The larger placebo-controlled report did not confirm the efficacy suggested by earlier studies. That conclusion should remain prominent in a balanced thymosin-alpha-1 evidence record. Primary report
What was reported
Ninety-seven participants with HBV-DNA-positive and HBeAg-positive chronic hepatitis B entered a multicenter double-blind trial. Treatment lasted six months, with an additional six months of follow-up. The defined complete response occurred in seven of 49 thymosin-alpha-1 participants and two of 48 placebo participants; P=0.084. Including later sustained responses did not establish the efficacy reported in previous clinical studies. Primary report
Our appraisal: a numerical advantage is not a confirmed result
The active group can have more responders without the comparison establishing the claimed benefit. Readers need the uncertainty and planned decision rule, not only the direction of the difference. Describing a near-threshold P value as a successful trial would replace the reported conclusion with a promotional interpretation.
Response definitions should also travel with the numbers. Sustained negativity, a single negative test, and negativity reached after the planned study period are different outcomes. Combining them without explanation makes comparisons across papers unreliable.
An evidence timeline should include this report even if other studies are more favorable. Selectively linking only positive studies would hide an important test of the earlier hypothesis. Conversely, this result alone does not prove zero effect in every conceivable setting; it describes what this study failed to establish.
Before release, inspect the response hierarchy, follow-up completeness, assay methods, safety results, and whether subsequent corrections affect interpretation.
Frequently asked questions
Was there a placebo group?
Yes, with double-blind treatment assignment.
Did the report confirm earlier efficacy claims?
No. The authors explicitly said it did not.
Was the active-group response numerically higher?
Yes, but numerical direction alone is insufficient.
Does P=0.084 prove no possible effect?
No. It does not establish equivalence or a universal absence of effect.
Are delayed and planned-period responses identical?
No. Their assessment windows differ.
Should unfavorable trials remain in a library?
Yes. Excluding them would distort the evidence set.
What is the review scope?
The indexed primary abstract, with full-report appraisal pending.