TFT is employed by Novartis Vaccines. The nested vaccine study is funded by Novartis Vaccines and Diagnostics GmbH, Marburg, Germany, who provided vaccines and funding as an unrestricted institutional grant. The main Copenhagen Prospective Study on Asthma in Childhood birth cohort study is funded by private and public research funds (see
Conceived and designed the experiments: HB ALB NVF. Performed the experiments: ALB NVF CGC LP JS MH AB. Analyzed the data: ALB SB. Contributed reagents/materials/analysis tools: HB TFT. Wrote the paper: ALB NVF CGC LP JS MH AB SB TFT HB.
Pregnant women were suspected to be at particular risk when H1N1pnd09 influenza became pandemic in 2009. Our primary objective was to compare the immune responses conferred by MF59®-adjuvanted vaccine (Focetria®) in H1N1pnd09-naïve pregnant and non-pregnant women. The secondary aims were to compare influences of dose and adjuvant on the immune response.
The study was nested in the Copenhagen Prospective Studies on Asthma in Childhood (COPSAC2010) pregnancy cohort in 2009-2010 and conducted as a single-blinded block-randomised [1∶1∶1] controlled clinical trial in pregnant women after gestational week 20: (1) 7.5 µg H1N1pnd09 antigen with MF59-adjuvant (Pa7.5 µg); (2) 3.75 µg antigen half MF59-adjuvanted (Pa3.75 µg); (3) 15 µg antigen unadjuvanted (P15 µg); and in non-pregnant women receiving (4) 7.5 µg antigen full adjuvanted (NPa7.5 µg). Blood samples were collected at baseline, 3 weeks, 3 and 10 months after vaccination, adverse events were recorded prospectively.
58 pregnant women were allocated to Pa7.5 µg and 149 non-pregnant women were recruited to NPa7.5 µg. The sero-conversion rate was significantly increased in non-pregnant (NPa7.5 µg) compared with pregnant (Pa7.5 µg) women (OR = 2.48 [1.03–5.95], p = 0.04) and geometric mean titers trended towards being higher, but this difference was not statistically significant (ratio 1.27 [0.85–1.93], p = 0.23). The significant titer increase rate showed no difference between pregnant (Pa7.5 µg) and non-pregnant (NPa7.5 µg) groups (OR = 0.49 [0.13–1.85], p = 0.29).
Our study suggests the immune response to the 7.5 µg MF59-adjuvanted Focetria®
ClinicalTrials.gov
Pregnant women experience increased influenza related morbidity and mortality during seasonal influenza epidemics,
We had the unique opportunity to study the
We conducted a randomized controlled clinical trial primarily to compare immunogenicity of the
This study is reported in accordance with the CONSORT guidelines
This study was nested in the novel COPSAC2010 cohort; an on-going, unselected, prospective clinical pregnancy cohort study of 743 women recruited in Zealand, Denmark, during 2009–2010. The recruitment was previously described in detail
The women participating in the COPSAC2010 cohort were invited to be enrolled in this phase IV randomized, participant-blinded study in 2009–2010. The pregnant women were recruited from gestational week 20 and women continuing in the birth cohort with their children were recruited up to 8 months after birth. Key exclusion criteria were; chronic endocrinological, nephrological or cardiac diseases; severe asthma; history of anaphylaxis or other serious vaccine reactions; or hypersensitivity to influenza viral proteins, any excipients, eggs (including ovalbumin), or chicken proteins.
Monovalent influenza A/California/2009 (
The vaccines were portioned, in identical syringes, into two different individual dosages. One for the 7.5 µg full MF59-adjuvanted and 3.75 µg half MF59-adjuvanted dosages and one for the 15 µg unadjuvanted dose. The vaccine was administered as an intramuscular injection into the deltoid muscle. The women were observed for 30 minutes after the injection.
The study was conducted in accordance with the guiding principles of the Declaration of Helsinki and approved by the Ethics Committee of Copenhagen (H-B-2008-093), the Danish Medicines Agency (EudraCT 2009-016877), and the Danish Data Protection Agency (2009-41-4031). The validity of the data was ensured by complying with Good Clinical Practice guidelines and quality-control procedures. (ClinicalTrials.gov number, NCT01012557). All participants gave their written informed consent prior to enrollment.
The women visited the research clinic after three weeks for a structured clinical interview performed by the research doctors interviewing for specific local reactions (pain, erythema, swelling, and bruising), systemic symptoms (chills, malaise, headache, myalgia, nausea, and vomiting), fever, and use of analgesics. Symptoms and reactions were graded as: none; mild (did not interfere with daily activity); moderate (interfered with daily activity); and severe (prevented daily activity). At two further clinical visits, after three and 10 months, unsolicited events affecting mother or child were recorded. Any reaction that resulted in hospitalization or was life-threatening was considered as a serious adverse event.
Serum was sampled prior to vaccination (baseline), 3 weeks, 3 months, and 10 months after vaccination. Serum plasma levels of antibodies were determined in twofold dilutions in a conventional hemagglutination-inhibition (HI) assay
Sero-conversion was defined as a change from a pre-vaccination HI titer <10 to a post-vaccination HI titer ≥40; significant increase as fourfold or higher post-vaccination titer from a pre-vaccination HI titer ≥10, and sero-protection as the proportion of subjects achieving an HI titer of 40 or greater. See
Sero-protection: | HI titer ≥40 | |
Pre-vaccination: | HI titer ≥40 | |
Post-vaccination: | HI titer ≥40 | |
Pre-vaccination: | HI titer ≥10 | |
Post-vaccination: | HI titer ≥4 x pre-vaccination titer |
The HI antibody response-criteria for success in young adults as defined by the EMEA criteria
The study power estimate was based on a minimal detectable Geometric Mean Titer (GMT) ratio of 1.5 between the Pa7.5 µg and the NPa7.5 µg groups, a coefficient of variation (CV) of 1.0, a ratio of 1:3 between number of pregnant and number of non-pregnant women, a power of 80%, and an two-sided alpha-level of 5%. This required us to recruit a total number of 180 women in the Pa7.5 µg and the NPa7.5 µg groups.
GMT were computed from the log10-transformed mean. GMT in the vaccination groups are compared in levels and trends over time using general estimating equations (GEE) analysis (repeated-measures analysis) with an independent working correlation structure. The motivation for using GEE analysis is to take advantage of the full data in order to report one overall conclusion on the difference in sero-conversion rates between the groups over time. Differences in GMT between groups are reported as ratios (R). Confidence intervals are reported in squared brackets. GMT at 3 weeks, 3 and 10 months are adjusted for baseline GMT, using baseline GMT as a covariate. Analysis on the Pa7.5 µg, Pa3.75 µg and P15µg groups were adjusted for gestational age at time of vaccination.
GMR are the ratio between baseline titer and titer at any time point.
Percentage of women sero-converted is calculated as the percentage of women with a pre-vaccination HI titer <10 who converted to a post-vaccination HI titer ≥40. Percentage of women with significant titer increase is calculated as the percentage of women with a pre-vaccination HI titer ≥10 who had a fourfold or higher post-vaccination titer. Percentage of women with sero-conversion or significant increase is calculated as percentage of the total number of women in the vaccination group.
Sero-protection, sero-conversion and significant titer increase in the vaccination groups are compared in levels and trends over time using general estimating equations (GEE) analysis (repeated-measures analysis) with an independent working correlation structure. Differences in sero-conversion and significant titer increases between groups are reported as longitudinal odds ratios (OR). Confidence intervals are reported in squared brackets. Due to large scatter in gestational age at time of vaccination in the pregnant women, analyses on the Pa7.5 µg, Pa3.75 µg and P15 µg groups were adjusted for gestational age at time of vaccination.
Local and systemic post-vaccination reactions are reported as percentages based on number of women and severity. We used Fisher's exact test to compare vaccine groups. Exact (Clopper-Pearson) confidence intervals are reported for all proportional endpoints in squared brackets.
A p-value <0.05 was considered as significant.
Analyses were done using SAS version 9.2 (SAS Institute, Cary, NC).
296 women were included in this study from November 2009 to August 2010 of which 149 were non-pregnant (NPa7.5 µg). The 147 pregnant women were randomly assigned to the three groups Pa7.5 µg (58 women), Pa3.75 µg (28 women), and P15 µg (61 women). The Pa3.75 µg group was closed after four months due to a lower-than-expected recruitment rate to the study. There was a follow-up of 87% of the women 10 months after vaccination and the study was closed July 2011. Flow of subjects into the study-analysis is illustrated in the
Pa7.5µg MF59- adjuvanted | Pa3.75µg MF59- adjuvanted | P15µg Not- adjuvanted | NPa7.5µg MF59- adjuvanted | P | |
N | 58 | 28 | 61 | 149 | |
Age – year Median (Range) | 32(19–40) | 34(26–40) | 31(20–41) | 32(23–49) | 0.12 |
Gestational age at vaccination – weeks Median (Range) | 25+0 (21+4 – 40+1) | 29+0 (21+1 – 38+6) | 24+6 (20+0 – 37+0) | ||
Time from birth to vaccination – days Median (Range) | 127 (6–244) | ||||
0.08 |
|||||
<400.000 Dkkr | 9(5) | 4(1) | 22(13) | 13(19) | |
400.000–1.000.000 Dkkr | 82(45) | 96(27) | 73(43) | 84(125) | |
>1.000.000 Dkkr | 9(5) | 0 | 5(3) | 3(5) | |
0.21 |
|||||
Yes | 48(28) | 68(19) | 64(39) | 55(82) | |
No | 51(30) | 32(9) | 36(22) | 45(67) | |
Living with partner %(n) | 0.69 |
||||
Yes | 100(58) | 100(28) | 98(60) | 97(144) | |
No | 0 | 0 | 2(1) | 3(5) |
ANOVA.
Fisher's exact test.
The gestational age at time of vaccination is different in the Pa3.75 µg group compared with the Pa7.5 µg group (P = 0.01). There is no significant difference between Pa7.5 µg and P15 µg (P = 0.82). Wilcoxon Rank Sum Test (t approximation).
We saw an overall successful sero-protective response from the H1N1pnd09 vaccination in more than 91% of pregnant and 99% of non-pregnant women as defined by one of the EMEA criteria
Vaccine group | ||||
Pa7.5µg | Pa3.75µg | P15µg | NPa7.5µg | |
N | 58 | 28 | 61 | 149 |
GMT |
10 |
11 |
9 |
12 |
Sero-protection |
17 |
17 |
13 |
13 |
N | 56 | 28 | 59 | 148 |
GMT |
345[244–487] | 206[126–335] | 202[145–283] | 465[374–575] |
GMR [95%CI] | 33.2[22.5–49.2] | 18.8[10.8–32.8] | 21.7[14.8–31.8] | 40.5[31.8–51.5] |
Sero-protection |
96[88–100] | 89[72–98] | 88[77–95] | 99[95–100] |
Sero-conversion %[95% CI] | 95[82–99] | 86[57–98] | 83[67–93] | 98[91–100] |
Significant increase %[95% CI] | 95[74–100] | 93[66–100] | 89[67–99] | 86[76–94] |
Sero-conversion or Significant increase %[95% CI] | 95[85–99] | 89[72–98] | 85[73–93] | 93[87–96] |
N | 49 | 27 | 58 | 147 |
GMT |
139[93–208] | 87[51–150] | 111[77–161] | 207[164–261] |
GMR [95%CI] | 12.5[8.1–19.3] | 8.0[4.4–14.4] | 11.5[7.7–17.1] | 18.0[14.0–23.1] |
Sero-protection |
82[68–91] | 74[54–89] | 76[63–86] | 94[89–97] |
Sero-conversion %[95% CI] | 70[51–85] | 64[35–87] | 74[58–87] | 91[83–96] |
Significant increase %[95% CI] | 84[60–97] | 69[39–91] | 63[38–84] | 79[67–88] |
Sero-conversion or Significant increase %[95% CI] | 76[61–87] | 67[46–83] | 71[57–82] | 86[79–91] |
N | 47 | 24 | 50 | 135 |
GMT |
75[48–118] | 63[34–118] | 85[55–130] | 78[60–101] |
GMR [95%CI] | 6.8[4.2–10.8] | 5.5[2.9–10.6] | 8.4[5.3–13.2] | 6.8[5.1–8.9] |
Sero-protection |
70[55–83] | 67[45–84] | 74[60–85] | 75[66–82] |
Sero-conversion %[95% CI] | 59[39–76] | 55[23–83] | 70[51–84] | 69[57–79] |
Significant increase %[95% CI] | 67[41–87] | 54[25–81] | 53[28–77] | 59[46–71] |
Sero-conversion or Significant increase %[95% CI] | 62[46–75] | 54[33–74] | 64[49–77] | 64[56–72] |
P: Pregnant.
NP: Non-pregnant.
GMT: Geometric mean titer.
GMR: Geometric mean ratio.
CI: Confidence interval.
Sero-protection: Titer ≥40.
Sero-conversion: Pre titer<10, Post titer ≥40.
Significant increase: Pre titer ≥10, Post titer 4 fold Pre titer.
Sero-conversion or significant increase: Percentage of total number of women in each vaccine group.
At day 0 GMT were the same in all groups (ANOVA, p = 0.51).
The number of women with sero-protection did not differ in any of the groups compared over time (GEE).
GMT at 3 weeks, 3 months and 10 months were adjusted for baseline titer.
The CV for GMT was 1.8. Subjects with baseline titers >40 were found in all groups. Therefore, GMT at 3 weeks, 3 months and 10 months were adjusted for baseline titer (
The trends over time for geometric mean titers in the vaccine groups were not significantly different using general estimating equations.
The interactions with time of the vaccine groups were not significant; i.e. the trends over time for the number of women sero-converted, and for the women with significant titer increase were not significantly different when comparing pregnant (Pa7.5 µg)
The trends over time for the number of women sero-converted were not significantly different. The number of non-pregnant women (NPa7.5 µg) who sero-converted was 2.48-fold higher than among pregnant (Pa7.5 µg) women using general estimating equations.
GMT numerical values (
There was no significant difference between the Pa3.75µg group and the Pa7.5 µg group with respect to GMT (R = 0.63 [0.33-1.24], p = 0.18) (
There was no significant difference between the P15µg group and the Pa7.5µg group with respect to GMT (R = 0.83 [0.48–1.42], p = 0.49) (
Local and systemic reactions in the first 3 weeks following vaccination are shown in
Pa7.5µg MF59-adjuvanted | Pa3.75µg MF59-adjuvanted | P15µg Not-adjuvanted | NPa7.5µg MF59-adjuvanted | ||
N | 58 | 28 | 61 | 149 | |
Missing data %(n) | 5(3) | 7(2) | 8(5) | 2(3) | |
Pain |
None | 18 |
27 |
82[70–91] |
24 |
Mild | 56[42–70] | 69[48–86] | 16 |
61[52–69] | |
Moderate | 22 |
4[0–20] | 2[0–10] | 14 |
|
Severe | 4[0–13] | 0[0–13] | 0[0–6] | 1[0–4] | |
Erythema |
None | 89[78–96] | 96[80–100] | 98[90–100] |
89[83–94] |
Mild | 7 |
4[0–20] | 0[0–6] | 10 |
|
Moderate | 4[0–13] | 0[0–13] | 2[0–10] | 1[0–4] | |
Severe | 0[0–6] | 0[0–13] | 0[0–6] | 0[0–2] | |
Swelling |
None | 85[73–94] | 81[61–93] | 98[90–100] |
87[81–92] |
Mild | 15 |
19 |
0[0–6] | 11 |
|
Moderate | 0[0–6] | 0[0–13] | 2[0–10] | 2[0–6] | |
Severe | 0[0–6] | 0[0–13] | 0[0–6] | 0[0–2] | |
Bruising | None | 93[83–98] | 96[80–100] | 96[88–100] | 95[90–99] |
Mild | 5 |
0[0–13] | 2[0–10] | 4 |
|
Moderate | 2[0–10] | 4[0–20] | 2[0–10] | 1[0–4] | |
Severe | 0[0–6] | 0[0–13] | 0[0–6] | 0[0–2] |
Mild: Not interfering with daily activity.
Moderate: Interfering with daily activity.
Severe: Preventing in engaging in daily activity.
The number of women who experienced pain as an adverse event were significantly less in the group vaccinated with the 15 µg non-adjuvanted vaccine, than in the other groups (P<0.000001, Fisher's exact test).
The number of women who experienced erythema as an adverse event were significantly less in the group vaccinated with the 15 µg non-adjuvanted vaccine, than in the other groups (P = 0.03, Fisher's exact test).
The number of women who experienced pain as an adverse event were significantly less in the group vaccinated with the 15 µg non-adjuvanted vaccine, than in the other groups (P = 0.008, Fisher's exact test).
Pa7.5µg MF59-adjuvanted | Pa3.75µg MF59-adjuvanted | P15µg Not-adjuvanted | NPa7.5µg MF59-adjuvanted | ||
N | 58 | 28 | 61 | 149 | |
Missing data %(n) | 5(3) | 7(2) | 8(5) | 2(3) | |
Chills | None | 95[85–99] | 100[87–100] | 93[83–98] | 91[85–95] |
Mild | 5 |
0[0–13] | 4[0–12] | 8 |
|
Moderate | 0[0–6] | 0[0–13] | 2[0–10] | 0[0–2] | |
Severe | 0[0–6] | 0[0–13] | 2[0–10] | 1[0–4] | |
Malaise | None | 78[65–88] | 92[75–99] | 86[74–94] | 73[66–80] |
Mild | 16 |
8 |
9 |
19 |
|
Moderate | 5 |
0[0–13] | 4[0–12] | 6 |
|
Severe | 0[0–6] | 0[0–13] | 2[0–10] | 1[0–4] | |
Headache | None | 89[78–96] | 96[80–100] | 89[78–96] | 88[81–93] |
Mild | 9 |
4[0–20] | 4[0–12] | 7 |
|
Moderate | 2[0–10] | 0[0–13] | 4[0–12] | 3 |
|
Severe | 0[0–6] | 0[0–13] | 4[0–12] | 2[0–6] | |
Myalgia | None | 91[80–97] | 100[87–100] | 96[88–100] | 88[81–93] |
Mild | 9 |
0[0–13] | 2[0–10] | 10 |
|
Moderate | 0[0–6] | 0[0–13] | 0[0–6] | 2[0–6] | |
Severe | 0[0–6] | 0[0–13] | 2[0–10] | 1[0–4] | |
Nausea | None | 100[94–100] | 100[87–100] | 96[88–100] | 93[88–97] |
Mild | 0[0–6] | 0[0–13] | 0[0–6] | 5 |
|
Moderate | 0[0–6] | 0[0–13] | 4[0–12] | 1[0–4] | |
Severe | 0[0–6] | 0[0–13] | 0[0–6] | 1[0–4] | |
Vomiting | None | 100[94–100] | 100[87–100] | 98[90–100] | 99[95–100] |
Mild | 0[0–6] | 0[0–13] | 2[0–10] | 0[0–2] | |
Moderate | 0[0–6] | 0[0–13] | 0[0–6] | 1[0–4] | |
Severe | 0[0–6] | 0[0–13] | 0[0–6] | 1[0–4] | |
Fever | None | 98[90–100] | 96[80–100] | 93[83–98] | 95[90–98] |
>38° C | 2[0–10] | 4[0–20] | 7 |
5 |
|
Analgesics | None | 95[85–99] | 96[80–100] | 95[85–99] | 90[85–95] |
Yes | 5 |
4[0–20] | 5 |
10 |
Mild: Not interfering with daily activity.
Moderate: Interfering with daily activity.
Severe: Preventing in engaging in daily activity.
The sero-conversion OR was 2.48-fold in non-pregnant women (NPa7.5 µg) compared with pregnant women (Pa7.5 µg) after receiving the same standard H1N1pnd09 vaccine. Likewise GMT was nominally higher in the non-pregnant than in the pregnant women.
As young adults were at high risk for illness during the pandemic, it was of interest that even after 10 months, 70% −74% of pregnant (Pa7.5 µg and P15 µg) and 75% non-pregnant women (NPa7.5 µg) were protected against H1N1 according to the EMEA criteria with a HI titer of 40 or greater.
Women receiving the non-adjuvanted vaccine had significantly fewer local reactions but similar rates of systemic reactions as women receiving the adjuvanted vaccine. There were no reports of serious adverse events in any of the groups.
It is a major strength to this study that it is nested in the on-going well established COPSAC2010 cohort study, which ensured a high follow-up rate and close observations in a clinical research centre.
The non-pregnant subjects were post-partum women drawn from the parent study and had delivered within 8 months prior to vaccination. As immune changes associated with pregnancy may be carried over transiently to the post-partum period, this would have had the effect of minimizing differences between pregnant and non-pregnant subjects, meaning that our observations were conservative.
It is a limitation of this study, that the recruitment was hampered by two main factors; primarily the Danish Health Authorities were ambiguous in their recommendations for vaccination of pregnant women changing their recommendations during the pandemic. This ambivalence caused the women to doubt the necessity and safety of the offered vaccination. Secondly the clinical symptoms of the pandemic proved less serious than expected. Therefore, failing to recruit at the scheduled rate, we chose to close the Pa3.75 µg treatment arm halfway into the recruitment period. The low number of participants in this study arm limits the power of the conclusion on this particular treatment.
As a consequence of lower than expected recruitment the power to detect differences between study arms was lower than planned.
The response to the unadjuvanted vaccine containing the usual 15 µg of antigen in the pregnant women can only be compared to the adjuvanted vaccine containing 7.5 µg of antigen in the non-pregnant women, since this study did not include a non-pregnant 15 µg dose unadjuvanted group. Hence we could not report on the adequacy of the usual unadjuvanted 15 µg antigen dose in pregnancy. It is a limitation of the study that it was not observer blind; but the extensive interviews for adverse events were part of the comprehensive clinic interviews in the main study with a focus on asthma, eczema and allergy and were integrated within the framework of the on-going COPSAC2010 study.
The serum plasma levels of antibodies were determined by only the HI assay. Using a second method e.g. the microneutralization assay
Finally, it is a limitation to our study (though inevitable), that the baseline GMT was raised in certain individuals in all four groups, indicating that some individuals in the population were already protected against the novel influenza H1N1pnd09. This could be due to the cross-reactivity from other influenza viruses
The increased risk for severe influenza in pregnant women was demonstrated again in the 2009 pandemic, underscoring the importance of protecting this vulnerable group by vaccination and with antiviral therapy
Vaccinating pregnant women against influenza may have a secondary benefit by passively protecting the parturient woman's infant,
The increased risk for severe influenza in pregnant women is not fully understood. Physiological changes associated with pregnancy play a role in increased severity of influenza. Immunity against secondary infections is also potentially important. At least one third of the deaths due to the pandemic were complicated by secondary bacterial infection
The response of pregnant women to unadjuvanted vaccine (containing 15 µg of HA) seemed adequate, though we observed numerically higher GMTs in response to adjuvanted vaccine containing one half the antigen, and equal responses to adjuvanted vaccine containing one fourth the antigen (3.75 µg). This degree of antigen-sparing from adjuvants has been observed consistently in previous studies in children, young and older adults
We find a significantly lower number of pregnant women who sero-converted than among non-pregnant women. Our study suggests a trend towards a reduced immune response to the 7.5 µg MF59 adjuvanted Focetria
(DOC)
(DOC)
We gratefully express our gratitude to the children and families of the COPSAC2010 cohort study for all their support and commitment. We acknowledge and appreciate the unique efforts of the COPSAC research team. Novartis Vaccines and Diagnostics, Marburg, Germany are appreciated for technical support.
The protocol, the technical appendix, statistical code, and dataset are available from the corresponding author at