Design: Cochrane systematic review and meta-analysis with GRADE certainty assessment, searched to June 2026
Population: 10 randomised trials, 1,349 adults with chronic hepatitis B across six countries
Primary outcome: all-cause mortality, serious adverse events and health-related quality of life
Effect: all-cause mortality RR 0.53 (95% CI 0.29-0.96, very low certainty); serious adverse events RR 0.72 (0.53-0.99, low certainty); no effect on quality of life or histology

A Cochrane review searched to June 2026 for randomised trials of thymosin-alpha-1 in chronic hepatitis B, as monotherapy or added to interferon, lamivudine, entecavir or tenofovir. Ten trials met criteria, 1,349 participants, conducted in Bangladesh, China, Italy, Korea, Singapore and Taiwan and published between 1991 and 2018.

The point estimates look favourable. All-cause mortality risk ratio 0.53 (95% CI 0.29 to 0.96) across three trials and 907 participants; serious adverse events 0.72 (0.53 to 0.99); HBV-related mortality 0.53 (0.29 to 0.96); non-serious adverse events 0.47 (0.27 to 0.83). Quality of life and histological improvement showed no effect, and HBV-related morbidity was uncertain at 0.86 (0.54 to 1.40). The certainty grading is where the review lands: very low for every outcome except serious adverse events, which is low. The downgrades were for risk of bias, imprecision and heterogeneity, and sixteen further studies could not be assessed because reporting was incomplete and the authors received no reply to their enquiries.

Thymosin-alpha-1 is used in parts of Asia, including in Indian practice, largely on the strength of these same point estimates. The review's conclusion is not that it does not work; it is that after 35 years of trials nobody can say. In a disease with effective, cheap, well-evidenced nucleos(t)ide analogues, that is the argument against spending on it.

Do not add thymosin-alpha-1 to a nucleos(t)ide analogue outside a trial
Where a patient is already on it, review whether it is adding cost without evidence
Confirm the patient is on an appropriate first-line analogue - that is the intervention with the evidence
Note the age of the underlying trials: most predate current standard therapy
Report the cost of unproven add-ons explicitly when discussing long-term treatment

Why it matters

A drug in real use across Asia turns out to rest on evidence that a systematic appraisal grades as very low for every outcome a patient cares about.

The statistics, in plain English

A risk ratio of 0.53 for mortality with an upper bound of 0.96 would normally be persuasive, but certainty grading asks a different question from statistical significance: how much should you believe this estimate at all, given how the trials were run and reported? Very low certainty means the true effect could be substantially different in either direction. Three trials and 907 participants is also a thin base for a mortality claim, and the confidence interval nearly touches 1.0. The consistent zero heterogeneity across these outcomes reflects how few trials contributed, not agreement among many.