Atazanavir/r vs lopinavir/r in treatment-naive patients: 48 week results
1 April 2008. Related: Conference reports, Antiretrovirals, CROI 15 (Retrovirus) 2008.
Simon Collins, HIV i-Base
Although widely used off-label, atazanavir/r is not currently approved in Europe for use in first-line combinations, nor recommended in European guidelines, due to limited data in naive patients. Results from a large randomised international head-to-head study against lopinavir/r, sponsored by BMS, are therefore important to report.
Jean-Michel Molina from St Louis Hospital, Paris, presented the 48-week analysis from this 96-week CASTLE study. This was a non-inferiority study (10% margin) and the primary endpoint was the proportion of patients with HIV RNA <50 copies/mL at week 48.
The trial randomised 883 treatment-naive patients to either atazanavir 300 mg / ritonavir 100 mg once-daily or lopinavir/ritonavir 400 mg/100 mg twice-daily, both in combination with fixed-dose tenofovir/FTC once-daily.
Baseline demographics and characteristics included median CD4 count 205 cells/mm3 (range 2-810) with 12% less than 50 cells/mm3; and viral load around 5 log copies/mL (range 2.6-5.9), with 50% patients starting above 100,000 copies/mL. Only around 5% had CDC class C diagnosis, and 12% were coinfected with hepatitis B or C.
Only around 10% of patients discontinued prior to week 48, with a balance between each arm, detailed in Table 1.
Table 1: Patient disposition at week 48
| ATV/r n=121 | LPR/r n=127 | |||||
|---|---|---|---|---|---|---|
| BL | Mo12 | Change | BL | Mo12 | Change | |
| TG > 500mg/dL | 3% | 4% | +1% | 10% | 17% | +7% |
| TC > 240mg/dL | 23% | 20% | -3% | 20% | 26% | +6% |
| LDL-c > 130mg/dL | 20% | 25% | +5% | 20% | 30% | +10% |
| HDH-c < 40mg/dL | 16% | 32% | +16% | 21% | 29% | +8% |
The percentage of patients with viral load <50 copies/mL at week 48 in the atazanavir/r and lopinavir/r arms was 78% vs 76% (estimated difference 1.9, 95% CI 3.6 to 7.4), by intent-to-treat analysis.
Stratified by baseline viral load, the results were 82% vs 81% (<100,000 copies/mL) and 74% vs 72% (>100,000 copies/mL), in the ATZ/r and LPV/r groups respectively, with no statistical difference between arms.
A post-hoc analysis of results by baseline CD4 counts showed no impact for ATZ/r, but a statistically significant poorer response for the LPV/r arm, ranging from 80% <50 copies/mL in patients with >200 CD4 cells/mm3 to 63% for those starting with <50 cells/mm3, p = 0.0085.
CD4 response was similar in each arm (+203 vs + 219 cells/mm3).
Side effects generally reflected the known profile of each drug, with ATZ/r having higher incidence of jaundice, and LPV/r reporting greater GI-related aes, that are detailed in Table 2.
Table 2: Adverse events in CASTLE study
| ATZ/r
n=441 (%) |
LPV/r
n=437 (% |
|
|---|---|---|
| Serious AEs | 54 (12%) | 42 (10%) |
| All grade 2-4 | 115 (26%) | 129 (30%) |
| – Jaundice* | 16 (4%) | 0 |
| – Nausea* | 17 (4%) | 33 (8%) |
| – Diarrhoea* | 10 (2%) | 50 (11%) |
| – Rash* | 14 (3%) | 9 (2%) |
| Renal (all grades) | 14 (3%) | 9 (2%) |
| Total bilirubin >2.5xULN | 146 (34%) | 1 (<1%) |
| Total cholesterol (>240 mg/dL) | 30 (7%) | 77 (18%) |
| Triglycerides (>750 mg/dL) | 2 (<1%) | 15 (4%) |
| Hyperglycemia (>251 mg/DL) | 1 (<1%) | 1 (<1%) |
* Grade 2-4 in >3% of patients
With laboratory AEs, grade 3-4 ALT/AST elevations were low (>2%) in both arms.
Lipid profile favoured ATZ/r with fewer patients at week 48 having total cholesterol >240 mg/dL (7% vs 18%), and significantly lower changes from baseline for TC, non HDL and TG (all p<0.0001). Lipid lowering drugs were used by 2% and 7% of the ATZ/r and LPV/r arms respectively.
Comment
The study concluded that atazanavir/r is an appropriate treatment for first line therapy. Regulatory and guidelines committees will hopefully review these data carefully and promptly.
Reference:
Molina JF et al. Efficacy and safety of once-daily atazanavir/ritonavir compared to twice-daily lopinavir/ritonavir, each in combination with tenofovir and emtricitabine in ARV-naive HIV-1-infected subjects: the CASTLE study, 48-week results. Abstract 37.
http://www.retroconference.org/2008/Abstracts/31137.htm
This oral presentation is available to view online from the conference website (Monday 4 February).