November 13th, 2016

Poll: Should We Be Starting HIV “Elite Controllers” On Antiretroviral Therapy?

(Note to readers: This will have nothing to do with the electionfor obvious reasons. Yes, there’s a poll at the end, but your political views will not matter one bit. There, I feel better.)

Just received this email from a longstanding leader in HIV care and research:

I would like to ask for you a favor. I am preparing an elite controllers talk for the next GESIDA meeting that will take place in Madrid on December 2nd 2016 and I am doing a very short survey to international experts [thanks very much, Jose!]. I would appreciate very much if you can answers these 2 questions:

1)      Would you treat with ART elite controllers (patients with CD4>500 and longstanding HIV VL<50 copies/mL)?

–          Yes

–          No

2)      Please, explain your answer in 1-2 short sentences

Thanks again for considering my proposal and for your time.

Warm regards,

Jose Miro

This is a tough question, and if you’re wondering why, let me provide a bit of background for those who don’t do ID/HIV medicine regularly.

22065301079_237858c25a_m“HIV controllers” are those rare individuals who have HIV infection, but their immune systems somehow “controls” viral replication without the need for antiretroviral medications. How common is it? Estimates vary, but it’s certainly no more than 1% of those with HIV. (Here’s one representative study from France.) Rarity notwithstanding, every experienced HIV/ID clinician has seen at least one of these patients, and those of us doing this for a while have several, most of whom we’ve been following for years.

At the extreme end of the controller phenotype are people sometimes termed “elite” controllers, meaning we can’t detect HIV in their blood using even our current highly sensitive viral load assays — their results always come back “Target not detected” (in other words, “undetectable”), and their CD4 cell counts remain normal.

Not surprisingly, these elite controllers are the source of immense interest to researchers, who have long believed that these patients’ immunologic response to HIV — a usually progressive, lethal infection — might hold the clue for immunotherapy, or an effective vaccine, or both. Not only that, but their “elite” status allows them to board the plane first even when traveling coach, and grants them free WiFi and an automatic room upgrade when checking into fine hotels.

(Just wanted to see if you were listening.)

Back to the science:  But all is not completely normal in these lucky HIV controllers. Some have elevated levels of immune activation and inflammation — perhaps a marker that their immune systems are working overtime to control HIV. Might this insalubrious inflammatory milieu cause lymphoid fibrosis? Or increase the risk of cardiovascular disease? One provocative study showed that HIV controllers were admitted to the hospital more often than HIV infected patients who were on antiretroviral therapy.

Ultimately, this raises the question Jose posed in his email — should we start HIV controllers on antiretroviral therapy? Ten years ago, the answer was easy — NO — and we told them this repeatedly. Your immune system is doing what our antiretrovirals set out to do, which is control viral replication. Not only that, it’s doing so without side effects. You’re lucky — you don’t need treatment.

But times most definitely have changed. In addition to the above data on immune activation and inflammation, HIV treatment is now so much safer. Plus, we have the results of the START study and HPTN 052, showing that HIV treatment is beneficial even to those with CD4 > 500, and that suppressive therapy makes people with HIV essentially non-contagious. As a result, all guidelines now explicitly state that HIV treatment is indicated for all people with HIV. Doesn’t that include HIV controllers?

The reason this isn’t so clear comes down to three remaining issues, all of them significant. Two are uncertainties from clinical science, the third an emotional one from our patients.

  1. For elite controllers, there’s no proof of clinical benefit or reduction in transmission risk. The START study showed that those with CD4 > 500 benefited from starting therapy rather than waiting until the CD4 fell to < 350. However, the median HIV RNA of study participants was 13,000, and while it’s not in the primary paper, only a small fraction of participants could have had HIV RNA < 500. Imagine how few (if any?) were “elite controllers”. Similarly, would HPTN 052 demonstrate a reduction in the risk of HIV transmission even among those who started the study with undetectable viral loads? Doubtful.
  2. Patients who start ART often have some residual abnormalities in immune activation and inflammation — will the same hold true for elite controllers? In patients on long-term suppressive ART, abnormalities in inflammatory and immunologic markers persist in many patients. Moreover, the number of elite controllers who have been treated with ART and studied intensively is very small, limiting our ability to make firm conclusions. This study looked at 4 such patients treated with TDF/FTC and raltegravir, showing a non-significant trend downward over 24 weeks of therapy. More such patients are being studied in ACTG A5308, which is an ongoing prospective clinical study of ART in HIV controllers (I’m on the study team) — but until we have those results, the benefits of ART even on surrogate markers for this population are unknown.
  3. Most HIV controllers don’t want to start treatment. We’ve been telling some of them for years — even decades — that they are special, and that they don’t need treatment since their immune systems are doing all the work. Not surprisingly, after hearing this for so long, it’s hard for them to hear otherwise, in particular since they still feel fine. It’s difficult to convey the benefits of HIV therapy by using the concepts of “T-cell activation” or (even worse), “%CD38+HLA-DR+ CD4+ T cells”.
    Our experience with A5308 when we approached HIV controllers about participating was fascinating — even those who have been enormously generous to the research community, providing blood and tissue samples to numerous investigators for years, simply didn’t want to take HIV meds. One even told me he would prefer something more experimental than ART, such as an HIV vaccine or novel combination of cytokines.

OK, that’s enough — much more than the 1-2 sentences requested by Jose, sorry about that!

But it’s time to decide, please take the poll below (all are encouraged to participate). And this time (ahem), there’s no wrong answer.

Should HIV "elite controllers" start antiretroviral therapy?

View Results

16 Responses to “Poll: Should We Be Starting HIV “Elite Controllers” On Antiretroviral Therapy?”

  1. Pablo Ryan says:

    Elite controllers have served as a natural model for the ‘functional’ cure of HIV. It will be interesting to ask this same question, once we have functional cure.

  2. Robert Cannon says:

    Treat those who volunteer for treatment in a study and follow both groups with the same end points. We will never know if treatment has a net positive effect until we study its effects.

  3. Ryo Miyakawa says:

    Would the answer to this question depend on sexual behavior of each patient?
    1) seropositivity of the partner
    One can argue for starting treatment to eliminate possibility of transmission (though no proven benefit for this elite controllers?) if the partner is seronegative. You can argue for starting treatment if the partner is seropositive, because the different genotype of HIV may not be well controlled with immunity. In such patient, I would assume treatment regimen, rather than prophylaxis regimen is more reasonable due to fear of resistance.

    2) sexual activity of the patient
    If the patient is engaged in unprotected sex (despite physicians strongly encouraging to use protection), would it be more reasonable to start treatment? But would such patient be agreeable to start treatment?

    • Stacy Lane says:


      I treat elite controllers that don’t consistently use condoms as we know there can be discordance in viral load in serum vs semen/vaginal fluid, and if your viral load does bump up no light bulb will appear over your head and it will likely be a few months before we catch it.

  4. Carlos Del Río says:

    At this point I think if an elite controller is considered for treatment it should be within the context of a clinical trial or some other research program

  5. Kevin Lapin says:

    Paul, I have been trying to find a way to send you an inquiry but can’t seem to find a link or your email.

    I have a patient who started a relationship with an HIV+ partner. He previously tested negative using the HIV 4th gen test. He now repeatedly tests positive to the HIV ag portion of the test but negative to the ab- and negative viral loads. No symptoms of acute infection of course. Puzzling. I have put him on PrEP and counseled about the need for condoms, of course. But the converstion to ag+ is still troubling.

    Any ideas?

  6. Joel Gallant says:

    If it were me, I’d want to be treated, and when my patients ask me that, that’s what I tell them. But I don’t feel the data are strong enough to push a reluctant patient toward treatment.

    • Jack says:

      I test every six months and rarely have CD4 counts of less than 1,200. I was diagnosed HIV-Positive 31-years ago in London and continue to show undetectable viral loads in the absence of any treatment. I avoid unprotected sex, whether active or passive, and have been fortunate enough to remain perfectly healthy throughout. Is there any good reason why I should want to be on medication? I can’t think of one.

      Also, l would be grateful for information on any UK-based clinical trial or other research program that considers elite controllers. I’d love to take part.

  7. Liz Jenny says:

    What a great subject. I’m writing a chapter for a primary care text and wondered if I should just omit this admittedly not infrequent practical conundrum. I feel much better knowing I’m not alone in my confusion. I guess the million dollar question is what surrogate markers could we use for patients with high CD4 and undetectable VL–due to ART or elitely controlled. Clearly there is a lot of residual morbidity despite favorable numbers but we can neither monitor it with the existing surrogate markets nor could we clearly intervene even if we did have a practical measure of immune activation. As for treat or not treat the elite, I try to figure out what the patient wants and then recommend that option. Question is easier if CD4 is low. Can see if it rises on ART though I don’t think it does.

  8. Federico Tomasella MD says:

    As Dr. Sax thoroughly stated, we don´t have enough evidence that elite controllers would benefit with ART. Nevertheless, HIV is an independent risk factor for CVD, whether on ART or not. Also the immune activation and proinflammatory state persists whether patients take ART or not; so the only issue that disturbs me is the higher hospital admission rates of ECs.
    Anyways, as today, I will vote for “Primum non nocere”.
    So, Dr. Sax, can we know what´s your point of view on this matter? (Ahem, to be more accurate, what did you answered to Dr. Miro? If of course it´s not top secret!) And another question: how frequently do you test these patients?
    Thanks in advance. It´s always a pleasure to read your posts!

  9. Osman Ebrahim says:

    I have a cohort of 20 Elite controllers. Some I have followed for 15 years!. They are clinically well. Non have developed any complications of inflammation. We are looking at functional cure. Treating such patients would delay the understanding of the mechanisms of virological control without drugs.

  10. Beatriz Mothe says:

    Dear Dr Sax and Miró
    What a great question! I think we need to explain to our few elite controllers patients the state of data today, as exactly you discuss on 1-2-3 points and let patients decide with us. (I’m more on the No…) We could also argue that multiple factors may intervene on elite control, and not the same factors are probably relevant for each patient. In some, a beneficial HLA allele is probably associated to an extremely efficient CD8 T cell mediated response. This Tcell response may exhaust and decrease over time and drive a loss in HIV control, which could be prevented by cART initiation.. But, in others, infection with replicative-defective viruses may be the reason for that elite control and we do not know what cART would imply. Do we now the rate of CD4 recovery after cART suppression on elites? I agree that there’s not enough data that shows how by reducing a portion of the %CD38+HLA-DR+CD4+ T cell with cART we could be preventing future hospital admissions or HIV-related morbidity; but hopefully prospective controlled trials will help on this… If we keep studying better these patients, we may be able to discern which patients would be at more risk of progressing or developing HIV-related comorbidities and could benefit from cART. In any case, I don’t think decision on initiating cART in elite controller individuals needs to be taken at HIV diagnosis (in contrast with any other HIV+) and better discuss it with patients over time as we monitor them. Also the consideration of behavioural risk assessment is important, as elite controllers can be superinfected with other virus strains and may no longer control it, so in these cases cART would also prevent from this (PrEP-like….)
    Looking forward seeing final results from poll and presentation at Gesida!

  11. David K says:

    Anecdotally, my elite controllers have all eventually developed bizarre health problems; e.g., several have had rare malignancies in middle age.

    • scott w in northern CA says:

      dear dr david k:

      i am an elite controller, over 30 years now living with hiv. i recently celebrated my 61st birthday on october 30th. i am only currently taking two prescriptions, levothyroxine and a ventolin inhaler for the very rare instance of shortness of breath.
      i used to take antidepressants, methadone for pain control, testosterone gel and lisinopril for hypertension but went off of them cold turkey over two years ago. it wasn’t easy but i am glad i did it (i had no doctor or health insurance at the time). now that you have mentioned “rare malignancies” in some of your patients, among other bizarre health problems, i am scared to death. what do you suggest i do?

  12. Well, I think we need to see from which point of view we are looking to elite controllers: From the point of view of the patient’s safety: Here the question is about comparing how much harm (low grade systemic inflammation, long time organ damage, …) exist, stigma of being HIV positive and psychological stress plus the treatment complications (which with new ARVs are few) compared to how much succesful would be the treatment cure or control? From this point of view one could compare the situation exactly with HBV carriers with the difference that target organ mostly with HBV infection is liver and with HIV hopefully lymph nodes and brain and tests (as reservoir organs!).
    The second point of view is its probablity of spread of HIV, and the fact that the infected person will NOT be more an elite controller. Then it’s clear that treatment of elite controllers out weights letting them without treatment.
    So, my answer is cleary this, that if this question was suggested 10 years ago (low ARVs efficacy and high toxicities), I was against treatment, however, maybe its now the time to suggest them to start to treat their infections (as Opt out: I mean to suggest them treatment and to allow to those who don’t want treatment, leave treatment plan).

  13. Maegan Jones says:

    Hi Paul,

    Congratulations! NEJM Journal Watch: HIV is nominated for Healthline’s Annual Best Health Blog Contest! Check it out here:

    Every year we take a look at the top health blogs to honor & recognize them. This year things may look a little different, as we’ve updated the contest and taken strides to make sure everything runs smoothly & everyone has a good time voting for their favorites!

    The contest has a nomination period, starting now, until November 21st. After that, you can vote once per day for your favorite blog up until December 12th when the winner will be announced.

    Once voting begins, we’ll send you a reminder email so you can share with your followers. Remember, the blog with the most votes will receive a $1000 cash prize!

    Let me know if you have any questions.

    Congrats and good luck!

    Maegan Jones | Content Coordinator
    Your most trusted ally in pursuit of health and well-being

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.