Cytokine Dysregulation & Lipodystrophy

Lipodystrophy in patients with HIV-1 infection: effect of stopping protease inhibitors on TNF-receptor levels, insulin resistance, and lipid profile. Although the follow-up was reported below as only 3 months, CT scans did not show body fat improvements after stopping PI therapy. In a second study reported at Lipodystrophy Workshop by Steve Deeks & Rebecca Hoh from UCSF, after 12 weeks following therapy interruption body changes did not improve,

Recently growing in popularity as a possible causative factor for lipodystrophy is cytokine dysregulation. It's possible that the dysregulation which might lead to lipodystrophy could be due to HIV or HIV therapy or both. It's also been suggested that TNF-alpha, a cytokine, dysregulation may be associated with lipodystrophy. TNF-a may lead to fat cell death. This has been recognized and studies are ongoing. The study below suggests an association of high TNF-a levels with lipodystrophy but does not conclude there is a cause-effect relationship. Moyle has written a report for NATAP web site discussing these issues. Here's the link to the article or to the web site:

Perspectives on Lipodystrophy: What have we learned about the cause?

B. Maher 1 , J. Lloyd 1 , E.G.L. Wilkins 3 , W.D. Fraser 2 , D.J. Back 1 , B.K. Park 1 and M. Pirmohamed 1
1 Department of Pharmacology and Therapeutics, and 2 Clinical Chemistry, The University of Liverpool, and 3 North Manchester General Hospital, Manchester, UK

Purpose of the study:
To determine whether the metabolic and morphological abnormalities seen in patients with lipodystrophy (LD) are reversed by stopping protease inhibitors (PIs).

Ten patients with lipodystrophy were studied while on PIs and 3

months after stopping PIs. In addition, 10 HIV subjects on PIs without lipo-dystrophy were also studied. All patients were assessed by CT scans and bio-impedance analysis for body fat content. Patients were also tested for insulin resistance, glucose intolerance, hyperlipidaemia, TNF-• and TNF receptor levels before and after stopping PIs.

Summary of results:
TNF receptor levels were lower in controls (P =0.02) than in patients with LD; however, there was a significant decrease (P =0.02) in the LD patients on stopping PIs (Table). TNF-a levels were undetectable. Insulin levels in LD patients (17.9 mU/l) on PIs were significantly (P =0.02) higher than in controls (6.6 mU/l), and did not alter (21.2 mU/l) when Pis were stopped, despite equivalent glucose tolerance. CT scan appearances did not change on stopping PIs, although bioimpedance analysis showed a significant decrease from 406 to 373 ohms (P =0.03). Apart from HDL levels there was no change in lipid parameters on stopping PIs.

Our results show a reversal in some but not all of the abnormalities associated with LD on stopping PIs. TNF-a activity may be modulated by PIs and may be responsible for some of the features of LD. Our results are also consistent with the hypothesis that many of the features of LD are due to an interaction between the disease and drug therapy rather than being an effect of PIs per se.



LD on PIs

LD off PIs

TNF receptor I

1406 377

2026 723

1680 585

TNF receptor II

3868 1042

5982 1897

5170 1876