Where Does That 2 Year UD Figure Come From?

[tednlou2]

Where does it come from?  I just read this Q&A from Dr. Gallant and he says there is no difference between 2 months and two years.  He does say that someone 2 years in is more likely to be taking their meds adherently and not missing doses.  Is that where the 2 year thing comes from-- someone more likely to have virologic failure in the beginning? 


Question

Hello and thank you Dr. Gallant, I've read a lot on your blog about undetectable. However, if someone recently becomes undetectable within 3-4 months are they more infectious than someone undectable for 2 years? Is that person still infectious? Or is the risk for transmission the same no matter how long or short you've become undetectable?

Answer

As far as we can tell, it’s the same.

The only caveat is that people whose viral loads have been undetectable for 2 years are likely to stay that way, because they’re clearly people who are on appropriate regimens and are taking their medications. 

We don’t always know that in the beginning. Someone who has just started ART may take it well for the first few months but then become less adherent. The viral load may become undetectable, but then rebound. When virologic failure occurs, it usually occurs early for that reason.

[AusShep]

There is also some data that viral load in semen can fall more slowly than blood levels.  I don't know if it's a meaningful amount though.

I think the more likely explanation for the 2 year number tossed around is the PARTNER study.  They required UD < 200 for at least one year, but as far as all of the gay couples, everyone enrolled had been UD for at least 2 years.  So for that we know it's very safe.

I've seen guys here and elsewhere try to jump on the no risk bandwagon a little too fast as well.  IMO, you need at least 6-12 months UD to show adherence and to be more certain that one reading wasn't a fluke and is a trend.  I'm not saying it's more risky if they are in fact UD, but as Dr G says, you really can't know long term adherence after one UD reading.

[eric48]

In some trials, patients come to UD, then, after a while they may have rebound (with or without resistance, hence failure).

That type of viral failure is different from the viral failure where patient can't reach UD. The later is obviously observed in the first 6-12 months. The former also appears more frequently in the first few months (year) of treatment.

Quite fortunatly, the risk of loss of control is not constant and diminishes with time (otherwise we would all of us remains under that same threat/risk at all times and treatment success would erode at 5-10 % a year)

Unexpected, 'unfair', loss of control after many years of control and adherence to the stategy used, is (fortunatly) very very rare. It happens... But very rarely.

I think that this is what was meant by lower risk in having unprotected sex with someone UD for a long time (say 2 y.) vs a shorter time (say 2 m.)

[Gruff]

There is also some data that viral load in semen can fall more slowly than blood levels.  I don't know if it's a meaningful amount though.

I remember when a study to this effect was released, and it was all the rage at HIV service Agency fundraisers and meetings: suddenly having an undetectable viral load was more dangerous than thought, because the virus might be much higher in other bodily fluids, particularly semen.  Everyone nodded in agreement, and we were all worried again....so I did some of my own research.

The study that indicated that viral loads in semen could be higher than in the blood and caused the firestorm and new paradigm was very specific: it found that VL in semen could be as much as 18% higher than in the blood.

That's an important number, because at the time of discovery, the VL tests were only sensitive enough to detect 40 copies/mL of blood; anything under 40 was considered "undetectable."

Today's tests are sensitive down to 20 copies/mL.  So, let's assume you have an undetectable blood Viral Load, but it is at the highest possible end of the undetectable spectrum: 20 copies.

And let's assume that, in accordance with the semen study, your viral load is the highest-ever identified variance in your semen - 18% higher in your semen than in your blood.

That would make the VL in your semen a whopping....23.6...which is still far below the 40 copies/mL that used to define undetectable.

All of which says to me (and I'm no doctor...) that even assuming that viral loads can be slightly different from blood to semen, if you have a consistent undetectable blood viral load and you're sticking with your meds...you really don't have to worry about  some dastardly super-collection of VL in your semen.