Abstract
Introduction
The effectiveness of AZD7442 (tixagevimab/cilgavimab) against COVID-19 hospitalizations was determined at 3 and 6 months among immunocompromised individuals in Israel during different variant circulations.
Methods
This was a retrospective cohort study using data from Clalit Health Services in Israel. Immunocompromised individuals eligible to receive AZD7442 300 mg between 15 February and 11 December 2022 were identified. Immunocompromised individuals receiving AZD7442 300 mg as pre-exposure prophylaxis (PrEP) were propensity score (PS)-matched 1:1 to unexposed individuals using a “rolling cohort” approach. Calendar time Cox proportional hazards regression models were performed with adjustment for post-matched unbalanced covariates to estimate hazard ratios (HRs) and 95% confidence intervals (CIs).
Results
Overall, 2444 AZD7442-exposed immunocompromised individuals were PS-matched to unexposed individuals. In the matched population, up to 6 months of follow-up, AZD7442 300 mg presented an unadjusted HR (without adjustment for the unbalanced covariates) of 0.68 (95% CI 0.43–1.08) and covariate-adjusted HR of 0.64 (95% CI 0.40–1.03) against COVID-19 hospitalization. Covariate-adjusted instantaneous hazards plots showed that the effectiveness of AZD7442 300 mg waned from Day 90. Up to 3 months of follow-up, the unadjusted HR was 0.43 (95% CI 0.21–0.91) for AZD7442 300 mg against COVID-19 hospitalization in the matched population; there were insufficient events to allow covariate-adjusted analysis.
Conclusion
Our results suggest that AZD7442 300 mg reduced COVID-19 hospitalizations among immunocompromised individuals; however, the findings are limited by a lack of sufficient events to produce conclusive results.
Supplementary Information
The online version contains supplementary material available at 10.1007/s40121-024-01100-3.
Keywords: AZD7442, COVID-19, Effectiveness, Immunocompromised, Monoclonal antibodies
Key Summary Points
| Why carry out this study? |
| It is important to understand the clinical protection provided by COVID-19 pre-exposure prophylaxis monoclonal antibodies to inform health policy recommendations. |
| This study aimed to determine the real-world effectiveness of AZD7442 300 mg against COVID-19 hospitalizations among immunocompromised individuals, analyzed within 3-month and 6-month follow-up periods and by different variant circulations. |
| What was learned from the study? |
| A lack of effect after Day 90 suggests waning of AZD7442 300 mg effectiveness and/or emergence of circulating resistant variants. |
| AZD7442 300 mg showed potential for reducing COVID-19 hospitalization, especially in the first 3 months after administration. |
| These findings confirm the potential utility of COVID-19 monoclonal antibodies and the rationale for assessing redosing. |
Introduction
Since December 2019, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has resulted in more than 775 million cases and 7 million deaths [1]. COVID-19 vaccines were first introduced into mass vaccination programs in December 2020 and have helped reduce the burden of the pandemic [2]. However, many people remain unvaccinated for reasons such as lack of vaccine access, prior medical conditions, or vaccine hesitancy [3]. Fully vaccinated individuals can also be at risk of breakthrough SARS-CoV-2 infections [4] and waning COVID-19 vaccine-induced immunity [5].
Immunocompromised individuals have suboptimal vaccine response [6–8] and are more likely to become severely ill when infected with SARS-CoV-2 [9]. Despite recently observed declines in COVID-19 cases and severe outcomes among the general population, immunocompromised individuals continue to be at disproportionately high risk of severe COVID-19 outcomes (COVID-19 death and hospitalization) relative to the general population, even during the Omicron era [10–13].
In December 2021, the US Food and Drug Administration (FDA) granted emergency use authorization (EUA) to AZD7442, a combination of two long-acting SARS-CoV-2–neutralizing monoclonal antibodies (mAbs)—tixagevimab and cilgavimab—to prevent COVID-19 as pre-exposure prophylaxis (PrEP) [14, 15]. AZD7442 was authorized for adults and children ≥ 12 years of age who are immunocompromised and may not produce an adequate response to COVID-19 vaccines, as well as for those for whom COVID-19 vaccination is not recommended [14, 15]. AZD7442 was authorized by the US FDA at a dose of 300 mg, but was subsequently modified in February 2022 to 600 mg due to the emergence of the Omicron BA.1 and BA1.1 subvariants [15, 16]. On 15 February 2022, the Israeli Ministry of Health (IMoH) approved AZD7442 300 mg as PrEP against COVID-19 among individuals ≥ 12 years of age who have at least one moderate/severe immunocompromised condition or who are in receipt of immunosuppressive medications or treatments and who may not mount an adequate immune response to the COVID-19 vaccine [17]. On 26 January 2023, the US FDA paused the EUA for AZD7442 due to its apparent inability to neutralize > 90% of circulating Omicron variants in vitro (including BQ.1 and XBB.1) [18].
The global authorizations and approvals of AZD7442 for PrEP were based on efficacy and safety data from the Phase 3 PROVENT clinical trial, showing that intramuscular AZD7442 as PrEP was associated with risk reductions for developing symptomatic COVID-19 versus placebo of 77% at the primary analysis and 83% at a median of 6 months [19]. The PROVENT trial was conducted during the Alpha (B.1.1.7) and Delta (B.1.617.2) variant waves (enrollment from November 2020 to March 2021), and included unvaccinated individuals and a small number of immunocompromised individuals (approximately 3.8%) [19]. Subsequent studies based on real-world data reported the effectiveness of AZD7442 at reducing the risk of symptomatic COVID-19 and COVID-19-related hospitalizations in immunocompromised individuals during the Omicron era, usually against specific variants [20–23]. However, there are limited data on the impact of AZD7442 on COVID-19–related hospitalizations over longer periods of follow-up.
There are few available preventative measures against COVID-19 for immunocompromised individuals and the variant landscape continues to evolve. Therefore, it is important to understand the clinical protection provided by COVID-19 PrEP mAbs, even during times when susceptible and resistant variants were circulating, to inform health policy recommendations for the use of new COVID PrEP mAbs. This study aimed to determine the real-world effectiveness of AZD7442 against COVID-19 hospitalizations among immunocompromised individuals, analyzed within 3-month and 6-month follow-up periods and by variant circulations.
Methods
Study Design and Data Source
This analysis is based on the eVusheld Assessment Real World Effectiveness (VALOR-C19) observational, retrospective cohort study, which assessed the effectiveness of AZD7442 PrEP in reducing COVID-19 hospitalizations among AZD7442-eligible immunocompromised populations in Israel (ClinicalTrials.gov: NCT05712096) [24]. We used real-world data from the Clalit Health Services (CHS) electronic health record (EHR) database, the largest of Israel's four integrated payer/provider healthcare organizations, encompassing more than 4.7 million members (52%) of the Israeli population. CHS contains clinical and administrative inpatient and outpatient data, including COVID-19-specific data curated at the national level, polymerase chain reaction test results, and COVID-19–related admissions and deaths. COVID-19 data are collected by the IMoH from multiple sources and shared daily with healthcare providers, including Clalit. These data repositories have been previously described in detail [2, 25].
Ethical Approval
This study used de-identified patient data and was approved by the CHS institutional review board.
Study Population
Individuals were eligible for inclusion if they were aged ≥ 12 years, weighed ≥ 40 kg during 2 years prior to the eligibility date, and met the IMoH criteria for AZD7442 PrEP during the study period (from 15 February to 11 December 2022). The study start date reflects the approval date of AZD7442 in Israel. The study end date was selected to restrict analysis to the year 2022 due to the study being divided into 4-week periods. Eligibility was defined as having at least one comorbidity or treatment causing moderate/severe immunosuppression [e.g., chimeric antigen receptor T-cell therapy, solid organ transplant, autologous or allogenic bone marrow transplant (within the previous year if allogenic), hypogammaglobulinemia, active lymphoma, active multiple myeloma, diagnosis of chronic lymphocytic leukemia (CLL) or receiving treatment for CLL, and B-cell depleting therapies]. The full list of disease definitions for immunocompromised individuals is included in Supplementary Table S1. A separate analysis of this study has previously characterized AZD7442 uptake among immunocompromised individuals eligible for AZD7442 administration as PrEP as per IMoH recommendations [24].
Study Outcome
The primary outcome of interest was COVID-19 hospitalization, defined as any hospitalization reported by the IMoH due to SARS-CoV-2 infection, at 6 months of follow-up. An exploratory analysis also assessed COVID-19 hospitalization at 3 months of follow-up.
Individuals who received AZD7442 300 mg as PrEP against COVID-19 were considered AZD7442-exposed, and individuals who did not receive AZD7442 were considered unexposed.
Statistical Analysis
We divided the study into 4-week calendar time periods. Within each of these 4-week intervals, newly AZD7442 exposed individuals were contemporaneously matched (1:1 ratio) to eligible AZD7442 unexposed individuals.
To minimize the impact of confounding by indication, matching was performed based on propensity scores (PS) estimated using the extreme gradient boosting model (XGBoost), and the matching was set up prospectively on retrospective data. The following covariates were included, based on relevant baseline characteristics: age, sex, district, peripheral rank, smoking status, chronic kidney failure, chronic obstructive pulmonary disease, asthma, other respiratory diseases, heart disease, cerebrovascular disease, body mass index, hypertension, type 1 diabetes, type 2 diabetes, neurological disease, liver disease, thalassemia, hematological cancer, organ transplant, bone marrow transplant, primary immunodeficiency, secondary immunodeficiency, number of physician visits in the last year, number of hospitalizations in the last year, and number of emergency department visits in the last year (the full definitions of these covariates are presented in Supplementary Table S2). Nearest-neighbor matching was performed using a caliper width between participants of 0.2 times the pooled standard deviation of the logit of the PS. After matching, a covariate was considered sufficiently balanced if the standardized mean difference (SMD) between the AZD7442-exposed group and the matched unexposed group was < 0.1. Data from 221 AZD7442-exposed individuals were utilized to construct the first PS model for the rolling cohort and consequently were excluded from the final analysis.
The index date for each PS-matched pair was the date of AZD7442 PrEP administration of the exposed unit. Each PS-matched pair was followed up from the index date until the earliest of COVID-19 hospitalization or censoring (death, 3 or 6 months from the index date, end-of-study period, or initiation of a subsequent dose of AZD7442 in the exposed or unexposed unit). For the primary analysis, we included a maximum follow-up time of up to 6 months from the index date. Furthermore, we conducted an additional analysis including follow-up of up to 3 months from the index date. Calendar time Cox proportional hazards regression models were used with adjustment for post-matched unbalanced covariates to estimate hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) for the association between AZD7442 exposure and COVID-19 hospitalization; effectiveness was estimated as 1 − HR. Two sets of HRs with 95% CI and the corresponding effectiveness measures were reported: (1) from the PS-matched population with no additional regression adjustment; and (2) from the PS-matched population with additional regression adjustment for unbalanced variables (SMD > 0.1).
In a pre-specified, exploratory analysis, we plotted non-parametric smoothed estimates of the instantaneous hazards for the AZD7442-exposed group and the PS-matched unexposed group to further determine the duration of protection conferred by AZD7442.
AZD7442 effectiveness against COVID-19 hospitalization was also estimated during pre-defined SARS-CoV-2 variant periods, by subdividing the study period into smaller calendar time units based on the dominant circulating SARS-CoV-2 variant at the time: BA.2 (from 13 March to 20 June 2022), BA.5 (from 21 June to 7 November 2022), BQ.1 (from 8 November to 31 December 2022), and BA.2 and BA.5 combined (from 13 March to 7 November 2022). Follow-up time within SARS-CoV-2 variant periods was calculated for each PS-matched pair to only include the person-time at risk that overlapped with that dominant-variant period. COVID-19 hospitalizations that occurred outside the dominant-variant period were censored.
Results
Participants
The initial dataset included 19,528 individuals who were eligible for AZD7442 as PrEP; 3500 received AZD7442, and 16,028 were unexposed (Table 1). Of the 3500 exposed individuals, 717 did not meet the eligibility criteria, 118 were excluded due to meeting one or more exclusion criteria, and 221 were utilized to construct the first PS model for the rolling cohort. Therefore, 2444 individuals who met the eligibility criteria were included in the final analysis and were PS-matched to 2444 unexposed individuals. The PS-matched population was well balanced (SMD < 0.1) for most of the covariates included in the PS estimation model (Supplementary Table S2). The covariates that remained unbalanced (SMD > 0.1) even after PS matching were chronic kidney disease, organ transplant, age (12–29 years), sex, district (Southern district and Eilat), and number of physician visits in the prior year (0–10 and ≥ 31).
Table 1.
Population demographics and baseline characteristics
| Characteristic | Overall (unmatched) populationa | PS-matched population | SMD | ||
|---|---|---|---|---|---|
| AZD7442 exposed n = 3500 |
Unexposedb n = 16,028 |
AZD7442 exposed n = 2444 |
Unexposedb n = 2444 |
||
| Age, years, mean (SD) | 65.6 (13.5) | 61.6 (17.9) | 65.4 (13.3) | 63.3 (15.7) | NA |
| Female, no. (%) | 1513 (43.2) | 7837 (48.9) | 1024 (41.9) | 1153 (47.2) | 0.106 |
|
Immunocompromising condition, no. (%) NA | |||||
| Thalassemia | 5 (0.1) | 21 (0.1) | 5 (0.2) | 5 (0.2) | 0.000 |
| Hypogammaglobinemia with IVIG | 83 (2.4) | 168 (1.0) | 51 (2.1) | 26 (1.1) | NA |
| Hypogammaglobinemia with CLL | 25 (0.7) | 59 (0.4) | 16 (0.7) | 15 (0.6) | NA |
| CLL-treated | 95 (2.7) | 186 (1.2) | 62 (2.5) | 42 (1.7) | NA |
| Anti-CD20 | 835 (23.9) | 2,561 (16.0) | 577 (23.6) | 594 (24.3) | NA |
| Allogenic BMT | 96 (2.7) | 425 (2.7) | 74 (3.0) | 61 (2.5) | NA |
| Autologous BMT | 50 (1.4) | 133 (0.8) | 38 (1.6) | 23 (0.9) | NA |
| CAR-T therapy | 9 (0.3) | 14 (< 0.1) | 8 (0.3) | 1 (< 0.1) | NA |
| Aggressive lymphoma | 717 (20.5) | 7,985 (49.8) | 606 (24.8) | 861 (35.2) | NA |
| Multiple myeloma treated | 385 (11.0) | 1,165 (7.3) | 347 (14.2) | 169 (6.9) | NA |
| Hematologic malignancy | 1,195 (34.1) | 9,330 (58.2) | 1,013 (41.4) | 1,071 (43.8) | 0.048 |
| BMT | 141 (4.0) | 547 (3.4) | 107 (4.4) | 82 (3.4) | 0.053 |
| Organ transplant | 977 (27.9) | 3,973 (24.8) | 908 (37.2) | 792 (32.4) | 0.100 |
| Primary immunodeficiency | 88 (2.5) | 209 (1.3) | 55 (2.3) | 34 (1.4) | 0.064 |
| Secondary immunodeficiency | 841 (24.0) | 2,570 (16.0) | 582 (23.8) | 594 (24.3) | 0.011 |
| Comorbidity, no. (%) | |||||
| CKD | 666 (19.0) | 2,214 (13.8) | 565 (23.1) | 435 (17.8) | 0.132 |
| COPD | 95 (2.7) | 344 (2.1) | 56 (2.3) | 48 (2.0) | 0.023 |
| Asthma | 27 (0.8) | 106 (0.7) | 17 (0.7) | 14 (0.6) | 0.015 |
| Other respiratory disease | 60 (1.7) | 142 (0.9) | 39 (1.6) | 33 (1.4) | 0.020 |
| Heart disease | 278 (7.9) | 1,221 (7.6) | 206 (8.4) | 196 (8.0) | 0.015 |
| Cerebrovascular disease | 125 (3.6) | 580 (3.6) | 100 (4.1) | 91 (3.7) | 0.019 |
| Hypertension | 226 (6.5) | 969 (6.0) | 153 (6.3) | 159 (6.5) | 0.010 |
| Diabetes type 1 | 17 (0.5) | 85 (0.5) | 21 (0.9) | 18 (0.7) | 0.014 |
| Diabetes type 2 | 406 (11.6) | 1,681 (10.5) | 318 (13.0) | 305 (12.5) | 0.016 |
| Neurological disease | 195 (5.6) | 1,047 (6.5) | 132 (5.4) | 143 (5.9) | 0.020 |
| Liver disease | 153 (4.4) | 633 (3.9) | 123 (5.0) | 104 (4.3) | 0.037 |
| Overweight | 1381 (39.5) | 5751 (35.9) | 951 (38.9) | 947 (38.7) | 0.003 |
| Obesity | 780 (22.3) | 3775 (23.6) | 553 (22.6) | 566 (23.2) | 0.013 |
| Severe obesity | 58 (1.7) | 339 (2.1) | 35 (1.4) | 49 (2.0) | 0.044 |
| Previously smoked | 1190 (34.0) | 4359 (27.2) | 845 (34.6) | 778 (31.8) | 0.058 |
| Currently smokes | 285 (8.1) | 2104 (13.1) | 201 (8.2) | 237 (9.7) | 0.052 |
BMT bone marrow transplant, CAR-T chimeric antigen receptor T-cell, CKD chronic kidney disease, CLL chronic lymphocytic leukemia, COPD chronic obstructive pulmonary disease, IVIG intravenous immunoglobulin, NA not applicable, PS propensity score, SD standard deviation, SMD standardized mean difference
aThe overall population included 18,811 Clalit members who were eligible to receive AZD7442 and 717 Clalit members who received AZD7442 despite being ineligible
bThe unexposed population were those who were eligible to receive AZD7442 but did not receive it; they are considered the comparison group
For the PS-matched population, mean age was approximately 64 years and approximately 45% were female (Table 1). The most common immunocompromising conditions were hematological malignancy, organ transplant, aggressive lymphoma, anti-CD20 therapy, and secondary immunodeficiency. Common non-immunocompromising comorbidities were chronic kidney disease, type 2 diabetes, and heart disease.
Effectiveness of AZD7442 as PrEP Against COVID-19 Hospitalization
In the primary analysis of 6 months of follow-up, there were 30 (1.2%) COVID-19 hospitalization events in the AZD7442-exposed group [incidence rate (IR): 3.15 per 100 person-years (PY); 95% CI 2.02–4.28] and 44 (1.8%) events in the matched unexposed group (IR: 4.64 per 100 PY; 95% CI 3.27–6.01) (Fig. 1). In the PS-matched population, the unadjusted HR (without adjustment for the unbalanced covariates) for COVID-19 hospitalization among AZD7442-exposed versus unexposed individuals was 0.68 (95% CI 0.43–1.08), and the covariate-adjusted HR was 0.64 (95% CI 0.40–1.03) (Table 2). Results by circulating-variant period are shown in Table 2. Analyses of the BA.2 and BA.5 time periods showed observed reductions in COVID-19 hospitalization among AZD7442-exposed versus unexposed individuals based on HRs, although 95% CI crossed 1.0 (i.e., they were not statistically significant) (Table 2). Analysis of the BQ.1 period did not show reductions in COVID-19 hospitalization, although both groups experienced a very small number of events during this follow-up period (four events in each group).
Fig. 1.
Kaplan–Meier of COVID-19 hospitalizations for the AZD7442-exposed and unexposed groups up to 6 months of follow-up
Table 2.
Effectiveness of AZD7442 as PrEP against COVID-19 hospitalization
| Matched population | Covariate-adjusteda HR (95% CI) |
|||||
|---|---|---|---|---|---|---|
| AZD7442 exposed | Unexposed | Unadjusted HR (95% CI) | ||||
| N | COVID-19hospitalizations, no. (%) | N | COVID-19 hospitalizations, no. (%) | |||
| Overall (up to 6 months’ follow-up) | 2444 | 30 (1.2) | 2444 | 44 (1.8) | 0.68 (0.43–1.08) | 0.64 (0.40–1.03) |
| Overall (up to 3 months’ follow-up) | 2444 | 10 (0.4) | 2444 | 23 (0.9) | 0.43 (0.21–0.91) | NEb |
| Variant analysisc | ||||||
| BA.2 period | 1547 | 3 (0.2) | 1547 | 7 (0.5) | 0.43 (0.11–1.65) | NEb |
| BA.5 period | 2154 | 23 (1.1) | 2151 | 33 (1.5) | 0.69 (0.41–1.18) | 0.66 (0.39–1.13) |
| Combined BA.2 and BA.5 period | 2284 | 26 (1.1%) | 2284 | 40 (1.8) | 0.65 (0.40–1.06) | 0.62 (0.38–1.02) |
| BQ.1 period | 1068 | 4 (0.4) | 1065 | 4 (0.4) | 1.00 (0.25–3.99) | NEb |
CI confidence interval, HR hazard ratio, NE not evaluable, PrEP pre-exposure prophylaxis
aAdditional regression adjustment for unbalanced covariates. Variables are: chronic kidney disease, organ transplant, age 12–19 years, gender (male), southern district and Eilat region, category of 0–10 physician visits in the previous year, and category of 31+ physician visits in the previous year
bInsufficient events for adjusted analysis
cBA.2 period: 13 Mar to 20 Jun 2022; BA.5 period: 21 Jun to 7 Nov 2022; BQ.1 period: 8 Nov to 31 Dec 2022
A plot of the covariate-adjusted (excluding the covariate of age 12–19 years, for which 95% CIs were not estimable) instantaneous hazards for the AZD7442-exposed and unexposed groups against COVID-19 hospitalization up to 6 months showed that the AZD7442-exposed group was associated with lower instantaneous hazards for most of the follow-up period (Fig. 2). Effectiveness of AZD7442 was seen until approximately Day 90, after which point the 95% CI of the two instantaneous hazards plots overlapped. By Day 180, the exposed and unexposed groups experienced COVID-19 hospitalization events at a similar rate.
Fig. 2.
Instantaneous hazards plots of COVID-19 hospitalizations for the AZD7442-exposed and unexposed groups up to 6 months of follow-up; dashed lines represent 95% confidence intervals
In the exploratory analysis of up to 3 months of follow-up, there were 10 (0.4%) COVID-19 hospitalization events in the AZD7442 exposed group (IR: 1.79 per 100 PY; 95% CI 0.68–2.91) and 23 (0.9%) events in the unexposed group (IR: 4.14 per 100 PY; 95% CI 2.45–5.83) (Fig. 3). The unadjusted model yielded a HR for COVID-19 hospitalization of 0.43 (95% CI 0.21–0.91) for AZD7442-exposed versus unexposed individuals (Table 2). A covariate-adjusted analysis was not feasible due to the small number of events observed.
Fig. 3.
Kaplan–Meier of COVID-19 hospitalizations for the AZD7442-exposed and unexposed groups up to 3 months of follow-up
Discussion
This retrospective cohort study described the real-world effectiveness of AZD7442 300 mg against COVID-19 hospitalizations in a large integrated payer/provider healthcare organization in Israel. The study was conducted across multiple immunocompromised populations and follow-up periods and during different circulating variant periods (BA.2, BA.5, and BQ.1). The PS-matched analysis of COVID-19 hospitalizations up to 6 months suggested a lack of effectiveness of AZD7442 300 mg over the entire follow-up period, although this may have been impacted by the small number of events observed. However, for the same follow-up time (i.e., a maximum of 6 months’ follow-up) and adjustment for unbalanced covariates, plots of instantaneous hazards showed reduced COVID-19 hospitalization associated with the AZD7442-exposed group, compared with the unexposed group, until about Day 90. By Day 180, no difference in event rate between AZD7442-exposed and unexposed individuals was observed, suggesting early effectiveness of AZD7442 300 mg that waned over time.
In the PS-matched population, the exploratory unadjusted analysis up to 3 months of follow-up suggested a reduction in the rate of COVID-19 hospitalizations for individuals who received AZD7442 300 mg compared with unexposed individuals, although the small number of events that were observed precluded covariate-adjusted analysis of this association. Therefore, the results from this exploratory analysis are subject to greater residual confounding, especially from the unbalanced covariates that could not be adjusted for.
Analyses of the BA.2 and BA.5 variant periods also suggested reductions in COVID-19 hospitalizations, although analyses were also impacted by the small number of events observed.
Our study used a PS-matching procedure to balance the cohorts on potential baseline confounders and minimize the impact of confounding. Covariate balance was achieved for most baseline demographic and clinical characteristics, although a few covariates showed deviations, requiring additional adjustment in the Cox proportional hazards regression model. However, there is a risk that residual confounding remained even after covariate adjustment.
We present plots of instantaneous hazards in the AZD7442-exposed and unexposed groups to complement the fully adjusted HR (95% CI) from the Cox proportional hazards regression for the primary analysis over 6-months’ follow-up. Instantaneous hazards plots help visualize the time-varying nature of AZD7442 effectiveness, something that is missed if the HRs from the Cox proportional hazards regression model are used as the sole summary measure of treatment effect from a cohort study [26].
Previous real-world studies have showed that AZD7442 was effective at reducing the risk of SARS-CoV-2 infection and COVID-19 hospitalization for specific Omicron variants [20–23]. The VALOR-C19 observational study included data from the US Department of Veterans Affairs Health Care system and covered a 12-month period, including several variant waves [27]. Using similar methods to the present analysis, effectiveness against COVID-19 hospitalization was shown among immunocompromised individuals who received AZD7442 600 mg versus matched unexposed controls [27]. Furthermore, the VALOR-C19 study showed that AZD7442 600 mg was effective against COVID-19 hospitalizations in immunocompromised individuals when susceptible variants dominated (BA.2), but was less effective against other circulating variants (BA.5, BQ.1, and XBB) [27], which is consistent with reported in vitro neutralization data [28, 29].
Strengths of this study include that it was undertaken using a comprehensive database from the largest health maintenance organization in Israel, representing > 50% of the population. The database includes many individuals with a broad range of immunocompromising conditions, suggesting our findings are generalizable to a wider population of immunocompromised individuals. Additionally, the study used pre-defined standards-based operational definitions of immunocompromised conditions to determine participant eligibility.
Limitations include a low number of COVID-19 hospitalization events for assessment of the primary outcome. Information was not available to explain why the matched control group of immunocompromised individuals did not receive AZD7442, despite being seemingly eligible according to the IMoH criteria for AZD7442 as PrEP, although this may be related to limited uptake of AZD7442 in Israel [24]. The comparatively limited uptake of AZD7442 in Israel [24] is also reflected in the study sample size. As for all retrospective analyses, accuracy of the results is dependent on the availability and accuracy of the information from EHRs. Finally, this study only included the 300-mg dose of AZD7442, as was indicated in Israel, whereas a higher (600-mg) dose was introduced post-authorization in the US [30] and other countries.
Conclusion
These results suggest that AZD7442 300 mg as PrEP may have potentially reduced the risk of COVID-19 hospitalizations among immunocompromised individuals up to approximately 3 months of follow-up across circulating variant periods, after which effectiveness waned, which aligns with in vitro neutralization data [28, 29] and effectiveness results from similar studies [27]. The suspension of AZD7442 clinical use limits the direct applicability of our findings, although this study provides valuable insights that may inform the design of future mAb PrEP observational studies in immunocompromised individuals. Moreover, our findings highlight critical lessons learned about the effectiveness of specific interventions during the early phases of the COVID-19 pandemic, which can inform future therapeutic strategies and their development. Transparency in science is vital, and we hope this paper underscores the importance of rapid innovation and implementation during global health crises.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgments
Medical Writing/Editorial Assistance
Medical writing support was provided by Shaun W. Foley, BSc (Hons), and editorial support by Jess Fawcett, BSc, both of Core (a division of Prime, London, UK), supported by AstraZeneca according to Good Publication Practice guidelines.
Author Contributions
Samah Hayek, Joseph Levy, Galit Shaham, Noa Dagan, Sabada Dube, Idit Livnat, Sylvia Taylor, Sudhir Venkatesan, and Ran D. Balicer conceived and designed the study. Joseph Levy and Galit Shaham participated in data extraction and analysis. Danielle Serby, Hadar Duskin-Bitan, Doron Netzer, and Alon Peretz contributed to data analysis and clinical interpretation. Samah Hayek, Sabada Dube, and Sudhir Venkatesan contributed to drafting the manuscript. Noa Dagan, Danielle Serby, Hadar Duskin-Bitan, Ran D. Balicer, Doron Netzer, and Alon Peretz provided clinical guidance. All authors critically reviewed the manuscript and decided to proceed with publication.
Funding
The study and the Rapid Service Fee were funded by AstraZeneca.
Data Availability
National and organizational data privacy policies prohibit the sharing of individual-level data, such as those used for this study, even if anonymized.
Declarations
Conflicts of Interest
The authors declare the following potential conflicts of interest with respect to the research, authorship and/or publication of this article: Samah Hayek, Noa Dagan, Danielle Serby, Hadar Duskin-Bitan, Ran D. Balicer, Doron Netzer, and Alon Peretz are employees of Clalit Health Services. Clalit Research Institute received funding from AstraZeneca to support the execution of this study. Joseph Levy and Galit Shaham were employees of Clalit Health Services at the time of this study. Dr Levy’s current affiliation is the Hebrew University of Jerusalem. Dr Shaham’s current affiliation is J2 Health. Sabada Dube, Cátia Ferreira, Idit Livnat, Carla Talarico, Sylvia Taylor, Sudhir Venkatesan, and Adva Yarden are employees of, and may hold stock and/or stock options in, AstraZeneca. The AstraZeneca team was blinded to the raw data and results during the data extraction and analysis phases and only had access to aggregated data as a Word document. Data extraction, analysis, and interpretation were conducted independently by the Clalit Health Services team, and discussions between both teams took place during the protocol development and manuscript preparation phases to ensure transparency and scientific rigor.
Ethical Approval
This study used de-identified patient data and was approved by the Clalit Health Services institutional review board.
Footnotes
Prior presentation: This analysis was previously presented at ECCMID 2024, 27–30 April 2024, Barcelona, Spain.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.World Health Organization. WHO COVID-19 dashboard. https://covid19.who.int/. Accessed November 22 2024.
- 2.Dagan N, Barda N, Kepten E, et al. BNT162b2 mRNA Covid-19 vaccine in a nationwide mass vaccination setting. N Engl J Med. 2021;384(15):1412–23. 10.1056/NEJMoa2101765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Denford S, Mowbray F, Towler L, et al. Exploration of attitudes regarding uptake of COVID-19 vaccines among vaccine hesitant adults in the UK: a qualitative analysis. BMC Infect Dis. 2022;22(1):407. 10.1186/s12879-022-07380-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bergwerk M, Gonen T, Lustig Y, et al. Covid-19 breakthrough infections in vaccinated health care workers. N Engl J Med. 2021;385(16):1474–84. 10.1056/NEJMoa2109072. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Menegale F, Manica M, Zardini A, et al. Evaluation of waning of SARS-CoV-2 vaccine-induced immunity: a systematic review and meta-analysis. JAMA Netw Open. 2023;6(5): e2310650. 10.1001/jamanetworkopen.2023.10650. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Prendecki M, Clarke C, Edwards H, et al. Humoral and T-cell responses to SARS-CoV-2 vaccination in patients receiving immunosuppression. Ann Rheum Dis. 2021;80(10):1322–9. 10.1136/annrheumdis-2021-220626. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lee ARYB, Wong SY, Chai LYA, et al. Efficacy of COVID-19 vaccines in immunocompromised patients: systematic review and meta-analysis. BMJ (Clinical research ed). 2022;376: e068632. 10.1136/bmj-2021-068632. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Tartof SY, Slezak JM, Puzniak L, et al. Immunocompromise and durability of BNT162b2 vaccine against severe outcomes due to omicron and delta variants. Lancet Respir Med. 2022;10(7):e61–2. 10.1016/S2213-2600(22)00170-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Williamson EJ, Walker AJ, Bhaskaran K, et al. Factors associated with COVID-19-related death using OpenSAFELY. Nature. 2020;584(7821):430–6. 10.1038/s41586-020-2521-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Evans RA, Dube S, Lu Y, et al. Impact of COVID-19 on immunocompromised populations during the Omicron era: insights from the observational population-based INFORM study. Lancet Reg Health Eur. 2023;35: 100747. 10.1016/j.lanepe.2023.100747. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Bahremand T, Yao JA, Mill C, Piszczek J, Grant JM, Smolina K. COVID-19 hospitalisations in immunocompromised individuals in the Omicron era: a population-based observational study using surveillance data in British Columbia, Canada. Lancet Region Health Am. 2023;20: 100461. 10.1016/j.lana.2023.100461. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Agrawal U, Bedston S, McCowan C, et al. Severe COVID-19 outcomes after full vaccination of primary schedule and initial boosters: pooled analysis of national prospective cohort studies of 30 million individuals in England, Northern Ireland, Scotland, and Wales. Lancet. 2022;400(10360):1305–20. 10.1016/S0140-6736(22)01656-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Brosh-Nissimov T, Hussein K, Wiener-Well Y, et al. Hospitalized patients with severe coronavirus disease 2019 during the Omicron wave in Israel: benefits of a fourth vaccine dose. Clin Infect Dis. 2023;76(3):e234–9. 10.1093/cid/ciac501. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Kmietowicz Z. Covid-19: Monoclonal antibodies authorised in US as alternative to vaccines for certain groups. BMJ. 2021;375: n3064. 10.1136/bmj.n3064. [DOI] [PubMed] [Google Scholar]
- 15.Keam SJ. Tixagevimab + cilgavimab: first approval. Drugs. 2022;82(9):1001–10. 10.1007/s40265-022-01731-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.United States Food and Drug Administration. Fact sheet for healthcare providers: emergency use authorization for Evusheld™ (tixagevimab co-packaged with cilgavimab). https://www.fda.gov/media/154701/download. Accessed November 8 2024.
- 17.Israeli Ministry of Health. The Ministry of Health has instructed the HMOs to vaccinated immunosuppressed individuals with AstraZeneca's Evusheld vaccine. https://www.gov.il/en/pages/15022022-02. Accessed August 5 2024.
- 18.United States Food and Drug Administration. FDA announces Evusheld is not currently authorized or emergency use in the U.S. https://www.fda.gov/drugs/drug-safety-and-availability/fda-announces-evusheld-not-currently-authorized-emergency-use-us. Accessed April 23 2024.
- 19.Levin MJ, Ustianowski A, De Wit S, et al. Intramuscular AZD7442 (tixagevimab-cilgavimab) for prevention of COVID-19. N Engl J Med. 2022;386(23):2188–200. 10.1056/NEJMoa2116620. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Najjar-Debbiny R, Gronich N, Weber G, Stein N, Saliba W. Effectiveness of Evusheld in immunocompromised patients: propensity score-matched analysis. Clin Infect Dis. 2023;76(6):1067–73. 10.1093/cid/ciac855. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Chen B, Haste N, Binkin N, et al. Real world effectiveness of tixagevimab/cilgavimab (Evusheld) in the Omicron era. PLoS ONE. 2023;18(4): e0275356. 10.1371/journal.pone.0275356. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Al Jurdi A, Morena L, Cote M, Bethea E, Azzi J, Riella LV. Tixagevimab/cilgavimab pre-exposure prophylaxis is associated with lower breakthrough infection risk in vaccinated solid organ transplant recipients during the omicron wave. Am J Transplant. 2022;22(12):3130–6. 10.1111/ajt.17128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kertes J, Shapiro Ben David S, Engel-Zohar N, et al. Association between AZD7442 (tixagevimab-cilgavimab) administration and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, hospitalization, and mortality. Clin Infect Dis. 2023;76(3):e126–32. 10.1093/cid/ciac625. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Hayek S, Levy J, Shaham G, et al. Implementation of AZD7442 (tixagevimab/cilgavimab) COVID-19 pre-exposure prophylaxis (PrEP) in the largest health maintenance organization in Israel: real-world uptake and sociodemographic and clinical characteristics across immunocompromised patient groups. Infect Dis Ther. 2024;13(6):1379–89. 10.1007/s40121-024-00981-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Barda N, Dagan N, Ben-Shlomo Y, et al. Safety of the BNT162b2 mRNA Covid-19 vaccine in a nationwide setting. N Engl J Med. 2021;385(12):1078–90. 10.1056/NEJMoa2110475. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Hernán MA. The hazards of hazard ratios. Epidemiology. 2010;21(1):13–5. 10.1097/EDE.0b013e3181c1ea43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.DuVall SL, Lynch JA, Alba PR, et al. Clinical effectiveness of AZD7442 (tixagevimab/cilgavimab) against COVID-19 hospitalisation among immunocompromised patients in the US Department of Veterans Affairs health system: overall and during periods of susceptible vs resistant Omicron variants. ECCMID 2023. Copenhagen, Denmark; 2023.
- 28.Arora P, Kempf A, Nehlmeier I, et al. Omicron sublineage BQ.1.1 resistance to monoclonal antibodies. Lancet Infect Dis. 2023;23(1):22–3. 10.1016/S1473-3099(22)00733-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Imai M, Ito M, Kiso M, et al. Efficacy of antiviral agents against Omicron subvariants BQ.1.1 and XBB. N Engl J Med. 2023;388(1):89–91. 10.1056/NEJMc2214302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.United States Food and Drug Administration. Frequently asked questions on the Emergency Use Authorization for Evusheld (tixagevimab co-packaged with cilgavimab) for pre-exposure prophylaxis (PrEP) of COVID-19. https://www.fda.gov/media/154703/download. Accessed 19 August 2024.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
National and organizational data privacy policies prohibit the sharing of individual-level data, such as those used for this study, even if anonymized.