Enfuvirtide

A synthetic 36-amino-acid peptide copied from the HR2 region of HIV-1 gp41 that blocks fusion of the virus with CD4 cells, granted accelerated approval as Fuzeon on March 13, 2003 for treatment-experienced HIV-1 infection and discontinued in the United States in 2025.

Also referenced as: Fuzeon, T-20, DP-178, Pentafuside

Also appears in: Other

COA Public product evidence No certificate records are bound to a tracked Enfuvirtide product page A record counts here only when a vendor publishes a certificate against one of its own tracked Enfuvirtide product pages. Other certificates naming this compound may still exist — search the wider certificate database. →

---

What is Enfuvirtide?

Enfuvirtide is a linear synthetic peptide of 36 amino acids, built from naturally occurring L-residues with the N-terminus acetylated and the C-terminus a carboxamide, empirical formula C204H301N51O64 and molecular weight 4,492. Its sequence is copied from the second heptad-repeat region (HR2) of the HIV-1 transmembrane glycoprotein gp41, which is why it was first described under the laboratory name DP-178 and then developed as T-20 and pentafuside before receiving the nonproprietary name enfuvirtide. It is an antiretroviral, not a signaling peptide: it has no human receptor target and works entirely by interfering with the virus’s entry machinery.

Marketed as Fuzeon, it became the first HIV fusion inhibitor to reach market approval and defined a new antiretroviral class. Because its indication was always limited to treatment-experienced patients and it required subcutaneous injection, its use fell as oral integrase inhibitors and other better-tolerated options arrived. It has no meaningful presence in the research-peptide vendor market, appearing, where it appears at all, only as a milligram-scale laboratory reagent in biochemical catalogs.

How it works

HIV-1 enters a CD4 cell when gp41 refolds into a six-helix bundle that pulls the viral and cellular membranes together. Enfuvirtide inserts itself into that refolding step:

• A decoy HR2 helix — the peptide corresponds to residues 643–678 of the HIV-1 LAI isolate, in the C-terminal heptad repeat of gp41. In the original cell-culture work this peptide, then called DP-178, blocked 100% of virus-mediated cell-cell fusion below 5 ng/mL (IC90 about 1.5 ng/mL) and cut the infectious titer of cell-free virus roughly tenfold at about 80 ng/mL, at concentrations 10,000 to 100,000 times below those producing cytotoxicity. Inhibition was HIV-1 specific, requiring 100 to 1,000 times more peptide against an HIV-2 isolate, and blocked a step before reverse transcription (Wild et al., Proceedings of the National Academy of Sciences, 1994, 91(21):9770–9774).
• Binding the exposed HR1 groove — after gp120 engages CD4 and a coreceptor, the N-terminal heptad repeat (HR1) of gp41 is transiently exposed. Enfuvirtide binds this HR1 region and prevents the native HR2 segment from packing against it, so the six-helix bundle never forms and the membranes never fuse. The approved labeling describes the mechanism in these terms: inhibition of the fusion of HIV-1 with CD4 cells, through a block on gp41 structural rearrangement at a pre-fusion conformation.
• Proof of antiviral effect in humans — in the first clinical study, 16 HIV-infected adults received intravenous T-20 as monotherapy for 14 days in four dose groups (3, 10, 30 and 100 mg twice daily), with dose-related declines in plasma HIV RNA. All four participants in the highest group fell below 500 copies/mL on the bDNA assay, with a median 1.96 log10 decline on a more sensitive RT-PCR assay, the first demonstration that blocking viral entry works in vivo (Kilby et al., Nature Medicine, 1998, 4(11):1302–1307).
• A narrow, site-specific resistance pathway — because the drug binds a short, defined stretch of HR1, substitutions in a ten-residue motif spanning gp41 positions 36 to 45 (most often at G36, V38 and N43) reduce susceptibility and emerge readily under incomplete suppression. Primary resistance has not been observed in enfuvirtide-naive isolates, resistant mutants replicate poorly compared with wild type, and reversion toward a drug-sensitive state has been reported after withdrawal; isolates resistant to the reverse transcriptase and protease inhibitor classes remain susceptible to enfuvirtide in cell culture (Greenberg and Cammack, Journal of Antimicrobial Chemotherapy, 2004, 54(2):333–340; FDA label, microbiology section).

Research status

Enfuvirtide has a complete Phase 3 program, more than two decades of postmarketing experience, and no active development.

TORO 1 (NCT00008528, labeled T20-301). This open-label Phase 3 trial in North and South America randomized 501 patients with prior exposure or documented resistance to all three then-available antiretroviral classes, 2:1, to enfuvirtide plus an optimized background regimen or the background regimen alone; 491 received at least one dose and had a post-treatment viral load. At week 24 the least-squares mean decrease in HIV-1 RNA was 1.696 log10 copies/mL with enfuvirtide versus 0.764 with background alone (P < 0.001), and CD4 counts rose by 76 versus 32 cells/mm³ (P < 0.001). Injection-site reactions were reported by 98% of enfuvirtide recipients, and pneumonia was more frequent in the enfuvirtide group (Lalezari et al., New England Journal of Medicine, 2003, 348(22):2175–2185).

TORO 2 (NCT00021554, labeled T20-302). The companion trial in Europe and Australia randomized 512 patients on the same design; 335 who received enfuvirtide plus background and 169 who received background alone were evaluable. Patients entered with a median of 7 years of prior treatment and 12 antiretrovirals received. The least-squares mean change in HIV-1 RNA at week 24 was −1.429 versus −0.648 log10 copies/mL (difference 0.781; P < 0.001), and CD4 counts rose by 65.5 versus 38.0 cells/mm³ (P = 0.02) (Lazzarin et al., New England Journal of Medicine, 2003, 348(22):2186–2195). The two trials were published together in the May 29, 2003 issue, after the FDA had already granted accelerated approval on the 24-week data.

48-week results. The two trials randomized 1,013 patients between them, and the pooled 48-week analyses drew on 997 of them (663 assigned to enfuvirtide, 334 to background alone). A total of 73.7% of enfuvirtide-randomized patients remained on their assigned treatment through week 48 versus 21.3% of control patients, 220 of whom switched to enfuvirtide at virologic failure under the protocol’s switch provision. The approved label reports the pooled week-48 outcomes as a virologic response of at least 1 log10 below baseline in 304 patients (46%) versus 61 (18%), HIV-1 RNA below 50 copies/mL in 154 (23%) versus 27 (8%), and a mean CD4 gain of 91 versus 45 cells/mm³ (Nelson et al., Journal of Acquired Immune Deficiency Syndromes, 2005, 40(4):404–412; FDA label). The companion safety analysis found that the week-24 safety profile held through week 48, with injection-site reactions leading to discontinuation in 4.4% of patients (Trottier et al., Journal of Acquired Immune Deficiency Syndromes, 2005, 40(4):413–421).

Injection-site pathology. Biopsies from seven patients receiving enfuvirtide showed an inflammatory response consistent with a localized hypersensitivity reaction in every specimen, including one from a patient with no clinical reaction, with a pattern resembling granuloma annulare and interstitial granulomatous drug reaction and the greatest changes where enfuvirtide had been deposited (Ball and Kinchelow, Journal of the American Academy of Dermatology, 2003, 49(5):826–831). The label’s clinical picture matches: erythema, induration, nodules or cysts and mild to moderate pain, lasting three to seven days in 41% of patients and more than seven days in 24%, with injection-site infection in 1.7% of adults.

Pediatric experience. The label describes 63 pediatric participants aged 5 through 16 years in two open-label trials, T20-204 (11 participants aged 6 to 12) and T20-310 (52 participants aged 5 and older). Injection-site infections (cellulitis or abscess) were more frequent in adolescents than adults, with four events in 3 of 28 (11%) participants. The pediatric indication is set by weight, at least 11 kg, and rests on studies in participants 6 years of age and older.

Later use and regulatory record. The March 13, 2003 approval was an accelerated approval under 21 CFR 314.510, conditioned on submitting the 48-week reports of T20-301 and T20-302; the efficacy supplement carrying those data was approved October 15, 2004. As integrase inhibitors became available, investigator-led switch studies asked whether patients suppressed on an enfuvirtide-containing regimen could replace the injection with an oral agent — among them ANRS 138 EASIER (NCT00454337), a completed French Phase 3 non-inferiority trial of 170 patients switching enfuvirtide to raltegravir. The label addresses administration with the Biojector 2000 needle-free device and carries a specific warning for it, covering nerve pain lasting up to six months at sites where large nerves run close to the skin, along with bruising and hematomas. The last approved US labeling supplement is dated August 15, 2019. Drugs@FDA now lists the 90 mg vial as discontinued, and the current DailyMed listing gives the kit’s marketing end date as October 31, 2025; NDA 021481 has not been withdrawn. The European marketing authorization, granted May 27, 2003 under exceptional circumstances (lifted July 8, 2008), is still listed as authorized. As of November 2026, a registry search returns no active study of enfuvirtide as an investigational product; its only appearance in an ongoing trial is as one of the complex baseline regimens participants leave behind in a Gilead switch study (NCT05502341, ARTISTRY-1). No new indication, formulation, or generic was approved in 2025 or 2026.

Common dosage forms

• Approved product — Fuzeon was supplied as a convenience kit of 60 single-dose vials of lyophilized enfuvirtide powder, each labeled 90 mg and filled with 108 mg, packaged with 60 single-dose vials of sterile water for injection, 60 3 mL reconstitution syringes, 60 1 mL administration syringes, and the package insert, patient package insert and instructions for use.
• Needle-free jet-injector presentation — the Biojector 2000 needle-free device is named in the approved labeling as an administration option, with its own warning section, though the kit’s instructions for use describe conventional syringes.
• Intravenous formulation — used only in the 1998 Phase 1 proof-of-concept study; never developed commercially.
• Research supply — where enfuvirtide is listed, it is as lyophilized reagent-grade powder in biochemical reagent catalogs, typically in 1 mg to 100 mg quantities labeled for laboratory use only.

Key considerations

• An approved drug with no US supply. The FDA Approved status reflects that NDA 021481 remains an approved application for a drug product containing enfuvirtide. Drugs@FDA lists the 90 mg vial as discontinued rather than withdrawn; the end of US supply was a commercial outcome, not a safety action, but it means the molecule is now of historical and mechanistic interest in the United States rather than a clinical option. Marketing authorizations may persist in other jurisdictions after supply ends.
• Injection-site reactions were nearly universal. Reactions occurred in 98% of enfuvirtide recipients in the pivotal trials, with the localized hypersensitivity pathology described above, and injection-site infections were a particular problem in adolescents. This tolerability burden is the reason usually given for its displacement once oral alternatives for multidrug-resistant HIV became available.
• Bacterial pneumonia and systemic hypersensitivity are labeled signals. Bacterial pneumonia occurred in 6.9% of treated participants, 6.7 events per 100 patient-years versus 0.6 on control, in the pooled Phase 3 analysis; about half of those patients required hospitalization and three deaths in the enfuvirtide arm were attributed to pneumonia, all in patients with serious concomitant AIDS-related illness. A required postmarketing observational cohort found 3.2 events per 100 patient-years in both treated and untreated groups, and the label concludes it is unclear whether the increased incidence is drug-related while advising monitoring. Systemic hypersensitivity reactions occurred in under 1% of participants, and the label directs that therapy not be restarted after one.
• Resistance is fast and site-specific, and the drug was never indicated alone. Among 277 patients with protocol-defined virologic failure at 48 weeks, the median decrease in susceptibility was 33.4-fold (range 0.4- to 6,318-fold), and all but three of the 249 isolates with more than a fourfold loss carried substitutions in gp41 HR1 positions 36 to 45. Enfuvirtide was therefore only ever indicated as part of a combination regimen built with resistance testing. There are no data in antiretroviral-naive patients, and the drug has not been studied in people without HIV-1 infection, where anti-enfuvirtide antibodies could in theory produce a false-positive HIV ELISA.
• Naming and class confusion. T-20, DP-178, pentafuside, enfuvirtide, and Fuzeon are the same molecule. Albuvirtide, a long-acting fusion inhibitor approved in China, is a different and chemically distinct peptide; results from albuvirtide trials do not describe enfuvirtide.