Flying back from Vancouver (coach) nearly 2 years ago to the day, Pulitzer prize winning journalist Laurie Garrett -- after an understandable rant over the questionable personal hygiene of her activist seat mate ("Is it chic to reek?") -- noted that the summer's advances in antiretroviral therapy could not be sustained without easier-to-take regimens. "They've done it with allergy medicine and antihypertensives," she observed, "surely they can do it with AIDS meds." Chelsea Westminster's irrepressible Brian Gazzard is fond of relaying the fact that a full 50% of transplant patients, "whose own sister or brother has had a kidney surgically removed in order to save his sibling's life," will fail to take the medication they need (once-a-day azathioprine or twice-a-day cyclosporine) in order to avoid transplant rejection. In Europe at large, the overarching principle to prescription writing has always been, "If it needs to be taken more than twice a day, it's not given."
Two years hence, we have edged considerably closer to user-friendly treatment regimens, but we are not quite there yet. Glaxo Wellcome, ever first to jump on a new marketing angle, had twice daily Combivir (AZT and 3TC in one pill) on the shelves before you could say "under-investigated pharmacokinetics." ("It's not AZT; it's Combivir!") Not to worry that, up until Combivir's launch, AZT had been generally accepted as a TID (and, prior to that, as a q6h) dosed medicine. Glaxo answered our concerns with a quick-and-dirty 16-week viral load equivalency study. (Why fuss with messy pharmacokinetics? AZT'd been dosed BID in Europe since day one.) The FDA looked the other way.
More recently, Merck and Agouron, under pressure from the onslaught of the twice daily "convenience" of a SQV/RTV dual protease combo, are out to prove that their imminently more popular PIs can also conform to the demands of an increasingly mutinous (and un- "undetectable") patient population. But how are patients and care givers to evaluate the legitimacy of the new BID -- and QD -- claims? Are short-term viral suppression results sufficient? This month Mike Barr and Spencer Cox take a look a the dizzying array of dosing modifications in the works and attempt to sort the science from the sleaze.
While no one from Roxane or Glaxo has yet gone knocking door to door hyping the use of once-daily nevirapine and lamivudine, clinicians in Europe and Canada -- arguably ahead of the learning curve -- are already openly discussing the possibilities of the brave, new and vastly simplified antiretroviral regimens. In a recent presentation to a group of HIV care providers in Scottsdale, (AZ), British Columbia's Julio Montaner had this to say:
"Restrictions regarding what can and cannot be eaten, what medicines may or may not be used, and oral intake of fluids may sound trivial at first glance; but they often become insurmountable obstacles in the long term. It is for this reason that attempts have been initiated towards the development of simplified and user friendlier treatment regimens. Recent data (all requests to Dr. Montaner for specific citations went unanswered) demonstrate that several drugs of proven antiviral efficacy can be used on simplified once daily regimens. Among them, 3TC in doses of 300 mg once daily, ddI in doses of 400 mg once daily and nevirapine in doses of 200 mg once daily for the first two weeks increasing to 400 mg once daily thereafter. When nevirapine is used once daily, it is generally recommended that it be given at bedtime as it may have a slight sedative effect."
Along those same lines, several new drugs such as adefovir -- possibly PMPA -- and efavirenz are currently being developed for once daily therapy. It is also possible that Abbott's second generation protease inhibitor ABT-378, used with 15 mg daily of ritonavir, may be given once daily.
Similarly, physicians in Frankfurt have released interim results from their all QD study of once daily ddI+3TC+nevirapine (N=70). At 20 weeks, according to Goethe University's Schlomo Staszewski, 90% of the study participants' plasma HIV RNA levels have fallen below the limit of quantification (400 copies/mL). In another study, dubbed "Atlantic," at Amsterdam University's Academic Medical Center, Joep Lange and colleagues have been comparing three different ddI+d4T-based regimens (the two nukes plus 3TC, the two nukes plus nevirapine, and the two nukes plus Crixivan). And to their surprise, the group which has fared the worst (judged by the percent of patients with "undetectable" plasma viral loads at week 24) turns out to be the Crixivan-containing group. "It's all about compliance," sighs Lange.
On the protease inhibitor front, Merck and Agouron both recently presented data from studies of their inhibitors in twice-a-day regimens. The current recommended dosing of both Crixivan and Viracept is three times daily. Neither of the companies, though, locked in a life-or-death struggle for market share, is willing to let its rival gain any sort of competitive edge.
Viracept, with a plasma half-life of 4.5 hours, enjoys a pharmacokinetic profile slightly more amenable to BID dosing than does that of Crixivan, with a plasma half-life somewhere around 3.2 hours. When Viracept was dosed at 1,250 mg BID, drug levels stayed about the minimum concentration thought necessary for therapeutic effect (the so-called IC50) and were comparable to the drug plasma levels achieved using the standard dosing schedule of 750 mg TID. In a small 6-month pilot study of BID dosed nelfinavir in combination with d4T and 3TC, viral load responses were good, with 14 out of 15 patients achieving plasma HIV RNA levels, 500 copies/mL by week 24.
For Merck, the situation is a little dicey. Knowing that Crixivan's pharmacokinetics were likely to look bad on a BID dosing schedule, the company initiated pilot studies that focused on plasma HIV RNA responses rather than on drug levels in the blood. Three groups of patients who were naïve to protease inhibitors and to 3TC were assigned to receive Crixivan 800 mg TID, 1,000 mg BID or 1,200 mg BID. All patients were given AZT+3TC.
About 30 patients were enrolled in each of the three groups, with an average CD4 count of 275 and a baseline viral load of 4.7 log copies/mL. After 32 weeks, 50% of patients treated with the standard 800 mg TID dose had viral load counts <500 copies/mL. Forty percent went below 50. For those treated with the BID regimen, 70% in both the 1,000 BID and 1,200 BID groups had viral loads <500 copies/mL. Sixty percent went below 50. So the BID dosing actually appeared superior to the TID schedule. (Echoes of Lange's Atlantic experience.)
All this was well and good until sneaky, envious Abbott Laboratories arrived on the scene. Still smarting from Norvir's less-than-stellar showing in the protease pageant, Abbott scientists are determined to figure out a way to make the viciously unpalatable ritonavir significantly less so. The latest brainstorm is to use tiny bits of the drug to potentiate the effects of their up-and-coming PI ABT-378 or, in the meantime, Crixivan itself. So in a PK study of Crixivan (1,200 mg BID) compared to Crixivan alone (also 1,200 mg BID) in combination with very low doses of ritonavir, Abbott carefully examined the pharmacokinetic profiles of the two regimens.
In combination with Norvir, levels of Crixivan stay well above the minimal acceptable concentrations throughout the dosing -- and the effects of food on Crixivan are eliminated. (So conceivably Crixivan and Norvir could be taken together on a full stomach.) The troubling news came from the PK data of BID Crixivan alone: at the 1,200 mg BID dose, blood levels of indinavir drop well below the IC50 for several hours between each dose. Certainly, more than one devil's advocate has pointed out that 1) The IC50s for these drugs are somewhat arbitrarily determined; and 2) The exact relationship between protease inhibitor plasma concentration and the quantity of active drug available for virus inhibition has not been well characterized, but the pharmacokinetic data on twice daily Crixivan is nevertheless far from reassuring.
Merck responds to these concerns by noting -- accurately -- that blood levels of Crixivan have not been correlated to antiviral response. (Although they were very quick to note that that low blood levels of Invirase, hard gel saquinavir, led to poor antiviral efficacy.) Merck believes that the demonstrated 32-week antiviral potency of Crixivan BID speaks for itself -- and that blood levels of drug are of little consequence. If the drug works, who cares if the pharmacokinetics are less than ideal? Furthermore, ever eyeing an opportunity to snipe at its chief competitor's product, Merck sales reps note that 90% of Viracept is bound to plasma proteins in the bloodstream (and is therefore unavailable for any protease inhibiting) whereas only 60% of Crixivan is so plasma protein bound. As a result, they say, simple blood levels of drug may not accurately reflect levels of active drug.
While TAG's Antiviral Project director, Spencer Cox, argues that it's probably too early to convert these TID meds into BID ones, physicians around the country have already begun doing so. Little did Newsday's intrepid science reporter know how effortlessly the money-minting pharmaceutical houses would be able to grant her wish. But then, maybe PK sleight-of-hand and dose-modification-by-fiat are not exactly what she had in mind. For TAG, the gold standard would be equivalent efficacy outcomes shown to endure out to one year -- blood levels of drug notwithstanding. But that sort of long-term information is simply not available yet. In the meantime, care givers and patients alike seem willing to wager that the trade-off between improved adherence and less-than-perfect pharmacokinetics will tip the balance in favor of more durable viral suppression -- rather than against it. Perhaps it is "all about compliance."
As part of TAG's new monthly meeting format (first Tuesday of each month at the Michael Callen/Audrey Lourde Community Health Center, 356 West 18th Street), St. Luke/Roosevelt's Dr. Donald Kotler was invited to share with us his latest thoughts on the fat redistribution phenomenon showing up in HIV-infected individuals on HAART. Just when you thought you'd heard it all before, Don arrives on the scene to challenge the conventional wisdom -- and turn the whole paradigm on its head. Mike Barr prepared this summary for TAGline.
The basis for Kotler's new hypotheses come from two recently completed studies he has conducted up at St. Luke's. (One's "sitting at AIDS"; the other's three-quarters completed and will be an oral presentation in Geneva.) Following his hunch that the "new" reports of strange fat redistribution in people on protease inhibitors were something we've seen since the early days of the epidemic -- just not paid much attention to ("When someone comes in with sunken cheeks or a pot belly and sight-threatening or gastric CMV, you tend to focus on the immediately life-threatening condition," he explains.), Kotler went back into his computer database of MRI scans of dozens and dozens of patients from the past year and carefully matched them to MRI scans of HIV-negative controls. What he unearthed was telling: some 40% of patients -- none of them on protease inhibitors -- exhibited what we now commonly refer to as "truncal obesity." In women, the syndrome manifested itself more often as increased breast size and loss of fat in the upper thighs, but it was there -- regardless of gender. And, interestingly, it was more evident in people with higher CD4+ T cell counts than in those with lower ones.
In a second, less high-tech study, Kotler used simple calipers and a tape measure to perform anthropometric assessments of patients' waist size and skin thickness. Of some 300 patients, Kotler divided them up into what he called the "old HIV" era (pre-1996) and the "new HIV" era (1996 to the present). (Of the "new HIV" patients, about 50 were taking protease inhibitors, 20 were taking non-protease inhibitor-containing antiretroviral regimens and another 20 or so were receiving no therapy at all.) He then matched these to a group of HIV negative controls.
In a multivariate analysis, taking into account treatment or no treatment, protease or no protease, CD4+ cell count, plasma viral load, age, sex, weight and height, the effect of plasma viral load was significantly predictive of the presence of truncal obesity -- even after controlling for all other parameters. Use of protease inhibitors and CD4+ T cell count were not. And the correlation was an inverse one: the patients with the lowest plasma HIV RNA levels were the most likely to exhibit truncal obesity!
So what's going on here? Kotler asks his compact audience rhetorically. First of all, let me say that "lipodystrophy" is a misnomer -- because it means "loss of fat." What we're seeing is not a loss of fat, but rather a redistribution of fat -- away from the peripheral and subcutaneous fat and towards the viscera. What can, all at the same time, cause hyperlipidemia, hypercholesterolemia, hypertriglyceridemia, hypertension, hyperglycemia, lowered testosterone levels, lowered progesterone levels, high uric acid levels and gout? The answer, in word: cortisol.
Cortisol, which the body converts from the naturally occurring hormone cortisone, is responsible for the metabolism of glucose, protein and fats. It also regulates the immune system and affects many other bodily functions. Cortisol levels in the body typically display a diurnal patter: elevated in the early morning and then dropping gradually but steadily throughout the day. Cortisol levels are lowest at night. Cortisol and its competitor hormone, adrenaline (a/k/a epinephrine), are produced by the body in response to stress -- be it outside stress (such as loud noise or the threat of bodily harm) or internal stress (such as anxiety or disease). Cortisol also causes resistance to insulin, which can lead to a kind of diabetes.
As Kotler tells it, these elevated cortisol levels ("hypercortisolism," if you will) are one of the last remaining legacies of a life-threatening condition which has become less life-threatening. ("I'm not saying this is hypercortisolism, just that that's certainly what it looks like.") "You see the same thing in women who have survived breast cancer, children who have survived childhood leukemia. I even had a Vietnam vet who had survived agent orange. They all had chronically elevated cortisol levels. They were all people with life-threatening illnesses who got better. It's the body's natural response to stress." "We're just seeing [a big wave of this] in HIV right now because everything's synchronized," Kotler exclaims. "Everybody started AZT at the same time. Then everybody 'failed' at the same time. Now everybody has started protease inhibitors at the same time."
But woe to s/he who would order a simple "fasting serum cortisol" and, upon receiving a reasonable level on the lab print-out, proclaim all is well. "The only accurate way to measure cortisol," Kotler cautions, "is a 24-hour urine cortisol." But you have to be willing to collect your urine for a 24-hour period and take it to the lab. (In a study of his own which examined the 24-hour urine cortisol levels of 9 HIV+ patients, 7 of 9 were above normal.)
What of the Sydney (Carr, Cooper et al. AIDS 1998, 12:F51-8) study at last February's retrovirus conference that declared a full 64% of protease patients to suffer from various manifestations of "lipodystrophy" you ask? And what of the knee-jerk reaction to categorically link the lipid problem to the protease inhibitors? About the Australian study, Kotler states matter-of-factly that the data were simply "over-interpreted." ("All the patients took protease inhibitors. All the patients' viral load fell dramatically. And they attributed it to the use of protease inhibitors -- rather than to the rapid and sustained reduction in plasma virus.")
As for the Falloon study (an NIH study published in the March 21, 1998 issue of The Lancet which unequivocally linked the fat redistribution problem to protease inhibitors -- indinavir in particular), her team analyzed a random sample of patients which included 10 individuals on indinavir who had evidence of truncal obesity, another 10 individuals on indinavir who had no evidence of truncal obesity and 10 controls. "Even though fully half of the patients taking indinavir had no evidence of truncal obesity, the NIH team concluded that the truncal obesity was caused by indinavir!" "It's possible that protease inhibitors are doing something on top of this," Kotler concedes, "but I really don't believe this is anything new."
A letter is due to come out soon in The Lancet describing "early onset cardiovascular disease" in individuals living with HIV. With more and more HIV-infected individuals walking around with elevated triglyceride and cholesterol levels, Kotler warns that we're looking at "heart attack city." He thinks a Clinical Alert is warranted. Or a letter to patients:
"As the prospects for long-term health have changed, so have the potential complications. If you have noticed any of the following, you should talk to your health care provider:
At the very least, the activist community (or NIH?) should convene a summit of sorts composed of leading HIV physicians. But what therapeutic interventions exist? Lipid lowering drugs abound (lovastatin, pravastatin, simvastatin, gemfibrozil), and have been met with mixed results. A year or so ago, Kotler recommended steroids and regular exercise. That didn't seem to work. ("So I was wrong," he bats back.) Testosterone supplementation, effective in the non-HIV arena, has also seen little success in HIV. Then there's human growth hormone (at $20,000-30,000 a year). Kotler says he's heard good things about it for the treatment of buffalo hump and truncal obesity but doubts it will help to replace fat lost peripherally. (It could also lead to diabetes.)
Arguably the most efficient mechanism would be to block the cortisol receptor itself. To date, the most effective agent for this is the controversial French "abortion pill," RU-486. "You'll have to put together that old AL-721 train again -- this time from France," Kotler notes wryly but half serious. "Maybe you should just eat more fish, less French fries and (as he fingers the Marlboro reds in the pocket of his dungarees) stop smoking."