Safety and immunogenicity report from the Com-COV study – A single-blind randomised non-

2

inferiority trial comparing heterologous and homologous prime-boost schedules with an

3

adenoviral vectored and mRNA COVID-19 vaccine.

4

Xinxue Liu*1, PhD; Robert H Shaw*1,2, MRCP; Arabella SV Stuart*1,2, MSc; Melanie Greenland1, MSc;

5

Tanya Dinesh1, MSci; Samuel Provstgaard-Morys1, BSc; Elizabeth A. Clutterbuck, PhD1; Maheshi N

6

Ramasamy1,2, DPhil; Parvinder K Aley1, PhD; Yama F Mujadidi1, MSc; Fei Long1, MSc; Emma L

7

Plested1, Hannah Robinson1, RN; Nisha Singh1, DPhil; Laura L Walker1; Rachel White1, RN; Nick J.

8

Andrews3, PhD; J. Claire Cameron4, FFPH; Andrea M Collins5, PhD; Daniella M Ferreira5, PhD; Helen

9

Hill5, PhD; Christopher A Green6, DPhil; Bassam Hallis3, PhD; Paul T Heath7, FRCPCH; Saul N Faust8,

pe
er
re
v

iew
ed

1

10

PhD; Adam Finn9,PhD; Teresa Lambe10, PhD; Rajeka Lazarus11, DPhil; Vincenzo Libri12, MD; Mary

11

Ramsay3, PhD; Robert C Read8 PhD; David PJ Turner13, PhD; Paul J Turner, PhD14; Jonathan S Nguyen-

12

Van-Tam15, DM; Matthew D Snape1,16^, MD; and the Com-COV Study Group†.

13
14

1.


Oxford Vaccine Group, Department of Paediatrics, University of Oxford, Oxford OX3 9DU, UK

15

2.

Oxford University Hospitals NHS Foundation Trust, Oxford, UK

16

3.

Public Health England

17

4.

Public Health Scotland

18

5.

Liverpool School of Tropical Medicine, University of Liverpool, Pembroke Place, Liverpool, L3

6.

21
22


Foundation Trust, Birmingham B15 2TH, UK
7.

23
24

NIHR/Wellcome Trust Clinical Research Facility, University Hospitals Birmingham NHS

tn

20

5QA, UK

ot

19

The Vaccine Institute, St. George's University of London, Cranmer Terrace, London SW17 0RE,
UK

8.

NIHR Southampton Clinical Research Facility and Biomedical Research Centre, University
Hospital Southampton NHS Foundation Trust, Southampton, SO16 6YD, UK; Faculty of Medicine

26

and Institute for Life Sciences, University of Southampton, Southampton, SO17 1BJ, UK


28
29
30

Schools of Population Health Sciences and Cellular and Molecular Medicine, University of
Bristol, Bristol, UK

10. Jenner Institute, University of Oxford, Old Road Campus Research Building, Roosevelt Drive,
Headington, Oxford OX3 7DQ, UK

11. North Bristol NHS Trust, Southmead Road, Bristol BS10 5NB, UK

Pr

31

9.

ep

27

rin

25

32

33


12. NIHR UCLH Clinical Research Facility and NIHR UCLH Biomedical Research Centre, University
College London Hospitals NHS Foundation Trust, London W1T 7HA, UK

This preprint research paper has not been peer reviewed. Electronic copy available at: />

35
36

iew
ed

34
13. University of Nottingham, Nottingham, NG7 2RD, UK; Nottingham University Hospitals NHS
Trust, Nottingham, NG7 2UH, UK

37

14. National Heart & Lung Institute, Imperial College London, Dovehouse St, London SW3 6LY, UK

38

15. Division of Epidemiology and Public Health, University of Nottingham School of Medicine,

39
40
41

Nottingham, NG7 2UH, UK


16. Oxford NIHR – Biomedical Research Centre, Oxford University Hospitals NHS Foundation Trust,
Oxford, OX3 9DU, UK
*Contributed equally

43

^ Corresponding author - Matthew D Snape, Oxford Vaccine Group, Department of Paediatrics,

44

University of Oxford, Oxford OX3 9DU, UK, , Phone 01865

45

611400

46

†Com-COV Study Group authorship - appendix

Pr

ep

rin

tn

ot


pe
er
re
v

42

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Abstract

48

Background

49

Use of heterologous prime-boost COVID-19 vaccine schedules could facilitate mass COVID-19

50

immunisation, however we have previously reported that heterologous schedules incorporating an

51

adenoviral-vectored vaccine (ChAd, Vaxzevria, Astrazeneca) and an mRNA vaccine (BNT, Comirnaty,

52

Pfizer) at a 4-week interval are more reactogenic than homologous schedules. Here we report the


53

immunogenicity of these schedules.

54

Methods

55

Com-COV (ISRCTN: 69254139, EudraCT: 2020-005085-33) is a participant-blind, non-inferiority trial

56

evaluating vaccine reactogenicity and immunogenicity. Adults ≥ 50 years, including those with well-

57

controlled comorbidities, were randomised across eight groups to receive ChAd/ChAd, ChAd/BNT,

58

BNT/BNT or BNT/ChAd, administered at 28- or 84-day intervals.

59

The primary endpoint is geometric mean ratio (GMR) of serum SARS-CoV-2 anti-spike IgG levels (ELISA)

60


at one-month post boost between heterologous and homologous schedules given the same prime

61

vaccine. We tested non-inferiority of GMR using a margin of 0.63. The primary analysis was on a per-

62

protocol population, who were seronegative at baseline. Safety analyses were performed amongst

63

participants receiving at least one dose of study vaccines.

64

Findings

65

In February 2021, 830 participants were enrolled and randomised, including 463 with a 28-day prime-

66

boost interval whose results are reported in this paper. Participant mean age was 57.8 years, 45.8%

67

were female, and 25.3% from ethnic minorities.


68

The geometric mean concentration (GMC) of day 28 post-boost SARS-CoV-2 anti-spike IgG in

69

ChAd/BNT recipients (12,906 ELU/ml) was non-inferior to that in ChAd/ChAd recipients (1,392 ELU/ml)

70

with a geometric mean ratio (GMR) of 9.2 (one-sided 97.5% CI: 7.5, ). In participants primed with

71

BNT, we failed to show non-inferiority of the heterologous schedule (BNT/ChAd, GMC 7,133 ELU/ml)

72

against the homologous schedule (BNT/BNT, GMC 14,080 ELU/ml) with a GMR of 0.51 (one-sided

74

pe
er
re
v

ot


tn

rin

97.5% CI: 0.43, ). Geometric mean of T cell response at 28 days post boost in the ChAd/BNT group
was 185 SFC/106 PBMCs (spot forming cells/106 peripheral blood mononuclear cells) compared to 50,
80 and 99 SFC/106 PBMCs for ChAd/ChAd, BNT/BNT, and BNT/ChAd, respectively. There were four

Pr

75

ep

73

iew
ed

47

76

serious adverse events across all groups, none of which were considered related to immunisation.

77

Interpretation

This preprint research paper has not been peer reviewed. Electronic copy available at: />


Despite the BNT/ChAd regimen not meeting non-inferiority criteria, the GMCs of both heterologous

79

schedules were higher than that of a licensed vaccine schedule (ChAd/ChAd) with proven efficacy

80

against COVID-19 disease and hospitalisation. These data support flexibility in the use of heterologous

81

prime-boost vaccination using ChAd and BNT COVID-19 vaccines.

82

Funding

83

Funded by the UK Vaccine Task Force (VTF) and National Institute for Health Research (NIHR)

Pr

ep

rin

tn


ot

pe
er
re
v

iew
ed

78

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Introduction

85

COVID-19 has severely impacted the world in terms of health, society and economy.(1) Immunity

86

through vaccination is fundamental to reducing the burden of disease, the emergence from current

87

public health measures and the subsequent economic recovery. Multiple vaccines with proven

88


effectiveness are being deployed globally, including the mRNA vaccine Comirnaty (BNT, Pfizer) and

89

the adenoviral vectored vaccine Vaxzevria (ChAd, AstraZeneca), both of which are approved as two-

90

dose homologous schedules in the UK and elsewhere.(2)

91

As of June 2021, around 2 billion COVID-19 vaccines were administered worldwide,(3) but many more

92

people remain unimmunised. Heterologous vaccine schedules may ease logistical problems inherent

93

in some national and international vaccine programmes. This could prove of particular importance in

94

low- and middle-income countries(4) as well as in countries which have adopted age-specific

95

restrictions for the use of ChAd.(5–7)


96

While the Sputnik V vaccine programme, which deploys a heterologous prime-boost schedule using

97

Ad26 and Ad5 vectored COVID-19 vaccines, induces a robust humoral and cellular response and has

98

shown 91.6% efficacy against symptomatic disease,(8,9) there are currently no efficacy data using

99

heterologous schedules incorporating COVID-19 vaccines across different platforms. Nevertheless,

100

pre-clinical studies support evaluation of this approach,(10,11) and a randomised study in Spain

101

suggested that there is an increase in binding and neutralising antibody after boosting ChAd primed

102

participants with BNT, compared with not having a boost dose.(12) Additionally, early results from an

103


observational study in Germany show that humoral responses are similar in the cohort receiving

104

BNT/BNT at a 3-week interval to those receiving ChAd/BNT at 10-week interval, with cellular responses

105

appearing to be higher in the ChAd/BNT cohort.(13)

106

Robust data on the safety and immunogenicity of heterologous vaccine schedules will help inform the

107

use of these schedules in individuals who develop a contraindication to a specific vaccine after their

108

first dose, and for vaccine programmes looking to mitigate vaccine supply chain disruption or changes

109

in guidance for vaccine usage. In addition, there remains the possibility that mixed schedules may

110

induce an enhanced or more durable humoral and/or cellular immune response compared to licensed


111

schedules, and may do so against a greater range of SARS-CoV-2 variants.

pe
er
re
v

ot

tn

rin

Accordingly, we have undertaken a randomised controlled trial to determine whether the immune
responses to heterologous schedules deploying ChAd and BNT are non-inferior to their equivalent

Pr

113

ep

112

iew
ed


84

114

homologous schedules.

115

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Methods

iew
ed

116

Trial Design

117

Com-COV is a participant-blinded, randomised, phase II, UK multi-centre, non-inferiority study

119

investigating the safety, reactogenicity and immunogenicity of heterologous prime-boost COVID-19

120

vaccine schedules (See supplementary or for full protocol). Four


121

permutations of prime-boost schedules using the ChAd and BNT vaccines are compared, at two

122

different prime-boost intervals (28 and 84 days) to reflect both ‘short’ and ‘long’ interval approaches

123

to immunisation. The majority of participants were enrolled into the ‘General cohort’ in which

124

participants could be randomised to receive the four vaccine schedules at either a 28 or 84 day

125

interval, while a subset (N=100, selected on the basis of site capacity and participant availability) were

126

enrolled into an immunology cohort that only randomised individuals to vaccine schedules with a 28

127

day interval and had four additional blood tests.

128


Here we report data from all participants randomised to vaccine schedules with a prime/boost interval

129

of 28 days.

pe
er
re
v

118

Participants

130

COVID-19 vaccine-naïve adults aged 50 years and over, with no or well-controlled mild-moderate

132

comorbidities were eligible for recruitment. Key exclusion criteria were previous laboratory confirmed

133

SARS-CoV-2 infection, history of anaphylaxis, history of allergy to a vaccine ingredient, pregnancy,

134


breastfeeding or intent to conceive, and current use of anticoagulants. Full details of the inclusion and

135

exclusion criteria can be found in the protocol (supplementary file).

tn

ot

131

Interventions and Procedures

136

Participants who met the inclusion and exclusion criteria via the online screening and/or the

138

telephone screening were invited to the baseline visits (D0), where randomisation occurred for those

139

passing the final eligibility assessment and providing informed consent.

140

Two COVID-19 vaccines were used in this study. ChAd is a replication-deficient chimpanzee adenovirus


141

vectored vaccine, expressing the SARS-CoV-2 spike surface glycoprotein with a leading tissue

143

plasminogen activator signal sequence. Administration is via 0.5ml intramuscular (IM) injection into
the upper arm. BNT is a lipid nanoparticle-formulated, nucleoside-modified mRNA vaccine encoding
trimerised SARS-CoV-2 spike glycoprotein. Administration is via a 0.3ml IM injection into the upper

Pr

144

ep

142

rin

137

145

arm.

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Vaccines were administered by appropriately trained trial staff at trial sites. Participants were


147

observed for at least 15 minutes after vaccination. During the D0 visit, participants were given an oral

148

thermometer, tape measure and diary card (electronic or paper) to record solicited, unsolicited, and

149

medically attended adverse events (AEs) with instructions. The study sites’ physicians reviewed the

150

diary card regularly to record AEs, adverse events of special interest (AESIs), and serious adverse

151

events (SAEs). The time-points for subsequent visits for immunogenicity blood sampling are shown in

152

the supplementary protocol. During the study visits, AEs, AESIs and SAEs that had not been recorded

153

in the diary card were also collected.

154


Participants testing positive for SARS-CoV-2 in the community were invited for an additional visit for

155

clinical assessment, collection of blood samples and throat swab, and completion of a COVID-19

156

symptom diary.

Randomisation and Blinding

pe
er
re
v

157

iew
ed

146

Computer-generated randomisation lists were prepared by the study statistician. Participants were

159

block randomised (block size four) 1:1:1:1 within the immunology cohort to ChAd/ChAd, ChAd/BNT,


160

BNT/BNT and BNT/ChAd schedules (boost interval of 28 days). General Cohort participants were block

161

randomised (block size eight) 1:1:1:1:1:1:1:1 to ChAd/ChAd, ChAd/BNT, BNT/BNT and BNT/ChAd

162

schedules at boosting intervals of both 28 and 84 days. Randomisation was stratified by study site.

163

Clinical research nurses who were not involved in safety endpoint evaluation performed the

164

randomisation using REDCapTM (the electronic data capture system) and prepared and administered

165

vaccine.

166

Participants and laboratory staff processing the immunogenicity endpoints were blinded to vaccines

167


received, but not to prime-boost interval. Participant blinding to vaccines was maintained by

168

concealing randomisation pages, preparing vaccines out of sight and applying masking tape to vaccine

169

syringes to conceal dose volume and appearance. The clinical team assessing the safety endpoints

170

were not blinded.

171

Outcomes

173
174

tn

rin

those with a prime-boost interval of 28 days in participants who were seronegative for COVID infection
at baseline.

Secondary outcomes include safety and reactogenicity as measured by solicited local and systemic


Pr

175

The primary outcome is serum SARS-CoV-2 anti-spike IgG concentration at 28 days post boost for

ep

172

ot

158

176

events for 7 days after immunisation (reported previously for the 28-day prime-boost interval

177

groups),(14) unsolicited AEs for 28 days after immunisation and medically attended AEs for 3 months

This preprint research paper has not been peer reviewed. Electronic copy available at: />

after immunisation. Blood biochemistry and haematology assessments were measured at baseline

179

(day 0), on day of boost and 28 days post-boost, with an additional day 7 post-boost time-point (D35)


180

for the immunology cohort only. AESIs (listed in protocol as a supplementary file) and SAEs were

181

collected throughout the study.

182

Immunological secondary outcomes include SARS-CoV-2 anti-spike binding IgG concentration, cellular

183

responses (measured by IFN-gamma ELISpot) in peripheral blood, and pseudotype virus neutralisation

184

titres at D0, D28 and D56. The immunology cohort had additional visits at D7, D14, D35 and D42 to

185

explore the kinetics of the immune responses further.

186

Laboratory methods

187


Sera were analysed at Nexelis, (Laval, Canada) to determine SARS-CoV-2 anti-spike IgG concentrations

188

by ELISA (reported as ELISA Laboratory Unit (ELU)/ml) and the 50% Neutralising Antibody Titre (NT50)

189

for SARS-CoV-2 pseudotype virus neutralisation assay (PNA), using a vesicular stomatitis virus

190

backbone adapted to bear the 2019-nCOV SARS-CoV-2 spike protein(15). Sera from day 0 were

191

analysed at Porton Down, Public Health England, by ECLIA (Cobas platform, Roche Diagnostics) to

192

determine anti-SARS-CoV-2 nucleocapsid IgG status (reported as negative if below a cut off index of

193

1.0). NT50 for live SARS-CoV-2 virus (Victoria/01/2020) was determined by microneutralisation assay

194

(MNA) also at Porton Down, on day 0 and 56 samples in the AZ-primed groups only.(15) Interferon-


195

gamma secreting T-cells specific to whole spike protein epitopes designed based on the Wuhan-Hu-1

196

sequence (YP_009724390.1) were detected using a modified T-SPOT-Discovery test performed at

197

Oxford Immunotec (Abingdon, UK) within 32 hours of venepuncture, using the addition of T-Cell Xtend

198

reagent to extend PBMC survival.(16) T cell frequencies were reported as spot forming cells (SFC) per

199

250,000 PBMCs with a lower limit of detection of one in 250,000 PBMCs, and these results multiplied

200

by four to express frequencies per 106 PBMCs.

pe
er
re
v

ot


tn

Statistical analysis

rin

201

iew
ed

178

The primary analysis of SARS-CoV-2 anti-spike IgG was carried out in participants boosted at D28 on a

203

per-protocol basis. The analysis population was participants who were seronegative for COVID at

204

baseline (defined by anti-nucleocapsid IgG negativity at Day 0 and no confirmed SARS-CoV-2 infection

205
206

within 14 days post prime vaccination), whose primary endpoint datawere available and who had no
protocol deviations. The geometric mean ratio (GMR) was calculated as the antilogarithm of the
difference between the mean of the log10 transformed SARS-CoV-2 anti-spike IgG in the heterologous


Pr

207

ep

202

208

arm and that in the homologous arm (as the reference), after adjusting for study site and cohort

209

(immunology/general) as randomisation design variables in the linear regression model. The GMRs

210

were reported separately for participants primed with ChAd and those with BNT with a one-sided

This preprint research paper has not been peer reviewed. Electronic copy available at: />

97.5% confidence interval. The criteria for non-inferiority of heterologous boost compared to the

212

homologous boost was for the lower limit of the one-sided 97.5% CI of the GMR to lie above 0.63; this

213


was chosen on a pragmatic basis to approach the WHO criterion of 0.67 for licencing new vaccines

214

when using GMR as the primary endpoint, while still allowing rapid study delivery.(17)

215

According to recommended practice for non-inferiority trials,(18) we also present the two-sided 95%

216

CI of the adjusted GMRs among the modified intent-to-treat (mITT) population by including

217

participants with protocol deviations as secondary analyses. The heterologous arm was considered

218

superior to the homologous arm if the lower limit of the two-sided 95% CI lay above one, and the

219

homologous boost arm superior to the heterologous boost arm if the upper limit of the two-sided 95%

220

CI lay below one. As an exploratory analysis, subgroup analyses were conducted stratified by age (50-


221

59, and 60+), sex (male and female) and baseline comorbidity (presence/absence of cardiovascular

222

disease, respiratory disease or diabetes).

223

The geometric means of secondary immunological outcomes were reported in the mITT population.

224

The proportions of participants with responses higher than the lower limit of detection (LLOD) or

225

lower limit of quantification (LLOQ) were calculated by vaccine schedule, with 95% CIs calculated by

226

the binomial exact method for each secondary immunological outcome, and compared between

227

heterologous and homologous arms using Fisher's exact test. Censored data reported as below the

228


LLOD/LLOQ were imputed with a value equal to half of the threshold before transformation. Between-

229

schedule comparisons of immunological outcomes were evaluated by linear regression models

230

adjusting for study site and cohort as secondary analyses. Correlations between different

231

immunological outcomes were evaluated by Pearson correlation coefficients.

232

Participants who received at least one dose of study vaccines were included in the safety analysis. The

233

proportion of participants with at least one safety event was reported by vaccine schedule. Fisher's

234

exact test was used to compare the difference between schedules.

235

The sample size calculation was done assuming the standard deviation (SD) of the primary endpoint


236

to be 0.4 (log10) and the true GMR to be one. The study needed to recruit 115 participants per arm to

237

achieve 90% power at a one-sided 2.5% significance level, after adjusting for an attrition rate of 25%

239

pe
er
re
v

ot

tn

rin

due to baseline SARS-CoV-2 seropositivity or loss to follow-up.
All the statistical analyses were carried out using R version 3.6.2 (2019-12-12).

Trial oversight and safety monitoring

Pr

240


ep

238

iew
ed

211

241

The trial was reviewed and approved by the South-Central Berkshire Research Ethics Committee

242

(21/SC/0022), the University of Oxford, and the Medicines and Healthcare Products Regulatory Agency

This preprint research paper has not been peer reviewed. Electronic copy available at: />

(MHRA). An independent data safety monitoring board (DSMB) reviewed safety data, and local trial-

244

site physicians provided oversight of all adverse events in real-time. The trial is registered at

245

www.isrctn.com as ISRCTN: 69254139.


246

Results

247

Between 11th February 2021 and 26th February 2021, 978 participants were screened at eight study

248

sites across England, among whom 830 were enrolled and randomised into the study. 463 participants

249

were randomised to the four arms with a 28-day prime-boost interval reported here including 100

250

participants enrolled into the immunology cohort. The mean age of the participants was 57.8 years

251

(SD 4.7) with 45.8% female participants and 25.3% from ethnic minorities. Baseline characteristics

252

were well balanced across the four arms in both the general and immunology cohorts (Table 1). At

253


baseline, 20 (4.3%) participants were positive for anti-nucleocapsid IgG (cut-off index ≥1.0), evenly

254

distributed across groups. The numbers of participants included in the modified intent-to-treat and

255

per-protocol analyses were 432 and 426, respectively (Figure 1).

pe
er
re
v

iew
ed

243

256

Immune responses at 28 days post boost vaccination: Primary outcome and key secondary

257

outcomes.

Among participants primed with ChAd, the GMCs of SARS-CoV-2 anti-spike IgG at 28 days post boost


259

vaccination was 1,392 ELU/ml (95%CI: 1,188-1,630) and 12,906 ELU/ml (95%CI: 11,404-14,604) in

260

the homologous arm (ChAd/ChAd) and heterologous arm (ChAd/BNT), respectively, with a GMR of 9.2

261

(one-sided 97.5% CI: 7.5, ) between heterologous and homologous arms in the per-protocol analysis

262

(Table 2). Similar GMCs were observed in the modified ITT analysis with a GMR of 9.3 (two-sided 95%

263

CI: 7.7-11). The GMR of PNA NT50 (secondary outcome) between heterologous and homologous arms

264

was 8.5 (two-sided 95% CI: 6.5, 11) in the modified ITT analysis. These results indicate that the

265

ChAd/BNT schedule was not only non-inferior, but statistically superior to ChAd/ChAd schedule for

266


both the SARS-CoV-2 anti-spike IgG and PNA NT50. The secondary outcome of cellular responses by

267

T-cell ELISpot revealed 50 SFC/106 PBMCs (39-63) for ChAd/ChAd and 185 SFC/106 PBMCs (152-224)

268

with a GMR of 3.8 (2.8-5.1) (Table 2).

269

In the two schedules with BNT as the prime vaccine, the GMCs of SARS-CoV-2 anti-spike IgG at 28 days

271

tn

rin

post boost vaccination were 14,080 ELU/ml (95%CI: 12,491-15,871) and 7,133 ELU/ml (95%CI: 6,4157,932) for the homologous and heterologous arms in the per-protocol analysis. The GMR in the perprotocol analysis was 0.51 (one-sided 97.5% CI: 0.43, ). The study therefore failed to show non-

Pr

272

ep

270


ot

258

273

inferiority of the heterologous arm (BNT/ChAd) to its corresponding homologous arm (BNT/BNT). In

274

addition, BNT/ChAd was statistically inferior for both SARS-CoV-2 anti-spike IgG (p<0.0001) and PNA

275

NT50 (p=0.0041), compared with BNT/BNT. The geometric mean SFC frequency (T-cell ELISpot) was

This preprint research paper has not been peer reviewed. Electronic copy available at: />

higher in the heterologous arm compared with the homologous arm (99 vs 80 SFC/106 PBMCs), though

277

did not reach a level of statistical significance (GMR: 1.2, two-sided 95% CI: 0.88-1.7).

278

Similar patterns of GMRs were seen in all subgroup analyses with SARS-CoV-2 anti-spike IgG and PNA

279


NT50 consistently higher in the ChAd/BNT compared with ChAd/ChAd and BNT/BNT higher than

280

BNT/ChAd (Figure 2). Strong correlations were seen between SARS-CoV-2 anti-spike IgG and PNA NT50

281

at 28 days post boost in all vaccine schedules (Pearson correlation coefficients of 0.6-0.7), while the

282

correlations between humoral responses and cellular response were weak (Pearson correlation

283

coefficients <0.4) (Figure 3).

284

Additional secondary outcomes

pe
er
re
v

iew
ed


276

285

Immunology cohort: Humoral & cellular immune responses at 7 and 14 days post boost

286

vaccination

Across all four schedules an increase in SARS-CoV-2 anti-spike IgG was seen from day 28 to day 35 (day

288

7 post boost), contrasting with a lack of response at day 7 post prime, suggesting that both vaccines

289

induced immunological priming that was augmented by either homologous or heterologous boost

290

(Figure 4 and Appendix Figure 1). No further increase in SARS-CoV-2 anti-spike IgG was seen at day 28

291

post boost, suggesting the peak response post-boost is likely to be earlier than 28 days. For all

292


schedules except ChAd/ChAd, peak T cell response was observed at 14 days post boost; no further

293

increase was seen in ChAd/ChAd post boost. (Appendix Figure 1).

294

ot

287

Humoral & cellular immune responses: Post-prime vaccination
In participants primed with ChAd and BNT, the SARS-CoV-2 anti-spike IgG GMCs were 129 (95% CI: 80-

296

206) and 843 (95% CI: 629-1130) ELU/ml at 14 days post prime (p<0.0001), and 555 (95% CI: 458-673)

297

and 1,597 (1,399-1,822) ELU/ml at 28 days post prime (p<0.0001), respectively .

298

In contrast, ChAd induced significantly higher cellular responses at 14 days (p<0.0001) and 28 days

299

(p<0.0001) post prime vaccination compared with BNT: Geometric mean at 14 days was 160 (95% CI:


300

100-256) vs 35 SFC/106 PBMCs (95%CI: 26-47), and at 28 days was 54 (95% CI: 44-65) vs 16 SFC/106

301

PBMCs (95%CI: 14-18), respectively.

303

rin

Humoral & cellular immune responses: Cross-schedule comparisons

When BNT was given as the boost vaccine, similar levels of SARS-CoV-2 anti-spike IgG (p=0.43) and
PNA NT50 (p=0.40) at 28 days post-boost were observed among participants primed with ChAd

Pr

304

ep

302

tn

295


305

(ChAd/BNT) and BNT (BNT/BNT). Participants boosted with ChAd following BNT prime (BNT/ChAd)

306

had significantly higher SARS-CoV-2 anti-spike IgG (p<0.0001) and PNA NT50 (p<0.0001) than those

307

primed with ChAd (ChAd/ChAd). Homologous BNT/BNT immunisation generated higher binding

This preprint research paper has not been peer reviewed. Electronic copy available at: />

antibodies at day 7 (p<0.0001) and day 28 (p<0.0001) post boost compared with ChAd/ChAd, with a

309

difference also observed in PNA at day 28 post boost (p<0.0001).

310

In contrast to the lack of further response following a homologous second dose of ChAd (Figure 4,

311

Appendix Figure 1), a significant increase in cellular response was seen after a homologous boost with

312


BNT, such that those receiving BNT/BNT had significantly higher number of SARS-CoV-2 specific T cells

313

per 106 PBMCs than ChAd/ChAd (p=0.0045) at 28 days post boost with a four week interval (Figure 4).

314

Safety

iew
ed

308

The results of the solicited adverse events in the week following immunisation have been reported

316

previously.(14) In summary, we observed an increase in systemic reactogenicity after boost in

317

participants receiving heterologous schedules in comparison to homologous schedules with the same

318

prime vaccine. In participants randomised to 28-day interval groups there were 316 adverse events

319


from 178 participants up to 28 days following boost immunisation (Supplementary Table 1). No

320

significant difference was observed between the vaccine schedules (p=0.89). Adverse events of Grade

321

≥3 are described in Supplementary Table 2.

322

Amongst all participants up to 6th Jun 2021 (date of data-lock) there were seven AESIs, of which four

323

were COVID-19 diagnoses (Supplementary Tables 3 & 4). The non-COVID-19 AESIs were not

324

considered related to immunisation. Four participants across all groups developed COVID-19. Three

325

were within 7 days of prime immunisation, one was 54 days later, and had not received their planned

326

28 day boost due to travel. (Supplementary Table 4)


327

There were four SAEs across all groups in the study up to the data lock, and none was considered

328

related to immunisation (Supplementary table 5).

tn

ot

pe
er
re
v

315

Pr

ep

rin

329

This preprint research paper has not been peer reviewed. Electronic copy available at: />


Discussion

331

We present here, for the first time in a randomised controlled clinical trial, the immunogenicity of

332

heterologous and homologous ChAd and BNT vaccine schedules with a 28-day prime-boost interval.

333

The findings demonstrate that all the schedules studied induced concentrations of SARS-CoV-2 anti-

334

spike IgG concentrations at least as high as those induced after a licensed ChAd/ChAd schedule, which

335

is effective in preventing symptomatic COVID-19 when administered at a 4-12 week prime-boost

336

interval.(19) Nevertheless, it is notable that the BNT containing schedules were more immunogenic

337

than the homologous ChAd/ChAd schedule, and none of the heterologous schedules generated


338

binding or pseudotype virus neutralising antibodies above those induced by BNT/BNT immunisation.

339

Cellular immune responses in the BNT vaccine containing schedules were likewise all at least as high

340

as ChAd/ChAd group with BNT/ChAd showing the greatest expansion of vaccine-antigen responsive T-

341

cells in the peripheral circulation at 28 days post boost.

342

Although the 28-day homologous ChAd/ChAd was the least immunogenic of the four schedules in our

343

trial, data from a phase 3 randomised clinical trial showed this regimen to be 76% efficacious against

344

symptomatic disease, and 100% against severe disease.(20) Additionally, when deployed in an 8 to 12

345


week schedule, ChAd/ChAd has been shown to be 86% and 92% effective against hospitalisation due

346

to the Alpha (B.1.1.7) and Delta (B.1.617.2) variants, respectively.(21–24) Given the established

347

associations between humoral responses and vaccine efficacy,(19) our findings indicate the two

348

heterologous schedules in this trial are also likely to be highly effective, and could be considered, in

349

some circumstances, for national vaccine programmes.

350

To the best of our knowledge, the current study is the first randomised controlled trial to report

351

immunogenicity of the BNT/ChAd heterologous schedule. Our results for the ChAd/BNT schedule build

352

on preliminary data from a Spanish randomised trial in which 18-60 year olds received a dose of BNT


353

two to three months after priming with ChAd and demonstrated a 37-fold increase in SARS-CoV-2 anti-

354

spike IgG at 14 days post-boost, higher than the 22-fold and 19-fold rises at 7 days and 28 days post

355

boost in this study.(12) Potential explanations for these differences include the longer prime-boost

356

interval, the different sampling time-points and a younger population in the Spanish study.(12) Fold

357

rises in the cellular response were, however, similar (4-fold vs. 3.5-fold). Early results from a

359

pe
er
re
v

ot

tn


rin

prospective cohort study in Germany, which compared healthcare workers immunised with BNT/BNT
at a 3-week interval or ChAd/BNT at an 8-12 week interval, showed similar concentrations of binding
antibody at 3 weeks post-boost and higher cellular responses in the ChAd/BNT recipients.(25) Another

Pr

360

ep

358

iew
ed

330

361

German cohort study of 26 participants aged 25-46 years receiving a ChAd/BNT schedule with an 8-

362

week prime-boost interval also reported a robust humoral immune response, with a suggestion of

This preprint research paper has not been peer reviewed. Electronic copy available at: />


better retention of neutralising activity against Beta and Delta variants than that observed in a non-

364

randomised cohort receiving BNT/BNT.(26)

365

Together with the evidence that the T cell ELISpot readouts are similar between schedules, the

366

immunological data presented here provide reassurance that ChAd/BNT and BNT/ChAd are

367

acceptable options. However, in contrast with recent non-randomised and non-blinded studies, we

368

did observe increased reactogenicity in the 28-day ChAd/BNT schedule (14), compared with

369

ChAd/ChAd. This discrepancy may be due to the variation in the prime-boost interval, and the

370

forthcoming data from the 84-day prime-boost interval participants in this trial will help to delineate


371

this difference. Although these mild-moderate symptoms were transient, this does need to be taken

372

into consideration when deploying this schedule, especially in those younger than the participants

373

enrolled in this study, given the reported trend towards increased reactogenicity with decreasing

374

age.(27,28) Additional considerations for deployment of mixed schedules include potential logistical

375

challenges within the healthcare infrastructure as well as the complex public communications

376

surrounding this.

377

Numerous other randomised heterologous prime/boost COVID-19 vaccine studies are now underway

378


or planned,(29) including Com-COV2, which incorporates vaccines manufactured by Moderna and

379

Novavax.(30) Crucially, several of these studies include vaccines manufactured by CanSinoBIO,

380

Gamaleya Research Institute and Sinovac that are extensively used in low- and middle-income

381

countries, which are potentially more likely to rely on mixed schedules. These data on heterologous

382

vaccination will also inform ‘3rd dose’ booster immunisation programmes, currently being considered

383

in preparation for the Northern Hemisphere 2021/2022 winter(31) and being studied in the ongoing

384

‘Cov-Boost’ study.(32)

385

There are a number of limitations of this study. Firstly, as an immunogenicity and reactogenicity study


386

the sample size is not adequate to assess vaccine schedule efficacy. Although there is evidence that

387

both binding and neutralising antibodies correlate well with protection against symptomatic

388

disease,(19,33,34) it is less clear to what extent variations in these measures above a certain,

389

unknown, threshold impact on protection against severe disease. Similarly, we are unable, at this

390

point, to determine whether higher antibody concentrations measured at 28 days post boost

392

pe
er
re
v

ot

tn


rin

immunisation will result in a more sustained elevation of vaccine-induced antibodies, and this will be
evaluated at ongoing study visits up to one-year post enrolment. An additional limitation is the
generalisability of these results to a younger population given the age (50 – 70 years old) of

Pr

393

ep

391

iew
ed

363

394

participants in this trial. Previous RCTs on homologous schedules of viral vector and mRNA vaccines

395

reported similar post boost immunogenicity between younger (18-55 years) and older (>55 years)

This preprint research paper has not been peer reviewed. Electronic copy available at: />


adults,(27,35,36) and higher reactogenicity in younger cohorts, (27,35,36) and there is no reason to

397

expect this would be different for the heterologous schedules but this has not been extensively

398

demonstrated. Lastly, the data presented here were from schedules with a 28-day prime-boost

399

interval, whereas the WHO recommended interval for ChAd/ChAd is 8-12 weeks.(37) There is

400

evidence that a longer prime-boost interval results in a higher post-boost SARS-CoV-2 anti-spike IgG

401

response for ChAd/ChAd,(19) and for BNT/BNT (38) but it is unknown how lengthening the prime-

402

boost interval will affect the heterologous schedules in this study. This question will be addressed

403

when the immunogenicity data for the schedules including boosting at 84 days become available.


404

In conclusion, our study confirms the heterologous and homologous schedules of ChAd and BNT can

405

induce robust immune responses with a 4-week prime boost interval. These results argue for allowing

406

for flexibility in deploying mRNA and viral vectored vaccines, subject to supply and logistical

407

considerations, and emphasise the importance of obtaining information on other mixed schedules

408

with different prime boost intervals, especially using vaccines being deployed in low- and middle-

409

income countries.

pe
er
re
v

iew

ed

396

Pr

ep

rin

tn

ot

410

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Research in context

412

Evidence before this study

413

National regulatory authorities have granted emergency use authorizations for more than 15 vaccines,

414


among which six vaccines have been approved for emergency use by the World Health

415

Organisation.(2) Although >2 billion COVID-19 vaccines have been administered as of June 2021,(3)

416

only approximately 20% of the global population has received at least one dose of COVID-19 vaccine,

417

with less than 1% of the population in low-income countries having received a vaccine dose.(39)

418

Heterologous COVID-19 vaccine schedules have the potential to accelerate vaccine roll-out

419

worldwide, especially in low and middle income countries. We searched PubMed for research articles

420

published between database inception and 22nd June 2021 using the search terms (COVID) AND

421

(Heterologous) AND (Vaccin*) NOT (BCG) with no language restrictions. Beside our previously


422

published reactogenicity results,(14) we identified two animal studies using combinations of

423

messenger RNA, adenoviral vectored, inactivated and recombinant protein vaccines as prime boost

424

schedules. Both studies showed robust humoral and cellular responses induced by heterologous

425

schedules in mice.(10,11) In addition, there were two clinical trials on the rAd26 and rAd5 vector-

426

based heterologous prime-boost schedule (Sputnik V, Gamaleya Research Institute of Epidemiology

427

and Microbiology), showing good safety profiles, strong humoral/cellular responses and a 91.6%

428

vaccine efficacy.(8,9) A further clinical trial, which randomised participants primed with ChAd to

429


received BNT as the boost vaccine or no boost vaccination, reported robust immune response and

430

acceptable reactogenicity profile, but with no comparison to a homologous vaccine schedule.(12)

431

There were another two cohort studies evaluating ChAd prime and BNT boost schedules on medRxiv,

432

showing similar results.(13,26)

433

Added Value of this study

434

We report the results on the safety and immunogenicity of the first participant-blinded randomised

435

clinical trial using two vaccines approved by WHO for emergency use, ChAd and BNT, when

436

administered at a 28-day interval in heterologous and homologous vaccine schedules (ChAd/ChAd,


437

ChAd/BNT, BNT/BNT, BNT/ChAd). The cellular and humoral responses at 28 days post-boost of the

439

pe
er
re
v

ot

tn

rin

two heterologous vaccines schedules are no lower than the ChAd/ChAd schedule, which has shown
to be highly effective in preventing severe COVID-19 disease, and no safety concerns were raised.
Implications of all the available evidence

Pr

440

ep

438

iew

ed

411

441

In the era of multiple COVID-19 vaccines having approval for emergency use, the paramount issue in

442

solving the COVID-19 pandemic is now to optimise global vaccine coverage rate using the currently

This preprint research paper has not been peer reviewed. Electronic copy available at: />

available vaccines. The positive results from our study support flexibility in use of heterologous prime-

444

boost schedules using ChAd and BNT, which can contribute to the acceleration of vaccine roll-out.

445

Further studies are needed examining more heterologous schedules, especially those vaccines being

446

deployed in low and middle-income countries.

Pr


ep

rin

tn

ot

pe
er
re
v

447

iew
ed

443

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Author Contributions

449

MDS and JSN-V-T conceived the trial and MDS is the chief investigator. MDS, AS, RHS, and XL

450


contributed to the protocol and design of the study. AS, EP and RHS led the implementation of the

451

study. XL and MG conducted the statistical analysis and have verified the underlying data. AS, RHS,

452

MG, XL and MDS drafted the report. All other authors contributed to the implementation and data

453

collection. All authors reviewed and approved the final report.

454

Declaration of interests

455

MDS acts on behalf of the University of Oxford as an Investigator on studies funded or sponsored by

456

vaccine manufacturers including AstraZeneca, GlaxoSmithKline, Pfizer, Novavax, Janssen,

457

Medimmune, and MCM vaccines. He receives no personal financial payment for this work. JSN-V-T is


458

seconded to the Department of Health and Social Care, England. AMC and DMF are investigators on

459

studies funded by Pfizer and Unilever. They receive no personal financial payment for this work. AF is

460

a member of the Joint Committee on Vaccination and Immunisation and Chair of the WHO European

461

Technical Advisory Group of Experts (ETAGE) on Immunisation. He is an investigator and/or provides

462

consultative advice on clinical trials and studies of COVID-19 vaccines produced by AstraZeneca,

463

Janssen, Valneva, Pfizer and Sanofi and of other vaccines from these and other manufacturers

464

including GSK, VPI, Takeda and Bionet Asia. He receives no personal remuneration or benefits for any

465


of this work. SNF acts on behalf of University Hospital Southampton NHS Foundation Trust as an

466

Investigator and/or providing consultative advice on clinical trials and studies of COVID-19 and other

467

vaccines funded or sponsored by vaccine manufacturers including Janssen, Pfizer, AstraZeneca,

468

GlaxoSmithKline, Novavax, Seqirus, Sanofi, Medimmune, Merck and Valneva vaccines and

469

antimicrobials. He receives no personal financial payment for this work. PTH acts on behalf of St.

470

George’s University of London as an Investigator on clinical trials of COVID-19 vaccines funded or

471

sponsored by vaccine manufacturers including Janssen, Pfizer, AstraZeneca, Novavax and Valneva. He

472

receives no personal financial payment for this work. CAG acts on behalf of University Hospitals


473

Birmingham NHS Foundation Trust as an Investigator on clinical trials and studies of COVID-19 and

474

other vaccines funded or sponsored by vaccine manufacturers including Janssen, Pfizer, AstraZeneca,

476
477

pe
er
re
v

ot

tn

rin

Novavax, CureVac, Moderna, and Valneva vaccines, and receives no personal financial payment for
this work. VL acts on behalf of University College London Hospitals NHS Foundation Trust as an
Investigator on clinical trials of COVID-19 vaccines funded or sponsored by vaccine manufacturers
including Pfizer, AstraZeneca and Valneva. He receives no personal financial payment for this work. TL

Pr

478


ep

475

iew
ed

448

479

is named as an inventor on a patent application covering this SARS-CoV-2 vaccine and is an occasional

This preprint research paper has not been peer reviewed. Electronic copy available at: />

consultant to Vaccitech unrelated to this work. Oxford University has entered into a partnership with

481

AstraZeneca for further development of ChAdOx1 nCoV-19

482

Data sharing

483

The study protocol is provided in the appendix. Individual participant data will be made available when


484

the trial is complete, upon requests directed to the corresponding author; after approval of a proposal,

485

data can be shared through a secure online platform.

486

Acknowledgments

487

The study is funded by the UK Government through the National Institute for Health Research (NIHR)

488

and the Vaccine Task Force (VTF). This research was supported by the NIHR Oxford Biomedical

489

Research Centre and delivered through the NIHR funded National Immunisation Schedule Evaluation

490

Consortium (NISEC). MDS and SNF are NIHR Senior Investigators. The views expressed are those of the

491


author(s) and not necessarily those of the NIHR or the Department of Health and Social Care. The

492

investigators express their gratitude for the contribution of all the trial participants, the invaluable

493

advice of the international Data Safety Monitoring Board. We additionally acknowledge the broader

494

support from the various teams within the University of Oxford including the Department of

495

Paediatrics, Clinical Trials Research Governance, Research Contracts and Public Affairs Directorate.

pe
er
re
v

496

Reference

ot

497


iew
ed

480

Impact of COVID-19 on people’s livelihoods, their health and our food systems [Internet]. [cited 2021
Jun 10]. Available from: />
501
2.
502

World Health Organization W. Status of COVID-19 Vaccines within WHO EUL/PQ evaluation process.
2021.

503
3.
504

WHO Coronavirus (COVID-19) Dashboard | WHO Coronavirus (COVID-19) Dashboard With
Vaccination Data [Internet]. [cited 2021 Jun 5]. Available from: />
505
4.
506

India Situation Report [Internet]. [cited 2021 Jun 10]. Available from:
/>
rin

ep


Public Health Agency of Sweden F. Information on the use of the Astra Zeneca vaccine in the
vaccination of people 65 and older [Internet]. [cited 2021 May 20]. Available from:
/>
Pr

507
5.
508
509
510
511

tn

498
1.
499
500

This preprint research paper has not been peer reviewed. Electronic copy available at: />

French Health Authority HA de S. Covid-19 : quelle stratégie vaccinale pour les moins de 55 ans ayant
dộj reỗu une dose d’AstraZeneca ? [Internet]. [cited 2021 May 20]. Available from: />
516
7.
517
518

Danish Health Authority S. Denmark continues its vaccine rollout without the COVID-19 vaccine from

AstraZeneca [Internet]. [cited 2021 May 20]. Available from: />
519
8.
520
521
522

Logunov DY, Dolzhikova I V., Zubkova O V., Tukhvatullin AI, Shcheblyakov D V., Dzharullaeva AS, et al.
Safety and immunogenicity of an rAd26 and rAd5 vector-based heterologous prime-boost COVID-19
vaccine in two formulations: two open, non-randomised phase 1/2 studies from Russia. The Lancet.
2020 Sep 26;396(10255):887–97.

523
9.
524
525
526
527

Logunov DY, Dolzhikova I v., Shcheblyakov D v., Tukhvatulin AI, Zubkova O v., Dzharullaeva AS, et al.
Safety and efficacy of an rAd26 and rAd5 vector-based heterologous prime-boost COVID-19 vaccine:
an interim analysis of a randomised controlled phase 3 trial in Russia. The Lancet [Internet]. 2021 Feb
20 [cited 2021 Jun 20];397(10275):671–81. Available from:
/>
pe
er
re
v

iew

ed

512
6.
513
514
515

528
10. Spencer AJ, Mckay PF, Belij-Rammerstorfer S, Ulaszewska M, Bissett CD, Hu K, et al. Heterologous
529 vaccination regimens with self-amplifying RNA and adenoviral COVID vaccines induce robust immune
530 responses in mice.
531
11. He Q, Mao Q, An C, Zhang J, Gao F, Bian L, et al. Heterologous prime-boost: breaking the protective
532 immune response bottleneck of COVID-19 vaccine candidates. 2021;

ot

533
12. Borobia AM, Carcas AJ, Teresa Pérez Olmeda M, Castaño L, Bertrán J, García Pérez J, et al.
534 Reactogenicity and immunogenicity of BNT162b2 in subjects having received a first dose of 1
535 ChAdOx1S: initial results of a randomised, adaptive, phase 2 trial (CombiVacS) 2.

tn

536
13. Hillus D, Tober-Lau P, Hastor H, Helbig ET, Lippert LJ, Thibeault C, et al. Reactogenicity of homologous
537 and heterologous prime-boost immunisation with BNT162b2 and ChAdOx1-nCoV19: a prospective
538 cohort study. medRxiv. 2021 May 22;2021.05.19.21257334.
539

14. Shaw RH, Stuart A, Greenland M, Liu X, Van-Tam JSN, Snape MD, et al. Heterologous prime-boost
540 COVID-19 vaccination: initial reactogenicity data. Lancet (London, England) [Internet]. 2021 May 12
541 [cited 2021 May 21];0(0). Available from: />
rin

542
15. Bewley KR, Coombes NS, Gagnon L, McInroy L, Baker N, Shaik I, et al. Quantification of SARS-CoV-2
543 neutralizing antibody by wild-type plaque reduction neutralization, microneutralization and
544 pseudotyped virus neutralization assays. Nature Protocols. Nature Research; 2021. p. 1–33.

ep

545
16. T-Cell Xtend [Internet]. [cited 2021 Jun 16]. Available from:
546 />547
17. Guidelines on Clinical Evaluation of Vaccines: Regulatory Expectations. 2016.

Pr

548
18. D’Agostino RB, Massaro JM, Sullivan LM. Non-inferiority trials: Design concepts and issues - The
549 encounters of academic consultants in statistics. Statistics in Medicine. 2003 Jan 30;22(2):169–86.
550
19. Voysey M, Costa Clemens SA, Madhi SA, Weckx LY, Folegatti PM, Aley PK, et al. Single-dose
551 administration and the influence of the timing of the booster dose on immunogenicity and efficacy of
552 ChAdOx1 nCoV-19 (AZD1222) vaccine: a pooled analysis of four randomised trials. The Lancet

This preprint research paper has not been peer reviewed. Electronic copy available at: />

[Internet]. 2021 Mar 6 [cited 2021 May 21];397(10277):881–91. Available from:

/>
555
20.
556
557
558

AstraZeneca. COVID-19 Vaccine AstraZeneca effective against Delta (‘Indian’) variant [Internet].
AstraZeneca Press Release. 2021 [cited 2021 Jun 22]. Available from:
/>
iew
ed

553
554

559
21. Voysey M, Clemens SAC, Madhi SA, Weckx LY, Folegatti PM, Aley PK, et al. Safety and efficacy of the
560 ChAdOx1 nCoV-19 vaccine (AZD1222) against SARS-CoV-2: an interim analysis of four randomised
561 controlled trials in Brazil, South Africa, and the UK. Lancet (London, England). 2020 Dec 8;0(0).

pe
er
re
v

562
22. Lopez Bernal J, Andrews N, Gower C, Robertson C, Stowe J, Tessier E, et al. Effectiveness of the Pfizer563 BioNTech and Oxford-AstraZeneca vaccines on covid-19 related symptoms, hospital admissions, and
564 mortality in older adults in England: test negative case-control study. 2021;
565

23. Lopez Bernal J, Andrews N, Gower C, Gallagher E, Simmons R, Thelwall S, et al. Effectiveness of
566 COVID-19 vaccines against the B.1.617.2 variant.
567
24.
568
569
570
571
572
573

Stowe J, Andrews N, Gower C, Gallagher E, Utsi L, Simmons R, et al. Effectiveness of COVID-19
vaccines against hospital admission with the Delta (B.1.617.2) variant [Internet]. Public library - PHE
national - Knowledge Hub. 2021 [cited 2021 Jun 22]. Available from: />et_DLPortlet_INSTANCE_v2WsRK3ZlEig_redirect=https%253A%252F%252Fkhub.net%253A443%252F
web%252Fphe-national%252Fpublic-library%252F-%25

ot

574
25. Hillus D, Schwarz T, Tober-Lau P, Hastor H, Thibeault C, Kasper S, et al. Safety, reactogenicity, and
575 immunogenicity of homologous and heterologous prime-boost immunisation with ChAdOx1-nCoV19
576 and BNT162b2: a prospective cohort study.
Groß R, Zanoni M, Seidel A, Conzelmann C, Gilg A, Krnavek D, et al. Heterologous ChAdOx1 nCoV-19
and BNT162b2 prime-boost vaccination elicits potent neutralizing antibody responses and T cell
reactivity. medRxiv [Internet]. 2021 Jun 1 [cited 2021 Jun 5];2021.05.30.21257971. Available from:
/>
581
27.
582
583

584

Ramasamy MN, Minassian AM, Ewer KJ, Flaxman AL, Folegatti PM, Owens DR, et al. Safety and
immunogenicity of ChAdOx1 nCoV-19 vaccine administered in a prime-boost regimen in young and
old adults (COV002): a single-blind, randomised, controlled, phase 2/3 trial. The Lancet. 2020 Nov
19;396(10267):1979–93.

rin

tn

577
26.
578
579
580

585
28. Polack FP, Thomas SJ, Kitchin N, Absalon J, Gurtman A, Lockhart S, et al. Safety and Efficacy of the
586 BNT162b2 mRNA Covid-19 Vaccine. The New England journal of medicine. 2020 Dec 10;383(27).

ep

COVAX. Booster and Mix & Match COVID-19 Vaccine Strategies - Planning Ahead in an Environment
of Increasing Complexity [Internet]. [cited 2021 Jun 17]. Available from:
/>D_Bm5iywZ.pdf

Pr

587

29.
588
589
590

591
30. Study Protocol | Com-CoV [Internet]. [cited 2021 Jun 17]. Available from:
592 />
This preprint research paper has not been peer reviewed. Electronic copy available at: />

595
32. Home | COV-Boost [Internet]. [cited 2021 Jun 17]. Available from:
596 />597
33.
598
599
600

iew
ed

593
31. Mahase E. Covid-19: Booster vaccine to be rolled out in autumn as UK secures 60m more Pfizer
594 doses. BMJ (Clinical research ed). 2021 Apr 29;373:n1116.

Khoury DS, Cromer D, Reynaldi A, Schlub TE, Wheatley AK, Juno JA, et al. Neutralizing antibody levels
are highly predictive of immune protection from symptomatic SARS-CoV-2 infection. Nature Medicine
[Internet]. 2021 May 17 [cited 2021 Jun 22];1–7. Available from: />
pe
er

re
v

601
34. Feng S, Phillips Mmath DJ, White Phd T, Sayal Phd H, Aley PK, Phd SB, et al. Correlates of protection
602 against symptomatic and asymptomatic SARS-CoV-2 infection. [cited 2021 Jun 24]; Available from:
603 />604
35. Baden LR, El Sahly HM, Essink B, Kotloff K, Frey S, Novak R, et al. Efficacy and Safety of the mRNA605 1273 SARS-CoV-2 Vaccine. New England Journal of Medicine. 2020 Dec 30;

606
36. Walsh EE, Frenck RW, Falsey AR, Kitchin N, Absalon J, Gurtman A, et al. Safety and Immunogenicity of
607 Two RNA-Based Covid-19 Vaccine Candidates. New England Journal of Medicine. 2020 Oct
608 14;383:2439–50.
609
37. AstraZeneca ChAdOx1-S/nCoV-19 [recombinant], COVID-19 vaccine [Internet]. [cited 2021 Jun 9].
610 Available from: />611
38. Parry H, Bruton R, Stephens C, Brown, Zuo J, Moss P. Extended interval BNT162b2 vaccination
612 enhances peak antibody generation in older people. medRxiv [Internet]. 2021 May 17 [cited 2021 Jun
613 24];2021.05.15.21257017. Available from: />
ot

614
39. Coronavirus (COVID-19) Vaccinations - Statistics and Research - Our World in Data [Internet]. [cited
615 2021 Jun 20]. Available from: />616

Pr

ep

rin


tn

617

This preprint research paper has not been peer reviewed. Electronic copy available at: />

618

d
e

Table 1. Baseline demographics and characteristics by cohort and study arm in the 28-day boost study arms

w
e
i
v

Prime with ChAd

Cohort/
Characteristic

Prime with BNT

ChAd/ChAd-28

ChAd/BNT-28


BNT/BNT-28

BNT/ChAd-28

(N=90)

(N=90)

(N=93)

(N=90)

58.2 (4.81)

58.0 (4.76)

58.2 (4.85)

57.6 (50.1, 69.1)

57.6 (50.3, 68.1)

General
Age (years)
Mean (SD)
Median (range)

57.7 (50.2, 69.3)

56.1 (50.5, 68.9)


49 (52.7%)

41 (45.6%)

44 (47.3%)

49 (54.4%)

65 (72.2%)

76 (81.7%)

66 (73.3%)

1 (1.1%)

-

2 (2.2%)

15 (16.7%)

7 (7.5%)

9 (10.0%)

6 (6.7%)

8 (8.6%)


10 (11.1%)

3 (3.3%)

2 (2.2%)

3 (3.3%)

16 (17.8%)

18 (19.4%)

21 (23.3%)

16 (17.8%)

11 (12.2%)

11 (11.8%)

11 (12.2%)

7 (7.8%)

8 (8.9%)

-

2 (2.2%)


(N=25)

(N=24)

(N=26)

(N=25)

55.7 (4.26)

58.4 (4.60)

56.7 (5.04)

57.6 (4.65)

55.3 (50.7, 64.1)

58.9 (51.8, 68.3)

54.7 (50.1, 67.2)

55.8 (51.4, 67.0)

Gender
Female

38 (42.2%)


Male

52 (57.8%)

White

70 (77.8%)

Black

1 (1.1%)

Asian

13 (14.4%)

Mixed

6 (6.7%)

50 (55.6%)

-

Comorbidities
Respiratory
Diabetes
Immunology
Age (years)


r
P

Mean (SD)

p
e

Median (range)

t
o

n
t

Other

ir n

19 (21.1%)

r
e
e
p
40 (44.4%)

Ethnicity


Cardiovascular

e
r

57.3 (4.56)

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Prime with ChAd

Cohort/
Characteristic

d
e

Prime with BNT

ChAd/ChAd-28

ChAd/BNT-28

BNT/BNT-28

BNT/ChAd-28

Female

13 (52.0%)


9 (37.5%)

12 (46.2%)

10 (40.0%)

Male

12 (48.0%)

15 (62.5%)

14 (53.8%)

White

17 (68.0%)

17 (70.8%)

Black

-

-

Asian

6 (24.0%)


Mixed

2 (8.0%)

Other

-

w
e
i
v

Gender

Ethnicity

Cardiovascular

7 (28.0%)

Respiratory

5 (20.0%)

Diabetes

6 (24.0%)


619

3 (12.5%)

e
r
2 (7.7%)

3 (12.0%)

-

1 (3.8%)

-

6 (25.0%)

10 (38.5%)

7 (28.0%)

6 (25.0%)

6 (23.1%)

5 (20.0%)

1 (4.2%)


2 (7.7%)

1 (4.0%)

r
e
e
p
4 (16.7%)

Comorbidities

t
o

15 (60.0%)

17 (65.4%)
2 (7.7%)

18 (72.0%)
-

4 (15.4%)

4 (16.0%)

n
t


620

r
P

p
e

ir n

This preprint research paper has not been peer reviewed. Electronic copy available at: />

621

d
e

Table 2. Immune responses between heterologous and homologous prime/boost schedules at 28 days post boost dose in the 28-day boost study arms

Prime with ChAd
ChAd/ChAd-28
ChAd/BNT-28
Per-protocol analysis
N=104
N=104
SARS-CoV-2 anti-spike IgG,
1392 (1188-1630)
12906 (11404-14604)
ELU/ml
[n=104]

[n=104]
Modified ITT
N=105
N=108
SARS-CoV-2 anti-spike IgG,
1387 (1186-1623)
12995 (11520-14660)
ELU/ml
[n=105]
[n=108]
Pseudotype virus
61 (50-73)
515 (430-617)
neutralising antibody, NT50
[n=101]
[n=101]
Cellular response, SFC/106
50 (39-63)
185 (152-224)
PBMCs
[n=104]
[n=108]
622
* Data shown are geometric mean (95% CI) for continuous variables;
623

§

624


sided 95% CIs in the modified ITT analyses; non-inferiority margin is 0.63.

GMR§
9.2
(97.5% CI:7.5,∞)
9.3
(95% CI:7.7,11)
8.5
(95% CI:6.5,11)
3.8
(95% CI:2.8,5.1)

BNT/BNT-28
N=109
14080 (12491-15871)
[n=109]
N=110
13938 (12358-15719)
[n=110]
574 (475-694)
[n=102]
80 (63-102)
[n=110]

r
e
e
p

Prime with BNT

BNT/ChAd-28
N=109
7133 (6415-7932)
[n=109]
N=109
7133 (6415-7932)
[n=109]
383 (317-463)
[n=104]
99 (77-126)
[n=109]

e
r

w
e
i
v

GMR§

0.51
(97.5% CI:0.43, ∞)
0.51
(95% CI:0.44,0.6)
0.67
(95% CI:0.51,0.88)
1.2
(95% CI:0.88,1.7)


GMRs were adjusted for randomisation stratification variables, including study site and cohort, with one-sided 97.5% CIs in per-protocol analyses and two-

t
o

625

n
t

r
P

p
e

ir n

This preprint research paper has not been peer reviewed. Electronic copy available at: />