Skip to content

Habashi JP, Judge DP, Holm TM, Cohn RD, Loeys BL, Cooper TK, Myers L, Klein EC, Liu G, Calvi C, Podowski M, Neptune ER, Halushka MK, Bedja D, Gabrielson K, Rifkin DB, Carta L, Ramirez F, Huso DL, Dietz HC. Losartan, an AT1 antagonist, prevents aortic aneurysm in a mouse model of Marfan syndrome. Science. 2006;312:117-21. PMID 16601194

One-paragraph summary

Building on the finding that fibrillin-1 deficiency dysregulates TGF-β activation (Neptune 2003, PMID 12598898), the authors tested whether aortic aneurysm in a mouse model of Marfan syndrome is TGF-β–driven and therefore preventable. Aortic aneurysm in the Marfan mouse was associated with increased TGF-β signalling and was prevented both by a TGF-β–neutralising antibody and by the angiotensin II type 1 receptor blocker losartan. AT1 antagonism also partially reversed non-cardiovascular manifestations, including impaired alveolar septation. The authors concluded that losartan — already in clinical use for hypertension — merited investigation as a therapy for Marfan syndrome with the potential to prevent its major life-threatening manifestation.

Key findings

  • Aneurysm in the Fbn1^C1039G/+ Marfan mouse is associated with increased TGF-β signalling in the aortic wall.
  • TGF-β–neutralising antibody prevented aneurysm — establishing TGF-β as necessary in this model.
  • Losartan prevented aneurysm, phenocopying the antibody and implicating the angiotensin II–TGF-β axis.
  • Propranolol was included as a comparator arm (the standing clinical standard of care).
  • Effects extended beyond the aorta: AT1 antagonism partially reversed impaired alveolar septation.
  • Immediate translational hook: losartan was an approved, inexpensive, widely used drug.

Limitations

  • Treatment was preventive, initiated before or at the onset of aortopathy in a genetically uniform, short-lived animal; the human analogue would be lifelong treatment of an already-dilated aorta.
  • Mouse aortas dilate but do not reproduce the human trajectory of decades of slow growth punctuated by acute type A dissection in adulthood.
  • The endpoint was aortic dimension and wall architecture over weeks, not dissection, rupture or survival in a matched human timeframe.
  • Losartan lowers blood pressure; the design does not fully separate haemodynamic unloading from TGF-β–specific effects, which matters because the human comparator (atenolol) also unloads.
  • Whole-aorta TGF-β readouts cannot attribute signalling to a specific cell compartment.

Why it matters

This is the most-cited experiment in the field and the field's most instructive failure of translation. It generated the ARB hypothesis, multiple randomised trials, and a decade of guideline debate. Then the human data arrived: in 608 children and young adults with Marfan syndrome, the aortic-root z-score slope was −0.139 ± 0.013 (atenolol) versus −0.107 ± 0.013 (losartan) SD units/year, p=0.08, with no difference in 3-year rates of aortic surgery, dissection or death (Lacro 2014, PMID 25405392); a meta-analysis of 8 trials in 1381 patients confirmed no significant difference (Kang 2019, PMID 31003918). In parallel, loss-of-function experiments showed that SMC-intrinsic TGF-β signalling is protective: conditional Tgfbr2 deletion in postnatal smooth muscle causes the thoracic aorta to thicken, dilate and dissect (Li 2014, PMID 24401272); Marfan aortopathy develops without detectable alteration in SMC TGF-β signalling and is exacerbated by receptor deletion (Wei 2017, PMID 28119285); and TGF-β protects both thoracic and abdominal aorta by distinct mechanisms (Angelov 2017, PMID 28729364). The paper therefore functions in this knowledge base in two roles at once: the founding evidence for the TGF-β model of aortopathy, and the anchor case for why mouse rescue does not imply human efficacy.

Cited by wiki pages

  • pathophysiology
  • animal models