The Body: The Complete HIV/AIDS Resource
Follow Us Follow Us on Facebook Follow Us on Twitter Download Our App 
Professionals >> Visit The Body PROThe Body en Espanol
  • Email Email
  • Printable Single-Page Print-Friendly
  • Glossary Glossary

Case Management

John D., a Patient Who Has Failed All Conventional Therapies

What Options Do Patients -- and Providers -- Have When Salvage Therapy Is Indicated?

October 1998

A note from The field of medicine is constantly evolving. As a result, parts of this article may be outdated. Please keep this in mind, and be sure to visit other parts of our site for more recent information!

Over the past decade John D., a 42-year-old white male, has run the gamut of antiretroviral therapies. He received an AIDS diagnosis in 1987, when he was hospitalized with what proved to be P. carinii pneumonia. The PCP episode, diagnosed by bronchoalveolar lavage, responded well to a three-week course of intravenous TMP-SMX, and John began antiretroviral therapy after his discharge. He was one of your first patients to take zidovudine, and he was on sequential monotherapy -- first with ddI, then with ddC -- before switching to the combination of ZDV and 3TC, which produced a detectable improvement in his clinical symptoms and immunologic markers.

When protease inhibitors became available, John added saquinavir to his ZDV-3TC regimen -- and then switched from saquinavir to indinavir as soon as the more powerful protease inhibitor reached pharmacy shelves. Although his CD4 count never rebounded above 100 cells/mm3 on this three-drug antiretroviral regimen, John's viral load did drop below the level of detection of the first-generation Amplicor assay, which was 400 copies/mL. It remained undetectable for 16 months, and then began an inexorable rise. When John's viral burden reached 20,000 copies by bDNA, he started a regimen of nelfinavir, d4T, and nevirapine.

On this new regimen John did well for almost a year. His viral load never exceeded 500 copies/mL, and he reported that he felt fine. He still feels fine, but his CD4 count has been drifting slowly downward over the past six months, and his viral load -- which was 16,000 in November of last year -- jumped to 45,000 in March of this year. You have just obtained a new measurement of his viral burden. It is now 52,000.

When you called John to reconfirm his next appointment, you told him that you wanted to talk to him about his treatment options. When he comes into your office, he looks you in the eye and asks you to tell him what he should do now. How do you respond?


Dr. Judith A. Aberg, UCSF AIDS Program, San Francisco General Hospital, and a member of the editorial advisory board of HIV Newsline:

This case is atypical in many respects. John has lived longer -- and fared better -- with his AIDS diagnosis than the vast majority of patients diagnosed a decade ago. He has responded well to all of the antiretroviral therapies he has been assigned, beginning with ZDV monotherapy and progressing through sequential monotherapy, dual-nucleoside therapy, and triple-drug combination therapy.

John's case is atypical in another important respect: by the standard virological measurements he broke through on an indinavir-containing regimen after only 16 months, yet he responded for almost a year to a successor regimen that contained nelfinavir. Normally, indinavir failure is attributable to a series of critical genetic mutations that result in resistance to all of the currently approved protease inhibitors, including nelfinavir (see "The Problem of Protease Resistance," Vol. 3, No. 3). Given that John did respond to his nelfinavir-containing regimen, I would first explore the possibility that this patient "failed" indinavir because of adherence problems, rather than because he developed indinavir-resistant viral strains.

Faltering adherence to long-term therapy is the rule, not the exception, and as Dr. Margaret Chesney so tellingly observed in "Compliance: How You Can Help," in the June 1997 issue of HIV Newsline, the multidrug antiretroviral regimens that are now being prescribed, with their exceedingly strict schedules and their detailed directives regarding such factors as the timing of meals and intake of fluids, "are the most complicated that have ever been prescribed for continuous and open-ended treatment of such a large patient population."

It would hardly be surprising to see incomplete compliance in a patient who has been on antiretroviral therapy for more than eleven years, and if poor adherence was indeed a factor in John's apparent failure to sustain suppression of viral replication on ZDV, 3TC, and indinavir, one would naturally wonder if poor adherence had also played a role in the "failure" of his regimen of nelfinavir, d4T, and nevirapine. Did he develop true resistance to the nelfinavir element of his combination, or did he simply fail to comply with the dosing regimen for his new protease inhibitor?

We know that viral isolates from patients who develop resistance to nelfinavir remain partially susceptible to indinavir and ritonavir, so phenotypic testing would be of real value in John's case. If those tests tell us that John's virus has developed resistance to nelfinavir via the D30N mutation, then the possibility exists that any one of the other three protease inhibitors might still produce clinical benefits. At the very least, genotypic testing would help us determine which agents would not work in a salvage regimen for this patient.

Because I can think of no other explanation for why this patient had such a strong and durable response to nelfinavir after seeming to fail his indinavir-containing regimen, I would approach this case on the assumption that John's indinavir failure was not due to the development of indinavir-resistant virus. In such a situation the plausible therapeutic options would include switching John to the combination of ritonavir and the soft-gel formulation of saquinavir or reinstituting indinavir therapy. In either case I would add a new nucleoside analog or NNRTI to his regimen. This patient would be eligible for compassionate use of adefovir, abacavir, and/or efavirenz (with the latter obtainable only if he discontinued nevirapine). If John's previous exposure to ddI was relatively brief, one might well consider the use of ddI with hydroxyurea (see "Administering hydroxyurea with ddI increases intracellular concentrations and facilitates activation of this nucleoside," Vol. 4, No. 2).

If, however, John does prove to have a mutant strain of virus that is resistant to all of the currently available protease inhibitors, then his options are much more limited. Some investigators have proposed discontinuing all antiretrovirals for a washout period of two to four weeks and then initiating a regimen that consists of two powerfully synergistic protease inhibitors, an NNRTI, and at least one nucleoside analog, preferably one to which the patient has no previous exposure. The benefit of this approach is that when you resume therapy you encounter wild-type virus, which is generally susceptible to any combination of antiretroviral agents. The drawback is that in most instances resistant strains begin to reassert themselves in fairly short order.

It should be emphasized here that although this approach makes a certain amount of theoretical sense, no one really knows at this point how well such last-ditch salvage regimens will actually work in practice. All we know is that such regimens are being tried, sometimes with encouraging results (see "A six-drug regimen proves to be surprisingly effective in heavily pretreated patients," in AIDS Care, Vol. 2, No. 2).

Dr. James O. Kahn, Associate Professor of Medicine at UCSF Medical School, Associate Director of the AIDS Program at San Francisco General Hospital, and a member of the editorial advisory board of HIV Newsline:

John is living testimony to the success of antiretroviral therapy. He is also an example of the limitations of available treatment. He has lived 11 years beyond his initial AIDS diagnosis, all that time with low CD4 cell counts. His survival should therefore be attributed in part to his response to a succession of antiretroviral regimens and in part to intelligent prophylaxis against the opportunistic infections to which individuals with low CD4 counts are susceptible.

It appears that John may now have reached the limits of available therapy. The first order of business is to determine if the dominant virus in his system is still sensitive to any of the currently available medications -- including those that may be available to him under expanded-access programs. I would maintain John on his current regimen, which is still modestly effective in suppressing viral replication in his body, while awaiting the results of phenotypic analysis of his blood -- a process than can take a week or more to complete. The results of that test panel would help me determine what antiretroviral combination would work best for John at this stage.

Phenotypic testing, still in its infancy, is time-consuming and expensive, costing roughly $600 to $900 for each agent tested. Given the number of antiretroviral agents now available, a full panel of these tests can cost thousands of dollars. Moreover, these tests can be conducted only in specialized laboratories, and for all of these reasons phenotypic assays are unlikely to become widely used clinical tools any time in the near future.

If I could not obtain a phenotypic profile of John's virus, genotypic testing -- which is less costly and more widely available -- would help me establish which drugs (and drug classes) were least likely to have activity against John's viral strain. Although genotypic assays are readily reproducible and rapidly obtained -- usually in a single day and at a cost of $400 to $600 for the entire battery of tests -- they are limited in their usefulness by the current lack of consistent data correlating mutation patterns with clinical outcome. Such tests can tell us what won't work -- but so does a careful reading of the patient's drug history.

Absent both phenotypic and genotypic test results, I would use the following process to assess John's treatment options:

First, I would evaluate each class of medications for their likely degree of activity in this heavily pretreated patient. Over the past decade John has received zidovudine, ddI, and ddC monotherapy, followed by the combination of ZDV and 3TC. His current regimen includes d4T. Without phenotypic testing we have no way of knowing if any of the medications that John was given long ago as monotherapy would exert a consequential antiretroviral effect at this point, but it is likely that ZDV, 3TC, and d4T are now of limited value to John.

The non-nucleoside reverse-transcriptase analogs are a resistance-prone class of drugs. High-level resistance develops very rapidly when these drugs are given as monotherapy, and such resistance seems to exert a class effect. As a result the NNRTIs are approved for use only in combination with at least two other antiretroviral agents (see "Nevirapine may prove to be effective when used in combination with protease inhibitors," Vol. 3, No. 1, pages 16-18). John's current regimen includes a protease inhibitor, a nucleoside analog, and the oldest of the NNRTIs, nevirapine.

If John's virus is indeed resistant to nelfinavir, the protease inhibitor he is taking -- a not unlikely conclusion, given that he previously failed an indinavir-containing regimen -- and if he no longer derives any benefit from d4T, his nucleoside analog, then he has in effect been on nevirapine monotherapy for some months now. Under such circumstances, resistance to the NNRTI would develop rapidly.

Assuming John has in fact failed nevirapine, delavirdine is unlikely to have significant anti-HIV activity against his virus. Efavirenz may also be of limited therapeutic value in patients who have developed resistance to nevirapine. The same class effect would be anticipated for the protease inhibitors. John's treatment history includes brief exposure to the original formulation of saquinavir, and the clinical indications are that he has developed resistance to two more powerful protease inhibitors, indinavir and nelfinavir. My colleague Dr. Aberg wonders if these apparent therapeutic failures should be attributed to the development of drug-resistant viral strains or to imperfect adherence on John's part to his assigned dosing schedules. This astute hypothesis would account for John's transient but positive response when he was switched from an indinavir-containing regimen to one that included nelfinavir.

True high-level resistance to indinavir should confer resistance to nelfinavir. The fact that it did not in John's case is anomalous, and if Dr. Aberg's hypothesis is right, this patient may indeed derive some benefit from a salvage regimen that contains indinavir or the combination of saquinavir-ritonavir. Certainly such options are worth exploring, especially given how limited John's therapeutic choices are at this point. In such situations it is helpful to remember that individual patients have individual responses to individual drugs.

In a sense, all patients are anomalous -- and just because a particular salvage therapy has shown little efficacy to date does not necessarily mean that it will not help a particular patient. We have very little data on salvage regimens, and what we do have comes from small, retrospective, uncontrolled trials. What our own clinical experience tells us is that some things that should work -- like John's first triple-drug combination of ZDV, 3TC, and indinavir -- do not. And some things that should not work -- like a nelfinavir-containing regimen in a patient with apparent high-level indinavir resistance -- do work. With patients like John -- long-term survivors whose response to therapy does not conform to expectations -- we eventually find ourselves operating in a realm of guesswork and gut feeling. And if such patients are game to try untested salvage therapies, we should be willing to accede to their wishes.

In an effort to determine what will work for John at this point, I would spend time attempting to elicit John's treatment goals. Are they realistic goals? And if they are not realistic for a patient in John's position, can I help him to develop a more reasonable set of goals? The long process of helping a patient like John accommodate himself to the fact that he is approaching the limits of what current antiretroviral therapy can offer him is as much a part of the patient-provider partnership as the provision of optimal antiretroviral therapy.

At this juncture John needs to consider the full implications of his rising viral burden in light of his long treatment history. Long-term survivors like John have already triumphed over very long odds, and it is not uncommon to find that such patients have persuaded themselves that they can beat the odds altogether. Easing a patient like John away from that reality-defying certitude -- without undermining his determination to do the best possible job of maintaining his health and his mental outlook -- is a delicate and demanding task. It is, moreover, a primary responsibility of all primary-care providers, more important at this stage of John's disease than any adjustments in his medications. Unfortunately, it is one professional responsibility for which nothing in the PDR can prepare us, since it calls upon our humanitarian instincts, not our technical expertise.

At this stage a reasonable goal of treatment might be to keep John's viral burden under 100,000 copies/mL for as long as possible, and to do so without inducing side effects that would compromise his quality of life. With the expectation that John's CD4 count will continue to fall, it is vital to discuss the importance of maintaining effective prophylaxis against PCP, which poses an ever-graver threat as immune competence declines. At this visit or a subsequent visit, I would also want to prepare John for the likelihood that other forms of OI prophylaxis will be necessary down the road.

Assuming that the twin goals of John's therapy are to suppress virus replication as completely as possible while maintaining a strong shield against opportunistic infections, I would aim to develop a maximally suppressive antiretroviral regimen with different classes of agents. At this point, there are few choices, and if I could not obtain phenotypic information to guide my decision-making, I would try to enroll John in one of the many research protocols that are currently evaluating salvage regimens in people with drug-resistant virus who have failed other therapies (see "Selected Clinical Trials Open to Patients with HIV Disease," the Pull Out and Save feature in the February 1998 issue of HIV Newsline). For information on clinical trials for which patients like John may be eligible, call 1-800-TRIALS-A.

If I could not enroll John in one of these resistance studies and I could not obtain phenotypic data on his viral isolate, I would treat John with the following regimen, based on what I know about his past drug history:

  • ddI plus hydroxyurea (to which I would consider adding abacavir if I could obtain it for John through Glaxo's expanded-access program, at 1-800-501-4672)

  • Ritonavir plus the soft-gel formulation of saquinavir

Susan Shea, R.N., F.N.P., UCSF AIDS Program at San Francisco General Hospital, and a member of the editorial advisory board of AIDS Care:

It may seem ironic to say so about a patient who is failing yet another antiretroviral regimen, but John strikes me as a very fortunate patient. Not only is he alive and leading a largely uncompromised life more than a decade after an AIDS-defining bout of PCP, he seems to have experienced some benefit from each of the drug regimens he has been assigned over those years -- including several that are of such limited efficacy that we no longer use them at any stage in the treatment of any patients.

Moreover, John seems to have derived real virological and immunological benefit from his current salvage regimen -- which suppressed his viral load below 500 copies/mL for the better part of a year and which continues to exert an antiretroviral effect. I have not had very good luck using nelfinavir in patients who have failed indinavir, and the fact that John exhibited such a positive response lends weight to Dr. Aberg's theory that his earlier breakthroughs are attributable to adherence problems rather than to high-level resistance to the drugs John was assigned.

My first step would be to interview John in depth, to assess how he feels about the fact that he has once again broken through on what was supposed to be a maximally suppressive regimen. On the surface, it appears that this patient does tend to comply with daily dosing schedules -- his long-term survival would seem to be testament to that fact. If Dr. Aberg is right, and John did in fact have trouble adhering to the particular demands of indinavir dosing, then it may be possible to elicit this information in the course of a careful, unhurried interview.

In any case, it does absolutely no good to scold a patient who has found it difficult to comply with the exacting demands of combination antiretroviral therapy. John should be roundly congratulated on his clinical and emotional health, should be praised for his perseverance, and should be gently but firmly reminded that full compliance with his assigned regimen is essential to keeping the virus under control.

Given John's deteriorating immunologic and virologic status, I would also tell him that breakthroughs can occur even when patients adhere perfectly to their dosing schedules. A number of factors outside John's control can contribute to the development of viral resistance -- which makes it all the more important for John to exercise command over the one factor he can control absolutely: compliance with his assigned regimen.

I think all of this is very important for patients to hear, and to hear at each clinic visit. In such an atmosphere, patients are more likely to be willing to confide lapses in adherence, knowing that they will not be taken to task for those shortcomings. Forgiveness -- and even absolution -- for such all too human failings is part of the practice of primary-care medicine. It is by offering sympathy and understanding that we get continued cooperation, and renewed commitment, from all of our patients, especially those for whom adherence presents real problems.

The next step would be a detailed history of recent symptoms that could be related to progression of John's HIV infection. If I felt that he was still essentially asymptomatic, I would tell the patient that from my point of view there are a couple of strategies that he and I could pursue:

1. Do nothing, and follow his labs.

I do not know, from the thumbnail history offered above, either John's nadir CD4 count or his zenith viral load, so I have no parameters for making a judgment about the significance of his recent drop in CD4 count and rise in viral burden, but doing nothing is an option for some patients, especially if their zenith viral load was in the hundred-thousands or higher.

It must be borne in mind that John's current regimen is still working -- not as effectively as it once did, but perhaps as well as any therapy we might switch him to at this late stage. As my colleague Dr. Kahn has observed on numerous occasions, except in cases of acute HIV infection, time is on our side when we are making decisions about initiating, or switching, antiretroviral therapy. Waiting to see what John's next set of labs reveal about his current situation is certainly an option.

2. Stop all medications and give the patient a "drug holiday" while waiting for the results of genotypic testing.

Dr. Kahn has thoroughly reviewed the merits, and limitations, of both phenotypic and genotypic testing in patients like John. Over the next few years these diagnostic tools are going to become more widely available, and more clinically relevant, to the decisions we make about what drugs to choose for patients who break through on their first -- or second, or third -- multidrug antiretroviral regimen. Indeed, these assays are going to assume increasing importance as, with time, more and more patients experience breakthroughs on their initial regimens. In any event my practice is to bring the patient back in when I have the genotyping data, review the results in light of his past drug history, and work out the next treatment plan with him. I want patients to be part of this process. It helps them understand that the options are few, and the benefits are likely to be modest.

3. Assume that the patient's 11-year drug history is as accurate as genotyping and start him on a new regimen immediately.

Based on what information has been provided, I would choose the same salvage regimen that Dr. Kahn chose: ddI, hydroxyurea, and ritonavir plus the soft-gel formulation of saquinavir. Although Dr. Kahn has legitimate doubts about the likelihood that John will benefit from efavirenz, given the fact that he has probably developed resistance to another NNRTI, the nevirapine that is an element of his current regimen, I might want to consider adding that agent to his regimen. And, like Dr. Kahn, I would want to explore the possibility that John could obtain abacavir under Glaxo's expanded-access program.

4. Put the patient on a combination of recycled drugs that combines some antiretroviral potency with ease of dosing and wait until ABT-378 is available to add to ritonavir, ddI, hydroxyurea, and abacavir.

Genotyping can help the care provider determine which combination of antiretroviral agents will make the most effective "bridging" regimen, but the most important factor is going to be how the patient feels physically and psychologically. If he is developing symptomatic HIV disease -- or if he fears that he is on the verge of developing symptoms -- he may be motivated to take a more aggressive (and more demanding) approach to therapy such as the six-drug protocol that Dr. Aberg cites above. On the other hand, if he is still asymptomatic but really burned out after two years of intensive drug therapy he may elect to simplify his antiretroviral regimen and wait until ABT-378 and other investigational agents become available (see "The Next Generation of Antiretroviral Agents -- An Update," Vol. 3, No. 6).

As with all patients, I would reassure John that making a decision about the next phase of his therapy is not urgent. He needs to know that he and I can make this treatment decision down the road, after a thorough assessment of his current situation and a detailed discussion about his treatment options, both immediate and future. In the interim, he can count on some degree of antiretroviral effect from his current meds. Like Dr. Kahn, I would round out my interview with John by discussing his need for additional prophylaxis at some future date in light of his falling CD4 count.

Harvey J. Makadon, M.D., is Associate Professor at Harvard Medical School and Vice President of Medical Affairs, Beth Israel Deaconess Medical Center, Boston, MA.

A note from The field of medicine is constantly evolving. As a result, parts of this article may be outdated. Please keep this in mind, and be sure to visit other parts of our site for more recent information!

  • Email Email
  • Printable Single-Page Print-Friendly
  • Glossary Glossary

This article was provided by San Francisco General Hospital. It is a part of the publication HIV Newsline.
See Also
More on Third Line/Rescue HIV/AIDS Treatment