J wrote: << A patient with symptoms and/or signs suggesting liver dysfunction,
or in whom an
abnormal liver test has occurred, should be monitored carefully for evidence of
the
development of a more severe hepatic reaction while on therapy with CELEBREX.
If
clinical signs and symptoms consistent with liver disease develop, or if
systemic
manifestations occur (e.g., eosinophilia, rash, etc.), CELEBREX should be
discontinued." >>

That is true.  However, any medication which may have the slightest effect on
such lists those warning.  I have been on more than a few with just those
warning--including Betaseron for almot 10 years.  I didn't not start it because
it had the potential for liver damage.  It has kept my M.S. under complete
control during that time at a time when it had just become relapsing
progressive.  I started several other meds in the past year plus with those
warning, and so far, so good.  That includes the maximum safe dose for Celebrex
(800 mg/day) at the level approved for reduction/reduction of risk of colon
tumors with possible  propensity that they may work on b.c.  tumor activity as
well.

J wrote << Hypothesis/theories (there are thousands out there) don't help
Jennifer..

But additional, promising treatments may:
http://www.medscape.com/viewarticle/448279
Selective Cyclooxygenase-2 Inhibition: A Target in Cancer Prevention and
Treatment
from Pharmacotherapy
Posted 02/20/2003Suphat Subongkot, Pharm.D., David Frame, Pharm.D., William
Leslie, M.D., Deborah Drajer, Pharm.D.
Abstract and Introduction
Abstract
A major goal in the area of cancer prevention and treatment is to make rational
use of defined molecular targets in order to block carcinogenesis. Studies
conducted in experimental animal models for many human cancers, including those
of lung, skin, mammary gland, urinary bladder, colon, and pancreas, have
demonstrated that carcinogenesis often may be inhibited by the administration
of a highly diverse group of biologic and chemical agents. One very promising
and well-studied target is cyclooxygenase (COX)-2. Interestingly, a number of
cancers appear to overexpress the COX-2 enzyme, which may play several roles in
carcinogenesis. Recent clinical studies have demonstrated the effect of COX-2
inhibitors in the treatment of familial adenomatous polyposis, a genetic
disorder that increases the risk for developing colorectal cancer. Ongoing
clinical trials with COX-2 inhibitors will increase our understanding and may
give us profound insights into the general applicability of this new targeted
approach for cancer control.
Introduction
Cancer is the second leading cause of death in the United States, exceeded only
by heart disease. In the United States, one of every four deaths is from
cancer. During the past decade, about 16 million new cancer cases were
diagnosed. In 2002, approximately 1.3 million Americans were expected to
receive a diagnosis of cancer, of which 555,500 were expected to die,
accounting for more than 1500 deaths/day.[1] Scientific evidence suggests that
about one third of the 555,500 cancer deaths expected to occur in 2002 could
have been prevented,[2] and many molecular targets are being studied for both
cancer prevention and treatment.
Cyclooxygenase (COX)-2 inhibitors as a new target for cancer prevention and
treatment arise from substantial epidemiologic, experimental, pathologic, and
clinical evidence suggesting that nonsteroidal anti-inflammatory drugs (NSAIDs)
possess anticancer properties.[3-8] The NSAIDs exert their anti-inflammatory
effects by inhibiting the synthesis of prostaglandins through nonspecific
inhibition of COX enzymes. Prostaglandins, especially prostaglandin E2, appear
to be important in the pathogenesis of cancer secondary to the effects on
mitogenesis, cellular adhesion, immune surveillance, and apoptosis.[9]
Malignant tissues have been found to overexpress prostaglandins when compared
with normal tissues.[10-18] In several animal and human models, the inhibition
of prostaglandin formation by blocking COX-2 appeared to show protective
effects against many types of cancers, including breast, colon, esophageal,
lung, skin, and head and neck cancers.[3, 4, 9, 19-34]
There are two isoforms of the COX enzyme: COX-1 and COX-2. The COX-1 isoform
synthesizes prostaglandins that are required for normal physiologic function
like gastrointestinal cytoprotection and platelet activity.[35, 36] Inhibition
of COX-1 may have an important role in NSAID-induced toxicity in humans, such
as gastric ulcer formation. The second isoform, COX-2, is not detectable in
most normal tissues; however, it is induced at sites of inflammation by
cytokines, growth factors, tumor promoters, and other agents.[37, 38] Also,
COX-2 is overexpressed in neoplasia.[39-43] Several mechanisms have been
proposed to explain the important role of COX-2 in tumorigenesis, including
stimulating cancer cell proliferation,[44] inhibiting apoptosis,[45] and
enhancing angiogenesis.[46-48]
Section 1 of 8
------------------------------------------------------------------------

From the Department of Pharmacy (Drs. Subongkot, Frame, and Drajer) and the
RushCancer Institute (Dr. Leslie), Rush-Presbyterian-St.Luke's Medical Center,
Rush University, Chicago, Illinois.
Pharmacotherapy 23(1):9-28, 2003. © 2003 Pharmacotherapy Publications
Biology of Prostaglandins and the COX Pathway
Prostaglandins, leukotrienes, and related substances are called eicosanoids
because they are derived from 20-carbon (eicosa-) essential fatty acids.[49]
These substances are extremely potent, endogenous, regulatory substances that
are synthesized and released for immediate, local (autacoids) action.[49] For
eicosanoid synthesis to occur, arachidonic acid (eicosanoid precursors) first
must be mobilized from membrane phospholipids by the enzyme phospholipase A2
(Figure 1).[49, 50]
Figure 1. (click image to zoom) Biosynthesis of arachidonic acid and the
cyclooxygenase (COX) pathway. HPETE = hydroperoxyeicosatetraenoic acid; NSAIDs
= nonsteroidal anti-inflammatory drugs.After mobilization, arachidonic acid is
oxygenated by four distinct pathways: COX, lipoxygenase, P450 epoxygenase, and
isoprostane pathways.[50] The COX pathway is the rate-limiting step and has two
distinct activities: endoperoxide synthase activity that oxygenates and
cyclizes the arachidonic acid to form the cyclic peroxide prostaglandin G2, and
peroxidase activity that converts prostaglandin G2 to prostaglandin H2, a
common precursor for all prostanoids.[49] Prostaglandin H2 then is converted by
specific enzymatic pathways, including prostacyclin synthase, thromboxane
synthase, and isomerase, to yield three groups of cyclic prostanoids:
prostacyclin (prostaglandin I2), thromboxane A2, and prostaglandins (e.g.,
prostaglandin E2), respectively (Figure 1).[51]
The two isoforms of the COX enzyme, COX-1 and COX-2, are coded by the COX-1 and
COX-2 gene, respectively. These two isoenzymes share about 60% gene homology;
however, substantial differences exist between the gene and promoter structures
of COX-1 and COX-2 (Table 1).[35, 37, 52-56]
The most important difference between the two isoforms is their pattern of
tissue expression and regulation. Cyclooxygenase-1 is expressed constitutively
in virtually all tissues, most notably platelets, endothelial cells,
gastrointestinal tract, renal microvasculature, glomerulus, and collecting
ducts.[57] Its expression can increase 2-4-fold under stimulatory conditions
and is not affected by glucocorticoids. Conversely, COX-2 is normally
undetectable in most tissues under basal conditions, but its expression in many
cell types, including macrophages, fibroblasts, chondrocytes, and epithelial
and endothelial cells, is augmented 10-80-fold on stimulation with inflammatory
cytokines, growth factors, or endotoxins.[58] Unlike COX-1, COX-2 is inhibited
at the level of transcriptional expression, and its stability is affected by
glucocorticoids.[59] Interestingly, COX-2 also is expressed in a broad range of
non-inflammatory tissues such as kidney, brain, bone, and cartilage, as well as
neoplasias[60-63] (Table 2). These findings of COX-2 overexpression in cancer
tissues suggest that COX-2 may play a role in tumorigenesis.
------------------------------------------------------------------------
Section 2 of 8
Role of COX-2 in Carcinogenesis
Broad spectrums of human neoplastic tissues overexpress COX-2 messenger RNA
(mRNA) and COX-2 protein, which suggests a role for COX-2 in carcinogenesis.
The relationship of COX-2 and carcinogenesis may involve several pathways,
including conversion of procarcinogens into active carcinogens, inhibition of
apoptosis, increase in tumor cell invasiveness, promotion of angiogenesis, and
immunosuppression.[39, 45, 46, 112, 113]
The activation of procarcinogens, such as benzo [a] pyrene, a polycyclic
hydrocarbon found in tobacco and overgrilled foods, by COX has been related to
its peroxidase activity.[114] The carcinogen formed by peroxidase reaction is
detoxified by cytochrome P450 (CYP) enzymes in the liver.[114] Other organs,
such as colon, lung, bladder, and oral cavity, have lower CYP concentrations
and are more likely to be exposed to active carcinogen.
The role of COX-2 in apoptosis has been observed in rat intestinal epithelial
(RIE) cells.[45] The overexpression of COX-2 in RIE cells has been shown to
increase the proto-oncogene Bcl-2 and lead to inhibition of apoptosis. The
inhibition of apoptosis, a process of cell death, appears to be a key pathway
in the survival of cancer cells. Experimental models have been able to reverse
this inhibition of apoptosis by using sulindac sulfide, a nonspecific COX
inhibitor. This could result in increased cancer cell death and sensitization
of cancer cells to chemotherapy agents.[45]
The invasive ability of cancer cells also appears to be modulated by
COX-2.[115] Human colon cancer cells that overexpressed COX-2 were shown to be
more invasive when compared with vector transfected control cells.
Overexpression of COX-2 has been associated with the activation of
metalloproteinase-2 and increased RNA levels for the membrane-type
metalloproteinase to promote the invasion of cancer cells.[115] Prostaglandin
E2 suppresses both cellular and humoral immune responses and induces
immunosuppressive cytokines such as interleukin-10.[116] Cyclooxygenase-2
induces prostaglandin E2-mediated immune suppression, which typically is
associated with tumor growth.[117] Restoring the immune response by
downregulating production of interleukin-10 by means of COX-2 inhibition
improved antitumor immunity in a murine Lewis lung carcinoma model.[118]
Also, the growth of primary tumors and the progression of metastatic disease
depend on an adequate blood supply.[119] To increase their vascularity, the
tumor cells secrete growth factors such as vascular endothelial growth
factor[46] and basic fibroblast growth factor[48] to promote angiogenesis.
Experimental evidence in colon carcinoma cells indicates that angiogenesis was
inhibited by the selective COX-2 inhibitor NS-398.[46, 48] NS-398 was developed
in Japan as an arylsulfonamide derivative of the anti-inflammatory agent
nimesulide.[120] It demonstrates 2000-fold selectivity for COX-2 over COX-1;
this is approximately 2-fold higher COX-2 selectivity than that of rofecoxib
and 5-fold higher than that of celecoxib.[121-123] In animal models, NS-398 was
a potent anti-inflammatory agent[124, 125]; however, it has poor
bioavailability and produces hepatotoxic metabolites.[126] Thus, NS-398 was not
developed into a therapeutic agent. The structure, however, became the starting
point for the development of newer COX-2 inhibitors such as celecoxib and
rofecoxib, and NS-398 is used routinely as a COX-2 prototype in vitro.
Based on research demonstrating COX-2 involvement in carcinogen activation,
apoptosis inhibition, tumor invasion, and angiogenesis promotion, it is
reasonable that COX-2 inhibitors may offer an important and powerful target for
cancer prevention and treatment. Many clinical trials of COX-2 inhibitors in
cancer prevention and treatment are in progress (Table 3).[127]
------------------------------------------------------------------------
Section 3 of 8
Role of COX-2 Inhibitors in Various Cancers
Colorectal Cancer
Predisposing Genetics. Colorectal cancer is the third leading cause of cancer
death in the United States.[1] In 2002, approximately 148,300 new cases were
expected to be diagnosed, accounting for 56,600 deaths.[1] Approximately 15-20%
of all colorectal cancers are familial.[128] Among these, familial adenomatous
polyposis (FAP), caused by germline mutations in the adenomatous polyposis coli
(APC) gene, accounts for less than 1%; FAP is diagnosed in approximately
1/10,000 people. Patients with FAP develop numerous precancerous polyps
throughout their colon and rectum during adolescence. Left untreated, nearly
all patients eventually will develop colorectal cancer. The primary treatment
for FAP is surgical removal of most or all of the colon and rectum, with
subsequent surveillance of any remaining colorectal segment.
Hereditary nonpolyposis colorectal cancer (HNPCC) is a medical condition in
which people are at an increased risk for early onset colorectal cancer and
cancer at other sites (uterus, ovary, stomach, urinary tract, small bowel, and
bile duct). About 75-80% of people with HNPCC develop colon cancer. Also known
as the Lynch syndrome, HNPCC accounts for 5-8% of all cases of colorectal
cancer.[128] Approximately one half of the cases are caused by germline
mutations in the DNA mismatch repair genes (MLH1, MSH2, MSH6, PMS1, and PMS2).
Most of the remaining cases in patients belonging to HNPCC or HNPCC-like
families are still unexplained at the molecular level. Molecular genetic
findings have enabled hereditary colorectal cancer to be divided into two
groups. The first group consists of tumors that show microsatellite instability
and occur more frequently in the right side of the colon. They have diploid
DNA, harbor characteristic mutations such as transforming growth factor B type
II receptor and BAX,[129] and behave more indolently. An example of this group
is HNPCC. The second group consists of tumors with chromosomal instability that
tend to be left sided. They show aneuploid DNA; harbor characteristic mutations
such as K-ras oncogene, APC gene, and p53 tumor suppressor gene; and behave
more aggressively. An example is FAP.[130] Because development of colorectal
cancer involves many pathways, interference with these processes, especially in
high-risk individuals (e.g., those with FAP or HNPCC), may reduce the risk of
carcinogenesis.
Experiences Experiences with NSAIDswith NSAIDs
Substantial evidence from epidemiologic and observational studies suggests that
taking aspirin and other NSAIDs is associated with a decreased risk of
colorectal cancer.[3, 6, 29, 30, 131, 132] A prospective epidemiologic study of
635,031 adults who provided information on the frequency and duration of their
aspirin intake showed that death rates from colon cancer decreased with more
frequent aspirin intake.[3] The relative risks of death from colon cancer were
0.62 and 0.52, respectively, in men and women who took aspirin 16 or more times
a month for at least 1 year.[3] The risk reduction with more frequent aspirin
intake was strongest among persons who had taken aspirin for 10 years or
more.[29, 30] This observation also was made in other studies in which the
routine intake of aspirin and other NSAIDs was associated with a 40-50%
reduction in risk of polyps and cancer.[30, 133]
The intriguing results from epidemiologic studies led to clinical trials in
patients with FAP. A randomized, double-blind, placebo-controlled trial
compared sulindac 150 mg orally twice/day with placebo for 9 months in patients
with FAP.[32] A statistically significant reduction in the mean number of
polyps and polyp size occurred in patients treated with sulindac, with the peak
effect occurring at 6 months. After discontinuing sulindac for 3 months, both
the number and the size of the polyps increased but still remained
significantly lower than the values at baseline.[32]
Sulindac treatment is also effective in producing regression of colorectal
adenomas in patients with FAP who underwent previous subtotal colectomy,
regardless of baseline polyp number and size.[31, 134] Conversely, the results
from a recent primary prevention trial of sulindac were disappointing.[135]
This randomized, double-blind, placebo-controlled trial compared sulindac 75 or
150 mg orally twice/day with placebo for 48 months in subjects who were
genotypically affected with FAP but phenotypically unaffected. The number and
size of new adenomas were evaluated every 4 months for 4 years. No significant
differences were noted in the mean number (p=0.69) or size (p=0.17) of polyps
between the groups.[135] Sulindac at standard dosages appears to be effective
in the treatment of FAP; however, it did not prevent the development of
adenomas in subjects with FAP who were phenotypically unaffected.

Studies of COX-2 Expression in Colorectal Cancer

In normal human epithelium, COX-2 generally is downregulated[112] and is not
expressed in the gastrointestinal tract.[41] However, its expression is
upregulated by approximately 50% in colorectal adenoma[42] and 85-90% in
colorectal cancer.[112] The level of COX-1 expression remains normal in cancer
cells.[42, 136] The cause of COX-2 over-expression in colorectal cancer remains
unclear; however, some evidence demonstrates that a ras gene or HER2/neu
mutation may be involved.[137-141]

Preclinical Studies of COX-2 Inhibitors in Colorectal Cancer

Preclinical evidence concerning the role of COX-2 in colorectal tumorigenesis
originates from a mouse model for human FAP (Apc716 knockout mice).[142] A
COX-2 null mutation mouse model with absent COX-2 gene reduced the number and
size of the intestinal polyps dramatically compared with those of mice with an
active COX-2 gene. Furthermore, treating Apc716 mice with a specific COX-2
inhibitor reduced the polyp number more significantly than did sulindac, which
inhibits both isoenzymes.
In human colorectal cancer cell lines, studies were conducted to investigate
the relationship between inhibition of intestinal cancer growth and selective
inhibition of the COX-2 pathway. Colon cancer cells (HCA-7), which express high
levels of COX-2 protein constitutively, and HCT-116 cells, which lack COX-2
protein, were evalutated.[81] Treatment of nude mice implanted with HCA-7 cells
with a selective COX-2 inhibitor (SC-58125) reduced tumor formation by 85-90%.
Colony formation of cultured HCA-7 cells also was inhibited by SC-58125. On the
other hand, SC-58125 had no effect on HCT-116 implants in nude mice or colony
formation in culture.[81] This evidence suggests that a correlation may exist
between inhibition of colon cancer cell growth and selective inhibition of the
COX-2 enzyme.
In preclinical animal tumor models, the Apc mutant multiple intestinal
neoplasia mouse model was used to determine whether the selective COX-2
inhibitor celecoxib would be effective for adenoma prevention and/or
regression, and whether it might be safer than the nonselective NSAID,
piroxicam, which previously was shown to be most effective in this model.[82]
The results showed that celecoxib and piroxicam caused dramatic reductions in
both the number and size of tumors in a dose-dependent manner. Treatment with
celecoxib (1500 ppm) and piroxicam (50 ppm) decreased tumor multiplicity by 71%
and 77% in early phase, and 52% and 66% in late phase, respectively. In
contrast to the significant toxicity of piroxicam, celecoxib caused no
gastrointestinal adverse effects and did not inhibit platelet thromboxane A2.
These results provide evidence that selective inhibitors of COX-2 are safe and
effective for the prevention and regression of adenomas in a mouse model of
adenomatous polyposis.
Another animal study of celecoxib in colonic aberrant crypt foci formation
induced by azoxymethane in rats found that celecoxib inhibited overall colon
tumor burden by more than 87%.[83] The results of this study confirm evidence
that a specific COX-2 inhibitor, celecoxib, possesses strong chemopreventive
activity against colon carcinogenesis. A recent study of rofecoxib also showed
consistent results, with a significant inhibition of polyposis in the FAP
(Apc716) mouse.[84]
The most recent evidence from a murine model of human FAP showed that sulindac,
a nonspecific COX-2 inhibitor, when combined with a newly developed
irreversible inhibitor of the epidermal growth factor receptor kinase, EKI-569,
was more effective in reducing polyps than either agent alone.[143] Thus, the
antiangiogenic and biochemical properties of these agents may have additional
or synergistic activity. The encouraging results of these preclinical studies
have resulted in clinical studies evaluating COX-2 inhibitors in colorectal
cancer.

Clinical Studies of COX-2 Inhibitors in Colorectal Cancer

In a National Cancer Institute-sponsored trial, celecoxib helped reduce the
number of colon polyps that occurred in patients with FAP.[144] In this
double-blind, placebo-controlled study, 77 patients were randomly assigned to
treatment with celecoxib (100 or 400 mg twice/day) or placebo for 6 months.
After 6 months, the patients receiving celecoxib 400 mg twice/day had a 28.0%
reduction in the mean number of colorectal polyps (p=0.003) and a 30.7%
reduction in the polyp size (p=0.001), as compared with reductions of 4.5% and
4.9%, respectively, in the placebo group. The occurrence of adverse events was
similar among the groups. The results of the study led to the approval of
celecoxib by the United States Food and Drug Administration (FDA) as an adjunct
to usual care for patients with FAP.
A recent phase II trial of celecoxib with irinotecan, 5-fluorouracil, and
leucovorin showed activity in 23 previously untreated patients with
unresectable or metastatic colorectal cancer. Partial response was achieved in
28% of patients, whereas 56% had stable disease and 17% had disease
progression.[145] Interestingly, occurrence of grades 3 and 4 neutropenia
appeared to be less than expected (27%) compared with 54% reported with
standard irinotecan plus 5-fluorouracil plus leucovorin alone.[146] These
preliminary results suggest that celecoxib may both add to responses seen with
chemotherapy as well as decrease the adverse effect of bone marrow suppression
seen with the chemotherapy, possibly owing to decreased inflammatory factors.
Although responses in treatment are encouraging, ongoing clinical trials
investigating celecoxib in colorectal cancer prevention are perhaps the most
exciting.[127] In one of these studies, patients with sporadic polyps, who are
known to be predisposed to developing colon cancer, are randomized to take one
of two doses of celecoxib or placebo to see if the drug reduces the recurrence
of polyps after surgical removal. Another clinical phase I-II study[127] is
being conducted in patients with HNPCC. Patients are being assigned randomly to
one of two doses of celecoxib to evaluate its effects on a number of cellular
and molecular biomarkers in normal-appearing rectal mucosa. The results of
these clinical trials could create a tremendous effect on the long-term
incidence of and mortality from colorectal cancer.
Esophageal Cancer
Overexpression of COX-2 has been shown in up to 91% of Barrett esophagus
tissue, 78% of esophageal adenocarcinoma and squamous cell carcinoma, and 79%
of Barrett intestinal metaplasia as shown by immunohistochemistry.[88-90]
Experimental models have demonstrated reduced expression of an apoptosis
ligand, Fas (CD-95), in esophageal dysplastic and malignant tissues.[147] This
reduced expression is linked to overexpression of COX-2, which in turn
downregulates the expression of Fas ligand.[148] A mechanism for the inhibitory
effects of aspirin and NSAIDs on the occurrence of esophageal adenocarcinoma
could be the induction of apoptosis by COX-2 inhibition. Overexpression of
COX-2 also was associated with increased levels of bcl-2, a proapoptotic
protein that induced resistance to apoptosis.[45] Therefore, selective COX-2
inhibitors may upregulate the expression of Fas receptors on the cell surface
in subjects with Barrett dysplasia and have an inhibiting role in esophageal
carcinogenesis by influencing apoptosis and cellular proliferation.
Substantial epidemiologic and clinical evidence supports the role of
prostaglandins and COX-2 in esophageal cancer development and progression.[3-5,
7] A prospective epidemiologic study of aspirin intake in 635,031 adults found
that aspirin intake was associated with a reduction in death rates of
esophageal, stomach, and colorectal cancers.[3] The deaths from esophageal
cancer were approximately 40% lower among people who took aspirin 16
times/month or more for at least 1 year compared with those who did not take
aspirin.[3] Aspirin ingestion for 10 years or more also was associated with
greater reduction in risk of cancer deaths compared with no aspirin intake.[3]
This was confirmed by the National Health and Nutrition Examination Survey and
the National Epidemiologic Follow-up Studies.[4, 5, 7]
Selective COX-2 inhibitors may be preferable to NSAIDs as chemopreventive
agents because of fewer gastrointestinal adverse effects. A phase II trial in
patients with Barrett esophagus is under way.[127] In this trial, celecoxib is
being compared with placebo in people with Barrett esophagus who are at an
increased risk of developing esophageal cancer.

Gastric Cancer

Overexpression of COX-2 has been documented in 49% of gastric carcinomas
(intestinal and diffuse types), thus representing a possible target in gastric
carcinogenesis.[91, 149] Interestingly, the presence of Helicobacter pylori
also correlates with an upregulation of the expression of COX-2 mRNA and
prostaglandin E2 in gastric cancer cell lines.[92, 93] Overexpression of COX-2
protein by immunohistochemical staining was shown in resected gastric carcinoma
tissue obtained from 104 patients with primary gastric adenocarcinoma who
underwent radical gastrectomy.[94] Cytoplasmic COX-2 staining was seen not only
in tumor tissue but also in metaplastic and adenomatous cells. This suggests
that COX-2 may have a role in gastric carcinogenesis. Studies also suggest a
role for COX-2 in neoplastic angiogenesis because growth factors such as
hepatocyte growth factor, basic fibroblast growth factor, and epiregulin can
induce overexpression of COX-2 in gastric cancer cell lines.[113]
An experimental study evaluating COX-2 expression in 100 gastric cancers and 7
gastric cancer cell lines found that overexpression of COX-2 in gastric cancer
tissue was seen in 70% of the tumors compared with normal controls.[150]
Prostaglandin production was increased in the cancer cell lines, especially
when COX-2 was overexpressed.[151] Also, a significant association was noted
between COX-2 levels, lymph node status, and advanced stage of disease, which
may correlate with a poor prognosis.
Experimental evidence has shown that COX-2 influences key cellular events,
including apoptosis, cell cycle control, cell proliferation, and
angiogenesis.[39, 45, 46, 112, 113] It is clear that COX-2 and prostaglandins
are overexpressed in gastric cancer cells.[150] Selective COX-2 inhibitors
(NS-398 and JTE-522), indomethacin, and aspirin can suppress cell replication,
induce apoptosis, and reduce epidermal growth factor in gastric carcinoma cell
lines (KATO III).[152-154] Epidemiologic studies also have shown a decreased
frequency of gastric cancer in people who take NSAIDs, consistent with the
results demonstrated in colon and esophageal cancers.[3-5, 7] This evidence
suggests that inhibition of COX-2 may be an attractive target for treatment and
prevention of gastric cancer.
Hepatocellular Carcinoma
In hepatocellular carcinoma (HCC), the expression pattern of COX-2 protein has
been well correlated with the differentiation of the tumor, suggesting that
abnormal COX-2 expression may play an important role in carcinogenesis.[96, 97]
Immunohistochemical staining for COX-2 was performed on HCC tissues, and all of
the well-differentiated HCCs were positive, whereas 83% of the poorly
differentiated HCCs were negative. The efficacy of a selective COX-2 inhibitor,
NS-398, has been evaluated in five hepatoma cell lines by a cell viability
assay for growth inhibition. NS-398 suppressed the growth of all cell lines,
independent of the degree of COX-2 expression, and all cell lines exhibited
cell death by increasing apoptosis when treated with NS-398. Cyclooxygenase-2
may be a determinant of the degree of differentiation of HCC, and inhibition of
COX-2 can suppress the growth of hepatoma cell lines by induction of apoptosis.
Clinical studies will be required to determine the role of COX-2 inhibitors in
the chemoprevention and treatment of HCC in humans.

Breast Cancer
A direct correlation has been shown between the overexpression of COX-2 and
HER2/neu,[72, 155] a receptor protein tyrosine kinase found in approximately
30% of women with breast cancer and associated with a poor prognosis.
Overexpression of COX-2 may be a "downstream target" for HER2/neu-mediated
tumorigenesis. One study found that more than half of the women who had either
infiltrating ductal carcinoma or lobular carcinoma overexpressed both COX-2 and
HER2/neu. Eighty percent of women with mucinous carcinoma were both HER2 and
COX-2 negative, whereas 20% overexpressed COX-2 but were HER2 negative.[72]
Although the overall numbers are small, these results suggest that 3 of 3 women
with ductal carcinoma in situ and 10 of 12 women with ductal, lobular, or
infiltrating cancers overexpress both COX-2 and HER2/neu.[72] It also appears
that COX-2 is overexpressed in patients with HER2/neu-positive cancers
regardless of tumor type and that those with HER2/neu-negative cancers express
COX-2 much less frequently than those with HER2/neu-positive cancers.[72] To
determine if treatment with COX-2 inhibitors may affect tumor progression, an
ongoing phase II trial sponsored by the National Cancer Institute[127] is
evaluating the combination of celecoxib and trastuzumab in patients with
metastatic breast cancer who have disease progression with chemotherapy.
Results of an animal study also suggest the possible role of COX-2 inhibition
in chemoprevention of breast cancer.[23] This study demonstrated that celecoxib
reduced the frequency, multiplicity, and volume of breast tumors significantly
(p<0.001) more than did ibuprofen or placebo in female Sprague Dawley rats
given the carcinogen 7,12-dimethyl-benz [a] anthracene.[23]

Lung Cancer

Cyclooxygenase-2 appears to be overexpressed in approximately 90% of human lung
cancers.[48] Cyclooxygenase-2 expression, as detected by immunohistochemistry,
appears to be higher in adenocarcinomas than in squamous cell carcinomas.[99,
100] Studies in lung cancer cell lines have shown that COX-2 inhibitors can
induce apoptosis and inhibit cell proliferation.[156-158]
Several preclinical studies have been conducted in animal models to evaluate
COX-2 inhibitors in lung cancer prevention and treatment.[157, 159-161] In one
animal study, aspirin was compared with the specific COX-2 inhibitor, NS-398,
in lung tumorigenesis induced by
4-(methylnitro-samino)-1-(3-pyridyl)-1-butanone in A/J mice.[160] Aspirin (147
and 294 µg/g of food consumed) and NS-398 (7 µg/g of food consumed) inhibited
lung tumorigenesis by 31%, 44%, and 34%, respectively.[160]
The role of COX-2 inhibition in blocking angiogenesis had been studied in the
Lewis lung cancer model.[48] The study showed that celecoxib dramatically
delayed the growth of the primary tumor and reduced the size of lung metastases
in a dose-dependent manner. Furthermore, the expression of COX-2 in this model
was restricted primarily to the angiogenic blood vessels, the preexistent
vasculature adjacent to the primary tumor, and the blood vessels invading the
metastatic lesions in the Lewis lung cancer model, but not in the tumor cells
themselves.[48] These findings suggest that COX-2 contributed to tumor growth
in this model by inducing newly formed blood vessels that maintain tumor cell
viability and growth. The antiangiogenic activity of celecoxib appears to be a
key mechanism of action in the animal model of lung cancer and results in
substantial inhibition of tumor growth and metastasis.[48, 119]
In lung cancer, a strong correlation also exists among transforming growth
factor 1, epidermal growth factor, COX-2, and prostaglandin E2.[162]
Transforming growth factor 1 and epidermal growth factor can synergistically
induce COX-2 and prostaglandin production, leading to inhibition of apoptosis.
One phase II trial of celecoxib 400 mg twice/day plus concurrent weekly
paclitaxel (50 mg/m2)-carboplatin (area under the plasma concentration-time
curve 2 mg/ml/min) and chest radiation therapy (63 Gy) in patients with stage
III non-small cell lung cancer showed that three of five evaluable patients had
an objective response.[163] Blood and urine analysis of these patients
demonstrated that vascular endothelial growth factor levels declined in the
months after celecoxib treatment.[163] Inhibition of the COX-2 enzyme appears
to decrease vascular endothelial growth factor levels, which in turn may
inhibit tumor growth and invasion, and may reverse the inhibition of apoptosis,
leading to chemosensitization. We are conducting a clinical trial studying the
addition of a COX-2 inhibitor to a new investigational epidermal growth factor
receptor inhibitor in which epidermal growth factor levels and correlation to
COX-2 levels will be investigated.
Bladder Cancer
Recent studies suggest that the expression of COX-2 is elevated in urinary
bladder carcinoma tissue and that inhibition of COX-2 activity suppresses
bladder cancer in experimental animal models.[65-71, 164] Immunohistochemistry
in normal human urinary bladder tissue did not detect COX-2. However, COX-2 was
found to be overexpressed in 75% of carcinomas in situ and 86% of invasive
transitional cell carcinomas.[70, 71] Cyclooxygenase-2 may play a role in the
development of invasive bladder cancer, possibly related to activation of
carcinoma in situ.
In preclinical animal studies,[66, 69] selective COX-2 inhibitors such as
celecoxib were shown to prevent the development of superficial rat urinary
bladder carcinomas in a dose-dependent manner. These data correlate with those
of an epidemiologic study suggesting NSAIDs and COX-2 inhibitors may have
potential as preventive agents.[165] Individuals who took NSAIDs regularly were
shown to have a reduced overall occurrence of bladder cancer (odds ratio 0.81,
95% confidence interval [CI] 0.68-0.96).[165] Ongoing phase II-III
placebo-controlled trials are investigating celecoxib in patients with bladder
dysplasia, a precancerous condition that can lead to bladder cancer.[127] The
results of these clinical trials will show whether COX-2 inhibitors can play a
role in the prevention of human bladder cancer.
Prostate Cancer
The immunohistochemical expression of the COX-2 enzyme is strongly expressed in
the epithelial cells of high-grade prostatic intraepithelial neoplasia and
cancer.[105] Mean levels of COX-2 mRNA were found to be 3.4-fold higher in
prostate cancer tissue (12 specimens) than in benign tissue.[107]
Overexpression of COX-2 in human prostate cancer tissue also was found to
correlate with high bcl-2 protein expression.[106] Treatment of human prostate
cancer cell lines with a selective COX-2 inhibitor led to a downregulation of
bcl-2 protein level and increased apoptosis both in vitro and in vivo.[106] The
in vivo results also showed that COX-2 inhibition decreased tumor microvessel
density and angiogenesis.[106, 166] Cyclooxygenase-2 inhibitors can prevent the
hypoxic upregulation of vascular endothelial growth factor, a potent angiogenic
factor.[113, 167] These results indicate that COX-2 inhibitors may serve as
effective chemopreventive and therapeutic agents in prostate cancer.
Gynecologic Malignancies
The expression of COX-2 has been investigated in cervical cancer,[73-80, 168]
ovarian cancer,[101, 169] and endometrial cancer.[85-87, 102, 170]
Cyclooxygenase-2 was detected in cervical intraepithelial neoplasia and
cervical cancer tissue but was undetectable in normal cervical tissue. In
addition, the epidermal growth factor receptor commonly is overexpressed in
cervical cancer and can induce COX-2 in cultured human cervical carcinoma
cells.[171] The expression of COX-2 in cervical cancer may inhibit apoptotic
processes and thus enhance tumor invasion and metastasis.[79] Overexpression of
COX-2 is also associated with a poor response to neoadjuvant treatment and an
unfavorable prognosis.[80] More data to evaluate the therapeutic application of
COX-2 inhibitors in the treatment of carcinoma of the cervix are necessary.
In ovarian cancer, the expression of COX-2 mRNA and protein was detectable in
three of five ovarian carcinoma cell lines.[73] Increased COX-2 expression was
detected in 42% of 86 ovarian carcinomas and in 37% of 19
low-malignant-potential tumors, but not in normal tissue. Multivariate analysis
showed that COX-2 expression is an independent poor prognostic factor in
ovarian cancer (relative risk 2.74, 95% CI 1.38-5.47). Recently, the effect of
the COX-2 inhibitor NS-398 on the growth of cell lines of human ovarian cancer
in vitro was investigated.[102] A decrease in mitotic activity and increase in
apoptosis were observed in primary ovarian cancer cell cultures treated with
the COX-2 inhibitor. Clinical trials of COX-2 inhibitors in ovarian cancer,
especially early stage ovarian cancer, may show improved responses when given
with chemotherapy.
Hematologic Malignancies
Cyclooxygenase-2 expression has not been studied extensively in hematologic
malignancies. In experimental studies, COX-2 mRNA expression was studied in
various human T cell lines.[172] Human T cell leukemia virus (HTLV) type
I-infected T cell lines were treated with NS-398, a selective COX-2 inhibitor.
NS-398 inhibited proliferation and induced apoptosis of the HTLV-I-infected
T-cell lines. These data suggest that a selective COX-2 inhibitor could play a
role in the treatment of T cell leukemia or T cell lymphoma. Similar results
were found in a bovine leukemia virus model.[172] Other non-specific COX-2
inhibitors, such as aspirin and NSAIDs, have been shown to induce apoptosis of
chronic lymphocytic leukemia[173] and acute myeloid leukemia cell lines[174]
through the activation of COX-independent caspases. The synergistic effect of
these agents with anthracyclines[175, 176] may be considered a potential new
mechanism for trials in the treatment of leukemias.
Other Malignancies
In pancreatic adenocarcinoma, the levels of COX-2 mRNA detected by
immunohistochemistry were increased by more than 60-fold compared with normal
tissue.[175] The effect of COX-2 inhibition with the specific COX-2 inhibitor
NS-398 was investigated in four pancreatic cancer cell lines.[104] All four
human pancreatic cancer cell lines expressed COX-2, and their proliferation was
inhibited by NS-398 in a concentration- and time-dependent manner. NS-398
induced substantial apoptosis. These findings suggest that COX-2 contributes to
the growth and antiapoptosis of pancreatic cancer. Specific COX-2 inhibitors
may play a role in the treatment and prevention of pancreatic cancer.
In skin cancer, there is increasing evidence that a constitutive expression of
COX-2 plays a role in the development and progression of malignant epithelial
tumors. Expression of COX-2 plays an important role in ultraviolet
carcinogenesis of the skin.[103] In malignant melanomas, COX-2 overexpression
may be involved in the regulation of melanoma invasion.[177] In
ultraviolet-induced skin tumor development in hairless mice, celecoxib and
indomethacin showed a dose-dependent reduction (60% and 89%, respectively) in
tumor yield,[108] which was confirmed in another similar experimental
model.[109] In addition, the specific inhibition of COX-2 reduces parameters of
ultraviolet-B-mediated inflammation, including edema, dermal neutrophil
infiltration and activation, prostaglandin E2 levels, and the formation of
sunburn.[110] The protective effects of celecoxib suggest that specific COX-2
inhibitors may offer a way to reduce the risk of skin cancer safely in
humans.[178] A phase II-III study of oral celecoxib is under way in people with
actinic keratoses, a precancerous condition that can lead to squamous cell skin
cancer.[127] The results of this trial will clarify the role of COX-2
inhibitors in the treatment and prevention of this type of cancer.
In one study, levels of COX-2 mRNA were evaluated in 15 patients with head and
neck squamous cell carcinoma and 10 subjects with normal oral mucosa.[127] A
150-fold increase in amount of COX-2 mRNA was detected in those with carcinoma.
In addition, COX-2 protein was detected in six of six patients with this
carcinoma but was undetectable in those with normal mucosa. Because retinoids
protect against oral cavity cancer, the role of retinoids in suppressing
epidermal growth factor-mediated induction of COX-2 in cultured oral squamous
carcinoma cells was investigated. The study results demonstrated that the
anticancer properties of retinoids may be due to inhibition of COX-2
expression, which in part leads to antiangiogenic activity, cell cycle arrest,
and inhibition of telomerase activity of the tumor cells.[111] Based on the
results of this study, selective COX-2 inhibitors may be useful in preventing
or treating head and neck squamous cell carcinoma. Furthermore, the combination
of a selective COX-2 inhibitor and a retinoid may be more effective than either
agent alone in preventing and treating head and neck squamous cell
carcinoma.[179, 180]
In brain tumors, an increased COX-2 expression was investigated in 50 glioma
and 3 normal brain specimens by immunohistochemistry.[181] The COX-2 expression
was considerably higher in high-grade gliomas than in low-grade gliomas and
normal brain specimens. In addition, the effect of the specific COX-2 inhibitor
NS-398 on monolayer cell cultures and three-dimensional glioma spheroids was
evaluated in two human glioblastoma cell lines.[182] This study found that the
growth of spheroids, tumor cell migration, and proliferation of monolayer cell
cultures were inhibited in a dose-dependent manner by NS-398. A moderate
increase in the number of apoptotic cells in the treated spheroids also was
observed. These results provide evidence that COX-2 is upregulated in gliomas,
and that COX-2 inhibitors may have a role as adjuvant therapy for brain tumors.
------------------------------------------------------------------------
Section 4 of 8
COX-2 Inhibitors as a Combined Modality in Cancer Prevention and Treatment

Combined with Radiation Therapy

When combined with radiation, COX-2 inhibitors have been shown to enhance tumor
cell destruction by enhancing radiation-induced apoptosis, blocking cellular
repair mechanisms, and inhibiting angiogenesis.[182, 183] In experimental cell
lines, COX-2 is upregulated in prostate cancer cells (PC-3) after irradiation,
resulting in elevated levels of prostaglandin E2.[184] This effect can be
suppressed by the selective COX-2 inhibitor, NS-398. The radiation-enhancing
effect of NS-398 was investigated in rat intestinal epithelial cells
transfected with COX-2 complementary sense DNA (RIE-S).[185] This study showed
that NS-398 enhanced radiosensitivity in a concentration-dependent manner in
RIE-S. This effect may be due to enhancement of radiation-induced apoptosis.
The combination of the selective COX-2 inhibitor SC-236 with radiation was
investigated in both a human glioma cell line (U251) and in tumor xenografts
grown in the hind leg of nude mice.[186] The data indicate that SC-236
synergistically increases the time of tumor growth, induces U251 apoptotic cell
death, and enhances the radiosensitivity of the cancer cells. SC-236
consistently showed synergistic antitumor activity with radiation without
increasing radiation injury to normal tissue in an animal sarcoma model.[95]
The results of these studies suggest that the selective inhibition of COX-2
combined with radiation may both potentiate tumor cell kill and protect normal
tissue.

Combined with Chemotherapy

In recent years, the idea of combining COX-2 inhibitors with chemotherapy has
been investigated in vitro. In three non-small cell lung cancer and three small
cell lung cancer cell lines, the nonspecific COX inhibitor sulindac enhanced
the growth-inhibitory effects of both cisplatin and paclitaxel.[187] Induction
of apoptosis in non-small cell lung cancer cell lines has been demonstrated
when a selective COX-2 inhibitor was combined with cisplatin, etoposide,
docetaxel, or irinotecan.[157] In combination with chemotherapy, COX-2
inhibitors can reduce concentrations required to reach the 50% inhibitory
concentration by up to 77%.[157] The potentially synergistic combination of
COX-2 inhibitors and chemotherapy may be considered a strategy for the
treatment of cancer.
------------------------------------------------------------------------
Section 5 of 8

    Selective Cyclooxygenase-2 Inhibitionfrom Pharmacotherapy
Safety Considerations

The toxicities of NSAIDs can be troublesome, including gastrointestinal
bleeding, inhibition of platelet function, and renal toxicity with long-term
therapy. Cyclooxygenase-2 inhibitors are equally effective as nonselective
NSAIDs in patients with arthritis, with an improved safety profile particularly
in the gastrointestinal tract.[188, 189] In cancer prevention and treatment, it
is very important to consider agents that are effective yet possess minimal
toxicity for long-term therapy. Several NSAIDs with varying degrees of COX
enzyme inhibition are available[190] (Table 4).
Selective COX-2 inhibitors appear to be safer options for cancer prevention and
treatment when compared with traditional NSAIDs since they spare the COX-1
enzyme and its functions.[191, 192] Evidence of decreased gastrointestinal
adverse effects with COX-2 inhibitors was demonstrated in a preliminary study
of patients with osteoarthritis and rheumatoid arthritis treated with
short-term celecoxib at various doses compared with naproxen.[191, 192] The
gastrointestinal safety of rofecoxib was investigated in one randomized
trial,[193] which showed that rofecoxib caused significantly fewer ulcers
(p<0.001) when compared with ibuprofen. These results have been supported by a
meta-analysis[194] and two large randomized controlled trials, the Celecoxib
Long-term Arthritis Safety Study (CLASS)[189] and Vioxx Gastrointestinal
Outcomes Research (VIGOR)[188] trials. The gastrointestinal safety profile also
was shown to be better than that of nonselective agents in the newer generation
selective COX-2 inhibitors, valdecoxib and parecoxib.[188, 195]
Platelet inhibition has very important implications in patients with cancer,
especially those who have hematologic malignancies. These patients often have a
high risk of bleeding. Conventional NSAIDs exert their antiplatelet effect by
inhibiting the formation of thromboxane A2 by inhibiting COX-1.
Cyclooxygenase-2 specific inhibitors spare COX-1 and thereby spare some
platelet thromboxane A2 activity. A double-blind, randomized,
placebo-controlled study comparing the effects of celecoxib and naproxen on
platelet function showed that celecoxib is unlikely to interfere with normal
platelet function and hemostasis.[196] The safety results also were reported in
the CLASS trial, showing a lower frequency of significant reduction in
hemoglobin and hematocrit, as well as a lower frequency of bleeding-related
adverse events.[189]
Because the COX-2 enzyme is expressed in other organs such as the brain and
kidney, there may be some adverse effects of COX-2 inhibitors with long-term
therapy. Ongoing clinical trials of COX-2 inhibitors in cancer treatment and
prevention will help to evaluate the safety and efficacy of these drugs.
------------------------------------------------------------------------
Section 6 of 8
Further Clinical Considerations
As these clinical trials are being designed, it is apparent that the doses of
COX-2 inhibitors being used have been extrapolated from studies of COX-2
inhibitors in inflammation. In these trials, the plasma concentrations
generally range from 0.5-5 µmol/L.[121-123, 197] However, in most in vitro
studies the drug concentration in neoplastic tissues that achieve optimal
results is greater than 35 µmol/L.[198, 199] This suggests higher doses may be
needed for more optimal antitumor effect.
Also, COX-2 independent pathways may play a role in the anticancer effects, and
different COX-2 inhibitors may vary in their tumor activity. Celecoxib and
rofecoxib have been compared in hematopoietic and epithelial cell lines for
antiproliferative effects.[199] The antiproliferative effect of celecoxib
appears to be greater in both cell lines. In prostate cancer cells (PC-3),
different COX-2 inhibitors have similar concentration for 50% inhibition, but
celecoxib has been shown to be better in inducing apoptosis.[198] This may be
the result of differences in the involvement of COX-2 independent pathways,
possibly through the modulation of cyclic guanosine 3´,5´-cyclic monophosphate,
the activation of nuclear factor  B, or the binding of peroxisome
proliferator-activated receptors.[197, 200] These mechanisms need to be further
elucidated to determine dosing effects and preferred choice of agents.
------------------------------------------------------------------------
Section 7 of 8
Summary
Overexpression of the COX-2 enzyme has been shown in a variety of animal and
human malignancies.[10-18] At least five mechanisms by which COX-2 contributes
to tumorigenesis and the malignant phenotype of tumor cells have been
identified; they are inhibition of apoptosis, increased angiogenesis, increased
invasiveness, modulation of inflammation and immunosuppression, and conversion
of procarcinogens to carcinogens.[39, 45, 46, 112, 113]
The link between inhibiting the COX-2 enzyme and its potential roles in cancer
prevention and treatment has been suggested in epidemiologic studies, which
show a 40-50% reduction in frequency of colorectal cancer in individuals taking
NSAIDs.[30, 133] Inhibition of COX-2 in experimental models has resulted in
tumor prevention, tumor reduction, and chemosensitization.[3, 4, 9, 19-34] The
specific COX-2 inhibitor, celecoxib, has been shown to reduce the colorectal
adenoma burden significantly in high-risk patients.[144] This has led to the
FDA's approval of celecoxib in the treatment of FAP. It is hoped that this is
only the beginning of important findings of COX-2 inhibitors in the treatment
and prevention of cancer, as the preclinical, epidemiologic, and clinical
evidence supports the important involvement of COX-2 in many forms of cancer
(Table 5). Many clinical trials are being conducted to investigate the safety
and efficacy of selective COX-2 inhibitors in cancer.[127] These trials should
further define the role of COX-2 inhibitors in the prevention and treatment of
cancer.
Reprint Address
Address reprint requests to David Frame, Pharm.D., Department of Pharmacy,
A036, Rush-Presbyterian-St.Luke's Medical Center, Rush University, 1653 West
Congress Parkway, Chicago, IL 60612; e-mail: David_G_ Frame@rush.edu.
------------------------------------------------------------------------
Section 8 of 8