Two infant doses of hepatitis B vaccine prime as well as three

Design

randomised clinical trial of two 2-dose schedules, with a separately recruited non-randomised 3-dose comparison group and a challenge dose at 3 years

Population

children recruited at 2 months (2-dose groups, homologous or heterologous) or at 18 months (3-dose group)

Primary outcome

seroprotection (anti-HBs 10 mIU/mL or more) after the primary series, and anamnestic response to a challenge dose 3 years later

Effect

seroprotection 91.7%, 90.9% and 94.2%; anamnestic response 97.2%, 95.5% and 92.7%

Children were recruited at two months of age and randomised to a two-dose hepatitis B schedule at 2 and 12 months, either homologous (the same hexavalent vaccine twice) or heterologous (hexavalent then a combined hepatitis A and B vaccine). A separate control group, recruited at 18 months, had already received three doses at 2, 4 and 18 months. All groups were given a challenge dose three years later to test immune memory.

One month after the primary series, seroprotection with anti-HBs of 10 mIU/mL or more was 91.7% after the heterologous two-dose schedule, 90.9% after the homologous two-dose schedule and 94.2% after three doses. Geometric mean titres were lower in both two-dose groups. After the challenge dose, an anamnestic response occurred in 97.2%, 95.5% and 92.7% respectively, and geometric mean titres were numerically higher in the heterologous two-dose group than in the three-dose group (9380.5 against 6900.6 mIU/mL, P=0.4). Local reactions were less frequent with the heterologous than the homologous two-dose schedule.

Immune memory, not the titre a month after priming, is what protects against hepatitis B years later, and on that measure the schedules were indistinguishable. That is the argument for fewer doses where supply, cost or attendance is the limiting factor.

Two limits matter before this is carried anywhere. The randomisation was between the two two-dose schedules; the three-dose group was recruited separately at 18 months and was not randomised against, so that comparison is not protected by randomisation. More importantly for Indian practice, none of these schedules includes a birth dose. India's programme gives hepatitis B at birth precisely to interrupt perinatal transmission, which is the dominant route here and the reason the birth dose is not negotiable. This trial says nothing about removing it.

• Do not read this as an argument for dropping the hepatitis B birth dose, which prevents perinatal transmission and was not studied
• Judge hepatitis B protection by immune memory rather than by a single post-primary antibody titre
• Remember that routine post-vaccination serology is not recommended for healthy infants; test the groups where it matters, such as infants of HBsAg-positive mothers
• Where a two-dose schedule is used, note that the heterologous version caused fewer local reactions
• Treat the three-dose comparison cautiously: that group was not randomised alongside the others

The statistics, in plain English

Seroprotection rates of 90.9% to 94.2% are close enough that with these group sizes the difference could easily be chance, and the lower geometric mean titres after two doses matter less than they look because titre falls over time in everyone; what predicts protection is whether a later exposure provokes a rapid response, which it did in over 92% of every group. The P value of 0.4 on the titre comparison after challenge confirms there is no detectable difference there. The weak point is structural rather than statistical: comparing a randomised pair against a separately recruited group leaves room for the groups to differ in ways nobody measured.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.