ChAdOx1 nCoV-19 effectiveness during an unprecedented surge in SARS COV-2 infections

To,

The Editor

European Journal of Internal Medicine

An unprecedented surge in SARS-CoV-2 infections driven by the Delta variant was reported from India recently[1]. The total reported new cases from its capital, New Delhi, in April and May 2021 were more than those reported in a 13-month period since the onset of the pandemic (765,117 vs 661,123)[2]. We estimated ChAdOx1 nCoV-19 effectiveness during this surge.

Sir Gangaram Hospital is a tertiary care private hospital in New Delhi, having 4296 employees with equitable access to medical benefits, including investigations, medicines and hospitalization. Of these, between 16.1.21 to 30.4.21, 2716 received two doses, and 623 received a single dose of ChAdOx1-nCoV2. 927 remained unvaccinated till 30.4.21. 20 received the BBV152 vaccine or the BNT162b2 vaccine and were excluded from our analysis. We studied infection rates, moderate to severe disease rates, supplemental oxygen therapy rates and death rates in the ChAdOx1 2-doses and single-dose cohorts against the unvaccinated cohort in the period from 1.3.21 to 31.5.21. Disease severity was assigned as per the Indian Council of Medical Research guidelines[3]. Cumulative event rates were calculated using the Kaplan Meier estimator. Cox Proportional Hazard regression model was used to calculate Hazard Ratios (aHR) adjusting for age, gender, health-worker role, previous SARS-CoV-2 infections, active or retired status and comorbidities. Vaccine effectiveness was calculated as (1-aHR)x100.

In total, 560 (13.1%) employees out 4276 tested positive for COVID-RTPCR between 1.3.21 to 31.5.21. Of these, 9 (1.6%) were lost to follow-up, 10 (1.8%) were asymptomatic, 458 (81.79%) had mild disease, 57 (10.2%) had moderate disease, and 26 (4.64%) had severe disease. There were six deaths totally in the study population (1.06%).

Of those testing positive, 61 were hospitalized, and 499 remained in home quarantine(HQ). Notably, 57.5% of those with moderate to severe disease were in home quarantine (due to non-availability of hospital beds) and would have been missed had it not been through strict disease surveillance and follow up. Since hospitalization rates alone would not have been entirely indicative of disease severity, we did not take it as an outcome measure in our study.

Marginal lowering of the incidence of symptomatic infections [12.0% (327/2716) vs 14.2% (133/937) aHR:0.76 (95%CI: 0.62-0.94)] and significant lowering of moderate to severe disease [1.2% (33/2716) vs. 3.4% (32/937); aHR: 0.35 (95% CI: 0.21-0.58)] and supplemental oxygen therapy [0.4% (11/2716) vs. 1.8% (17/937); aHR: 0.25 (95%CI: 0.11-0.58)] was observed in the 2-dose as compared to the unvaccinated cohort. This effect persisted for all events occurring beyond 14 days from dose 2.

However, when analyzing for events beyond 21 days from the single dose, the incidence of symptomatic infections [12.3% (75/607) vs. 13.9% (130/934; aHR=0.82 (0.62-1.10)], moderate to severe disease [2.0% (12/607) vs. 3.3% (31/934); aHR=0.63; 95%CI=0.32-1.24)] or the need for supplemental oxygen (0.7% (4/607) vs. 1.7%(16/934); aHR: 0.47 (95%CI: 0.15-1.44) did not seem to differ significantly from the unvaccinated cohort.

Another noteworthy finding was that symptomatic infections were significantly reduced with prior infections with SARS-CoV-2 (aHR 0.07; 95%CI 0.04-0.13).

The median interval between the second dose and test positivity was 45 days (IQR: 35-55). For the single dose, it was 42 days (IQR: 26-58). The intervals between previous infection and current infection varied from 167 days to 370 days.

A total of six deaths (5 in the unvaccinated, 1 in the one dose cohort) occurred in the age group of 65-93 years in individuals with comorbidities, at an interval of 3-40 days from RTPCR test positivity.

Our study on vaccine effectiveness (VE) reveals three key findings. (Table 1 )

Table 1.

xxxxxxxxx

Outcomes	2 doses versus no dose		1 dose versus no dose		2 doses vs no dose >=14 days		1 dose vs. no dose >=21 days		Previous infection versus no infection
	Effectiveness	95%CI	Effectiveness	95%CI	Effectiveness	95%CI	Effectiveness	95%CI	Effectiveness	95%CI
Symptomatic infections	24%	6 to 38	4%	-26 to 27	28%	10 to 41	18%	-10 to 38	93%	87 to 96
Moderate to Severe Disease	65%	42 to 79	7%	-68 to 48	67%	44 to 81	37%	-24 to 68	89%	57 to 97
Need for oxygen therapy	75%	42 to 89	12%	-111 to 63	76%	37 to 89	53%	-44 to 85	85%	-9 to 98
Deaths	97%	43 to 99.8	70%	-157 to 97	97%	43 to 99.8	69%	-160 to 97	NA	NA

Firstly, VE for two doses of ChAdOx1 nCoV-19 given at a median interval of 30 days (IQR: 28-36) was 28% (10-41%) for symptomatic infections, 67% (44-81%) for moderate to severe disease, 76% (37-89%) for supplemental-oxygen-therapy and nearly 97% (43-99.8%) for deaths. Our study reports lower protection by two doses from symptomatic infections than that reported by CMC Vellore, India (67%)[4], or by Public Health England (65% against the delta variant)[5]. It appears closer to that reported by Public Health Scotland (25-60% against the delta variant)[6]. However, of note, is similar high protection offered by two doses, as in these studies against moderate to severe disease, supplemental-oxygen-therapy and deaths.

Secondly, a single dose offered no protection in our study against symptomatic infections (18% (-10 to 38) or any outcome of interest. This appears to be in line with the Public Health Scotland data on single-dose protection against the delta variant (7%; -7 to18% within 28 days and 18%; 9-27%; beyond 28 days). However, it is in contrast to a modestly reduced but significant protection offered by a single dose as seen in the study from CMC Vellore (50%) and from Public Health England (33%). We do not believe that a larger sample size in our study could have changed these outcomes. Our total sample size of 4276 with 63% coverage and an attack rate of 13% corresponds with the WHO sample size recommendations for cohort study designs.[7]

The third key finding is that previous infections with SARS-CoV-2 were significantly protective against all studied outcomes, with an effectiveness of 93% (87 to 96%) seen against symptomatic infections, 89% (57 to 97%) against moderate to severe disease and 85% (-9 to 98%) against supplemental oxygen therapy. All deaths occurred in previously uninfected individuals. This was higher protection than that offered by single or double dose vaccine.

A genomic sequencing study from New Delhi that sequenced 33 employees from our centre identified the B.1.167.2 variant, now known as the delta variant, in more than 50% of the sequenced samples.[8] As per this study, the delta mutation allows for a greater receptor binding between the spike protein of SARS-Cov-2 and the host immune cells allowing for greater transmissibility and a limited reduction in the neutralization of these mutant strains by convalescent sera or vaccine-elicited antibodies. This may also have a potential impact on infection rates and effectiveness of vaccines.

While data on vaccine effectiveness has been accruing steadily from different nations, our findings assume importance for countries facing the threat of another surge, now driven by the delta variant globally, while quickly working to vaccinate its large susceptible populations. This information may be useful in deciding vaccine policy. In countries facing acute vaccine shortages, administrators could consider early full vaccine coverage, prioritizing those never infected, before the onset of another SARS-CoV-2 outbreak. Relying on the protective effect of a single dose alone may not be prudent.

Funding

None

Declaration of Competing Interest

The authors declare no conflicts of interest. The Institutional Ethics Committee approved of this study vide no. EC/04/21/1905. Participant consent has been taken as per the protocol approved.