News Release

24-Wk data from CONTEXT trial comparing GW433908/ritonavir QD & BID to Lopinavir/ritonavir BID

Peer-Reviewed Publication

Public Communications Inc.

Boston (Feb. 14, 2003) – Preliminary 24-week data were presented today from the CONTEXT trial, an open-label, multi-center study evaluating the safety and efficacy of once-a-day (QD) or twice-a-day (BID) dosing of the investigational protease inhibitor (PI) GW433908 (908) boosted with ritonavir (908/r) compared to a third treatment arm with the PI lopinavir/ritonavir (LPV/r, Kaletra®) BID in treatment-experienced patients with prior virologic failure. The medications in all three treatment arms were administered as part of combination therapy that included two reverse transcriptase inhibitors (RTIs). In this study, 908 was administered without food or fluid restrictions.

908 is the calcium phosphate ester pro-drug of amprenavir (APV) and was co-discovered by GlaxoSmithKline (GSK) and Vertex Pharmaceuticals (Nasdaq: VRTX).

The primary objective of CONTEXT is to assess antiretroviral response by measuring the time averaged change in viral load (vRNA) from baseline (AAUCMB) at 24 and 48 weeks. Similar efficacy responses were seen in both the 908/r regimens and the LPV/r regimen, meeting the primary endpoint of non-inferiority in this study at 24 weeks.

"In the preliminary analysis of 24-week data, the mean AAUCMB was comparable for all three treatment arms," said Edwin DeJesus, M.D., Infectious Disease Consultants, Altamonte Springs, Fla., lead investigator on the study.

"Although these results are from only 24 weeks of study, these Phase III data on the safety and efficacy of the investigational agent 908 as therapy for HIV in PI-experienced patients with prior virologic failure are encouraging," said Doug Manion, M.D., vice president of Clinical Development, GSK.

CONTEXT is the third of three pivotal trials to support regulatory submission of 908. The 24-week CONTEXT results add to the 48-week results of the NEAT trial, also presented today in Boston, and 48-week data from the SOLO trial, presented in November 2002 at the 6th International Congress on Drug Therapy in HIV Infection in Glasgow, United Kingdom.

Efficacy Results From CONTEXT

A total of 320 (315 treated) PI-experienced HIV+ patients experiencing virologic failure (plasma HIV-1 RNA viral load of ?1,000 copies/mL) were enrolled in the CONTEXT study. Forty percent of study participants had prior treatment experience with two or more PIs, 74 percent had experience with 3 or more nucleoside reverse transcriptase inhibitors (NRTIs), and 57 percent had received prior therapy with non-nucleoside reverse transcriptase inhibitors (NNRTIs).

Patients were randomized to receive either 1400 mg of 908 QD combined with low dose ritonavir (200 mg QD) (n=105), 700 mg of 908 combined with 100 mg ritonavir BID (n=107), or 400 mg lopinavir/100 mg ritonavir BID (n=103). All three groups took the medications in combination with two active NRTIs. Baseline vRNA (median 4.14 log10 copies/mL) and CD4+ cell counts (median 263 cells/mm3 ) were similar among the treatment arms.

By intent-to-treat analysis at 24 weeks, the mean AAUCMB (log10 c/mL) was -1.48 for patients on 908/r QD, -1.50 for those on 908/r BID and -1.66 for those taking LPV/r BID, according to DeJesus. Positive antiviral responses were seen with both the 908/r and LPV/r regimens, with 48 percent, 42 percent and 40 percent achieving <50 c/mL responses in the LPV/r, 908/r QD and 908/r BID arms, respectively. The median change from baseline in CD4 cells/mm3 was 72 for 908/r QD, 62 for 908/r BID and 63 for LPV/r BID.

Safety Results from CONTEXT

In the CONTEXT trial, the incidence of drug-related adverse events (AEs) of at least moderate severity was low, occurring in 19 percent of patients on 908/r QD (n=20), 35 percent on 908/r BID (n=37) and 34 percent on LPV/r BID (n=35). The most common AEs were diarrhea (5 percent, 10 percent and 8 percent, respectively) and nausea (4 percent, 3 percent and 9 percent, respectively). Additionally, more subjects discontinued LPV/r due to AEs (6 percent) than 908/rQD or 908/r BID (both <1 percent).

Likewise, Grade 3-4 lab abnormalities were similar between all three treatment arms, and all regimens had minimal impact on plasma lipids. The incidence of Grade 3-4 lipid abnormalities was low. No subject experienced a Grade 3-4 elevation in cholesterol, whereas elevated triglyceride levels occurred in 8 percent of patients taking 908/r BID (n=8,) versus 4 percent of patients taking 908/r QD (n=4) or LPV/r BID (n=4). Mean fasting cholesterol levels for all three study groups remained below the recommended National Cholesterol Education Program (NCEP) intervention guidelines.

Study Demographics

The study population comprised patients enrolled from 103 centers in 13 countries worldwide. Approximately 16 percent of patients enrolled in CONTEXT are female, and the age range of all patients is 24 –72 years (median 41 years). Sixty-seven percent of patients are Caucasian, 24 percent are of African descent and 9 percent are Hispanic. Criteria for entry into the study included treatment-experienced patients with current failure of antiretroviral therapy. To enroll in the CONTEXT trial, patients had to have a screening viral load >1,000 copies/mL and no active or acute Category C events (indicative of AIDS disease) according to guidelines established by the Centers for Disease Control and Prevention.

Summary of Clinical Trials with 908

More than 1,200 people have participated in three Phase III trials to test the safety and efficacy of 908: NEAT, SOLO and CONTEXT. The SOLO study was an open-label trial with 649 HIV+, treatment-naïve patients. Participants were randomized to receive either 1400 mg of 908 plus 200 mg ritonavir (908/r) QD or 1250 mg of nelfinavir BID. All also received the two NRTIs, ABC and 3TC. The trial, conducted at more than 100 research centers worldwide, was designed to assess the safety and efficacy of each regimen over a period of 48 weeks. Results from the SOLO trial were presented in November 2002 at the 6th International Congress on Drug Therapy in HIV Infection in Glasgow, United Kingdom.

NEAT was a phase III, randomized, open-label, parallel-group 48-week study that compared 908 BID and nelfinavir BID, both in combination with ABC and 3TC in 249 antiretroviral therapy-naïve patients. The primary endpoint of the NEAT study was the proportion of subjects with vRNA <400 c/mL at 24 and 48 weeks. Interim 24-week data from NEAT were presented in September at the 42nd Interscience Conference on Antimicrobial Agents and Chemotherapy (ICACC) in San Diego, and 48-week data are being presented here today.

Once approved, GSK will market 908 and GSK and Vertex will co-promote it in the United States and key markets in Europe.

GlaxoSmithKline is one of the world's leading research-based pharmaceutical and healthcare companies and an industry leader in HIV research and therapies. The company is engaged in basic research programs designed to investigate new targets to treat HIV.

This press release may contain forward-looking statements, including that 48-week interim analysis of the NEAT and SOLO studies in GlaxoSmithKline's pivotal program for 908 is indicative of a promising clinical and commercial outlook for 908, once approved. While management makes its best efforts to be accurate in making forward-looking statements, such statements are subject to risks and uncertainties that could cause Vertex's actual results to vary materially. These risks and uncertainties include, among other things, the risk that approval will be delayed or will not be obtained, that 908 will not be commercially successful, and those other risks listed under Risk Factors in Vertex's form 10-K filed with the Securities and Exchange Commission on April 1, 2002.

###

AT A GLANCE

Similar efficacy responses were seen in both the 908/r regimens and the LPV/r regimen, meeting the primary endpoint of non-inferiority in this study.

The CONTEXT study population included 320 patients who previously had been treated with one or two PIs and experienced virologic failure.

More subjects discontinued LPV/r due to adverse events (<1% 908/r QD or BID, 6% LPV/r BID). The most common adverse events for 908/r QD, 908/r BID and LPV/r BID were diarrhea (5%, 10%, 8%) and nausea (4%, 3%, 9%).

Data from CONTEXT add to the 48-week results of the NEAT trial also presented today and 48-week data from the SOLO trial, presented in November 2002.

Media contacts:
Mary Faye Dark, GlaxoSmithKline
Cell phone: 919-946-0190 – 2/10 through 2/14
Office: 919-483-2839

Beth Schlesinger, Public Communications Inc.
Pager: 800-759-8888, PIN 1050707 – 2/10 through 2/14
Office: 312-558-1770


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.