Treatment interruption: study found poor result for highly treated, highly resistant patients.

Patients whose antiretroviral treatment was no longer working well due to extensive drug resistance were randomly assigned to two different groups. In one, the control group, they changed their treatment immediately to a new antiretroviral regimen devised by their doctor with the help of viral resistance testing. In the other, the treatment-interruption group, patients also changed their treatment based on resistance testing, but first they went off all antiretrovirals for four months. The idea was to see if the virus would partly revert to more drug-sensitive strains during the treatment interruption, in the hope that the drugs would work better later when they were restarted.

The virus did tend to revert to more drug-susceptible strains in two thirds of the treatment-interruption patients. However those patients still did worse over all. While the two groups had the same number of deaths (8 in each), the treatment-interruption group had more cases of disease progression--defined as the first occurrence of a new AIDS-defining condition in that patient. At just under one year, the total number of cases of death or disease progression (the primary endpoint this study was set up to look for) was 22 of the 138 patients in the treatment-interruption group, vs. 12 of 132 patients in the control group that did not interrupt treatment. Only 2%, of the patients were lost to follow-up during this time.

In this trial the treatment interruption group had a CD4 count about 85 cells lower than the control group while treatment was discontinued, and a viral load 1.2 logs higher (about 16 times higher).

This study was conducted by the CPCRA--the Terry Beirn Community Programs for Clinical Research on AIDS. This group has very good patient follow-up, in part because patients are studied scientifically during the course of their regular medical care, instead of entering into a separate clinical trial. For example, in this study patients in both groups were given a regimen prescribed for them by their doctor, which could be changed if needed at any time--instead of having a standard regimen specified in a research protocol, which usually allows limited changes or none unless the patient leaves the study.

Comment

Even when a patient's virus does become more drug sensitive on the whole, plenty of resistant virus is still there. Today's resistance tests cannot detect low levels of any particular strain. And even if there were no viral copies, the DNA in some of the body's infected cells still contains the information, like a library of all viral strains the patient ever had. So when the drugs are started up again the susceptible virus may be suppressed better at first, but usually it does not take long for the resistant virus to come back.

Treating patients who are already resistant to most or all available HIV drugs remains difficult and controversial, with trade-offs between HIV progression, new resistance development, and drug toxicity. New drugs are being developed with different resistance patterns, and often the question is what to do now while waiting for enough good drugs to put together a viable suppressive regimen.

In any case this negative result should not prejudice people against clinical trials of other kinds of intermittent antiretroviral treatment, with very different patients, procedures, and goals.

References

(1.) Lawrence J, Mayers DL, Hullsiek KH and others. Structured treatment interruption in patients with multidrug-resistant human immunodeficiency virus. New England Journal of Medicine. August 28, 2003; volume 349, number 9, pages 837-846.