Skip directly to content

Recipes for Failure

'Early' Is Right, They Say, Unless of Course Your Regimen's Wrong; Then You're Shit Out of Luck


Burning Through Your Options One by One

October 1996

For this month's issue of TAGline, we decided to finish up last month's discussion with Aaron Diamond's Marty Markowitz and David Ho. In this second installment, a refreshing waft of realism breezes its way into the conversation. Perhaps only Elizabeth Kastor, staff writer for The Washington Post, has captured the complexities of recent developments ("The New 'Miracle' AIDS Drugs: A Dose of Hope and Hard Reality," 9/5/96) as accurately as our July TAG assemblage. "Hope," Ms. Kastor writes, "is a hungry thing, as easily fed by wishful thinking as by certainties."

Q: The difference [between the real world and clinical research] is that in the real world you have to follow patients who weren't able to stay on the regimen for a few weeks, patients who were non-compliant, patients who have trouble absorbing drugs; and we have to develop treatment strategies for them. We can't just shake our finger at them and say, "Bad patient! Sorry you're going to have to progress and die because you didn't stick to your regimen." We have to think of something more creative in terms of enhancing compliance and educating patients. But I still think there's going be a need to follow people who may have holes in their immunologic repertory that aren't going to be filled, so far, at least so far as we can tell. They're still going to have to be monitored for clinical events. The best way to do that is to compare the two very best regimens that we can think of now. I agree with you that maybe two nukes is not a good standard-of-care control arm.

MM: Fine. We can argue about what you want to compare; but again, unless you aim for undetectable viral load, you invite resistance into the host. That's the bottom line. And since you know you have regimens that can make people['s circulating virus] undetectable, how can you let people stay on regimens for 6 months, 8 months and 12 months, and they're not undetectable - and you don't even tell them? [As is planned in ACTG's "flagship" study, ACTG 320.]

Q: So that's your comment about the IAS [International AIDS Society] guidelines? That you think they're out-dated?

MM: I think that at this particular point in time, guidelines are probably not the best idea -- because it's a work in progress.

Q: If your goal is now to go for undetectable -- no virus replication whatsoever -- and it appears that resistance is forestalled, at least, I mean even John Coffin in his plenary on the last day of the conference said that, given the potency of these three-drug combinations, there must not be any three-drug-resistant virus in circulation -- unless it's in a latently infected cells, or we would have seen viral rebound within months. So it's not there in circulation; perhaps it's in latently infected cells; maybe it's not there at all. Anyway, given what you have found, what are the therapeutic implications? If people are not willing to wait for clinical data -- and generally they are not -- what are the therapeutic implications? You're saying that if you go in there with ritonavir + saquinavir and possibly nevirapine in advanced disease that you're going to see aviremia...

MM: [interrupts] Theoretically -- in a protease-naive person with a double protease combination. And nevirapine gives you even more potency.

Q: But is there a difference between populations? If you're sitting there with 500 CD4 cells now and a fairly low viral load, is there any difference between this chronic situation and acute infection? During the Wednesday [July 10] afternoon session on acute primary infection, a number of doctors said they were going to start screening for [HIV] p24 antigen in all of their gay male patients [in order to] initiate treatment. Should doctors start this sort of screening?

DH: There are no answers, but I can tell you my bias. You know, I'm trained as an infectious disease person. I've treated a lot of severe infections: bacterial, fungal, viral. When it comes to bacterial and fungal infections, wherever there is treatment, we don't accept a little bit of positive blood cultures. We happily accept that with HIV. Where's the difference here? Infectious disease has taught us for years -- especially with severe infections like systemic fungal infections, bacterial, TB -- you treat it early and you treat it at the outset with everything you've got. So if there are no answers, no data, to provide a concrete answer, you go back to the principles: what we've learned with acute leukemia and other infections. This is the only infection I know of that we are so happy to live with positive blood cultures. When I say "we," I mean, actually, the scientists and the physicians.

MM: The oncologist's point of view is the same. I mean, you don't say "Oh, there's a breast mass here. Let's check it again next year. It's not going to kill her in a year." Right? But at the same time, it's metastasizing and generating "drug resistant" tumor cells as well. I think, again, in two disciplines in medicine -- at least in diseases where there are either pathogens or diseases of cell turnover -- one would argue that when you treat, you treat aggressively. I think the big mistake is, "Well, it's not so bad right now so we'll just tease it a bit." I think that's the worse thing a doctor can do. I think you're better off not treating and waiting, but when you treat you hit it with a minimum of three drugs. And perhaps three drugs might not be enough; we don't know, but so far so good. But I think the wrong thing to do is two drugs at a time and then -- boom! -- you're in trouble.

DH: I don't see a lot of patients these days, but I've seen a few over the past two months, and I've seen too many that just go through one drug at a time. And it's really a shame. The cards for that patient are being burned up one after another every few months -- sometimes every few weeks. That's wrong. We know enough about this viral infection to know that sequential monotherapy is wrong. Marty made a very important point: that patient is actually better off not treated for a while until he can be put on a better regimen.

Q: So what can we do to educate doctors about this? I hear about this constantly: doctors who are simply adding on a protease inhibitor to a failing nucleoside regimen.

DH: I don't know how we can do it effectively. I was arguing with a physician in New York City who takes care of lots of HIV-infected patients who tells me that you don't need to prophylax for PCP with CD4 [cell counts] below 200 because it's easy to treat. He was serious!

Q: Maybe I just don't get it, but if the goal is to bring you down to "as low as you can go" -- to zero, to five, whatever it is you decide is low enough -- then doesn't what you start with depend upon where you are when you start? If my viral load is 20,000 when I start, and ddI has enough of an antiviral effect to bring me down below that level, why do I need a protease-based triple combination? Is there not a role for sequencing less aggressive one- or two-drug regimens, depending on what your viral load is?

DH: Didanosine alone is not potent enough. There will be residual replication.

MM: One protease inhibitor turns off viral replication 99.999% or so. You'll get close to a 2 log reduction. [You'll get about the] same [reduction] with [triple therapy,] AZT, 3TC and ritonavir, [for example.] The difference becomes durability -- and breakthrough. You can't achieve this with one drug -- whether you have [a pre-treatment viral load of] 5,000 copies or 15,000 copies or 150,000 copies. You just can't.

Q: What's the highest viral load you have seen go to undetectable?

MM: Six million. [awed hush]

Q: A few people I know who went on the triple combination two years ago had a very good response to it initially, and now they are seeing their viral loads again rise and their T-cells fall.

MM: Two years ago, not many people were truly on triple therapy. Most people were on sequential therapies that were added in a piecemeal fashion. I challenge you, I challenge you. There are very, very few people who had triple therapy initiated two years ago. They don't exist, essentially.

Q: But we know some.

MM: Well, then they cheated...

Q: [non-plussed] Yeh?

MM: -- because all the early protease studies were monotherapy.

Q: What are the implications from the research for people who have been through sequential therapies and are now in "late-stage AIDS" who are on the triple combination but have not achieved total viral suppression? What would you do about those people?

MM: This is a very difficult; I spend most of my time on the phone with people in this situation. Because, uh, the answer is, "I don't know." But what you have to push for ... you have to push for drug development: rapid assessment of potentially effective drugs and moving and shaking things at a grass roots level so that every drug company doesn't have to jump through the same hoops. I think if you heard Glaxo Wellcome's plan for 1592[U89], you'd all probably faint! They're not planning to file an NDA until late 1997 is what I heard! That's two years from now essentially.

Q: Would you, then, recommend that these people to go off drug for now until they can start a new combo regimen with, say, the Agouron protease [inhibitor]?

MM: No, I didn't say that at all. That's not the point.

Q: Then what would you do with a patient like that right now- whose viral load is still 600,000?

MM: I think there are a lot of people who unfortunately have participated in trials that allowed for the definition of these drugs' efficacy [to be determined], et cetera, et cetera. And I think that that's when you need to sit down with a physician who has some insight and ... You have viral load measurement available. You have a plethora of drugs available. You try to be creative, and you see what you can do. I think, you know, you can try to... If you're on a triple combination and your viral load is 600,000, it's not working, right? So it's time to jump ship, re-evaluate everything you've done, work with a doctor who seems to have some insight and see if you can make some changes. The bottom line is that a therapeutic change that works will be obvious in two weeks; that's the dynamics of viral replication. But then there's reality also. If you've gone through everything and your viral load is 600,000 when you're on drug, maybe you'd be better off going off drug and waiting for the landscape to change a bit. I don't know the answer. It's very difficult. [moment of silence]

Q: So you're saying that adding a new drug to a failing regimen is a recipe for failure?

MM: Absolutely. But I will say that, on the activist level, I think the real enemy is the way drugs are assessed, the way viral load measurements are wallowing -- still in committee is my understanding. I mean, where is the action on these? Where's the action on viral load measurements? I mean, I don't get it. I also don't understand why every drug company has to jump through the same hoop. It's like reinventing the wheel every single time a drug comes through. It's absurd! Nobody would say, "You know, you look like you have a very potent drug for breast cancer. You can just give it to a bunch of people with breast cancer as a single agent, and let's see if it works." No, they take CMF and CAF and they take that drug and they put them in combination. They don't do that to people. Why does HIV have to be singled out like this? I think that's wrong.