Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) is the virus responsible for the Coronavirus Disease 2019 (COVID-19) pandemic, which began in November, 2019. Most COVID-19 cases are asymptomatic to mild, but some cases lead to life-threatening pneumonia. As of mid-December, 2021, more than 5.4 million patients worldwide have died from the effects of COVID-19 (1).

To control this pandemic, various prophylactic and therapeutic approaches are being tried clinically, including vaccines, convalescent plasma therapy (CPT) and therapeutic monoclonal antibodies (Mabs) (2). Among these immunotherapies, the neutralizing antibodies (Nabs) that interrupt viral infection are essential components and are induced by natural infection or vaccination. Some SARS-CoV-2 vaccines, including messenger RNA-based vaccines and adenovirus-vectored vaccines, have a high efficacy at preventing symptomatic disease (3). Numerous different vaccines have been manufactured and distributed all over the world, but the production has not been sufficient to vaccinate populations. The rapid increase of vaccine supply remains the best hope for overcoming this pandemic.

On the other hand, passive immunization using CPT remains a therapeutic option and has been used for infectious diseases caused by severe acute respiratory syndrome coronavirus (4), Middle East Respiratory Syndrome Coronavirus (4), and Ebola virus (5). The efficacy of CPT for COVID-19 patients is controversial (6), although some studies provided good evidence that CPT was safe and reduced mortality when COVID-19 patients were treated in combination with antiviral drugs, steroids, and other supportive care (7). Treatment with therapeutic Mabs is another option as a passive immunotherapy with no risk of post-transfusion infection and no need to collect convalescent sera from many recovered patients. Mabs are important for immunocompromised people and unvaccinated people to be protected from infection. Currently, Mabs for five diseases (respiratory syncytial virus, anthrax, Clostridium difficile, human immunodeficiency virus 1, and Ebola virus) are approved by the US Food and Drug Administration (FDA) (8, 9). As for SARS-CoV-2, three anti-SARS-CoV-2 Mabs products (sotrovimab, combination of bamlanivimab plus etesevimab, and combination of casirivimab plus imdevimab) have Emergency Use Authorizations (EUAs) from the FDA (10). These Mabs products are used for treatment of mild to moderate COVID-19 nonhospitalized patients to prevent hospitalization and death. Combination of casirivimab plus imdevimab has obtained fast-track approval of Japan’s Ministry of Health, Labour, and Welfare on 19 July 2021 (11).

In the face of these new therapies, SARS-CoV-2 has evolved from its original strain. The D614G mutation of spike protein (S protein) was found worldwide by the end of June, 2020 (12). Recently, the emergence of SARS-CoV-2 variants with mutations that can enhance transmissibility and reduce neutralization activity has brought new challenges to the management of COVID-19. The World Health Organization (WHO) classified five variants (B.1.1.7, P.1, B.1.351, B.1.617.2, and B.1.1.529) as variants of concern (VOC) (13). B.1.1.7, which was firstly detected in the United Kingdom, has an N501Y mutation in the receptor binding domain (RBD) of S protein. P.1, which was identified in Brazil, has three mutations (K417T, E484K, and N501Y) in the RBD. B.1.351, which was found in South Africa, has three mutations (K417N, E484K, and N501Y) in the RBD (14). B.1.617.2, which was confirmed in India, has mutations (L452R and T478K) in the RBD (15), leading to higher viral load in infected individuals, in addition to the P681R mutation which increases the virus transmissibility (16). Finally, B.1.1.529, which was detected in Botswana on November 11, 2021 and South Africa on November 14, 2021, has 15 mutations (G339D, S371L, S373P, S375F, K417N, N440K, G446S, S477N, T478K, E484A, Q493R, G496S, Q498R, N501Y, and Y505H) in the RBD (17). The N501Y mutation shared among B.1.1.7, B.1.351, P1 and B.1.1.529 have influence on the affinity between the angiotensin-converting enzyme 2 (ACE2) receptor and the RBD of the S protein, which causes high transmissibility of the virus. The variant B.1.1.7 has proven to be more transmissible than original strains (18). The E484K mutation in the RBD of B.1.351 and P1 is involved in immune escape (19, 20). The K417N mutation of B.1.351 and B.1.1.529 and K417T mutation of P.1 are suggested to change the conformation of S protein, allowing escape from Nabs (19, 20). Nabs have reduced activity on the variants B.1.351 and P.1 (14). Especially, the variant B.1.351 is much more resistant to neutralization by Nabs, possibly because of differences in the mutations of the N terminal domain (NTD), which are associated with escape from immunity (21). Nabs against B.1.1.529 from sera of convalescent patients infected with the ancestral SARS-CoV-2 and sera of vaccinated people with two dose of ChAdOx1 nCoV-19 or BNT162b2 is also lower than that against the other variants (22, 23).

As elsewhere in the world, Japan is facing COVID-19 pandemic. As of mid-December 2021, nearly 1,730,000 Japanese people had been infected with COVID-19 and about 18,000 people had died (1). So far, the country has been affected by five waves of exponential increase in new cases. D614G_KR, which had D614G mutation of S protein and 203_204delinsKR mutation of nucleocapsid protein, and its lineage were predominant during the first to third waves, from March 2020 to February 2021 (24). VOCs have also spread to Japan. B.1.1.7 was first detected on 25 December 2020 and rapid community spread occurred from March 2021. As of June 2021, B.1.1.7 became the predominant variant in Japan. In contrast, B.1.351 and P.1 were not spread widely in Japan, although they were found at the end of December 2020 and the beginning of January 2021. B.1.617.2 was detected on 20 April 2021 and has rapidly replaced B.1.1.7. Then, B.1.617.2 has become the predominant variant at the fifth wave from July 2021 in Japan (25, 26). Recently, B.1.1.529 was detected in Japan as in other countries. Then, the Japanese government is on the alert for the sixth wave (27).

Hence, it is important to know about the longevity of the neutralizing activity of convalescent sera against SARS-CoV-2 to estimate the possibility of reinfection, to select good donors for CPT, and to make therapeutic Mabs products. Previous studies have shown that neutralizing activity of convalescent sera is maintained up to six to twelve months post onset, although follow-up studies for longer duration are still needed (28–30). Furthermore, the breadth and longevity of cross-neutralizing activities against VOCs have been minimally tested (28, 31). In this study, we analyzed the longevity and breadth of neutralizing activity of COVID-19 convalescent sera across the VOCs (B1.1.7, P.1, and B.1.351) and the D614G.

The purpose of this study was to examine the longevity of Nab activity of COVID-19 convalescent sera against D614G, and their neutralizing breadth against B.1.1.7, P.1, and B.1.351. We performed live virus neutralization assays against D614G and three VOCs to assess the Nab titers of COVID-19 convalescent sera. Infection of SARS-CoV-2 to the convalescent patients were confirmed from March 2020 to January 2021. Although B.1.1.7, P.1, and B.1.351 had already detected in Japan at the end of December 2020, only B.1.1.7 had spread in Japan and it rapidly increased from February 2021 (34, 35). A local surveillance in Japan also suggested that the B.1.1.7 became dominant after mid-April (36). These observations suggested that all participants in this study were likely to be infected with D614G. However, we could not completely exclude the possibility of VOCs infection due to the loss of the information about the viral sequence from patients’ nasopharyngeal specimens. In Addition, the imbalance of samples and small sample size were also our limitation.

This study revealed that all sera of ‘patients with pneumonia’ maintained cross-neutralizing activity against B.1.1.7, P.1, and B.1.351 until 3-8 months post infection. On the other hand, 14 out of 19 sera of ‘patients without pneumonia’ could not get or keep cross-neutralizing activity against three VOCs ( Figures 1 , 2A, B ). This reason might be that weak immune-response against SARS-CoV-2 lead to the low Nab titers of ‘patients without pneumonia’ against D614G compared to those of ‘patients with pneumonia’ (21, 37–40). The Nab titer against B.1.351 was lower than that against the other variants ( Figure 3A ), as similarly reported by other studies (14, 28, 41). The neutralizing titer against D614G significantly decreased in sera of 6-8 months post onset compared to those of 1-3 months post onset ( Figure 3B ). Several studies have reported that the peak neutralizing antibody titer is three to five weeks post onset, and decreases rapidly, then is sustained at low level for several months (42–44). The rapid early decay was shown to be caused by the short half-life of serum antibodies and by the short life of antibody-secreting cells; the maintenance of neutralizing antibody titers was supported by long-lived plasma cells to produce long-term antibodies (45). Further follow-up will be needed to confirm whether the specific Nab titer against D614G is maintained or not.

Interestingly and surprisingly, Nab titers against the three VOCs did not decrease until 6-8 months post onset ( Figure 3B ), possibly indicating that Nabs that recognize common epitopes were produced after infection, were selected for survival and sustained for a long time, while Nabs that recognized specific epitopes for a variant were produced in greater numbers after infection and decreased rapidly. Our results may reflect the increasing neutralizing breadth of antibodies which recognize common epitopes among VOCs (46–48). Importantly, SARS-CoV-2 RNA and proteins were detected in intestinal enterocytes several months post onset (46). These persisting viral antigens may stimulate memory B cells continuously. Then, B cells that produce neutralizing antibodies targeting the common epitope among variants can undergo further maturation in germinal center and as a result, produce high-affinity antibodies several months later. B cells that produce neutralizing antibodies targeting the common epitope among variants can undergo further maturation in germinal center and as a result, produce high-affinity antibodies several months later. Simultaneously, low-affinity antibodies disappear over time (48). Similar to our study, other study also showed that sera from convalescent patients infected with the ancestral SARS-CoV-2 maintained the cross-neutralizing antibody titers against B.1.1.7, P.1 and B.1.351 variants (49). Additionally, they compared cross-neutralizing activity in sera from convalescent patients infected with the ancestral strain, B.1.1.7, B.1.351 or B.1.617.2 SARS-CoV-2 variants. It was shown that the sera of convalescent patients infected with the ancestral strain maintained higher level of cross-neutralizing antibody titers than those infected with the other VOCs. The increasing neutralizing potency and breadth will help us develop effective hyper immunoglobulin and monoclonal variant-resistant antibodies like sotrovimab, which targets the surface of the RBD not overlapping with the ACE2 binding site and neutralizes B.1.1.7, P.1, B.1.351, and other variants (2). Neutralizing antibodies that recognize a common epitope for variants may be kept for a long time. SARS-CoV-2 specific functional CD 4⁺ T cell has also an important role to help the long lived S-specific B cell to produce high-affinity antibodies (50, 51). Recent study showed that mild COVID-19 patients induced fewer but functionally superior B cell than critical patients with mechanical ventilation (52). The diversity of B cell might be brought by CD4⁺ T cell. Further follow-up would be required to clarify the mechanism to get the long-lived immunity and cross-neutralizing activity against VOCs.

We should be careful in interpreting the meaning of ND. It remains unclear how much Nab titer determined by our method is required to protect reinfection. In this study, we did not examine whether fragment crystallizable (Fc) portions worked to recruit immune cells or serum complement as effectors. Some studies have shown that the Fc-mediated effector function of neutralizing antibodies against SARS-CoV-2 was essential for optimal therapy (53, 54). Therefore, the recoverees with low titer or ND in our study may still be protected against reinfection or severe disease after infection. Furthermore, we need to evaluate not only humoral immunity but also cell-mediated immunity. Cell-mediated immunity might be obtained because few mutations in the T-cell epitope of VOCs are known (41, 55), although L452R mutation, which is present in some variants (B.1.167 and B.1.427/429), escapes from HLA-24 cell-mediated immunity (56). Therefore, even if the neutralization activity falls below the detection limit in the long term, convalescent COVID-19 patients might be protected from VOCs.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by The ethics committee of Kobe University Graduate School of Medicine (ID: B200200) and Hyogo Prefectural Kakogawa Medical Center. The patients/participants provided their written informed consent to participate in this study.

All authors contributed to the concept of this article. YK drafted the manuscript. MN and YM provided revisions. YK, JA, MN, and YM analyzed the data. YK, SSu, KF, and LT did the experiments. YM and YN supervised the experiments. SI, YT, SSa, SN, TK, MY, and TN collected the samples. YM conducted the project. All authors approved the final version of the manuscript.

This work was supported by Hyogo Prefectural Government and the Kansai Economic Federation (KANKEIREN). The funders had no role in this study.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.