A Retrospective Review of Paclitaxel-Associated Gastrointestinal Necrosis in Patients with Epithelial Ovarian Cancer

A Retrospective Review of Paclitaxel-Associated Gastrointestinal Necrosis in Patients with Epithelial Ovarian Cancer
of 4
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  GYNECOLOGIC ONCOLOGY  67,  137–140 (1997) ARTICLE NO.  GO974842 A Retrospective Review of Paclitaxel-Associated GastrointestinalNecrosis in Patients with Epithelial Ovarian Cancer Victoria L. Seewaldt,* Joanna M. Cain,† Barbara A. Goff,‡ Hisham Tamimi,‡ Benjamin Greer,‡ and David Figge‡ *  Division of Medical and Gynecologic Oncology, Box 358080, University of Washington, Seattle, Washington 98195;  †  Department of Obstetricsand Gynecology, Pennsylvania State University, Milton S. Hershey Medical Center, Hershey, Pennsylvania 17033;and   ‡  Division of Gynecologic Oncology, University of Washington, Seattle, Washington 98195 Received April 11, 1997 ated with paclitaxel administration [4–6]. We now review Seven patients with gastrointestinal necrosis following paclitaxel  the occurrence of paclitaxel-associated bowel perforation at chemotherapy are reported. Four of seven patients had platinum the University of Washington. From 1991 to 1995 seven refractory disease, while 3/7 patients received primary paclitaxel cases of gastrointestinal necrosis have been observed in pa- therapy. Complications occurred 5 to 16 days following paclitaxel tients treated with paclitaxel for epithelial ovarian cancer at therapy. The most common clinical presentation was fever (7/  the University of Washington. Four of seven cases have 7 patients), neutropenia (6/7 patients), and abdominal pain (6/7 occurred in women receiving paclitaxel for platinum refrac- patients). All seven patients developed gastrointestinal necrosis tory disease. Three patients received primary paclitaxel ther- following the first cycle of paclitaxel chemotherapy. The exact apy. We postulate that gastrointestinal necrosis following mechanism by which this complication occurs is poorly under- paclitaxel administration is a direct drug effect. This compli- stood. We postulate that gastrointestinal necrosis may be the result cation may represent a synergistic effect of paclitaxel on of a direct drug effect on the gastrointestinal epithelium and mightinvolve a synergistic interaction between compromised bowel and  an already compromised gastrointestinal tract because all paclitaxel-induced mitotic arrest. We observe that the incidence  complications have occurred following the first cycle of  of gastrointestinal necrosis in patients with platinum refractory paclitaxel. We were, however, not able to identify a specific disease is 4 of 108 patients (3.7%). The incidence of this complica- risk factor which predicted gastrointestinal necrosis. tion in patients receiving primary paclitaxel at our institution is3 of approximately 128 patients (2.3%). Eighteen cases to date MATERIALS AND METHODS have been identified in the literature. A high index of suspicion of this complication should be considered forpatients presenting with We reviewed the medical records of the four cases of  neutropenic feverand abdominal pain following paclitaxel chemo- gastrointestinal necrosis from our initial cohort of 108 pa- therapy.   1997 Academic Press tients receiving paclitaxel under National Cancer InstituteTreatment Referral Center Protocol 9103 [3] and subsequent INTRODUCTION  records of approximately 128 women receiving paclitaxel atthe University of Washington from 1991 to 1995. Threeadditional cases were identified. Records were reviewed ret-Epithelial ovarian cancer remains the leading cause of rospectively for pertinent medical history including a historygynecologic cancer deaths in the United States [1]. Over theof abdominal pain, bowel perforation, fistulae formation,past decade, survival has been improved through aggressivetyphilitis, ischemic colitis, and neutropenic sepsis.surgical resection and combination chemotherapy [2]. Thissurvival is not without additional morbidity, as many womenundergo multiple abdominal operations, extensive courses  RESULTS ofchemotherapy,and,insomeinstances,intraperitonealche-motherapy or whole abdominal irradiation. The cumulative Patient characteristics and complications observed in theseven patients who developed gastrointestinal necrosis aftereffect of these treatment modalities, as well as the naturalhistory of ovarian cancer, puts patients at high risk for bowel receiving paclitaxel are summarized in Tables 1 and 2, re-spectively. Patient presentations are summarized in Table 3.complications.In 1993, we first reported that bowel perforation might be Seven of seven patients presented with fever, 6/7 patientspresented with neutropenia, and 6/7 patients presented witha direct toxicity of paclitaxel [3]. Since that time there havebeen several other reports of gastrointestinal necrosis associ- abdominal pain. The mortality following this complication 137 0090-8258/97 $25.00Copyright    1997 by Academic PressAll rights of reproduction in any form reserved.  138  SEEWALDT ET AL. TABLE 1  cycle of paclitaxel (Tables 2 and 4). Four of seven patient Patient Characteristics  had received multiple prior courses of chemotherapy andthree of seven patients received primary paclitaxel chemo- Courses of Prior therapy (Tables 1 and 4). We hypothesize that these compli- chemotherapy number Prior cations might be the direct effect of paclitaxel on the gastro- Patient prior to abdominal abdominal Intraperitoneal intestinal mucosa. number Age paclitaxel surgery radiation chemotherapy Gastrointestinal complicationsoccur frequentlyinpatients 1 53 3 1 Yes No with epithelial ovarian cancer but the incidence of bowel 2 40 3 2 No No perforation and fistula formation appears to be significantly 3 63 3 2 No Yes lower. The true incidence of fistula formation in patients 4 54 3 2 No No with epithelial ovarian cancer is difficult to estimate because 5 43 0 1 No No6 58 0 2 No No it is poorly described in the literature. Ruben  et al.  [12] 7 70 0 1 No No reported the frequency of bowel complications in all patientswith gynecologic malignancy from 1983 to 1985 at Memo-rial Sloan-Kettering Cancer Center. Of 1614 patients under-was 43% (3/7). Potential risk factors for the developmentgoing exploratory laparotomy for ovarian cancer, 51 patientsof gastrointestinal necrosis are summarized in Table 4. Allrequired further surgery for gastrointestinal complications.patients developed complications following the first cycleOf these 51 procedures, 44 (86%) were performed for bowelof paclitaxel chemotherapy. Four of seven received priorobstruction and 7 (14%) for fistula formation. One of thesechemotherapy but 3/7 patient received primary paclitaxel.7 patients had received abdominal radiation. It was not notedWe observe that the incidence of the development of gastro-whether the fistulas were thought to have occurred as theintestinal necrosis in patients who received paclitaxel forresult of postoperative complications or as a direct conse-platinum refractory epithelial ovarian cancer was 4 of 108quence of progressive disease. These data were reported aspatients (3.7%). The incidence of this complication in pa-a percentage of all surgeries performed and not in relation-tients receiving primary paclitaxel chemotherapy is 3 of ap-ship to a defined population of patients with epithelial ovar-proximately 128 patients (2.3%) at our institution.ian cancer. Therefore, it is not possible to estimate a trueincidenceoffistulaformationinpatientswithadvancedovar- DISCUSSION ian cancer from these data.The gastrointestinal complications we observe do not ap-Paclitaxel is a unique antineoplastic agent with a novelpear to be solely the result of disease progression. If thismechanism of action. Paclitaxel binds preferentially towere the case, ovarian cancer should be present at the sitepolymerized tubulin and thus blocks cell division at M-of gastrointestinal perforation and fistula formation. Whilephase [6]. Unlike cisplatin, the gastrointestinal side effectsthere was a correlation between the anatomic location of of paclitaxel are mild, and grade III to IV toxicity is infre-disease and the site at which complications occurred forquent [7–11].Patients 2, 3, 4, and 6; this was not the case for Patient 1.We observed seven cases of gastrointestinal necrosis inPatient 1 had disease present in the pleural cavity and adja-patients treated with paclitaxel from 1991 to 1995. Patientscent chest wall but had no disease present at the site of boweltypically presented with fever (7/7), neutropenia (6/7), andperforation. The absence of peritoneal disease in Patient 1abdominal pain (6/7) (Table 3). Three of seven patients died(Table 2). All complications occurred following the first was confirmed by surgical exploration and by microscopic TABLE 2Summary of Complications Patient Cycle of Day of Day of Tumornumber paclitaxel complication nadir Complication Surgery Outcome response1 1 16 7 Perforation Yes Improved SD2 1 14 6 Perforation No Improved PD3 1 13 7 Fistula No Death PD4 1 5 6 Fistula No Improved PD5 1 7 7 Perforation Yes Improved SD6 1 9 9 Perforation No Death ?7 1 8 7 Perforation? Yes Death ?  Note.  PD denotes progressive disease; SD denotes stable disease.  139 GASTROINTESTINAL NECROSIS AND PACLITAXEL TABLE 3  sis by demonstrating the formation of abnormal spindle Patient Presentation  asters during mitosis [14, 15]. Three of the gastrointestinalcomplications we describe were detected at a time period Patient Abdominal later than what is described by Hruban  et al.  [13]. This, number Fever Neutropenia pain however, does not exclude the possibility that mitotic arrestis an initiating event in the development of gastrointestinal 1  / 0 / 2  / / /  epithelial necrosis, as it may play an important role in a 3  / / 0 multistep process. A second possibility is that the complica- 4  / / / tion might not have been immediately recognized and there 5  / / / was a delay in diagnosis. 6  / / / It is interesting that the gastrointestinal perforation and 7  / / / fistula formation we describe were observed only after thefirst cycle of paclitaxel and did not occur with subsequentcycles. If gastrointestinal necrosis were solely the result of examination of the resected perforation site. We were notpaclitaxel, complications should occur at equal frequencyable to identify whether Patients 5 and 7 had disease presentwith each cycle. If it were a cumulative effect, complicationsat the site of perforation.should be seen at increasing frequency with each additionalGastrointestinal perforation and fistula formation were notcycle of therapy. The fact that gastrointestinal necrosis wasthe result of rapid tumor lysis. Patients 2, 3, and 4 hadseen only as a first cycle toxicity in patients with epithelialdisease progression and Patients 1 and 5 had stable diseaseovarian cancer suggests that paclitaxel might unmask a pre-(Table 2). It is difficult to determine whether Patients 6 andexistingcondition. Heavilypretreatedpatientswithepithelial7 had any significant clinical response to paclitaxel becauseovarian cancer are likely to have compromised bowel mu-they died before an exact clinical response could be deter-cosa due to the cumulative effects of multiple surgeries,mined; however, significant residual disease was observedprogressive disease, whole abdominal radiation therapy, andduring the autopsy of Patient 6. In no case was there aintraperitoneal chemotherapy. While paclitaxel affects thedramatic decrease in any disease parameter (measurableentire gastrointestinal tract, necrosis may occur at areas of disease by physical exam, computerized tomography, orpreexisting subclinical bowel injury.CA-125).It is difficult to define a specific risk factor which predictsThe seven gastrointestinal complications we report werethe development of gastrointestinal necrosis. Patients 1, 2,detected between 5 and 16 days following paclitaxel admin-3, and 4 were typical of heavily pretreated patients whoistration. Hruban  et al.  [13] describe three cases of paclitaxelreceived paclitaxel under National Cancer Institute Treat-gastrointestinal epithelial necrosis in patients with acute leu-ment Referral Center Protocol 9103 [11] but they did notkemia, which they postulated was the result of transienthave an exceptionally high number of surgical proceduresmitotic arrest occurring 1 to 11 days following treatment.or courses of chemotherapy relative to other patients placedMitotic arrest was not observed in autopsy specimens ob-on this protocol. Patient 3 had a prior history of intraperito-tained fromtwo additional patients 17 and 27 days followingneal chemotherapy and Patient 1 received cobalt radiationpaclitaxel treatment. The mechanism for this metaphase ar-therapy in 1963. It is unclear what role these treatmentsrest is not completely understood but it is postulated to resultplayed in the subsequent development of gastrointestinalfromaninabilityof cellstoformacompetent spindleappara-perforationandfistulaformation. Multiplecoursesofchemo-tus or to disassociate a drug bound spindle. Electron micros-copy and immunofluorescence studies support this hypothe- therapy did not appear to be a risk factor as Patients 5, 6, TABLE 4Potential Risk Factors for the Development of Gastrointestinal Necrosis Prior Necrosis atPatient Prior abdominal the site of Disease Other medicalnumber chemotherapy radiation tumor progression risk factors1  / / 0 0 0 2  / 0 / / 0 3  / 0 / / 0 4  / 0 / / 0 5  0 0  ?  0 0 6  0 0 /  ?  0 7  0 0  ? ?  /  140  SEEWALDT ET AL. and 7 received primary paclitaxel. The risk factors for these perforation (Patient 7). It is difficult to identify specific risk factors, however, and one patient (Patient 5) did not havethree patients are unclear. Patient 5 did not appear to haveany identifiable risk factors. Patient 6 developed abdominal any identifiable risk factors for gastrointestinal necrosis. Ahigh index of suspicion for bowel perforation should bepain and fever prior to the diagnosis of epithelial ovariancancer, suggesting that she might have had an antecedent maintained for any patient presenting with abdominal painfollowing paclitaxel.bowel injury prior to receiving paclitaxel; however, this wasnot documented at the time of diagnostic surgical explora- REFERENCES tion. Patient 7 had a history of polyarteritis nodosa andchronic steroid use which would put her at risk for bowel 1. Wingo PA, Tong T, Bolden S: Cancer statistics, 1995. CA Cancer J perforation. The remaining six of seven patients were other- Clin 45:8–20, 1995 wise healthy and did not have medical conditions which 2. Thigpen JT, Blessing JA, Vance RB, Lambuth BW: Chemotherapy in would put them at risk for gastrointestinal necrosis. ovarian carcinoma: present role and future prospects. Semin Oncol Gastrointestinal necrosis following paclitaxel therapy ap-  16:58–65, 1989 pears to be a rare but significant complication associated  3. Seewaldt V, Greer B, Tamimi H, Figge D, Brown W, Cain J: Bowelperforation following Taxol in heavily pre-treated ovarian cancer pa- with a 43% mortality (3/7). The true incidence of this com- tients. J Clin Oncol 11:1198, 1993 plication, however, is difficult to estimate because the pre- 4. Pestalozzi BC, Sotos GA, Choyke PL, Fisherman JS, Cowan KH, O’S- senting symptoms may be confused with disease progres- haughnessy JA: Typhilitis resulting from treatment with Taxol and sion. In addition to the 7 cases we report, there have been doxorubicin in patients with metastatic breast cancer. Cancer 71:1797– 11cases of gastrointestinal necrosisreported intheliterature,  1800, 1993 bringing the total number of cases to 18. Gastrointestinal  5. National Institute of Health: Communication, May 3, 1994 necrosis has been observed in three patients with leukemia  6. Rose PG, Piver MS: Intestinal perforation secondaryto paclitaxel. Gyn-ecol Oncol 57:270–272, 1995 [13]. Two cases of ischemic colitis with paclitaxel chemo- 7. Schiff P, Faut J, Horowitz S: Promotion of microtubule assembly in therapy in head and neck cancer [5] and three cases of typhi- vitro by taxol. Nature 277:665–667, 1979 litis or ishemic colitis in patients with breast and lung cancer 8. RowinskyEK, Cazenave LA, Donehower RC: Taxol: a novel investiga- have been observed [4]. Three cases of bowel perforation tional anti-microtubule agent. J Natl Cancer Inst 82:1247–1259, 1990 were observed in patients with epithelial ovarian cancer [6]. 9. McGuire WP, Rowinsky EK, Rosenshein NB: Taxol: a unique anti- In summary, we reviewed the incidence of gastrointestinal neoplastic agent with significant activity in advanced ovarian epithelial necrosis following paclitaxel therapy at the University of  neoplasm. Ann Int Med 111:273–297, 1989 Washington from 1991 to 1995 and observed that seven  10. Thigpen T, Blessing J, Ball H, Hummel S, Barret R: Phase II trial of  patientsdevelopedgastrointestinal necrosis5to16daysafter  Taxol as second-line therapy for advanced ovarian adenocarcinoma: AGynecologic Oncology Group Study. Proc Am Soc Clin Oncol receiving their first course of paclitaxel. We estimate that 10:1748–1753, 1992 the incidence of gastrointestinal necrosis is approximately 11. Trimble EL, Adams DV, Hawkins MJ, Friedman MA, Fisherman JS, 2.7 to 3.6%. The most common clinical presentation was Christian MC, Canetta R, Onetto N, Hayn R, Arbuck SG: Paclitaxel fever, neutropenia, andabdominal pain. These complications for platinum-refractory ovarian cancer: Results from the first 1,000 occurred in both heavily pretreated patients and patients re-  patient registered to National Cancer Institute treatment referral center9103. J Clin Oncol 11:2405–2410, 1993 ceiving primary paclitaxel therapy and are associated with a 12. Rubin SC, Benjamin I, Hoskins WJ: Intestinal surgery. Gynecol Oncol significant mortality (43%). We feel that these complications 34:30–33, 1988 might be a direct toxic effect of paclitaxel on the gastrointes- 13. Hruban RH, Yardly JH, Donehower RC: Taxol toxicity. Cancer tinal epithelium. Since all observed cases of gastrointestinal 63:1944–1950, 1989 necrosis have occurred following the first cycle of paclitaxel, 14. Manfredi JJ, Horwitz SB: Taxol: an antimitotic agent with a new mech- it is possible that there is a synergistic effect between pacli- anism of action. Pharmacol Ther 25:83–125, 1984 taxel and a gastrointestinal tract compromised by multiple 15. Rowinsky EK, Donehower RC, Jones RJ: Microtubule changes and courses of chemotherapy and prior surgeries (Patients 1–4)  cytotoxicity in leukemic cell lines treated with taxol. Cancer Res48:4093–4100, 1988 or by medical conditions which might predispose to bowel
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks