Novel Tuberculosis Vaccine Falls Short

An investigational tuberculosis (TB) vaccine didn’t top the standard bacillus Calmette-Guérin (BCG) vaccine at preventing infections in infants, a multicenter, phase III trial showed.

Among nearly 6,900 newborns in sub-Saharan Africa, QuantiFERON-TB Gold Plus (QFT) conversion, indicating Mycobacterium tuberculosis infection or exposure, occurred in 5.3% of those who received the VPM1002 vaccine compared with 4.3% of those who received the BCG vaccine, reported Sina Brückner, PhD, of Serum Life Science Europe in Hanover, Germany, and colleagues in Lancet Infectious Diseases.

For events meeting the first-case definition for QFT conversion in the VPM1002 group versus the BCG group, the HR was 1.23 (95% CI 0.99-1.53, P=0.057). Noninferiority required the upper bound of the confidence interval to be less than 1.25.

The trial also delivered uncertainty about VPM1002’s relative efficacy at preventing TB disease. Among 588 participants evaluated for suspected TB disease during follow-up, 3.2% of those who received VPM1002 and 1.7% of those in the BCG group developed microbiologically confirmed or unconfirmed TB.

With fewer QFT conversions overall than predicted (344 vs 632), the investigators terminated the trial early in October 2024.

“These findings highlight the methodological challenges of using QFT-based infection endpoints in infant vaccine trials and the importance of accounting for differential tuberculosis exposure in the design and analysis of future prevention of infection trials,” Brückner and colleagues wrote. “Given these findings, tuberculosis disease is a more appropriate endpoint in infant vaccine trials.”

BCG is the only licensed TB vaccine. While it protects young children against severe TB, it delivers waning and inconsistent protection against TB disease and transmission in adolescence and adulthood. VPM1002 is a recombinant BCG vaccine strain that showed similar immunogenicity to BCG, with fewer adverse injection-site events in early-phase trials.

Finding a more effective TB vaccine than BCG “is an urgent global health priority,” noted Helen McShane, PhD, of the University of Oxford in England, in an accompanying editorial. But better TB trial designs and endpoints are crucial to achieving that goal, she added.

Results from interferon-γ release assays such as QFT have been used widely as surrogate Mycobacterium tuberculosis infection markers in clinical trials. However, the gold-standard endpoint in TB trials is microbiologically confirmed disease, a challenging metric that requires large sample sizes and long follow-up periods.

VPM1002’s relative performance was worse when measured using microbiologically confirmed disease as the endpoint. That result shows that the trial’s surrogate infection endpoint of incident QFT conversion “might not be a sufficiently robust endpoint for tuberculosis vaccine efficacy trials,” McShane wrote.

“Overall, this trial teaches us that there are no obvious short cuts in tuberculosis vaccine efficacy trials,” she noted. “We need to continue to explore viable efficacy endpoints and trial design.”

There are relatively few early-stage TB vaccine candidates and only a handful of late-stage vaccine candidates, McShane pointed out. “As we encourage vaccine developers and funders to develop and support diverse and innovative vaccine candidates, we must in parallel work to design robust but feasible efficacy trials,” she added.

This double-blind, active-controlled trial randomized 6,940 newborns to vaccination with either VPM1002 or BCG, of whom 6,897 received shots. The trial ran from November 2020 to June 2022. Eligible infants were ages 0 to 14 days and had a birthweight of at least 2.3 kg. Mothers were at least 18 years old, had no active TB, and had no household contacts with TB in the 3 months prior to study enrollment.

Median post-vaccination follow-up was 35 months, 50.3% of infants were female, median enrollment age was 3 days, and 95.1% of infants were Black African.

The study’s primary efficacy endpoint was the occurrence of QFT conversion, indicating M. tuberculosis infection or exposure. Secondary endpoints included protective efficacy against microbiologically confirmed and unconfirmed TB disease.

Among 720 HIV-exposed, uninfected infants, 7% of those receiving VMP1002 had a QFT conversion compared with 8% of those receiving BCG. The 6,220 infants who were HIV-unexposed had a 5.1% QFT conversion rate with VPM1002 and a 3.9% rate with BCG.

Although more of the VPM1002 group had household TB exposures than the BCG group, adjustment for that increased exposure status showed similar QFT conversion rates between the two groups.

Safety profiles were similar for both vaccines. There was at least one solicited adverse event in 75% of those in the VPM1002 group versus 67% of those in the BCG group; nearly all events were mild. Rates of serious adverse events were also similar (10.7% vs 9.4%, respectively).