Understanding and managing chemotherapy-induced diarrhea

Understanding and ManagingChemotherapy-Induced DiarrheaLeonard B. Saltz, MD M anagement of chemotherapy-related Abstract Diarrhea is particularly problematic in patients receiving flu-
oropyrimidines and/or irinotecan. Careful patient monitoring, patient ed- ucation, and good patient-provider communication are the primary tools of prevention. The patient must be carefully evaluated on a regular basis sential if optimal efficacy, safety, and patient satis- early in treatment, so that mid-course corrections, dose adjustments, or faction are to be achieved. Diarrhea is a potential dose delays can be instituted early on if indicated. Diet need not be complication of a number of antineoplastic agents.
modified as a preventive measure, but once diarrhea occurs, a number It is particularly problematic with the fluoropyri- of modifications must be made. Maintenance of fluid intake is critical midines [1, 2] and with irinotecan (CPT-11,Camp- and inability to maintain adequate hydration would be a primary indi- tosar) [3], drugs that are central to the manage-
cation for parenteral fluid support. Oral intake of fluids should not be ment of colorectal cancer and other malignancies limited to plain water only, since electrolytes need to be replenished.
of the gastrointestinal tract. Other chemotherapy Early recognition of diarrhea and early pharmacologic intervention can agents used in other diseases are also prone to cause greatly facilitate successful resolution of this treatment complication.
diarrhea, although to a lesser degree. This manu-script will focus primarily on the management ofdiarrhea in patients with gastrointestinal malignan- cies, although the principles and practices outlined on the change above baseline. Thus, for example, a are potentially applicable to all oncology patients.
patient having six loose movements per day could not be evaluated accurately unless the baseline Defining the Problem
bowel-movement frequency was already known.
Surprisingly, there is no universal agreement on If the patient’s baseline frequency, especially after a definition of diarrhea. The most common crite- cancer surgery, was determined to be four loose ria for reporting of diarrhea in clinical trials are movements per day, then the change from base- the Common Toxicity Criteria of the National line is modest and the toxicity is low-grade. If the Cancer Institute (NCI-CTC) [4]. These criteria baseline was one formed bowel movement per day, define diarrhea on the basis of the number of bow- then this would constitute a more severe treat- el movements experienced per day above baseline. The issue of baseline activity is a critical one, Important Subtleties
and one which is prone to lead to misunderstand-ings and inaccurate reporting. Patients who have These subtleties are of critical importance not undergone intraabdominal and/or pelvic surgery, only to the interpretation of the clinical trial data or patients who have intraabdominal or pelvic pri- on which we base our practice recommendations mary or metastatic disease, or patients with non- but also are central to the practical day-to-day malignant gastrointestinal comorbidities often have management of individual patients. If we do not a baseline bowel function that includes multiple accurately characterize the extent of the problem, movements per day or movements that are soft, we cannot develop the most appropriate response.
semiformed, or even liquid. It is imperative to note We would thereby run the risk of either over or that the grading and reporting of toxicity is based underreacting to the problem, neither of whichwould permit the rendering of optimal care. Un- Correspondence to: Leonard B. Saltz, MD, Gastrointestinal fortunately, the careful characterization and quan- Oncology Service, Department of Medicine, Memorial Sloan- titation of diarrhea is not routinely undertaken by Kettering Cancer Center, 1275 York Avenue, New York, NY many oncology care providers. For example, one 10021; telephone: (212) 639-2501; fax: (212) 794-7186 study investigating this issue reported that a sub- stantial number of oncology nurses simply note diarrhea as being either present or absent, without An Ounce of Prevention…
attempting to provide quantification [5].
A number of pharmacologic maneuvers, to be practice to quantify diarrhea is a useful and discussed below, can be employed in the treatment constructive first step toward managing this of chemotherapy-related diarrhea. However, care- toxicity. First, such a practice encourages a ful patient monitoring, patient education, and good more detailed interview of the patient regard- patient-provider communication are the primary ing the toxicity. Second, since the studies on tools of diarrhea prevention. Diarrhea often has early warning signs, especially with weekly treat- based report their toxicities using NCI-CTC, ment regimens, and often presents as a mild-to- the use of these criteria in practice facilitates moderate problem before advancing to a more se- the placing of the toxicities noted in the clinic vere toxicity. Early recognition of the problem and into appropriate context for the reported reg- early intervention provide the best chance for a favorable therapeutic outcome. The potential ad- One important, but often overlooked, issue vantages of a weekly treatment, be it 5-fluorou- regarding the NCI-CTC criteria for diarrhea, racil (5-FU), irinotecan, or a combination of the however, is that the criteria changed slightly two, can be lost if the patient is not carefully eval- when they were updated in 1998 from CTC ver- uated on a weekly basis early in the treatment sion 1.0 to CTC version 2.0, and again, slightly, course, so that midcourse corrections, dose adjust- in 2003 with the introduction of CTC version ments, or dose delays can be instituted early on if 3.0. It is necessary to bear this in mind when reviewing the toxicities reported in older stud- The first component of good communication ies; for example, grade III or IV diarrhea report- and toxicity assessment is a careful, directed, pa- ed in older studies may not mean exactly the tient interview. Many patients often have a desire same thing as it does in more current trials. The to please, to be “good patients,” as well as a desire CTC criteria (versions 1.0, 2.0, and 3.0) are to think positively and to deny setbacks or com- plications. As such, there may be a predispositionnot to “complain” about side effects that are con- Mechanisms of Diarrheal Toxicity
sidered tolerable. There may also be a general re- sistance on the part of some patients to volunteer duced diarrhea, as well as other chemotherapy- a discussion of such usually private information as related diarrheas, are not entirely understood.
the frequency and consistency of their bowel move- Animal models and pathological analyses have ments. The clinician must take a proactive pos- suggested that cytotoxicity to the rapidly divid- ture and specifically question the patient regard- ing crypt cells of the intestinal epithelium may ing changes in bowel habits. A simple “how are lead to a relative loss of intestinal-absorptive you feeling?” or “are you having any problems?” capacity, as compared to secretory capacity may not elicit the information desired.
[6,7,8]. Cytotoxic destruction or augmentationof the enzymes involved in digestion of both pro- The Need to Be Specific
teins and carbohydrates may also alter osmotic Specific questions, such as “have you experi- gradients in the gut and thereby contribute to enced any increase in the number of bowel move- both decreased reabsorption and increased se- ments, or had any soft or liquid bowel movements, cretion of fluid and electrolytes in the stool [9].
or abdominal cramping, since the last treatment?” Attempts to identify specific risk factors for may be necessary. It is also important to specifical- diarrhea have thus far been inconsistent at best, ly ask if the patient has experienced any diarrhea and management of diarrhea has superseded pre- within the past 24 hours prior to the planned treat- vention in the clinical arena. Strategies have fo- ment dose. Patients with any diarrhea above their cused on nonspecific maneuvers to modulate di- pretreatment baseline should have their treatment arrhea through slowing of gastrointestinal transit delayed, usually for a week, but at least until the time, plus some more specific, though perhaps patient’s bowel-movement frequency has returned somewhat less established, attempts to deal with to the pretreatment baseline for a minimum of 24 drug-specific mechanisms of diarrhea.
hours. Patients who report “minor” or “some” di- Comparison of NCI Common Toxicity Criteria for Reporting of Diarrhea arrhea since the previous treatment need to be ber of modifications must be made. Patients expe- more thoroughly interviewed in order to better riencing diarrhea should avoid greasy, spicy or fried define the actual degree of toxicity that has been foods. Patients should also avoid milk and milk experienced, so that doses can be adjusted or de- products during and for a week or more after diar- rheal episodes, since the protracted diarrhea of Many patients are reluctant to “admit to” tox- chemotherapy can lead to a transient loss of lac- icity because they are afraid that such events will tase activity in the large bowel, resulting in tem- lead to a reduction in their treatment doses, and that this, in turn, will lead to less effective thera- Initially, use of the so-called BRAT (bananas, py. At the initiation of treatment, it is important rice, apple sauce, toast) diet, along with clear liq- to communicate to patients that side effects and uids, is indicated, until the diarrhea begins to re- dose modifications are common, and that accu- solve. This diet can then be slowly expanded, add- rate reporting of side effects will lead to the safest ing pasta without sauce, white-meat chicken and most effective therapy. It is important for pa- without the skin, scrambled eggs, and other easily tients to understand that failure to adjust doses digestible foods, as tolerated. Vegetables, especial- early enough may lead not only to severe and even ly cruciferous vegetables such as cabbage, Brussels life-threatening toxicity but could also lead to sub- sprouts, and broccoli, should be avoided as they stantial treatment delays, and even deeper dose re- may produce gas and lead to increased abdominal ductions, which in the long run could further de- crease the amount of medication that he or sheactually receives.
Fluid Intake Is Critical
A key component to management of diarrhea is maintenance of adequate hydration. It should be Patients do not need to modify their diets in emphasized to patients and their caregivers that food anticipation of possible diarrhea. A patient on po- intake is of minimal importance during acute diar- tentially diarrheogenic chemotherapy who is not rhea, but maintenance of fluid intake is critical. Pa- experiencing symptoms can eat a full, unrestrict- tients should be instructed to drink clear liquids in ed diet. If diarrhea occurs, however, then a num- adequate quantity to make up for the volume lost in diarrhea, plus their usual daily maintenance intake.
imbibe adequate quantities of fluid due to nausea, This will often necessitate the intake of 3–4 L or more anorexia, stomatitis, or other causes, should be of fluid per day. It is important to emphasize that this treated with intravenous fluid resuscitation and should not be all in the form of plain water, but rather maintenance. Clinical judgment should be exer- that fluids containing some salt and sugar, such as cised regarding whether this requires inpatient clear broth, gelatin desserts, Gatorade, soft drinks with management or not. Often, early institution of some of the carbonation removed, or similar fluids, outpatient fluid resuscitation can obviate the need should be consumed. Replacement of diarrheal loss- es with free water alone can lead to hyponatremiaand hypokalemia due to lack of replacement of lost Pharmacologic Management
LOPERAMIDE
Patients who are unable to maintain adequate hydration, as evidenced clinically by oral dryness Loperamide (Imodium and others) is the ini- and decreased urine production or by inability to tial drug of choice for management of most che- Commentary by Steven J. Cohen, MD, and nal toxicity was experienced [5] led to a system- atic review, resulting in specific recommenda-tions for the management of chemotherapy-in- In the accompanying article, Dr. Leonard Saltz duced diarrhea [6].
presents a thorough overview of issues con- PANEL RECOMMENDATIONS
fronting practitioners who treat patients An independent panel reviewed the medical As noted by Dr. Saltz, diarrhea is a common side records of 44 individuals who died while receiv- effect of chemotherapy. Although many chemo- ing therapy with either IFL (29 patients), 5-FU + therapy drugs cause diarrhea, those targeting gas- leucovorin (5 patients), oxaliplatin with bolus and trointestinal cancers are the most frequent offend- infusional 5-FU (FOLFOX 4, five patients), or ox- ers. Diarrhea can result in hospitalization, decreased aliplatin plus irinotecan (5 patients) in two large quality of life, and even death. Early studies evalu- studies [6]. Of these deaths, 23 were considered ating 5-fluorouracil (5-FU) modulated by leucovor- to be caused by a gastrointestinal syndrome in on a weekly schedule noted severe diarrhea in characterized by diarrhea, nausea/vomiting, and nearly one third of patients [1, 2]. Dose reductions abdominal cramping. These deaths represented and hospitalizations for intravenous hydration were 1% of all patients treated with the above regi- required in half of these patients, with a 4% death mens in these studies. The mechanism was like- ly loss of integrity of gut mucosa, followed by bacterial translocation and resulting sepsis. Sev- ing either irinotecan (Camptosar) or oxaliplatin eral important therapeutic and preventative rec- (Eloxatin) with 5-FU have improved overall sur- vival compared to 5-FU + leucovorin [3, 4]. These upon below. Although established for a specific regimens, however, also commonly cause severe chemotherapy regimen (IFL), the general prin- diarrhea. Saltz and colleagues compared a week- ciples of these guidelines are applicable to oth- ly bolus regimen of irinotecan, 5-FU, and leuco- er regimens that result in diarrhea.
vorin (IFL) to a daily x 5 schedule of 5-FU + leu- FREQUENT MONITORING
covorin [3]. They reported a 23% incidence ofgrade 3/4 diarrhea with IFL compared to 13% for As highlighted in Dr. Saltz’s review, patients 5-FU + leucovorin. A randomized study evaluat- at risk for chemotherapy-induced diarrhea ing infusional 5-FU + leucovorin with or without should be monitored weekly, as symptoms are oxaliplatin noted a 12% incidence of grade 3/4 cumulative. Patients experiencing diarrhea with- diarrhea in the oxaliplatin arm compared to 5% in 24 hours of their office visit should not receive for 5-FU/leucovorin [4]. Recent cooperative therapy. Abdominal cramping may herald diar- group experiences in which fatal gastrointesti- rhea and should be similarly managed with treat- motherapy-related diarrhea [10, 11]. It is oral high-dose loperamide has been expanded to and inexpensive, and it can be obtained with- management of diarrhea in general, and 5-FU- based diarrhea in particular [14]. Results sug- gest that loperamide is useful for management agent by slowing gastrointestinal peristalsis, of mild-to-moderate diarrhea and may also be thereby increasing gastrointestinal transit time useful as an early intervention in terms of keep- and promoting water reabsorption [12]. Studies ing moderate diarrhea from escalating. Howev- done early in the development of irinotecan in- er, for severe diarrhea, the effectiveness of lop- dicated that aggressive use of loperamide, start- ing with 4 mg by mouth at the first episode ofdiarrhea, followed by 2 mg every 2 hours until DIPHENOXYLATE/ATROPINE COMBINATIONS
diarrhea abated, was successful in ameliorating Diphenoxylate/atropine combinations (Lo- irinotecan-induced diarrhea significantly [13].
motil, Lonox) have been used empirically as an Subsequently, the approach of using frequent alternative to loperamide, although there are few ment delay. In addition, patients requiring antid- As Dr. Saltz emphasizes, studies of octreotide iarrheal therapy within 24 hours of treatment (eg, in chemotherapy-induced diarrhea are plagued loperamide) should not be treated, even if their by small sample sizes and lack of control arms.
diarrhea has resolved. The requirement for an- Taken in sum, however, most support the effec- tidiarrheal therapy indicates incompletely re- tiveness of octreotide in chemotherapy-induced diarrhea. Two randomized studies enrolling atleast 40 patients demonstrated a benefit of oct- ANTIBIOTICS
reotide therapy over loperamide as initial che- Appropriate use of prophylactic antibiotics is motherapy-induced diarrhea therapy in patients another important aspect of chemotherapy-in- with grade 2/3 [7] or grade 3/4 diarrhea [8]. In duced diarrhea management. In the panel’s review addition, single-arm studies of octreotide dem- above, toxic deaths related to gastrointestinal onstrate benefit, albeit without comparative symptoms were frequently attributed to sepsis, arms [9,10]. Importantly, several studies demon- sometimes with concurrent neutropenia. The panel strate a dose response with octreotide in both recommended institution of oral fluoroquinolone the phase I and randomized settings [11,12].
therapy for patients with persistent diarrhea last- In our practice, we routinely administer oct- ing more than 24 hours and for patients with diar- reotide to patients who require hospitalization rhea who develop either fever or neutropenia while for severe chemotherapy-induced diarrhea, giv- on chemotherapy. This recommendation derived en the significant morbidity associated with this in part from the European experience with 5-FU + condition. This is supported by the recommen- leucovorin and irinotecan, in which antibiotics are dations of another expert panel convened to routinely prescribed with diarrhea onset, and toxic consider the management of chemotherapy-in- deaths are rare (Eric van Cutsem, personal commu- duced diarrhea [13]. We agree that the current nication). Caregivers who infrequently care for can- data do not support prophylactic or early use of cer patients (eg, rotating residents or nursing staff ) octreotide for mild to moderate diarrhea.
should be informed of the appropriate use of anti- In addition to nonspecific chemotherapy-in- duced diarrhea therapies discussed above, an in- Dr. Saltz discusses several treatment options creasing understanding of chemotherapy-in- for chemotherapy-induced diarrhea, including duced diarrhea physiology is promoting study the controversial role of octreotide (Sandostatin).
of more targeted agents. For example, irinote- We believe that the current evidence supports its can administration has been demonstrated to in- use in patients with severe diarrhea unresponsive crease colon prostaglandin E and thromboxane to loperamide or diphenoxylate/atropine. In gen- A synthesis in rat models [14,15], prompting eral, this applies to most patients hospitalized for evaluation of COX-2 inhibition for chemothera- py-induced diarrhea. In mouse xenograft mod- published data on the effectiveness of this treat- OCTREOTIDE
ment for chemotherapy-related diarrhea.
As with loperamide, the mechanism of action Octreotide (Sandostatin) is an 8-amino-acid is a slowing of gastrointestinal-transit time in polypeptide that is a synthetically engineered an- order to promote improved water reabsorption alog of the natural 14-amino-acid polypeptide hor- from the bowel, thereby desiccating the stool and mone somatostatin [15]. Octreotide has been dem- promoting solidity. The presence of atropine in onstrated to be highly effective in the control of the combination may limit the dose escalation hormonal diarrheal syndromes due to serotonin of this combination as compared to using loper- production from carcinoid tumors or resulting from stimulation by vasoactive intestinal peptide, gas- Anecdotally, diphenoxylate/atropine appears trin, or glucagon produced by certain pancreatic to have similar efficacy to loperamide in mild- islet-cell tumors [16, 17]. In these circumstances, to-moderate diarrhea, and a similarly unsatis- diarrhea is caused by overproduction of a specific factory response in more severe cases.
bioactive substance; octreotide selectively binds els derived from HT-29 and colon-26 cells, admin- quired to clarify whether these agents will have istration of the COX-2 inhibitor celecoxib (Cele- a role in the prevention or treatment of chemo- brex) with irinotecan significantly reduced the incidence of diarrhea, compared to treatment Chemotherapy-induced diarrhea is a persis- with irinotecan alone [16]. COX-2 inhibition is tent source of morbidity and mortality. Recog- also being explored clinically in the context of nition of this problem has led to rational clinical recommendations for supportive management.
Further understanding of the pathophysiology POSSIBLE PHYSIOLOGIC MECHANISMS
of intestinal injury will hopefully lead to more As Dr. Saltz points out, loss of intestinal ab- effective methods of prevention and treatment.
sorption is a proposed mechanism of chemo- therapy-induced diarrhea. In pig models of diar- rhea, supplementation with the amino acid glutamine improves electrolyte absorption and nutrient transport [18]. Clinically, glutamine mayreduce the duration of diarrhea and loperamide requirement in patients treated with 5-FU + leu- Director, Gastrointestinal Cancer Program covorin [19]. As loss of intestinal absorption is theorized to occur through chemotherapy-me-diated injury to gut epithelial cells, therapies REFERENCES
derived to stimulate intestinal epithelial cell pro- 1. Petrelli N, Herrera L, Rustum Y, et al. A prospective liferation have been tested. Although primarily randomized trial of 5-fluorouracil versus 5-fluorouracil evaluated to reduce mucositis [20], keratinocyte and high-dose leucovorin versus 5-fluorouracil and growth factor (KGF) may inhibit chemotherapy- methotrexate in previously untreated patients with induced diarrhea. KGF stimulates growth of epi- advanced colorectal carcinoma. J Clin Oncol1987;5:1559–1565.
thelial cells, with its effect analogous to the im- 2. Buroker TR, O’Connell MJ, Wieand HS. Random- pact of hematopoietic growth factors on bone ized comparison of two schedules of fluorouracil and marrow recovery. KGF is protective against che- leucovorin in the treatment of advanced colorectal motherapy-induced diarrhea in mouse models, cancer. J Clin Oncol 1994;12:14–20.
decreasing weight loss and improving survival 3. Saltz LB, Cox JV, Blanke C, et al. Irinotecan plus flu- [21]. Although its mechanism of action is unclear, orouracil and leucovorin for metastatic colorectal can-cer. Irinotecan Study Group. N Engl J Med interleukin 15 has demonstrated a protective effect in animal models against chemotherapy- 4. de Gramont A, Figer A, Seymour M, et al. Leucov- induced diarrhea related to both irinotecan [22] orin and fluorouracil with or without oxaliplatin as and 5-FU [23]. Further clinical studies will be re- first-line treatment in advanced colorectal cancer. J to type II somatostatin receptors on the tumor cells in, when given in conjunction with bowel rest and and suppresses hormone production, thereby lim- intravenous hydration [20]. However, it is diffi- cult to distinguish the effect of the octreotide from Octreotide has a number of properties that the effects of bowel rest, intravenous fluids, oral make it potentially active against nonspecific, non- antidiarrheal medication, and the time that was hormonally generated diarrhea as well [15, 17, 18, allowed to transpire. Improvement in the setting 19]. Somatostatin has been shown to increase gas- of octreotide appears to be largely a time-depen- trointestinal transit time and cause a relative in- dent process. Very few patients respond rapidly to crease in intestinal water reabsorption and decrease the first one or two doses. It is highly likely that in intestinal secretions. These factors would tend octreotide is contributing to the control of diar- rhea in this setting; however, the extent of its con- Octreotide, given as a continuous intravenous tribution remains the subject of debate.
infusion, was noted to be effective in treatment of A trial of octreotide at two different dose lev- diarrhea caused by 5-FU plus high-dose leucovor- els (100 and 500 µg three times a day) resulted in intestine in rats. Jpn J Pharmacol 1997;75:399–405.
5. Sargent DJ, Niedzwiecki D, O’Connell MJ, et al.
15. Sakai H, Sato T, Hamada N, et al. Thromboxane Recommendation for caution with irinotecan, fluo- A , released by the anti-tumour drug irinotecan, is a rouracil, and leucovorin for colorectal cancer. N Engl novel stimulator of Cl– secretion in isolated rat colon.
6. Rothenberg ML, Meropol NJ, Poplin EA, et al.
16. Trifan OC, Durham WF, Salazar VS, et al. Cycloox- Mortality associated with irinotecan plus bolus flu- ygenase-2 inhibition with celecoxib enhances antitu- orouracil/leucovorin: summary findings of an inde- mor efficacy and reduces diarrhea side effect of CPT- pendent panel. J Clin Oncol 2001;19:3801–3807.
7. Cascinu S, Fedeli A, Fedeli SL, et al. Octreotide 17. Lin EH, Morris J, Chau NK, et al. Celecoxib atten- versus loperamide in the treatment of fluorouracil- uated capecitabine induced hand-and-foot syndrome induced diarrhea: a randomized trial. J Clin Oncol (HFS) and diarrhea and improved time to tumor pro- gression in metastatic colorectal cancer (MCRC). In: 8. Gebbia V, Carreca I, Testa A, et al. Subcutaneous Program/Proceedings of the 38th Annual Meeting of octreotide versus oral loperamide in the treatment the American Society of Clinical Oncology; May 18– of diarrhea following chemotherapy. Anticancer 21, 2002; Orlando, Fla. Abstract No. 2364.
18. Rhoads JM, Keku EO, Quinn J, et al. L-glutamine 9. Cascinu S, Fedeli A, Fedeli SL, et al. Control of stimulates jejunal sodium and chloride absorption in chemotherapy-induced diarrhoea with octreotide in pig rotavirus enteritis. Gastroenterology patients receiving 5-fluorouracil. Eur J Cancer 19. Daniele B, Perrone F, Gallo C, et al. Oral glutamine 10. Petrelli NJ, Rodriguez-Bigas, M, Rustum Y, et al.
in the prevention of fluorouracil induced intestinal Bowel rest, intravenous hydration, and continuous toxicity: a double blind, placebo controlled, ran- high-dose infusion of octreotide acetate for the treatment of chemotherapy-induced diarrhea in pa- 20. Meropol NJ, Somer RA, Gutheil J, et al. Random- ized phase I trial of recombinant human keratinocyte growth factor plus chemotherapy: potential role as 11. Wadler S, Haynes H, Wiernik PH. Phase I trial of mucosal protectant. J Clin Oncol 2003;21:1452–1458.
the somatostatin analog octreotide acetate in the 21. Farrell CL, Bready JV, Rex KL, et al. Keratinocyte treatment of fluoropyrimidine-induced diarrhea. J growth factor protects mice from chemotherapy and radiation-induced gastrointestinal injury and mortal- 12. Goumas P, Nazakis S, Christopoulou A, et al. Oc- treotide acetate in the treatment of fluorouracil-in- 22. Cao S, Black JD, Troutt AB, et al. Interleukin 15 duced diarrhea. Oncologist 1998;3:50–53.
offers selective protection from irinotecan-induced in- 13. Wadler S, Benson AB, Engelking C, et al. Rec- testinal toxicity in a preclinical animal model. Cancer ommended guidelines for the treatment of chemo- 23. Cao S, Troutt ABm Rustum YM. Interleukin 15 pro- tects against toxicity and potentiates antitumor activ- 14. Kase Y, Hayakawa T, Togashi Y, et al. Relevance of ity of 5-fluorouracil alone and in combination with leu- irinotecan hydrochloride-induced diarrhea to the lev- covorin in rats bearing colorectal cancer. Cancer Res el of prostaglandin E and water absorption of large a statistically superior outcome at the higher dose, ed. The considerable expense of octreotide rela- with 90% of 31 patients experiencing resolution tive to that of available oral agents and the fact of diarrhea within 5 days, versus 61% of 28 pa- that it must be injected do bear consideration in tients achieving resolution within 5 days at the determining the appropriateness of its use.
100-µg dose [21]. Again, the small study size, lackof blinding, and lack of a control somewhat weak- Pathophysiology as a Guide for
en the interpretability of the data.
Targeted Antidiarrheal Therapy
A phase I trial by Wadler et al [9] explored oc- The exact pathophysiology of chemotherapy- treotide in a dose range of 50–2,500 µg in patients related diarrhea has not been definitively identi- with 5-FU-induced diarrhea. Allergic reactions and fied, and it is undoubtedly multifactorial. It is likely hypoglycemia were noted at the higher doses; how- that some general mechanisms are consistent be- ever, antidiarrheal efficacy also improved with tween different drugs, while other mechanisms are drug-specific. A detailed investigation of the eti- A very small randomized trial of octreotide ver- ology of irinotecan-induced diarrhea was under- sus loperamide for treating 5-FU-induced diarrhea taken and demonstrated the diarrhea to have a also has been reported [22]. A total of 41 patients secretory mechanism with an exudative compo- with grade 2 or 3 diarrhea were randomized to ei- nent [25]. Fecal electrolyte measurements dem- ther octreotide 100 µg three times daily or lopera- onstrated a negative or small osmotic gap in 9/9 mide at a starting dose of 4 mg, followed by 2 mg patients studied, and an increased alpha-1-antit- every 6 hours. Both the small study size and the rypsin clearance in 6/6 patients. Blood levels of relatively small doses of both octreotide and loper- vasoactive intestinal peptide, glucagon, gastrin, amide that were used cloud the interpretability of somatostatin, prostaglandin E2, and carboxyl es- this trial. We do not know, for example, what the terase demonstrated no changes with the devel- outcome would have been if loperamide had been opment of diarrhea, nor were any changes in bac- given every 2 hours, as is now frequently advocat- ed. Conversely, a larger dose of octreotide might In patients who have undergone colonoscopic have shown better efficacy, although a small phase evaluation while experiencing severe irinotecan- II trial did report efficacy for the 100 µg dose of related diarrhea, areas of inflammation or colitis octreotide in this clinical setting [23]. What is also have been noted. On the basis of identification of notable is the lack of inclusion in this trial of any inflammation in the ileocecal region in one such patients who suffered from grade 4 diarrhea. An- patient, one group initiated a trial of the oral, top- ecdotally, octreotide may have better efficacy in ically active steroid, budesonide, in patients with moderate diarrhea, but has neither been reported either irinotecan-based, or 5-FU–based diarrhea, nor noted anecdotally to be rapidly effective in with some resolution of symptoms [26]. While res- olution of diarrhea was seen over time, it is diffi- The use of octreotide as a preventive agent cult to quantify what degree this reflected the nat- has been explored in a modest-sized cohort of ural history of the process. Randomized trials of patients receiving 5-FU with high-dose leucov- this approach will be needed to establish its rela- orin but was not found to be useful in this pro- Another group of investigators has postulated In summary, while it is clear that octreotide has that augmentation of glutamine metabolism could some significant antidiarrheal activity in chemo- be responsible for chemotherapy-related diarrhea therapy-induced diarrhea, its role in the manage- [27]. Glutamine is an amino acid that plays a key ment of this complication remains unclear, with role in whole-body nitrogen metabolism. It is a glu- different investigators having differing levels of coneogenic amino acid that serves as a primary enthusiasm for the drug in this setting, but with oxidative fuel for the enterocyte. As such, studies failing to definitively answer the important glutamine has a trophic effect on the bowel mu- questions. While some investigators advocate its cosa and is necessary for maintenance of the nor- routine use [19], others feel that the case for rou- mal intestinal structure. Patients with cancer have tine use of octreotide has not been made in this a tendency toward glutamine depletion. Animal setting and that further investigations, preferably studies have demonstrated that glutamine reple- of a randomized, double-blind design, are warrant- tion stimulates sodium and chloride absorption in a porcine jejunum model [28, 29]. Other animal ed patients are homozygous for an additional TA studies have shown a glutamine-supplemented diet in the TATAA box of the proximal promotor of to decrease methotrexate-induced enterocolitis the UGT1A1 gene [37]. The homozygous pres- [30]. Savarese et al have reported promising re- ence of this mutation, now called UGT1A1*28, sults in five patients treated with a glutamine sup- reduces transcription of this gene to approxi- plement for irinotecan-induced diarrhea [27].
mately 20% of normal [37]. Classically, this syn- Larger prospective trials are in progress to further drome manifests itself as a mildly to moderately evaluate the use of this dietary supplement for con- elevated bilirubin level, and fractionation of this trol of chemotherapy-related diarrhea.
value indicates a normal-to-low direct (conju-gated) bilirubin level but an elevated indirect Special Cases
(unconjugated) level. Consequently, patients donot manifest symptoms of jaundice. Although DIHYDROPYRIMIDINE DEHYDROGENASE
virtually all patients with Gilbert’s syndrome will DEFICIENCY
have elevated bilirubin levels under situations Dihydropyrimidine dehydrogenase (DPD) is of insufficient caloric intake, many will express the rate-limiting enzyme in the catabolism of 5- bilirubin levels within the normal range under FU and other fluorinated pyrimidines [31].
normal caloric-intake conditions, making a pri- Qualitative and quantitative variations in the ori identification of these individuals difficult.
expression of this enzyme lead to wide variations A patient may thus have a normal documented in 5-FU metabolism among patients. It has been bilirubin level on one day and a mild-to - demonstrated that a small percentage of the moderate elevation at a later or earlier time.
population has a clinically significant deficien- SN-38, the active metabolite of irinotecan, cy in DPD activity [32, 33]. This renders these is normally conjugated by UGT1A1 and secret- patients particularly sensitive to 5-FU, resulting ed in the bile. Patients with Gilbert’s syndrome in profound manifestations of multiple toxici- have a markedly reduced ability to conjugate and ties, including diarrhea, neutropenia, and mu- therefore are unable to secrete SN-38. This leads cositis. A patient experiencing early, severe, and to potentially profound toxicity from exposure prolonged multiple toxicities of 5-FU would be to irinotecan, resulting in both neutropenia and assumed on clinical grounds to have a relative diarrhea. Once toxicity occurs, management is DPD deficiency. Although some research labo- supportive. For patients with chronic indirect bi- ratories will perform DPD analyses on peripher- lirubin elevations, irinotecan should not be rou- al blood mononuclear cells, routine clinical lab- tinely used. For patients who experience substan- oratory tests for DPD are not available at present.
tial (grade 3 or greater) toxicity from their initial However, DPD testing is not really useful in most dose of irinotecan, Gilbert’s syndrome, or some variant thereof, should be suspected, and irino- If a patient experiences a toxicity profile con- tecan doses should be either discontinued or sistent with DPD deficiency, then that patient should be regarded as having a clinical diagno-sis of relative DPD deficiency. Such a patient BILIARY OBSTRUCTION
will require cessation of fluoropyrimidine thera- Somewhat analogous to the situation with py or a very substantial and empiric dose reduc- Gilbert’s syndrome, patients with biliary obstruc- tion. Laboratory quantification of DPD levels tion have severely impaired clearance of SN-38, would not, at this time, provide useful guidance the active metabolite of irinotecan. This leads to substantial increases in diarrhea and othertoxicities associated with irinotecan. Patients GILBERT’S SYNDROME
with elevated bilirubin levels on the basis of bil- Gilbert’s syndrome is an inherited disorder iary obstruction should not be routinely treated characterized by an inability or a deficiency in with irinotecan. No solid data have been pub- the process of glucuronidation [34]. Studies in- lished regarding dose attenuation recommenda- dicate that Gilbert’s syndrome may be present tions for irinotecan in patients with elevated in as much as 5% to 16% of the Caucasian pop- bilirubin levels, and treating such patients with ulation [34–36]. Within this population, affect- irinotecan usually is not recommended.
PREEXISTING BOWEL CONDITIONS
Conclusion
Patients with existing limitations in bowel func- Diarrhea is a serious potential consequence tion prior to initiation of chemotherapy may be of chemotherapy. It has an undeniable, nega- more prone to chemotherapy-induced diarrhea and tive impact on patient satisfaction with treat- may require especially close monitoring, or even ment and quality of life, and can be dose limit- consideration of attenuated starting doses. Patients ing, or even life threatening, in some extreme who have undergone extensive bowel resection and cases. Proper management requires an appreci- have short gut syndrome may have poorer toler- ation of the potential causal mechanisms of di- ance of diarrheogenic chemotherapy. Patients with arrhea, as well as early and aggressive interven- inflammatory bowel disease may be more prone to tion, both with antidiarrheal therapy and with exacerbation of diarrhea by chemotherapy, al- dose attenuation or interruption of chemother- though evidence suggests that some patients with apy. Proper attention to communication with pa- inflammatory bowel disease tolerate fluoropyrim- tients regarding bowel symptoms during treat- idines reasonably well [38]. The concurrent use ment and rapid institution of appropriate or history of radiation therapy to the bowel may supportive and dietary interventions are key fac- also exacerbate or predispose the patient to che- tors in optimizing control of chemotherapy-in- References
1. Grem, JL, Showmaker DD, Petrelli NJ, et al.
Severe life threatening toxicities observed in 13. Abigerges D. Armand J-P, Chabot GG, et al.
23. Zidan J, Haim N, Beny A, et al. Octreotide in study using leucovorin with 5-fluorouracil. J Clin Irinotecan (CPT-11): High-dose loperamide to con- the treatment of severe chemotherapy-induced trol diarrhea. J Natl Cancer Inst 1994;86:446–449.
diarrhea. Ann Oncol 2001;12:227–229.
2. Leichman CG, Fleming TR, Muggia FM, et al.
14. Cascinu S, Bichisao E, Amadori D, et al. High- 24. Meropol NJ, Blumenson LE, Creaven PJ .
Phase II study of fluorouracil and its modulation in dose loperamide in the treatment of 5-fluorou- Octreotide does not prevent diarrhea in patients advanced colorectal cancer: A Southwest Oncolo- racil-induced diarrhea in colorectal cancer pa- treated with weekly 5-fluorouracil plus high-dose gy Group study. J Clin Oncol 1995;13:1303–1311.
tients. Support Care Cancer 2000;8:65–67.
leucovorin. Am J Clin Oncol 1998;21:135–138.
3. Rothenberg ML. Efficacy and toxicity of 25. Saliba F, Hagipantelli R, Misset JL, et al .
irinotecan in patients with colorectal cancer.
Pathophysiology and therapy of irinotecan-in- Semin Oncol 1998;25(suppl 11):39–46.
16. Kvols LK, Moertel CG, O’Connell MJ, et al.
duced delayed-onset diarrhea in patients with Treatment of the malignant carcinoid syndrome: advanced colorectal cancer: A prospective as- 5. Engelking C, Rutledge D. Cancer-related di- Evaluation of a long-acting somatostatin ana- sessment. J Clin Oncol 1998;16:2745–2751.
arrhea: Oncology nurse perceptions of neglect- logue. N Engl J Med 1986;315:663–666.
26. Lenfers BH, Loeffler TM, Droege CM, et al.
ed symptoms and its management. Oncol Nurs 17. O’Dorisio TM, Mekhjian HS. Clinical experi- Substantial activity of budesonide in patients ence of somatostatin analog (compound 201- with irinotecan(CPT-11) and 5-fluorouracil in- 6. Baskerville A, Batter-Hatton D. Intestinal le- 995, Sandostatin) as an antidiarrheal agent. Dig duced diarrhea and failure of loperamide treat- sions induced experimentally by methotrexate.
ment. Ann Oncol 1999;10:1251–1253.
18. Cello JP, Grendell JH, Basuk P, et al. Effect of 27. Savarese D, Al-Zoubi A, Boucher J.
7. De Roy van Zuidewijn DBW, Schillings PHM, octreotide on refractory AIDS-associated diar- Glutamine for irinotecan diarrhea (letter). J Clin Wobbes TH, et al. Morphometric analysis of the rhea: A prospective, multicenter clinical trial. Ann effects of antineoplastic drugs on mucosa of 28. Rhoads JM, Keku EO, Bennett LE, et al . De- normal ileum and ileal anastomoses in rats. Exp 19. Wadler S, Benson AB, Engelking C, et al. Rec- velopment of L-glutamine-stimulated electro- ommended guidelines for the treatment of che- neural sodium absorption in piglet jejunum. Am 8. Milles SS, Muggia AL, Spiro HM. Colonic his- tologic changes induced by 5-fluorouracil. Gas- 29. Rhoads JM, Keku EO, Quinn J, et al. L- 20. Petrelli NJ, Rodriguez-Bigas M, Rustum Y, glutamine stimulates jejunal sodium and chlo- 9. Wadler S, Haynes H, Wiernik PH. Phase I trial et al. Bowel rest, intravenous hydration, and con- ride absorption in pig rotavirus enteritis. Gastro- of the somatostatin analog octreotide acetate in tinuous high-dose infusion of octreotide acetate the treatment of fluoropyrimidine-induced diar- for the treatment of chemotherapy-induced di- 30. Fox AD, Kripke SA, DePaula J, et al . Effect rhea. J Clin Oncol 1995;12:222–226.
arrhea in patients with colorectal carcinoma.
of a glutamine-supplemental enteral diet on 10. Physician’s Desk Reference, 57th Ed, 2003.
methotrexate-induced enterocolitis. J Parenter Thompson PDR, Montvale, NJ, pp. 1188–1189.
21. Goumas P, Naxakis S, Christopoulou A, et 11. Ericsson CD, Johnson PC. Safety and effi- al. Octreotide acetate in the treatment of fluo- 31. Diasio RB. Clinical implications of dihydro- cacy of loperamide. Am J Med 1990;20:10–14.
rouracil-induced diarrhea. The Oncologist pyrimidine dehydrogenase on 5-fluorouracil 12. Sthal KS, van Bever W Janssen P, et al. Re- pharmacology. In: Colorectal Cancer: Multimo- ceptor activity and pharmacological potenten- 22. Cascinu S, Fedeli A, Fedeli SL, et al. Oct- dality Management. Saltz LB (ed). Humana Press, cy of a series of narcotic analgesic, antidiarrheal, reotide versus loperamide in the treatment of fluorouracil-induced diarrhea: A randomized tri- with rates of 23%–33% of severe or life-threat-ening diarrhea.
C hemotherapy-induced diarrhea is the This problem came sharply into focus during most common serious toxicity faced by the the conduct of two National Cancer Institute (NCI)–sponsored studies, NCCTG N9741, an inter- cancer. The problem came sharply into focus 20 group trial comparing irinotecan-based therapy years ago, when high-dose leucovorin was add- with oxaliplatin (Eloxatin)-based therapy in pa- ed to standard fluorouracil (5-FU) regimens for tients with advanced colorectal cancer, and CAL- advanced colorectal cancer, based on observa- GB 89803, a trial of adjuvant therapy with 5-FU + tions by Hakala that leucovorin potentiated the leucovorin with or without irinotecan in patients activity of fluorouracil by stabilizing the ternary with surgically resectable colorectal cancer. Both complex. In the first study of fluorouracil admin- trials were halted because of excess deaths in the istered in combination with high-dose leucov- irinotecan-based therapy arms—more than 4% orin by the Gastrointestinal Study Group, there within 60 days of initiation of therapy in the ad- were eight toxic deaths attributable to a syn- vanced disease study and more than 2% in the drome of watery diarrhea followed by sepsis and vascular collapse. In large part, this resulted from An analysis of the excess deaths by an out- lack of experience with the regimen; investiga- side panel determined that the majority result- tors often continued to treat patients who pre- ed from a gastrointestinal (GI) syndrome consist- sented with diarrhea. With greater experience, ing of watery diarrhea, cramps, vomiting, sepsis, the toxicity of this regimen was reduced to ac- and vascular collapse [2]. Following this analysis, ceptable levels by appropriate dose modifica- drug doses were lowered in the advanced-dis- ease trial, and the adjuvant trial was suspended pending final analysis of the data. Nevertheless, py-induced diarrhea has remained problematic.
these lethal toxicities required a reassessment The incidence of grade 3–5 diarrhea in patients of the management of chemotherapy-induced receiving 5-FU and high-dose leucovorin re- mains at about 30%–40% [1], although among In this month’s issue, Saltz’s thoughtful re- patients receiving 5-FU on an infusional sched- view of the ambiguities and uncertainties in ule, the incidence is somewhat lower. The intro- duction of oral fluoropyrimidine compounds, therapy-induced diarrhea is a helpful contri- such as capecitabine (Xeloda), has not reduced bution to the literature on the subject. He cor- the incidence of severe chemotherapy-induced rectly focuses on the dilemma of defining, diarrhea; it has remained nearly identical to that quantifying, and grading the severity of che- The introduction of irinotecan (Camptosar), measurements of bowel dysfunction, includ- a novel topoisomerase-I inhibitor, which has ing parameters such as stool volumes over 24 demonstrated reproducible activity in ad- hours, are the most quantitative measure, these are not practical tools in clinical prac- the problem. As a single agent, irinotecan tice. Therefore, other, less-quantitative ap- treatment results in prolonged survival com- proaches have evolved to grade diarrhea. The pared with either best supportive care or in- sequential versions of the NCI Common Toxic- fusional 5-FU in patients with advanced col- ity Criteria have been problematic, making comparison of diarrhea severity between stud- first-line therapy with 5-FU + leucovorin. In the ies difficult. Of equal import are the problems first-line setting, combinations of irinotecan associated with the criteria themselves. It is with bolus 5-FU and leucovorin administered not clear, for example, that removing cramps on a weekly basis prolong survival, compared from the overall assessment of diarrhea sever- with 5-FU + leucovorin or irinotecan alone.
ity from version 1.0 to version 2.0 was an ad- Nevertheless, the introduction of irinotecan vance. Furthermore, subjective criteria, such as into the clinical arena has accentuated the “need for parenteral support,” and “mild” ver- sus “moderate increases in colostomy output,” while pragmatic, are less reproducible criteria double-blind design, since half the patients could than quantitative assessment of stool output.
receive up to 15 placebo injections as treatment for diarrhea. Furthermore, an NCI-sponsored, ran- rhea, synthetic opioids, such as loperamide, re- domized, intergroup study of loperamide versus main the mainstay of therapy for mild diarrhea, octreotide received little support, in part because as noted by Saltz. Unfortunately, synthetic opio- of a randomization between enteral and parenter- ids have limited utility in the treatment of mod- al therapy. Despite the absence of definitive data, erate, severe, or life-threatening diarrhea or in multiple experts in the field, after reviewing the the management of persistent diarrhea. This is a available clinical literature, have endorsed routine serious problem, and alternative therapies are of treatment with octreotide in patients with mod- great importance in reducing the high mortali- erate to life-threatening chemotherapy-induced ty rate associated with these syndromes.
In 1997, a panel of investigators devised In summary, chemotherapy-induced diarrhea re- guidelines to assist the community physician in mains a serious clinical problem, which is routinely the recognition and treatment of the GI syn- undertreated by the community physician and drome resulting from fluoropyrimidine and irino- which has a high mortality rate. It is clear that tecan treatment. The guidelines, published in the prompt recognition of this syndrome and aggres- Journal of Clinical Oncology in 1998, define an al- sive early intervention, employing an algorithm gorithm for the management of the GI syndrome such as that previously published, can reduce the [3], and have been recently updated (Benson et mortality from this syndrome and improve the clin- al, manuscript submitted for publication). Unfor- ical outcome for patients. Octreotide remains a use- tunately, Saltz dismisses these without defining ful adjunct in reducing the duration and severity of an alternative approach, leaving the practicing chemotherapy-induced diarrhea. Newer agents, clinician without a reproducible strategy for the such as thalidomide (Thalomid) and cyclooxygen- management of these lethal syndromes.
ase-2 inhibitors, are currently in clinical trials; prelim- Equally unfortunate is his dismissal of the ther- inary results have been promising, but they need to apeutic worth of octreotide acetate (Sandostatin) in the management of advanced chemotherapy- induced diarrhea. Octreotide, a synthetic, stabilized Richard T. Silver Distinguished Professor of analog of the endogenous enzyme somatostatin, has been shown to decrease fluid output from the Weill Medical College of Cornell University upper GI tract, increase transit time through the gut, and increase fluid and electrolyte absorption in thelarge bowel. It remains a useful agent for reducing REFERENCES
vomiting in patients with bowel obstruction by de- 1. O’Dwyer P, Ryan L, Valone F, et al. Biochemical mod- creasing fluid output. Repeated clinical trials, in- ulation of fluorouracil: a randomized phase III trial of cluding several randomized trials, have demon- five regimens (ECOG study 2290). J Clin Oncol strated the utility of this agent in the management of chemotherapy-induced diarrhea and its superi- 2. Rothenberg ML, Meropol NJ, Poplin EA, et al.
ority over synthetic opioids for persistent diarrhea.
Deaths associated with irinotecan and bolus 5-fluorou-racil/leucovorin: report of an independent panel. J Clin While Saltz’s call for a randomized, double- blind study of octreotide versus standard thera- 3. Wadler S, Benson AB, Catalano R, et al. Recom- py is commendable, investigators in a recent NCI- mended guidelines for the treatment of chemothera- sponsored trial rejected the feasibility of a py-induced diarrhea. J Clin Oncol 1998;16:3169–3178.
References Continued from page 44.
32. Lu Z, Zhang R, Carpenter JT, Diasio RB. De- 37. Bosma PJ, Chowdhury JR, Bakker C, et al . The creased dihydropyrimidine dehydrogenase ac- 34. Bosma PJ. Inherited disorders of bilirubin genetic basis of the reduced expression of biliru- tivity in a population of patients with breast can- metabolism. J Hepatol 2003;38:107–117.
bin UDP–glucuronosyltransferase 1 in Gilbert’s cer: Implications for 5-FU-based chemotherapy.
35. Owens D, Evans J. Population studies on Gil- Syndrome. N Eng J Med 1995;333:1171–1175.
bert’s Syndrome. J Med Genet 1975;12:152–156.
38. Tiersten A, Saltz L. Influence of inflamma- 33. Diasio RB, Beavers TL, Carpenter JT. Famil- 36. Monaghan G, Ryan M, Seddon R, et al. Ge- tory bowel disease on the ability of patients to ial deficiency of dihydropyrimidine dehydroge- netic variation in bilirubin UPD–glucuronosyl- tolerate systemic fluorouracil–based chemo- nase: Biochemical basis for familial pyrimidine- transferase gene promotor and Gilbert’s syn- therapy. J Clin Oncol 1996;14:2043–2046.
mia and severe 5–fluorouracil–induced toxicity.

Source: http://www.innovares.com/download/idraton/Chemotherapy-Induced%20Diarrhea.pdf