Medical Consultant
|
6th Conference on Retroviruses and Opportunistic InfectionsJanuary 31 - February 4, 1999 - Chicago, IL (PART TWO)
|
|
6th Conference on Retroviruses and Opportunistic InfectionsJanuary 31 - February 4, 1999 - Chicago, IL
|
Paul Blanchard, ATP
Much current debate is centred on the necessity of including a protease inhibitor in triple therapy regimens for "first-line" treatment. Concerns over adherence, tolerability, lipodystrophy and other metabolic complications associated with protease inhibitor use have prompted investigation of the utility of non-PI containing combinations. This approach is often referred to as "protease sparing". It should be recognised, however, that "sparing" can only occur if the protease based combination used second-line is not adversely affected in any way by the prior exposure to the non-PI containing regimen. The effects of such sequencing on long-term durability of response is still unknown.
The 6th Retrovirus conference saw the presentation of data on triple combinations using non-nucleoside reverse transcriptase inhibitors (NNRTIs) with nucleoside analogues as well as triple nucleoside analogue combinations.
The ATLANTIC trial is a multicentre randomised open label study of three different antiretroviral combinations in treatment naive patients. The three arms consist of a triple nucleoside combination of d4T/ddI/3TC, an NNRTI based combination of d4T/ddI and nevirapine (NVP), and a PI based combination of d4T/ddI and indinavir (IND). This is the first time that these three starting strategies have been compared head-to-head in one trial. All drugs were administered in standard daily dosages with the ddI and NVP taken just once daily. Safety, activity and durability of the therapies are to be compared over a planned 72 weeks. Data on 234 subjects who had reached 24 weeks of follow-up was presented at this conference [1]. At baseline these 234 subjects had a median plasma HIV-RNA of 4.19 log copies/mL and mean CD4 count of 444 cells/mm3.
This interim analysis did not allow for statistical testing for significance and caution should therefore be used in interpreting the results.
Using an intent to treat analysis, percentages of patients with viral loads below 500 and below 50 copies/mL at 24 weeks were:
| <500 | <50 | |
| d4T/ddI/3TC (n=88) | 71% | 56% |
| d4T/ddI/NVP (n=68) | 69% | 67% |
| d4T/ddI/IND (n=78) | 78% | 71% |
All three treatment arms appeared to be well tolerated. A separate poster presentation [2] compared the initial clearance rate of HIV-RNA in plasma from participants in the ATLANTIC study. The preliminary data suggested that no large differences exist in the clearance rates and lag times between the three different regimens.
Abacavir (ZiagenTM) is an unlicensed nucleoside analogue currently in clinical studies, which is being developed by Glaxo Wellcome. It is dosed as one 300 mg tablet twice daily with no dietary restrictions. Data was presented at Chicago on a study in antiretroviral naive subjects comparing the activity of the triple nucleoside combination of abacavir (ABC)/ZDV/3TC versus the PI based combination of indinavir (IND)/ZDV/3TC. This trial (CNA3005) is a blinded randomised study and interim data at 24 weeks of this 48 week study were presented.
Unblinded data available at 24 weeks using an intent to treat analysis showed the percentage of subjects with plasma viral loads below the limits of detection as:
| ABC/ZDV/3TC | IND/ZDC/3TC | |
| <400 | 87% | 85% |
| <50 | 65% | 65% |
Baseline median plasma HIV-RNA was 4.8log and median CD4 count 360 cells/mm3. Median CD4 cell increase at 24 weeks was similar in both arms at 103 cells/mm3 for the ABC arm and 105 cells/mm3 for the IND arm. Discontinuation at 24 weeks was 33% and 34% in the IND and ABC containing groups respectively with approximately 15% in each group due to adverse events. 13 cases of hypersensitivity to abacavir were reported (5%) and one death following rechallenge early on in the trial.
Margaret Fishel gave follow up data on a trial of ABC/ZDV/3TC versus ZDV/3TC in treatment naive subjects (CNA3003) in a separate presentation. This study was blinded until week 16 when subjects could elect to receive open label ABC/ZDV/3TC. Results were analysed on an intent to treat basis according to baseline viral load:
| less than 400 at 48 weeks | less than 50 at 48 weeks | |
| All subjects | 61% | 56% |
| <10,000 | 62% | 62% |
| 10-100,000 | 71% | 64% |
| >100,000 | 33% | 28% |
A median increase of 150 CD4 cells/mm3 was achieved at 48 weeks.
High proportions of subjects with starting viral loads greater than 100,000 copies/mL failing to achieve viral loads below assay quantification levels suggests caution in the use of this triple nucleoside analogue combination for this group.
References
1. Katlama C, Murphy R, Johnson V, et al: The Atlantic Study: a randomised open-label study comparing two protease inhibitors (PI)-sparing antiretroviral strategies versus a standard PI-containing regimen [Abstract 18]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
2. Van Heeswijk R, Lipniacki A, Nowak P, et al: Initial HIV-1 clearance rate in antiretroviral therapy with D4T plus DDI and either NVP, IDV, or 3TC [Abstract 634]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999
3. Staszewski S, Keiser P, Gathe J, et al: CNA3005 Intl Study Team. Ziagen/Combivir is equivalent to indinavir/combivir in antiretroviral therapy (ART) naive adults at 24 weeks (CNA3005) [Abstract 20]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
4. Fischl M, Greenberg S, Clumeck N, et al: Ziagen (Abacavir, ABC, 1592) Combined with 3TC & ZDV is highly effective and durable through 48 weeks in HIV-1 infected antiretroviral-therapy-naive subjects (CNAA3003) [Abstract 19]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
|
Mean baseline viral load of the trial participants in ATLANTIC was comparatively low (15,700 copies/mL) with high CD4 counts (mean 444 cells/mm3). They are representative, therefore, of a patient population that many still consider should be cautious about initiating treatment at all. Additionally, being open-label, differences in administration and pill volume across the arms needs consideration (especially in the d4T/ddI/IND group which was essentially four-times daily dosing with extra food restrictions). The CNA3005 study is better for a head-to-head comparison, being blinded, but longer term data are needed here. The CNA3003 data raise the concern that longer term data may reveal inadequacies in performance (for both abacavir or a single PI) in those with higher starting viral loads It is also unfortunate that the abacavir trial analysis used a somewhat idiosyncratic form of the intent to treat analysis (all missing data was not considered failure - those "failed" but not due to virological failure or toxicity were assigned as 50% failure, 50% "success). This despite Glaxo Wellcome's Andrew Hill describing the standardisation of analytical methods as "...essential for reliable comparison of treatment effects in different clinical trials." [Abstract 394].
|
Paul Blanchard, ATP
G Carvelain of Brigitte Autrans group in Paris presented an evaluation of long-term immune reconstitution of specific CD4+ T cell responses to HIV and other pathogens after 1 to 3 years of therapy with 2NA's+PI or 3NA's.
55 HIV-1 infected patients were enrolled and CD4 T cell responses to recall cytomegalovirus (CMV), tuberculin (PPD), vaccinal tetanus toxoid (TT) and HIV-1 p24 antigen measured with a standard T cell proliferation assay performed on fresh samples of PBMC's. Subjects had achieved "relative" viral control (<200-10,000 copies/mL).
A subgroup of subjects who had achieved VL <200 copies/mL receiving a protease inhibitor + 2 NA regimen were compared with those receiving a 3NA regimen (also <200 copies) after one year of treatment. The CD4 count was 817(252 cells/mm3 and 734380 cells/mm3 for each group respectively. No advantage was observed in the group of patients treated with PI - the frequencies of positive responses to the antigens being similar in both groups and no responses against HIV-1 p24 were observed.
The study concluded that sustained recovery of CD4 T cell reactivity against the pathogens tested is feasible in immunosuppressed patients but remains below normal levels and does not appear to occur for HIV-1 antigens even after long-term treatment. A PI based regimen appeared to offer no advantage over a triple nucleoside regimen in this study.
Ref: Carvelain G, Tubiana R, Li TS, et al: Long-term recovery in CD4+ T cell function with highly efficient antiretroviral therapy [Abstract 324]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
|
Encouraging early data (if somewhat limited) that qualitative immune reconstitution may occur with triple nucleoside combinations. |
Paul Blanchard, ATP
Efavirenz (SustivaTM), formerly known as DMP-266 is a non-nucleoside reverse transcriptase inhibitor (NNRTI) currently being developed by DuPont Pharmaceuticals. Data from a trial comparing the triple combination of efavirenz (EFV)/ZDV/3TC to indinavir (IND)/ZDV/3TC and IND/EFV in treatment na夫e subjects has been previously reported to week 36. The Sixth Retrovirus meeting saw the presentation of week 48 data for this trial known as Study 006. [1]
* EFV + ZDV + 3TC
* EFV + IND
* IND + ZDV + 3TC
Mean baseline CD4 count was 345 cells/mm3 and mean HIV-RNA 4.77 log copies/mL.
At 48 weeks percentages of subjects with plasma HIV-RNA below the level of quantification using and intent to treat (NC=F) analysis were:
| <400 | <50 | |
| EFV/ZDV/3TC (n=142) | 71%* | 65%* |
| EFV/IND (n=141) | 54% | 48% |
| IDV/ZDV/3TC (n=143) | 48% | 43% |
*statistically significant difference from IDV/ZDV/3TC, p<0.05
A further analysis was performed to look at the effect of baseline viral load on the different regimens abilities to suppress plasma HIV-RNA. Subjects were not stratified by viral load at entry making this a retrospective analysis, but the results were encouraging with the EFV/ZDV/3TC combination performing equally well at all strata, including that of >100,000 copies/mL. [2] References
1. Tashima K, Staszewski S, Stryker R, et al: A phase III, multicenter, randomized, open-label study to compare the antiretroviral activity and tolerability of efavirenz + indinavir, versus EFG + zidovudine + lamivudine, versus IDV + ZDV + 3TC at 48 weeks [LB-16]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
2. Manion D, Labriola D, Ruiz N. Efficacy of efavirenz in containing regimens in patients with baseline plasma HIV-RNA viral loads exceeding 100,000 copies/mL. [Abstract 383]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
|
Efavirenz has just received approval for marketing in Europe and the supportive data of efficacy in viral loads >100,000 may well support it's choice as a standard of care over PI's. Single PI based combinations have been shown to perform less well with these higher starting viral loads (eg. SPICE study). Additionally, pharmacologic barriers so problematic to PI efficacy are less of a problem with efavirenz. This may be a major factor in the higher percnetages of suppression seen with efavirenz based combinations. As with triple nucleoside combinations additional data on lymph node clearance and qualitative immune reconstitution is needed as well as a better understanding of the CNS effects of this drug - effects which can cause significant morbidity in users of this compound.
|
Paul Blanchard, ATP
Francois Raffi presented interim results from the VIRGO trial on behalf of the VIRGO study team. This is an ongoing open label, non-comparative study evaluating the safety and efficacy of nevirapine (NVP)/d4T/ddI in antiretroviral na夫e adults with CD4 counts >200 cells/mm3.
In addition to viral load and CD4 response to this combination, measurements of numbers of na夫e (CD4/45+RA) and memory (CD4/45+R0) T cells were performed. Although numbers are small (n=12 at week 52), a rapid increase of both na夫e and memory CD4 cells (with these phenotypic markers) was observed during the first month of therapy. This was followed by a plateau in the number of memory cells and a sustained increase in the number of na夫e cells.
Ref: Raffi F, Reliquet V, Hascoet C, et al. The VIRGO Trial: d4T/ddI (qd) /nevirapine (bid or qd) in antiretroviral na夫e HIV-1 infected patients - a convenient and potent regimen. [Abstract 632]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
|
Interesting observations hinting at qualitative immune restoration with nevirapine. Numbers are small and more data obviously needed.
|
Isolated reports linking protease inhibitor use to heart disease have left most doctors uncertain as to the long-term risk for their patients, particularly those with existing contributory factors. Looking at potential risk factors for atherosclerosis, Dr Carl Grunfeld reminded his audience during a special symposium that lipid abnormalities (particularly hypertriglyceridaemia) have been a common phenomenon in at least 50% of HIV-infected patients long before the advent of HAART (and that these had not been associated with wasting) [1]. For example, people who are HIV-infected have shown significantly raised triglyceride (TG) levels (5-fold increase) following a high fat meal compared to a 'normal' 2-fold increase in HIV negative people. These contributory factors act in a synergistic way, so an HIV-infected person who starts off with high TG levels who eats a high fat meal can have TG elevated to levels which put them into the risk category for pancreatitis.
High density lipoprotein (HDL) levels almost always drop early in HIV infection, which is followed by a drop in low density lipoproteins (LDL) and a rise in very low density lipoproteins (VLDL) triglycerides. About 70% of the plasma total cholesterol is carried in the LDL fraction, and about 25% in the HDL fraction. The major atherogenic potential of the plasma triglycerides appears to be mediated by the LDL cholesterol. HDL has an inverse relationship to the development of coronary heart disease and is generally regarded as exerting a protective effect. VLDL does not appear to have a direct effect on the risk of CHD. The British Hyperlipidaemia Association and the European Atherosclerosis Society consider a desirable value of total cholesterol to be no more than 5.5 mmol/l, and for LDL levels to be 3.5 mmol/l or less for those with CHD risk factors, and 4.0 mmol/l or less for those without risk factors. Total triglycerides should be no more than 2.3 mmol/l. Accurate determination of triglycerides (but not cholesterol) requires a 12 hour fast before blood draw. Dr Grunfeld explained that low levels of both HDL and LDL results in a higher risk factor than when both levels are high but that if HDL remains low and LDL (and therefore total) levels rise - which is the pattern seen - then risk also rises. However, he calculated the risk of PI-related CHD (using data from the Framingham Heart Study, National Heart, Lung and Blood Institute, U.S.A) without other contributory factors, to be 1.41 new cases per 100 patients per 10 years. Adding in risk factors of hypertension and smoking raised this by 3.2 cases per 10 years. Even allowing for age at 55 adding PI treatment only increased the risk from 24.1 to 27. All these risks are far outweighed by the benefits of PI therapy, which have reduced deaths from HIV by at least 50%.
Three poster studies [2, 3, 4] retrospectively looked for CHD incidence and the relationship to PI therapy in over 1,325, 4,000 and 4,600 patients. All studies compared patients on treatment with controls not receiving PI treatment and either failed to find an increased risk, or found small differences which were not significant.
Several studies recorded similar metabolic and fat redistribution changes in patients not using protease combinations. Five patients naive to protease inhibitors were reported in a study from the Royal Free Hospital, London [5]. Of the five, (3 women and 2 men) 4 were treated with nucleoside combinations only and one patient was treated with ZDV/3TC/nevirapine. Median time on treatment was 22 months. Median current cholesterol level was 6.4 (range 4.7-6.7) and median current triglyceride value was 3.18 (range 1.97-4.9). Median weight loss was 5.2 kg (3.6-6.2) despite the relatively mild hypertriglyceridaemia in these subjects. Another study of 17 patients (10 male, 7 female) who had developed generalised lipodystrophy after mean 15 months (range 4 - 24) on nucleoside combination therapy (d4T/3TC in 8, d4T/ddI in 9) concluded that LD could also be a complication of nucleoside analogue therapy, possibly linked to d4T [6]. For further examples see the notes below on PI use in women.
Clinical responses to lipid disturbances in HIV-infection should include closer monitoring, use of diet/lifestyle modifications and treatments to address blood/lipid changes directly, and altering the HAART regimen which is assumed to cause the problem.
As the introduction of routine lipid and glucose monitoring has been integrated into care for patients on protease combinations, there is an argument that baseline DEXA and/or MRI baseline scans should also be made more accessible, particularly for early reporting of changes in body fat redistribution. There would also be considerable benefit in discovering a predictor of risk for development of lipodystrophy.
A study by Andrew Carr looked at the presence and severity of LD in 116 patients who had received PIs for a mean 21 months [7]. Standardised patient self-reporting questionnaire results (absent 17%, mild 42%, moderate 30%, severe 11%) correlated with total body fat monitored by DEXA scan (19.4, 20.0, 15.7 and 16.2% respectively; P=0.02). Patient rating correlated more to peripheral (P=0.005) than to central (P=0.09) fat. Not only did fasting triglyceride and C-peptide levels distinguish between mild and no LD (P=0.05) but prior levels predicted current LD severity (P=0.03). No difference was noted between different protease inhibitors. Use of C-peptide monitoring could therefore provide an effective indicator for risk of LD progression and a possible early indicator of expected benefits when switching from PI-based regimens.
[Note: C-peptide is a biologically inactive peptide produced when proinsulin is cleaved to yield insulin. Measurements of C-peptide may be useful in documenting pathologically low insulin secretion. Raised levels might suggest insulin resistance.]
Three posters and one late-breaker presented preliminary results from studies where patients were switched to either nevirapine in three studies or efavirenz in one. An arm switching to nelfinavir (from existing PIs) was also included in one of the nevirapine studies. Andrew Carr presented this study of 16 patients who switched from a PI (after a mean 16 month prior use) to nevirapine and compared body composition (DEXA), fasting lipid and glycaemic parameters, CD4 count and plasma viral load over 6 months. 47 HIV negative controls were also used for comparison [8]. After 3 months improvements were recorded in median fasting triglyceride, total cholesterol, C-peptide and insulin resistance (but not insulin or HDL cholesterol) although improvements were not sufficient to match the controls. Accumulated abdominal fat was reported to be declining. Peripheral fat decreased for 3 months and then increased. 4/15 experienced a viral load rebound above 400 copies/mL after switching. Twelve patients in the same study who switched (from indinavir or ritonavir/saquinavir) to nelfinavir showed either no change or a slight worsening in metabolic parameters and body fat.
E. Martinez presented a Spanish study with a similar design (n=23, mean baseline CD4 = 514 copies/mm3 (range 83-994), prior PI treatment median 12 months) [9]. Viral load was suppressed <200 copies/mL in all patients for at least 9 months before substituting for the PI. Six months after replacement of the PI with nevirapine (median 7 months, range 6-10 months) there was a significant improvement (p<0.05) in cholesterol (-21%), triglycerides (-56%), glucose (-16%) and fasting insulin resistance (-46%). Only one patient became detectable (at 546 copies m/L) and 21 patients reported some improvement in fat redistribution (particularly peripheral LD) although not back to pre-PI levels.
Another Spanish study by L. Ruiz [10] presented as a late breaker showed 12 week data from 21 out of 60 currently enrolled patients with lipodystrophy who were randomised to switch their PI for nevirapine (n=11) or continue with their current PI regimen (n=10). All patients had HIV RNA <400 copies m/l for at least the previous 6 months and monitoring throughout the trial used the ultra-sensitive (<50 copies) assay. Cholesterol diminished in the nevirapine group (23046 to 19654, p <0.05) and was normalised in 7/11 patients. Only one pt in the PI group achieved normal cholesterol levels. Although improvements were also reported in triglyceride levels, BIA and DEXA parameters in the nevirapine arm they hadn't reached statistical significance by week 12. No patients experienced rebound over this (short) period and 17/21 patients who had <50 copies /mL at baseline have so far maintained this level.
Graeme Moyle of the Chelsea and Westminster Hospital, London looked at switching the protease inhibitor of HAART directly for efavirenz in patients with LD symptoms and presented data on 12 patients who had reached 12 weeks and 8 patients out to 24 weeks [11]. All patients with less than 500 copies/mL viral load before the switch have remained <500. Extensive blood monitoring together with anthropometrics, DEXA and single-slice abdominal MRI were performed at 0, 12 & 24 weeks. Weight increases averaging 2kg and 3.5kg after 12 and 24 weeks respectively were accompanied by very modest reductions in abdomen girth (1cm and 0.3cm) although improvement in appearance and well-being have been reported by most patients. As 5 patients experienced a reduction of more than 5cm, other patients must have experienced increases in abdominal girth. It was also suggested that the increase in weight might be related to relaxed diet restrictions as 10/12 patients prior to the switch had been receiving indinavir. Of some concern was the finding that substitution of efavirenz actually resulted in a significant elevation of both fasting cholesterol (5.88 mmol/l to 7.78, p<0.02) and triglycerides (5.1mmol/l to 8.0) at week 12. This change was less pronounced by week 24, with a suggestion of a return to baseline values. Individual patient variability would suggest the necessity for intensive lipid monitoring for patients looking at this approach. Incidentally, patients in this study experienced a mean CD4 rise from 251 cells/mm3 at baseline to 290 (n=12) at week 12 and 342 cells/mm3 at week 24 (n=8).
Focusing on the symptoms of central adiposity (CA) and the hyperinsulinaemia often found to be associated with lipodystrophy. Thierry Saint-Marc from the Herriot Hospital, Lyon looked at using metformin, a 30-year old biguanide anti-diabetic medication that lowers insulin levels, to determine if it might help other LD metabolic effects [12]. 29 patients with CA were monitored for 8 weeks and then randomised to receive 850mg metformin orally TID (n=16) or to a no-treatment control arm (n=13). Two patients from the treatment group withdrew from the study because of GI adverse events (including diarrhoea and cramps) within the first 10 days.
Statistically significant biochemical results from the treatment group include mean lowered levels of glucose (5.10.36 to 4.50.22, p<0.05), insulin (31.48.4 to 18.44.1, p<0.02), sum insulin concentration (mlU/l) (478.6104 to 316.259, p<0.02), C-peptide (3.210.4 to 2.61, p<0.05) and triglycerides (3.772.36 to 2.511.64, p<0.05). Body fat distribution changes after 8 weeks which reached statistical significance included change in waist-to-hip ratio (0.90.05 to 0.870.06, p<0.05), visceral fat (194.965 to 121.849, change -13%, p<0.02). Mean total and HDL cholesterol levels did not differ significantly between the treatment and control arm. However 12/14 patients receiving metformin reported large reductions in body weight over 8 weeks of 1-6kg (mean weight loss of 2.8kg), the weight of two patients remained stable.
Side effects of metformin can include anorexia, nausea, and diarrhoea, so without dietary analysis it is not possible to be sure that the results seen were not linked to decreased energy intake which should be included in further studies.
Although rHGH is rarely made available to patients in Europe, even for wasting, small U.S. studies have reported a degree of success in reversing buffalo hump syndrome. R. Torres reported on two patients at last years Geneva conference and expanded this at Chicago to include effects on 8 patients (2 female, 6 male) [13]. All had been on protease including treatments for at least 12 months. Four patients reported 25-75% reduction in buffalo hunt and abdominal girth after 3 months treatment (4-6mg/day). One subject was lost to follow up and one has received less than 8 weeks treatment (although observations report notable reductions in buffalo hump and abdominal girth. Discontinuation of rHGH in one patient due to carpal tunnel syndrome resulted in a reoccurrence of symptoms.
C. Gervasoni and colleagues from the Institute of Infectious Diseases in Milan found fat redistribution (defined by increased breast and abdominal girth and wasting of glutei, thighs and calves) and confirmed by physical examination and DEXA scan in 32/306 women (10.2%) receiving anti-retroviral therapy [14]. All subjects found with fat redistribution were taking 3TC including regimens (p<0.0017) and 12/32 were taking dual nucleoside combinations without PIs. The risk of fat redistribution was lower with concomitant use of ZDV (p=0.0171) and higher with d4T (p=0.0118). In this study use of PI was not independently associated with fat redistribution. It was observed that these side effects are becoming a cause of poor adherence among women on anti-retroviral treatment.
In another study interviews with 127 unselected women (who were neither pregnant nor on hormone treatment) between May-July 1998 regarding body shape changes while on ART were analysed into those using either PI (group 1, n=95) or non PI (group 2, n=32) containing regimens. [15] Additional information was provided by chart review:
| Group 1 (PI) N=95 | Group 2 (non-PI) N= 32 | |
| Increase in breast size (n=45) | 35/95 (37%) | 10/32 (31%) |
| Mean bra size | + 1.8 | + 0.9 |
| Mean breast weight gain | 6.3 lbs. | 0.9 lbs. |
| Increase in waist size | 53/95 (56%) | 13/32 (41%) |
| Mean pant size | + 3.6 | + 2.6 |
| Mean overall weight gain | 7.8 lbs. | 3.8 lbs. |
Changes were sufficiently marked in 7 patients (all in group I), including 2 women with distinctive buffalo hump, to discontinue PI use after 2-6 months of treatment.
References:
1. Grunfeld, C. Distribution in Lipid Metabolism due to HIV Infection and its Therapy. [Symposium Session S3]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
2. Juette, A et al. Increased Morbidity from Severe Coronary Heart Disease in HIV-Patients Receiving Protease Inhibitors. [Abstract 656]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
3. Klein, D et al. Do Protease Inhibitors Increase the Risk for Coronary Heart Disease among HIV Positive Patients? [Abstract 657]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
4. Coplan, P et al. No Association Observed Between Indinavir Therapy for HIV/AIDS and Myocardial Infarction in 4 Clinical Trials with 2,825 Subjects. [Abstract 658]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
5. Madge, S et al. lipodystrophy Syndrome In Patients On Reverse Transcriptase Inhibitors. [Abstract 654]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
6. Saint-Marc, T et al. A Syndrome of Lipodystrophy in Patients Receiving a Stable Nucleoside-Analogue Therapy. [Abstract 653]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
7. Carr, A et al. Diagnosis and prediction of HIV protease inhibitor (PI)-induced lipodystrophy and impaired glucose tolerance. [Abstract 641]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
8. Carr, A et al. Reversibility of protease inhibitor (PI) lipodystrophy syndrome on stopping PIs or switching to nelfinavir. [Abstract 668]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
9. Martinez, E et al. Reversion of lipodystrophy after Switching HIV-1 Protease Inhibitors to Nevirapine. [Abstract 670]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
10. Ruiz, L et al. A multi-centre, randomised, open label comparative trial of the clinical benefit of switching the protease inhibitor by nevirapine in HAART-experienced patients suffering lipodystrophy. [LB14]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
11. Moyle, G et al. Management of Indinavir-associated metabolic changes by substitution with Efavirenz in virologically controlled HIV+ persons [Abstract 669]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
12. Saint-Marc, T et al. Effects of Metformin on Insulin Resistance and Central Adiposity in Patients Receiving Effective Protease Inhibitor (PI) Therapy. [Abstract 672]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
13. Torres, R et al. The Effect of Recombinant Human Growth Hormone on Protease-Inhibitor-Associated Fat Maldistribution Syndrome. [Abstract 675]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
14. Gervasoni, C et al. Nucleoside Reverse Transcriptase Inhibitors Associated Fat Redistribution in HIV-Infected Women Undergoing Combined Antiretroviral Therapy. [Abstract 660]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
15. Sutinen, J et al. Changes in Body Shape during PI Therapy in HIV+ Women. [Abstract 662]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
|
A predominant link between PI use and lipodystrophy would hopefully lead to reversal of symptoms in studies in which patients change to non-protease combinations. However the possibility remains that these phenomena may not be entirely reversible (ie. if due to loss of fat cells by apoptosis or other mechanisms). Signs of resolution certainly seem to take greater than 6 months to become manifest.To some extent, benefits have been shown in reduction of lipid levels, although documented physical improvement are less dramatic they may become clearer after a longer follow up.
|
Simon Collins, ATP
Vaccinations against hepatitis A (HAV) and hepatitis B virus (HBV) are generally recommended for patients from high risk groups who are also HIV-infected because of the complications hepatitis can cause - often requiring interruption of HAART and/or clinically relevant increases in viral load. Concerns remain, however, over the risk of significant HIV viral load rises which may accompany the vaccinations themselves. Although previous studies with the HAVRIX vaccine for HAV showed a 77-88% sero-conversion rate with no adverse effect on CD4 count, viral load measurements were not performed.
A study from the Navel Medical Centre in San Diego looked at determining the safety of the VAQTA vaccine for HAV. In a blinded, randomised study, 90 HAV IgG negative patients were given either vaccine or placebo (ratio 2:1) at weeks 0 and 24. Equal numbers of patients had CD4 >300 and <300 and changes in background HAART were discouraged. HAV sero-conversion rates were 63% at week 4 and 96% at week 28 (88% CD4 <300, 100% CD4 >300) and any increase in HIV viral load levels, where they were detected, were transitory.
Ref: Wallace M, Tasker S, Earhart K, Chamberlain-Brandt C. Hepatitis A Immunisation in HIV-Infection. [Abstract 195] 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
There has always been uncertainty about the benefit of IFV vaccination for HIV-infected patients, both from community and academic perspectives. Additionally, the likelihood of generating a significant immune response decreases with lower CD4 count. Patients with a CD4 count >300 are most likely to respond to the vaccine whilst those with a CD4 <100 will not. It is those with the more severe immune dysfunction, however, who are in need of the chance of protection the vaccine may offer.
Since the introduction of HAART, there have been conflicting reports addressing the risks of elevated HIV plasma RNA levels which can result from vaccination and the implications this has for developing resistance to a HAART regimen.
A study of 21 patients (mean CD4 count 404 copies/mm3) received IFV vaccination and were analysed according to their baseline viral load: group 1 (n=11) had <50 copies m/L, group 2 (n=5) had 50-400 copies m/L and group 3 (n=3) had >400 copies m/L. There were 5 HIV negative controls to compare the results on immune response. The results showed differences in levels of plasma HIV-RNA after vaccination between the different groups.
After vaccination only 3/11 from group 1 showed transient rises in viral load at week 1 or 2 (range 76-89 copies/mL). All five subjects in group 2 showed a rise, which persisted out to week 8 in 4/5. The 3 subjects from group 3 experienced a rise in both HIV-RNA and HIV-DNA. (HIV DNA levels had changed little in groups 1 and 2). The control group also had a transient mean CD4 increase of 175 cells/mm3 at week 2 which was statistically significant (p< 0.05) compared to a minimal CD4 response in all 3 HIV positive groups, although the response from group 1 was closer to the control than group 2 or 3.
Ref: Gunthard H, et al. Impact of Influenza Vaccination (IFV) on HIV plasma RNA and patterns of immune response (IR) in patients (pts) on HAART. [Abstract 337] 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
Joan Tallada, GTT, Barcelona
The development of effective anti-HIV drugs in Western countries has led to an improvement in the life expectancy for HIV-infected people. When well planned, used and tolerated, antiretrovirals can achieve a suppression of the viral replication to undetectable levels for a variable period of time. This fact, which in itself is good news, allows the consideration of other viral infections such as hepatitis that were considered secondary when HIV itself was life threatening. Among these hepatotropic viruses, the hepatitis C virus (HCV) is emerging pandemically due to its spread in some populations such as IVDU, and regions, such as Southern European countries. Therefore, HIV-HCV co-infection deserves more and deeper attention in all HIV research.
Until now, HCV-infected individuals, like other so-called special populations, were excluded from clinical trials to test anti-HIV drugs to avoid deviations in resulting data. The same trials were used to approve and market antiretrovirals, and are the background for therapeutic recommendations and drug use guidelines. Today we have to find out if strategies tested and thought to become a generalised standard of care have, for some geographical areas or population groups, too many exceptions.
Most antiretroviral drugs are metabolised through the liver. It is rational to question if persons with liver disease could show different pharmacokinetic (PK) profiles compared to the standard ones. Two interesting abstracts were presented in Chicago relating PK variability of anti-HIV drugs in subjects with liver disease. The study by Khalid et al. from the Ottawa Hospital, Canada [1], analysed nelfinavir (NFV, Viracept(tm)) PK in 5 patients with different degrees of liver disease (LD): 2 were severe, 1 moderate and 2 mild. The findings showed a wide variability in NFV PK in these 5 patients after >10 days of NFV dosing: 1 with severe LD and the one with moderate LD required dose adjustments in order to meet the usual drug levels in patients without LD. The conclusion stated that "hepatitis B or C patients with LD show wide variability in NFV PK. Therapeutic drug monitoring permits individualisation of NFV dose in LD, which may reduce hepatotoxicity and preserve anti-HIV activity".
Fiske et al., from DuPont Pharmaceuticals, presented a small, controlled study of 20 participants: 10 had chronic liver disease (CLD) and 10 were healthy [2]. Both groups were matched, subject by subject, for age, gender, weight and smoking status. All individuals were given a 400 mg single dose of efavirenz (EFV, Sustiva(tm)) and indocyanine green (ICG) to determine their hepatic blood flow prior to receiving EFV. The results showed a significant reduced blood clearances of ICG in CLD patients. The conclusion points out that "efavirenz pharmacokinetics may be altered in subjects with liver disease compared to healthy subjects. It is possible that the absorption of efavirenz is reduced in subjects with liver disease, and this effect is offset by the reduced rate of elimination which results in small differences in AUC between the two groups".
These two studies suggest that anti-HIV drugs could have different PK profiles in subjects with liver disease (including HCV) that was not foreseen when standards of care have been drawn up. Consequences could be of two types: too high drug levels, leading to increased toxicity, and too low levels, leading to reduced suppression of viral replication and higher risk of resistance. The possibility of antiretroviral drug-related increased toxicity in HIV-HCV coinfected individuals has been investigated by another study presented in Chicago. Raffi et al., from the ANRS EP11 Study, in Paris, France, investigated the incidence and characteristics of Serious Adverse Events (SAE), i.e. WHO grade 3 or requiring hospitalisation, occurring in a multicentre prospective cohort of 1,200 consecutive unselected HIV-infected patients started on a protease inhibitor-containing regimen [3]. As of October 1, 1998, 166 SAE (median occurrence delay: 76 days) have been observed in 150 patients; incidence: 34.4 for 100 pt-years (95% CI: 27.4-37.3). After death (n=22), the main SAE was hepatic cytolysis, present in 26 individuals, of whom 17 were HCV+. Among the 75 SAE reviewed by the validation committee, 46 were considered related to HAART, of which 38 were possibly (n=22) or probably (n=16) related to PI. Authors concluded that "high incidence of SAE in patients started on PI deserves careful evaluation of prognosis factors related to SAE occurrence, particularly hepatic cytolysis, because of possible multiple drug and viral interactions".
On the other hand, liver disease-related insufficient drug levels could be a key factor to explain virologic failure in co-infected individuals, and not only lack of or poor adherence, resistance mutations or intolerance.
These data suggests the possible need of individualised dosing of anti HIV drugs depending on hepatic function. PK individual variability should be researched both in the HIV-infected population and the so called special populations, particularly HIV-HCV co-infected subjects. Furthermore, drug strategies and official therapeutic guidelines have to be adapted to the diverse HIV-infected population groups and geographical areas.
References
1. Khaliq Y, Gallicano K, Seguin I, et al: Therapeutic drug monitoring of nelfinavir in HIV patients with liver disease [Abstract 369]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
2. Fiske W, Benedek I, Brennan J, et al: Pharmacokinetics of efavirenz in subjects with chronic liver disease [Abstract 367]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
3. Raffi F, Saves M, Moing VLE, et al: Serious adverse events (SAE) in a prospective cohort of HIV-infected adults started on protease inhibitor therapy (ANRS EP11 Study) [Abstract 676]. 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Ill, 1999.
|
Individual titration of antiretroviral drugs using therapeutic blood level monitoring would seem sensible as an attempt to reduce adverse effects seen with drug administration in this group of patients. Specific prospective studies of safety and efficacy of antiretrovirals in hepatitis co-infected patients are urgently needed. Clearly there are risks of increases in LFT's with some PI's (particularly ritonavir) and some NNRTI's (certainly nevirapine, probably efavirenz) and lacticacidosis/steatosis and fulminant hepatic failure secondary to nucleoside analogues has also been associated with HBV and HCV co-infection. Hepatitis is proving to be the fatal illness, not HIV, for some co-infected patients.
|