Summary Basis of Decision for Ervebo

Clinical Pharmacology

Ervebo consists of a live, attenuated recombinant vesicular stomatitis virus-based vector expressing the envelope glycoprotein gene of Zaire Ebola virus (ZEBOV). Immunization of subjects with the vaccine results in an immune response and protection from Zaire Ebola Virus Disease (EVD). The relative contributions of innate, humoral, and cell-mediated immunity to protection from ZEBOV are unknown.

Clinical pharmacology studies with Ervebo were not conducted and are typically not required for vaccines.

Immunogenicity (see Immunogenicity section) was assessed as part of the clinical efficacy evaluation of Ervebo.

For further details, please refer to the Ervebo Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Efficacy

The authorization of Ervebo was supported by efficacy, immunogenicity, and safety data from thirteen clinical trials which encompassed eight Phase I trials, one Phase II trial, one Phase II/III trial, and three Phase III trials that were designed and conducted independently by a number of different sponsors. Of the thirteen trials, V920-010 was considered the pivotal trial for the demonstration of the clinical efficacy of Ervebo.

Clinical Efficacy

V920-010, conducted in Guinea during the 2014-2016 EVD outbreak, was a field-based, Phase III, open-label, cluster-randomized, controlled, ring vaccination trial designed to evaluate the efficacy, effectiveness, and safety of one dose of Ervebo in the prevention of EVD when implemented as ring vaccination. Ring vaccination is a strategy to vaccinate individuals at highest risk of infection due to their social network connection to a patient with laboratory-confirmed EVD. Individuals who met the criteria for a defined ring were randomized (1:1) at the ring level into either the immediate or delayed (21 days) vaccination arms, then vaccinated with Ervebo. Subjects were followed up to 84 days post vaccination for EVD.

The primary objective of V920-010 was to assess vaccine efficacy against laboratory-confirmed EVD between 10 and 31 days following randomization by performing a clinical trial comparing immediate versus (vs.) delayed ring vaccination. The final primary efficacy analysis compared all vaccinated subjects in the immediate arm (2,108 subjects in 51 rings) to all eligible subjects who consented on Day 0 in the delayed arm (1,429 subjects in 46 rings). The vaccine efficacy with Ervebo was 100% (95% confidence interval: 63.5% to 100%), with 10 cases (4 rings) of confirmed EVD observed in subjects in the delayed vaccination arm vs. 0 cases in the immediate vaccination arm. Cases of EVD occurring less than 10 days following randomization were excluded from the final primary efficacy analysis to maintain the comparability of the populations with respect to exposure to the index case. Cases of EVD occurring more than 31 days following randomization were also excluded from the final primary efficacy analysis to account for vaccination in the delayed arm on Day 21. The true vaccine efficacy and the duration of protection are unknown given the inherent biases with such studies as well as the challenges of conducting a ring vaccination study in the setting of an outbreak.

Given the data evaluated, the vaccine is expected to be efficacious to a great degree and is expected to confer protection for a disease that is highly contagious with a high mortality rate. Taking into account the high mortality rate of the disease, individuals at risk should continue to protect themselves from exposure to ZEBOV following vaccination.

Immunogenicity

The clinical development program consisted of twelve clinical trials that provided immunogenicity data. Eight Phase I trials (V920-001 through V920-008) were conducted in North America (United States and Canada), Europe (Switzerland and Germany), and non-epidemic regions of Africa (Gabon and Kenya). These trials evaluated dose ranges from 3 x 10³ to 1 x 10⁸ plaque forming units (pfu). Antibody titers, measurable by glycoprotein-enzyme-linked immunosorbent assay (GP-ELISA) at Day 14 in most recipients of Ervebo, had the greatest fold rise at Day 28 post vaccination for most of the dosing regimens and declined slightly but plateaued thereafter through Day 180 or through Day 360/365 (depending on the study). Plaque reduction neutralization test (PRNT) antibody titers were elevated as early as Day 14 and peaked at Day 28. The results revealed that the greatest fold rise occurred 1 to 2 months post vaccination and were sustained through Day 360. Immunogenicity data from these studies supported the selection of the 2 x 10⁷ pfu dose that was used in the pivotal trials.

Additionally, four Phase II/III trials (V920-009, V920-011, V920-012, and V920-018) provided immunogenicity data from subjects who received Ervebo as a single nominal dose of 2 x 10⁷ pfu (all four trials) or 1 x 10⁸ pfu (V920-012 only). Three of the trials were conducted in African countries (Liberia, Sierra Leone, and Guinea) and the other was conducted in Canada, the United States, and Spain (V920-012).

Analyses for an integrated summary of immunogenicity (ISI) were performed to provide a broad evaluation of the immunogenicity data collected in the Ervebo Phase II/III clinical program. The ISI analyses include the calculation of geometric mean titers (GMTs) at baseline, Month 1, Month 6, and Month 12, and the percentages of subjects with a seroresponse defined as a ≥2–fold increase from baseline and ≥200 ELISA units [EU]/mL for the GP-ELISA and a ≥4–fold increase from baseline for both the GP–ELISA and PRNT. Seroconversion was defined as a post-vaccination ELISA titre of at least 200 EU/mL that was also a ≥4–fold increase over baseline in ZEBOV immunoglobulin G (IgG). The values for the GP–ELISA and PRNT GMTs were higher than baseline at the Month 1, Month 6, and Month 12 post-vaccination time points for the vaccination groups.

In V920-012, the durability of immunogenicity was evaluated through Month 24 post vaccination. The values for GP–ELISA and PRNT GMTs for the Ervebo groups were higher than baseline through Month 24 post vaccination. Whereas GP–ELISA

GMTs peaked at Day 28 post vaccination with an approximate 25% decline observed at Month 24 post vaccination, PRNT GMTs peaked at Month 18 post vaccination and no appreciable decline was observed thereafter. The GMTs for the placebo group were less than the lower limit of quantification of the GP–ELISA and the PRNT at all time points.

The GP–ELISA geometric mean fold rises (GMFRs) demonstrated that GP–ELISA titers for the Ervebo groups increased from baseline by approximately 64–fold at Day 28 post vaccination and remained elevated approximately 48–fold at Month 24 post vaccination. The PRNT GMFRs demonstrated that PRNT titers for the Ervebo groups increased from baseline by approximately 11–fold at Day 28 post vaccination, peaked by approximately 16–fold at Month 18 post vaccination, and remained elevated approximately 15–fold at Month 24 post vaccination.

Immunogenicity specimens from subjects in the African trials underwent gamma irradiation to allow for processing in a biosafety level 2 facility. A study conducted by the sponsor showed that gamma irradiation results in an approximate 20% elevation in measured antibody response for negative clinical samples and an approximate 20% reduction in post-vaccination antibody response in the GP–ELISA. The effect of gamma irradiation was not consistently observed for the PRNT. The sponsor reasons that this is why GP–ELISA and PRNT GMTs were generally lower for subjects in the African trials compared with those in V920-012. For the GP–ELISA, 95.2% and 98.1% of subjects had a seroresponse in the African and ex-African subpopulations, respectively, at any time post vaccination. For the PRNT, 57.4% and 83.3% of subjects had a seroresponse in the African and ex-African subpopulations, respectively, at any time post vaccination. From a proof-of-principle perspective (as the data were not correlated to individual efficacy data), the robust increases of the GP–ELISA binding and PRNT neutralizing antibodies across the main and supportive studies suggest that the efficacy findings are explained by the immunogenicity responses. The sponsor concluded that although the GP–ELISA may be a correlate of protection as a result of vaccination, it is not prudent to define a threshold value for protection at this time given a number of considerations and limitations associated with the analyses conducted. The immunogenicity data are considered informative but in the absence of an immunocorrelate of protection, the clinical relevance to efficacy is unknown.

Key efficacy issues and uncertainties

Important limitations of the data at this time include the lack of information on the long-term efficacy and safety of the vaccine. Namely, the duration of protection, the level of protection, and the protection against other Ebola virus strains or viral variants of the Zaire strain are unknown. There are also limited data on efficacy in special populations (e.g., pediatrics and individuals who are significantly immunocompromised, pregnant/lactating, or with severe comorbidities). Additionally, the need for and timing of a booster dose is unknown, and antibodies against the viral vector were not measured, leaving a lack of understanding of the kinetics of antibodies against the viral vector and their potential impact on the immune response to repeated vaccinations. Furthermore, effectiveness of the vaccine when administered concurrently with antiviral medication, immunoglobulin, and/or blood or plasma transfusions is unknown. Given the high mortality rate of the disease, individuals at risk should continue to protect themselves from exposure to ZEBOV following vaccination. These limitations are considered adequately managed through labelling associated with authorization, and the Risk Management Plan for Ervebo.

Indication

The New Drug Submission for Ervebo was filed by the sponsor with the following indication, which Health Canada subsequently approved:

Ervebo (Ebola Zaire Vaccine [rVSVΔG-ZEBOV-GP, live]) is indicated for active immunization of individuals 18 years of age or older to protect against Ebola Virus Disease (EVD) caused by Zaire Ebola virus.

For more information, refer to the Ervebo Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Safety

The clinical safety evaluation of Ervebo focused on potential reactogenicity as assessed by solicited and unsolicited injection site and systemic adverse events (AEs) after vaccination with Ervebo. The review of vaccine safety data provided in the submission was complex due to a heterogeneous group of thirteen blinded and open-label trials (V920-001 to V920-012, and V920-018) conducted in Africa, North America, and Europe. These trials were conducted among a diverse set of study populations using different methods of measuring solicited and unsolicited AEs, and were initiated by different organizations, including the World Health Organization and the sponsor. As such, the sponsor was unable to pool safety data from across the different trials (16,765 adult subjects) but used a study-by-study strategy for evaluation of safety data.

The Phase III blinded V920-012 trial, mostly conducted in the United States with some subjects from Canada and Spain, was identified as the main safety study by Health Canada. V920-012 was supported by two earlier trials; the Phase I/II blinded V920-005 trial (Switzerland) and the Phase I blinded V920-004 trial (United States). V920-005 and V920-004 were considered key safety studies that confirmed AEs of special interest (AESI) clustering around delayed post-vaccination arthritis syndrome and delayed post-vaccination rash syndrome. The main efficacy study, V920-010 (Guinea), was excluded from the safety analysis due to a high baseline morbidity that would not be expected in a Canadian population (e.g., malaria, EVD, etc.), as well as being an open-label design without a placebo control group for comparison.

Solicited local (injection-site) AEs were collected in V920-012, V920-005, and V920-004, with the majority of Ervebo recipients experiencing local reactions. In V920-012, 71.8% of 1,051 Ervebo recipients vs. 14.3% of 133 placebo recipients experienced at least one injection site AE during the first 5 days post vaccination, with the most frequently reported being injection site pain (69.5% vs. 12.8%), injection site swelling (16.5% vs. 3.0%), and injection site erythema (11.9% vs. 1.5%), all with a median duration of 4 days. In the first 14 days post vaccination, solicited local treatment-emergent adverse reactions (TEAEs) were experienced by 52.0% of 102 Ervebo recipients vs. 23.1% of 13 placebo recipients in V920-005 and by 43.1% of 418 Ervebo recipients vs. 10.6% of 94 placebo recipients in V920-004.

Solicited systemic AEs were not specifically collected in V920-012 but were collected in V920-005 and V920-004 within 14 days post vaccination. In V920-012, systemic reactogenicity was captured in the data for unsolicited AEs within 42 days post vaccination. Common solicited systemic AEs included headache, fatigue, myalgia, chills, arthralgia, fever (subjective and objective), nausea, loss of appetite, and diarrhea. Overall, V920-005 reported more systemic reactogenicity, occurring in 91.2% of Ervebo recipients vs. 69.2% of placebo recipients compared with 59.6% vs. 41.5% of recipients, respectively, in V920-004.

In V920-012, the majority (82.8%) of 1,051 Ervebo recipients compared with 44.4% of 133 placebo recipients experienced one or more common AE from Day 1 to Day 42 post vaccination. The most common System Organ Classes with unsolicited AEs were (Ervebo vs. placebo) “general disorders and administrative site conditions” (78% vs. 18.8%), “musculoskeletal and connective tissue disorder” (29.9% vs. 6.8%), “nervous system disorders” (26.1% vs. 14.3%), “gastrointestinal disorders” (12.0% vs. 6.0%), and “infections and infestations” (7.5% vs. 9.0%). Adverse events, which occurred in 1% or more of subjects, included: diarrhea, nausea, vomiting, asthenia, chills, influenza-like illness, peripheral swelling, malaise, pyrexia, nasopharyngitis, anorexia, arthritis, muscle spasms, stiffness, lethargy, paraesthesia, rhinorrhea, blister, and rash.

During the implementation of the Phase I/II V920-005 blinded trial, investigators initiated a safety hold in response to the appearance of delayed-onset arthritis and rash syndromes (or both) among vaccine recipients during the first month of enrollment. Overall, 23.5% of 102 Ervebo recipients experienced arthritis TEAEs vs. 0% of placebo recipients, with a median time of onset of 10.5 days post vaccination and a median duration of 34.5 days. All but 2 subjects had no prior history of arthritis. Arthritis resolved in 50% of subjects by 1 year post vaccination, and in another 25% of subjects by 2 years post vaccination.

In V920-004, 4.5% of 418 Ervebo recipients experienced “temporally associated post-vaccination arthritis” compared with 3.2% of 94 placebo recipients. All vaccine-related cases resolved by the end of the study, with the exception for 1 subject with recurrent arthritis. The median time to onset was 12 days in the Ervebo group vs. 15 days in the placebo group, with a median duration of 8 days vs. 47 days, respectively. In the follow-up Phase III V920-012 trial, 3.5% of 1,051 Ervebo recipients and 0% of 133 placebo recipients experienced an “arthritis” AE (composite term for arthritis, monoarthritis, polyarthritis, osteoarthritis, joint swelling, or joint effusion). The median onset was 10 to 11 days post vaccination with a median duration of 5 to 6 days and no dose relationship (regular vs. high dose). In V920-005, the joint fluid samples of 3 subjects were positive for vaccine virus as determined by reverse transcription-polymerase chain reaction (RT-PCR) on Day 14 or Day 28, whereas 1 of 3 subjects tested positive in V920-012, which could suggest a vaccine aetiology.

Similarly, in V920-005, “skin- and mucosal-related treatment-emergent adverse events”, or skin- and mucosal-related TEAEs, were noted in 29.4% of 102 Ervebo recipients vs. 31% of 13 placebo recipients, with a median onset of 8 days post vaccination for Ervebo recipients vs. 8.5 days post vaccination for placebo recipients, and lasting 10.5 days vs. 11 days, respectively. However, skin- and mucosal-related TEAEs in the Ervebo group were different, and included the following conditions not experienced in the placebo group: maculopapular rash, skin lesions, cutaneous vasculitis, as well as papule, generalized rash, macular rash, popular rash, and vesicular rash. Five subjects who were vaccinated had a total of eight skin samples that were positive for virus vector, suggesting a vaccine aetiology. Overall, 10.8% of Ervebo recipients had concurrent arthritis and skin-related TEAEs, compared with 0% of placebo recipients.

In V920-004, 5.7% of 418 Ervebo recipients vs. 3.2% of 94 placebo recipients experienced “temporally associated post-vaccination dermatitis” (e.g., dermatitis, petechiae, macular rash, papular rash, vesicular rash, skin lesion, skin mass, or skin ulcer) with an onset within 56 days of vaccination. The overall median time to onset was 9 days for Ervebo recipients and 5 days for placebo recipients, with a median duration of 7 days and 54 days, respectively. A total of 1.2% of Ervebo recipients experienced concurrent post-vaccination arthritis and dermatitis with no similar cases within the placebo group.

In the main safety study, V920-012, 3.8% of 1,051 Ervebo recipients and 1.5% of 133 placebo recipients experienced “rash” AEs (composite term for petechiae, purpura, rash, rash generalized, rash macular, rash popular, and rash vesicular but not dermatitis) from Day 1 to Day 42 post vaccination. The median time to onset was 7.5 to 10.5 days for Ervebo recipients and 3.5 days for placebo recipients, with a median duration 6 to 18 days and 14 days, respectively. Additionally, a total of 1.5% of 1,051 Ervebo recipients experienced vesicular (blistering) lesion from Day 1 to Day 42 post vaccination vs. 0% of placebo recipients. Vaccine virus was detected in skin biopsies by RT-PCR in 1 of 6 subjects, suggesting a vaccine aetiology.

In V920-012, V920-005, and V920-004, no serious adverse events or deaths were considered to be vaccine related. Of the 408 Ervebo recipients with PCR data in V920-004, 20% had detectable vesicular stomatitis virus viremia at one or more time points post vaccination (vs. 0% in placebo). Viremia was most prevalent at Day 1 and Day 2 post vaccination, with 13% of Ervebo recipients testing positive for vaccine virus at each of these time points. The number of Ervebo recipients who were viremic dropped to 2% by Day 3 and to 0.5% by Day 4, with viremia resolving in all subjects by Day 28. Higher doses of Ervebo were associated with higher percentages of subjects experiencing vaccine viremia. In V920-005, 44% of Ervebo recipients had detectable vaccine viremia at Day 1 post vaccination (vs. 0% of placebo recipients). At Day 7, only 2% of the 92 subjects with available PCR data had detectable vaccine viremia. No vaccine viremia was detectable on Day 14 or Day 28. This suggests that the detection of administered vaccine product in the blood shortly after administration was short-lived. In V920-004, vaccine virus shedding/excretion in vaccine viremic subjects was infrequent as only 1 Ervebo recipient had detectable vaccine virus in saliva and urine within the first week. Similarly, in V920-005, it was planned that all subject samples were to be tested for vaccine virus ribonucleic acid shedding in saliva and urine following vaccination, but testing was stopped after the first safety analysis (i.e., after the first 10 subjects were tested) when it was noted that few samples from subjects were yielding positive results. Vaccine viral shedding was not assessed in the Phase III V920-012 trial. Overall, this suggests that the risk of transmission of vaccine virus by blood, saliva, or urine is low and short-lived. To avoid the theoretical risk of vaccine virus transmission, the Ervebo product monograph includes a precaution for vaccine recipients to avoid close contact with, and exposure of, high-risk individuals to blood and bodily fluids for at least 6 weeks following vaccination.

Key safety issues and uncertainties

Important limitations of the data at this time include the lack of information on the long-term efficacy and safety of Ervebo. Ervebo has been extensively used in field trials and during ZEBOV outbreaks subsequent to the trial data submitted within this submission. Fewer than the 3,000 vaccine exposures from the Phase III V920-012 blinded trial and the supportive Phase I blinded trials (V920-004 and V920-005) were used as the appropriate safety analysis dataset for the Canadian Product Monograph. However, this was considered adequate for the following reasons: participants were representative of the Canadian population, rates of AEs in vaccine recipients were properly controlled by a concurrent placebo group, and AESIs of arthritis and rash were common enough to sufficiently characterize the potential risk. Whether the post-marketing rates of AESIs will be higher, lower, or similar to trial rates is unknown. Similarly, solicited information from the smaller Phase I studies suggests that the risk of transmission of vaccine virus in blood, saliva, and urine appears to be low and short-lived. Thus, AESIs and transmission risk should continue to be monitored.

Appropriate warnings and precautions are in place in the approved Ervebo Product Monograph to address the identified safety concerns.

For more information, refer to the Ervebo Product Monograph, approved by Health Canada and available through the Drug Product Database.