HIV Vaccine Fails - Another Clinical Trial on Horizon?

A new HIV vaccine, supported by WHO and UNAIDS, being pushed by Merck hit a roadblock this week.  The STEP program, which has participants in North America, Australia, South America, and the Caribbean, has, "...cease[d] administering the investigational vaccine but continue scheduled follow-up visits with all volunteers until the data can be more thoroughly evaluated and a course of action is developed." [source]

The vaccine, Ad5 HIV-1 gag/pol/nef, has been safely used in animals and preliminary human subjects, but its efficacy in large populations has yet to be demonstrated.  The STEP program began canvassing participants in 2004 as part of the NIH/Merck funded program, and reached 3000 participants.

The vaccine works as follows: use a 'vector' (i.e. a genetically engineered of virus or bacteria -  a bacteria in this case) to expose your body to the products (i.e. proteins that exist on the outside of HIV viruses) of certain genes associated with AIDS.  In essence, it primes your immune system for any future HIV infections.

High risk individuals were selected, including gay men and female sex trade workers.  Of 741 volunteers who received one dose of the vaccine, 24 became HIV+.  However, in the control group receiving the placebo, 21 of 762 also became infected.  A second dose had no statistically significant effect on infection rates.

After a bit of sleuthing on the NIH website, I found a second study being performed by Merck, "This study will test the safety and immunogenicity of an investigational HIV vaccine. Immunogenicity will be measured by evaluating the immune response to several different dose levels." [source]  And the kicker, part of the inclusion criteria: Low risk of acquiring HIV infection.

How effective will this be?  If over the 3 year period, less than 3% of the high risk individuals became infected, how many individuals must they enroll to have any meaningful, statistically significant results?  Assuming high risk individuals are 10 time more likely to become infected, that's 30,000 people just to have two dozen infected.  I'm not a medical professional, and I have never conducted any such studies, but this seems ludicrous, does it not?

With Merck already in the news concerning the safety and efficacy of their HPV vaccine, Gardasil, this is surely to irk their shareholders.  A common concern associated with Gardasil is how vaccinated individuals will change their sexual habits?  My question for you is, could a successful HIV vaccine do the same?  Will people become more premiscuous?

If so, it looks like Merck will have a hay-day developing STD-related vaccines.