This report includes reviews of two published articles about 1592U89 in the May 1997 issue of Antimicrobial Agents and Chemotherapy (pgs 1082-1098). Discussions below include results from studies of both in vitro and human cross-resistance relating to 1592.
Summary Overview. Some of the information contained in this report may be highly technical so first is a brief and more simple summary. This report discusses in vitro (laboratory or test tube) research of 1592U89 resistance conducted by Glaxo Wellcome. It also reports in vitro results of studies for cross-resistance between 1592U89 and other nucleosides: AZT, d4T, 3TC, ddI, and ddC. The results of in vitro experiments do not necessarily predict the outcome when real people take these drugs. Our understanding of the effect of mutations, resistance and cross-resistance is still in the developmental stages. However, the in vitro results reported below indicate that there is some degree of cross-resistance between 1592 and other nucleosides.
If you have extensive prior experience with nucleosides your response may be less from 1592 compared to if you had never before taken a nucleoside. There will probably be varying degrees of responses by individuals. If you only have low level resistance to other nucleosides and/or limited prior experience you may have only low level or no cross-resistance to 1592. Although, an individual's response to 1592 may relate to the amount of resistance and/or experience they've had with other nucleosides, it is very difficult to predict now how one will respond to 1592.
In the near future, results from human studies which are now in progress will be available. These results will give us more information about 1592 resistance and cross-resistance from real people taking the drug in combination with other nucleosides including AZT/3TC. With this additional information you should be better equipped to decide how you may want to use 1592. NATAP will report this information when it's available.
If an individual is totally cross-resistant to 1592 but uses it anyway, it may be more harmful to them by causing more resistance. If an individual has limited sensitivity to 1592 and merely adds it onto a current therapy resistance is likely to develop quickly. For example, if you start therapy with nelfinavir +1592 and are not responsive to 1592, you are risking the possibility that this might be like taking nelfinavir monotherapy.
New HIV Treatments. Aside from 1592 a number of new opportunities for treatment are in development. Next year, if all goes according to plan, several new drugs will be available: DMP-266, PMEA, 141W94. It is expected that Abbott will soon begin human studies of their new protease inhibitor, ABT-378, which will be combined with a small amount of ritonavir to increase and sustain 378's concentration in the blood; once-a-day dosing may be explored. Perhaps more importantly, in preliminary animal studies ritonavir resistant virus was suppressed. In addition, a number of preliminary double-protease studies have already started. They are listed in the current issue of our newsletter, NATAP Reports. If the data is good, it is expected that these double-protease combinations will be studied to explore their potential for suppressing protease resistant virus.
Human trials for MKC-442, a new NNRTI, are expected to begin by the Fall of 1997, and will include studies designed to treat individuals who've developed protease inhibitor resistance. Furthermore, the first human study of PMPA has just started. If all proceeds well, it is expected that larger scale trials will begin in early 1998. Preliminary data suggests both 141W94 and PNU-140690 (a new protease inhibitor in early study) may possess unique resistance profiles which may limit cross-resistance from prior protease inhibitor resistance.
Finally, it is expected that other pharmaceutical industry and ACTG sponsored studies will be started exploring novel drug combinations aimed at suppressing protease inhibitor resistant virus. The problem that many individuals have already become resistant to protease inhibitors is high on the radar screen for researchers and will be addressed.
Results of 1592U89 Cross-Resistance Studies
Cross-resistance data from an in vitro experiment was reported, exploring the potential for cross-resistance between 1592 and other nucleosides. The data also describes the effects on resistance to nucleosides of mutant viruses with specific mutations created in the laboratory. Initially a wildtype HIV-1 strain, HXB2, and an AZT resistant virus were passaged in increasing concentrations of 1592, a classic way to create resistance for lab experiments. For the AZT-resistant virus, which already possessed slightly less susceptibility to 1592, a 4-5 fold reduction in susceptibility to 1592 was observed after passage 8, where the only mutation occurring was 184V in the background of mutations 67N, 70R, 215F, and 219Q; and indications were that the AZT mutations were contributing to resistance.
Human study. Richard Harrigan, of Glaxo Wellcome, has reported that in a human study of 1592, after 12 weeks of monotherapy, mutations were observed primarily at position 184, with some individuals also having mutations at positions 65, 74 and 115. The combination of AZT/1592 (participants were randomized to 1592 monotherapy or 1592/AZT) appeared to prevent emergence of these mutations in the trial which lasted 12 weeks. The combination of AZT/1592 resulted in the emergence of the 65R mutation, but appeared to delay or prevent the emergence of the 184 mutation. In the experiment discussed below, the 184 mutation was the first to emerge followed by the 65, 74 and 115 after passing the virus in increasing concentrations of 1592.
For this type of experiment a specific virus type is used. In this case, as described above, the wild-type virus used was the HXB2. Mutant viruses with single, double and triple mutations were created and their susceptibility to nucleosides were compared to the susceptibility of a wild-type virus to the same drugs.
In the following table, the data associated with wild-type virus are the amount of drug (1592, ddI, ddC, d4T, AZT, 3TC) needed to suppress the wild-type virus as measured by the IC50. The IC50 represents the amount of drug necessary to suppress viral replication of the wild-type virus (the 1st line in the table) by 50% and is a standard measure used in this type of experiment. The other numbers are expressed in a form indicating the decrease in susceptibility or sensitivity, but reflect the increase in drug amount necessary to suppress the mutant viruses created for this experiment by 50% (IC50). Except where indicated all changes are a decrease in susceptibility compared to wildtype.
The Fold Decrease in Susceptibility to 1592 as Measured by Changes in IC50 when Compared to wild-type virus without mutations 1592U89 ddI ddC 3TC d4T AZT wild type virus 5.3 3.9 0.54 0.71 0.87 0.10 65R decreased 3.10-fold 2.69 1.98 5.26 1.62 1.0** 74V decreased 3.73 2.79 3.88 2.42 1.08 2.0 115F decreased 2.06 * * * 1.08 2.0 184V decreased 2.92 1.35 3.11 >70.4 2.56 3.0 65R/74V decreased 3.56 3.43 3.51 12.39 *** 1.0** 65R/184V decreased 6.88 5.32 7.29 >70.4 1.05 1.0** 65R/74V/184V decreased 10.15 5.48 6.66 >70.4 1.10 1.0** 74V/184V decreased 8.50 3.71 4.81 >70.4 1.62 ND 74V/115F decreased 3.44 3.48 1.64 1.26 1.14 3.0 115F/184V decreased 6.78 4.82 2.59 >70.4 2.10 2.0 74V/115F/184V decreased 10.99 5.05 3.88 >70.4 1.96 1.0** * The virus containing the 115F single mutation was not resistant to ddC or ddI, but this virus was 1.54-fold more susceptible to 3TC. However, statistically all of the data in this experiment are subject to accepted deviations from the results used to reach the conclusions of fold changes. For some of the data the deviations are more than for other data.
** None of these virus mutants were cross-resistant to AZT; no change in amount of drug necessary to suppress virus replication by 50% (IC50).
*** The virus containing the double mutant 65R/74V was not resistant to d4T.
> means decrease of greater than.
ND - not done.
A single mutation caused a decrease in susceptibility or increase in resistance to 1592 of < 4-fold. This includes the 184V single mutant virus which was only 3-fold resistant to 1592. The viruses containing double mutations had 3.5-8.5 fold resistance, which may or may not be considered significant. The two viruses containing a triple mutation had 10-11 fold resistance to 1592. 10 fold resistance to a NRTI also may or may not be significant, but some individuals may have a higher level of cross-resistance. Since the 184 mutation appears to cause resistance to both 3TC and 1592, it may be detrimental to combine the two drugs. As noted earlier in the report, an AZT-resistant virus with a 184 mutation was 4-5 fold less sensitive to 1592.
Several resistance researchers I've consulted with say it is difficult to predict the response to 1592 in humans based on this in vitro data; that is, for an individual who has extensive nucleoside experience and may have the same mutations described above which are associated with in vitro resistance to 1592 it is difficult to predict how they will respond to 1592. They said some individuals with 6-8 fold or 10 fold resistance may be responsive to 1592, some may not be as responsive, and for some the response will be a matter of degree.
For AZT and d4T, 3-fold cross-resistance was observed with the mutant clones (viruses) used in this experiment. Maximum reductions in susceptibility to ddC were observed with the 65R/184V double mutant and the 65R/74V/184V triple mutant. The 115F alone or in combination with other mutations did not cause reductions in susceptibility to ddC but appeared to suppress resistance due to the 74V mutation.
With ddI, however, the 115F mutation alone did not cause resistance but in combination with other mutations, particularly with the 184V, marked increases in resistance were observed.
The 65R mutation caused partial resistance to 3TC as did the 65R/74V, while of course any virus mutant containing the 184V mutation was resistant to 3TC.
All three mutations (65R, 74V, 184V) have been observed both in vitro and in the clinic during ddI or ddC treatment.
The authors of one article concluded, based on their resistance profiles, that it may not be optimal to use these NRTIs together in combination therapy. The authors of the other article suggested from their in vitro experiments that 1592 was synergistic with AZT, nevirapine and 141W94; and, additive and/or some synergy was observed in combinations of 1592 with ddI, ddC, 3TC, and d4T. Although different types of experiments were used the investigators reached differing conclusions indicating to me their uncertainty about nucleoside cross-resistance with 1592. It appears that additional drugs were not selected to be explored for synergy in this experiment.
Further in vitro studies are ongoing examining 1592 and 3TC to evaluate resistance and synergy. Potentially, the AZT/1592 combination may be beneficial because prior experiments reported by Harrigan and discussed above have suggested that the 65R, 74V and 184V mutations may suppress AZT resistance. The authors suggested that 1592 may suppress or slow the development of AZT resistance.
1592U89 Compassionate Use/Expanded Access. Researchers have been talking lately about multi-drug resistant virus (MDR). They have identified a mutation at position 151 as causing multi-drug resistance. Although its a minority of individuals who may have such a MDR virus, I think it's been said that only 3.5% of individuals may have this MDR virus, some individuals may in fact have this virus. Due to the presence of a MDR virus or just due to extensive prior nucleoside experience some individuals may have more limited response to 1592, as we've already stated .
It's worth repeating, in consideration of limited amount of available resistance information, if an individual does not need to use 1592 and has other options it may be in their interest to wait until there is more information. After we learn more about 1592 and its use in individuals with varying degrees of prior nucleoside experience it will be easier to judge how you might be able to optimize its use for yourself.
Each drug is a precious commodity that should not be used inappropriately but should be preserved. Proper adherence or compliance to taking a drug or regimen is also very crucial to success with your current therapy and treatments you may use in the future. NATAP's Protease Inhibitor Users Guide -Revised July 1997 is a compliance manual available for you and your organization's staff and clients. The 36-page booklet contains adherence information for 5 protease inhibitor therapies including drug interactions, side effects, dietary, hydration and storage requirements, patient assistance programs, dosing, viral load monitoring, and more.
1592 CNS Penetration in Animals. 1592U89 drug levels were explored in the cat brain and the monkey CSF. The investigators concluded that 1592 penetration in both the cat brain and the CSF of the monkey was similar to that of AZT. The penetration of 1592 into the cat brain was compared to that of AZT after intraperitoneal injection. After .5 hr, 1 hr, and 2 hrs post-dose the 1592 drug levels were 6%, 6.7% and 7.9% of the 1592 concentration in the plasma, respectively. With AZT, after .5 hr, and 1 hr the AZT drug levels were 5% and 10% of the drug levels in plasma, respectively. But after 2 hrs post-dose the AZT drug level in all three animals was below quantification. However, the plasma concentration of AZT by 2 hrs post-dose was much lower so it appears that it is not possible to actually measure the brain/plasma ratio at 2 hrs. Therefore, it is not possible to compare the penetration of 1592 and AZT at two hours.
Using doses of 25, 70 and 210 mg/kg twice daily of 1592 in a 28-day toxicology study in monkeys the mean CSF/plasma ratio after 1 hr post-dose was 0.17. Similar ratios were reported averaging 0.21 in a different monkey species dosed with AZT.