icon-folder.gif   Conference Reports for NATAP  
 
  11th Annual Retrocirus Conference
(CROI-Conference on Retroviruses and Opportunistic Infections)
San Francisco
Feb 8-11, 2004
Back grey_arrow_rt.gif
 
 
 
Kaletra Once Daily (48 Week results)
 
 
  Tenofovir 144 Week Results; Triple & Quad Nuke Studies: The treatment of antiretroviral naive patients
 
Reported for NATAP by Dr Graeme Moyle, Chelsea & Westminster Hospital, London, UK
Editing by Jules Levin

 
Although they have been substantial advances in antiretroviral therapy over the last five years, there is still room for improvement. Patient surveys indicate that there is a desire to move from twice a day to once daily dosing and to reducing the number of pills per dose. However, these surveys also indicate that individuals taking antiretroviral therapy are not willing to increase the risk of drug-related adverse events to achieve improved convenience. A second issue which remains unanswered regarding therapy of antiretroviral naive individuals is the timing of starting therapy. Several cohort studies at the conference investigated this issue but mainly provided information on the risks of starting therapy with very advanced disease (CD4 <100cells/cu mm). This report will focus on the results of clinical trials reported during the conference.
 
Once vs Twice Daily Kaletra, 48 Week Results: Abbott Study 418
 
This open-label comparative study randomised 190 treatment naive individuals to receive lopinavir/r once daily (n = 115) or twice-daily (n = 75) with a backbone of tenofovir/emtricitabine (FTC). The patients were well matched the baseline characteristics with a median CD4 count of 4.6-4.8 log copies/ml and a median CD4 count of 214-232 cells/cu mm. Pharmacokinetic data and a previous small pilot study had supported the idea that lopinavir/r can be given once daily but that the pharmacokinetics are such that the ‘inhibitory quotient’ (IQ) for the drug (an estimate of the plasma trough drug exposure in relation to virus susceptibility) is reduced from approximately 94 with twice daily dosing to around 49 with once daily. Although no standardized target for IQ is established, these data would suggest that once daily dosing of lopinavir/r is likely to be only suitable to treatment naive or protease inhibitor naive individuals.
 
Data from the study through week 48 were presented. 19% of patients in the once daily and 25% of patients in the twice daily dosing group discontinued therapy before week 48. In the once daily dosing group 12 % of patients discontinued due to an adverse event compared with 5% in the twice-daily group who discontinued for this reason. Patients in the twice-daily dosing group were more likely to discontinue the study due to non-adherence, consent withdrawal or being lost to follow-up. Only one individual (in the twice-daily group) was defined as a virological failure. The majority of adverse events did not differ between groups, However moderate to severe diarrhea (>6 stools/day) was significantly more common in the once daily dosing group (16 %) compared with the twice-daily group (5%)(p = 0.04).
 
Efficacy differences were not observed. At 48 weeks the proportion of individuals with a viral load less than 50 copies by intention to treat missing = failure analysis was 70 % in the once daily and 64 % in the twice-daily group (Difference 6.4%, 95% CI -7.3%, 20.1%). For observed data, the response rates to <50 copies/ml were 90 % for once daily and 85 % for twice-daily. Impressive CD4 count rises of 185 and 188 cells/cu mm were observed for the once daily and twice-daily groups, respectively. Modest changes in fasting total cholesterol, a rise of 27 milligrams per dl in both groups, were observed. More substantial increases in fasting triglycerides were reported with a rise of 82 mg/dl in the once daily and 76 mg/dl in the twice-daily group. These rises are smaller than those reported in the Abbott 863 study which used a backbone of stavudine (d4T) plus lamivudine (3TC). However, one must be cautious at making comparisons as samples in the 863 study were not taken in a fasted state.
 
Given the increase in gastrointestinal side-effects observed with the once daily administration of lopinavir/r and the relatively high number of pills per intake for once daily dosing of this agent (six lopinavir/r capsules plus NRTIs), it is unclear that there will be a high demand amongst patients for this regimen. An improved formulation of lopinavir/r is planned which would reduce the intake to four tablets per day. If this new formulations is better tolerated from gastrointestinal standpoint the once daily dosing of lopinavir/r will begin to demand more attention.
 
Tenofovir vs d4T plus Efavrienz, 144 Week Results: Gilead Study 903
 
Although the study was not presented at the Retroviruses conference, data were released to the press in the days before. This randomized placebo-controlled study provides comparative information of tenofovir (n = 299) versus d4T (n = 301) in treatment naive individuals in combination with 3TC and efavirenz. Details of results through 144 weeks were reported. Only 18% of individuals in the tenofovir group and 21 % of individuals from the d4T group discontinued from the study. In particular, discontinuation is due to adverse events has occurred in only 1% of tenofovir and 3% of d4T treated patients.
 
At 144 weeks the proportion of patients with a viral load less than 50 copies/ml by intention to treat missing = failure analysis is 73% in the tenofovir and 69% in the d4T arms; these differences are not statistically significant. Important differences between the drugs emerged when adverse events were considered. In the group receiving d4T fasting triglycerides rose by 134 mg/dl over 48 weeks whereas in the tenofovir treated group they had only risen by 1 mg/dl. Changes for fasting total cholesterol were 58 mg/dl in the d4T group and 30 mg/dl in the tenofovir group. These differences in lipid changes were highly statistically significant at weeks 48, 96 and 144.
 
Substantial differences in the numbers of patients diagnosed by the investigators as having clinical lipodystrophy were also observed. Given that the study is placebo-controlled through 144 weeks these result should represent unbiased reports from investigators. The proportion of patients diagnosed with lipodystrophy in the d4T group at weeks 48, 96 and 144 rose steadily from 4% to 12% to 19%, respectively. In the tenofovir group only a few lipodystrophy events were reported throughout the study with just 1% diagnosed as having lipodystrophy at weeks 48 and 96 and 3% at week 144. DEXA scan data at weeks 96 and 144 showed highly significant differences in the total limb fat between tenofovir and d4T groups, with the amount of limb fat rising in the tenofovir group from 7.9 to 8.7 kg between these two-time points but falling from 5 to 4.4 kg in the d4T group. ‘Normal’ limb fat is around 8kg or more. Taken together these data suggest that the combination of tenofovir/3TC/efavirenz maybe protective against the development of peripheral lipoatrophy.
 
Changes in bone mineral density in the spine and hip were also determined by DEXA scanning. The bone mineral density was observed to decline in both groups during the first 48 weeks of study but then appeared to stabilize through week 144. Initial declines in bone mineral density were greater in the tenofovir recipients as compared with the d4T treated patients. These statistically significant differences persisted through week 144. These modest changes in bone mineral density were not associated with differences in fracture rate. In fact, numerically more fractures were reported in the d4T group and fractures reported were consistently associated with significant physical trauma (such as car accidents).
 
Assessments of renal function were also performed during the study. Grade one elevations in serum creatinine were observed in 11 tenofovir recipients and 7 d4T recipients. Grade two events were observed in 2 tenofovir but no d4T recipients and grade three events were reported in 2 d4T recipients but no tenofovir recipients. No patients experienced grade 4 shifts in serum creatinine and, of note only one new renal event, a grade one creatinine elevation, was reported in the tenofovir group after week 96.
 
Editorial note from Jules Levin: failure due to viral failure in the tenofovir regimen in this study was low. But in a portion of patients who do experience viral failure the K65R mutation emerges. At this point we are not certain about the treatment implications if patients develop the K65R. This question deserves attention and further research to understand treatment approaches if the K65R mutation is present.
 
Nucleoside/nucleotide only therapy: 33% viral failure (ABC/3TC/TDF); 90% viral failure (ddI/3TC/TDF); interim encouraging data on 4 nukes (AZT/3TC/abacavir) plus tenofovir
 
During the course of 2003 and several studies were reported which suggested triple nucleoside regimens are inferior to regimens which contained two NRTIs and efavirenz. In ACTG5095 patients initiating with AZT/3TC/ABC (as Trizivir tablets) were less likely to achieve and sustain virological suppression in initial treatment relative to regimens that contained two or three NRTIs plus efavirenz. A subsequent study, ESS 30009 which evaluated the once daily triple NRTI regimen of ABC/3TC/tenofovir was stopped prematurely due to low rates of virological response and a significantly inferior performance relative to the control arm of ABC/3TC/efavirenz. Two further studies reported at this conference report poor performance of triple NRTI regimens as initial therapy.
 
A small French study reported results from a non-comparative study of ABC/3TC/tenofovir in initial therapy, the same regimen that was studied in the comparative ESS 30009 trial. Following the presentation of ESS30009 this trial was prematurely interrupted for an unplanned interim analysis and was subsequently terminated. In the French study virological failure was defined as patients who never achieved undetectable VL <400 copies/mL or experienced a rebound in viral load >0.7 log10 copies/mL from nadir. The study included 38 antiretroviral naïve patients with a medium baseline CD4 count of 221 cells/mm3 (range 61 to 348) and median VL of 4.9 log10 copies/mL (range 2.0 to 5.9 log).
 
Virological failure were observed in 12/36 (33%) patients. The author noted that none of the eight patients with a baseline plasma viral load <4 log10 copies/mL experienced virological failure; all 12 failures arose from the group of 28 patients who commenced with higher viral load (p = 0.03). Suppression of viral load to <50 copies/mL was observed in 12/34 (35%) patients and 17/26 (65%) patients at month 3 and 6, respectively.
 
Genotypes available from 12 patients with virological failures between month 3 and month 6, found the presence of both K65R and M184V mutations in 11/12 with the remaining sample having just the M184V mutation.
 
Pharmacokinetic sampling performed after one month of therapy indicated that 32/37 patients had plasma Cmin exposure thought to be adequate for all the 3 drugs. In the remaining five patients tenofovir Cmin values below the limits of assay detection were present with three patients having undetectable ABC and three patients undetectable 3TC plasma concentrations. The data are consistent with the results of ESS3009 both in terms of the high rates of virological failure and the frequency in which both the K65R and M184V mutations are observed. The pharmacokinetic data suggest that whilst in occasional patients poor adherence or insufficient drug exposure are associated with failure, the majority of patients who experienced virological failure on this triple NRTI regimen have adequate plasma drug exposures. This suggests that an inadequate genetic barrier is the likely cause of insufficient response and rapid resistance emergence with this regimen.
 
A group of US investigators (Jemsek et al) reported a result of a 24-week pilot study to evaluate the potency and safety of a once-daily triple NRTI regimen of didanosine EC (ddI) 250 mg, 3TC 300 mg, and tenofovir 300 mg in treatment-naïve patients. The authors defined patients as responders if a ≥2 log10 reduction in viral load from baseline was observed by week 12 of treatment. Genotyping was performed at baseline with genotyping and phenotyping (Phenosense GT) performed at week 24 or the time of study discontinuation.
 
The study included 24 patients of whom 22 provided evaluable data. At baseline the median viral load was 4.91 (range 3.31 to 6.02) copies/mL and the CD4 count 133 (range 4 to 475) cells/mm3; 38% of the participants had a viral load >100,000 copies/mL, and 58% had CD4 < 200 cells/mm3 at baseline. Of the 22 evaluable patients, 2 completed the 24 weeks of study while 20 patients (91%) discontinued treatment early (after a median of 16 weeks [range 7 to 23]). These discontinuations were all secondary to a suboptimal anti-viral response. At week 12, the change from baseline in viral load ranged from –2.66 to 0.73 log with a median decline of just -0.61 copies/mL (n = 20). At week 24 or ET (n=22) mdian HIV RNA decline was –0.49 (range –2.66 to 0.89 log). 20% of patients had an increase in viral load at week 12. 25% of patients showed 0 to 0.5 log decrease at week 12. About 22% of patients had >1 log decrease in viral load. And 30% showed >0.5 to 1.0 log decrease.
 
Resistance testing (n = 20) showed M184I/V in all patients (100%) with 10 of these patients (50%) also having K65R. Of 19 patients who had phenotype results available:
--19 samples showed reduced susceptibility to 3TC
--6 samples showed reduced susceptibility to ddI (>1.7 x wt); most had both M184V & K65R
--3 samples showed reduced susceptibility to abacavir (>4.5 x wt)
--0 samples showed reduced susceptibility to TDF (>1.4 x wt), AZT (>1.9 x wt), or d4T (>1.7 x wt); presence of 184V increases susceptibility to these agents
These results are remarkably similar to those reported in ESS30009 and indicate that this regimen should not be considered as sole therapy.
 
Investigations into the K65R mutations indicate that this mutation leads to reduced binding and incorporation of all the available NRTIs, although this defect is greatest for tenofovir, ddI and abacavir. The mutation, however, also leads to reduced excision of AZT from the growing viral DNA chain. This reduced excision is seen to a lesser extent with d4T and abacavir. As a result viruses with the K65R mutation alone remain susceptible to d4T and abacavir and appear to have increased susceptibility to AZT. In vitro, addition of the K65R to viruses containing AZT resistance associated mutations (TAMs) resulted in increases in AZT susceptibility. Additionally, the presence of TAMs reduced the impact of K65R on the susceptibility of viruses to tenofovir but also to abacavir and ddI. These data suggest that there is "bidirectional phenotypic antagonism" between K65R and TAMs. This hypothesis, that the presence of thymidine analogue mutations may prevent or reduce the appearance of K65R was supported by interrogation of the Virco database. In particular the presence of K65R was uncommon when mutations 41L, 67N, 210 W and 215Y/F were present.
 
A single arm cohort study (called COL40263; R Elion et al) in 123 treatment naive individuals further investigated this issue by evaluating the efficacy and resistance outcomes in individuals treated with AZT/3TC/ABC/tenofovir. Following the result of the ESS30009 study a planned interim analysis was performed using data from the 88 individuals who had received eight weeks therapy and 54 individuals who had completed 24 weeks.
 
The median entry viral load was 5.1 log copies/ml (range 4.2-6.3) and CD4 count 226 cells/cu mm (range of 19-816). For the 88 subjects who had received at least eight weeks of therapy 11 (13%) prematurely discontinued the study, the majority (9%) having been lost a follow-up. Observed analysis of the 54 individuals who had completed 24 weeks of therapy indicated that 79% of individuals have achieved a viral load < 400 copies/ml with 67% of individuals having a viral load < 50 copies/ml. For the 35 individuals who had commenced with viral load > 100,000 copies/ml the response rate by observed analysis at week 24 was 74% of individuals < 400 copies/ml and 60% less than 50 copies/ml. These results are a striking improvement on those observed in ESS3009. Using the early virological nonresponse criteria from the ESS30009 study (< 2 log dropped in baseline viral load by week 8 and/or a 1 log increase in viral load from nadir after week 8) 76% of patients in COL40263 were responders compared with 51 percent of individuals in ESS30009.
 
Eight individuals in COL40263 had evaluable genotypes of whom 1/8 (13%) had the K65R. mutation, 2/8 and one or more Tams, 3/8 Tams plus M184V with the remaining two samples having wild type virus. These data support further investigation of this quadruple NRTI regimen and indicate that the presence of thymidine analogue may partially protect against the emergence of K65R.