Emerging Options with Coxib Therapy

As the size of the aging population increases, primary care physicians, who practice at the front line of medical care, can expect to see more patients with Alzheimer's disease (AD) or colorectal cancer (CRC) in their clinical practice.(1-4) Perhaps surprisingly, cyclooxygenase (COX)-2-selective inhibitors (coxibs) may have a role in treating these diseases in addition to their established utility in the management of arthritis and other painful conditions.

AD is an age-related neurological disorder leading to progressive dementia. The number of patients in the United States with primary dementia (AD and vascular dementia) is approximately 4 million, and an estimated 100,000 new patients are expected to be diagnosed each year.(5) Slowing or preventing the neurodegenerative process in AD is one of the major challenges facing healthcare professionals today.(6)

Similarly, the risk of developing CRC grows with advancing age. The American Cancer Society estimates that in 2001 approximately 135,400 new cases of CRC will have been diagnosed and 56,700 Americans will have died from CRC.(4) While risk minimization recommendations exist,(4,7) researchers continue to search for an effective agent that could prevent or limit the progression of CRC.

Another area of clinical concern is the control of malignant pain associated with cancer, a primary clinical objective when caring for cancer patients. The role of primary care physicians is essential in preserving patients' quality of life, as they can coordinate treatment and patient evaluation with oncologists and palliative care clinicians.(8) Strategies utilizing nonsteroidal anti-inflammatory drugs (NSAIDs), alone or in association with an opioid, can effectively manage most cancer pain. However, their use is limited by side effects typically associated with NSAID therapy.

The clinical benefits of coxibs for the treatment of AD and chemoprevention of CRC are being evaluated as a result of an increased understanding of the pathophysiology of both AD and CRC. The unique pharmacology of coxibs has already demonstrated potential value in these areas, in addition to their use in the management of cancer pain. This article will review the potential COX-2-related therapeutic targets that have been revealed in these diseases and that may offer unique treatment options for sufferers and physicians alike.

Alzheimer's Disease

A loss of neuronal function, most likely in glutamatergic neurons in neocortical and hypothalamic structures, is believed to be responsible for the signs and symptoms of AD.(6,9-12) The etiology of AD is not fully understood, but three interactive developments -senile plaques, neurofibrillary tangles, and inflammation-have been identified as pathogenic factors.(10,11) Notably, markers of local inflammation, such as activated microglia, reactive astrocytes, complement proteins, cytokines, and reactive mediators of oxygen and nitrogen (free radicals), all occur in close proximity to senile plaques and neurofibrillary tangles containing beta-amyloid (Ab) and tau (t) proteins.(9-11) Furthermore, senile plaques associated with activated complement factors, activated microglia, and reactive astrocytes-without any apparent influx of leukocytes-are strongly suggestive of a locally-induced, nonimmune-mediated inflammatory response.(9,10)

Inflammation in Alzheimer's disease

The inflammatory hypothesis of AD suggests that these inflammatory processes either directly or indirectly promote neurotoxicity and neurodegeneration.(11-15) The markers of a neuroinflammatory response detected in AD brain tissue represent a protective reaction to neuronal stress, but most likely contribute to neuronal stress as well.(9,11) One pharmacologic approach to retard AD progression, therefore, would be to suppress inflammation with anti-inflammatory treatment using nonselective NSAIDs or the COX-2-selective inhibitors.(6,9)

Epidemiological surveys have proven to be quite useful in investigating the pathogenesis of AD since circumstances associated with a decreased prevalence of disease may help to identify factors that may be providing a protective influence.(11) Several epidemiological surveys have identified chronic exposure to an anti-inflammatory agent as a protective factor for the development of AD.

Understanding the evidence

The first line of epidemiological inquiry entailed case-controlled studies of medical parameters in individuals diagnosed with AD.(16-23) In all but one of seven studies,(16) a lower prevalence of concomitant arthritis was consistently identified as a "protective" factor against AD.

Cross-sectional surveys of elderly individuals have measured the prevalence of concurrent diagnoses of AD and rheumatoid arthritis (RA), a disease typically managed by chronic anti-inflammatory treatments. Three large, population-based surveys all found a significantly lower prevalence of AD among patients with RA, providing some evidence of a positive benefit conferred by anti-inflammatory treatment.(14,24,25) Two smaller studies gave somewhat conflicting results. One study showed a significantly lower prevalence of RA among a cohort of patients with AD compared with the prevalence of RA in a cognitively intact cohort (2% vs 13%; odds ratio [OR] = 0.17; P <.005).(23,26) The second study reported no difference in the prevalence of RA among patients with AD than in those who were cognitively intact (6% vs 4%; OR = 1.18; 95% confidence interval [CI], 0.35-3.91).(23,27)

The impact of chronic exposure to steroid therapy on the development of AD has also been reviewed in epidemiological studies. Four case-control studies all found that exposure to steroid treatment provided a protective effect, if not as numerically large an effect as seen in studies evaluating the impact of a diagnosis of arthritis or RA.(20-23,27)

Bestowing a bit more favor on the inflammatory hypothesis of AD and the putative role for COX- 2-selective inhibitors are results from studies that found a protective effect with NSAID use on the development of AD.(20-22) Notably, the overall OR for these studies was 0.50 (95% CI, 0.34-0.72; P = .0002), compared with that for those studies evaluating the impact of steroid therapy (0.66; 95% CI 0.43-1.00; P = .049). These data suggest that NSAID use (which directly targets COX activity as contrasted to steroids, which act on the acute phase of the inflammatory response) confers a greater degree of protection against the development of AD than does steroid exposure.(20,22,23)

A comparable level of protection with NSAID use was seen in two population-based surveys. In one, the prevalence of NSAID users in the population of two AD clinical trials was 0.5% compared with a 22% prevalence of NSAID users in a control cohort from three surveys of elderly patients.(23,28) In the second population-based survey, 1.4% of 365 NSAID users were found to have AD compared with 2.5% in a cohort of 5,893 institutionalized and community-living individuals over 55 years of age.(29)

A third population-based study, the Baltimore Longitudinal Study on Aging (BLSA), followed 1,686 individuals prospectively to evaluate the effect of analgesic agents on the risk of AD.(30) With less than 2 years of nonaspirin NSAID use, the relative risk (RR) was 0.65 (95% CI, 0.33-1.29), and with 2 years or more of nonaspirin NSAID use, the RR dropped to 0.40 (95% CI, 0.19-0.84). Less benefit was seen with low-dose aspirin use: the RR was 0.74 (95% CI, 0.46-1.18). No benefit was seen with acetaminophen use, which has no anti-inflammatory properties: the RR was 1.35 (95% CI, 0.79-2.30).(30)

Another longitudinal survey was conducted at The Johns Hopkins Alzheimer's Disease Research Clinic. Among 209 patients entering the research clinic, only 15% claimed prior or current NSAID use.(31) During a 1-year period, the 32 NSAID users experienced later onset, reduced severity, and slower progression of AD symptoms when compared with age-matched and disease duration-matched patients not taking an NSAID.(23,31)

Lema MJ. Emerging options with coxib therapy. Cleve Clin J Med 2002;69:SI76-84.