Safety of heterologous ChAdOx1-S/BNT162b2 primary schedule versus homologous BNT162b2 vaccination: Insights from an Italian post-marketing study, 2021

ABSTRACT

During COVID-19 vaccination campaign, possible ChAdOx1-S-associated risks of thrombosis with thrombocytopenia syndrome led to implement ChAdOx1-S/BNT162b2 heterologous vaccination, despite the limited information on its reactogenicity and safety. We conducted a prospective observational post-marketing surveillance study to assess the safety of this heterologous schedule. A casually selected sample of recipients (n: 85; age: 18–60 years) of ChAdOx1-S/BNT162b2 at the vaccination hub of the Foggia Hospital, Italy, was matched with an equal sample of recipients of homologous BNT162b2. Safety was evaluated 7 days, 1 month and 14 weeks after the primary vaccination series using an adapted version of the “V-safe active surveillance for COVID-19 vaccine safety” CDC standardized questionnaire. After 7 days, local reactions were highly frequent (>80%) in both groups, and systemic reactions were less common (<70%). Moderate or severe pain at the injection site (OR = 3.62; 95%CI, 1.45–9.33), moderate/severe fatigue (OR = 3.40; 95%CI, 1.22–9.49), moderate/severe headache (OR = 4.72; 95%CI, 1.37–16.23), intake of antipyretics (OR = 3.05; 95 CI%, 1.35–6.88), inability to perform daily activities and work (OR = 2.64; 95%CI, 1.24–5.62) were significantly more common with heterologous than homologous vaccination. No significant difference in self-reported health status was recorded 1 month or 14 weeks after the second dose with BNT162b2 or ChAdOx1-S/BNT162b2. Our study confirms the safety of both heterologous and homologous vaccination, with a slight increase in some short-term adverse events for the heterologous regimen. Therefore, administering a second dose of a mRNA vaccine to the recipients of a previous dose of viral vector vaccine may have represented an advantageous strategy to improve flexibility and to accelerate the vaccination campaign.

Introduction

During the Spring of 2021, some uncertainties on the ChAdOx1-S-associated risk of thrombosis with thrombocytopenia syndrome^1,2 led several countries to recommend the use of mRNA vaccines as the second dose in individuals who had previously received an adenoviral vector vaccine as their first dose.^3–6 In June 2021, also the Technical and Scientific Committee of the Italian Medicines Agency (AIFA) expressed a favorable opinion on the use of the heterologous schedule in subjects under 60 years of age who had received the first dose of ChAdOx1-S.⁷

National decisions of completing the primary course of COVID-19 vaccination with an alternative vaccine as second dose have been progressively corroborated by an increasing number of clinical studies^4,8−12 and real-world evidence.^9–17 Emerging data indicate that heterologous primary vaccination has the potential to improve the immune response compared with homologous vaccination. Indeed, the combination of a viral vector vaccine and mRNA vaccine produces good levels of antibodies against SARS-CoV-2 and a higher T-cell response than the homologous vaccination. These results support the consideration of combination strategies to maximize the protection level.^{9–17,18,19–21} On the other hand, information on the reactogenicity and safety of heterologous vaccine regimens is still limited due to the scanty availability of data from both dedicated trials and real-life studies. Few studies^4,5,16 have focused explicitly on this issue, the majority of which evaluated reactogenicity only as a secondary outcome, mostly analyzing short-term adverse events (AEs) and not considering the impact on long-term quality of life.^8–15 To cover these unmet needs, we conducted a prospective observational post-marketing study. Short- and long-term safety profiles, including self-reported impact of daily activities, of heterologous primary schedule (first-dose ChAdOx1-S followed by a second dose BNT162b2) compared with homologous two-dose BNT162b2 vaccination were assessed.

Materials and methods

Study design, setting, and participants

Between 15 June and 30 October 2021, we conducted a prospective observational post-marketing surveillance study at the Policlinico Foggia Hospital, the largest COVID-19 vaccination hub in the province of Foggia, Italy, serving approximately 680,000 inhabitants. Over 400 doses per day were administered during the vaccination campaign, which amounts to a total of approximately 125,000 doses until November 2022.

Eligible subjects for the study were between 18 and 60 years of age and had received a heterologous vaccination (a ChAdOx1-S priming dose followed by secondary vaccination with the BNT162b2 vaccine), according to the Ministry of Health recommendations,⁷ or a homologous schedule using two doses of BNT162b2. The time interval between the first and the second dose was determined as per national guideline⁷ and was 8–12 weeks for the heterologous ChAdOx1-S/BNT162b2 regimen and 3 weeks for the homologous BNT162b2 regimen. Individuals with a known history of SARS-CoV-2 infection were not able to be enrolled in the study.

A sample of individuals who had received ChAdOx1-S/BNT162b2 vaccination was casually selected to be enrolled in the study among all those who had received the second dose of COVID-19 vaccination during the first month since the heterologous vaccine regimen had been authorized (15 June−14 July 2021). An equal number of persons who had received homologous BNT162b2 vaccination were matched to ChAdOx1-S/BNT162b2 recipients by the date of the second dose, sex, and age. A flow diagram describing the selection of participants and their enrollment into the study is reported in Figure 1.

Figure 1.

Flow diagram describing the selection of participants and their enrollment into the study. Foggia, Italy, 15 June−14 July 2021.

Data sources and outcomes

Demographic and clinical characteristics (i.e. comorbidities, allergy to drugs and/or other vaccines, etc.) and vaccination data of the enrolled subjects were retrieved through the regional Immunization Information System.

Safety was evaluated in terms of reactogenicity and impact of adverse events (AEs) on daily activities. Reactogenicity was monitored as the occurrence of AEs following the second vaccine dose within 7 days after the vaccination. The impact of the AEs on the ability to perform normal daily activities and work, and the need to receive medical care were evaluated 7 days, 1 month, and 14 weeks after the primary vaccination series. Subjects who had provided informed consent to participate in the study at the time of the second dose were contacted via telephone for the follow-up interviews. In case of missed response, they were recontacted twice over a day to minimize the possibility of a response bias.

The interviews were conducted using a structured questionnaire based on the US CDC web-based survey “V-safe active surveillance for COVID-19 vaccine safety” - version 2,²² translated into Italian and culturally adapted for the purposes of the study. The main outcome of the questionnaire was to investigate the severity (mild, moderate, and severe) of self-reported AEs, both local (pain, redness, swelling, itching) and systemic (chills, headache, joint pain, muscle or body aches, fatigue/tiredness, nausea, vomiting, diarrhea, abdominal pain, rash, fever, any other symptoms/health conditions that the subject had put in relation to the vaccination). The questionnaire also explored the impact of adverse reactions on the ability to perform normal daily activities and work, and the need to receive medical care. According to the questionnaire, symptoms were classified as follows: mild = “you notice symptoms, but they aren’t a problem”; moderate = “symptoms that limit of your normal daily activities”; severe = symptoms make normal daily activities difficult or impossible.” Additionally, the use of antipyretic medications (i.e., acetaminophen and nonsteroidal anti-inflammatory drugs) after vaccination was investigated and recorded. Frequencies of mild or moderate/severe local and systemic AEs, of their impact on normal daily activities and work, and the need to receive medical care (including use of antipyretic medications) 7 day after having received the second dose of vaccine were calculated as the ratio of participants with the self-reported condition to the total number of participants for both heterologous and homologous vaccination schedules. Similarly, the AEs impact on normal daily activities and work and the need to receive medical care at 1 month and 14 weeks after the primary vaccination series were calculated.

Statistical analysis

Descriptive statistics were performed. Categorical variables (sex, comorbidities, allergy to drugs, food and/or vaccines, pregnancy, breastfeeding, self-reported local and systemic adverse events, impact of adverse reactions on the ability to perform normal daily activities and work, use of analgesic/antipyretic medications) were expressed as counts and percentages in each category. Frequencies were expressed as percentages with 95% confidence interval (CI) calculated using the Clopper−Pearson method. Continuous variables (age, days of fever, temperature) were defined as means (standard deviation [SD]).

Direct comparisons of self-reported AEs frequency between heterologous and homologous regimen receiving groups were done in a univariate analysis computing the chi-square test (χ²) and the odds ratio (ORs) with 95% CIs or Fisher’s exact test. Differences in continuous variables were tested with Student’s t test for normally distributed ones, or the Mann–Whitney U-test when variables showed a non-normal distribution.

When the univariate analysis retrieved significant differences for some variables, these were included in multivariate logistic regression models to evaluate whether the heterologous vs homologous vaccine regimen, as well as demographics and clinical characteristics (sex: male vs female; age group: above vs below the median age, comorbidity: yes vs no, allergy history: yes vs no) were independently associated with the self-reporting AEs.

The sample size was calculated based on the assumption that ≥80% of individuals who received ChAdOx1-S/BNT162b2 would have reported ≥1 AEs, with a two-sided type I error of 0.001 and a power of 90%.²³ The sample estimation considered a dropout rate of 15%. The statistical power was calculated and tabulated for those outcomes for which a statistical association in the multivariate models was observed. The level of statistical significance was set at p < .05. All analyses were conducted with STATA/SE 15.0.

Ethics approval and consent to participate

The study was conducted in accordance with the Guidelines for Good Clinical Practice and the ethical principles that originate in the Declaration of Helsinki and within the Italian law. According to the Italian Data Protection Authority, this observational study took advantage of a derogation regimen for the duration of the COVID-19 emergency (art. 110 of the Code regarding the protection of personal data).

Informed consent was obtained from all participants prior to enrollment in the study. Data were provided and analyzed anonymously.

Results

Between 15 June and 14 July 2021, a total of 244 individuals received heterologous ChAdOx1/BNT162b2 vaccination during COVID-19 vaccination practice; of these, 85 (34.8%) were casually selected to be enrolled in the study and 74 agreed to participate in the follow-up telephone interviews; 11 (12.9%) subjects did not agree to be enrolled into the study. A total of 85 to 2,734 (3.1%) individuals who had received homologous BNT162b2 vaccination were matched to ChAdOx1-S/BNT162b2 recipients by date of the second dose, sex, and age; of these, 65 agreed and provided informed consent to participate in the study; 20 (23.5%) refused to be included (Figure 1). Baseline characteristics of the study population are presented in Table 1.

Table 1.

Participants’ demographic characteristics and health status before vaccination. Foggia, Italy, 15 June−14 July 2021.

	Individuals who received heterologous ChAdOx1/BNT162b2 vaccination (n = 74)		Individuals who received homologous BNT162b2 vaccination (n = 65)		Univariate analysis
	n	%	n	%	Odds Ratio (95% CI)	p
Sex
Male	21	28.38	20	30.77	0.89 (0.40–1.97)	.7578
Female	53	71.62	45	69.23
Age, years
Mean age (SD)	34.7 (11.75)		36.1 (12.06)			.2448
Median age (IQR), years	32.5 (23–46.25)		38 (24–47)
Reported allergies
At least one allergy	18	24.32	15	23.08	1.07 (0.45–2.54)	.8631
Drug allergies (yes)	11	14.86	8	12.31	0.75 (0.24–2.37)	.5858
Food allergies (yes)	8	10.81	6	9.23	1.20 (0.33–4.42)	.7574
Others (yes)*	5	6.76	2	3.08	2.28 (0.35–24.63)	.2775
≥1 Chronic health problem (yes)**	13	17.57	11	16.92	1.04 (0.39–2.81)	.9201
≥1 Immune system disorders (yes)***	0	0.00	0	0.00
Medications in the past 3 months (yes)****	0	0.00	1	1.54
Vaccinations in the past 4 weeks (yes)	0	0.00	2	3.08

n, number of participants in the group; 95% CI, 95% confidence interval.

* e.g., a vaccine component, or latex.

** Chronic health problem included: heart, lung, kidney, or metabolic (e.g., diabetes) diseases; asthma; blood disorders; asplenia; complement component deficiencies; having a cochlear implant or a spinal fluid leak.

*** Immune system disorders included: cancer; leukemia; lymphoma; HIV/AIDS.

**** Medications that weaken the immune system: prednisone, other steroids, or anticancer drugs; drugs for the treatment of rheumatoid arthritis, Crohn’s disease, or psoriasis.

None of participants reported serious AEs after a vaccination. None of female participants reported to be pregnant or breastfeeding.

Participants with limitations in daily-life and ability to work reported only moderate or severe AEs, therefore they were grouped together in the analysis, also in order to minimize the limited sample size effect. Seven days after having received the second dose, both vaccination regimens were associated with a relatively high frequency of local reactions, with ≥80% of the participants in each group reporting at least one local event. Specifically, local reactions were usually mild in both study groups. Pain at the injection site was the most frequently reported local reaction in both vaccine regimens (83.78% in heterologous and 87.69% homologous group), with a higher frequency of moderate/severe pain after heterologous ChAdOx1-S/BNT162b2 vaccination compared with the homologous one (OR = 3.62; 95% CI, 1.45 to 9.33). The frequencies of other local AEs were not different between heterologous and homologous regimen receiving groups (Table 2a).

Table 2.

Self-reported local and systemic AEs (a, b), impact of AEs on daily activities and work, and need to receive medical care (c) 7 days after having received the second dose of the heterologous ChAdOx1/BNT162b2 vaccination compared with homologous BNT162b2 regimen. Foggia, Italy, 15 June−30 October 2021.

	Individuals who received heterologous ChAdOx1/BNT162b2 vaccination		Individuals who received homologous BNT162b2 vaccination		Univariate analysis			Multivariate analysis		Power
	(n = 74)		(n = 65)
Local adverse events (a)	n	%	n	%	Odds Ratio (95% CI)	X2	p	Odds Ratio (95% CI)	p	%
At least one local AE	62	83.78	57	87.69	0.72 (0.23–2.09)	0.43	.5124
Pain (yes)	62	83.78	57	87.69	0.72 (0.23–2.09)	0.43	.5124
Mild	37	50.00	48	73.85	Ref.					80.92
Moderate/Severe	25	33.78	9	13.85	3.6 (1.40–9.77)	8.76	.0031	3.62 (1.45–9.33)	.006
Redness (yes)	5	6.76	2	3.08	2.28 (0.35–24.63)	0.98	.322
Mild	2	2.70	1	1.54	Ref.
Moderate/Severe	3	4.05	1	1.54	1.5 (0.01–156.4)	0.06	.7143
Swelling (yes)	10	13.51	4	6.15	2.38 (0.64–10.9)	2.07	.1503
Mild	5	6.76	3	4.62	Ref.
Moderate/Severe	5	6.76	1	1.54	3 (0.15–188.4)	0.73	.3932
Itching (yes)	3	4.05	1	1.54	2.70 (0.21–144.12)	0.78	.3761
Mild	2	2.70	0	0.00	-	-	-
Moderate/Severe	1	1.35	1	1.54	-	-	-
Injection-site reaction (yes)	62	83.78	57	87.69	0.72 (0.23–2.09)	0.43	.5124
Systemic adverse events (b)
At least one systemic reaction	52	70.27	42	64.62	1.29 (0.59–2.80)	0.51	.4771
Chills (yes)	11	14.86	4	6.15	2.70 (0.74–12.21)	2.82	.0785
Mild	5	6.76	4	6.15	-	-	-
Moderate/Severe	6	8.11	0	0.00	-	-	-
Headache (yes)	30	40.54	24	36.92	1.13 (0.54–2.39)	0.13	.7151
Mild	12	16.22	18	27.69	Ref.					99.15
Moderate/Severe	19	25.68	7	10.77	4.07 (1.15–14.95)	6.17	.0130	4.72 (1.37–16.23)	.014
Joint pain (yes)	21	28.38	21	32.31	0.83 (0.37–1.82)	0.81	.6147
Mild	11	14.86	16	24.62	Ref.
Moderate/Severe	10	13.51	5	7.69	2.90 (0.65–13.78)	2.59	.1074
Muscle or body aches (yes)	21	28.38	18	27.69	1.03 (0.46–2.33)	0.01	.9284
Mild	12	16.22	13	20.00	Ref.
Moderate/Severe	9	12.16	5	7.69	1.95 (0.42–9.54)	0.96	.3278
Fatigue or tiredness (yes)	44	59.46	30	46.15	1.71 (0.83–3.54)	2.46	.1167
Mild	13	17.57	17	26.15	Ref.					91.87
Moderate/Severe	31	41.89	13	20.00	3.11 (1.06–9.20)	5.44	.0196	3.40 (1.22–9.49)	.019
Nausea (yes)	7	9.46	6	9.23	1.04 (0.28–3.98)	0.01	.9427
Mild	4	5.41	5	7.69	Ref.
Moderate/Severe	3	4.05	1	1.54	3.75 (0.18–235.48)	1.04	.3427
Vomiting (yes)	3	4.05	3	4.62	0.87 (0.1–6.76)	0.03	.5962
Mild	0	0.00	1	1.54	-	-	-
Moderate/Severe	2	2.70	2	3.08	-	-	-
Diarrhea (yes)	4	5.41	1	1.54	3.65 (0.35–182.8)	1.49	.2261
Mild	2	2.70	0	0.00	-	-	-
Moderate/Severe	2	2.70	1	1.54	-	-	-
Abdominal pain (yes)	3	4.05	3	4.62	0.87 (0.11–6.76)	0.03	.5962
Mild	1	1.35	2	3.08	Ref.
Moderate/Severe	2	2.70	1	1.54	0.25 (0.002–14.83)	0.67	.5000
Rash (yes)*	2	2.70	3	4.62	0.57 (0.046–5.19)	0.37	.4386
Mild	2	2.70	3	4.62	-	-	-
Moderate/Severe	0	0.00	0	0.00	-	-	-
Fever (yes)	17	22.97	12	18.46	1.26 (0.51–3.20)	0.31	.5751
Mean (SD) days of fever	2.1 (0.87)		1.91 (0.90)				.3382
Highest temperature (°C), Mean (SD)	37.9 (0.51)		37.7 (0.47)				.2511
Other symptoms/health conditions (yes)**	9	12.16	6	9.23
Impact on daily activities/work, need to receive medical care (c)
Inability to normal daily activities (yes)	34	45.95	16	24.62	2.60 (1.19–5.78)	6.84	.0089	2.64 (1.24–5.62)	.012	77.84
Inability to work (yes)	34	45.95	16	24.62	2.60 (1.19–5.78)	6.84	.0089	2.64 (1.24–5.62)	.012	77.84
Need to receive medical care (yes)	6	8.11	6	9.23	0.86 (0.21–3.43)	0.06	.8141
Use of medications to relieve AEs (yes)	29	39.19	12	18.46	2.97 (1.28–7.15)	7.75	.0054	3.05 (1.35–6.88)	.007	84.07
Acetaminophen	29	39.19	10	15.38	3.16 (1.34–7.78)	3.16	.0038	3.29 (1.29–8.37)	.012	85.20
Non-steroidal anti-inflammatory drugs (NSAIDs)	0	0.00	1	1.54	-	-	-	-	-

n, number of participants in the group; 95% CI, 95% confidence interval.

* Not including the immediate area around the injection site.

** Swollen lymph nodes, Leg paresthesia, Abdominal pain, Chest pain, Capillary fragility, Bruises on legs, A bruise at the vaccine injection site, Syncope 4 days after vaccination.

Multivariate analysis included only those variables that were statistically significant in univariate analysis.

On days 0–7 after second dose, systemic reactions were less common than local reactions (70.27% in heterologous and 64.62% in homologous group). Moderate/severe fatigue and headache were the systemic reactions most frequently reported in all groups, being significantly more common with the heterologous ChAdOx1-S/BNT162b2 regimen than with the homologous BNT162b2/BNT162b2 regimen (OR fatigue = 3.40; 95% CI, 1.22 to 9.49; OR headache = 4.72; 95% CI, 1.37 to 16.23). The frequencies of other AEs are reported in Table 2b. Fever had a similar frequency between the two groups (OR = 1.26; 95% CI, 0.51 to 3.20), for an average duration of about 2 days after vaccination (2.1 days after ChAdOx1-S/BNT162b2; 1.91 days after BNT162b2/BNT162b2) and a maximum average temperature below 38°C (37.9°C after ChAdOx1-S/BNT162b2; 37.7°C after BNT162b2/BNT162b2) with no significant difference between the two vaccination schedules (Table 2b). Intake of antipyretic medication (mainly acetaminophen) to relieve side effects following vaccination was significantly more frequent after heterologous ChAdOx1-S/BNT162b2 (OR = 3.05; 95% CI, 1.35 to 6.88) compared with homologous vaccination (Table 2c).

During the 7 days after the second dose, inability to perform daily activities and inability to work were reported more frequently among heterologous ChAdOx1-S/BNT162b2 recipients (OR = 2.64; 95% CI, 1.24 to 5.62) (Table 2c). One month and 14 weeks after the second dose, no significant differences in self-reported health conditions were reported between recipients of the heterologous and recipients of the homologous regimen (Table 3). No potentially life-threatening reactions were reported after any of the vaccine regimens in this study.

Table 3.

Impact of AEs on daily activities and work and need to receive medical care 1 month and 14 weeks after having received the second dose of the heterologous ChAdOx1/BNT162b2 vaccination compared with homologous BNT162b2 regimen. Foggia, Italy, 15 June−30 October 2021.

	Individuals who received heterologous ChAdOx1/BNT162b2 vaccination (n = 74)		Individuals who received homologous BNT162b2 vaccination (n = 65)		Univariate analysis
	n	%	n	%	Odds Ratio (95% CI)	X2	p
1 month apart
Inability to normal daily activities (yes)	2	2.70	2	3.08	0.875 (0.062–12.04)	0.02	.6398
Inability to work (yes)	2	2.70	2	3.08	0.875 (0.062–12.04)	0.02	.6398
Need to receive medical care (yes)	1	1.35	1	1.54
14 weeks apart
Inability to normal daily activities (yes)	3	4.05	0	0.00	-
Inability to work (yes)	3	4.05	0	0.00	-
Need to receive medical care (yes)	2	2.70	1	1.54	-

n, number of participants in the group; 95% CI, 95% confidence interval.

Discussion

Crucially, this prospective observational post-marketing study that we conducted on COVID-19 vaccines at the Policlinico Foggia hub between June and October 2021 showed that both the primary schedule with the heterologous ChAdOx1-S/BNT162b2 vaccine regimen and the homologous regimen with two doses of BNT162b2 were safe.

Our analysis was based on a limited sample and exclusively focused on comparing the safety between primary heterologous ChAdOx1-S/BNT162b2 and homologous BNT162b2/BNT162b2 regimens. Overall, local and systemic adverse events reported in this study were mostly of mild intensity with similar frequency between groups, except for moderate/severe injection-site pain, fatigue and headache, more frequent in the heterologous regimen. This finding was consistent with several previously available studies.^8,9,11,13,15 In their clinical trial, Borobia et al.⁹ conducted the reactogenicity analysis at 7 days showing that 68% of the reactions were mild and 30% moderate, with injection site pain (88%) and headache (44%) among the most reported AEs. In another clinical trial, Banki et al.⁸ retrieved good tolerability and similar level of reactogenicity in heterologous ChAdOx1-S/BNT162b2, homologous ChAdOx1-S/ChAdOx1-S and BNT162b2/BNT162b2 groups, with a slightly higher frequency at 30 days of adverse events at the injection site, general discomfort, and self-reported feverishness/fever with heterologous ChAdOx1-S/BNT162b2 compared with the homologous ChAdOx1-S/ChAdOx1-S, but not with BNT162b2/BNT162b2. Among the observational studies, Hillus et al.¹³ did not find major differences in reactogenicity between heterologous and homologous regimens, while Shaw et al.,⁴ Stuart et al.,¹² Powell et al.,⁵ and Warkentin et al.¹⁶ observed a broader spectrum of local and systemic symptoms (fever or feverishness chills, fatigue, headache, joint pain, malaise, and muscle ache) after the second heterologous dose. Moreover, comparing the results of our study with safety data reported in the summary of product characteristics (SmPC) of BNT162b2²⁴ showed that the type and frequency of AEs after the heterologous vaccination are similar to those reported in the regulatory document and considered acceptable by the regulatory agencies. The differences found relate to diarrhea and skin rashes. In detail, diarrhea was reported as “very common” (≥1/10) in the SmPC and had a lower frequency in our sample (5.41%). Conversely, skin rash was reported as “uncommon” (≥1/1,000 to <1/100) in the SmPC but had a higher frequency (2.70%) after heterologous regimen ChAdOx1-S/BNT162b2 in our study.

However, we found that inability to perform daily activities and inability to work were more frequent among recipients of the heterologous ChAdOx1-S/BNT162b2 vaccination in the week following the second dose (days 0–7). Similarly, Powell et al.,⁵ on the basis of real-world data with a follow-up of at most 21 days, demonstrated that previously uninfected individuals who received heterologous schedules were 2.4 times (27.8% vs 11.6%) more likely to report severe inability to perform normal daily activities after their second dose, than those receiving homologous schedules.

The detection of post-vaccination antipyretic use in our study revealed a statistically significantly higher intake of acetaminophen after the heterologous ChAdOx1-S/BNT162b2 vaccination compared with homologous BNT162b2 vaccination, as also reported by Shaw et al.⁴ (57% of heterologous regimens vs 41% of homologous). On the one hand, the higher antipyretic use could be attributable to the increasing hands-on experience following the first vaccine dose. Indeed, it is known that the first dose of ChAdOx1-S is more reactogenic than the first dose of a mRNA vaccine¹⁴ and this may have affected the tendency to take antipyretic drugs when BNT162b2 was used for the second dose. On the other hand, this aspect is particularly relevant since taking antipyretic medication could have influenced the reactogenicity profile of vaccine regimens.

The key strength of the present study is the collection of real-world data through a validated questionnaire to assess the safety of different vaccine schedules. Secondly, our study investigated not only the frequency of adverse reactions following vaccination among the study groups but also the impact of these events on the quality of life, both in the short and mid-term confirming safety comparability between the two regimes. Several studies^5,13,14 have focused only on the assessment of local or systemic adverse events in the short term, i.e. the first week after completion of the primary course, and have come to non-univocal conclusions when comparing heterologous and homologous vaccination regimens. Hence, although they have good methodological quality and even larger sample sizes, despite the small sample size, the present study added value as it explored the safety of the heterologous and homologous vaccine schedules over a longer period of time and considered outcomes (impact on daily activities and ability to work) meaningful for the patient and potentially capable of modifying his/her attitude toward the vaccine offer. Only Warkentin et al.¹⁶ have carried out short (14–19 days) and medium-term (40–56 days) questionnaires, potentially able to provide insight into the impact of vaccination on quality of life. In the mid-term, comparable proportions of recipients in all groups reported requiring medical care (ChAdOx1-S/ChAdOx1-S vs ChAdOx1-S/mRNA vs mRNA/mRNA, respectively, 15% vs 18% vs 16%) in line with our results, from which there were no differences in self-reported reactions and health status 1 month and 14 weeks after vaccination between the regimens under study.

Although composed of a clinically and demographically homogeneous population, the main limitation of this study is the small sample size used for the analysis. Indeed, our study was conducted during the first month since the heterologous vaccine regimen had been authorized and individuals who had received ChAdOx1-S/BNT162b2244 vaccination were only 244. Moreover, it must be emphasized that the small sample size may have limited the ability to detect rare severe AEs. Indeed, due to the small sample size, we chose to group moderate with severe AEs for the statistical analysis. Furthermore, our findings specifically refer to the primary heterologous immunization scheme with adenoviral vector and mRNA vaccine. Further properly designed studies are warranted to extend our conclusions to all heterologous regimens. Although not discussed in this paper, other available heterologous schemes do not seem to be associated with any particular safety concerns.^25,26 In addition, in our study we do not discuss the safety of heterologous booster doses, which require separate detailed evaluations. Lastly, it must be considered that our study design did not include a placebo group and was not blinded. Consequently, it cannot be totally excluded that AEs reporting was influenced by the expectation of adverse reactions in some participants, especially in those receiving heterologous vaccination.

Overall, our results confirm the safety of heterologous vaccination with ChAdOx1-S followed by BNT162b2, supporting its use, especially when greater flexibility or acceleration in vaccination campaigns was required.

Conclusion

The adverse events observed in our study were mainly local and of mild intensity, and slightly more frequent with the heterologous vaccination, confirming the safety of both the heterologous ChAdOx1-S/BNT162b2 and homologous vaccine regimens for the completion of the primary schedule. Administering a second dose of mRNA vaccine to recipients of a previous dose of viral vector vaccine has been an advantageous strategy in those circumstances requiring greater flexibility or vaccination campaign acceleration.

Funding Statement

The author(s) reported that there is no funding associated with the work featured in this article.

Authors’ contributions

Conceptualization, FF, DM, and RP; methodology FF and DM; validation AL; formal analysis GI; investigation AL, GI, DL; data curation GI; writing-original draft preparation FF, DM; writing-review and editing, FF, DM, RP, PL; supervision PL and DM. All authors read and approved the final manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

All relevant data are within the manuscript.