April 10th, 2012

A Skeptical Look at “Test and Treat”

Over in Journal Watch AIDS Clinical Care, Abbie Zuger has written a fascinating perspective on the recent enthusiasm for universal HIV treatment.

Her take? Let’s just say she doesn’t share the enthusiasm of public health officials and members of guidelines committees. Well, that’s a huge understatement.


This strategy, which calls for universal voluntary HIV testing and immediate antiretroviral therapy (ART) for all who test positive, has been widely hailed as a breakthrough in the fight against AIDS. I think it would be a big mistake.

After citing extensive data on how adherence to recommended medical regimens is notoriously poor, she offers her anecdotal experience:

In our busy urban clinic, some patients are as precise with their medications over years and decades as anyone might wish. The rest offer up a true symphony of reasons for nonadherence. Some are too depressed, distracted, drug-addled, disengaged, blasé, or suspicious to take their medications consistently. Some sell the drugs on the black market, and some share them with friends and partners. Some work out their own schedule of treatment interruptions, contrary to all medical advice.

I highly recommend reading the full piece — it is of course extremely well written (it is Abbie Zuger, after all), and provides a healthy counterpoint to the recommendation to treat everyone.

And she’s right that human beings don’t always do what their doctors recommend. It’s a long way from the exciting results of Study 052 to making “treatment for all” work in clinical practice. Just look at the huge number of people who know they have HIV but aren’t even in care.

But will the universal treatment strategy also engender a rise in HIV drug resistance?

Making treatment the default position means that we will just see more of all these behaviors, with a rapid rise of resistance to all drug classes … Sometimes I look at the multidrug-resistant gram-negatives in our intensive care unit and think ahead a few decades to the day when fearsome multidrug-resistant HIV strains begin to land in our clinics.

Here’s where we disagree. (For the record, I hate disagreeing with Abbie.)

I actually don’t think expanding treatment will significantly increase drug resistance, ironically for the exact reasons she cites in her piece — the same people who don’t take their medications and don’t show up for clinic visits usually don’t get resistance, or at least not much of it.

No meds = no selective pressure.

Just look at the people in HIV clinics today with horribly multi-drug resistant virus. They are predominantly the highly adherent, aggressively treated patients from the 1990s/early 2000s who received “serial monotherapy”, that inadvertent adding-on of a single active drug to a failing regimen.

Sometimes the serial monotherapy was out of clinical necessity — we had to do something when our patients were about to die of AIDS even if we didn’t have two active drugs — and sometimes it was the result of incomplete understanding HIV drug resistance. For example: did we really once think that d4T resistance was uncommon? Yikes.

But the good news is that many of these patients with multi-drug resistant HIV are now virologically suppressed, thanks to the great wave of drug development from 2006-2008 and their excellent adherence. See Steve Deeks discuss this phenomenon here.

And diagnosing a new case of high-level resistance to NRTIs, NNRTIs, and PIs — or even worse, to all six drug classes — has become incredibly rare. I strongly suspect it’s likely to stay that way, at least in settings with access to viral load monitoring, resistance testing, and the full range of antiretroviral drugs.

2 Responses to “A Skeptical Look at “Test and Treat””

  1. Jon says:

    Paul, it’s always informative to hear your take on things. I have a feeling that there won’t be a clear winner in this debate. Levels and patterns of adherence that influence the rate of resistance accumulation are likely to differ from region to region and within certain subpopulations. As data from 2nd-line failures in developing countries show, rates of drug resistance may rise in one place (India), but not another (South Africa). In the end, it’s hard to get too excited about a universal treatment strategy when the economics of it seem so unfavorable for patients in the developing world, even for those who we can all agree need to be treated.

    • Paul Sax says:

      Jon, thanks for your comment.

      I agree that poor adherence can lead to resistance — just not the kind of truly untreatable viruses we saw from the serial monotherapy days. Use of boosted PIs and a better understanding of resistance has made this so much less common.

      If you define highly-resistant virus as follows:

      1. pan-NNRTI resistance
      2. multiple TAMS or K65R + 184V
      3. Five or more PI mutations

      How many newly-diagnosed HIV+ patients in the last 10 years developed this degree of resistance due to poor adherence? I’m sure they exist, but I’ve never seen one.


HIV Information: Author Paul Sax, M.D.

Paul E. Sax, MD

Contributing Editor

NEJM Journal Watch
Infectious Diseases

Biography | Disclosures | Summaries

Learn more about HIV and ID Observations.