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
The 2NN Study: a comparison between efavirenz and nevirapine
By Dr Graeme Moyle MD, MBBS, Chelsea & Westminster Hospital, London, UK
  Several cohort studies, both in treatment naive and treatment experienced patients, have observed differences in efficacy between efavirenz and nevirapine. In all 5 published cohorts differences have favoured efavirenz in terms of both treatment success and time to virological failure. Cohort studies, however, have potential biases, which may, to some degree, be corrected for in analysis, but are never the 'gold standard' comparison that a randomized trial provides. The one randomised comparison of efavirenz and nevirapine to date is the NEFA study, in which PI treated patients with viral loads <200copies/ml were switched to either efavirenz, nevirapine or abacavir. No differences in viral efficacy or discontinuations were observed between efavirenz and nevirapine, although the reasons for discontinuation differed. For the most part, efavirenz treated patients who stopped this drug complained of so called central nervous system (CNS) effects (such as vivid dreams, concentration difficulties) whereas nevirapine treated patients were troubled by liver toxicity and skin rashes.
The 2NN study was an open-label, randomised, comparative trial of first-line antiretroviral therapy with regimens based on d4T plus 3TC plus either efavirenz, nevirapine dosed either at 400mg once or 200mg twice daily or both NNRTIs simultaneously (with a dose adjustment of 800mg of efavirenz dosed with 400mg nevirapine QD). The study randomised 1216 individuals of whom approximately 84% completed the study through 48 weeks. The study focused on the comparison of the approved dosing of efavirenz (600mg QD) and nevirapine (200mg BD, after an initial 200mg QD for 2 weeks), so that these groups contained approximately twice the number of participants as the other two 'exploratory' groups. Therefore, the number of participants who started treatment in each group were efavirenz 381, nevirapine BD 378, nevirapine QD 208 and efavirenz plus nevirapine 199 (2NN group). Participants were well matched at baseline with 63 percent of the study participants being men, the median age 34 years, median CD4 count 190 cells/mm3 and median viral load of 4.7 log copies/ml (~55,000). Treatment success was defined as completing 48 weeks on their original randomised treatment with a viral load less than 50 copies/ml and without experiencing HIV disease progression. By this definition, the number of individuals who were considered a treatment success at 48 weeks were 62.3% on efavirenz, 56.3% on nevirapine twice daily, 56.4% on nevirapine once daily, and 46.9% in the 2NN group. There was no statistically significant difference reported between the results when comparing efavirenz to nevirapine once or twice daily. The only statistical difference was between the efavirenz and 2NN group. Based on the safety and activity results from 2NN, using of nevirapine plus efavirenz together is not recommended.
A separate analysis of focusing on virological success/failure was also presented. Virological failure was defined as viral load values greater than 50 copies/ml having previously had a viral load less than 50 copies/ml, or never achieving a viral load less than 50 copies/ml. Virological success by this definition was observed in 67.8% on efavirenz recipients, 63.6% on nevirapine twice daily, 65% of those randomised nevirapine once daily, and 61.7% on the 2NN arm. These differences were not statistically significant. The 2NN study evaluated drug performance in patients with high (>100,000cps/ml) viral load. Analysis by baseline CD4 was not presented. All groups saw some decline in efficacy from low to high viral load. In the subset with a baseline viral load > 100,000 copies/ml, virological success was observed in 61.3% of efavirenz patients, 53.7% of nevirapine twice a day, 51.5% of nevirapine once and 57.1% of 2NN recipients. These differences were not subjected to statistical scrutiny. Relative to those with baseline viral loads <100,000 copies/ml, the 'step down' in efficacy in the efavirenz group was 9.8%, for nevirapine BD was 14.5% and for nevirapine QD was 19.6%. Other analyses also found no differences between groups. For the on-treatment analysis 86.8% on efavirenz recipients, 81.5% of nevirapine twice daily, 88.7% of nevirapine once daily, and 79.5% of 2NN recipients were virologic successes. Changes in CD4 count did not differ at between groups.
As the treatments were similar in efficacy, we need to consider adverse events in choosing between drugs. The proportion of patients who changed off their randomised NNRTI during the course of the study were 20% in efavirenz group, 22% with nevirapine twice daily, 29.1% with nevirapine once daily, and 34.5% in the 2NN group. Differences in the types of adverse events were seen between groups. In part, the interpreter of the results needs to 'rank' their concern about the different toxicities: efavirenz can lead to vivid dreams, sleep disturbance, impaired concentration, while nevirapine can cause abnormalities in liver function tests, clinical hepatotoxicity, infrequently liver toxicity; nevirapine has displayed modest advantages in terms of lipid changes.
Clinical hepatotoxicity events (i.e becoming jaundiced or other physically evident liver problems) were more common in nevirapine recipients with 1.4% of nevirapine QD and 2.1% of nevirapine BD recipients experiencing at least one grade 3-4 (serious or potentially life-threatening) event compared with 0.3% of efavirenz recipients. At a laboratory level, grade 3-4 liver function tests abnormalities (values of at least >5-times the upper limits of the normal range) were significantly more common in the nevirapine QD group as compared with the efavirenz group. Whereas 4.5% of efavirenz recipients experienced grade 3-4 AST or ALT elevations, these changes were seen in 13.2% of nevirapine QD, 7.8% of nevirapine BD and 8.6% of 2NN recipients.
Skin problems occured more often in the nevirapine group. Grade 3 rashes are those which include skin blisters and grade 4 rashes involved the mucosa such as mouth, genitals or eyes. Grade 3-4 rash was reported in 4.1% of nevirapine QD and 3.1% of nevirapine BD recipients and 1.8% of efavirenz treated patients as well as 3.8% of those on the 2NN arm.
CNS effects such as sleep disturbance, abnormal dreams and anxiety were reported in 1.8% of nevirapine QD, 4.9% of nevirapine BD, 6.5% of efavirenz and 8.1% of 2NN recipients. The difference between nevirapine QD and the 2NN arm was statistically significant.
Perhaps most important to patients and physicians is how many people stopped their NNRTI therapy due to toxicity. The percentage of patients in each group who discontinued therapy due to adverse events was statistically significant (p < 0.001) with 15.5% of efavirenz patients compared to 24.1% of nevirapine QD, 21.2% of nevirapine BD, and 29.7% of 2NN patients discontinuing due to adverse events. Two deaths were attributed to nevirapine; one secondary to fuminant hepatic failure (a woman with no hepatitis co-infection) and one to MRSA sepsis (blood infection with an antibiotic resistant bacteria) following a severe blistering rash (known as Steven's Johnson Syndrome). One death was att ributed to d4T due to lactic acidosis. Eleven deaths related to HIV disease. Eleven deaths were non-treatment related and non-HIV related; none of the suicides were attributed to study medications and there were no deaths attributed to efavirenz.
If we consider the differences in harm done by each drug, harm including virologic failure and causing side effects that led to discontinuation, differences emerge. Between efavirenz and nevirapine BD, patients on nevirapine experienced 4.2% more viral failures (7.6% in high viral load patients) and 5.7% more drug related discontinuations: 9.9% more individuals impacted by the choice nevirapine BD. Comparing efavirenz and nevirapine QD, viral differences were 2.8% (9.8% at high viral load) and 8.6% more discontinuations: a difference of 11.4%. It is up to physicians and patients to decide if these differences are clinically relevant.
The results of 2NN are in keeping with studies which have gone before. In the Parkland cohort (Keiser et al 2002) which followed patients at 3 US centers treated with nevirapine or efavirenz as first therapy, differences between the drugs at week 52 favoured efavirenz by 6%. A result very similar to 2NN. Over time this difference was important as the median time to viral rebound ('failure') was substantially different: 583 days with efavirenz, 303 days with nevirapine. No time to toxicity or time to treatment failure data have been presented from 2NN. Similarly, time to treatment related discontinuation or time to grade 3-4 adverse event have not been reported, data which will help inform issues around treatment monitoring as a means of anticipating and preventing adverse events. Breakdown of adverse events by Hepatitis B and C co-infection and other baseline parameters would also be informative.
Hepatotoxicity with nevirapine was also similar in the 2NN study to previous studies of nevirapine BD (BI1090 study 8%, 2NN 7.8% for grade 3-4) and nevirapine QD (Atlantic study 13%, 2NN 13,2%). Some but not all cohort studies have been consistent with these findings. Of note, Sulkowski et al (Hepatology 2002) observed a sharp increase in risk of grade 3-4 liver transaminase disturbance with both efevirenz and nevirapine in the presence of hepatitis co-infection. In both co-infected and HIV only infected patients, however, risk of liver transaminase elevation was greatest in nevirapine treated patients. More data are needed on the relationship between hepatitis co-infection and hepatotoxicity in this study.
Lipid profiles in 2NN are also consistent with the NEFA switch study and results of other studies in treatment naive patients, that nevirapine increases HDL-c and reduces total:HDL-c ratio more than efavirenz. In a 40 year old male smoker with hypertension and a family history of cardiovascular disease, one example presented by Dr Peter Reiss in the plenary session on cardiovascular disease, enrolment in the 2NN study would have reduced his 10 year risk of a cardiovascular event from 5.7% at baseline to 5.1% with nevirapine and kept it stable at 5.7% with efevirenz. For this individual the improvement appears modest, based on the assumption that the therapy and lipids changes remain stable over a 10 year period. Additionally, some studies, such as Gilead 903, have indicated that the d4T/3TC backbone may have greater effects on lipids that alternative backbones (such as Tenofovir/3TC), thus lipid changes observed in this study may underestimate the favourable impact of NNRTI based regimens on lipid profiles.
In summary, the first comparative study of efavirenz and nevirapine does not indicate statistical superiority for efavirenz over nevirapine. The drugs have distinctive adverse event profiles with CNS disturbances, which are sometimes transient and sometimes they are not, being characteristic of efavirenz whereas rash and liver toxicity are more characteristic of nevirapine. Nevirapine once daily is as effective as twice daily although some questions remain about the use of once daily nevirapine in high viral load patients and that some trends to more liver and skin toxicities were seen. Combining nevirapine 400mg QD and efavirenz 800mg QD leads to lower treatment efficacy due to increased adverse events.
van Leth F, E Hassink , Phanuphak P, et al. Results of the 2NN Study: A Randomized Comparative Trial of First-line Antiretroviral Therapy with Regimens Containing Either Nevirapine Alone, Efavirenz Alone or Both Drugs Combined, Together with Stavudine and Lamivudine. 10th Conference on Retroviruses and Opportunistic Infections, Boston, Feb 10-14, 2003; abstract 176.
Other Studies Mentioned
Keiser P, Nassar N, White C, Koen G, Moreno S. Comparison of nevirapine- and efavirenz-containing antiretroviral regimens in antiretroviral-naive patients: a cohort study. HIV Clin Trials. 2002;3:296-303.
Cozzi-Lepri A, Phillips AN, d'Arminio Monforte A, et al. Virologic and immunologic response to regimens containing nevirapine or efavirenz in combination with 2 nucleoside analogues in the Italian Cohort Naive Antiretrovirals (I.Co.N.A.) study. J Infect Dis. 2002;185:1062-1069.
Matthews GV, Sabin CA, Mandalia S, et al. Virological suppression at 6 months is related to choice of initial regimen in antiretroviral-naive patients: a cohort study. AIDS. 2002;16:53-61.
Phillips AN, Pradier C, Lazzarin A, et al. Viral load outcome of non-nucleoside reverse transcriptase inhibitor regimens for 2203 mainly antiretroviral-experienced patients. AIDS. 2001;15:2385-2395.
Moyle GJ, Wilkins E, Leen C, Cheeesbrough A, Reynolds B, Gazzard BG. Salvage therapy with abacavir plus efavirenz or nevirapine in HIV-1-infected persons with previous nucleoside analogue and protease inhibitor use. AIDS. 2000;14:1453-1454.
Robbins G, Shafer R, Smeaton L, et al. Study of initial antiretroviral strategies for the treatment of HIV infection: a randomized 6-arm trial utilizing a factorial design [Abstracts LbOr20A and LbOr20B]. XIV International AIDS Conference, July 7-12, 2002; Barcelona.
Squires K. The Atlantic Study: a randomized, open-label trial comparing two protease inhibitor-sparing anti-retroviral strategies versus a standard PI-containing regimen, final 48-week data [Abstract LbPeB7046]. XIII International AIDS Conference; July 9-14, 2000; Durban.
Pollard R and the 1090 Team. Factors predictive of durable HIV suppression in randomized double blind trial with nevirapine (NVP), zidovudine (ZDV) and lamivudine (3TC) in treatment-naive (ARV-n) patients with advanced AIDS [Poster]. 7th Conference on Retroviruses and Opportunistic Infections, January 30-February 2, 2000; San Francisco.