Plasma after both SARS-CoV-2 boosted vaccination and COVID-19 potently neutralizes BQ.1.1 and XBB.1

Abstract

Objectives

Recent 2022 SARS-CoV-2 Omicron variants, have acquired resistance to most neutralizing anti-Spike monoclonal antibodies authorized, and the BQ.1.* sublineages are notably resistant to all authorized monoclonal antibodies. Polyclonal antibodies from individuals both vaccinated and recently recovered from Omicron COVID-19 (VaxCCP) could retain new Omicron neutralizing activity.

Methods

Here we reviewed BQ.1.* virus neutralization data from 920 individual patient samples from 43 separate cohorts defined by boosted vaccinations with or without recent Omicron COVID-19, as well as infection without vaccination.

Results

More than 90% of the plasma samples from individuals in the recently (within 6 months) boosted VaxCCP study cohorts neutralized BQ.1.1, and BF.7 with 100% neutralization of WA-1, BA.4/5, BA.4.6 and BA.2.75. The geometric mean of the geometric mean 50% neutralizing titers (GM (GMT₅₀) were 314, 78 and 204 for BQ.1.1, XBB.1 and BF.7, respectively. Compared to VaxCCP, plasma sampled from COVID-19 naïve subjects who also recently within 6 months received at least a third vaccine dose had about half of the GM (GMT₅₀) for all viral variants.

Conclusions

Boosted VaxCCP characterized by either recent vaccine dose or infection event within 6 months represents a robust, variant-resilient, passive immunotherapy against the new Omicron BQ.1.1, XBB.1 and BF.7 variants.

Introduction

In immunocompromised (IC) patients both passive immunotherapies and small molecule antivirals are often necessary to treat COVID-19 or eliminate persistently high SARS-CoV-2 viral load. Chronic, persistent viral loads increase both transmission and mutation risk, and prevent administration of the required immunosuppressive/antineoplastic therapies(1). Small molecule antivirals have not been formally validated for IC patients, who often have contraindications, and the convergent evolution of the Omicron variant of concern (VOC) has led to inefficacy of all the anti-Spike monoclonal antibodies (mAbs) authorized so far for both treatment or prevention, e.g. in the highly prevalent BQ.1.* sublineages(2). The other rapidly growing XBB.* and BF.7 sublineages are also highly resistant to anti-Spike mAbs(3). Polyclonal plasma from individuals who are both vaccinated and had COVID-19 (VaxCCP) has more than ten times the antibody levels capable of neutralizing pre-Omicron variants as well as Omicron variants BA.1 through BA.4/5(4, 5). Polyclonal COVID-19 convalescent plasma (CCP) has thousands of distinct antibody specificities of different isotypes, including many capable of SARS-CoV-2 neutralization. High-titer pre-Omicron CCP contains Omicron neutralizing activity despite being collected before variant appearance(4, 5).

Given that CCP remains a recommended therapy for IC(1, 6, 7), we systematically reviewed recent primary research for neutralization results against BQ.1.1 by plasma collected from vaccinated subjects with or without COVID-19 or after recent Omicron infection alone.

Results

Ten articles were included (Figure 1) which contained virus neutralizations with WA-1, BQ.1.1, BA.4/5, BA.4.6, XBB.1 and BF.7, assessed with either live authentic SARS-CoV-2 or SARS-CoV-2 pseudovirus neutralization assays and represented data from 920 patients (Supplementary Table 1). Qu et al. in the USA reported on Spring and Summer 2022 breakthrough infections with BA.1 and BA.4/5 in two sampled cohorts with predominantly unvaccinated individuals, as well as a third cohort of healthcare workers after a single monovalent booster vaccination in the Fall of 2021(8) (Table 1). Zou et al. in the USA in the Summer and Fall of 2022 sampled individuals who had already received 3 mRNA BNT162b2 vaccinations with or without previous COVID-19, both before and about 4 weeks after a 4^th monovalent or bivalent vaccine booster vaccination(9). Miller et al. also in the USA sampled both before the 3^rd vaccination dose and about 4 weeks after monovalent mRNA vaccination in the Fall of 2021, as well as with the 4th vaccine dose in the Summer or Fall of 2022, with either monovalent or bivalent booster vaccinations in Fall of 2022 in those with no documented COVID-19(10). Cao et al. in China investigated BQ.1.1 neutralizations from plasma of 4 cohorts after 3 doses of CoronaVac (Fall 2021) without COVID-19 or 2–12 weeks after BA.1, BA.2 and BA.5 infection(3). Planas et al. in France evaluated GMT₅₀in plasma from individuals both 4 and 16 weeks after a third monovalent mRNA vaccine dose in the Fall of 2021 as well as 12 and 32 weeks after vaccine breakthrough BA.1/2 or BA.5 infection(11). Davis et al in the USA sampled after the 3^rd mRNA vaccine monovalent dose in the Fall of 2021 and also after either a 4th monovalent mRNA dose or a bivalent (wild-type + BA.4/5) vaccine dose in the Summer and Fall of 2022(12). Kurhade et al in the USA also compared GMT₅₀after the 4^th monovalent vaccine dose or 3 mRNA doses with the 4^th the bivalent dose without COVID-19 and also after bivalent boost with recent COVID-19(13). Wang et al in the USA compared GMT₅₀ after three vaccine doses, the 4^th monovalent vaccine dose or 3 mRNA doses with the 4^th the bivalent dose without COVID-19, and also after 2–3 vaccine doses and recent BA.2 breakthrough infection or 3–4 mRNA vaccine doses and recent BA.4/5 breakthrough infection(14). Ito et al in Japan compared breakthrough infections after BA.2 and BA.5 after 2–3 doses of mRNA vaccines in the Spring and Summer of 2022(15). Akerman et al in Australia characterized neutralizing antibodies in four groups 1) sampling one to three months after 3 doses of mRNA vaccines with an Omicron infection in 2022; 2) sampling 3 months after 4 doses of mRNA vaccine; 3) sampling 6 months after 3 doses of mRNA vaccine and 4) sampling 3–6 months after last vaccine in a larger cohort who had the original WA-1 infection in early 2020 as well as 3 more doses of mRNA vaccine(16).

Figure 1.

PRISMA flowchart for the current study. Number of records identified from various sources, excluded by manual screening, found eligible and included according to subgroup analyses.

Table 1.

Synopsis of included studies, reporting plasma sources, epoch of sampling, region, time since vaccination/infection to plasma sampling, and sample size. The cohorts were split into three groups-1) boosted vaccinations and recent COVID-19 (VaxCCP), 2) boosted vaccines only without documented COVID-19 (Vac only) and 3) infection alone or pre-boosted sampling before 3^rd or 4^th vaccine dose (Infection only or pre-boost)

Study	Vaccine and COVID-19 history at sample time	Group	Time period of plasma sampling	Geography	Sampling time mean or median (min-max)	Sample number
Cao(3)	3xCorVac+BA.1 inf	VaxCCP	Spring 2022	China	5–7 weeks post hosp admit (42 weeks avg)	50
Cao(3)	3xCorVac +BA.2 inf	VaxCCP	Summer 2022	China	3–11 weeks post hosp admit (8 weeks mean)	39
Cao(3)	3xCorVac +BA.5 inf	VaxCCP	Summer/Fall 2022	China	2–11 weeks (mean 5 weeks)	36
Zou(9)	4xBNT162b2+BTI	VaxCCP	Summer/Fall 2022	USA	4 weeks post dose	20
Planas(11)	mRNAvacx3+BA.1/2 inf	VaxCCP	Spring/Fall 2022	France	32 weeks post BTI BA.1/2	13
Wang(14)	2–3xmRNAvac+BA.2 BTI	VaxCCP	Spring/Fall 2022	USA	2–23 weeks (mean 6 wk (3over 90 days))	14
Wang(14)	3–4xmRNAvac+BA.4/5 BTI	VaxCCP	Summer/Fall 2022	USA	2–8 weeks (mean 4 weeks)	20
Kurhade(13)	3xmRNAvac+bivalent+BTI	VaxCCP	Fall 2022	USA	4 weeks post with infection history	23
Ito(15)	2–3xmRNAvac+BA.2 BTI	VaxCCP	Spring 2022	Japan	2–8 weeks	14
Ito(15)	2–3xmRNAvac+BA.5 BTI	VaxCCP	Summer 2022	Japan	2–3 weeks	20
Zou(9)	3xBNT162b2+bivalent+BTI	VaxCCP	Summer/Fall 2022	USA	4 weeks post dose	19
Akerman(16)	3xmRNA +bivalent	VaxCCP	Fall 2022	Australia	4–12 weeks post BTI	29
Planas(11)	3xmRNAvac+BA.1/2 inf	VaxCCP	Spring/Fall 2022	France	12 weeks post BTI BA.1/2	16
Planas(11)	3xmRNAvac+ BA.5 inf	VaxCCP	Fall 2022	France	8 weeks post BTI BA.5	15
Davis(12)	3xmRNAvac	Vac only	Fall 2021	USA	1–4 weeks post boost	12
Kurhade(13)	4xmRNAvac	Vac only	Summer 2022	USA	4–12 weeks	25
Cao(3)	3xCorVac	Vac only	Fall 2021	China	4 weeks	40
Zou(9)	4xBNT162b2	Vac only	Summer/Fall 2022	USA	4 weeks post dose	20
Planas(11)	3xmRNAvac	Vac only	Winter 2021/2022	France	16 weeks post 3rd dose	10
Wang(14)	3xmRNAvac	Vac only	Fall 2021	USA	2–12 weeks (mean 5 weeks)	14
Wang(14)	3xmRNAvac+monovalent	Vac only	Summer/Fall 2022	USA	3–4 weeks	19
Wang(14)	3xmRNAvac+bivalent	Vac only	Summer/Fall 2022	USA	3–4 weeks	21
Davis(12)	3xmRNAvac+monovalent	Vac only	Summer/Fall 2022	USA	10–15 weeks post boost	12
Akerman(16)	4xmRNA	Vac only	Fall 2022	Australia	12 weeks	23
Akerman(16)	3xmRNAvac after 2020 WA-1	Vac only	Summer/Fall 2022	Australia	3–6 months	47
Kurhade(13)	3xmRNAvac+bivalent	Vac only	Fall 2022	USA	4 weeks post	29
Davis(12)	3xmRNAvac+bivalent	Vac only	Summer/Fall 2022	USA	2–6 weeks post booster (8 no vacc. 10 no infection)	12
Qu(8)	3xmRNAvac	Vac only	Fall 2021	USA	2–13 weeks	15
Zou(9)	3xBNT162b2+bivalent	Vac only	Summer/Fall 2022	USA	4 week post dose	18
Planas(11)	3xmRNAvac	Vac only	Fall/Winter 2021	France	4 weeks post 3rd dose	18
Miller(10)	3xBNT162b2	Vac only	Fall 2021	USA	2–4 weeks	16
Miller(10)	3xmRNA+monovalent	Vac only	Spring/Fall 2022	USA	2–9 weeks	18
Miller(10)	3xmRNA +bivalent	Vac only	Fall 2022	USA	2–3 weeks	15
Qu(8)	BA.4/5 inf (17-unvac)	Inf only	Summer 2022	USA	not stated	20
Qu(8)	Hosp BA.1 (6-unvac;5-2xmRNAvac)	Inf only	Spring 2022	USA	1 week post hospitalization	15
Zou(9)	3xBNT162b2+BTI	preboost with BNT162b	Summer/Fall 2022	USA	preboost with BNT162b (6–11 months post last dose)	20
Zou(9)	3xBNT162b2 +BTI	preboost with bivalent	Summer/Fall 2022	USA	preboost with bivalent (6–11 months post last dose)	19
Zou(9)	3xBNT162b2	preboost with bivalent	Summer/Fall 2022	USA	preboost with bivalent (6–11 months post last dose)	18
Zou(9)	3xBNT162b2	preboost with BNT162b	Summer/Fall 2022	USA	preboost with BNT162b (6–11 months post last dose)	20
Akerman(16)	3xmRNA	Preboost	Fall 2022	Australia	preboost (6 months)	47
Miller(10)	2xBNT162b2	preboost with BNT162b	Fall 2021	USA	preboost (6–11 months post last dose)	16
Miller(10)	3xmRNA	preboost with bivalent	Fall 2022	USA	preboost with bivalent (6–11 months post last dose)	15
Miller(10)	3xmRNA	preboost with monovalent	Spring/Fall 2022	USA	preboost with monovalent (6–11 months post last dose)	18

These diverse cohorts were assembled into 3 groups, 1) plasma after both 2–4 vaccine doses and COVID-19 (VaxCCP); 2) plasma from subjects after administration of 3–4 vaccine doses (i.e. boosted), but either self-reported as COVID-19-naïve or anti-nucleocapsid negative; and 3) Omicron infection without vaccination (CCP) as well as participants sampled 6 to 11 months after previous vaccine dose and before the booster vaccination. Boosted VaxCCP neutralized BQ.1.1, XBB.1 and BF.7 with approximately 3 times the dilutional potency of the vaccine-only or 2–6 times CCP/pre-boost vaccination groups for all viral variants (Table 2 and Figure 2). Importantly, while there was a 19-fold reduction in neutralization by boosted VaxCCP against BQ.1.1 compared to WA-1, more than 90% of the boosted VaxCCP samples neutralized BQ.1.1 as well as XBB.1 and BF.7 (Table 2 and Figure 2c). Three cohorts within the boosted VaxCCP group were below at 90% neutralization with one sampled late, 8 months after BA.1/2 breakthrough infection(11) and the other two from a single study after BA.2 and BA.5 (Supplementary Table 2 and 3). Except for the GMT (GMT₅₀) against XBB.1 at 78, the other viral variant neutralizations were in the same range as pre-Alpha CCP neutralizing WA-1 (i.e., 311)(4). By comparison the large randomized clinical trial which effectively reduced outpatient COVID-19 progression to hospitalizations had a GMT₅₀ of 60 for WA-1 with pre-Alpha CCP(17). Boosted vaccinations at 3–4 doses without COVID-19, showed GM (GMT₅₀) of 118 for BQ.1.1, with only 6 of 23 cohorts over 90% neutralizations, for 79% overall (i.e. 326 of 414 individuals). Four separate studies(8),(13),(12),(10) characterized BQ.1.1 virus neutralizations with plasma after the new bivalent (wild-type + BA.4/5) mRNA vaccine booster in the Fall of 2022, with 88% (103 of 117 samples) neutralization activity within 4 weeks of bivalent booster (Supplementary Table 3).

Table 2.

GM (GMT₅₀) of plasma from three different sources against recent Omicron sublineages.

Neutralization virus	WA-1	BQ.1.1	BA.4/5	BA.4.6	BA.2.75	XBB.1	BF.7
Post COVID-19/vaccine (study cohorts)	12	16	14	7	9	9	4
GM (GMT50)	5876*	314	987	346	303	78	204
Fold reduction from WA-1	ref	19	6	17	19	75	32
Samples tested	294	237**	328	135	231	125	148
Samples neutralizing	285	221	326	135	230	111	146
Percent neutralizing	97	93	99	100	100	89	99
Boosted vaccine (study cohorts)	19	23	19	10	14	10	7
GM (GMT50)	3766	118	346	126	107	52	357
Fold reduction from WA-1	ref	32	11	30	35	72	12
Samples tested	384	414	384	206	261	217	158
Samples neutralizing	383	326	363	191	231	125	149
Percent neutralizing	100	79	95	93	89	58	94
Infection only or preboosted vaccine (study cohorts)	10	12	10	7	9	5	5
GM (GMT50)	870	47	96	103	57	23	110
Fold reduction from WA-1	ref	19	9	8	15	38	9
Samples tested	184	220	184	136	162	101	84
Samples neutralizing	182	139	144	104	104	50	63
Percent neutralizing	99	63	78	76	64	50	75

^*Pre-Alpha CCP from 27 different studies had a GM (GMT₅₀) of 311 from 707 samples with 315 or 45% neutralizing omicron BA.1(4).

^**percent neutralizations after CoronaVac and Omicron COVID-19 in the paper by Cao et al could not be retrieved from the manuscript. 237 samples from the 6 other cohorts were used for percent neutralization.

Figure 2.

Neutralizing GMT (GMT₅₀) against WA-1, BQ.1.1, BA.4/5, BA.4.6, BA.2.75, XBB, BF.7. A) post boosted vaccinations and COVID-19 and B) boosted vaccinated plasma without COVID-19. Geometric standard deviation for error bars, fold reduction (FR) below data, and number of studies above x-axis. Geomeans statistically significant in difference by multiple comparison in Tukey’s test are indicated. C) The percent of total samples within a study which neutralized Omicron BQ.1.1 graphed in increasing percentages on left y axis with the total number of samples tested on the right y axis.

Many studies performed virus neutralizations on samples drawn before the 3^rd or 4^th vaccine dose which were 6 to 11 months after last vaccine dose. The GM (GMT₅₀)’s for BQ.1.1 and BA.2.75 were about 6 times reduced compared to VaxCCP even though the fold reductions were similar (Figure 3, Table 2). In agreement with lower GMT₅₀ for neutralizations was the low percent neutralizing BQ.1.1 (63%), XBB.1 (50%), and BF.7 (75%) at 6 to 11 months after vaccination (Figure 3, Table 2 and Supplementary Table 3).

Figure 3.

Geometric mean neutralizing titers (GMT₅₀) against WA-1, BQ.1.1, BA.4/5, BA.4.6, BA.2.75, XBB, BF.7 A) plasma Omicron infection alone or pre-boosted-6 to 11 months after last vaccine dose sampled in 2021 or 2022. Geometric standard deviation for error bars, fold reduction (FR) above data, and number of studies above x-axis. GM (GMT₅₀) statistically significant in difference by multiple comparison in Tukey’s test are indicated. B). The percent of total samples within a study which neutralized Omicron BQ.1.1 graphed in increasing percentages on left y-axis with the total number of samples tested on the right y-axis.

Five studies used the lentiviral pseudovirus assays, with diverse Spike proteins cloned in, while the other four were live virus assays using different cell types (Supplementary Table 1). Notably, Planas et al employed the IGROV-1 cell type for better growth of Omicron sublineages(11). While the single study fold reductions (FR) and percent neutralizations normalize the results between studies, the GMT₅₀ can vary between studies even amongst the live authentic viral neutralization studies (e.g., mNeonGreen^™ reporter assays versus cytopathic effects)(9, 13). We sorted the live authentic viral neutralizations from the pseudoviral neutralizations, plotting also the minimum and maximums (Supplementary Figures 1–3). In general, the live authentic SARS-CoV-2 neutralization assays for VaxCCP appeared to have similar antibody neutralization levels, with the single study by Cao et al(3) employing lentiviral pseudovirus with lower dilutional titers. In contrast, the GMT₅₀ achieved with pseudoviral assays in the boosted vaccinations without COVID-19 appeared slightly higher than the ones achieved with authentic virus.

Discussion

The FDA deemed CCP safe and effective for both immunocompetent and IC COVID-19 outpatients(6, 7, 18), and further extended its authorized use in the IC patient population in December 2021(7, 18), at a time when oral antiviral therapy promised a no transfusion outpatient solution and many anti-Spike mAbs were still effective.

Up until the present, CCP remained a backup bridge for IC patients, durable against the changing variants and as a salvage therapy in seronegative IC patients. With the recent advent of Omicron XBB.* and BQ.1.* defeating the remaining anti-Spike mAbs, boosted VaxCCP, recently collected within the last 6 months of either a vaccine dose or SARS-CoV-2 is likely to be the only viable remaining passive antibody therapy in the 2022–23 Winter for IC patients who have failed to make antibodies after vaccination and still require B-cell depleting drugs or immunosuppressive therapy. In a literature review of CCP from diverse VOC waves as well as boosted vaccinees and VaxCCP up to BA-1, VaxCCP showed higher neutralization titers against Omicron at levels above 300 dilutional GMT₅₀⁴.

The accelerated evolution of SARS-CoV-2 VOCs has created the problem that the pharmaceutical development of additional mAbs is not worth the effort and cost given their expected short useful clinical life expectancy, so the anti-Spike mAb pipeline has remained stuck in 2022. High levels of antibodies in donor plasma from both boosted vaccinations and COVID-19 convalescent plasma (VaxCCP) neutralizing more than 93% of BQ.1.1and BF.7, with XBB.1 at 89%. Recently collected plasma within a 6 months window from those boosted vaccinees without prior documented COVID-19 had a 20–30% reduction in neutralization percent for BQ.1.1and XBB.1 with 10% reduction for the others and a third of the GM (GMT₅₀) neutralizing antibody levels compared to VaxCCP. In those vaccinated with last dose more than 6 months prior to sample collection, both the neutralization percent and neutralizing antibody titers fell further, compared to the recently boosted VaxCCP group. Four studies (Planas(11), Zou(9), Cao(3) and Kurhade(13)) had directly comparative cohorts in the three groups which increases the robustness reduction in neutralizations with the vaccine only or more than 6 months to last vaccine or infection event compared to VaxCCP. The main limitation of our systematic review is the small number of studies reporting virus neutralization with BQ.1.1 with most available as pre-preprints without peer-review yet. However, we note that peer-review itself does not change GMT₅₀ or neutralization numbers and the authors of these papers have considerable expertise in the topic.

Boosted VaxCCP has full potential to replace anti-Spike mAbs for passive antibody therapy of IC patients against recent Omicron sublineages, in the meanwhile polyclonal IgG formulations can be manufactured. VaxCCP qualification in the real-world will likely remain constrained on high-throughput serology, whose correlation with GMT₅₀ is not perfect(19, 20). Nevertheless, the very high prevalence (93%) of Omicron-neutralizing antibodies and the high GM (GMT₅₀) in recently boosted VaxCCP reassure about its potency, and further confirm that exact donor-recipient VOC matching is dispensable. Overall, our findings urge WHO to revise its guidelines and recommend boosted VaxCCP for therapy of COVID-19 in IC patients.

Search strategy and selection criteria

On November 19, 2022 we initially searched PubMed, medRxiv and bioRxiv for manuscripts reporting BQ.1.1 neutralization, using English language as a restriction. Search of bioRxiv with same keywords now yields 17 records of which only 10 contained plasma viral neutralization data. Search of medRxiv produced 3 records which did not have BQ.1.1 neutralizations. PubMed retrieved 3 entries using (“BQ.1.1”) and (“neutralization”), one of which was focused on anti-Spike mAb alone(2) and the other 2 were duplicates from bioRxiv(8, 12). Articles underwent evaluation for data extraction by two assessors (DS and DF) with disagreements resolved by third assessor (AC). Articles lacking plasma BQ.1.1 virus neutralizations were excluded. The process of study selection is represented in the PRISMA flow diagram (Figure 1).

The type of viral assay (live or pseudovirus), time interval to blood sample, GMT₅₀, minimum and maximum neutralizing 50% dilutional titer for WA-1 (pre-Alpha wild-type) and Omicron sublineages BQ.1.1, BA.4/5, BA.4.6, BA.2.75, XBB.1 and BF.7 and number out of total that neutralized Omicron were abstracted from study text, graphs and tables. Two studies (Wang(14) and Qu(8)) reported BQ.1 and those were separate cohorts in addition to BQ.1.1. Prism v. 9.4 (GraphPad Software, San Diego, CA, USA) was used for data analysis. While all manuscripts included neutralization data against WA-1, BQ.1.1, BA.4/5 and BA.2.75, only a subset of manuscripts included neutralization data for BA.4.6, XBB.1 and BF.7 which were assembled for relevance to present circulating variants. Historic early Omicron partial neutralization data on variants like BA.1 or BA.2 were excluded because of the full set data with BA.4/5 and BA.2.75.

Statistical significance between log₁₀ transformed GMT₅₀ was investigated using Tukey’s test. The multiple comparison test was a two-way ANOVA with Alpha 0.05 on log transformed GMT₅₀. The log normal test was performed on WA-1, BQ.1.1, BA.4/5, BA.4.6, XBB.1 and BF.7 virus GMT₅₀. Two studies(10, 11) reported the median titer rather than the GMT₅₀. Compiled data abstracted from the published studies is available in the supplementary dataset.

Abbreviations:

VOC

variant of concern

VaxCCP

plasma from both vaccinated and COVID-19 convalescent subjects

Data availability statement:

all the data used for this manuscript are available in the Supplementary Dataset file.