ABO blood types and SARS-CoV-2 infection assessed using seroprevalence data in a large population-based sample: the SAPRIS-SERO multi-cohort study (2024)

The SAPRIS (“SAnté, Perception, pratiques, Relations et Inégalités Sociales en population générale pendant la crise COVID-19”) study was set up in April 2020 to investigate various aspects of the COVID-19 crisis (COVID-19 infection/diagnosis and experience of lockdown), based on a consortium of existing French prospective cohort studies. In May 2020, participants from the Constances, E3N-E4N and NutriNet-Santé cohort studies answering SAPRIS questionnaires were also invited to take part in the SAPRIS-SERO project which aimed to estimate the seroprevalence of antibodies against SARS-CoV-2 at the population level, as previously described^12,13. Briefly, between May and November 2020 (during or at the end of the first wave, pre-vaccine period), participants received self-sampling kits by mail to collect dried blood spots from which were detected anti-SARS-CoV-2 antibodies (IgG) directed against the S1 domain of the spike protein (S) and the nucleocapsid protein (NP) using ELISA (Euroimmun®, Lübeck, Germany), as well as neutralizing anti-SARS-CoV-2 antibodies (SN) using an in-house micro-neutralization assay. ELISA-S and ELISA-NP tests were considered to be positive with an optical density ratio ≥ 1.1, indeterminate between 0.8 and 1.1, and negative < 0.8. The SN test was considered to be positive with a titre ≥ 40. ELISA-NP and SN tests were performed whenever the optical density ratio for ELISA-S was ≥ 0.7. Positivity (vs. negativity) to ELISA-S test and positivity (vs. negativity) to all three ELISA-S, ELISA-NP and SN tests (associated to more symptomatic COVID-19 in a previous study in SAPRIS-SERO¹³)were considered as outcomes. Characteristics of the participants including ABO and RhD blood types were collected from questionnaires as part of each cohort follow-up or from the shared SAPRIS questionnaire. Associations between the seroprevalence of antibodies against SARS-CoV-2 and blood types were studied using multi-adjusted logistic regression models stratified by cohort and including the following covariates: sex (men/women), age (≥ 18– < 40, 40–49, 50–59, 60–69, ≥ 70 years old), month of blood collection, residential area during the lockdown: city size (rural, < 20,000, 20–100,000, ≥ 100,000 inhabitants) and French administrative region. Additional covariates linked to the risk of SARS-CoV-2 infection were included in sensitivity analyses: weight status, smoking status, educational level, professional activity during lockdown, socio-professional category, number of individuals at home during lockdown and alcohol intake. All tests were two-sided and P < 0.05 was considered statistically significant. Analyses were carried out using SAS 9.4 (SAS Institute Inc., USA).

Ethics approval and informed consent

The SAPRIS-SERO study (registered #NCT04392388) was conducted in accordance with the relevant guidelines and regulations and was approved by the ethics committee CPP Sud-Méditerranée III on April 27th 2020 and by the CNIL #920,193. All participants provided written or electronic informed consent for participation in each cohort and a specific electronic informed consent for participation in the SAPRIS-SERO study.

Our study included 67,340 participants, among which 2964 with a positive seroprevalence of anti SARS-CoV-2 antibodies according to the ELISA-S test and 64,376 participants who tested negative (respectively 678 triple positive and 64,011 triple negative to ELISA-S, ELISA-NP and SN tests). Characteristics of participants are shown in Table 1.

Higher odds of presenting anti SARS-CoV-2 antibodies (Table 2) were observed in participants belonging to non-O blood types (ELISA-S: OR _{non-O vs. O} = 1.09 [1.01–1.17], ELISA-S/ELISA-NP/SN: OR _{non-O vs. O} = 1.19 [1.02–1.39]) and especially to blood type A (ELISA-S: OR_{A vs. O} = 1.08 [1.00–1.17], ELISA-S/ELISA-NP/SN: OR _{A vs. O} = 1.19 [1.01–1.41]) and AB (ELISA-S/ELISA-NP/SN: OR _{AB vs. O} = 1.43 [1.01–2.03]). In analyses combining ABO and RhD blood types, higher odds were observed in participants with blood types A- (ELISA-S: OR _{A- vs. O+}  = 1.16 [1.00–1.35]), A + (ELISA-S/ELISA-NP/SN: OR _{A+ vs. O+}  = 1.24 [1.03–1.48]) and AB + (which represented the majority of AB participants, ELISA-S: OR _{AB+ vs. O+}  = 1.24 [1.02–1.51], ELISA-S/ELISA-NP/SN: OR _{AB+ vs. O+}  = 1.60 [1.11–2.30]). No association was observed comparing RhD positive and negative blood types overall.

Additional adjustments in sensitivity analyses did not change the results of increased odds of positive seroprevalence in non-O participants (ELISA-S: OR _{non-O vs. O} = 1.09 [1.01–1.18], ELISA-S/ELISA-NP/SN: OR _{non-O vs. O} = 1.19 [1.02–1.39]).

In this study, we used seroprevalence data obtained from screening a large population-based sample from 3 French prospective cohorts to assess the associations between ABO blood types and the risk of SARS-CoV-2 infection. Our results showed a decreased seroprevalence of anti-SARS-CoV-2 antibodies in type O individuals compared to all others. Among non-O blood types, types A and AB were more likely positive for SARS-CoV-2 infection. No association was observed with type B or RhD types.

These results confirmed prior observations obtained in other population settings, namely a decrease risk of infection for type O individuals and an increase risk for type A individuals, with some indication of increased risk for type AB individuals as well, thus strengthening the evidence towards a differential susceptibility of SARS-CoV-2 infection according to ABO blood types. Yet, most studies so far involved registry or hospital data of SARS-CoV-2 positivity from PCR test results in patients or blood donors in comparison to the general population^8,9,10 and fewer studies used seroprevalence data, mostly in small samples, with the exception of a recent report in 32,600 blood donors¹¹. In turn, our study used seroprevalence data and a screening design whereby a large population-based sample of 67,340 participants involved in the SAPRIS-SERO project received a dried blood spot sampling kit. This allowed us to test for SARS-CoV-2 infection in a single general population, regardless whether the participants sought testing or had symptoms, allowing detecting a large panel of infected cases.

This differential susceptibility to SARS-CoV-2 infection according to blood types could be due to the incorporation of ABH antigens on the envelop of SARS-CoV-2 virions, leading to a likelihood of infection from one individual to another following ABO incompatibility^9,10. As type O individuals carry both anti-A and anti-B antibodies, they would be less susceptible to primary infection by SARS-CoV-2 viruses enveloped with type A or B-like patterns such as those excreted by non-O individuals. In contrast, type A individuals, carrying anti-B antibodies, would be susceptible to infection by viruses excreted by type O and type A individuals (i.e., 42% and 44% of the French population respectively), while type AB individuals carrying neither anti-A or anti-B antibodies would be susceptible to infection by individuals of all group types, which is why type A or AB are particularly at risk for SARS-CoV-2 infection. Type B individuals carrying anti-A antibodies are susceptible to infection by viruses excreted by type O and type B individuals but type B is less represented in the French population (10%) which may explain why no significant association was observed for these individuals. Another hypothesis, yet less supported by available data relates to the entry of SARS-CoV-2 in host’s cells, that would be facilitated by antigen A^9,14. Finally, non-O individuals are at higher risk of thromboembolism, a common complication of COVID-19, hence beyond infection, non-O individuals could also be at higher risk of severe cases of COVID-19^9,10.

Our study strengths pertained to the systematic screening for anti-SARS-CoV-2 antibodies targeting a large well-characterized sample, hence providing data from positive and negative cases within the same population, as well as the use of seroprevalence allowing the detection of cases that may or may not have symptoms or sought testing. Yet, some limitations should also be acknowledged. First, although large, our sample size may have still been limited to have sufficient power in analyses combining ABO and RhD blood types, especially for rarer blood types. Next, misclassification of individuals as ‘negative’ may have arisen from imperfect sensitivity of the ELISA-S test (85–90%, the specificity being > 95%)^15,16 and from the decrease of anti-SARS-CoV-2 antibodies over time (with no detectable antibodies 4–5 months post-infection in half of participants in a recent report¹¹). Yet, the sample collection was performed between May and November 2020, the first and second waves of the pandemic occurring in spring and fall 2020 in France, so that antibodies should still be detectable in most participants. Finally, the observational nature of the study makes it difficult to assess the extent to which participants have actually been exposed to the virus. Although lifestyle-related exposure to the virus is mostly unlikely to have been differential according to blood types, we adjusted our models on several covariates related to the spatial distribution of the epidemic since it could also relate to blood types, considering the distribution of ABO blood types in different regions or populations. However, we did not have access to racial and ethnic data as the collection of such data is generally not permitted in the framework of French cohort studies. Finally, our data only reflect infections occurring before November 2020, that is, before the emergence of variants but also before the large-scale vaccination campaigns.

To conclude, this study adds to the current body of evidence providing data on the association between ABO blood types and the risk of SARS-CoV-2 infection from screening a large sample from the general population using seroprevalence data. Our results confirmed a higher susceptibility of infection for individuals of blood types A and AB and a lesser risk for blood type O, thereby providing additional insights into the dynamic of SARS-CoV-2 infection. However, considering the magnitude of association and the current mechanistic hypotheses related to blood type incompatibility (hence a collective rather than individual type of protection), type O individuals are not without risk of infection and protective measures remain the best way of preventing SARS-CoV-2 transmission and infection.

Author notes

1. A list of authors and their affiliations appears at the end of the paper.

Authors and Affiliations

1. Inserm U1153, Inrae U1125, Cnam, Nutritional Epidemiology Research Team (EREN), Epidemiology and Statistics Research Center – Université Paris Cité (CRESS), Sorbonne Paris Nord University, Bobigny, France

   Mélanie Deschasaux-Tanguy,Fabien Szabo de Edelenyi,Nathalie Druesne-Pecollo,Younes Esseddik,Julien Allègre,Bernard Srour,Pilar Galan,Serge Hercberg,Mathilde Touvier,Fabien Szabo de Edelenyi,Julien Allègre,Nathalie Druesne-Pecollo,Younes Esseddik,Serge Hercberg,Mathilde Touvier&Mathilde Touvier

2. UVSQ, Inserm, Gustave Roussy, “Exposome and Heredity” team, CESP UMR1018, Paris-Saclay University, Villejuif, France

   Gianluca Severi,Gianluca Severi,Fanny Artaud,Pascale Gerbouin-Rérolle,Mélody Enguix,Camille Laplanche,Roselyn Gomes-Rimav,Lyan Hoang,Emmanuelle Correia,Alpha Amadou Barry,Nadège Senina&Gianluca Severi

3. Department of Statistics, Computer Science and Applications “G. Parenti”, University of Florence, Florence, Italy

   Gianluca Severi,Gianluca Severi&Gianluca Severi

4. Population-based Cohorts Unit, INSERM, UMS 011, Paris Saclay University, Université de Versailles Saint-Quentin-en-Yvelines, Université Paris Cité, Paris, France

   Marie Zins,Emmanuel Wiernik,Marie Zins,Sofiane Kab,Adeline Renuy,Stéphane Le Got,Céline Ribet,Mireille Pellicer,Emmanuel Wiernik,Marcel Goldberg&Marie Zins

5. Unité des Virus Emergents (UVE), IRD 190, INSERM 1207, Aix Marseille Univ, Marseille, France

   Stéphane Priet,Paola Mariela Saba Villarroel,Toscane Fourié,Souand Mohamed Ali,Abdenour Amroun,Morgan Seston,Nazli Ayhan,Boris Pastorino,Xavier de Lamballerie&Xavier de Lamballerie

6. Inserm, Institut Pierre-Louis d’Epidémiologie et de Santé Publique, Sorbonne Université, Paris, France

   Fabrice Carrat,Clovis Lusivika-Nzinga,Gregory Pannetier,Nathanael Lapidus,Isabelle Goderel,Céline Dorival,Jérôme Nicol,Olivier Robineau,Fabrice Carrat&Fabrice Carrat

7. Département de Santé Publique, APHP, Sorbonne Université, Paris, France

   Fabrice Carrat,Fabrice Carrat&Fabrice Carrat

8. Obstetrical, Perinatal and Pediatric Research Team (EPOPé), Centre for Research in Epidemiology and StatisticS (CRESS), Inserm, INRAE, Université de Paris, Paris, France

   Pierre-Yves Ancel&Pierre-Yves Ancel

9. Centre for Research in Epidemiology and StatisticS (CRESS), Inserm, INRAE, Université de Paris, Paris, France

   Marie-Aline Charles&Marie-Aline Charles

10. IRIS, UMR CNRS 8156, EHESS, Inserm U997, Aubervilliers, France

    Nathalie Bajos

11. EPIPAGE-2 Joint Unit, Paris, France

    Valérie Benhammou,Anass Ritmi&Laetitia Marchand

12. ELFE Joint Unit, Paris, France

    Cecile Zaros,Elodie Lordmi,Adriana Candea,Sophie de Visme,Thierry Simeon,Xavier Thierry,Bertrand Geay,Marie-Noelle Dufourg&Karen Milcent

13. Inserm, Paris, France

    Cindy Lai,Liza Belhadji,Hélène Esperou&Sandrine Couffin-Cadiergues

14. Aviesan, Inserm, Paris, France

    Jean-Marie Gagliolo

15. Fondation Jean Dausset-CEPH (Centre d’Etude du Polymorphisme Humain), CEPH-Biobank, Paris, France

    Hélène Blanché,Jean-Marc Sébaoun,Jean-Christophe Beaudoin,Laetitia Gressin,Valérie Morel,Ouissam Ouili&Jean-François Deleuze

Consortia

Contributions

The authors’ contributions were as follows – M.D.T., M.T. conceptualized the study and defined the analytical strategy; F.S.E.: performed statistical analyses; M.D.T. supervised statistical analyses and drafted the manuscript; M.T.: supervised statistical analyses and writing; F.S.E., N.D.P., Y.E., J.A., P.G., S.H., G.S., M.Z., E.W., X.dL, F.C., M.T., SAPRIS-SERO study group: played a key role in the acquisition of the data. All authors critically helped in the interpretation of results, revised the manuscript and provided relevant intellectual input. They all read and approved the final manuscript. M.D.T. and M.T. had primary responsibility for the final content.

Corresponding author

Correspondence to Mélanie Deschasaux-Tanguy.

Competing interests

The authors declare no competing interests.

Publisher's note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions