A report by Jules Levin, Executive Director of NATAP (16 April 1998)

Dr. Franco Lori, who has helped to drive hydroxyurea (HU) research, reported data from this study of 24 individuals who received HU+ddI+Indinavir. Ten of these individuals initiated treatment prior to seroconversion. Using a more sensitive test than usual, 3/24 individuals were tested and found to have low level replication competent proviral DNA. Initially, using a more standard assay replication competent DNA was not found. One person who discontinued treatment and remained off therapy for about 10 months had undetectable plasma viral load; but, using a RNA analysis allowing investigators to view 40 million cells, some cells expressing RNA were found.

The dosing regimen was: HU 300 mg tid if <60kg or 400 mg tid if >60 kg; ddI 200 mg bid; indinavir 800 mg 3x/day. In the USA, the standard HU dose is 500 mg bid but in Europe different mg pills are available. Lori has suggested that tid dosing may be superior to bid dosing. It is possible that different mg HU pills will be available soon in the USA. Bristol Myers is planning a study to explore tid vs bid dosing.

Lori said this combination of drugs (protease, NRTI and HU) were selected not just because they have different targets (protease enzyme, reverse transcriptase enzyme, and a cellular target by HU of ribonucleotide reductase) but also because they are hoping two of these drugs (ddI and HU) would efficiently target the cells, such as, macrophages and quiescent lymphocytes which are supposed to be the main reservoir for longer term HIV. A protease inhibitor is added to target activated lymphocytes.

The addition of a protease inhibitor with d4T or other drugs will help to increase CD4.

* by in situ hybridization

** of these two patients one also had undetectable HIV RNA by in situ hybridization while the other was detectable

Among the group who were treated prior to seroconversion 5 or 6 had Western Blot never became completely positive. One patient remains that way after 18 months. Lori said that is consistent with a strong reduction of virus replication which could be due to less antigen presenting itself.

Lori briefly discussed the concern that absolute neutrophil count can decline from HU treatment. In this study he said it was not a problem. He believes that the HU dosing regimen is important to the effect on ANC. He has suggested that a ANC of 1700 before initiating HU therapy might be a cutoff below which could develop a problem after starting HU therapy. He said, with regards to developing an ANC problem, a person’s baseline ANC is more relevant than their CD4. Thrombocytopenia, which is an abnormal lowering of your platelet count, can be a concern from HU. Lori said he saw 1 case in a study of 40 patients receiving HU. If it occurs, he said it should emerge in the first 6-8 weeks.

Lori discussed one interested patient in the study. This person was infected 57 days prior to starting treatment. Data suggests that by this time there is a steady state viral load (set point). Lori said this patient’s viral load would not have declined this much without treatment. Their baseline plasma viral load was about 800,000 copies/ml. The viral load went down fast to undetectable. At 39 days after starting treatment, following reaching undetectable, the patient stopped treatment due to an episode of orchitis. Immediately the viral load started to rebound. After starting treatment again at day 42, viral load went back to undetectable. 141 days after starting treatment the patient developed an episode of hepatitis A and the person couldn’t take study drugs for 3 weeks. Although concomitant infections can cause viral load rebound, in this case viral load remained undetectable. Then he started treatment again but about a month later, which was about 200 days after starting study treatment, the patient decided to discontinue study drugs. Up until now, which is about 460 days after starting treatment, the patient’s plasma HIV RNA remains undetectable. 72 days after stopping treatment investigators analyzed a lymph node. They were unable to detect HIV DNA. They used a nested PCR system which allowed them to go down to one copy. But that testing system is limited to analyzing a maximum of 300,000 cells. Using a RNA analysis which allows them to look at over 40 million cells, and they found cells that were expressing low level RNA.

Lori said he didn’t know what to do with this patient until several papers were published saying that despite undetectable plasma RNA, resting CD4 T lymphocytes harboring latent DNA were detectable and HIV was recoverable in vitro. Lori sent samples from 3 acutely infected individuals in this study to Robert Siliciano’s lab. In Siliciano’s lab, it was their experience that they could routinely recover replication competent HIV with a frequency of 0.2 to 16.4 per 100,000 cells after HAART. Initially, no replication competent HIV was recovered in 2/3 patients. The other patient was the one who interrupted therapy 12 months before testing. The sensitivity of the test was increased 10 times (60 million cells screened) in order to detect HIV. HIV was recovered in these two patients with a frequency of 1 cell/10 million. Although low levels of RNA and replication competent DNA were detected, Lori said, it is possible the virus may not be replication competent in vivo, or some immunological changes are able to keep the virus under control. This might explain the one person who discontinued treatment. (ed note: for those still on treatment, remaining undetectable in plasma may be due to the treatment and/or one of the reasons Lori suggested).

Siliciano said the number of samples examined were too small and the assay variability could explain the difficulty in detecting replication competent DNA. It is to soon to draw any conclusions from this data.

Lori went on to caution that we should not draw conclusions because a small number of individuals were observed in this study. He suggested that the cytostatic effect of HU is important to the benefits of using HU in acute infection. That effect is to inhibit cell division. The low rate of cell division may cause a slower rate of HIV replication. If you have less activated cells you should have less virus. There are less cells available to be infected. If the theory is true that following HIV infection there is an excess of CTL activation which exhausts the CD4 and CD8 pool, then cytostatic effect of HU may limit the pool of cells available for infection. As well, using HU with ddI could prevent ddI resistance from being an obstacle. Data has suggested that even if ddI resistance develops the combination of HU+ddI might allow ddI to be just as effective as if ddI resistance did not occur.

Potential explanations for the mechanism of action: productive virus replication requires cell activation. CD4s and other lymphocytes are activated and proliferate in response to infection. The immune system is activated upon infection. A greater number of cells are available for infection. HIV increases its replication and infection of new cells. The massive proliferation or activation may exhaust the HIV specific (CTL) immune response. Early antiretroviral treatment may suppress HIV activity; but the cytostatic effect of HU may further inhibit HIV replication because it inhibits cell division, and therefore lowers the amount of cells infected. Its cytostatic effect prevents a robust CD4 increase but the same immunosuppressive effect can be exploited in the early stages of infection to suppress the activation of the immune system.

For this substudy a group of 8 treated participants were compared to 8 untreated individuals with a similar time of infection. The investigators reported the following improvements in immunological function.

The treated group’s CD4 counts increased, their CD8 counts decreased and therefore their CD4/CD8 ratio increased. The percentage of CD3-Zeta expression on CD4 and CD8 cells, which is supposed to represent t-cell functioning, was significantly higher in the treated group than the untreated group. The mean proliferation response to flu (twice as high) and allo antigen was higher in the treated than in the untreated groups. The treated group had significantly more naive CD4 and CD8 cells than the untreated group (ed note: although investigators didn’t report a baseline number for comparison). CD38 and HLA-DR expression on CD8 cells, which represents cell activation, were significantly lower in the treated group compared to the untreated group. CD28 expression, a co-stimulatory t-cell molecule which is essential for t-cell proliferation against antigen, was expressed at higher levels in the treated group than in the untreated group. (ed note: a comparison study might be necessary to detect if these immunological improvements are any different than would be seen with any fully suppressive regimen).

Lori concluded that further studies are required to evaluate these results. In fact, HU is getting much attention and many new studies are planned to explore its use in a variety of situations including salvage therapy, acute infection and early infection.

In the January issue of NATAP Reports, the 24 week data from this study was reported. 144 patients were initially randomized to receive ddI+d4T or ddI+d4T+HU. After 24 weeks, 23 of 26 evaluable patients initially randomized to the HU regimen had <200 copies/ml of viral load. At week 48, which was reported at Retrovirus, 16/20 were reported to have <200 copies/ml. But since 72 started in that arm, the remaining were no longer evaluable.