Content uploaded by Keith Henry
Author content
All content in this area was uploaded by Keith Henry on Mar 27, 2015
Content may be subject to copyright.
n engl j med
348;22
www.nejm.org may
29, 2003
2175
The
new england
journal
of
medicine
established in 1812
may
29
,
2003
vol. 348 no. 22
Enfuvirtide, an HIV-1 Fusion Inhibitor, for Drug-Resistant
HIV Infection in North and South America
Jacob P. Lalezari, M.D., Keith Henry, M.D., Mary O’Hearn, M.D., Julio S.G. Montaner, M.D., Peter J. Piliero, M.D.,
Benôit Trottier, M.D., Sharon Walmsley, M.D., Calvin Cohen, M.D., Daniel R. Kuritzkes, M.D., Joseph J. Eron, Jr., M.D.,
Jain Chung, Ph.D., Ralph DeMasi, Ph.D., Lucille Donatacci, M.S., Claude Drobnes, M.D., John Delehanty, Ph.D.,
and Miklos Salgo, M.D., Ph.D., for the TORO 1 Study Group*
abstract
From Quest Clinical Research, Mount Zion
Hospital, and the University of California,
San Francisco, San Francisco (J.P.L.); the
HIV Program, Hennepin County Medical
Center, Minneapolis (K.H.); the Oregon
Health and Science University, Portland
(M.O.); St. Paul’s Hospital and the Uni-
versity of British Columbia, Vancouver, Can-
ada (J.S.G.M.); the Clinical Research Initia-
tive, Albany Medical College, Albany, N.Y.
(P.J.P.); the Clinique Médicale l’Actuel, Mon-
treal (B.T.); the University of Toronto, To-
ronto (S.W.); the Community Research Ini-
tiative of New England, Boston (C.C.); the
Division of Infectious Diseases, University
of Colorado Health Sciences Center, Denver
(D.R.K.); the University of North Carolina at
Chapel Hill School of Medicine, Chapel Hill
(J.J.E.); Roche, Nutley, N.J. (J.C., L.D., M.S.);
and Trimeris, Durham, N.C. (R.D., C.D.,
J.D.). Address reprint requests to Dr.
Lalezari at Quest Clinical Research, 2300
Sutter St., Suite 202, San Francisco, CA
94115, or at drjay@questclinical.com.
*Members of the TORO 1 (T-20 vs. Opti-
mized Regimen Only Study 1) group are
listed in the Appendix.
This article was published at www.nejm.org
on March 13, 2003.
N Engl J Med 2003;348:2175-85.
Copyright © 2003 Massachusetts Medical Society.
background
The T-20 vs. Optimized Regimen Only Study 1 (TORO 1) was a randomized, open-label,
phase 3 study of enfuvirtide (T-20), a human immunodeficiency virus type 1 (HIV-1)
fusion inhibitor.
methods
Patients from 48 sites in the United States, Canada, Mexico, and Brazil with at least six
months of previous treatment with agents in three classes of antiretroviral drugs, re-
sistance to drugs in these classes, or both, and with at least 5000 copies of HIV-1 RNA
per milliliter of plasma were randomly assigned in a 2:1 ratio to receive enfuvirtide plus
an optimized background regimen of three to five antiretroviral drugs or such a regi-
men alone (control group). The primary efficacy end point was the change in the plas-
ma HIV-1 RNA level from base line to week 24.
results
A total of 501 patients underwent randomization, and 491 received at least one dose of
study drug and had at least one measurement of plasma HIV-1 RNA after treatment
began. The two groups were balanced in terms of the median base-line HIV-1 RNA
level (5.2 log
10
copies per milliliter in both groups), median CD4+ cell count (75.5
cells per cubic millimeter in the enfuvirtide group, and 87.0 cells per cubic millimeter
in the control group), demographic characteristics, and previous antiretroviral ther-
apy. At 24 weeks, the least-squares mean change from base line in the viral load (inten-
tion-to-treat, last observation carried forward) was a decrease of 1.696 log
10
copies
per milliliter in the enfuvirtide group, and a decrease of 0.764 log
10
copies per milli-
liter in the control group (P<0.001). The mean increases in CD4+ cell count were 76
cells per cubic millimeter and 32 cells per cubic millimeter, respectively (P<0.001).
Reactions at the site of the injections were reported by 98 percent of patients receiving
enfuvirtide. There were more cases of pneumonia in the enfuvirtide group than in the
control group.
conclusions
The addition of enfuvirtide to an optimized antiretroviral regimen provided significant
antiretroviral and immunologic benefit through 24 weeks in patients who had previous-
ly received multiple antiretroviral drugs and had multidrug-resistant HIV-1 infection.
Downloaded from www.nejm.org at Stanford University on September 29, 2003.
Copyright © 2003 Massachusetts Medical Society. All rights reserved.
n engl j med
348;22
www.nejm.org may
29
,
2003
The
new england journal
of
medicine
2176
he durability of suppression of
human immunodeficiency virus (HIV) in-
fection with antiretroviral therapy is often
limited, for reasons that include poor penetration
into protected sites containing a reservoir of HIV,
1,2
drug toxicity,
3,4
alterations in the bioavailability
and metabolism of antiretroviral drugs (e.g., inter-
actions between drugs),
5
and lack of adherence to
complex treatment regimens.
6,7
These factors con-
tribute to persistent viral replication in patients
receiving therapy, increasing the risk of viral resist-
ance, which can limit future treatment options.
8,9
Salvage therapy after viral rebound is more success-
ful if an agent from a class of antiretroviral drugs to
which the patient has not previously been exposed
is included in the regimen.
Enfuvirtide (also known as T-20) is a novel, syn-
thetic, 36-amino-acid peptide that binds to a re-
gion of the envelope glycoprotein 41 of HIV type 1
(HIV-1) that is involved in the fusion of the virus
with the membrane of the CD4+ host cell.
10
This
agent exhibits potent and selective inhibition of
HIV-1 in vitro without cytotoxicity
10
and is the first
inhibitor of HIV entry to show consistent potent ac-
tivity in persons infected with HIV-1.
11-14
In the T-20 vs. Optimized Regimen Only Study
1 (TORO 1), we compared the effect of enfuvirtide
in combination with an antiretroviral regimen that
was optimized with the aid of phenotypic and geno-
typic resistance testing with the effect of an opti-
mized regimen alone on plasma HIV-1 RNA levels
and CD4+ cell counts in patients who had previous-
ly received multiple antiretroviral drugs and carried
virus that was resistant to all three currently available
classes of antiretroviral drugs. A similar study (T-20
vs. Optimized Regimen Only Study 2 [TORO 2]) was
conducted in Europe and Australia.
15
In this article,
we present the results of the week-24 primary effi-
cacy and safety analyses of TORO 1.
study design
We conducted a randomized, open-label, multi-
center, phase 3 study comprising a 6-week screen-
ing phase followed by 48 weeks of treatment, with
an optional 48-week extension of treatment, and
4 weeks of follow-up for safety analyses. An initial
screening visit, occurring three to six weeks before
randomization, included a complete medical his-
tory taking, measurement of plasma HIV-1 RNA
(Amplicor HIV-1 Monitor, version 1.5, Roche), and
genotypic and phenotypic resistance testing (per-
formed by ViroLogic, San Francisco). A second
measurement of plasma HIV-1 RNA and safety as-
sessments were performed at a second screening
visit one to two weeks before randomization. An op-
timized regimen of three to five antiretroviral drugs
was selected by the investigator and the patient be-
fore randomization on the basis of the patient’s pre-
vious antiretroviral treatment and tolerance history
and the results of previous and screening genotypic
and phenotypic resistance testing.
Qualifying patients underwent randomization
according to a centralized, adaptive randomization
scheme and were assigned in a 2:1 ratio to one of
two groups: enfuvirtide plus the optimized back-
ground regimen (enfuvirtide group) or the opti-
mized regimen alone (control group). Randomiza-
tion was stratified according to the second plasma
HIV-1 RNA measurement (<40,000 or ≥40,000 cop-
ies per milliliter) and according to the use or nonuse
of newly approved or investigational antiretroviral
drugs (lopinavir–ritonavir, tenofovir, or both) in the
optimized background regimen.
Virologic failure was defined by either a decrease
from base line of less than 0.5 log
10
copies per mil-
liliter in plasma HIV-1 RNA on two or three consec-
utive measurements after week 6, with at least 14
days between the first and last measurements, a de-
crease from base line of less than 1.0 log
10
copies
per milliliter in plasma HIV-1 RNA on consecutive
measurements (as above) after week 14, or a de-
crease from base line of at least 2.0 log
10
copies per
milliliter in plasma HIV-1 RNA on consecutive
measurements (as above) followed by a rebound of
more than 1.0 log
10
copies per milliliter in plasma
HIV-1 RNA from the average of the two lowest val-
ues (not necessarily consecutive) after week 6. All
patients in whom the criteria for virologic failure
were met after week 8 underwent repeated genotyp-
ic and phenotypic resistance testing and were en-
couraged to change their background regimen. In
these cases, a background regimen of more than
five drugs was permitted. Patients in the enfuvirtide
group could continue to receive enfuvirtide, and pa-
tients in the control group were permitted to add
enfuvirtide to their revised regimen. Patients with
virologic failure who did not want to switch to or to
continue to receive enfuvirtide were allowed to re-
main in the study for a maximum of one month.
The results on the screening resistance tests were
used to define phenotypic and genotypic sensitivity
scores. The genotypic sensitivity score was the total
number of drugs in the optimized background reg-
imen to which a patient’s viral isolate showed geno-
t
methods
Downloaded from www.nejm.org at Stanford University on September 29, 2003.
Copyright © 2003 Massachusetts Medical Society. All rights reserved.
n engl j med
348;22
www.nejm.org may
29, 2003
enfuvirtide for drug-resistant hiv-1 in north and south america
2177
typic sensitivity according to a modification of a
previously published interpretation algorithm.
16
For tenofovir, genotypic resistance was defined
as the presence of K65R, or three or more of the
thymidine analogue–associated resistance muta-
tions (M41L, D67N, K70R, L210W, T215Y, T215F,
K219Q, K219E, or K219N), including either M41L
or L210W. The phenotypic sensitivity score was the
total number of drugs in the optimized background
regimen to which a patient’s viral isolate showed
phenotypic sensitivity. For tenofovir, the phenotypic
sensitivity was set equal to the genotypic sensitivity.
study population
Patients were older than 16 years of age and had at
least six months of previous treatment with at least
one nucleoside reverse-transcriptase inhibitor, at
least one nonnucleoside reverse-transcriptase in-
hibitor, and at least two protease inhibitors, docu-
mented resistance to these drugs, or both. Patients
were eligible for the study if they had been receiving
stable combination therapy or no antiretroviral ther-
apy for four or more weeks before inclusion and
had a plasma HIV-1 RNA level of at least 5000 cop-
ies per milliliter at both screening visits. Patients
were ineligible for inclusion if they had already re-
ceived treatment with enfuvirtide, the experimental
fusion inhibitor T-1249, or both. Female patients
were excluded if they were pregnant, breast-feeding,
or planning to become pregnant during the study.
consent, approvals, and conduct
of the study
Written informed consent was obtained from all
patients. Before the study began, the protocol and
the informed-consent provisions were reviewed and
approved by the independent ethics committee or
institutional review board at each of the centers in-
volved in the study.
Design of the trial protocol was the responsi-
bility of the study sponsors in collaboration with
various health authorities and advisory boards,
which included some coauthors of the article. All
statistical analyses were executed by employees of
the study sponsors. Data were collected by the study
sponsors. Interpretation of the data was performed
by the study sponsors in collaboration with the ad-
visory boards and the clinical investigators.
study medication
Enfuvirtide (90 mg) was administered twice daily
by subcutaneous injection into the abdomen, the
upper arm, or the anterior aspect of the thigh. The
first injection was administered by study personnel
at the study site, and patients were trained in sterile
technique and instructed in reconstituting and in-
jecting enfuvirtide.
The optimized background regimen could in-
clude tenofovir, lopinavir–ritonavir, or both. Both of
these agents were investigational at the start of the
study but were approved in most countries during
the course of the study. Ritonavir at doses of 266 mg
per day or less (i.e., booster doses) was not counted
as an active component in either a prestudy regi-
men or the background regimen. Changes in the
background regimen were permitted only in the
event of protocol-defined virologic failure or man-
agement of toxic effects. Adherence in both treat-
ment groups was calculated with the use of a patient
questionnaire that assessed the number of missed
doses of enfuvirtide or oral antiretroviral drugs dur-
ing the four days preceding each study visit.
efficacy analysis
The primary efficacy end point was the change from
base line to 24 weeks in the plasma HIV-1 RNA level
(measured on a logarithmic [base 10] scale). Sec-
ondary efficacy end points included the category of
virologic response, the time to virologic failure, and
the changes from base line to week 24 in the CD4+
cell count. Virologic response at week 24 was classi-
fied into three categories: an HIV-1 RNA level of less
than 50 copies per milliliter, an HIV-1 RNA level of
less than 400 copies per milliliter, or a decrease from
base line in the HIV-1 RNA level of at least 1.0 log
10
copies per milliliter; two consecutive measurements
were required for categorization.
safety analysis
Safety end points included adverse events, serious
adverse events (including death), adverse events
leading to premature withdrawal from the study,
local reactions at the site of the injections, and ab-
normal results on clinical laboratory tests (hema-
tology, serum chemistry, and urinalysis). The sever-
ity of adverse events was graded according to the
modified grading scale of the AIDS Clinical Trials
Group.
17
Causality was assessed in all cases; if the
study treatment was deemed to have caused the
event, investigators attributed it to the entire regi-
men, except in the case of serious adverse events,
which were attributed to individual agents.
Injection-site reactions were assessed according
to an overall grade that was based on the level of
pain and discomfort. In the event of a grade 4 injec-
tion-site reaction (severe pain that was clinically sig-
Downloaded from www.nejm.org at Stanford University on September 29, 2003.
Copyright © 2003 Massachusetts Medical Society. All rights reserved.
n engl j med
348;22
www.nejm.org may
29
,
2003
The
new england journal
of
medicine
2178
nificant or life-threatening or resulted in a new or
prolonged hospitalization or persistent or substan-
tial incapacity or death) or recurrent grade 3 injec-
tion-site reactions (severe pain requiring nontopical
analgesic agents or limiting the patient’s ability to
engage in usual activities), enfuvirtide therapy was
discontinued. All grade 4 injection-site reactions
and laboratory abnormalities except grade 4 triglyc-
eride values were also defined as serious adverse
events.
An additional updated safety analysis combined
data from the two phase 3 studies (TORO 1 and
TORO 2). This combination offered a larger pop-
ulation on which to base a characterization of the
safety profile of enfuvirtide and was appropriate
because the studies have similar designs, criteria for
patient selection, and protocol-specified analyses.
For this update, a separate analysis investigated the
incidence of combinations of adverse events that
might be considered clinically equivalent in order to
identify whether small increases in the incidence of
several adverse events would, when combined, show
a relevant difference between treatment groups.
statistical analysis
Data on efficacy were analyzed for an intention-
to-treat population, defined as all patients who un-
derwent randomization, received at least one dose
of study medication, and had at least one plasma
HIV-1 RNA measurement after treatment began.
HIV-1 RNA values obtained after confirmed viro-
logic failure were excluded from the analysis. Miss-
ing values were imputed for the analysis with the
use of a last-observation-carried-forward method.
The log
10
HIV-1 RNA value at week 24 was defined
as the mean of the last two log
10
HIV-1 RNA values
on completion of 24 weeks in the study, the mean
of the two HIV-1 RNA values that confirmed viro-
logic failure, or for patients who withdrew from the
study, the mean of the last two log
10
HIV-1 RNA val-
ues before withdrawal. Changes from base line to
week 24 in log
10
HIV-1 RNA levels and CD4+ cell
counts were evaluated by analysis of covariance in-
cluding terms for the randomization stratum (of
plasma HIV-1 RNA level and use or nonuse of new-
ly approved or investigational agents), the treatment
group, and the interaction between stratum and
treatment group, with the base-line phenotypic sen-
sitivity score (for the evaluation of the change in the
HIV-1 RNA level) or the base-line CD4+ cell count
(for the evaluation of the change in the CD4+ cell
count) as covariates.
To test the robustness of the results of the pri-
mary efficacy analysis based on the last-observa-
tion-carried-forward method of imputation, three
sensitivity analyses were performed as follows: the
change from base line in viral load was set at zero
for patients who withdrew before week 24; the
change from base line in viral load was set at zero
for patients who withdrew or who had virologic fail-
ure before week 24; and in a cohort analysis, sepa-
rate analyses were performed without the use of
the last-observation-carried-forward method for pa-
tients who completed 4, 8, 12, 16, 20, and 24 weeks
of treatment.
For analysis of categories of virologic response
(with patients who had missing data or virologic
failure treated as having had no response), a strati-
fied Mantel–Haenszel test was used. Time to viro-
logic failure was estimated by the Kaplan–Meier
method. A stratified log-rank test was used to com-
pare the time-to-event curves of the two treatment
groups.
study population
A total of 501 patients underwent randomization at
48 centers in the United States, Canada, Mexico, and
Brazil between December 2000 and June 2001. Of
these patients, 491 (326 in the enfuvirtide group and
165 in the control group) used the study medication
at least once, had a follow-up visit to record safety-
related data, and had an assessment of the HIV-1
RNA level after treatment began (Fig. 1).
demographic and base-line characteristics
of the patients
Demographic and base-line characteristics were
similar in the two groups (Table 1). The genotypic
sensitivity scores at base line (mean, 1.9 in each
group) and the phenotypic sensitivity scores at base
line (mean, 1.7 in the enfuvirtide group and 1.8 in
the control group) indicated that HIV-1 from the ma-
jority of patients in each group was sensitive to less
than two of the drugs used in the background reg-
imen. Previous treatment with at least five protease
inhibitors was reported for a slightly higher per-
centage of patients in the enfuvirtide group (49.4
percent) than in the control group (39.4 percent,
P= 0.04). The percentage of patients who had pre-
viously received lopinavir–ritonavir, classified as a
newly approved or investigational antiretroviral
drug in this study, was also higher in the enfuvirtide
results
Downloaded from www.nejm.org at Stanford University on September 29, 2003.
Copyright © 2003 Massachusetts Medical Society. All rights reserved.
n engl j med
348;22
www.nejm.org may
29, 2003
enfuvirtide for drug-resistant hiv-1 in north and south america
2179
group (38.7 percent) than in the control group (27.9
percent, P= 0.02). In both groups, there was a mean
of four drugs in the background regimen (Table 1).
changes in treatment and premature
withdrawal
In the control group, 106 patients (64.2 percent)
had protocol-defined virologic failure by week 24,
and 81 of these patients switched to enfuvirtide
(Fig. 1). By week 24, 37 patients in the enfuvirtide
group (11.3 percent), 18 patients remaining in the
control group (21.4 percent), and 6 of the patients
who had switched to enfuvirtide (7.4 percent) had
withdrawn from the study.
adherence
The mean level of adherence to the enfuvirtide com-
ponent of the regimen over the 24-week period was
at least 85 percent in 92.9 percent of patients. The
overall adherence to the entire regimen in the enfu-
virtide group was at least 85 percent in 88.0 percent
of patients. The mean level of adherence to the back-
ground regimen in the control group was at least
85 percent in 90.3 percent of patients.
efficacy
Changes in the Plasma HIV-1 RNA Level
The least-squares mean change from base line in
the plasma HIV-1 RNA level was a decrease of 0.764
log
10
copies per milliliter in the control group and
a decrease of 1.696 log
10
copies per milliliter in the
enfuvirtide group, representing a difference be-
tween groups of 0.933 log
10
copies per milliliter (P<
0.001) (Table 2). The least-squares mean differences
significantly favored the enfuvirtide group in all four
randomization strata (P<0.05 for all comparisons).
Figure 1. Disposition of All Randomized Patients to Week 24.
Virologic failure was defined as outlined in the Methods section.
501
Underwent randomization
332
Assigned to enfuvirtide group
4
Not treated
328
Treated
2
Not treated
167
Treated
17
Withdrew
119
Completed
study
20
Withdrew
170
Completed
study
8
Withdrew
51
Completed
study
2
With no
follow-up
136
With virologic
failure
190
Without virologic
failure
2
With no
follow-up
106
With virologic
failure
59
Without virologic
failure
169
Assigned to control group
10
Withdrew
15
Completed
study
25
Did not switch
to enfuvirtide
6
Withdrew
75
Completed
study
81
Switched to
enfuvirtide
Downloaded from www.nejm.org at Stanford University on September 29, 2003.
Copyright © 2003 Massachusetts Medical Society. All rights reserved.
n engl j med
348;22
www.nejm.org may
29
,
2003
The
new england journal
of
medicine
2180
The sensitivity analyses yielded least-squares mean
differences of 0.888 log
10
copies per milliliter (P<
0.001) when the change from base line in viral load
was set at zero for patients who withdrew from the
study and 0.802 log
10
copies per milliliter (P<0.001)
when the change from base line in viral load was set
at zero for both patients who withdrew and those
who had virologic failure. The least-squares mean
difference also remained significant (P<0.05) in
each of the cohort analyses performed at time points
up to week 24. At week 24, the proportion of pa-
tients in each category of response was significantly
greater in the enfuvirtide group than in the control
group (Table 2).
The percentage of patients with virologic failure
by week 8 was greater in the control group (33.3 per-
cent) than in the enfuvirtide group (16.0 percent),
and this difference persisted to week 24 (control
group, 64.2 percent; enfuvirtide group, 41.7 per-
cent). The distribution of the time to virologic failure
was significantly different between the two groups
(P<0.001 by the log-rank test) (Fig. 2). The median
time to virologic failure was 99 days in the control
group but could not be estimated in the enfuvirtide
group.
Changes in CD4
+
Cell Counts
At week 24, the increase from base line in the CD4+
cell count was significantly greater in the enfuvirtide
group than in the control group (Table 2).
safety
Injection-Site Reactions
At week 24, nearly all patients in the enfuvirtide
group (98.2 percent) had had at least one injection-
site reaction, with most having their first reaction
during week 1. Among patients who had pain or
discomfort from injection-site reactions, most had
either mild tenderness (49.7 percent) or moderate
pain without limitation of usual activities (41.7 per-
cent); 8.7 percent had pain or discomfort requir-
ing nontopical analgesic agents or limiting usual
activities, and none required hospitalization. Fre-
quent symptoms of injection-site reactions includ-
ed erythema (in 87.1 percent of patients), induration
(in 84.0 percent), and nodules and cysts (in 81.6 per-
cent). There was no evidence of an increase in the
severity of injection-site reactions over time. Only
small percentages of patients (2.8 percent in the en-
fuvirtide group and 1.2 percent in the group that
switched to enfuvirtide) discontinued treatment
with enfuvirtide because of injection-site reactions.
* There were 401 patients in the intention-to-treat population in the United
States, 66 in Canada, 15 in Mexico, and 9 in Brazil. The genotypic and pheno-
typic sensitivity scores were calculated as described in the Methods section.
AIDS denotes the acquired immunodeficiency syndrome.
Table 1. Demographic and Base-Line Characteristics of the Patients
in the Intention-to-Treat Population.*
Characteristic
Enfuvirtide
Group
(N= 326)
Control
Group
(N= 165)
Male sex — no. (%) 301 (92.3) 152 (92.1)
White race — no. (%) 274 (84.0) 135 (81.8)
Median age — yr 42.0 42.0
Median viral load (log
10
copies/ml) 5.2 5.2
Median CD4+ cell count (cells/mm
3
)75.5 87.0
Mean no. of antiretroviral drugs used previously 12.3 11.9
Mean duration of previous exposure to antiretroviral
drugs — yr
7.1 7.3
Previous antiretroviral treatment — no. (%)
≥5 Protease inhibitors
Lopinavir–ritonavir
Tenofovir
161 (49.4)
126 (38.7)
10 (3.1)
65 (39.4)
46 (27.9)
0
Previous AIDS-defining event — no. (%) 273 (83.7) 148 (89.7)
Stratum of HIV-1 RNA level and use or nonuse
of lopinavir–ritonavir, tenofovir, or both
in background regimen — no. (%)
<40,000 copies/ml
Nonuse
Use
≥40,000 copies/ml
Nonuse
Use
16 (4.9)
48 (14.7)
50 (15.3)
212 (65.0)
9 (5.5)
24 (14.5)
27 (16.4)
105 (63.6)
Phenotypic sensitivity score — no. (%)
0
1–2
3–4
≥5
Missing
90 (27.6)
137 (42.0)
80 (24.5)
15 (4.6)
4 (1.2)
40 (24.2)
72 (43.6)
44 (26.7)
6 (3.6)
3 (1.8)
Genotypic sensitivity score — no. (%)
0
1–2
3–4
≥5
Missing
52 (16.0)
169 (51.8)
90 (27.6)
13 (4.0)
2 (0.6)
22 (13.3)
93 (56.4)
45 (27.3)
3 (1.8)
2 (1.2)
Mean no. of antiretroviral drugs in background
regimen
4.0 4.0
Newly approved or investigational antiretroviral drugs
in background regimen — no. (%)
Lopinavir–ritonavir only
Tenofovir only
Lopinavir–ritonavir and tenofovir
201 (61.7)
26 (8.0)
33 (10.1)
102 (61.8)
11 (6.7)
16 (9.7)
Downloaded from www.nejm.org at Stanford University on September 29, 2003.
Copyright © 2003 Massachusetts Medical Society. All rights reserved.
n engl j med
348;22
www.nejm.org may
29, 2003
enfuvirtide for drug-resistant hiv-1 in north and south america
2181
Adverse Events at Week 24
Because of the 2:1 ratio for randomization, the de-
sign of the study allowing patients in the control
group to switch to enfuvirtide, and the lower rate
of virologic failure in the enfuvirtide group after
24 weeks of treatment, the total number of patient-
years of exposure to the randomly assigned treat-
ment regimen was approximately 2.5 times as high
in the enfuvirtide group (162.8 patient-years) as
in the control group (64.9 patient-years). After 24
weeks, 77.6 percent of patients in the enfuvirtide
group and 74.5 percent in the control group had
had an adverse event (excluding injection-site re-
actions) that was related to the treatment regimen.
Diarrhea, nausea, and fatigue were the most fre-
quently reported treatment-related adverse events
in both groups (Table 3). Peripheral neuropathy and
decreased appetite were the only treatment-related
adverse events that occurred with a frequency at least
5 percent higher in the enfuvirtide group than in the
control group.
Overall, 22 patients in the enfuvirtide group (6.7
percent) and 8 patients in the control group (4.8 per-
cent) had adverse events with onset before week 24
that led to withdrawal from the study. The most fre-
quent adverse events leading to withdrawal were
vomiting in both groups (in 1.2 percent of patients
in the enfuvirtide group and 1.2 percent of those in
the control group), nausea in both groups (in 0.9
percent of patients in the enfuvirtide group and 1.2
percent of those in the control group), and diarrhea
in the control group (1.2 percent). All other adverse
events leading to withdrawal occurred in 0.6 per-
cent of patients or less. Two patients who switched
to enfuvirtide (2.5 percent) had adverse events that
began after the switch and subsequently led to with-
drawal from the study; these events were diarrhea in
one patient and progression of the acquired immu-
nodeficiency syndrome in the other patient. Similar
percentages of patients in the two groups died (1.2
percent [four patients] in the enfuvirtide group and
2.4 percent [four patients] in the control group) or
* Plasma samples were obtained at every visit, and quantitative analysis of HIV-1 RNA levels was performed by a central laboratory (Covance Central
Laboratory Services, Indianapolis) with the Amplicor HIV-1 Monitor, version 1.5 (Roche). Samples found by this test to contain fewer than 400
copies of HIV-1 RNA per milliliter were retested with ultrasensitive preparation of the sample in order to detect as few as 50 copies per milliliter.
CD4+ cell counts were assessed with the use of standard techniques for flow cytometry. The last-observation-carried-forward method was
used for the analysis of least-squares mean changes. CI denotes confidence interval.
† A negative number represents a decrease.
Table 2. Efficacy at Week 24 in the Intention-to-Treat Population.*
Variable
Enfuvirtide
Group
Control
Group
Difference between
Groups (95% CI)
Odds Ratio
(95% CI)
P
Value
Least-squares mean change from base line in plasma HIV-1
RNA level (log
10
copies/ml)†
¡1.696 ¡0.764 0.933 (0.594– 1.271) <0.001
<50 Copies of HIV-1 RNA/ml of plasma (% of patients) 19.6 7.3 3.30 (1.70–6.39) <0.001
<400 Copies of HIV-1 RNA/ml of plasma (% of patients) 31.7 16.4 3.17 (1.96–5.13) <0.001
Reduction from base line of ≥1 log
10
copies of HIV-1 RNA
per milliliter of plasma (% of patients)
51.8 29.1 2.64 (1.77–3.95) <0.001
Least-squares mean increase in CD4+ cell count (cells/mm
3
)76.2 32.1 44.1 (22.5–65.8) <0.001
Figure 2. Time to Protocol-Defined Virologic Failure, as of Week 24.
Data were censored at the time of discontinuation of treatment. The analysis
was conducted using the intention-to-treat population.
No. at Risk
Enfuvirtide
Control
311
161
183
58
241
91
Proportion without Virologic Failure
Time to Virologic Failure (wk)
1.00
0.75
0.50
0.25
0.00 0 2 4 6 8 10 12 14 16 18 20 22 24 2826
Enfuvirtide group
Control group
Downloaded from www.nejm.org at Stanford University on September 29, 2003.
Copyright © 2003 Massachusetts Medical Society. All rights reserved.
n engl j med
348;22
www.nejm.org may
29
,
2003
The
new england journal
of
medicine
2182
had a serious adverse event while receiving the treat-
ment to which they were originally assigned (25.8
percent in the enfuvirtide group and 21.2 percent in
the control group).
Updated Safety Analysis
The updated safety analysis included 663 patients
in the enfuvirtide groups and 334 patients in the
control groups from TORO 1 and TORO 2. At the
time of the updated analysis, 229 patients originally
assigned to the control group had switched, adding
enfuvirtide to their regimen. Because of the 2:1 ratio
of randomization and the study design that allowed
switching to enfuvirtide, patients in the original en-
fuvirtide group had 813 patient-years of exposure
(median, 1.48 years per patient; range, <0.01 to
1.92) and patients in the control group had 163
patient-years of exposure (median, 0.35 year per
patient; range, 0.04 to 1.60; ratio, 5:1). Patients
exposed to enfuvirtide after switching had 214 pa-
tient-years of exposure (median, 1.08 years per pa-
tient; range, <0.01 to 1.71). Results were adjusted
for exposure (presented in terms of the number of
patients with an event per 100 patient-years of expo-
sure), with all the patients exposed to enfuvirtide
combined into one group (with a total of 1027 pa-
tient-years of exposure) and compared with the con-
trol group.
This update generally confirmed the safety pro-
file seen at 24 weeks, with the following observa-
tions. Although the overall incidence of bacterial
infections was similar in the two treatment groups
after adjustment for exposure (183 patients in the
combined enfuvirtide groups [20.5 percent], or 17.8
per 100 patient-years of exposure, and 30 patients
in the control group [9.0 percent], or 18.4 per 100
patient-years of exposure; P= 0.56), pneumonia, pri-
marily bacterial, occurred more frequently in the
combined enfuvirtide group (50 patients [5.6 per-
cent], or 4.9 per 100 patient-years) than in the con-
trol group (1 patient [0.3 percent], or 0.6 per 100
patient-years; P= 0.02). Sepsis occurred more fre-
quently in the combined enfuvirtide group (16 pa-
tients [1.8 percent], or 1.6 per 100 patient-years, vs.
2 patients [0.6 percent], or 1.2 per 100 patient-
years, in the control group), but the exposure-adjust-
ed rates were not significantly different (P= 0.37).
Two patients had cases of systemic hypersensi-
tivity reaction (both in TORO 1) that were consid-
ered to be related to enfuvirtide therapy, and both
recurred on rechallenge. In the first patient, who
was taking enfuvirtide in combination with didan-
osine, stavudine, amprenavir, and ritonavir, rash,
fever, and vomiting developed after eight days of
treatment. On rechallenge on days 17 and 22 of
the study, rash, fever, and vomiting recurred with-
in hours after the administration of enfuvirtide.
The eosinophil count increased progressively from
280 per cubic millimeter at base line (upper limit of
normal, 570 per cubic millimeter) to 350 per cubic
millimeter on day 12 of the study and 690 per cu-
bic millimeter on day 15. Membranoproliferative
glomerulonephritis developed in the second pa-
tient after 57 days of therapy with enfuvirtide in
combination with tenofovir, lamivudine, lopinavir–
ritonavir, amprenavir, and efavirenz. This patient
had a history of diabetes, seasonal allergies, pro-
teinuria, and hematuria. On rechallenge with all an-
tiretroviral drugs on day 223, severe respiratory dis-
tress developed. No eosinophilia or increase from
base line in the eosinophil count was noted.
Treatment-related eosinophilia (>700 cells per
cubic millimeter) occurred in a greater proportion
of patients in the enfuvirtide group (74 of 662 pa-
tients who could be evaluated [11.2 percent], or
11.5 patients per 100 patient-years) than in the
control group (8 of 332 patients who could be eval-
* Data are for events that are not otherwise specified or not elsewhere classified.
Table 3. Treatment-Related Adverse Events Occurring in at Least 5 Percent
of the Patients in Either Group.
Variable
Enfuvirtide Group
(N= 326)
Control Group
(N= 165)
number (percent)
Patients with ≥1 event 253 (77.6) 123 (74.5)
Adverse event
Diarrhea* 79 (24.2) 63 (38.2)
Nausea 72 (22.1) 48 (29.1)
Fatigue 64 (19.6) 28 (17.0)
Peripheral neuropathy* 36 (11.0) 9 (5.5)
Insomnia 32 (9.8) 10 (6.1)
Headache 29 (8.9) 15 (9.1)
Decreased appetite* 26 (8.0) 5 (3.0)
Vomiting* 25 (7.7) 21 (12.7)
Dizziness (except vertigo) 24 (7.4) 7 (4.2)
Weight loss 18 (5.5) 6 (3.6)
Flatulence 17 (5.2) 13 (7.9)
Downloaded from www.nejm.org at Stanford University on September 29, 2003.
Copyright © 2003 Massachusetts Medical Society. All rights reserved.
n engl j med
348;22
www.nejm.org may
29, 2003
enfuvirtide for drug-resistant hiv-1 in north and south america
2183
uated [2.4 percent], or 4.9 patients per 100 patient-
years). Eosinophilia was not associated with clini-
cal events suggestive of systemic hypersensitivity.
Aside from eosinophilia, differences between
the groups in the incidence of treatment-related
grade 3 or grade 4 laboratory abnormalities were
small, and no consistent pattern was evident to sug-
gest a definitive association of enfuvirtide with any
particular laboratory abnormality.
Our trial was designed specifically to evaluate a new
class of anti–HIV-1 compounds in patients who
have received treatment with multiple drugs. Resist-
ance testing was used to construct an optimized
background regimen for all patients in the trial.
Along with the criteria for switching to enfuvirtide
and the use of lopinavir–ritonavir and tenofovir, this
approach allowed patients access to the best possi-
ble treatment options. This fact is reflected by the
relatively high proportion of patients in the control
group with responses to treatment despite the ex-
tensive resistance to antiretroviral drugs in this pop-
ulation of patients. This positive outcome was also
due in part to the high level of adherence to treat-
ment among patients in the control group (≥85 per-
cent adherence in 90.3 percent of patients). The
benefit of enfuvirtide was demonstrated by signifi-
cant differences between the two treatment groups
at week 24 in the magnitude of the reduction in the
plasma HIV-1 RNA level, the proportion of patients
in each category of virologic response, the distribu-
tion of time to virologic failure, and the increase in
the CD4+ cell count.
Although the last-observation-carried-forward
method can overestimate the individual drug effect
if the HIV-1 RNA level rebounds quickly after viro-
logic failure, it was chosen to provide a consistent
rule for handling patients in either group who dis-
continued treatment or had virologic failure, as well
as for those who switched to enfuvirtide.
18,19
The
robustness of the primary result was confirmed
by three stringent sensitivity analyses that clearly
showed that the magnitude of the estimate of the
effect of enfuvirtide treatment was not determined
primarily by the method of imputation.
Overall, except for local injection-site reactions,
the safety and tolerability of enfuvirtide in combina-
tion with an optimized background regimen were
similar to those of the background regimen alone
over the course of 24 weeks of therapy. The safety
results obtained from the combined TORO 1 and
TORO 2 studies after longer exposure to enfuvirtide
showed a higher rate of pneumonia among patients
receiving enfuvirtide than among patients in the
control group, but the overall incidence of bacterial
infection was similar in the two groups.
Two patients had a hypersensitivity reaction that
was considered to be related to enfuvirtide thera-
py and that recurred with rechallenge. There was
a higher incidence of eosinophilia among patients
receiving enfuvirtide, even after adjustment for the
duration of exposure. Review of the cases of indi-
vidual patients with eosinophilia did not reveal any
clinical adverse events suggestive of hypersensitiv-
ity to enfuvirtide.
Injection-site reactions were the most common
events associated with enfuvirtide treatment, occur-
ring in most patients who received the drug, but
pain or discomfort requiring analgesics or limit-
ing usual activities occurred in only 8.7 percent. Only
a small number of patients discontinued enfuvirtide
therapy because of an injection-site reaction (2.8
percent of patients assigned to the enfuvirtide group
and 1.2 percent of patients who switched to enfu-
virtide). There was a high rate of adherence to en-
fuvirtide treatment, suggesting that injection-site
reactions were not treatment-limiting.
In our study, enfuvirtide resulted in significant
improvement in virologic and immunologic re-
sponses as compared with individualized, opti-
mized, combination antiretroviral therapy alone.
These findings are supported by the similar results
obtained in the TORO 2 trial.
15
These two studies
provide firm proof of principle that HIV-1 glyco-
protein 41 is a viable target for effective treatment
of HIV-1 infection. A week-48 analysis will be per-
formed in both trials to assess the durability of the
response to enfuvirtide. The introduction of enfu-
virtide as the first of this new class of antiretroviral
agent could make an important contribution to the
successful, individualized treatment of growing
numbers of patients who have limited remaining
treatment options.
Supported by Roche and Trimeris.
Dr. Lalezari reports having received consulting and lecture fees
from Roche and Trimeris. Dr. Henry reports having received con-
sulting or lecture fees from Abbott, Agouron, Bristol-Myers Squibb,
Gilead, GlaxoSmithKline, and Merck and grant support from Bris-
tol-Myers Squibb, GlaxoSmithKline, and Roche. Dr. Montaner re-
ports having received consulting or lecture fees or grant support from
Abbott, Agouron, Shire BioChem, Boehringer Ingelheim, Bristol-
Myers Squibb, Dupont Pharma, Gilead Sciences, Glaxo Wellcome,
Roche, Kucera Pharmaceutical, Merck Frosst, and Pharmacia &
Upjohn. Dr. Piliero reports having received consulting and lecture
fees from Roche Laboratories and grant support from Roche and
discussion
Downloaded from www.nejm.org at Stanford University on September 29, 2003.
Copyright © 2003 Massachusetts Medical Society. All rights reserved.
n engl j med
348;22
www.nejm.org may
29
,
2003
2184
The
new england journal
of
medicine
Trimeris. Dr. Trottier reports having received consulting or lecture
fees from Roche, Bristol-Myers Squibb, Agouron, and Abbott. Dr.
Walmsley reports having received consulting or lecture fees from
Abbott, Aguoron, Boehringer Ingelheim, Bristol-Myers Squibb,
Dupont, GlaxoSmithKline, and Merck and grant support from
Abbott, GlaxoSmithKline, and Boehringer Ingelheim. Dr. Cohen
reports having received consulting fees, lecture fees, or grant
support from Roche, Bristol-Myers Squibb, Abbott, Gilead,
Boehringer Ingelheim, GlaxoSmithKline, and Trimeris. Dr. Ku-
ritzkes reports having received consulting or lecture fees from
Abbott, Bayer–Visible Genetics, Bristol-Myers Squibb, Gilead–
Triangle, GlaxoSmithKline, Merck, Roche, Shire BioChem, Se-
rono, Trimeris, Ortho Biotech, and ViroLogic and grant support
from Roche, Trimeris, GlaxoSmithKline, Bristol-Myers Squibb,
and Tanox. Dr. Eron reports having received consulting and lec-
ture fees from Roche and grant support from Abbott, Merck,
Roche, and Trimeris.
We are indebted to the patient volunteers who took part in the trial.
appendix
In addition to the authors, the TORO 1 Study Group included the following institutions and persons: C. Farthing and E. Graham (AHF Re-
search, Los Angeles); M. Packard and L. Ngo (AIDS Clinical Research Unit, University of North Carolina at Chapel Hill, Chapel Hill); M. Le-
derman and J. Baum (AIDS Clinical Trials Unit, Case Western Reserve University, University Hospitals of Cleveland, Cleveland); R. Pollard,
S. Rauf, and W. Silkowski (AIDS Clinical Trials Unit, University of Texas Medical Branch, Galveston); M. Thompson and A. Rucker (AIDS
Research Consortium of Atlanta, Atlanta); M. Harris and G. Larsen (AIDS Research Program, St. Paul’s Hospital, Vancouver, B.C., Canada);
S. Preston and D. Cunningham (Albany Medical College, Albany, N.Y.); D. Guimaraes, A. Bertasso, and T. Kinchelow (Roche, Nutley, N.J.);
R. Myers and B. Casimir (Body Positive, Phoenix, Ariz.); P.R. Skolnik and B. Adams (Boston University Medical Center, Boston); O.H.M.
Leite and M. Oliveira (Casa da AIDS, Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, São Paulo, Brazil); E.
Lefebvre and B. Gomez (Clinique Médicale l’Actuel, Montreal); K.B. Foy (Community Research Initiative of New England, Boston); H. Lam-
piris and S. Charles (Department of Veterans Affairs Medical Center, San Francisco, and University of California, San Francisco); J. Dobkin
and M. Crawford (Division of Infectious Diseases, Columbia University, New York); T. Slom, R. Murphy, and T. Mikaitis (Division of Infec-
tious Diseases, Northwestern University, Chicago); J. Witek and R. Anthony (Drexel University College of Medicine, Philadelphia); G. Rich-
mond and V.F. Appleby (Fort Lauderdale, Fla.); F. Smaill and L. Kelleher (Hamilton Health Sciences, McMaster University Hospital, Hamil-
ton, Ont., Canada); L. Nieto and S. Trevino (Hospital Gabriel Mancera, Mexico City, Mexico); M. Schechter and B. Fonseca (Hospital
Universitario Clementino Fraga Filho, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil); E. DeJesus and R. Ortiz (Infectious
Diseases Consultants Research Initiative, Altamonte Springs, Fla.); J. Wheat, M. Goldman, and D.K. O’Connor (Division of Infectious Dis-
eases, Indiana University, Indianapolis); J.G. Sierra-Madero and S. Niño-Oberto (Instituto Nacional de Ciencias Médicas y Nutrición Salva-
dor Zubirán, Mexico City, Mexico); J.E. Gallant and L. Apuzzo (Johns Hopkins University School of Medicine, Baltimore); N. Basgoz and K.
Habeeb (Massachusetts General Hospital AIDS Clinical Trials Unit, Boston); P. Alpert and S. Thomas (Montefiore Medical Center, New
York); T. Miller and T. Kempner (Oregon Health and Science University, Portland); P.R. Wolfe and J. Bautista (Pacific Oaks Research, Beverly
Hills, Calif.); H.L. Martin and M.E. Morton (Park Nicollet Medical Center, St. Louis Park, Minn.); D. Henry and S. Kilcoyne (Pennsylvania
Oncology Hematology Associates, Philadelphia); E. Glutzer (Quest Clinical Research, San Francisco); C. Rivera-Vazquez and Z. Pomales
(Veterans Affairs Medical Center, San Juan, Puerto Rico); N. Bellos and L.A. Hoffman (Southwest Infectious Disease Associates, Dallas); B.
Olmscheid, O. Klein, and M. Miller (St. Vincent’s Medical Center, New York); C.R. Steinhart and A. Liebmann (Steinhart Medical Associ-
ates, Miami); S. Williams and L. Springate (The Vancouver Clinic, Vancouver, Wash.); K. Logue (Toronto Hospital, University of Toronto,
Toronto); L. Smiley and G.D. Miralles (Trimeris, Durham, N.C.); R. Haubrich and K. Nuffer (University of California, San Diego, San Di-
ego); G. Beatty and S. O’Leary (University of California, San Francisco, HIV Clinical Trials Group at San Francisco General Hospital, San
Francisco); D. Rouleau and S. Dufresne (Unité Hospitalière de Recherche, d’Enseignement et de Soins sur le SIDA du Centre Hospitalier de
l’Université de Montréal, Hôpital Notre-Dame, Montreal); J.M. Kilby, M. Saag, and K. Upton (University of Alabama at Birmingham, Bir-
mingham); J. Feinberg and P. Kohler (University of Cincinnati, Cincinnati); T.B. Campbell and B.A. Putnam (University of Colorado Health
Science Center, Denver); S.A. Riddler and R.R. Rosener (University of Pittsburgh, Pittsburgh); B.J. Barnett and I. Hansen (University of Tex-
as Houston Medical School, Houston); A.C. Collier and B.A. Royer (University of Washington AIDS Clinical Trials, Seattle); D.W. Haas and
M. Morgan (Vanderbilt AIDS Clinical Trials Center, Nashville); K. Sathasivam and J. Hersch (Whitman–Walker Clinic, Washington, D.C.).
references
1.
Hoetelmans RM. Sanctuary sites in
HIV-1 infection. Antivir Ther 1998;3:Suppl
4:13-7.
2.
Lafeuillade A, Solas C, Halfon P, Chada-
paud S, Hittinger G, Lacarelle B. Differenc-
es in the detection of three HIV-1 protease
inhibitors in non-blood compartments: clin-
ical correlations. HIV Clin Trials 2002;3:
27-35.
3.
Dieleman JP, Jambroes M, Gyssens IC,
et al. Determinants of recurrent toxicity-
driven switches of highly active antiretroviral
therapy: the ATHENA cohort. AIDS 2002;16:
737-45.
4.
Moyle G, Carr A. HIV-associated lipo-
dystrophy, metabolic complications, and an-
tiretroviral toxicities. HIV Clin Trials 2002;
3:89-98.
5.
Dresser GK, Spence JD, Bailey DG.
Pharmacokinetic-pharmacodynamic conse-
quences and clinical relevance of cytochrome
P450 3A4 inhibition. Clin Pharmacokinet
2000;38:41-57.
6.
Casado JL, Sabido R, Perez-Elias MJ, et
al. Percentage of adherence correlates with
the risk of protease inhibitor treatment fail-
ure in HIV-infected patients. Antivir Ther
1999;4:157-61.
7.
Raboud JM, Harris M, Rae S, Montaner
JS. Impact of adherence on duration of viro-
logical suppression among patients receiv-
ing combination antiretroviral therapy. HIV
Med 2002;3:118-24.
8.
Race E, Dam E, Obry V, Paulous S, Clavel
F. Analysis of HIV cross-resistance to pro-
tease inhibitors using a rapid single-cycle
recombinant virus assay for patients failing
on combination therapies. AIDS 1999;13:
2061-8.
9.
Van Vaerenbergh K, Van Laethem K,
Albert J, et al. Prevalence and characteristics
of multinucleoside-resistant human immu-
nodeficiency virus type 1 among European
patients receiving combinations of nucleoside
analogues. Antimicrob Agents Chemother
2000;44:2109-17.
10.
Wild C, Greenwell T, Matthews T. A syn-
thetic peptide from HIV-1 gp41 is a potent
inhibitor of virus-mediated cell-cell fusion.
AIDS Res Hum Retroviruses 1993;9:1051-3.
11.
Kilby JM, Hopkins S, Venetta TM, et al.
Potent suppression of HIV-1 replication in
humans by T-20, a peptide inhibitor of gp41-
mediated virus entry. Nat Med 1998;4:1302-7.
12.
Kilby JM, Lalezari JP, Eron JJ, et al. The
safety, plasma pharmacokinetics, and anti-
viral activity of subcutaneous enfuvirtide
(T-20), a peptide inhibitor of gp41-medi-
ated virus fusion, in HIV-infected adults.
AIDS Res Hum Retroviruses 2002;18:685-
93. [Erratum, AIDS Res Hum Retroviruses
2003;19:83.]
13.
Lalezari JP, DeJesus E, Northfelt DW, et
Downloaded from www.nejm.org at Stanford University on September 29, 2003.
Copyright © 2003 Massachusetts Medical Society. All rights reserved.
n engl j med
348;22
www.nejm.org may
29, 2003
enfuvirtide for drug-resistant hiv-1 in north and south america
2185
al. A controlled phase II trial assessing three
doses of enfuvirtide (T-20) in combination
with abacavir, amprenavir, ritonavir, and
efavirenz in non-nucleoside reverse trans-
criptase inhibitor-naive HIV-infected adults.
Antiviral Ther (in press).
14.
Lalezari JP, Eron JJ, Carlson M, et al.
A phase II clinical study of the long-term
safety and antiviral activity of enfuvirtide-
based antiretroviral therapy. AIDS 2003;17:
691-8.
15.
Lazzarin A, Clotet B, Cooper D, et al.
Efficacy of enfuvirtide in patients infected
with drug-resistant HIV-1 in Europe and
Australia. N Engl J Med 2003;348:2186-95.
16.
DeGruttola V, Dix L, D’Aquila R, et al.
The relation between baseline HIV drug
resistance and response to antiretroviral ther-
apy: re-analysis of retrospective and prospec-
tive studies using a standardized data analy-
sis plan. Antivir Ther 2000;5:41-8.
17.
AIDS Clinical Trials Group. Table of
grading severity of adult adverse experiences.
Rockville, Md.: Division of AIDS, National
Institute of Allergy and Infectious Diseases,
1996.
18.
Little R, Yau L. Intent-to-treat analysis
for longitudinal studies with drop-outs. Bio-
metrics 1996;52:1324-33.
19.
Le Corfec E, Chevret S, Costagliola D.
Visit-driven endpoints in randomized HIV/
AIDS clinical trials: impact of missing data
on treatment difference measured on sum-
mary statistics. Stat Med 1999;18:1803-
17.
Copyright © 2003 Massachusetts Medical Society.
apply for jobs electronically at the new nejm career center
Physicians registered at the new NEJM Career Center can now apply for jobs electron-
ically using their own cover letters and CVs. You can now keep track of your job-applica-
tion history with a personal account that is created when you register with the Career
Center and apply for jobs seen online at our Web site. Visit www.nejmjobs.org for
more information.
Downloaded from www.nejm.org at Stanford University on September 29, 2003.
Copyright © 2003 Massachusetts Medical Society. All rights reserved.
