The HIV Vaccine Chronicle
An independent continuation of hivpave.org · est. 2005
ENZHFRSWPT

Why it's so hard

SIX REASONS · GRADED

Not a mystery, not a conspiracy — a list. Each reason below is graded by evidence strength at our evidence page: Grade A (primary literature, verified), B (reviewed secondary), C (interpretation, labeled as such).

1It evolves faster than any virus we've vaccinated against

HIV mutates nearly every time it copies itself; a single infected person carries thousands of variants within weeks. Any antibody the immune system learns, the virus can escape. Compare: SARS-CoV-2 spiked global alarm with its first major variant two years in — HIV does that inside one patient in one month.

Grade A · Rambaut et al., Nat Rev Genet 2004; WHO COVID-19 variant timeline

2It integrates into your genome within days

HIV splices itself into the host's DNA early — often before symptoms, often before tests turn positive. A vaccine that arrives after exposure-week-one is chasing an entrenched enemy. There is no 'clearance' success story in 40 years of data.

Grade A · CDC; Chun et al., Nature 1997 (latent reservoir establishment)

3The good targets are hidden

HIV's envelope is spiked with sugars and shape-shifts on contact with immune cells; the rare patches antibodies can reach are concealed until the moment of infection. The strongest known antibodies — broadly neutralizing antibodies — take years of chronic infection to evolve, in the few people who make them.

Grade A · Kwong & Mascola reviews, Immunity 2018; IAVI protocol summaries

4No animal model predicts what works

Only humans and some non-human primates can be infected with immunodeficiency lentiviruses in ways that resemble human HIV. Monkey results have repeatedly failed to transfer: candidates that protected macaques (STEP, HVTN 702's preclinical record) did not protect people.

Grade B · Grant & Duggan reviews; HVTN 702 preclinical-to-clinical divergence

5Human proof is slow, expensive, and ethically demanding

Efficacy requires tens of thousands of uninfected volunteers followed for years, with full prevention counseling and PrEP access — as ethics requires. PrEP uptake in trial arms makes it harder to detect small vaccine effects. Each readout is a decade away from its design.

Grade A · HVTN ethics literature; HVTN 702 and AMP designs

6Money follows hope — and leaves in winters

Funding has surged after each flicker and frozen after each failure. The STEP winter (2007-2009) saw budgets cut and programs merged; the current quiet period follows 2020-2024's run of failures. Sustained funding for unglamorous Phase 1 immunology is the field's chronic shortage.

Grade C · NIH report language; AVAC investment tracking — interpretation, clearly labeled

Straight answers

FAQ
Why was COVID-19 vaccinated against in under a year but HIV never?
Three reasons: SARS-CoV-2 is a stable, slowly mutating target; its spike protein is exposed and easily trained against; and most infected people clear it, giving vaccine designers a natural blueprint. HIV does the opposite of all three — it mutates relentlessly, hides its targets, and once integrated, is never cleared naturally.
What would a successful HIV vaccine look like?
The leading strategy — germline targeting — tries to walk the immune system through a training sequence so it produces the rare 'broadly neutralizing antibodies' that some long-term survivors develop after years. Phase 1 results (IAVI G002, Science 2025) show the first steps working in humans; an efficacy trial is years away.
Did anyone ever get close?
RV144 (2009) remains the only efficacy success: 31.2% protection, fading within a year. It was enough to prove possibility — not enough to license. Its correlate hints still guide design.
The honest summary. A vaccine is not impossible — RV144 proved a vaccine can reduce acquisition — but nothing licensed is close. Meanwhile prevention tools that do exist have grown powerful enough that UNAIDS talks about ending the epidemic by 2030 without one. The two facts coexist: the wait continues, and the waiting has options.