That is one conclusion of an online article published this week in the Journal of Experimental Medicine, which seeks to explain how the cardiovascular side-effects of the drug occur and crucially, how they can also be controlled.
Timothy Hla, a professor of cell biology, and his team at the University of Connecticut, US, discovered that Vioxx prevents the cyclooxygenase (COX)-2 enzyme performing an anti-blood clot function.
Prof. Hla believes this important function can be restored using a certain unapproved class of drugs, of which a GlaxoSmithKline compound is the most clinically advanced.
He found that a cascade of enzymes working together to metabolise endocannabinoid molecules prevent the formation of blood clots: COX-2 in endothelial cells, together with prostacyclin synthase (PGIS) enzyme, activate a third protein called peroxisomal proliferator-activated receptor (PPAR)-delta. PPAR-delta, in turn, reduces the amount of tissue factor (TF, also called thromboplastin, or Factor III or CD142), which is an enzyme necessary for thrombin formation and hence the coagulation of blood.
By giving a patient a COX-2 inhibitor, this chain of events is blocked, PPAR-delta levels are reduced, the amount of circulating TF increases and blood clots become more likely.
This in turn increases the risk of side-effects such as heart attacks and stroke.
Prof. Hla believes this is at least partly why Vioxx can cause adverse cardiovascular events.
Vitally, he also thinks that if Vioxx, or a similar drug, were to be given in combination with a PPAR-delta activator, it is theoretically possible that these side-effects could be avoided and COX-2 inhibitors may be able to be used safely.
GlaxoSmithKline (GSK) has such a PPAR-delta activator in Phase II clinical trials: GW 50156.
It is currently the most advanced drug that selectively activates this target and was originally in-licensed from Ligand Pharmaceuticals.
The two companies have been collaborating for over a decade, in particular to develop small molecules capable of controlling the formation and development of blood cells.
One of the fruits of this alliance, Revolade (eltrombopag), is currently in Phase III clinical trials and a backup molecule called SB-559448 is in Phase I trials.
Two selective COX-2 inhibitors, Merck & Co.'s Vioxx and Pfizer's Bextra (valdecoxib), were withdrawn from the market in 2004 and 2005 following concerns over increased heart problems.
However the mechanistic basis of these effects is not well understood.
Whether GSK could prove to be an unlikely saviour for a Merck & Co drug would require a large, long-term clinical trial and even then, Vioxx might not sell, such is the public perception of the drug.
However, the news may come as a boost to other drug developers with COX-2 drugs still on the market - such as Pfizer with Celebrex (celecoxib), or to those who still believe COX-2 inhibitors have a place in pain therapy.