Tesamorelin First Phase 3 Trial: NCT00123253

NCT00123253 was a randomized, double-blind, placebo-controlled phase 3 trial in adults with HIV receiving antiretroviral therapy and having excess abdominal fat. Of 412 randomized participants, 410 received assigned treatment. The paper reported a mean 15.2% reduction in computed-tomography-measured visceral adipose tissue with tesamorelin versus a 5.0% increase with placebo over the 26-week main phase. Some lipid measures changed, but the average glycemic comparisons were not statistically significant.[1][2][6]

These findings concern a selected population, an imaging endpoint, and the sponsor's trial formulation. They do not establish general weight-loss efficacy, fewer cardiovascular events, permanent benefit, or equivalence of another tesamorelin-labeled material. The missing-data method, adverse-event withdrawals, and later finished-product warnings belong beside the efficacy result.

Study at a glance

The paper reports recruitment at 43 sites, whereas the current registry lists 42 locations. A present-day location list is not treated as a reconstruction of every historical recruiting site. The paper describes stratified randomization and masking of participants, investigators, and central CT readers.[2][6]

Endpoint, analysis population, and missing data

The primary endpoint was the percentage change in visceral adipose tissue area measured on a CT slice at the L4–L5 level. The statistical methods describe an efficacy analysis using participants who received at least one assigned treatment, with last observation carried forward (LOCF) for missing follow-up. A supportive per-protocol analysis restricted the population to completers without significant protocol violations.[6]

LOCF replaces an unavailable later measurement with an earlier observation. It does not recover the missing measurement or establish what would have happened under continued treatment. A per-protocol result also does not erase the consequences of excluding noncompleters. The report states that the supportive visceral-fat findings were similar, but those per-protocol numerical results were not displayed.[6]

The treatment comparison used analysis of covariance. Additional models considered stratification variables, center, and relevant concomitant therapies. Nonnucleoside reverse-transcriptase inhibitor use differed at baseline; the reported imaging/lipid findings remained significant after the stated adjustments. This does not establish that every possible source of bias was removed.[6]

All reported p-values were two-sided and unadjusted for multiple testing. The primary imaging question should be distinguished from the many secondary, safety, and exploratory comparisons. No standalone full protocol or statistical analysis plan was obtained to independently reconstruct all planned analyses. See trial estimands and missing data and confidence intervals, p-values, and multiplicity.

Primary visceral-fat result and the label's different analysis

The paper's Table 2 reports:

The tabulated between-group area difference was −32.9 cm², with a 95% confidence interval of −40.7 to −25.0; the primary comparison had p<0.001. The difference between the reported percentage changes was −20.2 percentage points. A change in fat area is not a change in total body weight.[1][6]

The current Egrifta WR label presents Study 1 using least-squares means and a logarithmic model for relative VAT change. It gives an area difference of −31 cm² (95% CI −39 to −24) and modeled percentage changes of −18% versus +2%. Those source-labeled estimates are not silently substituted for the paper's tabulated means or explained as rounding alone.[4][5]

The label also describes how Egrifta WR evidence relies on earlier Egrifta clinical studies plus demonstrated comparable bioavailability. That regulatory bridge does not mean the newer formulation was the product used in this 2005–2006 trial, or that an unrelated material has the same evidence.[4][5]

This is an imaging-based fat-distribution finding, not a direct test of myocardial infarction, mortality, or long-term cardiovascular benefit. The surrogate endpoints and clinical outcomes guide explains why those distinctions matter.

Metabolic, hormonal, and patient-reported outcomes

The paper reports several secondary changes:

The rise in insulin-like growth factor I is a hormonal finding, not inherently a benefit. The paper reported no significant between-group differences in fasting glucose, two-hour glucose, or insulin; their p-values were 0.28, 0.17, and 0.93. Body-mass index and C-reactive protein comparisons were also nonsignificant.[6]

Patient-reported “belly image distress” and “belly profile” comparisons nominally favored tesamorelin, each at p=.03, but the “belly size” comparison was null (p=.70). These are distinct questionnaire outcomes, with no multiplicity adjustment, not evidence that every aspect of body image improved.[6]

A nonsignificant mean glucose result does not establish no diabetes risk. Current labeling draws on a broader clinical program and explicitly warns about glucose intolerance and diabetes; its reported diabetes comparison is separate from the first paper's average glycemic measures.[4][5] Read what negative and null results mean before turning a null comparison into an equivalence or safety claim.

Participant flow and discontinuation

The paper's flow diagram distinguishes assignment, exposure, and completion:

Two participants assigned tesamorelin did not receive it. The reported discontinuation percentages, 22.7% versus 16.1%, 20.5% overall, use the treated denominators of 273, 137, and 410. If the question instead concerns everyone randomized, 86 of 412 did not complete, including the two who never received treatment. The denominators cannot be interchanged.[6]

The between-group discontinuation test had p=.12. That is not proof that missingness had no effect on the efficacy estimate. The current label also uses 273/137 in its Study 1 efficacy tables; these are not the paper's 275/137 randomized allocation.[4–6]

The reported high vial-count adherence among participants should not be substituted for a completion rate. The extension enrolled and treated 315 participants after the main phase, with a new assignment structure; its outcomes are not another observation on the unchanged original randomized groups.[6]

Adverse events and reporting limits

The paper's Table 3 uses the treated population:

Nonsignificant overall-event and serious-event comparisons do not establish equal safety. The greater frequency of treatment-related events and event-related withdrawal remains important. Joint symptoms, swelling, and local reactions were among the reported reasons for stopping.[6]

There is an unresolved reporting difference within the paper: Figure 1 assigns adverse events as the discontinuation reason for 25 active and three placebo participants, whereas Table 3 gives the percentages above. The supplementary adverse-event appendix was not obtained, so the two presentations are retained with their source labels rather than forced into one reconciled count.[6]

Four serious events were reported as possibly related to tesamorelin: peripheral neuropathy, febrile diarrhea with dehydration, decreased mobility, and congestive heart failure. No deaths occurred in the randomized main phase. The later extension reported two deaths among tesamorelin recipients; investigators did not attribute the extension's serious events to study drug. The main-phase no-death statement must not be expanded to the entire study.[6]

Six tesamorelin participants (2.2%) developed urticaria extending beyond the local site and stopped treatment. The paper reports anti-tesamorelin IgG antibodies in 48.6% of active participants and 2.7% of placebo participants. It did not display a complete antibody-analysis denominator in the accessible main report. The cited finding is not proof of long-term clinical harmlessness; the label separately explains assay-dependent immunogenicity interpretation.[5][6]

Sponsor role and conflicts

Theratechnologies funded and designed the study in consultation with investigators Falutz and Grinspoon. Quintiles Canada performed the statistical analyses; Phase V assessed body-image outcomes. Grinspoon drafted the manuscript, and the decision to publish was shared by Grinspoon and Theratechnologies. He vouched for the data and analyses.[6]

Several authors were sponsor employees with equity interests; other consulting, lecture, grant, and equity relationships were disclosed. A trial data-and-safety monitoring board was described, but neither that board nor the journal publication is an independent scientific review of this webpage. This summary does not reanalyze individual participant data.

Current FDA status and formulation boundaries

FDA originally approved Egrifta under NDA 022505 on November 10, 2010. The former NDA was deemed to be BLA 022505 on March 23, 2020; Drugs@FDA currently lists Theratechnologies. At this check, the latest linked supplement/label was S020 dated March 25, 2025. DailyMed's Egrifta WR page was updated July 29, 2026, while its prescribing text is revised March 2025. These are distinct regulatory and document dates.[3–5]

The finished-product indication concerns reduction of excess abdominal fat in HIV-infected adults with lipodystrophy. Label limitations include no indication for weight-loss management, no established long-term cardiovascular safety, and no evidence of improved antiretroviral adherence. Egrifta WR and Egrifta SV are expressly not substitutable.[4][5]

Current warnings include elevated IGF-I, fluid retention, glucose intolerance/diabetes, hypersensitivity, and malignancy-related concerns. The label contraindicates use in active malignancy, pregnancy, specified hypothalamic-pituitary disruption, and known hypersensitivity. This is context for interpreting a historical trial, not treatment advice or a list sufficient for clinical decision-making.[4][5]

Approval and formulation bridging concern the named finished products and their evidence. Neither the paper nor an ingredient name authenticates a lot or establishes quality or clinical equivalence of separately sourced tesamorelin. See approved drugs versus research materials, checking trial and FDA status, and the growth-hormone and endocrine research hub.

Frequently asked questions

What did the primary endpoint measure?

Percentage change in CT-measured visceral adipose tissue area from baseline to week 26, not total weight loss or cardiovascular events.

Were all 412 randomized participants treated?

No. The paper shows 275/137 assigned to tesamorelin/placebo and 273/137 receiving treatment. Analysis and withdrawal denominators must remain explicit.[6]

Why do the paper and label give different VAT numbers?

The label reports least-squares and logarithmic-model summaries, while the paper displays its own mean changes and area difference. The estimates should retain their source and method rather than be treated as rounding variants.[4–6]

Did average glycemic measures differ significantly?

The first paper's fasting glucose, two-hour glucose, and insulin comparisons were nonsignificant. That does not override later labeling's glucose-intolerance/diabetes warning.[4–6]

Does a nonsignificant withdrawal-rate test remove missing-data concerns?

No. LOCF and per-protocol analyses still rely on assumptions, and the .12 withdrawal comparison does not establish an absence of bias.

Were there no deaths throughout the whole study?

No deaths were reported in the main randomized phase, but two were reported among tesamorelin recipients during the extension. Keep phase and investigator attribution separate.[6]

Does finished-product approval cover another tesamorelin material?

No. The approved product, formulation, indication, and regulatory evidence are specific; a shared name does not establish equivalence or lot quality.

Primary and official sources

1. Primary paper record, PMID 18057338
2. ClinicalTrials.gov: NCT00123253
3. FDA Drugs@FDA: NDA-to-BLA 022505 history
4. FDA Egrifta WR label, S020, revised March 2025
5. DailyMed: Egrifta WR prescribing information, updated July 29, 2026
6. Published primary paper PDF hosted by the Association of Nurses in AIDS Care

Review and corrections

Sources checked September 2, 2026 (America/Edmonton) / September 3, 2026 (UTC). The complete 12-page published paper, relevant current labeling, approval history, and fresh official registry/PubMed records were reviewed. The external journal supplement, standalone protocol/SAP, and individual participant data were not obtained. The registry has no posted summary results or documentSection, so it does not independently confirm the numerical results. PubMed returned two commentary relationships, not a correction notice; that retrieval is not a guarantee about notices elsewhere.

Serum Research is responsible for this editorial summary. No independent scientific, clinical, statistical, analytical, or regulatory review of this page is claimed. It is not medical advice, a use recommendation, or a quality/equivalence conclusion. Read how we review evidence, browse the study index and research library, or report a correction.