Impact of Vaccination With the SCB-2019 Coronavirus Disease 2019 Vaccine on Transmission of Severe Acute Respiratory Syndrome Coronavirus 2 Infection: A Household Contact Study in the Philippines

Abstract

Background

An exploratory household transmission study was nested in SPECTRA, the phase 2/3 efficacy study of the adjuvanted recombinant protein-based COVID-19 vaccine SCB-2019. We compared the occurrence of confirmed COVID-19 infections between households and household contacts of infected SPECTRA placebo or SCB-2019 recipients.

Methods

SPECTRA participants at 8 study sites in the Philippines who developed real-time reverse transcriptase–polymerase chain reaction (rRT-PCR)–confirmed COVID-19 were contacted by a study team blinded to assignment of index cases to vaccine or placebo groups to enroll in this household transmission study. Enrolled households and household contacts were monitored for 3 weeks using rRT-PCR and anti–SARS-CoV-2 N-antigen IgG/IgM testing to detect new COVID-19 infections.

Results

One hundred fifty-four eligible COVID-19 index cases (51 vaccinees, 103 placebo) were included. The secondary attack rate per household for symptomatic COVID-19 infection was 0.76% (90% CI: .15–3.90%) if the index case was an SCB-2019 vaccinee compared with 5.88% (90% CI: 3.20–10.8%) for placebo index cases, a relative risk reduction (RRR) of 79% (90% CI: −28% to 97%). The RRR of symptomatic COVID-19 per household member was similar: 84% (90% CI: 28–97%). The impact on attack rates in household members if index cases were symptomatic (n = 130; RRR = 80%; 90% CI: 7–96%) or asymptomatic (n = 24; RRR = 100%; 90% CI: −76% to 100%) was measurable but the low numbers undermine the clinical significance.

Conclusions

In this prospective household contact study vaccination with SCB-2019 reduced SARS-CoV-2 transmission compared with placebo in households and in household members independently of whether or not index cases were symptomatic.

SARS-CoV-2, COVID-19, vaccination, household transmission, SCB-2019

Almost 628 million people have contracted coronavirus disease 2019 (COVID-19) due to infection by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [1], with more than 18 million excess deaths worldwide [2]. SARS-CoV-2 is transmitted through 3 main mechanisms: inhalation, direct inoculation onto mucosal surfaces, or touching contaminated surfaces [3]. Airborne transmission has been argued to be the dominant mechanism [4, 5], with infected individuals releasing virus in droplets during coughing, sneezing, talking, or singing to infect others [6], leading to the global implementation of mask-wearing policies in public places.

As infection risk directly relates to duration and closeness to an infected person, secondary attack rates (SARs) in household contacts (HHCs) have been estimated between 3% and 40% [7–11], with the highest SARs reported following symptomatic index cases (18%), spousal contacts (37.8%), and in households with only 1 contact (41%) [11].

Globally, the main strategies against COVID-19 have been implementation of masking and social-distancing measures and mass immunization campaigns to provide personal protection and decrease household and community transmission. Therefore, the impact of new SARS-CoV-2 vaccines on household transmission is an essential aspect of vaccine development. Several post-licensure, nonrandomized, field effectiveness investigations of vaccine prevention of SARS-CoV-2 infection among HHCs have reported a positive impact [12–18]. One database-link study reported an up to 50% reduction among HHCs of vaccinated versus unvaccinated index cases [13], but other studies have reported limited impact, especially for HHCs of index cases with the Delta (B.1.617.2) variant, potentially due to the high viral load in breakthrough cases [17, 18]. We report a prospective, blinded clinical study on transmission in households of index cases who were previously randomized to receive either the adjuvanted, recombinant protein-based COVID-19 vaccine SCB-2019 or placebo.

METHODS

This exploratory household transmission (HHT) study was conducted in 8 Philippine sites of the pivotal, randomized, placebo-controlled, phase 2/3 efficacy trial of the Clover Biopharmaceuticals SCB-2019 vaccine (study evaluating protective-efficacy and safety of Clover’s trimeric recombinant protein-based and adjuvanted COVID-19 vaccine [SPECTRA]; ClinicalTrials.gov NCT04672395) [19]. SCB-2019 contains the SARS-CoV-2 Spike protein (S-protein) from the ancestral strain (Wuhan-Hu-1) maintained in its trimeric structure using the proprietary Trimer-Tag (Clover Biopharmaceuticals) technology. Two 0.5-mL doses, containing 30 µg SCB-2019, 1.5 mg of the Toll-like receptor agonist CgG-1018 (Dynavax Technologies), and 0.75 mg aluminum hydroxide (Alhydrogel; Croda Health Care), were administered by intramuscular injection in the deltoid muscle 28 days apart. Placebo was 0.5 mL of 0.9% sodium chloride for injection. The objective was to assess and compare the reduction in COVID-19 infections in households and household members of SCB-2019 vaccine recipients with those of placebo recipients. Sites and numbers of placebo or vaccine recipient index cases enrolled by site are presented in Supplementary Table 1. The protocol was approved by the Single Joint Research Ethics Board (SJREB), the institutional review boards (IRBs) of the 8 study sites, and the International Vaccine Institute IRB and implemented in accordance with the principles of the Declaration of Helsinki, International Conference on Harmonisation - Good Clinical Practice (ICH-GCP), and Philippines’ ethical requirements. All participants or their parents/legal guardians provided written informed consent or assent.

Participants and Procedures

As part of the main SPECTRA trial, all participants, symptomatic or not, were prospectively followed for SARS-CoV-2 infection using symptom screening and weekly testing with rapid COVID-19 antigen testing kits (Roche Diagnostics, Basel, Switzerland) [19]. Any participant at the 8 Philippines study sites with a SARS-CoV-2 infection confirmed by positive real-time reverse transcriptase–polymerase chain reaction (rRT-PCR) within 24 hours of sample collection was identified as a potential index case for the HHT study and was contacted by the HHT study team, which was distinct from the SPECTRA study team, for their willingness to enroll in this study and allow follow-up of their household. Both index cases and the HHT study team were blinded to vaccine or placebo study group allocation. Once an index case agreed to participate, their HHCs were contacted within 3–5 days for their consent to enroll and adhere to 3 weeks of follow-up. Eligible HHCs were aged 6 years or older (caretaker consent/assent), with no known prior SARS-CoV-2 infection, with only the index case having been diagnosed with COVID-19 in the household. Key sociodemographic and vaccination information on HHCs, including age, sex, income status, comorbidities, COVID-19 risk (including at-risk occupations such as healthcare workers, bus drivers, etc), use of COVID-19–preventive efforts (masking, crowd avoidance, and hand sanitation), was documented. Vaccination history included vaccine type and date. Household contacts were classified as fully vaccinated, partially vaccinated (only 1 dose of a 2-dose regimen, or an interval of <14 days since the second dose), or unvaccinated. Case count started 14 days after the last dose in vaccinated HHCs.

At enrollment, HHCs were tested using lateral flow anti–SARS-CoV-2 N-antigen immunoglobulin M/immunoglobulin G (IgM/IgG) test kits (PCL, Inc, South Korea). IgM positivity at study start was considered as acute infection and confirmed by rRT-PCR testing of a nasopharyngeal (NP) swab sample. Those with IgG positivity or who were IgG/IgM-negative were actively followed up for 3 weeks for symptoms suggestive of COVID-19 (breathing difficulties, fever, chills, cough, fatigue, muscle/body aches, headache, new loss of taste, smell, sore throat, congestion, runny nose, nausea, vomiting, or diarrhea). A report of any symptom during follow-up triggered NP rRT-PCR testing. Those who were negative and without COVID-19 symptoms during follow-up were tested again at the study end. Asymptomatic infection was defined as a negative anti-N IgG/IgM test at enrollment but a positive result at study end and no COVID-19–suggestive symptoms during the 3-week follow-up.

Follow-up for COVID-19 Symptoms and Nasopharyngeal Sampling

Enrolled HHCs completed daily checklists for any COVID-19 symptoms for 3 weeks with instructions to promptly contact the study coordinator (available 24 hours, 7 days/week) if they developed any symptoms. Study coordinators telephoned households weekly to check for any symptoms that had been missed. Nasopharyngeal samples were collected at home or the study site on the day any symptoms were reported and transported to 1 of 2 certified molecular laboratories in the Philippines (Manila Doctors Hospital or the Asian Hospital and Medical Center) for rRT-PCR testing with molecular assays and specific laboratory instructions according to World Health Organization (WHO) guidelines [20].

The rRT-PCR-confirmed cases were considered as symptomatic COVID-19 and the HHC was followed for severity from the time of diagnosis until symptom resolution or outcome 14 days after symptom onset. For HHCs with symptomatic COVID-19 a detailed case report was completed to capture risk of infection, relationship to index case, any contact with individuals with known or suspected COVID-19 other than the index case within the previous 2 weeks, and details of symptoms including temporal relationship with the index case diagnosis.

Statistical Analysis

Baseline characteristics (COVID-19 risk, comorbidities, and vaccination status) of SCB-2019 and placebo index case HHCs were summarized and compared using chi-square or Fisher's exact test for categorical variables, and Student's t, Wilcoxon rank-sum, or nonparametric tests for quantitative covariates, depending on whether the data were normally distributed. Primary endpoints were symptomatic SARs of SARS-CoV-2 infection in households (hSARs) and household contacts (mSARs) of SCB-2019 and placebo recipients in the SPECTRA trial (index cases); secondary endpoints were the hSARs and mSARs for asymptomatic and overall SARS-CoV-2 infection. hSARs were estimated within strata of proportions of households (0–24%, 24–49%, 50–74%, 75–100%) who received any SARS-CoV-2 vaccine (household vaccination coverage); mSARs were also estimated according to vaccination status of the HHC at study enrollment (full, partial, or unvaccinated).

The mSAR was estimated as the proportion of HHCs in a stratum who experienced symptomatic, asymptomatic, or overall SARS-CoV-2 infection, with 90% confidence intervals (CIs) estimated assuming a binomial distribution. Relative risk reduction (RRR) for the mSAR was estimated as 1 minus the ratio of 2 mSAR estimates (eg, contacts in SCB-2019 households vs placebo households) × 100, with 90% CIs estimated using the PropCIs library (version 0.30) of R (version 4.1.2; R Foundation for Statistical Computing). To account for clustering of HHCs by household, hSARs and RRRs were estimated by generalized estimating equation (GEE) logistic regression modeling, using the geeglm function of the geepack library of R (version 4.1.2), assuming an exchangeable correlation structure, with standard errors estimated by the approximate jackknife variance.

RESULTS

During active surveillance for this HHT study from 9 July 2021 to 6 December 2021 we identified 598 rRT-PCR–confirmed index cases among SPECTRA trial participants. Of those, 42 were excluded because they were identified as an index case too late (>5 days after diagnosis), 118 due to lack of eligible household members, 184 did not consent to having their household members contacted, 98 households refused to participate, and 2 were excluded as they lived in the same household. Thus, 154 index cases representing 154 households (51 SCB-2019 and 103 placebo) were eligible (Figure 1). Of these index cases, 130 (42 SCB-2019 and 88 placebo) were symptomatic and 24 (9 SCB-2019 and 15 placebo) were asymptomatic at diagnosis. Sequencing analysis of 74 index case swab samples (24 SCB-2019, 50 placebo) identified all as the Delta (B.1.617.2) variant, the predominant variant in the Philippines during the study period (Table 1). The mean ages of the index cases were similar in the SCB-2019 (36.7 ± 12.2 years) and placebo (35.0 ± 10.9 years) groups as were sex, comorbidities, and adherence to the SPECTRA study vaccination regimen (Table 1).

Figure 1.

Disposition of households, HHCs, and household infections by vaccinated/placebo index case (SPECTRA). Abbreviations: COVID-19, coronavirus disease 2019; HHC, household contact(s).

Table 1.

Sociodemographic Information on the Index Cases and Their Household Contacts

	Index Cases (N = 154)a
	SCB-2019 SPECTRA Group (n = 51)	Placebo SPECTRA Group (n = 103)
Age, years
ȃMean (SD)	36.7 (12.2)	35.0 (10.9)
ȃMedian (Min, Max)	35 (19, 65)	33 (20, 65)
Sex, no. (%)
ȃMale	25 (49.0%)	40 (38.8%)
ȃFemale	26 (51.0%)	63 (61.2%)
Infection status, no. (%)
ȃSymptomatic	42 (82.4%)	88 (85.4%)
ȃAsymptomatic	9 (17.6%)	15 (14.6%)
Number who did not receive second dose (%)	3 (5.9%)	1 (1.0%)
Days from last vaccination to RT-PCR confirmation
ȃMean (SD)	77.9 (33.1)	73.5 (34.7)
ȃMedian (Min, Max)	81 (14, 158)	66 (17, 169)
Virus sequenced (all Delta variant), no. (%)	24 (47.1%)	50 (48.5%)
Households (n = 154)
ȃAll households	51	103
ȃȃWith 1 eligible household contact, no. (%)	11 (22%)	32 (31%)
ȃȃWith 2 eligible household contacts, no. (%)	11 (22%)	29 (28%)
ȃȃWith 3 or more eligible household contacts, no. (%)	29 (57%)	42 (41%)
ȃHousehold size
ȃȃMean (SD)	2.6 (1.3)	2.4 (1.1)
ȃȃMedian (Min, Max)	3 (1, 4)	2 (1, 6)
Household contacts of index cases (n = 384), n	134	250
ȃAge, years
ȃȃMean (SD)	28.8 (15.4)	28.8 (16.9)
ȃȃMedian (Min, Max)	24 (6, 70)	26 (6, 78)
ȃSex, no. (%)
ȃȃMale	75 (56.0%)	137 (54.8%)
ȃȃFemale	59 (44.0%)	113 (45.2%)
ȃRelationship to the vaccine trial participant, no. (%)
ȃȃParent	21 (15.7%)	37 (14.8%)
ȃȃChild	22 (16.4%)	51 (20.4%)
ȃȃSpouse	7 (5.2%)	18 (7.2%)
ȃȃGrandparent	1 (0.7%)	5 (2.0%)
ȃȃOther	83 (61.9%)	139 (55.6%)

Abbreviations: Max, maximum; Min, minimum; RT-PCR, reverse transcriptase–polymerase chain reaction; SD, standard deviation.

SARS-CoV-2 infection was confirmed in all index cases by RT-PCR.

A total of 388 HHCs in the 154 households enrolled; 4 were lost to follow-up, with 384 (134 SCB-2019 vaccine and 250 placebo index cases) completing the study (Table 1). Household composition and size and demographic characteristics of age, sex, and relationship to the index case were all comparable. The majority of HHCs were either not vaccinated or partially vaccinated; only 22.9% reported as being fully vaccinated, a proportion that was similar in households of SCB-2019 vaccinee or placebo index case index cases (Supplementary Table 2). Baseline seropositivity as an indication of past infection was similar between the SCB-2019 and placebo index case households (39.6% vs 34.8%). Of those who were negative at baseline, 21.8% and 26.9% seroconverted as an indication of infection during the study (Table 2). Most HHC infections occurred when the index case was symptomatic: 106 (79.1%) HHCs of 42 (82.4%) symptomatic SCB-2019 index cases and 212 (84.8%) HHCs of 88 (85.4%) placebo recipient symptomatic index cases. The remainder in both groups were contacts of asymptomatic index cases (Figure 1).

Table 2.

COVID-19 Infection History and Vaccine-Related Information in Household Contacts

	Household Contacts
	SCB-2019 Index Cases (n = 134)	Placebo Index Cases (n = 250)
COVID-19 tested in the previous 3 months, no. (%)
ȃTested	12 (9.0%)	29 (11.6%)
ȃȃPositive test result	1 (0.7%)	3 (1.2%)
Vaccination status for COVID-19, no. (%)
ȃUnvaccinated	85 (63.4%)	154 (61.6%)
ȃPartially vaccinated	19 (14.2%)	38 (15.2%)
ȃFully vaccinated	30 (22.4%)	58 (23.2%)
Baseline and end of follow-up COVID-19 antibody test resultsa
ȃTested at baseline, n	134	250
ȃȃSeropositive at baseline, no. (%)	53 (39.6%)	87 (34.8%)
ȃȃSeronegative at baseline, no. (%)	81 (60.4%)	163 (65.2%)
ȃPaired baseline seronegative samples tested for antibody at end of follow-up,b no. (%)	78 (96.3%)	145 (89.0%)
ȃȃSeropositive at end of follow-upc	17 (21.8%)	39 (26.9%)
ȃȃSeronegative at end of follow- upc	61 (78.2%)	106 (73.1%)
Time to symptomatic HHC diagnosis from index case diagnosis
ȃn	1	12
ȃMean (SD) days	16.0 (0)	(19.6)

Abbreviations: COVID-19, coronavirus disease 2019; HHC, household contact; IgG, immunoglobulin G; IgM, immunoglobulin M; SD, standard deviation.

Anti-N IgG or IgM.

Denominator is those with seronegative test result at baseline.

Denominator is those who had paired baseline and end-of-follow-up samples.

The symptomatic SARS-CoV-2 infection rate was lower in the SCB-2019 recipient index case households and among their HHCs compared with the infection rate in households and HHCs of placebo recipient index cases (Tables 3 and 4).

Table 3.

Symptomatic SARS-CoV-2 Infection Secondary Attack Rates in the Households of Index Cases

	Households of SPECTRA Index Cases
	SCB-2019 Index Cases (n = 51)				Placebo Index Cases (n = 103)				RRRb (90% CI*)
	Cases	Households	Households With ≥1 Case	Household SAR, % of Index Case's Contacts (90% CI)a	Cases	Households	Households With ≥1 Case	Household SAR, % of Index Case's Contacts (90% CI)a
ȃAll	1	51	1	.76 (.15 –3.90)	12	103	7	5.88 (3.20–10.82)	.79 (−.28 to .97)
Household COVID-19 vaccine coverage (%)c
ȃ75 to 100	0	9	0		2	25	2	4.27 (1.33–13.73)
ȃ50 to 74	0	12	0		0	18	0
ȃ25 to 49	0	7	0		8	16	3	18.3 (7.48 –44.95)
ȃ0 to 24	1	23	1	1.95 (.38 –9.87)	2	44	2	2.48 (.77–8.00)	.21 (−5.14 to .90)

Abbreviations: CI, confidence interval; COVID-19, coronavirus disease 2019; GEE, generalized estimating equation; hSAR, household secondary attack rate; RRR, relative risk reduction; SAR, secondary attack rate; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

Point estimates and 90% CI for hSARs and the indirect effect were estimated using a GEE logistic regression modeling approach with jackknifed standard errors (see Methods).

RRR calculated as 1 minus the relative risk comparing SARs in SCB-2019 and placebo households.

Coverage for any COVID-19 vaccine received estimated as the proportion of household members who received any SARS-CoV-2 vaccine over total size of household.

Table 4.

Symptomatic SARS-CoV-2 Infection Secondary Attack Rates Among Household Contacts of Index Cases

	Household Contacts of Index Cases
	SCB-2019 Vaccine Group		Placebo Group
	Cases Among Contacts/Total Number of Contacts	SAR, % (90% CI)	Cases Among Contacts/Total Number of Contacts	SAR, % (90% CI)	RRRa (90% CIb)
Secondary attack rates among household contacts of all index cases
ȃNo.	134		250		…
ȃAll household contacts	1/134	.75 (.04–3.49)	12/250	4.80 (2.79–7.66)	.84 (.28–.97)
ȃFully vaccinated	0/30	0 (0–9.50)	3/58	5.17 (1.42–12.8)	1.00 (−.66 to 1.00)
ȃPartially vaccinated	0/19	0 (0–14.6)	1/38	2.63 (.13–11.9)	1.00 (−4.21 to 1.00)
ȃUnvaccinated	1/85	1.18 (.06–5.46)	8/154	5.19 (2.61–9.18)	.77 (−.07 to .95)
Secondary attack rates among household contacts of symptomatic index cases
ȃNo.	106		212		…
ȃAll household contacts	1/106	.94 (.05–4.40)	10/212	4.72 (2.58–7.87)	.80 (.07–.96)
ȃFully vaccinated	0/21	0 (0–13.3)	3/53	5.66 (1.56–14.0)	1.00 (−1.13 to 1.00)
ȃPartially vaccinated	0/16	0 (0–17.1)	1/33	3.03 (.16–13.6)	1.00 (−4.34 to 1.00)
ȃUnvaccinated	1/69	1.45 (.07–6.69)	6/126	4.76 (2.09–9.18)	.70 (−.47 to .94)
Secondary attack rates among household contacts of asymptomatic index cases
ȃNo.	28		38		…
ȃAll household contacts	0/28	0 (0–10.2)	2/38	5.26 (.94–15.7)	1.00 (−.76 to 1.00)
ȃFully vaccinated	0/9	0 (0–28.3)	0/5	0 (0–45.1)	…
ȃPartially vaccinated	0/3	0 (0–63.2)	0/5	0 (0–45.1)	…
ȃUnvaccinated	0/16	0 (0–17.1)	2/28	7.14 (1.28–20.8)	1.00 (−1.21 to 1.00)

Abbreviations: CI, confidence interval; mSAR, secondary attack rate among household contacts of index cases; RRR, relative risk reduction; SAR, secondary attack rate; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

One minus the relative risk comparing the mSAR among the household contacts of SCB-2019 versus placebo index cases.

90% CIs estimated as 1 minus the score CI for the ratio of 2 proportions [relative risk], as implemented in the PropCIs package (version 0.3.0) of R (version 4.1.2; R Foundation for Statistical Computing).

Risk Reductions in Households

The hSAR for symptomatic COVID-19 infection was 0.76% (90% CI: .15–3.90%) in the 51 households where the index case was an SCB-2019 vaccinee compared with 5.88% (90% CI: 3.20–10.8%) in the 102 households of placebo index cases (Table 3), an RRR of 79% (90% CI: −28% to 97%).

Risk Reductions in Household Contacts

There was 1 symptomatic secondary COVID-19 case among 134 HHCs of symptomatic or asymptomatic SCB-2019 index cases, with an mSAR of 0.75% (90% CI: .04–3.49%), compared with 12 symptomatic cases reported from 250 HHCs in the placebo group, with an mSAR of 4.80% (90% CI: 2.79–7.66%); the RRR of transmission to household members was 84% (90% CI: 28–97%) (Table 4). Risk reduction was similar for vaccinated and unvaccinated HHCs. There were no cases of symptomatic SARS-CoV-2 infection in fully or partially vaccinated HHCs of SCB-2019 recipient index cases. The RRR in symptomatic COVID-19 disease among unvaccinated HHCs of SCB-2019 recipient index cases compared with unvaccinated HHCs of placebo recipient index cases was 77% (90% CI: −7% to 95%).

The pattern was similar when only assessing index cases with symptomatic COVID-19 (Table 4). The mSAR in the HHCs of SCB-2019 index cases was 0.94% (90% CI: .05–4.40%) compared with 4.72% (90% CI: 2.58–7.87%) for placebo index case HHCs, an RRR of 80.0% (90% CI: 7–96%). There were no secondary infections observed in 28 HHCs of SCB-2019 index cases with asymptomatic COVID-19 infections and only 2 cases in 38 HHCs of placebo index cases with asymptomatic infections (Table 4). These small numbers preclude any meaningful analysis.

The mSAR for asymptomatic infections in HHCs who were seronegative at baseline did not show any differences between those in households with SCB-2019 or placebo index cases, an RRR of 12% (90% CI: −32% to 43%) based on mSARs of 21.0% (90% CI: 13.8–29.8%) and 23.9% (90% CI: 18.5–30.1%) in SCB-2109 and placebo group households, respectively.

DISCUSSION

Measures such as mask-wearing, social distancing, and work-from-home policies are major components of preventative health measures in the COVID-19 pandemic, but domestic transmission within households cannot be subject to such regulations. Therefore, a better understanding of this aspect of viral dissemination, and especially of how it is impacted by vaccination campaigns, is particularly important for health authorities to control outbreaks.

Our study, conducted as part of the SPECTRA trial with blinded vaccine and placebo groups, is an important contribution to understanding the impact of vaccination in preventing transmission among household contacts. The efficacy of SCB-2019 in the Philippines as part of the SPECTRA study was 100% against severe disease and 69.3% (95% CI: 51.0–81.4%) against disease of any severity in the initial short-term analysis [19]; respective efficacies at 6 months post-vaccination were 100% and 48.3% (38.0–57.0%) when most cases would have been the Delta variant. After monitoring the occurrence of rRT-PCR–confirmed COVID-19 in household members of infected SPECTRA participants for 3 weeks, we found a 79% (90% CI: −28% to 97%) risk reduction in secondary symptomatic infection rates in the 51 included households where the COVID-19–positive index cases were vaccinated with the SCB-2019 vaccine versus the 103 households where the infected index cases had received placebo. An 84% (90% CI: 28–97%) reduction in secondary infections in the 134 household contacts of vaccinated index cases compared with 250 contacts of placebo recipients mirrored the trend seen in households. These results are particularly notable since they were obtained when the Delta (B.1.617.2) variant, which has more rapid symptom onset and higher viral load than earlier strains, predominated [18].

The study was conducted as part of a well-conducted, randomized, double-blind, placebo-controlled efficacy trial (“SPECTRA”) with the study team blinded to participant allocation throughout the study, providing a unique opportunity to minimize confounding while estimating the vaccine impact on preventing secondary infections among household contacts. Trial participants underwent intensive follow-up and testing for SARS-CoV-2 infection and disease. Any confirmed COVID-19 in SPECTRA participants in the Philippines was a potential household index case. Daily clinical follow-up of index case HHCs used methods established for the clinical efficacy trial including weekly telephone call interviews and daily self-reported symptom checklists assessing HHCs with signs and symptoms of any severity. All symptomatic cases were confirmed by testing NP samples using rRT-PCR following the WHO standard guidelines to reliably diagnose COVID-19 [20]. No cases of symptomatic SARS-CoV-2 infection occurred in fully or partially vaccinated contacts of SCB-2019 recipient index cases, indicating that SCB-2019 vaccine provides both direct protection to the vaccinee and indirect protection to HHCs by reducing transmission from the index case to their household members. Although we were not able quantify this in our study, a faster decline in viral load was reported in vaccinated people with breakthrough SARS-CoV-2 infection than in unvaccinated persons [18], which may impact transmission.

Household transmission has played a significant role in the scale of the COVID-19 pandemic [21–23]. There is a lack of conclusive evidence on the impact of vaccines to prevent secondary infections in HHCs; some studies have demonstrated a positive impact of vaccination on reducing household transmission of SARS-CoV-2 [12–16], whereas others reported limited impact in decreasing the risk of household transmission of COVID-19, especially for contacts of index cases with the Delta variant [17, 18]. A major barrier to obtaining conclusive evidence is the lack of control groups and the presence of significant confounding, so our study findings, with blinded vaccine and placebo groups, provides important information on the impact of vaccination on secondary COVID-19 infections among HHCs, and particularly on the impact of SCB-2019 on preventing such secondary infections in HHCs.

Elucidating the role of household transmission can have several policy and practice implications, in particular to inform vaccination practices and policies in resource-limited countries where the coverage is still low, such as sub-Saharan Africa where average vaccine coverage is approximately 10% [24]. In countries with limited access to COVID-19 vaccines, vaccines that block household viral transmission could enhance control of SARS-CoV-2.

As COVID-19 can transition to endemicity in many settings, understanding the mechanisms and developing prevention strategies for household transmission will provide several opportunities. Moreover, with the potential for future waves as vaccine immunity wanes or new variants emerge, vaccination strategies globally will need to consider household coverage of COVID-19 vaccination as an important aspect of limiting viral transmission.

Limitations

Our study findings should be interpreted with several important limitations. First, 441 index cases refused to participate due to fear of stigma and discrimination related to COVID-19 infection in the Philippines. This could have created a selection bias if there was a difference between SCB-2019 and placebo index cases. However, 155 (35%) and 286 (65%) of those who refused to participate had received SCB-2019 or placebo, respectively, mirroring the 51 (33%) and 103 (67%) who did enroll. Second, the detection of incident asymptomatic infections during follow-up of HHCs using rapid COVID-19 antibody testing kit may underestimate the burden of asymptomatic COVID-19 infections due to potentially limited sensitivity and specificity [25–27]. Third, pandemic-related lockdowns and travel restrictions disrupted onsite monitoring and quality-control/-assurance activities, but data quality was assured through centralized database-based monitoring and contracting local monitors for onsite visits. Secondary attack rate estimates presented here do not account for possible exposure to sources of SARS-CoV-2 other than the SPECTRA index case, an issue in practically all HHC studies. The lack of multiple variable adjustment makes it likely that these SAR estimates are affected by residual confounding. Finally, our observations apply only to the Delta variant, which predominated in the Philippines at the time of the study, and not to variants such as Omicron, which emerged subsequently.

Conclusions

Based on a limited number of households and contacts, we observed a trend of a 79% reduction in household transmission and 80% in HHC transmission for SCB-2019 vaccinated compared with placebo index cases in an environment of predominantly Delta variant circulation.

Supplementary Data

Supplementary materials are available at Clinical 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. The authors are grateful to all participants in the study and to their external service providers (contract research organizations, laboratories, clinical suppliers, and biostatistics) for their invaluable assistance in performing the trial. They thank Keith Veitch (keithveitch communications, The Netherlands) for editorial assistance with the manuscript (funded by Clover Biopharmaceuticals).

Financial support. This work was supported by the Coalition for Epidemic Preparedness Innovations (grants numbers RRCL 2001 and RCL2202) and Clover Biopharmaceuticals. P. L. reports support as a vaccine manufacture employee.

The Household Contact (HHC) Study Group. The Philippines: Camilo Roa Jr, Manila Doctors Hospital, Manila; Charissa Borja-Tabora, Asian Hospital and Medical Center, Alabang, Muntinlupa; Josefina Carlos, University of the East Ramon Magsaysay Memorial Medical Center, Quezon City; May Emmeline B. Montellano, Far Eastern University Hospital—Nicanor Reyes Medical Foundation, Quezon City; Mari Rose A. de Los Reyes, Las Pinas Doctors Hospital, Las Pinas City; Edison R. Alberto, Tropical Disease Foundation-HIMC, Cavite City; Milagros Salvani-Bautista, University of the East Ramon Magsaysay Memorial Medical Center, Quezon City. South Korea: Deok Ryun Kim, Hwa Young Kim, Sye Lim Hong, Sophie S. Y. Kang, and Irene Njau, International Vaccine Institute, Seoul.

Author notes

The Household Contact (HHC) Study Group is listed in the Notes section.

Potential conflicts of interest. P. L. and J. G. are full-time employees of the study sponsor, Clover Biopharmaceuticals, and receive an employment total award package as well as salary. R. C. reports participation as Chair Scientific Advisory Board for Clover Biopharmaceuticals, plays a role on the Board for Clover Biopharmaceuticals and the Board of Trustees for the International Vaccine Institute (IVI), and has stock or stock options in Clover Biopharmaceuticals. J. S. reports consulting fees from IVI and participation on a Data Safety and Monitoring Board (DSMB) for the OPTIMMS Trial (Oxford University, Andrew Pollard as Principal Investigator). F. R. reports personal fees from Clover Biopharmaceuticals during the conduct of the study, personal fees from Merck Research Labs, personal fees from Novartis, personal fees from Lilly, grants and personal fees from AstraZeneca, personal fees from Union Chimique Belge, personal fees from Sanofi, personal fees from Amgen, personal fees from Icosavax, personal fees from Gilead, grants from Eidos, grants from Bristol Myers Squibb, personal fees from Janssen, personal fees from Doctor Evidence, grants from American Regent, personal fees from Biogen, personal fees from Intercept, personal fees from Alkermes, and personal fees from Diurnal, outside the submitted work, and equity interest in GSK, Athira, Spencer Healthcare, Clover Biopharmaceuticals, Doctor Evidence, Odyssey Topco, and Adaptic Health. All other authors report no potential conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.