Important note: Information in this article was accurate in September 2002. The state of the art may have changed since the publication date.
There were several abstracts at the meeting on the controversial topic of when to start therapy in asymptomatic individuals. There appears to be a consensus that therapy should be initiated based primarily on CD4 count rather than viral load, and that the CD4 count should be between 200 and 350 cells/mm3. However, the optimal CD4 count and timing within this range remains unclear.
Concurrent with the meeting, two studies and the most recent International AIDS Society–USA recommendations for the initiation of therapy were published in the peer-reviewed literature (discussed below). These papers provided additional, though somewhat contradictory, infor-mation. Opravil and colleagues from the Swiss HIV Cohort Study found that initiation of therapy at CD4 counts >350 cells/mm3 improved survival compared to matched controls who delayed initiation of therapy for at least 12 months [AIDS 2002 Jul 5;16(10):1371-81]. However, the proportion of patients who developed toxicity and/or required a change in therapy was substantial. Interestingly, the IAS-USA guidelines, which had previously recommended starting therapy when the CD4 count was <350 cells/mm3, now recommend initiating therapy when the CD4 count is <200 cells/mm3 [JAMA 2002 Jul 10;288(2):222-35]. Thus, in spite of the Opravil paper, there are currently no recommendations to start therapy in asymptomatic persons with CD4 counts >350 cells/mm3. It should be noted that this does not pertain to patients who are symptomatic, who should initiate therapy regardless of CD4 count.
A study by Kaplan and colleagues confirmed previous studies and was consistent with initiating therapy at CD4 counts <350 cells/mm3 [Abstract TuPeB4661; Adult and Adolescent Spectrum of HIV Disease study of the Centers for Disease Control and Prevention (CDC)]. This study included 2,478 persons who initiated HAART and had at least one CD4 count within 12 months of starting therapy. Clinical disease progression was assessed according to baseline CD4 count and viral load at initiation of HAART. Compared with patients with CD4 counts >350 cells/mm3, patients with CD4 counts 200-349 cells/mm3 and viral load <55,000 c/mL did not have a statistically significant different risk of clinical disease progression.
The largest study to date on the topic was presented at the meeting and concomitantly published in The Lancet. In an observational study that assessed data from 13 HIV cohort studies, Chene and colleagues assessed the prognosis of 12,574 antiretroviral-naïve patients who started potent antiretroviral therapy [Abstract TuOrB1140; Lancet 2002 Jul 13;360(9327):119-29]. The median baseline CD4 count was 250 cells/mm3, and the median baseline viral load was 4.9 log10; there were 24,310 person-years of follow-up. The authors assessed 80 different risk strata in determining the probability of progression to AIDS and/or death. Baseline CD4 count was the most important predictor of disease progression, but baseline viral load >100,000 c/mL, age >50, and a history of injection drug use were also associated with an increased risk of disease progression. The risk of clinical disease progression was greatest among those with baseline CD4 count <100 cells/mm3, followed by those with CD4 count 100-199 cells/mm3. Although individuals initiating therapy at CD4 >350 cells/mm3 had a slower rate of disease progression than those who started at 200-350 cells/mm3, the difference was approximately 2% and was not statistically significant. Drug toxicity was not accounted for in this analysis. The authors did not make any specific suggestions regarding the CD4 count at which to initiate therapy, but provided data on the risk of disease progression according to several different clinical categories. These categories were based on CD4 count, viral load, and clinical factors, such as a history of stage C disease. Although not a controlled clinical trial (to date, none have been published on this issue), this is probably the most definitive study because of its large sample size.
A study by Brooks and colleagues assessed the durability of virologic response according to the baseline CD4 count [Abstract TuOrB1141; Adult and Adolescent Spectrum of HIV Disease Working Group of the CDC]. Of the 583 persons studied, 525 (90%) had a virologic response. The authors assessed the 525 responders and found that a baseline CD4 count <200 cells/mm3 was associated with a significantly less durable virological response than a baseline CD4 >350 cells/mm3. Persons starting at 200-349 vs >350 cells/mm3 did not have a statistically significant difference in durable virologic suppression. This study was limited by the relatively small number of participants, the median observation time of 12 months, and the lack of data on prior antiretroviral therapy, adherence, and viral resistance.
In a study by Peterson and co-workers, the risk of clinical disease progression was assessed according to adherence to therapy and CD4 count prior to starting therapy [Abstract TuPeB4664]. There were 1,211 patients who started combination antiretroviral therapy between 1996 and 2001, with 3,583 person-years of follow-up. Adherence was defined as an adherence score of at least 90%. Rates of clinical disease progression after starting therapy were higher among individuals with lower baseline CD4 counts. Adherence of <90% was also associated with an increased risk of disease progression. Among adherent patients who started therapy at >200 CD4 cells/mm3, initial CD4 count did not predict subsequent disease progression. In a multivariate model, independent predictors of disease progression on HAART included a prior opportunistic infection, low CD4 count, non-adherence, and psychiatric illness. Adherence did not affect disease progression among persons with baseline CD4 count >350 or <50 cells/mm3.
In a study from the British Columbia Center for Excellence in HIV/AIDS, Wood and colleagues found that factors associated with an increased risk of disease progression included a CD4 cell count <200 cells/mm3 prior to initiating HAART [Abstract TuPeB4672]. However, adherence to therapy and physician experience in the care of HIV infected patients also had a significant impact on survival. A total of 1,416 patients were included in this study. Adherent patients were defined as those who received at least 75% of their medications; physicians were considered experienced if they had previously treated at least five HIV infected patients.
In conclusion, disease progression may be slower in persons who start at CD4 >350 cells/mm3 than at 200-350 cells/mm3, but the difference is small. Adverse events from medications usually tip the balance in favor of not starting at >350 cells/mm3. Factors such as viral load >100,000 c/mL, age >50, prior AIDS-defining conditions, and a history of injection drug use increase the risk of disease progression, and may influence the decision to start earlier rather than later. Durable virologic suppression is more likely in persons who initiate therapy at CD4 >200 cells/mm3, and as long as patients are >90% adherent, further stratification of CD4 above this level may not be necessary. Adherence decreases the risk of disease progression, particularly among persons who start at CD4 <200 cells/mm3; receiving care from a more experienced clinician also improves outcome.
This topic dovetails nicely with the issue of when to start therapy, since factors such as adherence, durable virologic suppression, and the class of therapy selected all influence the response to therapy. There were several important studies reported in Barcelona that addressed the issue of what to start with in chronically infected, antiretroviral-naïve patients.
In one of the more important studies presented, JA Bartlett reported preliminary (48 week) data on the CLASS study, which is comparing the nucleoside combination of abacavir + 3TC plus either efavirenz (EFV), amprenavir/ritonavir (APV/RTV) (1200/200 mg qd) or d4T in treatment-naïve patients with CD4 >50 cells/mm3 and viral load >5,000 c/mL [Abstract TuOrB1189]. Patients were allowed to make in-class substitutions if necessary (e.g., due to toxicity). In an intention-to-treat (missing=failure) analysis at 48 weeks, the proportion of patients achieving viral load <50 c/mL was 76%, 59%, and 62% among those receiving EFV, APV/r, and d4T, respectively (p=0.047 for EFV vs APV/r or d4T comparisons). Among those with baseline viral load >100,000 c/mL, 77% of the EFV group achieved viral load <50 c/mL, compared with 53% of the protease inhibitor group and 55% of the triple nucleoside group. Abacavir hypersensitivity occurred in 6% of patients. Thus, while all three regimens performed well, the EFV-based regimen appeared to be superior. An additional 48 months of follow-up is planned and will include analysis of sequencing strategies, in which patients failing the triple-NRTI regimen switch to EFV/APV/r, and patients failing the EFV or APV/r regimens switch to AZT/ddI plus APVr or EFV, respectively.
Forty-eight-week data were also presented for Gilead 903, an assessment of the efficacy and safety of tenofovir DF vs d4T in combination with 3TC and EFV in naïve patients [Staszewski S, et al. Abstract LbOr17]. In an intention-to-treat (missing=failure) analysis after 48 weeks of follow-up, 82% of the patients in the tenofovir arm (n=299) had a viral load <50 c/mL, compared with 81% of the patients in the d4T arm (n=301)(P=NS). There were no differences in efficacy by baseline viral load. The drop out rate was low, and rates of adverse events and drug discontinuation were similar in both arms, though random cholesterol and triglycerides were significantly higher in d4T treated patients (p<0.001), and there was a suggestion of increased peripheral neuropathy in the d4T arm as well. The impressive efficacy seen in both arms of the study confirms the potency of EFV in previously naïve patients, and supports the use of tenofovir as first-line therapy. While this was not specifically designed as a study of once-daily therapy, the implications are obvious, since both tenofovir and 3TC can be administered once a day, and d4T will also be a once daily drug when Zerit XR is approved.
ACTG 384 was also presented at the late-breaker session [Robbins G, et al. Abstract LbOr20a; Shafer R, et al. Abstract LbOr20b]. This trial, which included 980 patients among 6 study arms, was designed to answer three questions:
The first component presented assessed the following four strategies (n=155 in each arm; nelfinavir and efavirenz were blinded):
ddI/d4T/EFV
AZT/3TC/NFV
ddI/d4T/NFV
AZT/3TC/EFV
AZT/3TC/NFV
The primary endpoint was time to regimen failure of the second 3-drug regimen. Regimen failure included virologic failure, toxicity, or premature discontinuation. Secondary endpoints included time to first regimen failure or virologic failure. The median baseline CD4 was 278 cells/mm3 and the median baseline viral load was 4.9 log10 c/mL. Median follow-up time was 28 months.
Compared with ddI/d4T, starting with AZT/3TC showed a strong trend toward delay in sequential regimen failure, and significantly delayed first regimen failure, when combined with EFV but not NFV. Compared with NFV, starting with EFV showed a strong trend to delayed sequential regimen failure and significantly delayed first regimen and virologic failure, when combined with AZT/3TC, but not ddI/d4T.
In comparing the strategy of using a 4-drug regimen vs two sequential 3-drug regimens, the following regimens were studied:
ddI/d4T/NFV/EFV
AZT/3TC/NFV/EFV
ddI/d4T/EFV
ddI/d4T/NFV
AZT/3TC/EFV
AZT/3TC/NFV
The primary endpoint was time to regimen failure of a single 4-drug regimen or two sequential 3-drug regimens. There was no significant difference in the primary endpoint between single 4-drug regimens and two sequential 3-drug regimens. Four-drug regimens significantly delayed first regimen failure and virologic failure compared with NFV-containing 3-drug regimens. Four-drug regimens also significantly delayed first regimen and virologic failure compared to EFV-based 3-drug regimens that contained ddI/d4T, but not AZT/3TC.
It should be noted that the drop-out rate in this study was high and that less than one-third of the endpoints were due to virologic failure. Thus, in an intent-to-treat analysis, the differences among the treatment arms may be more likely to reflect tolerability and toxicity than actual potency. On-treatment analyses will presumably be included in future presentations of the data.
Robert Hogg presented data from another study from the British Columbia Center for Excellence in HIV/AIDS, on the risk of clinical disease progression among patients who initiated therapy with a protease inhibitor-based vs NNRTI-based regimen [Abstract TuOrB1142]. Of note, 96% of the persons in the NNRTI group received nevirapine, and the protease inhibitor group received primarily indinavir (IDV). Very few people received either EFV or dual protease inhibitors; 431 persons (39%) received an NNRTI-based regimen. The primary endpoint was death. Although patients who received an NNRTI-based regimen had a slightly lower risk of progression to death (relative risk: 0.79), this was not statistically significant. Factors associated with an increased risk of death were baseline CD4 count <200 cells/mm3, increased age, and intermittent antiretroviral therapy.
In a study that assessed the durability of first-line HAART regimens, NNRTI-based HAART was associated with a more durable response and significantly lower rates of adverse events and changes in therapy than PI-based regimens [Moreno A, et al. Abstract TuPeB4673]. The PI-based regimens included either IDV or NFV, and the NNRTI-based regimens included either NVP or EFV. In this study, 256 anti-retroviral-naïve patients started HAART after April 1998 and were followed up for at least one year. In a multivariate model that controlled for prior AIDS-defining illness, baseline CD4 count, and viral load, the relative risk of disease progression was 2.69 for persons on PI-based therapy compared with patients on NNRTI-based therapy (p<0.01).
In a retrospective study from Arribas and colleagues, EFV-based HAART regimens were more effective than PI-based regimens in patients who initiated therapy at CD4 counts <100 cells/mm3 [Abstract TuPeB4444]. There were 92 patients taking EFV-based regimens and 218 taking PIs; the median baseline CD4 counts were 34 and 39 cells/mm3 in the two arms, respectively, and median baseline viral load was 350,500 and 254,000 c/mL. At 24 months, 69% of patients taking EFV had a viral load of <400 c/mL by ITT analysis (non-completer or change other than simplification=failure) vs 45% for patients taking PIs (p<0.05). While many clinicians have tended to prescribe protease inhibitors in patients with advanced HIV disease, this study supports the use of EFV in such patients. It should be noted, however, that 94% of the PI-treated patients in this study were taking single PIs, while the current trend is to use boosted PIs, especially in patients with high viral loads and/or low CD4 counts.
Fätkenheuer presented data from a similar study that assessed lopinavir/ritonavir (LPV/r, Kaletra) in 86 patients who initiated HAART at a CD4 count <100 cells/mm3, viral load >100,000 c/mL, or after having had an opportunistic infection [Abstract TuPeB4447]. The median baseline CD4 count was 90 cells/mm3, and the median baseline viral load was 321,561 c/mL. After six months of therapy, 13 of 15 (87%) who remained on the LPV/r-based regimen had a viral load <400 c/mL. In a study of 68 patients who had baseline CD4 count <200 cells/mm3 and who initiated HAART with an NNRTI based regimen, 84% of patients had viral load <200 c/mL at 12 months (N=49; intention to treat analysis) [Lonca M, et al. Abstract TuPeB4457].
Mallolas from Spain compared two HAART sequences in naïve patients with chronic infection [Abstract TuOrB1186]. In this multi-center, randomized, open-label study, patients with CD4 <500 cells/mm3 were randomized to receive either ddI/d4T/NFV, switching to AZT/3TC/NVP (group A; n=73), or ddI/d4T/NVP, switching to AZT/3TC/NFV (group B; n=85). The indication for switching regimens was virologic failure. The primary study endpoints were time to failure of the second regimen, and the proportion of patients with viral load <200 c/mL. After 18 months of follow-up, 65% of patients had failed (virologic failure or change in regimen). However, there were no statistically significant differences in the mean increase in CD4 count, or in the proportion of patients who had achieved viral load <200 c/mL, virologic failure, or had an adverse event. Thus, the two strategies had comparable efficacy and tolerability, though the study was under-powered to truly claim equivalence.
In a retrospective cohort study, Pulido and colleagues assessed virological, immunological, and clinical outcomes among patients with CD4 counts <100 cells/mm3 who were treated with either an EFV-based or protease inhibitor-based HAART regimen [Abstract TuOrB1187]. The median baseline CD4 count was 34 cells/mm3 in the EFV group and 39 cells/mm3 in the PI group; median baseline viral loads were 351,000 and 254,000 c/mL, respectively. The mortality rate was 4% in the EFV group and 2% in the PI group (p=0.5). However, the time to failure (defined as either 2 consecutive viral loads >400 c/mL or a change in regimen) was delayed in the EFV group (p=0.001). There was no difference in the increase in CD4 count, except at 24 months, at which time it was higher in persons treated with EFV. Although this summary pertains primarily to the treatment of chronically-infected persons, there was one study of short-course antiretroviral therapy (SCART) in primary HIV infection [Fidler SJ, et al. Abstract TuOrB1185]. Although therapy is currently recommended for persons with primary HIV infection, the duration of such therapy is unknown. These investigators from the United Kingdom identified 60 persons with primary HIV infection, of whom 52 agreed to take SCART. They all achieved a viral load of <50 c/mL after a median duration of treatment of 12 weeks (range: 4 to 32 weeks). Up to 48 weeks after discontinuation of therapy, the average viral load was lower than at baseline. Four of 52 discontinued therapy due to toxicity; 6/52 progressed to a CD4 count <300 cells/mm3, at which point therapy was resumed. SCART was also associated with preservation of HIV-specific immune responses.
In conclusion, for initial therapy in the antiretroviral-naïve patient, EFV-based regimens appear to be superior to the protease inhibitor-based regimens tested in these studies. A 3-drug regimen is sufficient, particularly if it contains EFV. If protease inhibitors are used, dual or boosted therapy now appears to be preferred over a single agent. Tenofovir appears to be a reasonable component of initial therapy, though it remains to be seen whether there will be differences in toxicity and/or resistance in tenofovir- compared to thymidine analog-based combinations. Longer-term follow-up, during which both efficacy and tolerability are assessed, will be helpful in determining the optimal regimens for initial therapy.
20020901
JH20020901
©1997,1998,1999,2000,2001, 2002. The Johns Hopkins University AIDS Service, Division of Infectious Diseases. Permission to use and reproduce portions of this newsletter is hereby granted provided that author and publication are fully credited and both copyright and permission notice appear with reprinted material. Inquiries may be directed to Sharon McAvinue, Managing Editor. Website: Johns Hopkins AIDS Service.
AEGiS is made possible through unrestricted grants from Boehringer Ingelheim, iMetrikus, Inc., the National Library of Medicine, and donations from users like you. Always watch for outdated information. This article first appeared in 2002. This material is designed to support, not replace, the relationship that exists between you and your doctor.
AEGiS presents published material, reprinted with permission and neither endorses nor opposes any material. All information contained on this website, including information relating to health conditions, products, and treatments, is for informational purposes only. It is often presented in summary or aggregate form. It is not meant to be a substitute for the advice provided by your own physician or other medical professionals. Always discuss treatment options with a doctor who specializes in treating HIV.
Copyright ©1980, 2002. AEGiS. All materials appearing on AEGiS are protected by copyright as a collective work or compilation under U.S. copyright and other laws and are the property of AEGiS, or the party credited as the provider of the content.
