• Session 287/12235: Hydroxyurea in combination regimens for the treatment of antiretroviral naive, HIV-infected adults
• Session 288/12205: Salvage of multi-drug resistant HIV infection with d4T/3TC/Hydroxyurea
• Poster Session 12371: Suppression of HIV replication and cell proliferation leads to minimal levels of proviral DNA and to immune recovery
• Late Breaker: Hydroxyurea containing combinations exhibit long-term efficacy, a novel resistance/rebound profile, and lead to anti-HIV specific immune response reconstitution

Hydroxyurea (HU) is a drug that has been used many years to treat patients with malignancies, and more recently, to prevent the painful crises of sickle cell disease. In 1993 and 1994 there were reports of its potential use to inhibit HIV replication, particularly in combination with nucleoside reverse transcriptase inhibitors. HU inhibits a cellular enzyme, ribonucleotide reductase, one of the most important enzymes in the synthesis of deoxynucleotide triphophates (dNTP), the building blocks of DNA. By lowering intracellular concentrations of dNTP, HU slows down the efficancy of reverse transcription. Nucleoside reverse transcriptase inhibitors are metabolized to their active triphosphate within the cell and then compete with the dNTPs for incorporation into the elongating HIV DNA strand. When dNTP concentrations are lowered within a cell the nucleoside reverse transcriptase inhibitor-triphoshate is more likely to be incorporated into proviral DNA, and when that occurs, the HIV DNA can not be elongated further. The early work with HU suggested that it would be most effectively used with reverse transcriptase inhibitors that are adenosine analogs, because HU lowers deoxyadenosine triphosphate concentrations to a greater extent than the other dNTPs. For that reason most of the studies to date have focused on HU with ddI, which is metabolized intracellularly to its active form, dideoxyadenosine triphosphate (ddATP). Several presentations of hydroxyurea's effects on inhibiting HIV replication were presented at this meeting. More importantly, perhaps, were some other presentations that describe some unique immunological benefits experienced by some patients that have been treated with hydroxyurea.

Session 287/12235: Hydroxyurea in combination regimens for the treatment of antiretroviral naive, HIV-infected adults.

A study conducted in South America and Canada compared two HU inclusive combinations versus combinations of reverse transcriptase inhibitors without HU in 183 antiretroviral naive patients with CD4+ counts between 201-500 cells/mm³. Results were presented up to week 24 with 118 of the patients reaching that timepoint. Patients received treatment with:

• ZDV + ddI
• d4T + ddI
• ddI + HU
• d4T + ddI + HU

HU was dosed at 500 mg bid.

The mean baseline viral load in these patients was about 50,000 copies/mL and the mean CD4+ count was about 350 cells/mm³. Viral load reductions, the proportion of patients with viral loads below quantifiable levels, and CD4+ count increases from baseline at study week 24 are shown below:

Not surprisingly, viral load reductions were best in the patients receiving the triple therapy. Although viral load reductions were slightly less in patients on ddI + HU compared to that experienced by the patients on dual nucleosides, the proportion of patients with viral loads <400 copies/mL was comparable in all of the dual therapy groups. In this study adverse events were few. The principal toxicity of hydroxyurea is bone marrow suppression. The white blood cell lineage is effected more than red blood cells and platelets. However, neutropenia occurred in only two patients on the study. Because hydroxyurea causes a relative lymphopenia, CD4+ counts do not rise as high in those patients on the drug. CD4+ percentages do rise, but the absolute counts increase to a more modest extent, despite better suppression of viral replication.

These results are similar to other studies that have compared ddI or d4T + ddI with and without HU. In general, the addition of HU has produced an additional 0.5-0.7 log reduction in viral load when used in a combination. Thus far, HU has not been studied in patients at more advanced stages of disease who have other therapeutic options. The limited increases in CD4+ counts that have been seen in almost all studies with the drug may make this an agent that we reserve for patients with higher counts, or one that we add into a regimen after CD4+ counts have increased. Neutropenia is the most commonly reported toxicity with HU. This side effect usually resolves within one or two weeks after the drug is held, and many patients can be restarted at a lower dose without recurrence. Other toxicities reported with hydroxyurea include hair loss and skin ulcers. Additional studies designed to identify the optimal dose and frequency of administration of HU are being planned.

Session 288/12205: Salvage of multi-drug resistant HIV infection with d4T/3TC/Hydroxyurea.

Although HU has been used almost exclusively with adenosine analogs to date, one presentation of hydroxyurea together with d4T and 3TC suggested that there may be some benefit to using the drugs with other nucleoside analogs as well. Dr. Steven Miles and colleagues from UCLA used this triple combination to treat highly experienced patients who had no other options for therapy. These patients had already been treated with an average of 7 other antiretroviral combinations, and all had experienced virologic or clinical failure on d4T and 3TC, and presumably had virus resistant to both drugs. In addition, all of the patients had received at least two different combinations including protease inhibitors and had failed on these drugs as well.

Eighteen patients were enrolled in the study and their mean baseline viral load and CD4+ counts were approximately 480,000 copies/mL and 20 cells/mm³. Follow-up was of varying duration. Three patients had no response to therapy. However, among those responders, the mean maximal viral load reduction seen at any point in time was 1.7 log, with several patients experiencing viral load reductions of more than 3 logs! Dr. Miles stated that viral loads were rebounding in many patients -- not surprising in this group with very high viral loads. CD4+ count increases were not observed. Adverse drug events were more commonly seen in this study than in others focusing on patients at earlier stages of disease. Neutropenia was seen in four patients; two patients needed blood transfusions; and one patient experienced bone marrow failure that had not resolved four months after the drug was stopped. Four patients had hair loss on study.

Why HU would have activity with d4T and 3TC in this patient population that has already failed on these two drugs is not clear. In vitro, HU has additive effects in combinaiton with non-adenosine analogs. Another possibility is that the concentration of those cellular enzymes responsible for metabolism of d4T and 3TC, particularly thymidine kinase and deoxycytidine kinase, increases after cells are exposed to HU. This could then lead to higher concentrations of d4T-triphosphate and 3TC-triphosphate, the active forms of the drugs, than those achievable in the absence of HU. These higher concentrations of the triphosphates may be able to overcome some degree of viral resistance such that the drugs regain some of their lost activity.

This report of HU with d4T and 3TC is intriguing, but should be considered preliminary. Dr. Miles stated that he would not treat these patients the same way if he could start this study again, but would take a more aggressive approach, adding additional drugs. Other studies investigating the use of HU as part of salvage regimens are being planned and some are underway.

Although CD4+ counts do not seem to rise in patients on HU, there is evidence that these cells regain lost function, and there was some early data presented suggesting that there may be some unique recovery of HIV-specific immunity in some patients receiving HU that has not been seen in patients treated with other maximally suppressive antiretroviral combinations.

Poster Session 12371: Suppression of HIV replication and cell proliferation leads to minimal levels of proviral DNA and to immune recovery.

Two abstracts were presented that explored the function of CD4+ cells in patients treated with HU inclusive combinations. One abstract reported on 24 patients, 10 treated prior to seroconversion during their acute HIV syndrome, 6 within one year and 8 after one year of seroconversion. All patients were treated with ddI, HU and indinavir. This cohort of patients includes the famous "Berlin patient," a patient who was treated during his acute HIV syndrome and eventually stopped therapy after approximately six months, only to maintain viral loads below quantifiable levels more than 18 months later.

Baseline CD4+ counts in these patients was 499 cells/mm³, and these counts increased on average by 168 cells/mm³ during the average 11 months of follow-up. This rise in CD4+ counts was atypical; as shown above, CD4+ counts don't usually rise in patients treated with HU. The elevation in these patients counts may be due to the fact that they received a protease inhibitor as part of their therapy, or it may be because a large proportion of these patients were treated during their acute infection, a time when CD4+ counts will transiently decline. These patients had significant increases in their memory CD4+ cell counts compared to a group of patients that elected not to receive treatment. In addition, the CD4+ cells of these patients were able to proliferate in response to exposure to influenza antigen and other non-specific lymphoproliferative stimuli; CD4+ cells of the untreated control group did not respond in this fashion. Markers of immune activation were also measured in treated and untreated patients. CD38 and HLA-DR expression on CD8+ cells was lower in the treated group compared to the untreated group (increased expression of these markers predicts more accelerated disease progression). Lastly, CD28, a costimulatory T cell molecule essential for the development of normal proliferative responses to antigen on CD8+ cells, was expressed at higher levels in the treated patients.

In addition to these favorable immunologic responses, 9 of 11 patients studied demonstrated robust CD4+ proliferative responses to HIV antigens. [A discussion focusing on the development of the HIV specific immune response following infection will continue after this section on hydroxyurea.] The reponse of CD4+ cells to HIV antigens is typically lost in patients treated much beyond their acute HIV infection, and a recent study has shown that patients treated during their acute retroviral syndrome are able to maintain these HIV+ specific lymphoproliferative responses. Because a portion of these patients were treated prior to seroconversion, one can not conclude from this data that the patients' maintenance of a CD4+ lymphoproliferative response to HIV antigens was related to the their treatment with HU.

Late-Breaker: Hydroxyurea containing combinations exhibit long-term efficacy, a novel resistance/rebound profile, and lead to anti-HIV specific immune response reconstitution.

A far more provocative abstract presented as a late-breaker described another group of 12 patients treated with ddI + HU. This group of patients had CD4+ counts between 250 and 500 cells/mm³ and a mean viral load of 51,795 copies/mL (range 602-173,166 copies/mL) prior to therapy, and all had established HIV infection. Viral loads declined progressively, but over a long period of time. After 40 weeks of treatment the patients' mean viral load was 1,847 copies/mL, and after 122 weeks it was 254, including 8 patients with detectable viral loads <50 copies/mL. A published report of virus analyzed from these patients (Lori, et al: AIDS Research and Human Retroviruses 1997; 13:1403) demonstrated that these patients had mutations associated with ddI resistance. However, this ddI-resistant virus was inhibited by ddI in the presence of HU. Therefore, these patients have experienced something unique -- long periods of sustained inhibition of HIV replication and no viral load rebound, despite the development of resistant virus.

These patients' naive and memory cell subsets of the CD4+ and CD8+ lymphocyte populations were analyzed and compared to two control groups, untreated patients and uninfected individuals. During this study, CD4+ counts rose from a mean of 376 cells/mm³ to 406 cells after 122 weeks, an increase of only 30 cells/mm³. However, naive cells comprised 54.1% of the CD4+ T cells in the treated patients, compared to 23.8% of the CD4+ T cells in untreated patients and 54.2% in the HIV uninfected controls. Therefore, after two years of therapy, naive CD4+ T cells were present at a similar percentage as was seen in a group of uninfected individuals. Naive CD8+ T cells also were present in higher percentages in the treated patients compared to the untreated patients (39.8% vs. 9.85%), but were still lower than in the uninfected individuals 48.8%).

The HIV specific immune response was also measured in these patients after 122 weeks of therapy. The CD4+ lymphoproliferative stimulation index to HIV antigens exceeded 10 in 5 of 12 patients. A stimulation index above 3 is considered a good response, and one above 10 is considered excellent. As will be discussed in the section on HIV immunity, this was an unexpected obervation in a group of patients with established infection at the time therapy was initiated, with CD4+ counts less than 500 cells/mm³, and with ongoing viral replication for the majority of the treatment period in most patients. Although this is an intriguing finding, the lack of baseline analyses limits the conclusions that can be made. Clearly, we will need additional studies to determine if HU has a singular capacity to reconstitue HIV specific immune function. However, the HU experience has offered us an opportunity to reexamine the aphorism that total inhibition of viral replication is required to maintain therapeutic benefit and promote immune restoration.