Egrifta (Tesamorelin) Evidence Base Graded by GRADE

What GRADE rates, and what it does not

GRADE assigns a certainty level (high, moderate, low or very low) to one outcome at a time, alongside a separate judgment about strength of recommendation [7]. It is not a single score for a drug. Applied to tesamorelin, that means separate verdicts for visceral fat, for triglycerides, for cardiovascular events and for long-term safety.

The grades below are this page's own application of the framework's domains (risk of bias, inconsistency, indirectness, imprecision and publication bias) to the trials.

The visceral fat endpoint: moderate to high certainty

Approval rests on two multicenter, randomized, double-blind, placebo-controlled trials that used 2 mg of the original 1 mg/vial formulation daily [5].

Study 1 enrolled 412 HIV-infected adults with excess abdominal fat, 86% of them men [1]. Visceral adipose tissue on CT fell 15.2% on tesamorelin and rose 5.0% on placebo over 26 weeks [1].

Study 2 randomized 404 patients 2:1 and found visceral fat down 10.9% (-21 cm2) versus 0.6% (-1 cm2) on placebo at six months (P<0.0001), with no change in glucose parameters [2]. Pooled across both trials (806 patients: 543 on tesamorelin, 263 on placebo) the treatment effect was -15.4%, with visceral fat down 24 cm2 against a 2 cm2 rise [8].

The replication matters. A trial run at Massachusetts General Hospital randomized 50 HIV-infected men and women, 28 to tesamorelin and 22 to placebo, for six months; visceral fat fell 34 cm2 versus a rise of 8 cm2, a treatment effect of -42 cm2 (95% CI -71 to -14) [3].

Domain by domain: both pivotal trials were randomized, double-blind and placebo-controlled. Consistency is good: the two pivotal trials and the academic trial all found visceral fat reductions [1][2][3]. The pooled analysis is an analysis of those two pivotal trials, not a third independent trial [8]. Precision is adequate for the primary endpoint, because the confidence intervals exclude zero [3][5], though the 50-patient trial has a wide interval [3]. Publication bias cannot be judged from the trial reports themselves.

That combination supports moderate to high certainty for the fat endpoint itself.

Where certainty falls: triglycerides, liver fat, body image

Triglycerides moved in the same direction across trials: -50 mg/dL versus +9 mg/dL in Study 1 (P<0.001) [1], and -37 mg/dL versus +6 mg/dL in the pooled analysis [8]. A lipid number is a marker, not a clinical event, so it carries less certainty than the fat endpoint. Self-assessed body image was a secondary endpoint in Study 1 [1].

Liver fat has two randomized trials behind it. In the 50-patient trial, median liver fat (lipid-to-water percentage) fell 2.0% on tesamorelin versus a 0.9% rise on placebo (P=.003) [3]. In the 12-month multicenter trial, participants with HIV and hepatic fat of at least 5% had an absolute liver-fat reduction of 4.1 percentage points more than placebo (95% CI -7.6 to -0.7), a 37% relative reduction [4]. Glucose and HbA1c did not differ [4]. This is not an approved use, and liver histology effects remain unproven [4].

Cardiovascular events and long-term safety: very low certainty

The trials covered visceral and liver fat, lipids, glucose, IGF-1, body image, other body-composition measures and safety outcomes. Cardiovascular events were not among the endpoints listed in these trial reports [1][2][3][4]. The labels state that long-term cardiovascular safety has not been established, and say to consider the risk and benefit of continuing treatment in patients who have not had a reduction in visceral adipose tissue [5][6]. For that reason this page rates any claim that tesamorelin protects the heart as very low certainty.

The label also states what the drug is not for: it is not indicated for weight loss management because its effect on weight is neutral, and there are no data showing improved adherence to antiretroviral therapy [5].

How certain are the harms?

Some harms are quantified tightly, others barely at all.

IGF-1: among patients treated 26 weeks, 47% had IGF-1 above 2 SDS and 36% above 3 SDS; among those treated 52 weeks, 34% and 23% [5]. That the hormone rises is high certainty. Whether the rise causes harm is very low certainty, and the label says the effects of prolonged IGF-1 elevation are unknown [5].

Glucose: HbA1c of 6.5% or higher developed in 5% on tesamorelin versus 1% on placebo by week 26, with a hazard ratio of about 3.3 (95% CI 1.4 to 9.6) [5]. The label says to evaluate glucose before starting and monitor during therapy, and to monitor patients with diabetes for retinopathy [5].

Fluid retention and injection sites: injection site reactions occurred in 25% versus 14% during the first 26 weeks [5]. In the label's combined table, joint pain was 13% versus 11% on placebo, peripheral edema 6% versus 2% and muscle pain 6% versus 2% [5]. Hypersensitivity reactions occurred in 4%; the label says patients should seek prompt medical attention and treatment should be discontinued if one is suspected [5].

Neoplasms: the label lists an increased risk of neoplasms, asks prescribers to weigh the higher background cancer risk in HIV-positive patients, and says to discontinue if malignancy recurs [5]. Certainty about the size of any excess risk is very low.

What happens when the drug stops

In the label's extension phase, visceral fat change from week 26 to week 52 was 0% to -5% in patients re-randomized to continue tesamorelin versus +16% to +22% in those re-randomized to placebo [5]. The fat loss held only while treatment continued; it is not a permanent change in body composition [5].

Off-label uses and their grades

Cognitive function in older adults: a randomized, double-blind, placebo-controlled trial in 152 adults aged 55 to 87, 66 of them with mild cognitive impairment, used tesamorelin 1 mg nightly for 20 weeks and reported a favorable overall effect on cognition (P=.03) [9]. This is one trial, at a different dose from either marketed product, and not an approved use [5]. Certainty for any clinical benefit is very low.

Use outside HIV-associated lipodystrophy is outside the label [5]; the visceral-fat and liver-fat trials described above enrolled people with HIV [3][4].

Which product the evidence actually covers

The pivotal trials used the original 1 mg/vial formulation at 2 mg daily, while the current labels describe Egrifta SV at 1.4 mg and Egrifta WR at 1.28 mg, each injected subcutaneously into the abdomen once daily [5][6]. The WR label says the two current products are not substitutable and differ in dose, number of vials, reconstitution and storage [6]. FDA accepted the SV 1.4 mg dose on the basis of comparable exposure to the 2 mg trial dose [5]. The trial numbers on this page therefore come from the formulation and dose used in the pivotal studies, not from the currently labeled doses [5][6].