogenicity and effectiveness of the third compared to the second BNT162b2 vaccine dose. Nat Immunol 2022; 23:940–6. [DOI] [PubMed] [Google Scholar]
- 24. Müller L, Andrée M, Moskorz W, et al. . Age-dependent immune response to the BioNTech/Pfizer BNT162b2 coronavirus disease 2019 vaccination. Clin Infect Dis 2021; 73:2065–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Coggins SA, Laing ED, Olsen CH, et al. . Adverse effects and antibody titers in response to the BNT162b2 mRNA COVID-19 vaccine in a prospective study of healthcare workers. Open Forum Infect Dis 2022; 9:ofab575. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Takeuchi M, Higa Y, Esaki A, Nabeshima Y, Nakazono A. Does reactogenicity after a second injection of the BNT162b2 vaccine predict spike IgG antibody levels in healthy Japanese subjects? PLoS One 2021; 16:e0257668. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Kanizsai A, Molnar T, Varnai R, et al. . Fever after vaccination against SARS-CoV-2 with mRNA-based vaccine associated with higher antibody levels during 6 months follow-up. Vaccines (Basel) 2022; 10:447. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Yamamoto S, Fukunaga A, Tanaka A, et al. . Association between reactogenicity and SARS-CoV-2 antibodies after the second dose of the BNT162b2 COVID-19 vaccine. Vaccine 2022; 40:1924–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Sabine C-R, Marcel M, Florence A, et al. . Evaluation of three quantitative anti-SARS-CoV-2 antibody immunoassays. Microbiol Spectr 2021; 9:e01376-21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Lee B, Ko J-H, Park J, et al. . Estimating the neutralizing effect and titer correlation of semi-quantitative anti-SARS-CoV-2 antibody immunoassays. Front Cell Infect Microbiol 2022; 12:822599. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Kohmer N, Rühl C, Ciesek S, Rabenau HF. Utility of different surrogate enzyme-linked immunosorbent assays (sELISAs) for detection of SARS-CoV-2 neutralizing antibodies. J Clin Med 2021; 10:2128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Moy JN, Anderson M, Shen X, et al. . Neutralizing antibody activity to severe acute respiratory syndrome coronavirus 2 Delta (B.1.617.2) and Omicron (B.1.1.529) after 1 or 2 doses of BNT162b2 vaccine in infection-naive and previously infected individuals [manuscript published online ahead of print 27 June 2022]. J Infect Dis. 10.1093/infdis/jiac261 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Saleh E, Moody MA, Walter EB. Effect of antipyretic analgesics on immune responses to vaccination. Hum Vaccin Immunother 2016; 12:2391–402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Ryan EP, Pollock SJ, Murant TI, Bernstein SH, Felgar RE, Phipps RP. Activated human B lymphocytes express cyclooxygenase-2 and cyclooxygenase inhibitors attenuate antibody production. J Immunol 2005; 174:2619–26. [DOI] [PubMed] [Google Scholar]
- 35. Purssell E. Cyclooxygenase inhibitors inhibit antibody response through interference with MAPK/ERK pathways and BLIMP-1 inhibition. Med Hypotheses 2014; 83:372–7. [DOI] [PubMed] [Google Scholar]
- 36. Folegatti PM, Ewer KJ, Aley PK, et al. . Safety and immunogenicity of the ChAdOx1 nCoV-19 vaccine against SARS-CoV-2: a preliminary report of a phase 1/2, single-blind, randomised controlled trial. Lancet 2020; 396:467–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Rosenblum HG, Gee J, Liu R, et al. . Safety of mRNA vaccines administered during the initial 6 months of the US COVID-19 vaccination programme: an observational study of reports to the Vaccine Adverse Event Reporting System and v-safe. Lancet Infect Dis 2022; 22:802–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.