icon-folder.gif   Conference Reports for NATAP  
  10th Conference on Retroviruses and Opportunistic Infections
Boston, Mass, Feb 10-14, 2003
Back grey_arrow_rt.gif
STI and immunology summary CROI
Mark Dybul, MD National Institute of Health
  Salvage therapy
Deeks (abstract 640) and colleagues continued their pioneering strategies in this field. 20 patients with a mean viral load of 3.9 log and CD4 of 336 receiving PI-based salvage therapy stopped either the NRTI (5) or the PI (15) component of their therapy. The point here was to see if patients could maintain good clinical outcomes while potentially reducing toxicity. The patients who stopped the PI had a mean log change in plasma HIV RNA of 0.005 log copies/week, a mean decrease in serum triglycerides of 90 mg/dl (P=0.02) and non-HDL cholesterol of 30 mg/dl (P=0.03) at week 12, and stable genotypic and phenotypic mutations. In contrast, the patients who stopped the NRTIs had a sustained increase in viremia of 0.03 log copies/week; 3 of 5 exhibited a delayed loss of M184V that was temporally associated with an increase in viremia.
Edit note: it's unclear how long lipid improvements may last; development of additional PI resistance may stop on NRTI regimen; see additional perspectives on STI/interruption studies presented at retrovirus in the NATAP David Margolis Retrovirus Review of STIs and Interruptions: http://www.natap.org/2003/Retro/day11.htm
The controversy regarding STI in salvage therapy rages on. Lawrence (abstract 67) presented data from the large CPCRA study that was terminated to new enrollment by the DSMB because the study could not demonstrate a virologic or clinical benefit to patients who underwent an interruption. 270 patients with mean viral load 5.0 log and CD4 180 were randomized to undergo a 16 weeks interruption and then resume therapy or to continue therapy. Genotypic resistance results were available for all patients; mean number of new drugs was 3.6 (2.7 active) in the STI arm and 3.8 (2.8 active) in the no STI arm. There were 22 progression of disease or death endpoints in the STI arm versus 12 in the no STI arm; there were 8 deaths in each group. Interestingly, the events occurred after the ARVs had been resumed in the STI arm. However, a significant difference in CD4 T cells remained between the STI and no STI arm to 8 months (47 cells, P<0.001) but not at 12-20 months (31 cells, P=0.11). There was no difference in quality of life between the groups.
Katlama (abstract 68) presented follow-up data to week 48 of the GIGHAART. 63 patients who were randomized to an 8 weeks interruption followed by mega-ARV therapy (median 11 drugs!) or to begin mega-ARV immediately. Despite the fact that the majority of patients did not have a return of drug-susceptible HIV during the relatively short off-ARV period, there remained a statistically significant difference at week 48 in change in viral load at week 48; however the P value changed from 0.013 at week 24 to 0.4 at week 48. In addition, at week 48 the p value for the percent of patients with <400 copies/ml was >0.05.
The difference in results may be due to the duration of the STI period or the large number of drugs used in the GIGHAART study. But it may be important to note the trend in the P values in the GIGHAART study in trying to resolves the clinical benefit of STI in salvage therapy.
Patients doing well on therapy
Less time on therapy

Two studies evaluated treatment interruptions based on CD4 T cell count. Ruiz (abstract 65) presented follow-up data on their study of a single interruption in patients with viral load < 80 copies/ml and CD4 > 500 for 1 year and nadir CD4 > 100. Patients were randomized to continue therapy or to stop. At 48 weeks of interruption, 35/61 (57%) had resumed therapy due to increased viral load > 100,000 c/ml (96%), decreased CD4 to <350 (4%) or both (23%). However, 43% remained off therapy for nearly 1 year. These results are similar to those that have been presented by Joel Gallant. The major predictors for resuming therapy were pre-ART therapy nadir CD4 and viral load. Of note, 41% of patients with interruption experienced minor events and 10% experienced acute retroviral syndrome.
Jean Ananworanich (abstract 64) presented data from their innovative randomized, controlled trial of continuous ARV therapy versus CD4-driven intermittent therapy, versus 7 days on/7 days off therapy. At week 48, none of the 23 patients in the CD4-driven arm had failed by criteria and the mean time off therapy was 67% patients resume therapy when CD4 drops <350 or by >30%; the mean decrease in CD4 T cell count was 178 and one patient had an acute retroviral syndrome. Of note, there was no significant reduction in serum triglyceride or total cholesterol levels between the arms. In the week on/week off arm, more than half of the 26 patients had virologic failure and the DSMB had closed the arm. Anaworanich noted that these patients were significantly different in several ways than the NIAD cohort in which failure has not been seen in pilot studies of 7 days on/7 days off : 1) patients in the is cohort were heavily pre-treated and may have entered with resistance, and 2) perhaps of greatest importance, the regimen used was SQV/RTV which may be less potent that IDV/RTV or EFV used in the NIAID studies. If one is off therapy for 50% of the time, potency matters. In this regard, Gunthard (abstract 639) reported that 8/14 patients, 7/14 of whom were treated with nelfinavir-based regimens, had > 100 copies/ml of plasma HIV RNA at Day 8 of treatment interruption. In contrast, Dybul (abstract 68lb) reported that of 76 patients in NIAID studies, 8 (11%) had > 50 copies/ml (range 62-955) at day 7 of treatment interruption: few patients received nelfinavir and none received SQV.
Two randomized, controlled trials of long cycle intermittent ARV therapy with pre-determined time periods on and off ART (as opposed to CD-driven where the off intervals are determined with each cycle based on the CD4 count). Stephano Vella (abstract 66) presented the largest trial: 273 patients have been randomized to receive continuous ARV therapy or intermittent therapy of 2 cycles of 1 month off and 1 cycle of 2 months off ARV therapy; the period on ARVs is 3 months regardless of the off-drug duration. An interim analysis at 6 months showed that of 136 patients randomized to intermittent therapy, 13% had dropped out versus 4% in the continuous arm. Resistance was seen in 21% of patients in the interruption arm; however, after analysis of resistance prior to interruption, 8/72 (11%) of patients had new resistance. Resistance was seen in both patients treated with PI or NNRTI-based therapy. The most frequent mutations were to NRTIs; of note, nevirapine was stopped 3 days prior to NRTIs and efavirenz was stopped 6 days prior to NRTIs given the longer half-life of NNRTIs. AT the end of the 3rd STI, 92% of patients had plasma HIV RNA < 400 (versus 100% in the continuous arm), but only a 40 patients had achieved this endpoint. There may a trend to decreased grade 3 or 4 toxicity in the interruption arm (19 events in continuous arm versus 8 in intermittent arm), but Vella noted that it was too early to evaluate the effects on toxicity.
Dybul (abstract 68lb) presented data from their randomized, controlled trial of continuous ARV therapy versus 4 weeks off/8 weeks on ARVs. The study, designed for 90 patients, was stopped to new enrollment after 52 patients had enrolled (26 per arm) due to new resistance to NNRTI (efavirenz; nevirapine was excluded) and/or 3TC in 3 of 8 patients in the intermittent arm after 4 to 6 cycles. In addition, 2 of 18 patients receiving a PI developed NRTI mutations during the second cycle off ART. Of the 52 patients who enrolled, 41 (19 intermittent and 22 continuous) had 48 week data. At week 48, there was no significant difference in serum triglyceride, total cholesterol or LDL cholesterol levels. At week 40 there was a significant difference in triglyceride and total cholesterol (P=0.4 for each), but by week 48 that was abrogated. Although the transient decreases may be important, there was no difference at any time in hs-CRP, the most sensitive marker for risk of cardiovascular disease. In addition, there was no difference ALT or AST at any time evaluated.
Auto-immunization and immune control of HIV infection
Bruce Walker set the tone for the potential role of STI for auto-immunization (abstract 164) when he stated that "the further one goes out the more depressing it gets." Longer follow-up in the cohort of patients treated during acute HIV infection has revealed that a significant number of the 14 patients have now failed to maintain control of plasma HIV RNA following multiple STIs. In a Kaplan-Meier evaluation of time to reaching greater than 30,000 copies/ml of plasma HIV RNA, 40% of patients who underwent STI had less than 30,000 copies/ml at 900 days compared to a control cohort of approximately 25%. Although these data may indicate an advantage to STI in acute HIV infection, the loss of control in a reasonable number of patients treated during acute infection when the potential for STI to have the greatest auto-immunization impact may not bode well for auto-immunization in chronic HIV infection.
In fact, there was not a single oral or poster presentation devoted to auto-immunization in chronic HIV infection. Two presentations on studies of long cycle treatment interruptions to reduce drug exposure did provide data on auto-immunization in chronic HIV infection. Stefano Vella (abstract 66) presented data on 3 interruptions (1 month off/3 months on, then 2 months off/3months on) in 136 patients. The mean log HIV rebound was 2.3, 2.1 and 2.2 for the 1st, 2nd and 3rd cycle. In a study of 1 month off/2 months on, Dybul (abstract 68lb), there was no significant difference in mean plasma HIV RNA levels from the 1st to the 2nd (-0.33 log), 1st to the 4th (-0.46 log) cycles off ARVs (P>0.5).
Therapeutic immunization and STI
Harrer (abstract 60) presented a safety and immunogenicity of an MVA vector expressing HIV-1 Nef in 14 patients with plasma HIV RNA < 50 copies/ml and CD4 > 400. Patients were given 3 doses of the immunogen at weeks 0, 4 and 16; 2 weeks following the final immunization, patients interrupted their ARV therapy. Other than local inflammation and systemic effects, the immunizations were well tolerated. Following 3 immunizations, 7/14 patients recognized new CTL epitopes and 3/14 had Nef-specific CD4 t cells. The mean historic pre-ARV plasma HIV RNA level was approximately 20,000 copies/ml; following immunization and STI the mean viral load was 46,000 copies/ml.
In a trial of ALVAC, Tubiana (abstract 61) administered 4 immunizations over 12 weeks in 48 patients receiving ARV therapy with a viral load < 200 copies/ml and CD4 > 400; at week 16 ARV therapy was interrupted. No significant clinical events were recorded. 36 patients with low CD4 proliferation SI (mean 3.3) to p24 antigen prior to immunization had a mean SI of 6.8 at the time of interruption. There was a 3-fold increase in CD8 responses to gag in 7 patients. 38/48 patients had significant increases in plasma viremia following interruption and were required to resume therapy.
In an interesting twist, Levy (abstract 62) presented data from 70 pts who received ARV therapy alone (37 pts) or 4 ALVAC immunizations every 4 weeks followed by 3 cycles of subcutaneous IL-2 every 8 weeks. At week 40 (weeks after final IL-2 dose), all patients stopped ARVs. 15/32 immunized versus 8/33 controls had sustained CD4 proliferation to p24 Antigen at week 36. Time to virologic failure was statistically different between the groups, but only by 13 days (42 versus 29). 2/37 (5%) controls versus 8/33 (24%) remained off ARVs at week 52.
Although therapeutic immunization remains the greatest hope for the future, and it was encouraging that the immunogens used were safe and well tolerated, the virologic results were less than impressive. In fact, there did not seem to be a difference in therapeutic immunization compared to what has been reported for several cycles of STI alone for auto-immunization.