Enhanced passive safety surveillance of a quadrivalent inactivated split virion influenza vaccine in Finland during the influenza season 2020/21

Seasonal influenza epidemics are estimated to cause between 291,243 and 645,832 deaths annually with an excess mortality rate of 0.1–6.4 per 100,000 individuals for those < 65 years of age and 17.9–223.5 per 100 000 for those older than 75 years [1].

Annual seasonal influenza vaccination is the primary strategy used globally to limit the burden of seasonal influenza [2, 3]. Annual vaccination is recommended by the World Health Organization (WHO) for high-risk groups including pregnant women, children aged < 5 years, the elderly, and individuals with chronic health conditions [4, 5]. However, current influenza vaccines generally induce strain-specific immunity. As influenza virus strains continuously evolve through antigenic drift resulting in new variants or strains [2], vaccines need to be frequently updated to match the strains most likely to be circulating during the upcoming season [2]. This leads to specific challenges for the continued monitoring of vaccine safety.

Since 2014, the European Medicines Agency (EMA) has required manufacturers of seasonal influenza vaccines to conduct annual enhanced safety surveillance, replacing previous requirements based on data from small-scale safety and immunogenicity clinical trials [6, 7]. Specific guidance has been produced by the EMA’s Pharmacovigilance Risk Assessment Committee (PRAC) for enhanced safety surveillance [6]. The purpose of this surveillance is to rapidly detect and evaluate potential new safety concerns before the peak immunisation period of the vaccine in any particular year. Increased rates of reactogenicity or allergic events compared to those expected or reported following vaccination during the the previous season may signal new safety concerns and may indicate a potential for increased severity with increased vaccine exposure [6]. The use of enhanced safety surveillance to monitor adverse drug reactions (ADRs) is recommended by the EMA for the purpose of regulatory submission and review [6]. Using a passive surveillance system based on individual and healthcare professional (HCP) reporting of ADRs has been shown to improve the rates of adverse event (AE) reporting in comparison to previous years where passive surveillance systems relied on ADR reporting by HCPs alone [8, 9].

In 2016, a quadrivalent inactivated split-virion influenza vaccine (IIV4; Vaxigrip Tetra®, Sanofi), which offers broad protection against influenza through the inclusion of two influenza A and two influenza B virus strains, was approved in Europe [10]. IIV4 is indicated for use in adults and children from 6 months of age [11].

Several adjustments have been made to methods of enhanced passive safety surveillance (EPSS) reporting which have been used for for IIV4 surveillance over previous years, with the aim of improving both the quality of the data and the reporting stimulus. The introduction of a combination of a paper reporting method, using a pre-paid envelope, and phone reporting to replace email reporting, during the 2019/20 EPSS for individuals vaccinated with IIV4 provided better quality reporting. This was demonstrated by the higher than expected rates of spontaneous reporting (6.0%) in comparison to the method of email reporting alone that was used during the northern hemisphere 2018/19 EPSS for IIV4 [4, 12].

Here we report an EPSS for individuals vaccinated with IIV4 during the northern hemisphere (NH) 2020/21 season, in which we used electronic reporting via an electronic data capture system (EDC), rather than email or mail, in addition to the standard reporting method by telephone.

The overall aim of this study was to detect any clinically significant change in the frequency or severity of expected reactogenicity, compared with what is known or expected with the vaccine as defined by the Summary of Product Characteristics (SmPC). We also described the estimated reporting rates of suspected ADRs occurring within 7 days following routine vaccination with IIV4 and compared them with previous data for the EPSS 2019/20 influenza season and with expected frequencies as per the SmPC.

A total of 1008 participants were included; most (99.8%) were ≥ 18 years old. In total, 29 reports were received for 82 suspected ADRs. The time to ADR onset was known for 79 (79/82; 96.3%) ADRs, all of which occurred within the first 7 days of vaccination. Of these, the majority (70/79; 88.61%) occurred on the same day or the following day after vaccination.

For the primary endpoint, 28 participants who received the IIV4 vaccine reported 79 suspected ADRs within 7 days following vaccination, corresponding to a vaccinee reporting rate of 2.78% (95% CI: 1.85, 3.99) (ADR reporting rate, 7.84% [95% CI: 6.25, 9.67%]) (Table 1). Among older participants (> 65 years), the vaccinee reporting rate was 6.25% (95% CI: 0.77, 20.81), while those aged 18–65 years had a vaccinee reporting rate of 2.67% (95% CI: 1.73, 3.91) (ADR reporting rates of 15.63% [95% CI: 5.28, 32.79] and 7.58% [95% CI: 5.97, 9.46], respectively). No ADRs were reported by those aged 13–18 years (number of VCs distributed, n = 4) and of only two participants in the 6–13 years age group, one reported 3 ADRs (Table 1). Seventy-two PRAC ADRs of interest occurred in 27 vaccinees within 7 days following vaccination with IIV4, corresponding to a vaccinee reporting rate of 2.68% (95% CI: 1.77, 3.87%) (PRAC ADR reporting rate: 7.14% [95% CI: 5.63, 8.91%]) (Table 1).

Reporting rates for all ADRs (including those without known time to onset) did not differ substantially from those known to have occurred within the first 7 days (Table 2). The most frequently reported ADRs were injection site reactions (vaccination site pain, vaccination site erythema and vaccination site swelling; n = 46, 2.28%), myalgia (n = 9, 0.89%) and headache (n = 8, 0.79%) (Table 3). No serious suspected ADRs were reported at any point post-vaccination. However, two ADRs (vaccination site swelling and headache) were of severe intensity; these were both PRAC ADRs of interest.

The total vaccinee reporting rates and ADR reporting rates tended to be lower in the current season (2020/21) than in the previous season (2019/20), for suspected ADRs occurring ≤ 7 days post vaccination (Table 2).

A number of PRAC ADRs of interest reported in the current EPSS were not observed in the previous year. These included 11 reports of vaccination site swelling reported by 11 vaccinees (vaccinee reporting rate, 1.09% [95% CI: 0.55, 1.94]): 1 in the ≥ 6 to < 13 years age group, 9 in the ≥ 18 to ≤ 65 years age group and 1 aged > 65 years. This was in line with the expected frequency stated in the SmPC, which lists vaccination site swelling as common (1% to 10%) in adults and the elderly and very common (≥ 10%) among children aged 6 to 13 years. The other PRAC ADRs of interest were reported in both previous and current seasons.

In this EPSS conducted in Finland during the NH 2020/21 influenza season, 2.78% of the vaccinees reported 79 suspected ADRs within 7 days following vaccination with IIV4. These findings were favourable compared with those reported in the EPSS from the previous year in Finland (overall vaccinee reporting rate for the previous year (2019/20), 5.96% [95% CI: 4.45, 7.48%] and ADR reporting rate, 17.36% [95% CI: 14.99, 19.94%]) [12]. Furthermore, the safety profile was comparable to that described in the SmPC [13]. For the primary objective, no safety concerns were observed during the surveillance period, in line with findings from EPSS studies conducted in previous influenza seasons [4, 12, 14]. Over half of the events reported during the current EPSS were of mild intensity and 2.5% of ADRs reported were of severe intensity (vaccination site swelling and headache).

For the current season, we implemented electronic reporting via the EDC using vaccinee-specific accounts in addition to the standard phone reporting method. While this digital reporting strategy provided the expected good quality of reporting, reporting rates were surprisingly lower versus the paper reporting method (pre-paid envelope) used in the 2019/20 season. Possible reasons for this include the inherent risks associated with using a digital solution, such as technical issues in accessing the EDC in time or a lack of vaccinee adherence to the EDC solution, thereby possibly hindering ADR reporting. This is however unlikely as reporting via telephone was provided as a back-up solution. Additionally, only one vaccinee reported facing technical difficulties with the EDC. Overall, the EDC solution was found to be well accepted by the vaccinees, with no negative feedback during follow-ups by the sites/ local PV. The current disruption to health-seeking behaviour and staffing/routines in clinics and to surveillance priorities and capacities as a result of the ongoing coronavirus disease 19 (COVID-19) pandemic has led to a decrease in influenza information sharing globally [15]. Such disruptions during the 2020/21 influenza season could be one explanation for the lower reporting rates observed in this study. It is also possible that confidence in flu vaccines has increased over time, with IIV4 being on the market for at least three seasons.

The vaccinee reporting rate reported here for IIV4 is in line with published reporting rates of spontaneous ADR reports after vaccination with other seasonal influenza vaccines, ranging from 20 to 90 reports per 1,000,000 people vaccinated [16‐20]. The ability of EPSS to improve reporting rates following vaccination has been previously demonstrated in a number of studies [8, 14, 16, 21, 22]. In an Australian investigation, the introduction of EPSS increased the reporting rates for vaccines administered in the state of Victoria from 2.6 reports per 100,000 in 2003 to 13.5 reports per 100,000 people vaccinated per annum in 2009 [21], while in an Italian study reporting rates for the seasonal influenza vaccine increased from 55 reports per 1,000,000 for the period 2009-2012 to approximately 140 reports per 1,000,000 doses in 2014 following the introduction of EPSS [8].

The surveillance carried out in this study enabled near real-time reporting of ADRs after vaccine exposure through the collection of data within a short time period at the start of the seasonal influenza vaccination. This was important to mitigate potential risks before the peak period of the influenza season. Spontaneous reporting was stimulated by HCPs informing vaccinees or vaccinees’ parents/legal guardians about the importance of reporting ADRs and was further facilitated using VCs and enabling the vaccinees or the parents/legal guardians to report the ADRs directly using the EDC system. The customised reporting form containing predefined ADRs and focused questions enabled the collection of accurate and complete information for case processing and follow-up by local PV.

In this EPSS, both younger and older age groups were under-represented. The study included individuals from private clinics, whereas influenza vaccines were distributed for free at public health clinics in Finland. Therefore, there may have been a lack of economic incentive for parents to bring their children or for older adults to attend private clinics to receive the vaccine. Although the enrollment was low in these age groups, considering the descriptive nature of the analyses, it is unlikely that this would have significantly impacted the findings of the EPSS.

It should also be noted that all cases were self-reported by vaccinees or their legal guardians and were not medically confirmed. However, all cases were non serious and it is unlikely that they would have triggered a medical visit; thus confirmation by an HCP would not have been obtained for most ADRs reported. As is inherent in passive surveillance, and to avoid bias in the collection of safety data, there was no possibility to follow-up with individuals at the site level. The interpretation of our findings should take into account the possibility of both under-reporting (where only a fraction of the total number of ADRs actually occurring after vaccination are reported) and differential reporting (where ADRs and ADRs with shorter time to onset after vaccination are more likely to be reported during the surveillance period than ADRs with longer time to onset).

In conclusion, the 2020/21 EPSS results did not suggest any clinically significant change in what is known or expected in terms of the safety of IIV4. The analysis of suspected ADRs did not identify any reporting pattern by type, frequency or severity, supporting the established acceptable safety profile of this vaccine. These findings will help to maintain public confidence in influenza vaccination.