key: cord-0323485-li5w01ey
authors: Vargas, L.; Valdivieso, N.; Tempio, F.; Simon, V.; Sauma, D.; Valenzuela, L.; Beltran, C.; Castillo-Delgado, L.; Contreras-Benavides, X.; Acevedo, M. L.; Valiente-Echeverria, F.; Soto-Rifo, R.; Gonzalez, R. I.; Lopez, M.; Osorio, F.; Bono, M. R.
title: Serological study of CoronaVac vaccine and booster doses in Chile: immunogenicity and persistence of anti-SARS-CoV-2 S antibodies
date: 2022-01-23
journal: nan
DOI: 10.1101/2022.01.14.22269289
sha: 30e1bf972bd8452adbfc4bfebe7974386be1e365
doc_id: 323485
cord_uid: li5w01ey

ABSTRACT Background: Chile was severely affected by COVID19 outbreaks but was also one of the first countries to start a nationwide program to vaccinate against the disease. Furthermore, Chile became one of the fastest countries to inoculate a high percentage of the target population and implemented homologous and heterologous booster schemes in late 2021 to prevent potential immunological waning. The aim of this study is to compare the immunogenicity and time course of the humoral response elicited by the CoronaVac vaccine in combination with homologous versus heterologous boosters. Methods and Findings: We compared the immunogenicity of two doses of CoronaVac and BNT162b2 vaccines and studied the effect of different booster regimes in the Chilean population. Our results demonstrate that a two-dose vaccination scheme with CoronaVac induces lower levels of anti-SARS-CoV-2 S antibodies than BNT162b2 in a broad age range. Furthermore, antibody production declines with time in individuals vaccinated with CoronaVac and less noticeably, with BNT162b2. Remarkably, analysis of booster schemes revealed that individuals vaccinated with two doses of CoronaVac generate immunological memory against the SARS-CoV-2 ancestral strain, which can be re-activated with homologous or heterologous (BNT162b2 and ChAdOx1) boosters. Nevertheless, the magnitude of the antibody response with the heterologous booster regime was considerably higher and persistent (over 100 days) than the responses induced by the homologous scheme. Conclusions: Two doses of CoronaVac induces antibody titers against the SARS-CoV-2 ancestral strain which are lower in magnitude than those induced by the BNT162b2 vaccine. However, the response induced by CoronaVac can be greatly potentiated with a heterologous booster scheme with BNT162b2 or ChAdOx1 vaccines. Furthermore, the heterologous booster regimes induce a durable antibody response which does not show signs of decay 3 months after the booster dose.

Background: Chile was severely affected by COVID19 outbreaks but was also one of the first 82 countries to start a nationwide program to vaccinate against the disease. Furthermore, Chile 83 became one of the fastest countries to inoculate a high percentage of the target population and 84

implemented homologous and heterologous booster schemes in late 2021 to prevent potential 85 immunological waning. The aim of this study is to compare the immunogenicity and time course 86 of the humoral response elicited by the CoronaVac vaccine in combination with homologous 87 versus heterologous boosters. 88

Methods and Findings: We compared the immunogenicity of two doses of CoronaVac and 89

BNT162b2 vaccines and studied the effect of different booster regimes in the Chilean population. 90 Our results demonstrate that a two-dose vaccination scheme with CoronaVac induces lower 91 levels of anti-SARS-CoV-2 S antibodies than BNT162b2 in a broad age range. Furthermore, 92

antibody production declines with time in individuals vaccinated with CoronaVac and less 93 noticeably, with BNT162b2. Remarkably, analysis of booster schemes revealed that individuals 94 vaccinated with two doses of CoronaVac generate immunological memory against the SARS-CoV-95 2 ancestral strain, which can be re-activated with homologous or heterologous (BNT162b2 and 96 ChAdOx1) boosters. Nevertheless, the magnitude of the antibody response with the 97 heterologous booster regime was considerably higher and persistent (over 100 days) than the 98 responses induced by the homologous scheme. 99

Conclusions: Two doses of CoronaVac induces antibody titers against the SARS-CoV-2 ancestral 100 strain which are lower in magnitude than those induced by the BNT162b2 vaccine. However, the 101 response induced by CoronaVac can be greatly potentiated with a heterologous booster scheme 102 with BNT162b2 or ChAdOx1 vaccines. Furthermore, the heterologous booster regimes induce a 103 durable antibody response which does not show signs of decay 3 months after the booster dose. 104 105 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101/2022.01.14.22269289 doi: medRxiv preprint

Chile is one of the several countries severely threatened by the COVID-19 pandemic in 2020, but 108 that had prompt access to vaccines for a large number of individuals since early 2021. The first 109 SARS-CoV-2 vaccine authorized in Chile for emergency use by the Health Ministry ( (https://covid19.trackvaccines.org/agency/who/), which is currently administered in 48 114 countries (https://covid19.trackvaccines.org/vaccines/7/). In Chile, vaccination with CoronaVac 115 began on February 1, 2021, with people over 55 years old, people with specific pathologies, and 116 essential services personnel. Progressively, the vaccination scheme extended to younger people 117 (target population over 18 years old: 15,200,840) . In this first phase of vaccination, the 118

CoronaVac vaccine was predominantly used across the population. Real-world data indicated 119 that the two-dose vaccination scheme with CoronaVac in Chile showed a 65.9% vaccine 120 effectiveness, 90.3% for prevention of ICU admission, and 86.3% for prevention of COVID-19 121 related death (1). To date, more than 86,8% of the Chilean population received their complete 122 vaccination schedule with any available vaccines (DEIS/MINSAL), and about 77% of the target 123 population received CoronaVac (Minsal / Deis). 124 However, around mid-2021, immunological studies reported a decline of antibody levels in 125 vaccinated individuals. These studies predicted a reduction in antibody titers directed against 126 SARS-CoV-2 over time, highlighting the requirement of an additional immunization (2) (3). In this 127 context, a group of countries, including Israel (4), and Chile authorized a booster vaccine dose. 128

On August 11, 2021, the vaccination with booster doses began for people who had received two 129 doses of Coronavac in Chile. Interestingly, Chile implemented a heterologous booster schedule 130 for most individuals including BNT162b2 and the ChAdOx1 vaccine from AstraZeneca as the most 131 used boosters. These schemes offer an important opportunity to assess the magnitude of the 132 immunological response to homologous and heterologous boosters schedules within the same 133 population. Furthermore, this issue is relevant considering that immunological studies of 134 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 heterologous booster schedules using CoronaVac as the first immunization vaccine have not been 135 extensively documented. 136

This study describes the production of IgG antibodies directed against the ancestral SARS-CoV-2 137 S protein induced by the two-dose scheme of the Coronavac vaccine in a Health Service of the 138 Hospital La Florida, Santiago. Our data shows that detectable levels of specific antibodies appear 139 in most vaccinated individuals. By comparing the humoral responses to CoronaVac and 140 BNT162b2 vaccines over time, we found that the antibody production elicited by CoronaVac 141 declined six months after vaccination, whereas people vaccinated with two doses of BNT162b2 142 maintained a noticeably higher level of antibodies over time. Next, we analyzed the impact of 143 the booster doses of CoronaVac, BNT162b2, or ChAdOx1 vaccines, administered to individuals 144 vaccinated with the two-dose scheme with CoronaVac six months earlier. Our data show that the 145 three types of boosters produce a noticeable increase in anti-spike IgG antibody production 146 twenty days after the booster administration, which was more strongly noticed in individuals 147 vaccinated with the heterologous booster regime. Antibody responses measured 100 days after 148 the booster dose revealed that the heterologous regime induced higher and persistent anti-SARS-149

CoV-2 S antibodies compared to the homologous regime. 150

In summary, our results show that the CoronaVac vaccine produces memory against the SARS-151

CoV-2 that can be greatly potentiated with a heterologous booster strategy. Moreover, the 152 persistent antibody titers obtained using the heterologous booster strategy may allow to space 153 subsequent booster doses in the population. Furthermore, these data suggests that Chile's 154 vaccination scheme has been efficient in avoiding contagion with the Delta variant, as predicted 155 by data derived from the epidemic in Chile. 156 This study is composed of four groups covering the period ranging from the beginning of the 183 vaccination program in February 2021 and months after the administration of the booster doses 184

in August 2021 (depicted as timelines in Figure 1A ). Group 1 corresponds to 104 individuals 185 belonging to the clinical staff from the Hospital Clinico Metropolitano La Florida "Dra. Eloisa 186

Diaz", which were among the first cohort to be vaccinated as a priority group CoronaVac vaccine. 187

In this group of individuals, the antibody response to the first and second dose of the CoronaVac 188 vaccine was assessed. 189 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101/2022.01.14.22269289 doi: medRxiv preprint Group 2 corresponds to 158 individuals from a broad range of age vaccinated with CoronaVac 190 and BNT162b2 vaccines. A comparison of IgG production against spike SARS-CoV-2 protein 191 induced by the vaccines was performed, and antibody evolution was followed over time. The ELISA was performed as detailed (5), and adapted from the group of Kramer (6). Briefly, 96-212 well ELISA plates were coated overnight at 4°C with 50 µl per well of a 2 µg/ml solution of 213 resuspended SARS-CoV-2 Spike protein (Recombinant SARS-CoV-2 S protein S1 from the original 214

Wuhan SARS-CoV-2 virus, Biolegend 796906) on PBS. Then, the coating solution was removed, 215 and the wells were blocked for one hour at room temperature with 150 µl of 3% skim milk 216 prepared in PBS-0.1% Tween-20 (TPBS). After this period, 100 µl per well of serial dilutions (from 217 1/200 to 1/1,600) of the sera prepared in 1% skim milk in 0.1% TPBS was added and incubated 218 for 2 hours at room temperature. The plates were washed three times, added 100 µl per well of 219 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101/2022.01.14.22269289 doi: medRxiv preprint HRP-conjugated anti-human IgG (HRP Donkey anti-human IgG Clone: Poly24109, Biolegend), and 220 incubated for 1 hour at room temperature. The plates were washed three times, after which 50 221 µl of TMB substrate solution (BD Biosciences) was added per well to reveal the reaction, which 222 was stopped by adding 50 µl per well of 1M orthophosphoric acid. Optical density at 450 nm was 223 measured on a Molecular Devices Emax ELISA plate reader. We tested the specificity of our ELISA 224 assay by analyzing serum samples from unvaccinated COVID19 patients at the time where Delta 225 variant was dominant in our country. Our data confirmed that the ELISA test we performed with 226 the S protein of the original coronavirus recognizes all the variants that have entered Chile at that 227 time, including the Delta variant. 228

Neutralization assay 230 231

HIV-1-based SARS-CoV-2 pseudotype production: Pseudotyped viral particles were produced by 232 transient transfection of HEK293T cells using polyethylenimine (PEI) and plasmids pNL4.3-DEnv-233

Firefly and pCMV14-3X-Flag-SARS-CoV-2 SD19C (lineage A) in a 1:1 ratio as we described (7). The 234 viral particles were diluted with 50% in fetal bovine serum (Sigma-Aldrich) and stored at -80°C. 235

Viral stock was quantified with the HIV-1 Gag p24 Quantikine ELISA kit (R&D Systems). 236

Neutralizations assays were performed as we previously reported (7). Briefly, inactivated serum 237 samples were diluted in DMEM with 10% fetal bovine serum (serial dilutions from 1:4 to 1:8748) 238 and incubated with 5 ng of p24 HIV-1-based SARS-CoV-2 pseudotyped particles during 1h at 37°C, 239 and 1x10 4 HEK-ACE2 cells were added to each well. HEK293T cells incubated with the 240 pseudotyped virus were used as a negative control. Cells were lysed 48 hours later, and firefly 241 luciferase activity was measured using the Luciferase Assay Reagent (Promega) in a Glomax 96 242

Microplate luminometer (Promega). Then the percentage of neutralization for each dilution was 243 calculated as previously described. All statistical analyses were performed using GraphPad Prism 244 version 8.0.1 (7) 245 246

Quantification and statistical analysis: 247 248

For the ELISA assay, the background value was established at OD 0.100, and Area Under the Curve 249 (AUC) was calculated from serum dilutions. To obtain a correlation between AUC and antibody 250 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101/2022.01.14.22269289 doi: medRxiv preprint titers, we used estimated values of antibody titers from 212 samples, and we established a curve 251 according to Padé's approximation (with R2=0.9636). Differences between clinical groups were 252 calculated using a one-way ANOVA with Freedman or Kruskal-Wallis test followed with Dunn's 253 multiple comparations test. Differences between the two groups were calculated using the 254 unpaired two-tailed t-test or Mann-Whitney test. Simple linear regression was performed, and 255 correlations were analyzed by calculating nonparametric Spearman's correlation. Statistical 256 analyses were performed using GraphPad Prism 9.1.0, and statistical significance was 257 represented by *p<0.05, **p<0.01, ***p<0.001, and **p<0.0001. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101/2022.01.14.22269289 doi: medRxiv preprint

The To evaluate the effect of the CoronaVac vaccine on antibody titers in individuals potentially 299 exposed to the SARS-CoV-2 virus, we first focused our study on clinical staff from group 1, who 300 treated COVID-19 patients in the first wave of the disease in Chile. We analyzed the serum of 301 these individuals by ELISA to detect IgG antibodies directed against the Spike (S) protein of the 302 SARS-CoV-2 virus. This test was developed with samples from hospitalized COVID-19 patients as 303 positive controls (13 samples) and pre-pandemic or negative samples (54 samples) for negative 304 controls (5) and developed as reported (6). Sera were diluted serially from 1/200 to 1/1,600, and 305 the area under the curb (AUC) was determined. These values were equivalent to the antibody 306 titer (see Methods). We established the negative limit of the test (AUC = 70 ± 51) from the 307 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101/2022.01.14.22269289 doi: medRxiv preprint analysis of 54 samples from people who had no history of COVID-19. We considered AUC values 308 between 120 and 300 as a weak response in the ELISA test. In contrast, an AUC of around 300 309 corresponds approximately to an antibody titer of 1/1,000. To analyze the SARS-CoV-2 antibody 310 response course in this group, we analyzed the antibody response in three-time points. The first 311 serum sample was obtained 1-3 days before the first dose of the vaccine (referred to as 312 Fig 2) ; the second sample was obtained 1-3 days before the second immunization 313 (referred to as 'First dose + 30d' , Fig 2) , and the third sample was collected one month after the 314 second dose (referred to as 'Second dose +30d' , Fig 2) . Regarding previous SARS-CoV-2 infection, 315 the individuals who participated in this study were laboratory staff, and primary clinical 316 caregivers in contact with COVID-19 patients. Many individuals in this group reported not 317 knowing whether they had been exposed to SARS-CoV-2 since they could have experienced the 318 asymptomatic disease. 319

Of a total of 104 people tested, 18 had high antibody titers (AUC> 300) before being vaccinated, 320

suggesting that these individuals were infected with SARS-CoV-2 in the first pandemic wave (Fig.  321 2A, empty circles). Of these 18 individuals, only two did not improve antibody titers with 322 vaccination (Fig. 2B orange lines) . For the remaining 16 individuals (orange lines), the first dose 323 of the vaccine led to an increase in anti-S IgG production. Interestingly, there were no statistical 324 differences when comparing the level of antibodies induced by the first and the second dose of 325 the vaccine (Fig. 2B, orange circles) . 326 For individuals who initially had an AUC>120 (weak positive reaction, Fig 2B green circles) , the 327 first dose showed an increase in the level of anti-S IgG. Although significant, there was a mild 328 difference between the first and the second doses. Interestingly, the group who initially had an 329 AUC<120 (negative reaction, Fig 2B blue circles) showed remarkable differences between the 330 first and the second dose of the vaccine. Of the total 104 people, only one person remained 331 unresponsive to the two doses of the vaccine. As such, we conclude that the two-dose 332 vaccination scheme with CoronaVac induces a good antibody response against SARS-CoV2, which 333 is particularly noticed in individuals who have not been previously exposed to the virus. 334 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101/2022.01.14.22269289 doi: medRxiv preprint Next, the amount of neutralizing antibodies from 34 samples obtained one month after the 335 second dose was determined. The results show a significant positive correlation between the 336 AUC values and the IC-50 of neutralizing antibodies (Fig. 2C) . These results demonstrate that the 337 CoronaVac vaccine induces the production of neutralizing antibodies. Furthermore, this data 338 suggests that high titers of total antibodies should represent a greater probability of having 339 neutralizing antibodies against the virus. vaccine, respectively (group 2). Figure 3A shows a comparison of the data from both vaccines in 349 individuals raging from 18-87 years old (IQR: 27-61 years). We observed that the BNT162b2 350 vaccine induces significantly higher antibody production than the CoronaVac vaccine (2060 ± 361 351 for BNT162b2 and 1041 ± 520 for CoronaVac). Given that people vaccinated with CoronaVac were 352 mainly older than 55 years in Chile and those vaccinated with BNT162b2 were people between 353 18 and 54 years old, we compared and plotted antibody production according to the age of the 354 individuals and the type of vaccine they received. Figure 3B shows a significant negative 355 correlation (p = 0.032, black circles) for antibody production with increasing age for the 356 CoronaVac vaccine. In contrast, a similar (but not statistically significant, green circles) trend is 357 shown for the BNT162b2 vaccine. These results show that the BNT162b2 vaccine induces twice 358 the amount of IgG against SARS-CoV-2 S protein compared to CoronaVac, independent of the age 359 of the individuals. 360

Overtime evolution of the humoral response to CoronaVac and BNT162b2 vaccines 362 363 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101/2022.01.14.22269289 doi: medRxiv preprint

So far, we have demonstrated the presence of neutralizing antibodies in a significant number of 364 individuals immunized with CoronaVac and demonstrated a positive correlation between the 365 amount of IgG against SARS-CoV-2 S antibodies and the production of neutralizing antibodies 366 (Fig. 2C) . Moreover, we showed that the BNT162b2 vaccine produces higher levels of antibodies 367 in vaccinated people than those elicited by the CoronaVac vaccine (Fig. 3A) . Therefore, we sought 368 to determine how antibody levels vary with these two vaccines over time. For this purpose, we 369 analyzed samples taken 15 to 200 days after the second dose of CoronaVac or BNT162b2 370 vaccines. One hundred and fifty-nine samples from individuals vaccinated with CoronaVac and 371 53 samples from individuals vaccinated with BNT162b2 were analyzed. Fig. 3D shows a significant 372 negative correlation for each of these vaccines (CoronaVac p <0.0001; BNT162b2 p = 0.0111). 373

The curve slope allows us to infer that around 200 days after the second dose of the CoronaVac 374 vaccine, most individuals vaccinated will present low antibody titers against the SARS-CoV-2 S 375 protein. In contrast, in individuals vaccinated with the BNT162b2 vaccine, antibodies slightly 376 decrease in most individuals, agreeing with data from the literature (2). 377

We then disaggregated the data to visualize the results (Fig. 3C ). Comparing the data obtained 378 13 to 45 days or beyond 80 days after the second dose from both vaccines, we observed a 379 significant loss of antibodies beyond 80 days after the second dose of the CoronaVac vaccine 380

(1,057±519 vs. 378±318) compared to the BNT162b2 vaccine (2,060±361 vs. 1,861±351). These 381 data suggest that the BNT162b2 vaccine is more efficient in inducing and maintaining the 382 production of antibodies against the SARS-CoV-2 virus S protein. Values range from 268±218 before the boost to 2,245±581 considering any booster, meaning an 393 8,37-fold change average. However, when we separated the data based on the type of booster 394 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101/2022.01.14.22269289 doi: medRxiv preprint vaccine, we observed that the CoronaVac booster vaccine-induced antibody production, which 395 was noticeable but milder (fold induction: 9.8x) than the antibody production induced by the 396 ChAdOx1 vaccine booster (fold induction: 12.4x) or the BNT162b2 vaccine booster (fold 397 induction: 11.2x). These results demonstrate that the CoronaVac vaccine combined with a 398 booster from CoronaVac or any other vaccine enables memory immune response to be activated, 399 in agreement with recent data (9). These authors showed that a booster with CoronaVac vaccine 400 eight months after the second dose increased neutralizing antibodies against the original virus 401 SARS-CoV-2. However, it is noteworthy to mention that the antibody response induced by the 402 third dose of the CoronaVac vaccine is lower than the two other boosters. 403

To obtain insights on the extension of the antibody response induced by the homologous and 404 heterologous booster regimes, we measured anti-SARS-CoV-2 S antibodies in 78 individuals 100 405 days after the booster dose (Median 128 days; IQR: 119-135 days) (Fig. 4C ). This analysis revealed 406 that the homologous booster with CoronaVac showed a trend towards a decline in antibody 407 production, which did not reach statistical significance (Fig. 4C) . However, the antibody response 408 CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101/2022.01.14.22269289 doi: medRxiv preprint

This work reports the dynamics of anti-S IgG after SARS-CoV2 vaccination with CoronaVac, a 426 vaccine used globally (10), a comparison with an mRNA vaccine over time, and an assessment of 427 homologous and heterologous booster schemes in Chile using CoronaVac as the basal vaccine. 428

The groups analyzed in this study span the entire vaccination program in Chile, from the 429 beginning of the vaccination schedule with priority groups, to the implementation of booster 430 schemes in late 2021. 431

Our data indicate that in individuals not exposed to SARS-CoV-2, a two-dose vaccination scheme 433 with CoronaVac induces a noticeable antibody response against SARS-CoV-2, in agreement with 434 additional reports (9). Furthermore, there is a positive correlation between the production of 435 neutralizing antibodies and those detected by ELISA (AUC). When comparing CoronaVac and 436

BNT162b2 vaccines, we found that the BNT162b2 vaccine is more efficient in inducing and 437 maintaining the production of antibodies against the SARS-CoV-2 virus S protein independent of 438 has been demonstrated to be immunogenic and safe in a double-blind, randomized, placebo-448 controlled phase-2 clinical trial (9). In this context, the homologous and heterologous booster 449 schemes analyzed in this work re-activated anti-S IgG production in individuals previously 450 vaccinated with the two-dose scheme of CoronaVac. Analysis over a more extended period of 451 time (more than 100 days) revealed that heterologous booster schemes are capable of inducing 452 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted January 23, 2022.

an elevated and long-lasting antibody response compared to two-doses plus a booster of 453

CoronaVac. Thus, these data suggest that the use of heterologous instead of homologous booster 454 regimes may allow to space the subsequent booster doses to achieve long-lasting humoral 455 response and protection against COVID19. These findings also provide evidence that will allow to 456 prioritize the subsequent booster doses in individuals that have lost optimal anti-SARS-CoV-2 457 antibodies, such as those with the homologous regime. 458

It remains to be observed if these heterologous regimes potentiate an immune response that 459 could provide protection (or partial protection) against novel variants. In this context, many 460 questions remain to be addressed. For instance, although we provide data of over 3 months after 461 the booster, it is unclear how long the protection mediated these booster schemes will last or if 462 these strategies will efficiently protect against novel variants such as delta and the recently 463 described omicron (11). In this regard, a very recent study of a heterologous booster scheme 464 based on CoronaVac + BNT162b2 in the Dominican Republic showed a reduced antibody 465 response towards the Omicron variant (12). One distinction between that study and the data 466 presented here relates to the timing between the second dose and the booster, which in Chile 467 was implemented after a six-month interval, whereas in the Dominican Republic study, the 468 heterologous booster scheme was implemented after four weeks (12). As such, the immune 469 response elicited under two different time schemes may differ in terms of the magnitude of 470 antibody production. Thus, future work combined with clinical studies are required to determine 471 the optimal time between vaccine and booster administration. Along these lines, the study of 472

Zeng et al demonstrates that extending the interval of eight months between the second and the 473 homologous booster dose with CoronaVac greatly increases antibody production (9). 474

Interestingly, our study also reports potent responses with the heterologous booster scheme 475 with the ChAdOx1 vaccine, requiring further assessment. In addition, our work is also in line with 476 a very recent report showing that heterologous booster regimes are superior to homologous 477 booster schemes based on the CoronaVac vaccine in a Brazil study (13). 478

One limitation of our study is that we assessed antibody production against the spike protein of 479 SARS-CoV2 but a relevant response mediating long-lasting immunity could also be carried out by 480 T cells, which are not analyzed in this work. However, a recent study with 15 volunteers with no 481 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101/2022.01.14.22269289 doi: medRxiv preprint suspected history of COVID-19, vaccinated with two doses of CoronaVac showed humoral and 482 cellular immune response 28d after the second dose (14). 483 As such, it is possible that a heterologous booster scheme based on CoronaVac as the basal 484 vaccine could lead to potent immunity, based on the diversity of viral antigens provided by an 485 inactivated virus formulation, followed by a booster with mRNA or adenoviral vector vaccines, 486 which trigger a superior degree of immunogenicity. The long-term immunological effects related 487 to protection against SARS-CoV-2' variants of concerns and variants of interests induced by 488 heterologous booster strategies should be determined with high priority in order to shed light on 489 the future management of the pandemic across the globe. 490   491   492  493  494  495  496  497  498  499  500  501  502  503  504  505  506  507  508  509  510  511  512  513  514  515  516  517  518  519  520 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 A) Chile had three waves of COVID-19. The first begun on April 2020, and ended in August 2020. 640

The second wave was mainly caused by gamma and lambda variants and it was more extensive, 641

beginning CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted January 23, 2022. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101/2022.01.14.22269289 doi: medRxiv preprint ! " Figure 1 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101/2022.01.14.22269289 doi: medRxiv preprint CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101/2022.01.14.22269289 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101/2022.01.14.22269289 doi: medRxiv preprint Figure 4 A B C . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted January 23, 2022. ; https://doi.org/10.1101/2022.01.14.22269289 doi: medRxiv preprint

Effectiveness of 551 an Inactivated SARS-CoV-2 Vaccine in Chile

Dynamics of 553 antibody response to BNT162b2 vaccine after six months: a longitudinal prospective study. and Salvador, Natalia and Franco

Randomized 597 Immunogenicity and Safety Study of Heterologous Versus Homologous COVID-19 Booster 598 Vaccination in Previous Recipients of Two Doses Of Coronavac COVID-19 Vaccine

3@A:* =4*58>?>8A:* >4*538?@@A:* 9@*534?;6A:* 3-412-! @@*5@8?=9A:* ;3*54@?@=A:* 9@*5=4?>9A:* 46*5=8?>6A:*