SEARCH

SEARCH BY CITATION

Keywords:

  • paclitaxel;
  • docetaxel;
  • colitis;
  • typhlitis;
  • breast carcinoma

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

BACKGROUND

Colitis is a rare but serious gastrointestinal complication associated with taxane-based chemotherapy in patients with cancer. The incidence, clinical presentation, and outcome of colitis in patients with breast carcinoma treated with taxane-based chemotherapy is not known.

METHODS

The authors searched their electronic database and identified patients with breast carcinoma who were treated with taxane-based chemotherapy between January 1997 and December 1999. Patients diagnosed with colitis were identified and their presentation and clinical outcomes were analyzed.

RESULTS

During the study period, 1350 patients received taxane-based chemotherapy. Sixty-four patients were admitted to The University of Texas M. D. Anderson Cancer Center (Houston, TX) (73 admissions) for gastrointestinal complications. Neutropenia and/or fever accounted for 56 of these admissions. Fourteen patients (16 admissions) were diagnosed with colitis. All had abdominal pain at median Day 6 of their chemotherapy cycle (range, 3–8 days), with or without other symptoms. Computed tomography scans of the abdomen and pelvis were abnormal for the 10 patients tested, whereas only 3 of the 9 patients who underwent radiographic abdominal series had abnormal findings. Two patients had a colonoscopy that confirmed the diagnosis of colitis. Blood cultures were positive in only 3 (20%) of 15 colitis events. All eight patients tested for Clostridium difficile toxin were negative. One patient died of sepsis. An autopsy subsequently revealed cecal ulceration. Two patients had bowel perforations requiring hemicolectomy. Colitis recurred in two patients after they resumed taxane-based chemotherapy without dose reduction. Dose reductions prevented the recurrence of colitis in seven patients. Treatment discontinuation of taxane-based chemotherapy also prevented disease recurrence.

CONCLUSIONS

Acute abdominal pain in patients with breast carcinoma treated with taxane-based chemotherapy signaled a potentially fatal colitis, and warrants aggressive supportive care. Dose reduction or discontinuation of taxane-based chemotherapy may prevent the recurrence of colitis. Cancer 2004. © 2004 American Cancer Society.

Typhlitis, a necrotizing inflammation of the cecum, typically occurs in patients with leukemia who have prolonged chemotherapy-induced neutropenia,1 as well as in patients with solid tumors who undergo chemotherapy.2 Typhlitis is characterized by neutropenic fever associated with stomatitis, nausea, emesis, diarrhea, and abdominal pain.3 Thickening of the cecal wall and aerobic gram-negative septicemia are also common features.3, 4

In contrast to patients with typhlitis, patients with breast carcinoma with metastatic disease may develop ischemic colitis when treated with docetaxel-based chemotherapy (DBC).5 Colitis is defined as occurring in any patient presenting with a symptom complex of acute abdominal pain and direct or rebound tenderness, with possibly associated neutropenia, fever, and/or diarrhea (with or without blood). In addition, suggestive computed tomography (CT) scan of the abdomen and pelvis and a negative Clostridium difficile titers support the diagnosis of hemic colitis.5 Patients who receive paclitaxel-based chemotherapy (PBC) may also present with a similar symptom complex (Nuhad Ibrahim, unpublished data). Because docetaxel and paclitaxel are widely used for treating breast carcinoma and other solid tumors,6–9 it is important to determine the frequency of this complication, as well as to recognize, diagnose, and manage it effectively. We, therefore reviewed, the medical records of patients with breast carcinoma who were admitted to The University of Texas M. D. Anderson Cancer Center (MDACC; Houston, TX) with gastrointestinal complications while receiving taxane-based chemotherapy to analyze the incidence, presentation, and clinical outcomes of colitis.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

We performed an Investigation Research Board-approved search of our institutional pharmacy database registry to identify patients with breast carcinoma who received taxane-based chemotherapy at MDACC between January 1997 and December 1999. We then searched the inpatient registry of the Department of Breast Medical Oncology to identify patients who were admitted to MDACC with gastrointestinal complications after they received taxane-based chemotherapy. We reviewed the patients' medical records and documented their age, disease stage, type of chemotherapy (including dose and schedule), presenting symptoms and their onset related to chemotherapy administration, diagnostic workup, final diagnosis at discharge, clinical management, and clinical outcome.

Docetaxel or paclitaxel was used either as a single agent or in combination with other chemotherapy agents. For most patients, the paclitaxel doses ranged from 175 to 250 mg/m2 every 3 weeks and the docetaxel doses ranged from 50 to 100 mg/m2 every 3 weeks. Each dose was counted as one cycle. A small percentage of patients received paclitaxel (80–175 mg/m2) or docetaxel (30–40 mg/m2) weekly for 3 weeks followed by 1 week off and this was counted as one cycle as well.

Standard premedications were administered to patients who received paclitaxel or docetaxel every 3 weeks. For paclitaxel, dexamethasone (20 mg) was taken orally 12 hours and 6 hours before paclitaxel infusion and then diphenhydramine (50 mg) and cimetidine (300 mg) were given intravenously (i.v.) 30 minutes before paclitaxel infusion. If patients forgot to take oral dexamethasone, then dexamethasone (20 mg) was administered i.v. 30 minutes before paclitaxel infusion. For docetaxel, dexamethasone (8 mg) was taken orally twice a day for 3 days, starting 1 day before chemotherapy infusion to help prevent or decrease the incidence of the capillary leak syndrome. If patients forgot to take dexamethasone, then dexamethasone (20 mg) was administered i.v. 30 minutes before docetaxel infusion. Patients resumed taking dexamethasone (8 mg) orally twice a day for 2 days after chemotherapy.

For patients receiving weekly paclitaxel at a dose > 100 mg/m2, dexamethasone (20 mg), diphenhydramine (50 mg), and cimetidine (300 mg) were given i.v. 30 minutes before paclitaxel infusion. If no hypersensitivity or allergic reaction occurred for the initial 3 doses of paclitaxel, then dexamethasone was reduced to 10 mg in subsequent cycles. For patients receiving weekly paclitaxel at a dose ≤ 100 mg/m2, dexamethasone (10 mg) was given i.v. 30 minutes before paclitaxel infusion. If no hypersensitivity or allergic reaction occurred for the initial 3 doses, then dexamethasone was reduced to 5 mg. For patients receiving weekly doses of docetaxel, dexamethasone (4 mg) was taken orally every 12 hours for 3 doses starting 12 hours before docetaxel infusion. If no hypersensitivity or allergic reaction occurred for the initial 3 doses, then dexamethasone was gradually reduced to one 4-mg dose given 12 hours before chemotherapy infusion. Antiemetics were not required for either paclitaxel or docetaxel infusion.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

In the current study, 1350 patients with breast carcinoma received taxane-based chemotherapy at MDACC between January 1997 and December 1999 (Table 1). Eight hundred and thirty patients received PBC, of whom 42 patients received PBC after they failed to respond to DBC. Conversely, 520 patients received DBC, of whom 45 patients received DBC after they failed to respond to PBC. Sixty-four patients were admitted to MDACC (a total of 73 admissions) for National Cancer Institute Grade III/IV gastrointestinal complications secondary to PBC (26 patients, 41%) or DBC (38 patients; 59%). Colitis was diagnosed in 14 patients, 2 of whom had recurrence of colitis when they were rechallenged with the same taxane-based chemotherapy at the same dosage.

Table 1. Number of Patients who Received Taxane-Based Chemotherapy and Type of Complications
Chemotherapy and complicationsNo. of patients (%)
  1. PBC: paclitaxel-based chemotherapy; DBC: docetaxel-based chemotherapy; GIC: gastrointestinal complications.

PBC830
DBC520
Total1350
Admitted for GIC64 (5)
 PBC26
 DBC38
Admitted for colitis14 (1)
 PBC4
 DBC10

Ten of 520 patients (1.9%) who received docetaxel treatment developed colitis, of whom 7 patients had no previous exposure to taxanes and 3 had docetaxel therapy after failing to respond to paclitaxel. Conversely, only 4 of 830 patients (0.5%) who received paclitaxel developed colitis—3 patients had no previous taxane exposure and the other 1 had paclitaxel after failing to respond to docetaxel. The risk ratio of developing colitis after PBC versus DBC was 0.35 (95% confidence interval = 0.19–0.64). Colitis events were encountered in patients receiving PBC or DBC as either neoadjuvant chemotherapy or as treatment for metastatic breast carcinoma. Although most colitis events occurred when chemotherapy was administered every 3 weeks, colitis also occurred when either docetaxel or paclitaxel was given as a single agent at weekly intervals. Colitis could occur after the patient received the first cycle of chemotherapy, or after any cycle thereafter. Patients' age, complications that precipitated admissions, mean taxane doses, and the number of cycles leading to colitis are summarized in Table 2. Five patients received docetaxel in combination with another agent (cyclophosphamide [n = 1], doxorubicin [n = 2], vinorelbine [n = 2]), and only 1patient received doxorubicin in combination with paclitaxel. All other patients received taxane as a single agent. No differences were observed between the different groups in terms of colitis presentation. There were no significant differences in age, doses, and number of treatment cycles among patients who tolerated chemotherapy without complications, patients who required admission for noncolitis gastrointestinal complications, and patients admitted with colitis. Development of severe gastrointestinal toxicities or colitis did not seem to be affected by the cumulative doses of the specific taxane administered (data not shown).

Table 2. Age, Dose, Type, and Number of Cycles of Taxane Chemotherapy for Patients with or without Gastrointestinal Complications Requiring Hospital Admission
Admission for GICNo. of patients (n = 1350)Mean age (range)Mean taxane doses (mg/m2/cycle) (range)No. of cycles (range)
  1. PBC: paclitaxel-based chemotherapy; DBC: docetaxel-based chemotherapy; GIC: gastrointestinal complications.

PBC    
 No80451 (23–86)212 (120–525)4 (1–18)
 Noncolitis GIC2151 (33–70)206 (165–525)4 (1–7)
 Colitis446 (42–67)206 (180–225)2 (2–4)
DBC    
 No48251 (25–58) 76 (48–120)4 (1–21)
 Noncolitis GIC2950 (28–61) 76 (56–100)6 (1–15)
 Colitis1050 (37–62) 75 (52–85)2 (1–17)

Although nausea and emesis were the most common gastrointestinal complications necessitating hospital admission (62% of admissions), acute abdominal pain was the most common presenting symptom in patients with colitis (100% of admissions) (Table 3). The median time to onset of colitis was 6 days after the adminstration of chemotherapy (range, 3–8 days). Neutropenia and/or fever were not consistently present in patients with colitis, manifested in 38% and 6% of events, respectively (Table 3).

Table 3. Symptoms at the Time of Admission for Gastrointestinal Complications Related to Chemotherapy
SymptomsNo. of admissions related to gastrointestinal complications (n = 73) (%)aNo. of admissions related to colitis (n = 16)b
  • a

    Sixty-four patients were admitted 73 times.

  • b

    Fourteen patients were admitted 16 times.

Nausea and emesis46 (63) 9 (56)
Abdominal pain35 (48)16 (100)
Rebound tenderness 5 (7) 5 (31)
Nonbloody diarrhea27 (37) 6 (38)
Bloody diarrhea 7 (10) 4 (25)
Stomatitis34 (17) 3 (19)
Constipation10 (14) 4 (25)
Bowel perforation 2 (3) 2 (12)
Neutropenia and fever47 (64) 8 (50)
Neutropenia without fever 3 (4) 1 (6)
Fever only 6 (8) 1 (6)
No neutropenia and no fever17 (23) 6 (38)

To further characterize taxane-associated colitis, we reviewed the diagnostic workup performed for all 14 patients (16 admissions) with colitis. CT scans of the abdomen and pelvis were abnormal for all of the 10 tested patients. CT scan findings consistent with colitis included diffuse or localized thickening of the colonic wall; pneumoperitoneum was evident in one patient. Abnormal findings on abdominal X-ray series, including ileus, intestinal obstruction pattern, and pneumoperitoneum with pneumatosis of the colonic wall, were present in three of the nine tested patients. Two patients underwent colonoscopies that showed ischemic colitis. Blood cultures were performed for 15 admissions for colitis (all 14 patients were cultured), and were positive in only 3 (20%) episodes. The bacteria isolated were coagulase-negative Staphylococcus from all three patients in addition to Stenotrophomonas maltophilia isolated in one of these patients. C. difficile toxin was not detected in stool samples collected from eight patients.

All patients received supportive care with fluid resuscitation, broad-spectrum antibiotics, and close surgical monitoring. One patient died of septic shock on Day 8 of the chemotherapy cycle. Necropsy revealed a 3 × 3-cm well demarcated ulceration in the cecum. Blood cultures from this patient grew both coagulase-negative Staphylococcus and S. maltophilia. Two patients underwent exploratory laparotomy and hemicolectomy. Histology revealed bowel perforations secondary to transmural necrosis.

Two patients had recurrences of colitis when they resumed taxane-based chemotherapy at the same dose level. One patient was receiving single-agent paclitaxel and the other was receiving docetaxel-based combination chemotherapy. Both of these patients fully recovered with supportive care. Of the 7 patients in whom the dose of the taxane was empirically reduced by 10%, none manifested the recurrence of their symptoms. Two of these patients were receiving single-agent paclitaxel, three were receiving single-agent docetaxel, one was receiving paclitaxel in combination with doxorubicin, and another was receiving docetaxel in combination with doxorubicin. There was no recurrence of colitis in four patients who discontinued taxane-based chemotherapy and switched to other chemotherapy regimens.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Gastrointestinal complications frequently occur in patients undergoing chemotherapy.10 Typhlitis, a serious gastrointestinal complication, is typically associated with prolonged neutropenic fever, stomatitis, emesis, diarrhea, and abdominal pain. Symptoms often occur 10–14 days after initiation of chemotherapy.3 Radiologically, typhlitis manifests as paralytic ileus and thickening of the cecum.3 Septicemia frequently occurs and the most common causative organism is aerobic gram-negative bacteria.4

A related clinical syndrome was described in patients with metastatic breast carcinoma treated with DBC.5 In the current study, we found that PBC also resulted in this complication. Although acute abdominal pain was a common presenting feature, other characteristics that are typically associated with typhlitis—neutropenia, fever, positive blood cultures, and late onset—may not necessarily be present. This complication occurs early (median Day, 6) in the treatment course, with or without neutropenia. In contrast, typhlitis usually occurs 10–14 days after initiation of treatment and neutropenia is always present.3 A histologic review of the available samples suggested ischemic colitis and bowel perforation secondary to transmural necrosis.5 These histologic changes are different from those that are typically associated with typhlitis, such as patchy inflammation of the bowel wall, well demarcated ulcers with a paucity of inflammatory cells, hemorrhages, and clusters of microorganisms.11.

Severe gastrointestinal complications, including bowel necrosis, perforation, or typhlitis, have been described previously in patients treated with paclitaxel or docetaxel chemotherapy. Colonic perforation and necrosis associated with neutropenic fever were reported in patients with ovarian carcinoma, occurring Days 5–16 after receiving paclitaxel as a single agent (9 patients)12, 13 or in combination with cisplatin (1 patient).12 Pathology revealed bowel necrosis with no evidence of tumor involvement at the site of perforation. Typhlitis was reported in patients with metastatic breast carcinoma treated with paclitaxel in combination with cyclophosphamide14 or with doxorubicin.15, 16 These patients presented with diarrhea, abdominal pain, and neutropenic fever. Radiography revealed typhlitis. Of note, no disease recurrences were observed in three patients who continued chemotherapy with reduced doses of paclitaxel and doxorubicin.16 In another study, a patient with squamous cell carcinoma of the lung died on Day 5 after the first dose of docetaxel when he developed emesis, abdominal pain, neutropenic fever, and septic shock.17 Blood cultures showed Clostridium septicum infection. Postmortem examination showed intense edema of the intestinal wall, air within the mucosa and submucosa, necrosis without inflammatory infiltrate, and extensive bacterial infiltration that affected primarily the terminal ileum and cecum. Two other taxane-related toxic deaths were reported in patients with metastatic breast carcinoma. One death was associated with the combination of docetaxel and vinorelbine18 and the other was associated with docetaxel/epirubicin treatment.19 In both patients, the cause of death was ascribed to typhlitis as well.

Paclitaxel and docetaxel have been used widely to treat solid tumors as single agents or in combination with other cytotoxic agents because of their broad spectrum of antitumor activity.9 In the current study of patients with breast carcinoma, colitis events were encountered with taxane-based chemotherapy in both the neoadjuvant setting and treatment for metastatic breast carcinoma. Colitis could occur when taxane-based chemotherapy was administered either every 3 weeks or at weekly intervals. Although colitis associated with taxane-based chemotherapy is a rare complication, it is important to recognize and manage this complication because it is potentially fatal.

The risk ratio of developing colitis after PBC versus DBC is 0.35. However, caution is indicated before any conclusions can be drawn from these data. The characteristics of the patients who received PBC or DBC were different. A significant portion of patients who received DBC had advanced-stage of disease and were heavily treated with different chemotherapy regimens. Conversely, a significant portion of patients who received PBC had early-stage disease and received paclitaxel as neoadjuvant or adjuvant treatment. It is not clear whether or not these differences may contribute to developing side effects of chemotherapy, including colitis.

The mechanism of colitis as a complication of taxane-based chemotherapy is still not known. Both paclitaxel and docetaxel bind to the β subunit of tubulin, which results in the formation of stable, nonfunctional microtubule bundles and interferes with mitosis.20, 21 In one study, the tissue samples obtained from patients within 11 days of receiving paclitaxel revealed mitotic arrest secondary to accumulation of polymerized microtubules in the gastrointestinal tract, liver, and bone marrow.22 However, mitotic arrest was not present in tissue samples obtained > 11 days after paclitaxel exposure. Therefore, it was postulated that gastrointestinal necrosis or bowel perforation after paclitaxel administration is a direct drug effect as a result of transient mitotic arrest due to nonfunctional microtubule bundles.13, 22 Paclitaxel also has strong antiangiogenic activity both in vitro and in vivo,23 and this antiangiogenic property is associated with a down-regulation of vascular endothelial growth factor in transgenic murine breast carcinoma.24 More recently, emerging evidence has shown that taxanes can also induce apoptosis.25 An in vitro study of hepatoma cell lines showed that both paclitaxel and docetaxel induced apoptosis mediated through G2/M-phase arrest, caspase activation, and DNA fragmentation.25 Whether the ischemia and transmural necrosis observed in the biopsy samples and resected bowels of our patients were related to such apoptosis and inhibition of angiogenesis needs to be validated.

An attempt was made to determine any potential predisposing factors in the few patients who developed colitis after taxane treatment. Among 14 patients with colitis, 7 patients had remote (> 5 years ago) abdominal surgery including hysterectomy, cesarean section, or appendectomy. Three patients were actively smoking tobacco for > 20 pack-years and two patients were ex-smokers who quit smoking > 5 years ago. None of the patients had diabetes mellitus or vascular diseases such as stroke, coronary artery disease, or peripheral vascular disease.

In summary, we identified 14 patients with breast carcinoma with 16 episodes of colitis (two recurrences) that occurred during taxane-based chemotherapy. Patients with breast carcinoma presenting with acute abdominal pain while receiving taxane-based chemotherapy should be suspected of having ischemic colitis. CT scans of the abdomen and pelvis may be necessary to establish a diagnosis. Colonoscopy may be associated with an increased risk of perforation and should be discouraged. Aggressive supportive care with i.v. fluids, broad-spectrum antibiotics, and close surgical monitoring should continue until complete resolution of the colitis. Colitis in patients with breast carcinoma receiving taxane-based chemotherapy appears to be dose related, and reduction in taxane doses or discontinuation of taxane therapy may successfully prevent recurrences of colitis.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The authors thank Judy Dillon and Lynn Bratton for secretarial assistance and Noelle Heinze of the Department of Scientific Publications for editing.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES