Dr. Sergio Lupo, from Argentina, presented the results of an international randomized controlled study comparing the triple combination ddI/d4T/HU to other anti-HIV combinations (abstract 12235). The study, which was carried out in Argentina, Brazil, Canada and Mexico, enrolled 183 treatment-naïve patients into one of the following four arms: ddI/AZT; ddI/d4T; ddI/HU; or ddI/d4T/HU. The HU dose was 500 mg twice a day. Baseline characteristics were as follows: CD4 counts between 200 and 500 cells/mm3, with median count of 345; median viral load of 33,000 copies/ml; median age of 30, and 68% were male.
At the time of the report, 75% of the patients had completed 24 weeks of treatment, and viral load results were available for 118 patients. Mild-to-moderate side effects including nausea and vomiting were reported in all the study arms, with a higher incidence in the AZT/ddI arm. At week 24, 76% of the patients in the ddI/d4T/HU arm had undetectable viral load (below 400 copies/ml), against 53% in the ddI/d4T arm, 43% in the ddI/HU arm, and 36% in the AZT/ddI arm. These differences are statistically significant. Mean drop of viral load in the triple-combination arm was reported to be 2.0 logs (99%). Changes in the absolute CD4 counts were minimal in the HU-containing arms, compared to some increases in the ddI/d4T and AZT/ddI arms.
Dr. Franco Lori, from the Research Institute for Genetic and Human Therapy (RIGHT) of Washington, D.C., and Pavia, Italy, has indicated that RIGHT's future research protocols for HU-based combinations will include d4T as well as ddI. Dr. Lori and RIGHT were the original promoters of hydroxyurea research.
Dr. Lori and other researchers at the Geneva conference presented a new theory to explain the additional benefit of this triple combination over the ddI/HU double combination. Hydroxyurea, which is known to inhibit the activity of the cellular ribonucleotide reductase enzyme, may also indirectly increase the activity of another group of cellular enzymes. The ribonucleotide reductase enzyme is responsible for the creation of the nucleotide pools that serve as building blocks for both cellular and HIV genes. Ribonucleotide reductase is especially important for the creation of adenosine, for which ddI serves as a defective replacement. By inhibiting the action of ribonucleotide reductase, HU slows the creation of new HIV genes as the virus attempts to infect a new cell.
The other cellular enzymes that HU may affect are responsible for the conversion of the nucleosides and defective nucleoside analogs such as AZT, d4T, 3TC and ddC into the active compounds in a process known as "phosphorylation." When the nucleoside pools are depleted by the action of HU, cells appear to enhance the activity of these enzymes, making more activated nucleoside analogs available inside the cell to stop viral replication.
In a World AIDS Conference satellite session sponsored by Bristol-Myers Squibb (producer of HU as well as of d4T and ddI), Dr. Jean-Pierre Sommadossi, from the University of Alabama at Birmingham, presented the results of a phosphorylation study in which CD4 cells were taken from patients who had been on long-term (3 to 4 years) nucleoside analog therapy and were failing. These cells were exposed to d4T and HU in the test tube, and the results showed a three- to four-fold increase in d4T phosphorylation when HU was present, regardless of the nucleosides to which the patient had been exposed in the past. (Last winter, Dr. Sommadossi reported the results of experiments indicating that prolonged exposure to AZT reduces cells' phosphorylation of that drug, and of d4T as well.)
Another Conference report (abstract 42262) described the results of a study at the Pharmacology University of Liverpool (U.K.) that investigated the effects of HU on the activation of nucleoside analogs AZT, 3TC and ddI. The study found a significant increase in the phosphorylation of 3TC and AZT in the presence of HU.
One of the most interesting studies that supports the use of HU in combination with the nucleotide analogs adefovir and PMPA (both made by Gilead Sciences) was presented at the 2nd International Workshop on HIV Drug Resistance and Treatment Strategies, which took place in Lake Maggiori, Italy, just before the Geneva Conference. In the study (see abstract 3 of the Resistance Workshop), researchers from Stanford University Center for AIDS Research measured the IC50 of adefovir, PMPA, and ddI, alone and in the presence of HU. (The IC50 is the drug concentration needed to reduce HIV activity in cell culture by half. This value goes up as the virus becomes more drug-resistant; a reduction means the drug is becoming more effective.) They measured the IC50 of these compounds for a wild-type laboratory isolate of HIV, and for two clinical ddI-resistant isolates known to be multidrug resistant. The concentrations of HU used in the study were low and correspond to what would be clinically tolerated in the body.
Researchers compared the IC50 values for these three drugs alone with the IC50 values when HU was added. The results were as follows: when HU was added to ddI, a seven-fold reduction on ddI IC50 occurred with the wild-type HIV. A 17-fold and a 22-fold reduction were observed in the comparatively high ddI IC50 for the two multidrug-resistant isolates. With adefovir, the reductions were eight-fold (with the wild-type isolate), four-fold and five-fold (with the multidrug-resistant isolates). The HU-associated reductions for PMPA were 26-fold with the wild-type and 6-fold and 35-fold with the multidrug-resistant HIV strains.
The Stanford lab results support the design of clinical studies using adefovir or PMPA in combination with HU. They also confirm results from previous clinical studies of ddI/HU, in which suppression of the virus continues despite the presence of ddI-resistant HIV mutations. The Center for AIDS Research at Stanford University is currently enrolling patients whose HIV has specific reverse transcriptase and protease gene mutations in a trial of a quadruple-drug combination with adefovir, efavirenz, ddI and HU (see Treatment Issues, June 1998).
Steven Miles, from the UCLA Care Center, presented the results of an open-label retrospective study (Geneva abstract 12205) of 18 heavily treated patients with advanced HIV (median CD4 count of 63 and median viral load of 427,325 copies/ml). They had all received at least 8 weeks of therapy with d4T/3TC/HU. These patients had clinical and genotypic resistance to d4T and 3TC. Dr. Miles reported a median peak 1.7 log (98%) viral load reduction at week 8 on these patients. A median weight gain of 4.3 kg was observed. Severe bone marrow suppression was common: four patients required blood transfusions, five patients required erythropoietin to increase red blood cell counts, and four others received G-CSF to increase their population of neutrophils (a type of infection-fighting white blood cell). The average length of time with at least a .7 log reduction was 17 weeks. The benefits of this HU regimen seem transient in advanced patients with a history of poor response to other regimens.
Despite the temporary nature of the response, large number of adverse effects and widespread skepticism, Dr. Miles is very positive about the results. He stresses that it is unheard of to get such a large viral load reduction in persons similar to this cohort, whose HIV contained a large number of mutations associated with drug resistance. As for the side effects, Dr. Miles thinks that many of these can be attributed to other drugs the patients were or had been taking.
New virologic and immunologic data were presented on 12 of the 57 patients in the original ddI versus ddI/HU study conducted by RIGHT. These 12 individuals have now been followed for an average of 2.5 years in the double-combination therapy. Researchers have named members of this cohort the "Panda" patients because of their limited number. They started with an average viral load of 51,795 copies/ml. Their viral load decreased gradually, and at week 122 the average remains at 254 copies/ml for the eight patients with measurable HIV (using the quantification limit of 200 copies/ml), with no rebound despite the presence of ddI-resistant virus. The average increase in CD4 counts has been only 40 cells.
Most importantly, Dr. Lori reported that 6 of the 12 trial participants exhibited HIV-specific CD4 cell responses. These responses have not been found in other patients whose viral load is fully suppressed with HAART, except those treated during initial, or primary, HIV infection. (The "Panda" patients started treatment during chronic stable infection, not during the initial period just after contracting HIV.) In fact, one of the main subjects of the Conference was the lack of such responses, and potential mechanisms to induce them.
It took a long time for the U.S. National Institutes of Health AIDS Clinical Trials Group to enroll its ddI/HU trial. Results from this study were presented at the ACTG meeting held August 1-4, 1998. This clinical trial compared ddI/HU for six months versus ddI alone for three months, with HU added for the second three months. Complete results were not available at press time, but our sources indicate that very good results were achieved in the combination arms with both the 1,000 mg and 1,500 mg daily doses of HU. The 1,500 mg arm fared slightly better, but as expected there was more toxicity associated with this high dose. Viral load reductions in the ddI/HU arms were similar in both naïve and experienced patients. This was not the case in the ddI monotherapy arm, where experienced patients did much worse than naïve patients. This is consistent with the results from other studies showing ddI-resistant virus is still sensitive to ddI in the presence of HU.
ACTG-307 participants will have the opportunity to roll over to a new trial in which they will choose to either add a new agent or stay in the ddI/HU combination. The ACTG is also planning rollover studies from several other ACTG trials that will include HU-containing regimens. Regimens that contain hydroxyurea plus adefovir, PMPA or d4T/3TC also are under ACTG consideration.
Robert Murphy, from Northwestern University in Chicago, is leading a clinical trial comparing ddI/d4T/efavirenz to ddI/d4T/efavirenz/HU. Participants can be experienced or naïve. Those in the HU arm will start with ddI/d4T/efavirenz for four weeks to raise the CD4 counts, before adding HU to the combination. This trial, which is funded by Bristol-Myers Squibb (BMS), is due to start next month in sites across the U.S., France and Canada.
Contrary to the usual new drug development process, clinical research of HU as an HIV therapy has proceeded before an optimal dose has been determined. No dose escalation study has been performed, essentially because, until recently, the drug was only available in the 500 mg size. This June, BMS started producing HU in 200 mg, 300 mg and 400 mg pills. These new pills will allow for further flexibility in dosing quantities and schedules. However, they are significantly more expensive.
Dr. Lori has a new study (RIGHT 702) to look at dosing amounts and intervals, specifically when used with ddI/d4T. The objective of this trial is to determine what the smallest daily dose of HU can be and still be effective. It will also provide information as to whether taking HU once a day is as good as twice or three times daily. The study has nine arms, with all patients on ddI and d4T at dosages related to their weight. There will be three total daily amounts of HU: 1,200 mg (once a day, 600 mg twice a day or 400 mg three times a day); 800-900 mg (800 mg once a day, 400 mg twice a day or 300 mg three times a day); and 600 mg (once a day, 300 mg twice a day or 200 mg three times a day). Patients are required to have a viral load of less than 50,000 copies/ml. The cohort will be 75% treatment-naïve and 25% drug-experienced.
This study will provide information that is crucial to physicians who are either already prescribing HU to some of their patients, or are planning to do so in the near future. Based on current data, most doctors usually prescribe HU at 500 mg twice a day. One of the first studies done by Julio Montaner in Canada suggested that a daily dose of 500 mg of HU had a lesser impact on viral load reduction than the 1,000 mg daily dose. Doses above 1,200 mg a day, on the other hand, generally have a greater incidence of side effects.