An ongoing dialogue on HIV/AIDS, infectious diseases,
October 19th, 2011
Going, Going, Gone … HIV Treatment Failure Is Disappearing in People Who Take Their Meds
World Series time, hence the baseball reference in the title.
(Doesn’t take much.)
But over in Lancet Infectious Diseases — which has turned out to be a terrific journal, by the way — there’s a study reminding us that advances in HIV treatment in the late 2000s were truly spectacular.
The goal of the paper was to track the outcome of patients with “triple-class virologic failure” (TCVF) over the course of the last decade. Using data from 24 European cohorts, the investigators had access to over 90,000 patients, out of whom 2722 failed treatment with regimens that at some point contained the original three drug classes: NRTIs, PIs, and NNRTIs.
Rates of virologic suppression after TCVF steadily increased over the decade, from 19.5% in 2000 to 57.9% in 2009, and both AIDS complications and deaths declined. Significant predictors of virologic suppression by multivariable analysis included later calendar year, transmission category (MSM did the best), low viral load, and high CD4.
Those who had previously been able to achieve virologic suppression before TCVF were also more likely to be successfully treated — this clearly being a marker for good adherence. And although “past performance is no guarantee of future results,” it sure can be useful regardless. You have to assume that most of the 40% or so who did not achieve virologic suppression simply didn’t take their meds.
So what’s missing from their analysis?
Notably, because our objective was mainly descriptive, we did not attempt to further adjust for time-dependent variables such as access to new drugs …
In other words, the single most important factor for improved treatment outcomes in the 2000s — the introduction of darunavir, raltegravir, maraviroc, and etravirine — is left out.
(They also didn’t include data on resistance or adherence, but I suspect the former was tough to find, and the latter probably didn’t change all that much.)
So these results are either truly remarkable (if you follow HIV from a distance) or, if you are actively practicing HIV medicine day-to-day, are so obvious you can legitimately wonder why they did the study in the first place.
Both reactions are appropriate.
October 4th, 2011
Hormonal Contraception MAY Increase Risk of HIV
From the pages of Lancet Infectious Diseases, a study from Africa:
We aimed to assess the association between hormonal contraceptive use and risk of HIV-1 acquisition by women and HIV-1 transmission from HIV-1-infected women to their male partners … Among 1314 couples in which the HIV-1-seronegative partner was female, rates of HIV-1 acquisition were 6·61 per 100 person-years in women who used hormonal contraception and 3·78 per 100 person-years in those who did not (adjusted hazard ratio 1·98, 95% CI 1·06—3·68, p=0·03). Among 2476 couples in which the HIV-1-seronegative partner was male, rates of HIV-1 transmission from women to men were 2·61 per 100 person-years in couples in which women used hormonal contraception and 1·51 per 100 person-years in couples in which women did not use hormonal contraception (adjusted hazard ratio 1·97, 95% CI 1·12—3·45, p=0·02).
We’ve heard this story before, that hormonal contraception may increase the risk of HIV acquisition. The mechanism is unclear — vaginal thinning, increased HIV replication in genital secretions, something else — but undoubtedly part of it may be the simple fact that women receiving hormonal contraception must be less likely to use condoms.
In fact, that last factor is pretty darn likely, and why the results cannot be considered definitive — even though condom use was controlled for in the study.
Lack of concrete proof notwithstanding, women should be counseled that hormonal contraception could increase the risk of HIV acquisition and transmission — and hence it’s even more important that condom use be emphasized, especially in high HIV prevalence regions.
October 4th, 2011
Spanish HIV Vaccine Story Gets Lots of Attention — Here’s Why
If you’re looking for a good way to pass the time while running errands, traveling, or walking to work, I highly recommend the Freakonomics podcasts, which have taught me all sorts of interesting things.
Such as the fact that suicide is more common than murder in the USA, but gets way less attention. And how a restaurant can recover from serving a salad with a mouse in it. (Yes, waiter, there’s a mouse in my salad.) Or how much do our efforts to be better parents really matter? Not as much as we’d like to think, I’m afraid.
Which brings me to their comprehensive review of people who predict the future — markets, politics, sports, agriculture, you name it. Turns out that the kind of people most likely to make an accurate prediction are the ones who pretty much tell you what you already expect. But they’re too boring — anyone can say things will turn out the obvious way — so we have an insatiable demand for people who go the opposite route, making surprising predictions that go far out on a limb, foretelling something shocking or incredible.
And paradoxically, rather than taking these bold pundits to task for being wrong, we mostly forget about them until they get it right — at which point, we proclaim them geniuses. What sports fan of a certain age can forget Joe Namath’s shocking prediction in 1969 that the Jets would beat the Colts in Super Bowl III? (They did.)
And it didn’t matter that the guy predicting the 2008 market collapse had been saying the same thing every year for more than a decade, now he’s known as the guy who got it right! What vision!
(As for baseball prognostication, don’t get me started. Hmm, could be trouble.)
Which brings me to this HIV vaccine research done by a group of investigators in Spain, which has generated a fair bit of news coverage:
Researchers at the Spanish Superior Scientific Research Council (CSIC) have successfully completed Phase I human clinical trials of a HIV vaccine that came out with top marks after 90% of volunteers developed an immunological response against the virus. The MVA-B vaccine draws on the natural capabilities of the human immune system and “has proven to be as powerful as any other vaccine currently being studied, or even more”, says Mariano Esteban, head researcher from CSIC’s National Biotech Centre.
With the caveat that I am not an HIV vaccine researcher, I was surprised at how much news these early data generated — mostly because the study only involved 26 people.
And, to be blunt, also because the first report of the research in English occurred in the on-line “journal” called “Gizmag“. Unless there’s a report at a scientific meeting or journal I’m missing.
Regardless, it’s safe to say it will be hard to know when one of these highly-touted advances in the HIV vaccine effort actually turns out to be the real thing.
Someone making this prediction will eventually be right. Problem is that these could be the people just as likely — or more likely — to get it wrong.
October 3rd, 2011
CASCADE: When to Start, (Yet) Another Take
As we await the enrollment, analysis, and results of the START study — which is randomizing patients with CD4>500 to start HIV therapy vs waiting until the CD4 falls to 350 — much of the research on “when to start” ART in patients with high CD4’s comes from observational studies. Several have already been published (NA-ACCORD, ART-CC, CAUSAL), but one limitation of each of them is that none could accurately assess duration of HIV infection.
Enter “CASCADE“, or “Concerted Action on SeroConversion to AIDS and Death in Europe”. CASCADE includes patients from Europe, Australia, and Canada only if they have a defined date of HIV acquisition, thereby limiting effects of lead-time bias. In order to get at the when-to-start question, the investigators constructed something called “nested subcohorts” between 1996 and 2009, comparing the outcomes of those who started ART vs. those who didn’t.
The results? Out of 9,455 patients, starting ART (vs deferring) was associated with a lower risk of developing AIDS or death for those with a CD4 cell count < 500 — but not for those who started between 500-799. An interesting aspect of this study is that they were able to calculate a “number needed to treat” (NNT) to prevent progression. Over 3 years of follow-up, 21 and 34 patients would need to start treatment to prevent one patient from progressing to AIDS or death in those with CD4s between 200-349 and 350-499 respectively.
Usual caveats of observational studies apply — most notably that no such study can control for all factors between those who started ART vs those who didn’t that would influence outcome — but the results are helpful nonetheless, as the benefits of therapy before the CD4 cell count falls to < 350 shown here have been consistently seen in multiple studies. Furthermore, the NNT data to prevent AIDS or death — 34 for CD4 between 350-499 — place the benefits of treating these patients as even greater than what we accomplish with statin therapy for hypercholesterolemia to prevent MIs, which is estimated as 40-70.
Bottom line: Treating patients with CD4 350-500 and no symptoms may not be as exciting giving ART to someone with advanced HIV disease, but it sure makes good clinical sense.
September 24th, 2011
Warning: Viral Replication is Hazardous to Your Health
When studies evaluate the prognostic importance of measuring HIV viral load, they generally do so by assessing a single measurement rather than values obtained longitudinally. One obvious limitation of this approach is that baseline VL poorly predicts outcome after ART initiation — a finding in stark contrast to the original description of VL from the MACS cohort prior to effective HIV therapy.
Now, a collaborative group of researchers report the effect of something they call “viremia copy-years,” a marker of cumulative exposure to viral replication. This “area under the viral load” curve was calculated in 2027 patients starting ART at eight US sites from 2000-2008. Total number of VL measurements was a whopping 21,665.
The results are striking: viremia copy-years strongly predicted all-cause mortality — and did so more powerfully than either cross-sectional VL measurements or CD4-cell count. Each 1-unit increase in log10 copy years/mL was independently associated with a 44% increase in mortality risk.
The take-home message from this skillfully done study is that viral replication is bad for your health — even if your CD4-cell count is OK. And why might this be the case? The authors write:
We speculate the number of viremia copy-years, as a measure of cumulative plasma HIV burden, serves as a surrogate for and perhaps is the underlying driver of cumulative inflammation and immune system activation that approximates such long-term inflammatory biomarker effects.
A few other thoughts:
- Is high viremia copy-years a lab proxy for poor medication adherence? Probably — but this doesn’t invalidate the results.
- Of course in clinical practice, few of us have the calculus skills (remember integrals?) actually to calculate something like an area under the viral load curve. OK, I don’t have the math skills.
- A study evaluating whether pre-ART viremia copy-years influences prognosis would be particularly fascinating — and highly relevant for the when-to-start question.
- The results reinforce the findings from this interesting study linking time of uncontrolled viral replication to an increased risk of lymphoma — paper summarized in Journal Watch AIDS Clinical Care.
So even though we are unlikely to start using this viremia copy-years value in clinical practice, these data certainly make intuitive sense — and reinforce the notion that getting this lethal virus under control is the logical way to improve outcomes.
September 22nd, 2011
Common Sense on HIV Testing
There’s an editorial in today’s Boston Globe that concisely (188 words) describes the problems with both the current and proposed HIV testing laws in Massachusetts.
I’ve not been shy about the fact that I agree with every word of this piece.
And though I strongly recommend reading the whole editorial — it’s very well written — if you don’t have time, just read the title:
HIV Testing: No Need for Special Rules
That says it all.
September 17th, 2011
Drinking Coffee Prevents MRSA
I follow the medical literature on coffee very closely.
Why? Because I’m completely addicted — and, judging from the lines at the Starbucks, Dunkin’ Donuts, etc at the airports before early morning flights, I am not alone.
(It’s just one cup a day. Any more and say hello to palpitations, jitters, sweats, and long sleepless nights. Is there ever such a thing as a short sleepless night?)
So I thank a former ID fellow for pointing out this key paper:
We performed a secondary analysis of data from the 2003–2004 National Health and Nutrition Examination Survey to investigate the relationship between the consumption of coffee, hot tea, cold tea, and soft drinks, and MRSA nasal carriage …Individuals who reported consuming coffee had about a one-half reduction in the risk of MRSA nasal carriage relative to individuals who drank no coffee (odds ratio = 0.47; 95% confidence interval, 0.24–0.93).
(Brits will take solace that drinking tea was similarly effective.)
And let the record show that despite various researchers trying to blame coffee for ulcers, high blood pressure, coronary artery disease, gout, birth defects, anxiety, and several cancers — most notoriously pancreatic cancer — the evidence that it causes any of these things is weak at best. In fact, the coffee/pancreatic cancer paper is taught in some statistics classes as an example of how poorly designed case-control studies can give misleading results.
So in defense of this one cup a day addiction, let’s bring on more articles on the health benefits of coffee.
September 11th, 2011
Must-Read Paper: “Antiscience” and Lyme Disease
As I’ve written before, there are few clinical encounters more challenging for Infectious Diseases specialists than the patient who, despite negative standard diagnostic testing, believes he/she has Lyme disease.
Now, in Lancet Infectious Diseases, comes a paper entitled “Antiscience and ethical concerns associated with advocacy of Lyme disease.” It meticulously describes the distinctive world of alternative diagnosis, treatment, and passionate advocacy related to Lyme.
Some highlights — first, the background:
As with other antiscience groups, some Lyme disease activists have created a parallel universe of pseudoscientific practitioners, research, publications, and meetings, arranged public protests and made accusations of corruption and conspiracy, used harassment and occasional death threats, and advocated legislative efforts to subvert evidence-based medicine and peer-reviewed science.
And those odd diagnostic tests, often paid for out-of-pocket by the patient?
Despite warnings from the US Food and Drug Administration and the CDC about the potential unreliability of unvalidated diagnostic tests for Lyme disease, many LLMDs [Lyme-literate MDs] continue to use such assays. [Specific assays are cited in the full article.] Lyme specialty laboratories are favoured by some activists and LLMDs because their non- standard testing methods and interpretation criteria often lead to more positive results than other laboratories that rely on validated methods.
What to do?
Many patients who have been labelled as having chronic Lyme disease arrive at this diagnosis as a consequence of inadequate or frustrating previous medical care for symptoms that are difficult to define. Patients who suspect or who have been diagnosed with chronic Lyme disease should consider seeking a comprehensive assessment from an empathetic physician..
My advice: Read the full article. (That’s the polite form of RTFA.)
September 4th, 2011
“Novel” Approaches to Initial HIV Therapy: Part II
Two studies were just published on alternative strategies for initial HIV therapy. I’ve already reviewed the first one
The second paper is a single-arm (n=112) study of darunavir/r (once daily) plus raltegravir, the latest riff on the “NRTI sparing” approach.
As I mentioned when I first covered this study, the high rate of virologic failure — 26% overall, and a whopping 43% (21 of 49) with viral loads > 100,000 — came as a complete shock. In fact, I would have bet good money that this regimen would work just fine. (Glad I’m not a betting man — except in poker.) Five of the 28 virologic failures developed integrase resistance, all of them from the high VL stratum.
Why is this so surprising? You have the antiviral potency of raltegravir plus the high genetic resistance barrier of boosted darunavir — so what went wrong?
The study authors offer some possible explanations for these disappointing results (suboptimal adherence, asymmetrical dosing, lowered darunavir concentrations from raltegravir, minority variants resistant to raltegravir), but the bottom line is that this regimen just didn’t work very well. A fully-powered comparative study (darunavir/r plus raltegravir or TDF/FTC) is ongoing in Europe; I’m sure their DSMB is keeping a close eye on the results.
So two non-standard approaches to initial HIV therapy, with TDF/FTC + etravirine looking promising, and darunavir/r + raltegravir much less so.
September 3rd, 2011
“Novel” Approaches to Initial HIV Therapy: Part I
It’s been several years since the “preferred” or “recommended” initial regimens for HIV treatment have been consolidated into one of the following four:
- TDF/FTC + efavirenz
- TDF/FTC + atazanavir/r
- TDF/FTC + darunavir/r
- TDF/FTC + raltegravir
Any room for improvement in this “TDF/FTC + key third drug” approach? With the recent approval of TDF/FTC/rilpivirine, certainly this will have a role in some patients, but the issues of excess virologic failure and resistance in those with high viral loads will likely limit its use.
And though extended release nevirapine has been available for months, I confess it hasn’t really occurred to me to use it in a patient starting treatment given the other options out there. (And even those who are currently stable on NVP twice daily don’t seem eager to switch — I’ve offered.)
Now along come two interesting papers testing alternative initial treatment strategies. Both have quite interesting results that definitely got my attention when presented earlier this year at major meetings, small size of the studies notwithstanding.
The first is a blinded, randomized trial comparing etravirine and efavirenz, both given with 2 NRTIs. Note that the etravirine was dosed once-daily — not the twice-daily manner in which it is approved — and given in the older 100 mg (very crumbly) tablet formulation. Note also that the primary endpoint was the percentage of patients with Grade 1-4 drug-related CNS side effects, and not virologic efficacy. Since it was powered off differences in this side effect, only 157 patients were enrolled, not the 600 or so typically seen in a comparative study of efficacy.
Nonetheless, the results are impressive: 76% vs 74% in the etravirine and efavirenz arms respectively have VL < 50 at week 48, and for the group with baseline VL > 100,000, the results are 74% vs 67% — which is perhaps the first time in HIV study history that a regimen is numerically superior to EFV in this high viral load stratum. Also of interest, among the four patients with virologic failure on etravirine, zero show emergent resistance mutations — very unusual for an NNRTI — vs 3 of 7 for efavirenz.
Oh, and the primary endpoint — the occurrence of CNS side effects — significantly favors etravirine, as do lipid changes.
These results have to be considered encouraging on multiple fronts, including efficacy (caveat: small sample size, I know), resistance, tolerability, and lipid effects. Plus, with the newer 200 mg formulation of etravirine, this is an initial regimen of just three pills once-daily. Consider me impressed.
(Nerdy HIV specialist historical quiz question: how many pills/day was etravirine when in phase I/II studies? Hint, it was a lot.)
For another novel approach to initial therapy, read
Paul E. Sax, MD
Contributing Editor
NEJM Journal Watch
Infectious Diseases
Biography | Disclosures | Summaries
Learn more about HIV and ID Observations.
