SAN DIEGO - On Sept. 30, Merck & Co Inc. announced that it was voluntarily withdrawing from the market Vioxx, its nonsteroidal anti-inflammatory pain reliever for osteoarthritis and rheumatoid arthritis. That removal was the largest ever of a marketed drug.

Vioxx works by inhibiting the enzyme Cox-2, and as is so often the case in biological systems, Cox-2 is a multitasker, exerting its effects in several realms. In fact, Vioxx's problems were discovered in a clinical trial for preventing recurring colorectal polyps, and Cox-2 inhibitors are receiving attention for their potential in the fight against a variety of neurological conditions.

Chronic Conditions: A New Risk-Benefit Analysis

Free radicals appear to be the culprit in Cox-2's deleterious effects in Parkinson's disease. At a symposium titled "Cox-2 and its Reaction Products: Regulation and Roles in Neurodegeneration and Neuroprotection" at the Annual Meeting of the Society for Neuroscience this week, Serge Przedborski, professor of neurology and pathology at Columbia University, presented data showing that in the case of Parkinson's disease, Cox-2's pathogenic contribution is related to oxidative damage, not neuroinflammation, as his group originally had hypothesized.

Przedborski's group found that after giving mice MPTP (a toxin that leads to degeneration of dopaminergic neurons, the hallmark feature of Parkinson's disease) they observed increased Cox-2 expression. However, they saw the increase not primarily in glial cells, as expected if Cox-2 were causing inflammatory damage. Instead, the researchers found increased levels of Cox-2 in dopaminergic neurons - "the exact opposite of what we would have expected," Przedborski told the audience. Analysis of post-mortem brain samples from Parkinson's patients confirmed that finding.

Further research by Przedborski and his colleagues suggested that Cox-2 was oxidizing dopamine, which in turn reacts with cysteine-containing proteins and can cause them to misfold. Usually, protein misfolding results from alterations in a protein's genetic code, but post-translational modification also can cause it. Przedborski's group showed that such misfolding does occur as a downstream consequence of Cox-2 oxidation of dopamine, though he cautioned that whether it contributes to Parkinson's pathology is unclear.

Asked how the Vioxx withdrawal feeds back onto basic and preclinical research, Przedborski drew a distinction between inhibiting Cox-2 in acute and chronic conditions. The heightened risk of averse cardiovascular events that led to Vioxx's withdrawal is "not an acute side effect, so if you are dealing with neuroprotective strategies in acute settings, that should have no bearing," he told BioWorld Today. However, in the chronic setting, the new findings on Vioxx change the risk-benefit analysis of treating patients, since "even if you have a fatal disease, that doesn't mean we should accelerate your demise by giving you a stroke or a heart attack."

Przedborski saw "two ways to circumvent this problem, assuming it's general." First, "antioxidant strategies targeting the downstream consequences of Cox-2 activation may be safer" than inhibiting Cox-2 directly.

The other possibility is to go upstream. In the MPTP model of Parkinson's disease, Cox-2 is activated by the JNK pathway, and directly targeting JNK might also provide an alternative to Cox-2. Indeed, the JNK pathway already is being investigated for that purpose. The Parkinson Study Group, a collection of Parkinson's disease experts from medical centers in the U.S. and Canada, is conducting Phase II/III trials on the JNK inhibitor CEP-1347, a compound developed by West Chester, Pa-based Cephalon Inc.

Stroke Treatment: Watch Your Timescale

Given that the reason Merck voluntarily withdrew Vioxx was a heightened risk of cardiovascular events, including stroke, it seems ironic that Cox-2 inhibitors are being investigated in protecting against stroke damage. The key to that particular paradox is that the heightened risk of stroke apparently comes late, after at least 18 months of using Cox-2 inhibitors, while the protective effects of inhibiting Cox-2 come from acute treatment.

Costantino Iadecola, professor of neurology and neuroscience at Cornell University's Weill Medical College, is studying Cox-2 inhibition in neurology. Iadecola is senior author of a paper titled "Prostanoids, Not Reactive Oxygen Species, Mediate Cox-2 - Dependent Neurotoxicity," which appeared in the May 2004 issue of the Annals of Neurology. In that paper, Iadecola and his colleagues investigated the mechanisms of Cox-2 neurotoxicity following stroke.

The researchers directly injected excitatory agents into the brains of rats, an "injury model that is related to stroke and other brain pathologies involving activation of glutamate receptors," such as trauma and neurodegeneration, Iadecola told BioWorld Today via email.

Treating the animals with a Cox-2 inhibitor attenuated the damage done by NMDA injection. Investigating the mechanisms by which Cox-2 does its damage, Iadecola's group found that, at least in the acute phase, less than 24 hours after induced stroke, Cox-2-dependent ischemic damage is mediated by prostanoids, rather than free radicals as previously thought.

"Radicals play an important role in the injury," Iadecola clarified, "but these radicals do not come from Cox-2." His and other research suggest that dysfunctional mitochondria might instead be producing those radicals.

Prostanoids, which the Annals of Neurology paper suggested are the alternate culprit in Cox-2-mediated stroke damage, result from Cox-2's metabolism of arachidonic acid. Like Cox-2 itself, they are multitaskers - in the brain their functions include fever responses, pain, memory and cerebral blood-flow regulation. They have multiple receptors, and Iadecola's group now is working to identify the specific receptors they activate to produce stroke damage.

Asked if the Vioxx withdrawal affected his research plans, Iadecola replied that "the problems with pathogenic effects of Cox-2 inhibitors have provided a boost to our research. There is no doubt that Cox-2 is a major player in stroke-mediated damage." He believes that more specific receptor targeting "will lead to treatments that will focus only on the deleterious effects of Cox-2, sparing the beneficial actions and the imbalance deriving from total block of Cox-2 enzymatic activity."