Inflammatory bowel disease (IBD) is a spectrum of chronic, idiopathic, inflammatory intestinal conditions. IBD causes significant gastrointestinal (GI) symptoms that include diarrhea, abdominal pain, bleeding, anemia, and weight loss. IBD also is associated with a variety of extraintestinal manifestations, including arthritis, ankylosing spondylitis, sclerosing cholangitis, uveitis, iritis, pyoderma gangrenosum, and erythema nodosum.
IBD conventionally is divided into two major subtypes: ulcerative colitis and Crohn's disease. Ulcerative colitis is characterized by confluent mucosal inflammation of the colon starting at the anal verge and extending proximally for a variable extent (e.g., proctitis, left-sided colitis, or pancolitis). Crohn's disease, by contrast, is characterized by transmural inflammation of any part of the GI tract but most commonly the area adjacent to the ileocecal valve. The inflammation in Crohn's disease is not necessarily confluent, frequently leaving "skip areas" of relatively normal mucosa. The transmural nature of the inflammation may lead to fibrosis and strictures or, alternatively, fistula formation.
Medical therapy for IBD is problematic. Because no unique abnormality has been identified, current therapy for IBD seeks to dampen the generalized inflammatory response; however, no agent can reliably accomplish this, and the response of an individual patient to a given medicine may be limited and unpredictable. Based on this variable response, clinical trials generally employ standardized quantitative assessments of efficacy that take into account both clinical and laboratory parameters (e.g., the Crohn's Disease Activity Index). The disease also exhibits marked fluctuations in activity—even in the absence of therapy—leading to a significant "placebo effect" in therapeutic trials.
Specific goals of pharmacotherapy in IBD include controlling acute exacerbations of the disease, maintaining remission, and treating specific complications such as fistulas. Specific drugs may be better suited for one or the other of these aims (Table 47–1). For example, glucocorticoids remain the treatment of choice for moderate to severe flares but are inappropriate for long-term use because of side effects and their inability to maintain remission. Other immunosuppressive agents, such as azathioprine, that require several weeks to achieve their therapeutic effect, have a limited role in the acute setting but are preferred for long-term management.
Table 47-1Medications Commonly Used to Treat Inflammatory Bowel Disease ||Download (.pdf) Table 47-1 Medications Commonly Used to Treat Inflammatory Bowel Disease
| ||CROHN'S DISEASE ||ULCERATIVE COLITIS |
| ||ACTIVE DISEASE ||MAINTENANCE ||ACTIVE DISEASE || |
|CLASS/Drug ||Mild-Moderate ||Moderate-Severe ||Fistula ||Medical Remission ||Surgical Remission ||Distal Colitis ||Mild-Moderate ||Moderate-Severe ||Maintenance |
|Mesalamine || || || || || || || || || |
| Enema ||+a ||− ||− ||− ||− ||+ ||+a,b ||− ||+ |
| Oral ||+ ||− ||− ||+/− ||+c ||+ ||+ ||− ||+ |
|Antibiotics || || || || || || || || || |
| (metronidazole ciprofloxacin, others) ||+ ||+ ||+ ||? ||+c ||− ||− ||− ||+c |
|Corticosteroids || || || || || || || || || |
| Enema, foam, suppository ||+a ||− ||− ||− ||+ ||+b ||− ||− ||− |
| Oral ||+ ||+ ||− ||− ||− ||+ ||+ ||+ ||− |
| Intravenous ||− ||+ ||− ||− ||− ||+d ||− ||+ ||− |
|Immunomodulators || || || || || || || || || |
| 6-MP/AZA ||− ||+ ||+ ||+ ||+c ||+d ||− ||+d ||+d |
| Methotrexate ||− ||? ||? ||? ||? ||− ||− ||− ||− |
| Cyclosporine ||− ||+d ||+d ||− ||− ||+d ||− ||+d ||− |
|Biological response modifiers || || || || || || || || || |
| Infliximab ||+d ||+ ||+ ||+c ||? ||+ ||− ||+ ||? |
| Adalimumab ||+ ||+ ||+ ||+ ||? ||? ||? ||? ||? |
| Certolizumab pegol ||+ ||+ ||? ||+ ||? ||? ||? ||+ ||? |
| Natalizumab ||− ||+ ||? ||+ ||? ||? ||? ||? ||? |
For many years glucocorticoids, sulfasalazine and 5-aminosalicylic acid were the mainstays of pharmacotherapy for IBD. More recently, agents used in other immune/inflammatory conditions, such as azathioprine and cyclosporine, have been adapted for IBD therapy. Advances in the understanding of the inflammatory response and improved biotechnology have led to the development of biological agents that can target single steps in the immune cascade, a successful and widely accepted strategy. Drug delivery to the appropriate site(s) along the GI tract also has been a major challenge, and second-generation agents have improved drug delivery, increased efficacy, and decreased side effects.
PATHOGENESIS OF INFLAMMATORY BOWEL DISEASE
Crohn's disease and ulcerative colitis are chronic idiopathic inflammatory disorders of the GI tract; a summary of proposed pathogenic events and potential sites of therapeutic intervention is shown in Figure 47–1. Although Crohn's disease and ulcerative colitis share a number of GI and extraintestinal manifestations and can respond to a similar array of drugs, emerging evidence suggests that they result from fundamentally distinct pathogenetic mechanisms (Xavier and Podolsky, 2007). Histologically, the transmural lesions in Crohn's disease exhibit marked infiltration of lymphocytes and macrophages, granuloma formation, and submucosal fibrosis, whereas the superficial lesions in ulcerative colitis have lymphocytic and neutrophilic infiltrates. Within the diseased bowel in Crohn's disease, the cytokine profile includes increased levels of interleukin (IL)-12, IL-23, interferon-γ, and tumor necrosis factor-α (TNFα), findings characteristic of T-helper 1 (TH1)– mediated inflammatory processes. In contrast, the inflammatory response in ulcerative colitis resembles aspects of that mediated by the TH2 pathway. Recently, this relatively simplistic classification has been revised in the light of the description of regulatory T cells and pro-inflammatory TH17 cells, a novel T-cell population that expresses IL-23 receptor as a surface marker and produces, among others, the pro-inflammatory cytokines IL-17, IL-21, IL-22, and IL-26. Several studies have demonstrated an important role of TH17 cells in intestinal inflammation, particularly in Crohn's disease (Cho, 2008; Strober et al., 2007).
Proposed pathogenesis of inflammatory bowel disease and target sites for pharmacological intervention. Shown are the interactions among bacterial antigens in the intestinal lumen and immune cells in the intestinal wall. If the epithelial barrier is impaired, bacterial antigens can gain access to antigen-presenting cells (APC) such as dendritic cells in the lamina propria. These cells then present the antigen(s) to CD4+ lymphocytes and also secrete cytokines such interleukin (IL)-12 and IL-18, thereby inducing the differentiation of TH1 cells in Crohn's disease (or, under the control of IL-4, type 2 helper T cells [TH2] in ulcerative colitis). The balance of pro-inflammatory and anti-inflammatory events is also governed by regulatory TH17 and TReg cells, both of which serve to limit immune and inflammatory responses in the GI tract. Transforming growth factor (TGF)β and IL-6 are important cytokines that drive the expansion of the regulatory T cell subsets. The TH1 cells produce a characteristic array of cytokines, including interferon (IFN)γ and TNFα, which in turn activate macrophages. Macrophages positively regulate TH1 cells by secreting additional cytokines, including IFNγ and TNFα. Recruitment of a variety of leukocytes is mediated by activation of resident immune cells including neutrophils. Cell adhesion molecules such as integrins are important in the infiltration of leukocytes and novel biological therapeutic strategies aimed at blocking leukocyte recruitment are effective at reducing inflammation. General immunosuppressants (e.g., glucocorticoids, thioguanine derivatives, methotrexate, and cyclosporine) affect multiple sites of inflammation. More site-specific intervention involve intestinal bacteria (antibiotics, prebiotics, and probiotics) and therapy directed at TNFα or IL-12 (see text for further details).
Important insights into pathogenesis have emerged from genetic analyses of Crohn's disease. Mutations in the gene NOD2 (nucleotide-binding oligomerization domain-2; also called CARD15) are associated with both familial and sporadic Crohn's disease in whites (Hugot et al., 2001; Ogura et al., 2001). NOD2 is expressed in monocytes, granulocytes, dendritic cells, Paneth cells, and epithelial cells. It is proposed to function as an intracellular sensor for bacterial infection by recognizing peptidoglycans, thereby playing an important role in the natural immunity to bacterial pathogens. Consistent with this model, other studies have identified bacterial antigens, including pseudomonal protein I2 (Dalwadi et al., 2001) and a flagellin protein (Lodes et al., 2004), as dominant superantigens that induce the TH1 response in Crohn's disease (shown as bacterial products in Figure 47–1). More recently, genome-wide studies have revealed other genes associated with IBD, whose identification may lead to the development of novel therapeutics (Cho, 2008)
Thus, these converging experimental approaches are generating novel insights into the pathogenesis of Crohn's disease that soon may translate into novel therapeutic approaches to IBD. The major therapeutic agents available for IBD are described next.
MESALAMINE (5-ASA)-BASED THERAPY
Chemistry, Mechanism of Action, and Pharmacological Properties. First-line therapy for mild to moderate ulcerative colitis generally involves mesalamine (5-aminosalicylic acid, or 5-ASA). The archetype for this class of medications is sulfasalazine (azulfidine), which consists of 5-ASA linked to sulfapyridine by an azo bond (Figure 47–2). Although this drug was developed originally as therapy for rheumatoid arthritis, clinical trials serendipitously demonstrated a beneficial effect on the GI symptoms of subjects with concomitant ulcerative colitis. Sulfasalazine is a prime example of an oral drug that is delivered effectively to the distal GI tract. Given individually, either 5-ASA or sulfapyridine is absorbed in the upper GI tract; the azo linkage in sulfasalazine prevents absorption in the stomach and small intestine, and the individual components are not liberated for absorption until colonic bacteria cleave the bond. 5-ASA is now regarded as the therapeutic moiety, with little, if any, contribution by sulfapyridine.
Structures of sulfasalazine and related agents. The red N atoms indicate the diazo linkage that is cleaved to generate the active moiety.
Mesalamine is a salicylate, but its therapeutic effect does not appear to relate to cyclooxygenase inhibition; indeed, traditional nonsteroidal anti-inflammatory drugs and selective inhibitors of cyclooxygenase-2 ("coxibs") may exacerbate IBD. Many potential sites of action have been demonstrated in vitro for either sulfasalazine or mesalamine: inhibition of the production of IL-1 and TNFα, inhibition of the lipoxygenase pathway, scavenging of free radicals and oxidants, and inhibition of NF-κB, a transcription factor pivotal to production of inflammatory mediators. Specific mechanisms of action of these drugs have not been identified.
Although not active therapeutically, sulfapyridine causes many of the adverse effects observed in patients taking sulfasalazine. To preserve the therapeutic effect of 5-ASA without the adverse effects of sulfapyridine, several second-generation 5-ASA compounds have been developed (Figures 47–2, 47–3, and 47–4). They are divided into two groups: prodrugs and coated drugs. Prodrugs contain the same azo bond as sulfasalazine but replace the linked sulfapyridine with either another 5-ASA (olsalazine, dipentum) or an inert compound (balsalazide, colazide). Thus, these compounds act at sites along the GI tract similar to those of sulfasalazine. The alternative approaches employ either a delayed-release formulation (pentasa) or a pH-sensitive coating (asacol; lialda/mezavant). Delayed-release mesalamine is released throughout the small intestine and colon, whereas pH-sensitive mesalamine is released in the terminal ileum and colon. These different distributions of drug delivery have potential therapeutic implications.
Metabolic fates of the different oral formulations of mesalamine (5-ASA). Chemical structures are in Figure 47–2.
Sites of release of mesalamine (5-ASA) in the GI tract from different oral formulations.
Metabolism of azathioprine and 6-mercaptopurine. HGPRT, hypoxanthine–guanine phosphoribosyl transferase; TPMT, thiopurine methyltransferase; XO, xanthine oxidase. The activities of these enzymes vary among humans because genetic polymorphisms are expressed differentially, explaining responses and side effects when azathioprine–mercaptopurine therapy is employed (see text for details).
Oral sulfasalazine is effective in patients with mild or moderately active ulcerative colitis, with response rates in the range of 60-80% (Prantera et al., 1999). The usual dose is 4 g/day in four divided doses with food; to avoid adverse effects, the dose is increased gradually from an initial dose of 500 mg twice a day. Doses as high as 6 g/day can be used but cause an increased incidence of side effects. For patients with severe colitis, sulfasalazine is of less certain value, even though it is often added as an adjunct to systemic glucocorticoids. Regardless of disease severity, the drug plays a useful role in preventing relapses once remission has been achieved. In general, newer 5-ASA preparations have similar therapeutic efficacy in ulcerative colitis with fewer side effects. Because they lack the dose-related side effects of sulfapyridine, the newer formulations can be used to provide higher doses of mesalamine with some improvement in disease control. The usual doses to treat active disease are 800 mg three times a day for asacol and 1 g four times a day for pentasa. Lower doses are used for maintenance (e.g., asacol, 800 mg twice a day). Although some studies have suggested that a given preparation may be superior in treating colonic disease, there is no consensus on this issue.
The efficacy of 5-ASA preparations (e.g., sulfasalazine) in Crohn's disease is less striking, with modest benefit at best in controlled trials. Sulfasalazine has not been shown to be effective in maintaining remission and has been replaced by newer 5-ASA preparations. Some studies have reported that both asacol and pentasa are more effective than placebo in inducing remission in patients with Crohn's disease (particularly colitis), although higher doses than those typically used in ulcerative colitis are required. The role of mesalamine in maintenance therapy for Crohn's disease is controversial, and there is no clear benefit of continued 5-ASA therapy in patients who achieve medical remission (Camma et al., 1997). Because they largely bypass the small intestine, the second-generation 5-ASA prodrugs such as olsalazine and balsalazide do not have a significant effect in small-bowel Crohn's disease.
Topical preparations of mesalamine suspended in a wax matrix suppository (rowasa) or in a suspension enema (canasa) are effective in active proctitis and distal ulcerative colitis, respectively. They appear to be superior to topical hydrocortisone in this setting, with response rates of 75-90%. Mesalamine enemas (4 g/60 mL) should be used at bedtime and retained for at least 8 hours; the suppository (500 mg) should be used two to three times a day with the objective of retaining it for at least 3 hours. Response to local therapy with mesalamine may occur within 3-21 days; however, the usual course of therapy is from 3-6 weeks. Once remission has occurred, lower doses are used for maintenance.
Pharmacokinetics. Approximately 20-30% of orally administered sulfasalazine is absorbed in the small intestine. Much of this is taken up by the liver and excreted unmetabolized in the bile; the rest (~10%) is excreted unchanged in the urine. The remaining 70% reaches the colon, where, if cleaved completely by bacterial enzymes, it generates 400 mg mesalamine for every gram of the parent compound. Thereafter, the individual components of sulfasalazine follow different metabolic pathways. Sulfapyridine, which is highly lipid soluble, is absorbed rapidly from the colon. It undergoes extensive hepatic metabolism, including acetylation and hydroxylation, conjugation with glucuronic acid, and excretion in the urine. The acetylation phenotype of the patient determines plasma levels of sulfapyridine and the probability of side effects; rapid acetylators have lower systemic levels of the drug and fewer adverse effects. By contrast, only 25% of mesalamine is absorbed from the colon, and most of the drug is excreted in the stool. The small amount that is absorbed is acetylated in the intestinal mucosal wall and the liver and then excreted in the urine. Intraluminal concentrations of mesalamine therefore are very high (~1500 μg/mL or 10 mM in patients taking a typical dose of 3 g/day).
The pH-sensitive coatings (methyl-methacrylate methacrylic acid copolymer) of asacol (eudagrit) and lialda/mezavant limit gastric and small intestinal absorption of 5-ASA, as assessed by urinary, ileostomal, and fecal measurements of the various metabolites. The pharmacokinetics of pentasa differ somewhat. The ethylcellulose-coated micro-granules are released in the upper GI tract as discrete prolonged-release units of mesalamine. Acetylated mesalamine can be detected in the circulation within an hour after ingestion, indicating some rapid absorption, but some intact micro-granules also can be detected in the colon. Because it is released in the small bowel, a greater fraction of pentasa is absorbed systemically compared with the other 5-ASA preparations.
Adverse Effects. Side effects of sulfasalazine occur in 10-45% of patients with ulcerative colitis and are related primarily to the sulfa moiety. Some are dose related, including headache, nausea, and fatigue. These reactions can be minimized by giving the medication with meals or by decreasing the dose. Allergic reactions include rash, fever, Stevens-Johnson syndrome, hepatitis, pneumonitis, hemolytic anemia, and bone marrow suppression. Sulfasalazine reversibly decreases the number and motility of sperm but does not impair female fertility. Sulfasalazine inhibits intestinal folate absorption and is usually administered with folate.
The newer mesalamine formulations generally are well tolerated, and side effects are relatively infrequent and minor. Headache, dyspepsia, and skin rash are the most common. Diarrhea appears to be particularly common with olsalazine (occurring in 10-20% of patients); this may be related to its ability to stimulate chloride and fluid secretion in the small bowel. Nephrotoxicity, although rare, is a more serious concern. Mesalamine has been associated with interstitial nephritis; although its pathogenic role is controversial, renal function should be monitored in all patients receiving these drugs. Both sulfasalazine and its metabolites cross the placenta but have not been shown to harm the fetus. Although less well studied, the newer formulations appear to be safe in pregnancy. The risks to the fetus from the consequences of uncontrolled IBD in pregnant women are believed to outweigh the risks associated with the therapeutic use of these agents.
The effects of glucocorticoids on the inflammatory response are numerous and well documented (Chapters 38 and 42). Although glucocorticoids are universally recognized as effective in acute exacerbations, their use in either ulcerative colitis or Crohn's disease involves considerable challenges and pitfalls, and they are indicated only for moderate to severe IBD.
The response to glucocorticoids in individual patients with IBD divides them into three general classes: responsive, dependent, and unresponsive. Glucocorticoid-responsive patients improve clinically, generally within 1-2 weeks and remain in remission as the steroidsare tapered and then discontinued. Glucocorticoid-dependent patients also respond to glucocorticoids but then experience a relapse of symptoms as the steroid dose is tapered. Glucocorticoid-unresponsive patients do not improve even with prolonged high-dose steroids. Approximately 40% of patients are glucocorticoid responsive, 30-40% have only a partial response or become glucocorticoid dependent, and 15-20% of patients do not respond to glucocorticoid therapy.
Glucocorticoids sometimes are used for prolonged periods to control symptoms in corticosteroid-dependent patients. However, the failure to respond to steroids with prolonged remission (i.e., a disease relapse) should prompt consideration of alternative therapies, including immunosuppressive agents and anti-TNFα therapies. Glucocorticoid are not effective in maintaining remission in either ulcerative colitis or Crohn's disease (Steinhart et al., 2003); their significant side effects have led to increased emphasis on limiting the duration and cumulative dose of corticosteroids in IBD.
The approach to glucocorticoid therapy in IBD differs somewhat from that in diseases such as asthma or rheumatoid arthritis. Initial doses are 40-60 mg of prednisone or equivalent per day; higher doses generally are no more effective. The glucocorticoid dose in IBD is tapered over weeks to months. Even with these slow tapers, however, efforts should be made to minimize the duration of therapy. Glucocorticoids induce remission in most patients with either ulcerative colitis or Crohn's disease (Faubion et al., 2001). Oral prednisone is the preferred agent for moderate to severe disease, and the typical dose is 40-60 mg once a day. Most patients improve substantially within 5 days of initiating treatment; others require treatment for several weeks before remission occurs. For more severe cases, glucocorticoids are given intravenously. Generally, methylprednisolone or hydrocortisone is used for intravenous therapy, although some experts believe that corticotropin (ACTH) is more effective in patients who have not previously received any steroids.
Topically acting agents (i.e., given by enema) have fewer adverse effects than systemic steroids but are also less effective in reducing remission. Glucocorticoid enemas are useful mainly in patients whose disease is limited to the rectum and left colon. Hydrocortisone is available as a retention enema (100 mg/60 mL), and the usual dose is one 60-mL enema per night for 2 or 3 weeks. When administered optimally, the drug can reach up to or beyond the descending colon. Patients with distal disease usually respond within 3-7 days. Absorption, although less than with oral preparations, is still substantial (up to 50-75%). Hydrocortisone also can be given once or twice daily as a 10% foam suspension (cortifoam) that delivers 80 mg hydrocortisone per application; this formulation can be useful in patients with very short areas of distal proctitis and difficulty retaining fluid.
Budesonide (entocort er) is an enteric-release form of a synthetic steroid that is used for ileocecal Crohn's disease (Greenberg et al., 1994; McKeage and Goa, 2002). It is proposed to deliver adequate steroid therapy to a specific portion of inflamed gut while minimizing systemic side effects owing to extensive first-pass hepatic metabolism to inactive derivatives. Topical therapy (e.g., enemas and suppositories) also is effective in treating colitis limited to the left side of the colon. Although the topical potency of budesonide is 200 times higher than that of hydrocortisone, its oral systemic bioavailability is only 10%. In some studies, budesonide was associated with a lower incidence of systemic side effects than prednisone, although data also indicate that systemic steroids are more effective in patients with higher Crohn's Disease Activity Index scores. Budesonide (9 mg/day for up to 8 weeks followed by 6 mg/day for maintenance of remission for up to 3 months) is effective in the acute management of mild-to-moderate exacerbations of Crohn's disease, but its role in maintaining remission has not been fully delineated (Hofer, 2003).
A significant number of patients with IBD fails to respond adequately to glucocorticoids and are either steroid-resistant or steroid- dependent. The reasons for this failure are poorly understood but may involve complications such as fibrosis or strictures in Crohn's disease, which will not respond to anti-inflammatory measures alone; local complications such as abscesses, in which case the use of glucocorticoids may lead to uncontrolled sepsis; and intercurrent infections with organisms such as cytomegalovirus and Clostridium difficile. Steroid failures also may be related to specific pharmacogenomic factors such as upregulation of the multidrug resistance (mdr) gene (Farrell et al., 2000) or altered levels of corticosteroid-binding globulin.
Several drugs developed initially for cancer chemotherapy or as immunosuppressive agents in organ transplants have been adapted for treatment of IBD. Although their initial use in IBD was based on their immunosuppressive effects, their specific mechanisms of action are unknown. Increasing clinical experience has defined specific roles for each of these agents as mainstays in the pharmacotherapy of IBD. However, their potential for serious adverse effects mandates a careful assessment of risks and benefits in each patient.
The cytotoxic thiopurine derivatives mercaptopurine (6-MP, purinethol) and azathioprine (immuran) (Chapters 51 and 52) are used to treat patients with severe IBD or those who are steroid resistant or steroid dependent (Prefontaine et al., 2009). These thiopurine antimetabolites impair purine biosynthesis and inhibit cell proliferation. Both are prodrugs: azathioprine is converted to mercaptopurine, which is subsequently metabolized to 6-thioguanine nucleotides that are the presumed active moiety (Figure 47–5).
These drugs generally are used interchangeably with appropriate dose adjustments, typically azathioprine (2-2.5 mg/ kg) or mercaptopurine (1.5 mg/kg). As discussed later, the pathways by which they are metabolized are clinically relevant, and specific assays can be used to assess clinical response and to avoid side effects. Because of concerns about side effects, these drugs were used initially only in Crohn's disease, which lacks a surgical curative option. They now are considered equally effective in Crohn's disease and ulcerative colitis. These drugs effectively maintain remission in both diseases; they also may prevent (or, more typically, delay) recurrence of Crohn's disease after surgical resection. Finally, they are used successfully to treat fistulas in Crohn's disease. The clinical response to azathioprine or mercaptopurine may take weeks to months, such that other drugs with a more rapid onset of action (e.g., mesalamine, glucocorticoids, or infliximab) are preferred in the acute setting.
The decision to initiate immunosuppressive therapy depends on an accurate assessment of the risk/benefit ratio. In general, physicians who treat IBD believe that the long-term risks of azathioprine–mercaptopurine are lower than those of steroids. Thus these purines are used in glucocorticoid-unresponsive or glucocorticoid-dependent disease and in patients who have had recurrent flares of disease requiring repeated courses of steroids. Additionally, patients who have not responded adequately to mesalamine but are not acutely ill may benefit by conversion from glucocorticoids to immunosuppressive drugs. Immunosuppressives therefore may be viewed as steroid-sparing agents.
Adverse effects of azathioprine–mercaptopurine can be divided into three general categories: idiosyncratic, dose related, and possible. Although the therapeutic effects of azathioprine– mercaptopurine often are delayed, their adverse effects occur at any time after initiation of treatment and can affect up to 10% of patients. The most serious idiosyncratic reaction is pancreatitis, which affects ~5% of patients treated with these drugs. Fever, rash, and arthralgias are seen occasionally, whereas nausea and vomiting are somewhat more frequent. The major dose-related adverse effect is bone marrow suppression, and circulating blood counts should be monitored closely when therapy is initiated and at less frequent intervals during maintenance therapy (e.g., every 3 months). Elevations in liver function tests also may be dose related. Although keeping drug levels in the appropriate range diminishes these adverse effects, they can occur even with therapeutic serum levels of 6-thioguanine nucleotides. The serious adverse effect of cholestatic hepatitis is relatively rare. Although the increased risk of infection is a significant concern with immunosuppressives, especially if pancytopenia occurs, infections are linked more closely to concomitant glucocorticoid therapy than to the immunosuppressives (Aberra et al., 2003).
Immunosuppressive regimens given in the setting of cancer chemotherapy or organ transplants have been associated with an increased incidence of malignancy, particularly non-Hodgkin's lymphoma. Definitive conclusions about the causative roles of azathioprine–mercaptopurine in lymphomas are complicated by the possible increased incidence of lymphomas in IBD per se and by the relative rarity of these cancers. The increased risk, if any, must be relatively small.
Metabolism and Pharmacogenetics. Favorable responses to azathioprine–mercaptopurine are seen in up to two-thirds of patients. Recent insights into the metabolism of the thiopurine agents and appreciation of genetic polymorphisms in these pathways have provided new insights into variability in response rates and adverse effects. As shown in Figure 47–4, mercaptopurine has three metabolic fates:
conversion by xanthine oxidase to 6-thiouric acid
metabolism by thiopurine methyltransferase (TPMT) to 6-methyl-mercaptopurine (6-MMP)
conversion by hypoxanthine–guanine phosphoribosyl transferase (HGPRT) to 6-thioguanine nucleotides and other metabolites
The relative activities of these different pathways may explain, in part, individual variations in efficacy and adverse effects of these immunosuppressives.
The plasma t1/2 of mercaptopurine is limited by its relatively rapid (i.e., within 1-2 hours) uptake into erythrocytes and other tissues. Following this uptake, differences in TPMT activity determine the drug's fate. Approximately 80% of the U.S. population has what is considered "normal" metabolism, whereas 1 in 300 individuals has minimal TPMT activity. In the latter setting, mercaptopurine metabolism is shifted away from 6-methyl-mercaptopurine and driven toward 6-thioguanine nucleotides, which can severely suppress the bone marrow. About 10% of people have intermediate TPMT activity; given a similar dose, these individuals will tend to have higher 6-thioguanine levels than the normal metabolizers. Finally, ~10% of the population is considered rapid metabolizers. In these individuals, mercaptopurine is shunted away from 6-thioguanine nucleotides toward 6-MMP, which has been associated with abnormal liver function tests. In addition, relative to normal metabolizers, the 6-thioguanine levels of these rapid metabolizers are lower for an equivalent oral dose, possibly reducing therapeutic response. Given this variability, some experts evaluate an individual's TPMT activity status prior to initiating treatment with thiopurines and also measure 6-thioguanine/6-MMP levels in individuals not responding to therapy. To avoid these complexities, treatment with 6-thioguanine was explored; unfortunately, 6-thioguanine is associated with a high incidence of an uncommon liver abnormality, hepatic nodular regeneration, and associated portal hypertension; 6-thioguanine therapy of IBD therefore has been abandoned.
Xanthine oxidase in the small intestine and liver converts mercaptopurine to thiouric acid, which is inactive as an immunosuppressant. Inhibition of xanthine oxidase by allopurinol diverts mercaptopurine to more active metabolites such as 6-thioguanine and increases both immunosuppressant and potential toxic effects. Thus, patients on mercaptopurine should be warned about potentially serious interactions with medications used to treat gout or hyperuricemia, and the dose should be decreased to 25% of the standard dose in subjects who are already taking allopurinol.
Methotrexate was engineered to inhibit dihydrofolate reductase, thereby blocking DNA synthesis and causing cell death. First used in cancer treatment, methotrexate subsequently was recognized to have beneficial effects in autoimmune diseases such as rheumatoid arthritis and psoriasis (Chapter 62 discusses the use of methotrexate in dermatological disorders). The anti-inflammatory effects of methotrexate may involve mechanisms in addition to inhibition of dihydrofolate reductase.
As with azathioprine–mercaptopurine, methotrexate generally is reserved for patients whose IBD is either steroid-resistant or steroid-dependent. In Crohn's disease, it both induces and maintains remission, generally with a more rapid response than that seen with mercaptopurine or azathioprine (Feagan et al., 1995). Its use in ulcerative colitis has not been thoroughly investigated.
Therapy of IBD with methotrexate differs somewhat from its use in other autoimmune diseases. Most important, higher doses (e.g., 15-25 mg/week) are given parenterally. The increased efficacy with parenteral administration may reflect the unpredictable intestinal absorption at higher doses of methotrexate. For unknown reasons, the incidence of methotrexate-induced hepatic fibrosis in patients with IBD is lower than that seen in patients with psoriasis. Use of methotrexate for treatment of IBD has largely been supplanted by biological therapies (such as anti-TNFα antibodies).
The calcineurin inhibitor cyclosporine is a potent immunomodulator used most frequently after organ transplantation (Chapter 35). It is effective in specific clinical settings in IBD, but the high frequency of significant adverse effects limits its use as a first-line medication.
Cyclosporine is effective in patients with severe ulcerative colitis who have failed to respond adequately to glucocorticoid therapy. Between 50% and 80% of these severely ill patients improve significantly (generally within 7 days) in response to intravenous cyclosporine (2-4 mg/kg per day), sometimes avoiding emergent colectomy. Careful monitoring of cyclosporine levels is necessary to maintain a therapeutic level in whole blood between 300 and 400 ng/mL.
Oral cyclosporine is less effective as maintenance therapy in Crohn's disease, perhaps because of its limited intestinal absorption. In this setting, long-term therapy with neoral or gengraf (formulations of cyclosporine with increased oral bioavailability) may be more effective, but this has not been studied fully. The calcineurin inhibitors can be used to treat fistulous complications of Crohn's disease. A significant rapid response to intravenous cyclosporine has been observed; however, frequent relapses accompany oral cyclosporine therapy, and other medical strategies are required to maintain fistula closure. Thus, the calcineurin inhibitors generally are used to treat specific problems over a short term while providing a bridge to longer-term therapy (Sandborn, 1995).
Other immunomodulators that are being evaluated in IBD include the calcineurin inhibitor tacrolimus (FK 506, prograf), mycophenolate mofetil and mycophenolate (cellcept, myfortic), inhibitors of inosine monophosphate dehydrogenase to which lymphocytes are especially susceptible (Chapter 35).