ICAAC, Part II: HIV Infection
Conference Coverage
Editor’s note: The following summaries conclude the coverage of a selection of papers from those presented at the Fourth Conference on Retroviruses and Opportunistic Infections held on January 22-26, 1997, in Washington, DC. It is important to recognize that many of these summaries are extracted only from the published abstract, and it is possible that some of the material presented at the conference may have differed.
The abstracts, as well as other information presented at the conference, are available on the internet at www.retroconference.org.scd
In December 1995, a protocol calling for use of "triple therapy" was introduced into a Los Angeles private practice, managed care specialty HIV clinic, and by October 1996 this was being administered to almost 300 of 482 patients. This followed the previous year when 1218 of the patients received "double therapy" with AZT plus 3TC, with most of the remainder presumably receiving monotherapy. The number of hospitalization days per patient per month during the period July 1994-June 1996 fell from 3.36 to 1.28, the rate of days in skilled nursing facilities decreased from 136.8 to 48.0, and the mean number of patients receiving hospital care for CMV disease decreased 90% with new cases absent in compliant patients. The number of patients receiving G-CSF and erythropoietin declined, respectively, by 70% and 74%. The frequency of consultations with other subspecialists also declined dramatically. Thus, it would appear that the appropriate use of the newer antiretrovirals is associated with considerable cost savings. (Abstract #262)
A marked decline in hospitalization rates was also reported from a group in New York City in association, however, with a three-fold increase in costs of antiretroviral therapy. (Abstract #264) A French study reported a significant monetary savings from decreased hospitalization attributed to protease inhibitor therapy. (Abstract #LB12)
Data continues to accumulate demonstrating that the quality of care received by HIV-infected patients is directly related to the experience of their physicians in dealing with individuals with this disease. A Canadian study found, in a multivariate analysis, that highly experienced physicians (defined as having seen > 20 HIV-infected patients) were more likely to provide complex supportive services (home-based and supportive care), primary clinic services (vaccinations, TB and anergy testing), and complex services (antiretroviral therapy and prophylaxis and treatment of opportunistic infections [OI]) than were lesser experienced physicians. (Abstract #189)
Similarly, a survey of 1166 providers yielding 345 evaluable surveys (42% response rate) found that experienced physicians (defined as those with > 20 HIV-infected patients) were more likely than lesser experienced physicians to intervene early with antiretroviral therapy, to be more aggressive in their management in asymptomatic patients, to treat OIs according to national guidelines, and to treat "difficult-to-treat" conditions (those for which there is no FDA-approved or standard therapy). (Abstract #255)
One reason for an HIV-infected patient to be cared for by a dedicated, experienced, and knowledgeable physician is the sheer volume of information, constantly evolving, which must be mastered in order to provide optimal care. There is no more confusing area than the management of antiretroviral therapy itself, in part because of the large range of pharmacokinetic drug interactions that may occur with use of HIV protease inhibitors.
A retrospective analysis of 165 HIV-infected patients found that 82 (50%) had at least one potential drug interaction at the time protease inhibitor therapy was initiated, with approximately one-third being potentially life-threatening. Only 22 (19.8%) of the total potential interactions were recognized at the time of protease inhibitor therapy initiation, while an additional 12 were subsequently recognized. (Abstract #610)
The observations that CCR-5, a chemokine receptor, also serves as an HIV-1 coreceptor and that a mutation in the CCR-5 gene (CCR-5D32) is associated with apparent absolute resistance of HIV-1 infection in homozygotes are exciting. Whether heterozygosity (CCR-5D32/CCR-5) is protective has remained a matter of debate. Thus, several studies found that CCR-5D32 heterozygosity has little or no effect on the rate of progression of HIV-1 infection and is not associated with lower plasma HIV-1 RNA levels, slower decline of CD4 counts, or SI phenotype evolution. (Abstracts #22, 37, 438) These conclusions are not, however, unanimous. (Abstracts #440, 738)
The coreceptor story is becoming more complex than initially perceived. Examination of sequential isolates from three HIV-1-infected subjects found that primary isolates obtained early in the course of infection primarily used CCR-5 as preferred coreceptor, while isolates from individuals with disease progression became able to also use CCR3, CCR2b, and fusin. Fusin use was seen only with primary isolates with SI phenotype. This phenotypic switch (NSI to SI) associated with fusin usage was also associated with loss of sensitivity to chemokines and decrease in CD4 count. (Abstract #18)
The increase in plasma HIV RNA frequently observed after inflammatory stimuli continues to be extensively examined. Retrospective analysis of a cohort of 2081 patients indicates that most opportunistic infections increase the risk of death independently of CD4 level, thus supporting the hypothesis that such infections, by enhancing HIV replication, cause progression of HIV disease. (Abstract #366)
A retrospective analysis found that nine of 10 patients had transiently increased plasma HIV RNA concentrations in association with the development of PCP, as well as a decrease in CD4 count. (Abstract #1) Observations of 40 subjects during a variety of common acute infections (such as sinusitis and urinary testing) demonstrated a mean increase of plasma HIV RNA over baseline of 1.5 log10 and a mean 29% decrease in CD4 count. (Abstract #116)
Tetanus toxoid vaccination of a single HIV-infected patient with a CD4 count of 350 was associated with a 7.5-fold increase in viral load eight days later, with a return to baseline by seven weeks, and was associated with an apparent shift of viral quasispecies from lymph nodes to plasma. However, no new major quasispecies were detected nor was virus with typical SI-associated sequences in the env gene. (Abstract #118)
Administration of tetanus toxoid or of pneumococcal vaccine to HIV-infected subjects on stable antiretroviral therapy did not induce significant virologic or immunologic changes. While emergence of new quasispecies and SI type virus in plasma was found in some vaccinees, such changes were also seen in unvaccinated controls, suggesting that these events were independent of immune activation occurring as the result of vaccination. (Abstract #428) This latter observation is consistent with that of another group that reported repeated phenotype switching over two years in patients with low CD4 counts. (Abstract #450)
Examination of lymph nodes from AIDS patients found remarkably high levels of HIV expression in macrophages from nodes coinfected with opportunistic pathogens such as MAC and P. carinii. This is consistent with the observations of increased plasma viral load during OIs and suggests that tissue macrophages may be an important source of that virus. (Abstract #386)
One of the mechanisms by which OIs or vaccination may lead to increased HIV viral load is by induction of pro-inflammatory cytokines, such as TNF-a, which may in turn upregulate HIV production. Interestingly, TNF-a also upregulates in vitro surface expression of fusin, an HIV-1 coreceptor. (Abstract #402) IL-10 is an "immunosuppressive" cytokine that, among other things, decreases TNF-a production. A single infusion of IL-10 to patients with CD4 counts between 100 and 200 caused a transient decrease in plasma HIV RNA. This was associated with an inhibition of production of pro-inflammatory cytokines, as well as a decrease in expression of fusin on peripheral blood mononuclear cells. (Abstract #37)
The role of the putative Th1 to Th2 "switch" in the evolution of HIV disease is becoming muddier. It was reported that plasma TNF-a, IL-6, and IL-10 concentrations were similar in HIV-infected subjects and in non-HIV-infected controls. TNF-a and IL-6 concentrations rose after endotoxin infusion to a much greater degree in HIV-infected than in non-HIV-infected subjects, while the reverse was true for IL-10. IL-12 concentration was higher pre-infusion in HIV-infected patients than in controls and remained so after endotoxin infusion. This in vivo study does not support the possibility that a deficiency of IL-12 and elevation of IL-10 account for the putative Th1 to Th2 switch and progressive immunodeficiency. (Abstract #82)
The risk of HIV transmission by sexual contact is increased in the presence of genital ulcer disease. HIV RNA was detected in 23 (96%) of 24 genital HSV-2 lesions in 12 patients. (Abstract #27) Individuals with advanced disease are more likely to transmit HIV infection than those at earlier stages. This may be related to the finding that the quantities of HIV RNA in paired plasma and vaginal secretion samples were found to be correlated (R=0.6; P < 0.001), and both were negatively correlated with CD4 count. (Abstract #25)
Previous reports of mostly individual cases had indicated that viral load decreases after splenectomy for thrombocytopenia in HIV-infected patients. Splenectomy in eight such patients was followed by a greater than three-fold decrease in plasma HIV RNA in four; no patient had a significant increase in viral load. (Abstract #1114)
HIV env-specific cytotoxic T-lymphocyte (CTL) activity was found in seven (35%) of 20 health care workers with parenteral HIV exposure (none of whom became infected) but in none of 20 exposed to HIV-negative blood and in none of seven healthy blood donors. CTL activity was found in only one (14%) of seven who took AZT preventive therapy but in six (46%) of 13 who did not take AZT. This suggests that AZT prevents viral replication in exposed individuals, and this, in turn, prevents the development of CTL against the virus. (Abstract #28)
A wide variety of studies of combination regimens were reported. Among those with apparently beneficial effects were d4T plus ddI plus hydroxyurea (Abstracts #549, #550), d4T plus ddI (Abstracts #552, 553, 554), and 3TC plus d4T (Abstracts #557, LB4).
AZT, 3TC, and either indinavir or ritonavir were administered to 24 patients within 90 days of primary HIV-1 infection. All compliant patients developed undetectable (< 100 copies/mL) plasma HIV RNA and negative viral cultures. Examination of mononuclear cells in semen failed to demonstrate HIV mRNA, and no evidence of HIV could be found in mucosal lymphoid tissue obtained by sigmoidoscopic biopsy. A reduction in env and gag specific antibody titers occurred. This suggests the possibility of viral elimination when aggressive antiretroviral therapy is initiated soon after HIV infection. Whether "cure" is possible, however, remains to be seen. (Abstract #LB8)
In ACTG 193A, 1313 patients with CD4 counts less than 50, 82% of whom were antiretroviral experienced, were randomized to one of the following regimens: alternating (monthly) AZT and ddI, AZT plus ddC plus ddI, or AZT plus ddI plus nevirapine. By pairwise comparison, the three-drug combination was associated with significantly longer survival than AZT plus alternating ddI (P = 0.012) or AZT plus ddC (P = 0.009), but not when compared with AZT plus ddI (P = 0.25). (Abstract #LB6)
A total of 1200 patients with CD4 counts of 200-500 were randomized to receive AZT alone or in combination with one of three doses (200, 300, or 400 mg tid) of delavirdine, a non-nucleoside reverse transcriptase inhibitor currently being considered for approval in the United States. Analysis of viral isolates from 190 randomly selected subjects revealed that 14% harbored resistant virus after six months of AZT monotherapy, while none among those receiving AZT together with the 400 mg tid dose of delavirdine developed AZT resistance. The delavirdine IC50 of 12% of these isolates was greater than the achieved median trough plasma concentration of this drug. Delavirdine/AZT therapy of AZT-experienced patients led to resensitization of virus to AZT, with isolates demonstrating a mean 85-fold decrease in AZT IC50 during the first six months of therapy. (Abstract #12)
In 10 patients randomized to receive either AZT plus 3TC or AZT plus d4T, the decrease in CSF HIV RNA was similar in both groups and was similar to the decrease observed in serum. The serum to CSF concentration ratios of AZT, d4T, and 3TC were 0.69, 0.28, and 0.12, respectively. (Abstract #LB5) The ability of the newer antiretrovirals to achieve adequate concentrations in the central nervous system to prevent the development of HIV encephalopathy is of concern. Nevirapine administration to six HIV-infected patients resulted in CSF concentrations of the drug equivalent to the free unbound fraction in plasma (45 + 5%). (Abstract #567)
The protease inhibitors were a major focus of a number of investigations. One of the problems with this class of drugs is patient intolerance. A retrospective analysis in France found that patients were four times more likely to discontinue saquinavir therapy and 2.8 times more likely to discontinue ritonavir therapy relative to indinavir therapy. (Abstract #197) Gradual institution of ritonavir, reaching a full dose of 600 mg bid, was associated with improved tolerancenone of 57 patients withdrew due to drug-related symptoms during the first year of study. (Abstract #193)
The addition of saquinavir to the antiretroviral regimens of patients with advanced HIV disease who were heavily pretreated with nucleoside analogs resulted in a decrease in plasma HIV RNA which, however, returned to baseline by week 12. (Abstract #192) Administration of AZT, ddI, 3TC, and saquinavir to 10 antiretroviral naïve subjects led to undetectable viral load in all 10. In addition, HIV-1-infectious titers in lymph node mononuclear cells decreased with a mean half-life of 5.8 ± 0.7 days, while the HIV RNA half-life decrease in these cells was 7.4 ± 0.9 days; the absolute decrease in lymph node viral load averaged more than 3.5 log10 in the first four months. There was an associated decreased apoptosis. (Abstract #235)
Patients (n = 279) with CD4 counts less than 50 were randomized to receive either indinavir or ritonavir in addition to two nucleoside analogs by the Picasso Trial Group in Brussels. The one month analysis found similar increases in CD4 count (+ 34 and + 52, respectively) and changes in viral load (-1.1 log10 and -1.4 log10) in the two groups. However, six patients assigned ritonavir crossed over to indinavir because of drug intolerance, while there were no crossovers in the other direction. (Abstract #196)
Six (18%) of 33 patients with CD4 counts between 50 and 350 given ritonavir alone for 14 days with d4T and ddI then added stopped therapy because of adverse events. Seventeen patients reaching 12 weeks of study had a mean increase in CD4 count of 138 and a mean decrease in viral load of 2.14 log10, with plasma HIV RNA being undetectable (< 200 copies/mL) in 12 (70%). (Abstract #246)
Indinavir, d4T, and 3TC were administered to 145 patients previously treated with nucleoside analogs for a mean of 36 months. The mean CD4 count increased from 84 ± 68 at baseline to 142 ± 92 at eight weeks of therapy, and mean viral load decreased by 1.62 log10, with 31 (43%) of 72 having indetectable plasma HIV RNA. (Abstract #247)
Nelfinavir is under consideration for approval by the U.S. Food and Drug Administration. It appears to be potent and well tolerated, with the most common side effect being diarrhea, which appears to occur in at least 20% of patients. Several studies were presented that demonstrated benefit from nelfinavir therapy. In these studies, nelfinavir was given alone or was combined with d4T or AZT plus 3TC, as well as with d4T plus ddI. (Abstract #240) In addition, nelfinavir was administered, together with d4T and ddI to 22 subjects who had not previously received d4T, ddI, or any protease inhibitor. The median CD4 increase in eight patients reaching eight weeks of study was 218, and their median decrease in plasma HIV RNA was 21og10. Seventeen (77%) of the 22 reported diarrhea. (Abstract #241)
Many studies examined the feasibility of administration of protease inhibitors in combination. However, it is important to first examine potential antiviral interactions in vitro. The antiretroviral activity of indinavir and saquinavir in combination was studied in vitro against a panel of HIV-1 clinical isolates. The drugs demonstrated activity ranging from synergy to antagonism against a pan-susceptible HIV isolate. However, antagonism was noted at all drug concentration combinations against an AZT-resistant isolate as well as against an isolate resistant to multiple reverse transcriptase inhibitors. This suggests the need for caution in the coadministration of these drugs. (Abstract #158)
Pharmacokinetic interactions, such as those between saquinavir and ritonavir, must be taken into account when considering the coadministration of drugs, including protease inhibitors. Either ritonavir or saquinavir was added to triple drug regimens that already included one or the other of these protease inhibitors (with subjects having added the inhibitor they were not already receiving) and which the 18 patients had been receiving for more than five months. After a mean duration of six months of quadruple drug therapy including both saquinavir and ritonavir, there was a mean increase of CD4 cells from 101 (5-354) to 163 (6-495), as well as a mean decrease in plasma HIV RNA of 0.85 log10. (Abstract #244)
A single-dose pharmacokinetic study found that coadministration of indinavir increased the area under the curve (AUC) and and maximum concentration (Cmax) of saquinavir five- to eight-fold and that no serious adverse effects were noted. (Abstract #608) The coadministration of nevirapine reduced saquinavir Cmax by 29% and AUC by 27%, while the nevirapine AUC decreased insignificantly by 3%. (Abstract #614)
Thirty-two patients with CD4 counts less than 250 and viral loads greater than 5000 or disease progression despite triple drug therapy with two nucleoside analog reverse transcriptase inhibitors (RTI) and one protease inhibitor were treated with two RTI plus both saquinavir and ritonavir. At 12 weeks, the mean increase in CD4 count was 72, and the mean decrease in viral load was 2.12 log10 with 96.6% having undetectable (< 400 copies/mL) plasma HIV RNA. Two (6.2%) of the 32 subjects withdrew because of adverse side effects. (Abstract #245)
An example of the exploitation of favorable pharmacokinetic interactions in the evaluation of antiretroviral therapies is the planned approach to the study of ABT-378. ABT-378 is a "second-generation" protease inhibitor that retains activity against protease resistant to ritonavir. (Abstracts #208, 212) While ABT-378 has limited bioavailability in some animals, including cynomolgus monkeys, the metabolism of ABT-378 by rat and human liver microsomes is markedly inhibited by ritonavir. As a result, these drugs will be examined in combination in human trials. (Abstracts #210, 211)
Also of interest is the use of protease inhibitors and non-nucleoside reverse transcriptase inhibitors in combination, although pharmacokinetic interactions are of concern. Twenty-one patients with CD4 counts less than 50 with extensive prior antiretroviral therapy were treated with indinavir, 3TC, and nevirapine given 200 mg daily for 14 days. No unexpected toxicity was observed. There was a greater than 2 log10 decrease in plasma HIV RNA at eight weeks, with five of 10 subjects completing that duration of treatment having viral loads less than 500 copies/mL. (Abstract #234)
The major problem with protease inhibitor therapy is the development of resistance, a problem that is likely to be increasing rapidly in the United States. Sequencing of the pol gene of HIV-1 isolates from 72 protease-inhibitor (PRI) naïve patients in rural Iowa found that 26% had mutations associated with resistance to PRIs; multiple mutations were present in 21%. Mutations associated with resistance to reverse transcriptase inhibitors (RTI) were present in 25%; 3% of RTI naïve patients were infected with isolates with such mutations. Studies such as this suggest that optimal antiretroviral management requires sequencing data. (Abstract #9)
The predominant mutation found in 55 subjects participating in a Phase II trial of monotherapy with nelfinavir was a change from aspartic acid (D) to asparagine (N) at position 30 (D30N) of the proteinase gene. This stable mutation was also found in subjects receiving D4T with nelfinavir. Observation for up to 44 weeks of study demonstrated that, with the exception of the rarely observed L90M change, mutations associated with therapy with other protease inhibitors were not found. The presence of the D30N mutation was associated with reduced susceptibility to nelfinavir. Strains with high level resistance to nelfinavir, however, maintained full susceptibility to indinavir, saquinavir, ritonavir, and VX-478. (Abstract #10)
Resistance to ABT-378 induced in vitro was associated with the sequential appearance of the following mutations: I84V -> L10F > M46I -> T91S -> V32I + I47V. (Abstract #209)
A major risk for the development of resistance to protease inhibitors is poor patient compliance. Of 202 patients who had been receiving antiretroviral therapy for at least six months at an HIV clinic in Baltimore, and despite the fact that 71% were receiving monotherapy, only 60% reported taking more than 80% of their antiretroviral regimen during the previous week. The more medications of all sorts taken, the less the compliance. Knowledge concerning the reason for taking antiretrovirals was associated with improved compliance. This study does not deal specifically with compliance with protease inhibitor therapy, but illustrates the major problem we have in assuring this important aspect of patient care, especially in patients receiving these potent, but often poorly tolerated agents. (Abstract #251)
The administration of potent combination therapies including protease inhibitors is commonly associated with marked increases in CD4 and CD8 lymphocyte counts. A number of studies sought to further characterize these cells by both functional studies and evaluation of surface markers.
Saquinavir and ritonavir were administered together to patients with CD4 counts of 100-500 with a resultant more than 4 log10 decrease in plasma HIV RNA after 12 weeks. At four weeks, CD4 lymphocyte count increased by a mean of approximately 200 cells and in vitro proliferative responses to PHA, p24Ag, and tetanus toxoid developed or improved in 27 (66%) of 41, 14 (34%) of 41, and three (38%) of eight patients, respectively. The majority of subjects demonstrated significantly improved secretion of IL-2, IL-12, and IL-10 after in vitro stimulation. (Abstract #33)
Ritonavir, AZT, and ddC were administered to 20 antiretroviral naïve patients with a mean CD4 count of 170 ± 73 with a resultant mean 3 log10 plasma RNA decrease and 60% CD4 increase. Phenotypic analyses demonstrated a peak increase (+ 35%) of activated memory cells (CD45RO+ RA-CD25+) at day 15 followed at 4-6 months by a significant decrease in activation (CD25, CD38, DR) markers, a 25% increase in CD4+CD28+ cells, and, at six months, an increase of naïve (CD45RA+62L+) CD4 cells in 4-8 subjects. These results suggest that the initial rise in CD4 count in patients on this regimen was the result of mobilization or proliferation of a mature memory cell population, followed by decreased T cell activation. The finding of an increase in the naïve cell population in some patients after six months of therapy is encouraging and suggests the possibility of significant immune reconstitution. (Abstract #34)
Other studies, however, have not found increases in naïve cells. For instance, administration of ritonavir to 12 patients with advanced HIV infection led to modest increases in in vitro lymphocyte production of IL-2 and IFN-g, mostly as the result of improved response by CD8, rather than CD4 lymphocytes. The increase in CD4 cells was the result of an increased number of memory (CD4+/CD45RO+/CD45RA-) cells without any increase in true naïve cells. Increased expression of CD28 on both CD4 and CD8 lymphocytes was found, as was decreased levels of the activation marker CD38. (Abstract #537)
In another study, the marked rise in CD4 counts seen in 24 patients given AZT and 3TC with or without indinavir (ACTG 320) was almost completely of the memory phenotype; no increase in naïve (CD4+/CD45RA+/CD62L+) cells was observed. (Abstract 249a)
Administration of rhIL-2 to some patients results in marked increases in CD4 counts. This was confirmed in one study, but, unfortunately, no improvement in lymphoproliferative response to mitogens (PHA, anti-CD3), to recall antigens (TT, SK/SD), or to nine T-helper epitopes from HIV gag and env was detected. (Abstract #35) It was reported that IL-2 administration was associated with an increase in both memory and naïve CD4 T cells over 12 months. (Abstract #369) The increase in naïve cells suggests the possibility of true immunologic reconstitution.
A number of clinical observations, some of which were outlined in the section on Opportunistic Infections (Infect Dis Alert 1997;16:81-86), indicate that many patients receiving potent antiretroviral therapy undergo clinically significant immunologic improvement. Among these are the improvement or clearance of opportunistic infections after institution of such therapy. For instance, three patients with previously intractable OIs had clinical improvement (a case of PML) or clinical and microbiological resolution (cryptosporidiosis, chronic parvovirus B19 infection), and one with intractable wasting gained 40 pounds after initiation of protease-inhibitor therapy. (Abstract #355) PML improved markedly in one patient, and KS improved in another after initiation of multiple-drug therapy that included a protease inhibitor. (Abstract #357) Chronic cryptosporidiosis resolved in 12 of 13 patients with chronic cryptosporidiosis or microsporidiosis after initiation of "triple therapy" with complete disappearance of the pathogen from the stool in most. (Abstract #358)
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content