- Email: [email protected]
Which factors predict bowel complications in patients with recurrent epithelial ovarian cancer being treated with bevacizumab?
Gynecologic Oncology 118 (2010) 47–51
Contents lists available at ScienceDirect
Gynecologic Oncology j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y g y n o
Which factors predict bowel complications in patients with recurrent epithelial ovarian cancer being treated with bevacizumab? D.L. Richardson, F.J. Backes, J.D. Hurt, L.G. Seamon, L.J. Copeland, J.M. Fowler, D.E. Cohn, D.M. O'Malley ⁎ Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, The Ohio State University College of Medicine, M-210 Starling Loving Hall, 320 West 10th Avenue, Columbus, OH 43210, USA
a r t i c l e
i n f o
Article history: Received 9 September 2009 Available online 10 April 2010 Keywords: Bevacizumab Bowel perforation Fistula Risk factors Epithelial ovarian cancer
a b s t r a c t Background. Increased rates of bowel perforation in patients with recurrent epithelial ovarian cancer (EOC) treated with bevacizumab have been reported, but the risk factors for this association are uncertain. We sought to identify factors associated with bowel perforation and fistula formation in recurrent EOC patients treated with bevacizumab. Methods. A chart review of all patients treated with bevacizumab for recurrent EOC at a single institution was performed. Pertinent patient characteristics and treatment information were collected. Univariate logistic regression was performed to analyze multiple variables. Results. One hundred twelve patients who were treated with 160 different bevacizumab regimens were identified. The median age was 60 years (range, 29–78 years). Patients had received a median of 4 prior chemotherapy regimens (range, 1–10). The median number of cycles was 4 (range, 0.5–31). Ten patients (9%) were diagnosed with bowel perforations, and another 2 patients (1.8%) were diagnosed with fistulas. The 30-day mortality following perforation was 50%, with 30% of patients dying within 1 week. Patients with rectovaginal nodularity were more likely to develop a bowel perforation or fistula than those who did not have this finding, OR = 3.64 (95% CI = 1.1 to 12.1, p = 0.04). None of the other variables were significantly associated with bowel perforations or fistula formation. Conclusions. Rectovaginal nodularity is associated with an increased risk of bowel perforation or fistula formation for patients with recurrent EOC treated with bevacizumab. Careful consideration should be given prior to initiating bevacizumab treatment in EOC patients with rectovaginal nodularity since the mortality rate with bevacizumab associated bowel perforations is 50%. © 2010 Elsevier Inc. All rights reserved.
Introduction Ovarian cancer is the leading cause of gynecologic cancer deaths in the United States [1]. While the majority of patients respond to front line chemotherapy with platinum and paclitaxel, most advanced stage patients will recur. Although there are many cytotoxic agents with activity in the recurrent setting, ultimately the disease becomes chemotherapy resistant. Therefore, drugs with novel mechanisms of action designed to target specific pathways, for example angiogenesis, are being evaluated for efficacy and safety in both the front line and recurrent setting. Bevacizumab (Genentech, San Francisco, CA, USA), a monoclonal antibody against vascular endothelial growth factor, is increasingly being utilized for the treatment of recurrent ovarian cancer. It has modest activity as a single agent, with reported response rates ranging from 10% to 21% and median progression-free survival (PFS)
⁎ Corresponding author. Fax: +1 614 366 7942. E-mail address: David.O'[email protected] (D.M. O'Malley). 0090-8258/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2010.01.011
of 4–5 months [2-5]. Combinations of bevacizumab and other novel compounds have also been studied. A phase II trial of bevacizumab combined with erlotinib (an EGFR tyrosine kinase inhibitor) demonstrated no improvement in response rates to single agent bevacizumab in the first phase of accrual with a concerning rate of bowel perforation. Therefore, enrollment was halted after 13 patients were treated with this combination [6]. In contrast, response rates of 24– 78% have been reported when bevacizumab has been combined with cytotoxic chemotherapy [7-10]. The large range of response rates may be attributable to different patient populations (platinum sensitivity and number of previous regimens), different cytotoxic drugs used, and different definitions for response (measurable disease versus CA125 levels). Serious bowel complications, including bowel perforations and fistulae, have been reported more commonly in ovarian cancer patients treated with bevacizumab than in most other types of cancer patients treated with this drug. Bowel perforation and fistula have been reported to occur in 0–11% of patients treated with single agent bevacizumab for recurrent ovarian cancer and in 0–9% of patients treated with bevacizumab in combination with cytotoxic
48
D.L. Richardson et al. / Gynecologic Oncology 118 (2010) 47–51
chemotherapy [2-4, 7-10]. The highest bowel perforation rate, 15%, was reported in recurrent ovarian cancer patients treated with combination bevacizumab and erlotinib who had previously received 2 or less prior regimens [6]. The risk factors for bowel perforation and fistula formation in recurrent ovarian cancer patients treated with bevacizumab are unclear. Possible risk factors include number of previous chemotherapy regimens, bowel wall thickening on CT, and bowel obstruction [3,10]. It is also unknown whether careful patient selection can avoid this complication. Although Simpkins et al. [11] suggested that bowel perforation can be avoided by restricting bevacizumab to patients without bowel obstruction, bowel involvement on CT scan, and no evidence of rectosigmoid involvement on pelvic exam, this has not been prospectively evaluated. In the present study, we sought to identify the risk factors for increased risk of gastrointestinal complications (bowel perforation and fistula formation) in consecutive patients with recurrent ovarian cancer treated with bevacizumab. Our secondary aim was to report the associated 30-day mortality rate of bowel perforation in this group of patients. Materials and methods A retrospective chart review was performed of all patients who were treated with bevacizumab with or without cytotoxic chemotherapy for recurrent epithelial ovarian, primary peritoneal, or fallopian tube cancer at a single institution from April 2005 (the first time a patient was treated with bevacizumab for recurrent ovarian cancer at our institution) through August 2008. Institutional review board approval was obtained. Patients were identified by reviewing chemotherapy log books. Patients who received at least one dose of bevacizumab were included. Multiple patients received more than one regimen of bevacizumab. Data from all regimens were abstracted, not only from the first regimen, as the patient remained at risk for bowel complications. Charts were abstracted for pertinent data, including patient characteristics, treatment, response, gastrointestinal complications (bowel perforation, fistula), and management and outcome of gastrointestinal complications. Bowel perforation was diagnosed by either free air on an abdominal radiograph or CT scan, evidence of contrast extravasation on a CT scan, or bowel contents in the abdomen on surgical exploration. Fistula was defined as either clinical or radiographic evidence of an abnormal communication between two epithelium-lined organs. Response rates were calculated for patients who received at least 2 cycles of bevacizumab, with or without cytotoxic chemotherapy. Measurable disease was defined as at least one lesion that was greater than 1 cm on spiral CT, as defined by Response to Treatment in Solid Tumors (RECIST) criteria [12]. In the absence of measurable disease, CA-125 levels were used to assess response per modified Rustin criteria [13]. Complete response was disappearance of gross evidence of disease, resolution of measurable disease on CT scan, or a normalization of CA-125 levels from an elevated level. Partial response was a 30% or greater reduction in measurement of longest dimension of each lesion or N50% decrease in CA-125 levels. Progressive disease was a 20% or greater increase in measurement of longest dimension of each lesion or the appearance of any new lesion or a doubling of CA-125 levels within 8 weeks of the start of chemotherapy. Stable disease was any condition which did not meet the above criteria. Best response was reported. CA-125 levels were routinely drawn with the standard pre-chemotherapy labs. CT scans were obtained as appropriate to follow measurable disease and at the physician's discretion. Simple descriptive statistics were used to describe response rates, including mean, median, and standard deviation. Logistic regression was used to analyze the risk of gastrointestinal complications for multiple variables p b 0.15 from the univariate analysis. Continuous
variables included total dose of bevacizumab (mg), total number of cycles of bevacizumab, age, and body mass index. The number of previous chemotherapy regimens was dichotomized at the median (N4 vs. ≤4), and also at 2 (N2 vs. ≤2) based on previous literature [3]. Binary variables included history of bowel surgery, previous bevacizumab, platinum resistance at time of treatment with bevacizumab, carcinomatosis, history of small bowel obstruction, rectovaginal nodularity, and bevacizumab treatment dose (15 vs. 10 mg/kg). Because of the number of different cytotoxics combined with bevacizumab and the small numbers of patients treated with some of these combinations, the type of cytotoxic combined with bevacizumab was not analyzed as a risk factor for perforation. The final model contained only variables which were significant at the p b 0.05 level. Stata release 10 (College Station, Texas) was used to perform statistical calculations. Results One hundred and twelve patients who were treated with 160 bevacizumab regimens were identified, and all patients were included in the analysis. Thirty-one percent (35/112) of the patients received more than one bevacizumab regimen. The median age was 60 years (range, 29–78 years), and the majority of patients were Caucasian. Sixty-four percent (103/160) were platinum resistant (primary or secondary) at the time of treatment. The median body mass index at the start of treatment was 26.6 kg/m2 (range, 14.9–45.6 kg/m2). The Table 1 Baseline demographics of the study population. Baseline demographics (N = 112) Median age (range), years Mean age (SD) Race Caucasian African-American Stage at diagnosis I II III IV Not staged Histology Serous Endometrioid Mucinous Mixed epithelial Clear cell Transitional Adenocarcinoma NOSb Unknown Grade at diagnosis 1 2 3 Unknown Neoadjuvant chemotherapy Residual disease at primary surgery No gross residual Optimal (b1 cm) Suboptimal (N 1 cm) No surgery Unknown Platinum response Sensitive Resistant/refractory Secondary resistant Median number of previous chemotherapeutic regimens Received more than one regimen of bevacizumab a b c d
Percentages may not add up to 100% due to rounding. NOS, not otherwise specified. Includes 5 patients who received neoadjuvant chemotherapy. Includes 2 patients who received neoadjuvant chemotherapy.
Number (%)a 60 (29–78) 59 (9.5) 107 (96) 5 (4) 4 (4) 7 (6) 89 (79) 11 (10) 1 (1) 83 (74) 6 (5) 5 (4) 5 (4) 2 (2) 1 (1) 7 (6) 3 (3) 4 (4) 14 (13) 85 (76) 10 (9) 36c (32) 53d (47) 17 (15) 1 (1) 5 (4) 75 (67) 37 (33) 30 4 (1–10) 35 (31)
D.L. Richardson et al. / Gynecologic Oncology 118 (2010) 47–51
patients were relatively heavily pre-treated, with a median of 4 prior chemotherapy regimens (range, 1–10) (Table 1). The most common regimen was weekly paclitaxel 60–70 mg/m2 and bevacizumab day 1 and day 15 every 28 days, with 55 patients receiving this regimen. Please refer to Table 2 for a listing of all the bevacizumab combinations. The dose of bevacizumab ranged from 5 to 15 mg/kg in all regimens. A quarter of the regimens (39/160) used bevacizumab 15 mg/kg. The median number of cycles of each regimen was 4 (range, 0.5–31). The median number of cycles of all regimens containing bevacizumab was 6 (range, 0.5–36). The median total dose of bevacizumab was 8313 mg (range, 430–54653 mg). Twenty-one patients (13%) were still undergoing treatment with bevacizumab at the completion of this study. Eighteen patients (16%) had a history of a small bowel obstruction prior to treatment with bevacizumab. Four of these patients were treated with 2 bevacizumab regimens. Twenty-nine percent (31/107) of the patients had rectovaginal nodularity prior to initiation of bevacizumab, 5 of whom were treated with more than one bevacizumab regimen . Rectovaginal exam was not performed or documented in 5 patients. CT scans were obtained about 50% (78/ 160) of the time prior to starting bevacizumab. Five patients had PET scans, and 2 patients had MRIs performed prior to starting treatment. In addition, 2 patients had undergone secondary cytoreduction prior to starting treatment with bevacizumab and had no gross residual disease. Bowel involvement was noted on 10 (13%) of the CT scans, and carcinomatosis was noted on radiologic imaging in 21/85 (25%) cases. In total, 56/160 (35%) regimens of bevacizumab would have been eliminated if bevacizumab was withheld from patients with either rectovaginal nodularity, bowel involvement on CT, or history of small bowel obstruction. Response rates were not evaluable for 16/160 (10%) because the subjects received less than 2 cycles. The response rate for all the evaluable regimens was 47% (67/144); there were 32 (22%) complete responses and 35 partial responses (24%). An additional 32% (46/144) had stable disease, and 22% (31/144) had progressive disease. There were 7 complete responses and 9 partial responses in patients with rectovaginal nodularity, for a response rate of 52% (16/31). Six patients who had rectovaginal nodularity were not evaluable for response, as they received less than 2 cycles of treatment. There was a 42% (8/19) response rate for patients with history of small bowel obstruction prior to the bevacizumab regimen, 2 complete responses and 6 partial responses. Three patients were not evaluable for response. The response rate was 50% (4/8) for patients who had bowel involvement on CT prior to treatment with bevacizumab, while 2 patients were not evaluable for response. Ten patients (9%) were diagnosed with bowel perforations, and 2 patients (1.8%) were diagnosed with fistulas (1 enterocutaneous, 1 fistula in ano). Considering all regimens of bevacizumab, the bowel perforation rate was 6% (10/160). Bowel perforations were diagnosed Table 2 Cytotoxic drugs combined with bevacizumab in this study. The most common combination was weekly paclitaxel and bevacizumab. Bevacizumab combinations (N = 160)
Number (%)
Weekly paclitaxel Gemcitabine, cisplatin Gemcitabine, carboplatin Gemcitabine Topotecan Cyclophosphamide Liposomal doxorubicin Docetaxel Carboplatin, paclitaxel Single agent bevacizumab Carboplatin, docetaxel Carboplatin Other
55 (34) 27 (17) 23 (14) 10 (6) 11 (7) 9 (6) 6 (4) 5 (3) 4 (3) 4 (3) 2 (1) 2 (1) 2 (1)
49
2 to 25 days after treatment with bevacizumab. Bowel perforations occurred after 0.5 to 5 cycles of the patient's current bevacizumab regimen; however, 3 patients had received 6, 10, and 15 previous cycles of bevacizumab during different bevacizumab regimens, respectively. The 30-day mortality following perforation was 50%, with 30% of patients dying within 7 days. Four patients were managed surgically, and 75% of them survived N30 days. Four of the patients opted for best supportive care given their disease status, previous treatment history, and condition at time of presentation, and all succumbed within 30 days. The 2 patients who were managed conservatively with intravenous antibiotics and percutaneous drain placement survived. However, one of these patients developed an enterocutaneous fistula 9 months after bowel perforation at the site of the perforation and required a right hemicolectomy. The other patient developed a large bowel obstruction in the region of previous perforation 4 months after perforation, requiring a sigmoid colectomy with side to side reanastomosis (Table 3). Both patients had biopsy proven recurrence. Patients with rectovaginal nodularity were more likely to develop a bowel perforation or fistula than those who did not have this finding, OR = 3.64 (95% CI = 1.1 to 12.1, p = 0.04). None of the other variables were significantly associated with bowel perforations or fistula formation (Table 4). None of the patients who had bowel perforation had bowel involvement on CT scan, so this was dropped from the model. Nineteen out of the 37 (51%) patients with rectovaginal nodularity had a CT scan prior to treatment with bevacizumab. Only 1 of these patients was noted to have rectal involvement on CT scan. No multivariate analysis was performed because the only variable which met inclusion criteria was rectovaginal nodularity. Rates of bowel perforation for different cytotoxic combinations were as follows: carboplatin, docetaxel, bevacizumab 50% (1/2), docetaxel, bevacizumab 20% (1/5), topotecan, bevacizumab 18% (2/11), gemcitabine, carboplatin, bevacizumab 3/23 (13%), paclitaxel, bevacizumab 5% (3/55). Discussion Bevacizumab is used to treat a variety of solid tumors, both as a single agent and in combination with cytotoxic chemotherapy. The rate of bowel perforation for patients treated with bevacizumab varies with primary disease site. Bowel perforations have been reported to occur in 1–2% of metastatic colorectal cancer patients treated with bevacizumab [14-16]. It has been recognized that ovarian cancer patients have a higher risk of bevacizumab associated bowel perforation than colorectal patients, with studies reporting rates of 0–15% [2-10]. The bowel complication rate of 9% in this study is consistent with these previous reports. Interestingly, pancreatic cancer patients also have a higher rate of bowel perforation (5–6%) than colorectal patients [17,18]. Badgwell et al. [19] reported on bowel complication rates (bowel perforation and fistula formation) for over 1400 patients treated with bevacizumab. The overall rate of perforation was 1.4% but varied by disease site as follows: colorectal 1.3%, renal cell 1.5%, lung 1.5%, unknown primary 1.7%, pancreatic 5%, and gastroesophageal 5.3%. Ovarian cancer patients had the highest perforation rate (6%) in their series. While the bowel perforation rate is similar in colorectal patients whether bevacizumab is used in the up front or recurrent setting (b2% versus 1.1%), this may not be the case in ovarian cancer, as Micha et al. [20] had no bowel perforations in 20 patients treated with carboplatin, paclitaxel, and bevacizumab as first line treatment. Ongoing studies such as GOG 218 (carboplatin, paclitaxel, with or without bevacizumab-NCT00483782) will help to better clarify the risk of bowel perforation in previously untreated ovarian cancer patients receiving bevacizumab [21]. The reason for this higher bowel perforation rate in ovarian cancer patients is not well understood. Cannistra et al. [3] noted that patients
50
D.L. Richardson et al. / Gynecologic Oncology 118 (2010) 47–51
Table 3 Characteristics of patients who experienced bowel perforation while receiving treatment with bevacizumab. Patient
Cytotoxic agent combined with bevacizumab
Number of previous regimens
Number of previous cycles of bevacizumab
Number of cycles completed on current regimen
Management
Died within 30 days of bowel perforation
1 2 3 4 5 6 7 8 9 10
Docetaxel Gemcitabine, carboplatin Paclitaxel Topotecan Gemcitabine, carboplatin Paclitaxel Gemcitabine, carboplatin Topotecan Carboplatin, docetaxel Paclitaxel
3 1 5 3 1 3 2 6 1 2
0 0 15 10 0 6 0 0 0 0
3 2 1 0.5 2 0.5 5 2 2 2
Surgical Surgical Surgical Supportive care Conservative Surgical Conservative Supportive care Supportive care Supportive care
Yes No No Yes No No No Yes Yes Yes
treated with 3 prior chemotherapy regimens compared to 2 was significantly associated with bowel perforation, and other authors have suggested that the risk of bowel perforation increases in heavily pre-treated patients [10]. However, perforations have been noted in patients receiving less than 3 prior regimens [8,9]. In the current study, we found no association between number of previous regimens and risk of bowel perforation. Cannistra et al. [3] also found a trend toward increased risk of bowel perforation with bowel wall thickening or bowel obstruction on CT scan (p b 0.1). Simpkins et al. [11] suggested that bowel perforation can be avoided by restricting bevacizumab to patients without bowel obstruction, bowel involvement on CT scan, and no evidence of rectosigmoid involvement on pelvic exam. However, we previously reported on 2 patients treated with combination gemcitabine, platinum, and bevacizumab with bowel perforation who had none of these potential risk factors [9]. In the current study, rectovaginal nodularity was the only variable associated with bowel perforation or fistula formation. Of note, rectovaginal nodularity is routinely assessed in patients with recurrent epithelial ovarian cancer receiving treatment at our institution. The 30-day mortality rate from bowel perforation in this study was 50%. This high mortality rate is partially attributable to the decision for best supportive care by 40% of the patients. Previous reports demonstrate a 30-day mortality rate of 0–100% [3,6,8-10]. This large range is probably due to a variety of factors, including patient characteristics, condition of patient at diagnosis of bowel perforation, and response to chemotherapy. There are several limitations to this study. It is retrospective and therefore subject to the biases associated with retrospective studies. Table 4 Results of univariate logistic regression analyzing the association of possible risk factors with bowel perforations and fistula formation in patients with recurrent EOC treated with bevacizumab. No multivariate analysis was done as only one factor (rectovaginal nodularity) reached statistical significance. Variable
Numbera
OR
95% CI
p value
Previous bowel surgery (yes or no) Number of previous regimens (N2 or ≤2) Number of previous regimens (N4 or ≤4) Previous bevacizumab (yes or no) Platinum resistant at time of treatment (yes or no) Total number cycles bevacizumab Total dose of bevacizumab (mg) Bevacizumab treatment dose (15 vs. ≤10 mg/kg) Body mass index (kg/m2) Age (years) History of small bowel obstruction (yes or no) Carcinomatosis (yes or no) Rectovaginal nodularity (yes or no)
44/159 108 59 47 103
1.35 0.66 0.85 1.22 0.95
0.39–4.7 0.2–2.2 0.25–3 0.35–4.3 0.27–3.39
0.64 0.51 0.8 0.76 0.93
6b 8313b 39
0.94 1 0.61
0.83–1.06 1–1 0.13–2.9
0.25 0.82 0.51
26.6b 60b 22
0.99 0.97 1.28
0.9–1.08 0.91–1.03 0.26–6.27
0.78 0.35 0.76
21/90 37/155
1.37 3.64
0.25–7.6 1.1–12.1
0.73 0.04
a b
Denominator equals 160 except where listed. Median—the variable entered into the model as a continuous variable.
To reduce the risk of selection bias, all patients who received bevacizumab for recurrent ovarian cancer were included. Because of the relatively few events, this study is underpowered to detect more than 1 risk factor, so it is possible that the failure to detect other risk factors is a type II error. In conclusion, rectovaginal nodularity is a risk factor for bowel perforation and fistula formation in patients with recurrent ovarian cancer. Because of the high 30-day mortality rate associated with bowel perforation in this population, we recommend using bevacizumab in patients with rectovaginal nodularity only after exhausting other treatment options or with comprehensive counseling regarding the risks and benefits of bevacizumab containing regimens.
Conflict of interest statement Dr. David O'Malley and Dr. Larry Copeland have received research support from Genentech. Some of the patients included in this study are part of an ongoing prospective trial of gemcitabine, carboplatin, and bevacizumab, supported by Lilly and Genentech.
References [1] Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, et al. Cancer Statistics, 2008. CA Cancer J Clin 2008;58:71–96. [2] Burger RA, Sill MW, Monk BJ, Greer BE, Sorosky JI. Phase II trial of bevacizumab in persistent or recurrent epithelial ovarian cancer or primary peritoneal cancer: a Gynecologic Oncology Group Study. J Clin Oncol 2007;25:5165–71. [3] Cannistra SA, Matulonis UA, Penson RT, Hambleton J, Dupont J, Mackey H, et al. Phase II study of bevacizumab in patients with platinum-resistant ovarian cancer or peritoneal serous cancer. J Clin Oncol 2007;25:5180–6. [4] Monk BJ, Han E, Josephs-Cowan CA, Pugmire G, Burger RA. Salvage bevacizumab (rhuMAB VEGF)-based therapy after multiple prior cytotoxic regimens in advanced refractory epithelial ovarian cancer. Gynecol Oncol 2006;102:140–4. [5] Wright JD, Secord AA, Numnum TM, Rocconi RP, Powell MA, Berchuck A, et al. A multi-institutional evaluation of factors predictive of toxicity and efficacy of bevacizumab for recurrent ovarian cancer. Int J Gynecol Cancer Jul 2007;21:1–7. [6] Nimeiri HS, Oza AM, Morgan RJ, Friberg G, Kasza K, Faoro L, et al. Efficacy and safety of bevacizumab plus erlotinib for patients with recurrent ovarian, primary peritoneal, and fallopian tube cancer: a trial of the Chicago, PMH, and California Phase II consortia. Gynecol Oncol 2008;110:49–55. [7] Chura JC, Van Iseghem K, Downs LS, Carson LF, Judson PL. Bevacizumab plus cyclophosphamide in heavily pretreated patients with recurrent ovarian cancer. Gynecol Oncol 2007;107:326–30. [8] Garcia AA, Hirte H, Fleming G, Yang D, Tsao-Wei DD, Roman L, et al. Phase II clinical trial of bevacizumab and low-dose metronomic oral cyclophosphamide in recurrent ovarian cancer: a trial of the California, Chicago, and Princess Margaret Hospital Phase II consortia. J Clin Oncol 2008;26:76–82. [9] Richardson DL, Backes FJ, Seamon LG, Zanagnolo V, O'Malley DM, Cohn DE, et al. Combination gemcitabine, platinum, and bevacizumab for the treatment of recurrent ovarian cancer. Gynecol Oncol 2008;111:461–6. [10] Wright JD, Hagemann A, Rader JS, Viviano D, Gibb RK, Norris L, et al. Bevacizumab combination therapy in recurrent, platinum-refractory, epithelial ovarian carcinoma. Cancer 2006;107:83–9. [11] Simpkins F, Belinson JL, Rose PG. Avoiding bevacizumab related gastrointestinal toxicity for recurrent ovarian cancer by careful patient screening. Gynecol Oncol 2007;107:118–23. [12] Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000;92:205–16. [13] Guppy AE, Rustin GJ. CA125 response: can it replace the traditional response criteria in ovarian cancer? Oncologist 2002;7:437–43.
D.L. Richardson et al. / Gynecologic Oncology 118 (2010) 47–51 [14] Giantonio BJ, Catalano PJ, Meropol NJ, O'Dwyer PJ, Mitchell EP, Alberts SR, et al. Bevacizumab in combination with oxaliplatin, fluorouracil, and leucovorin (FOLFOX4) for previously treated metastatic colorectal cancer: results from the Eastern Cooperative Oncology Group study E3200. J Clin Oncol 2007;25:1539–44. [15] Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004;350:2335–42. [16] Kabbinavar FF, Schulz J, McCleod M, Patel T, Hamm JT, Hecht R, et al. Addition of bevacizumab to bolus fluorouracil and leucovorin in first-line metastatic colorectal cancer: results of a randomized phase II trial. J Clin Oncol 2005;23: 3697–705. [17] Kindler HL, Friberg G, Singh DA, et al. Phase II trial of bevacizumab plus gemcitabine in patients with advanced pancreatic cancer. J Clin Oncol 2005;23: 8033–40.
51
[18] Ko AH, Dito E, Schillinger B, et al. A phase II study evaluating bevacizumab in combination with fixed-dose rate gemcitabine and low-dose cisplatin for metastatic pancreatic cancer: is an anti-VEGF strategy still applicable? Invest New Drugs 2008;26:463–71. [19] Badgwell BD, Camp ER, Feig B, et al. Management of bevacizumab-associated bowel perforation: a case series and review of the literature. Ann Oncol 2008;19: 577–82. [20] Micha JP, Goldstein BH, Rettenmaier MA, et al. A phase II study of outpatient firstline paclitaxel, carboplatin, and bevacizumab for advanced-stage epithelial ovarian, peritoneal, and fallopian tube cancer. Int J Gynecol Cancer 2007;17: 771–6. [21] U.S. National Institutes of Health. Clinical trials information. Available from URL http://clinicaltrials.gov/ct2/show/NCT00483782?term = ovarian+cancer +AND+bevacizumab and rank = 11 [accessed June 5, 2009].
Contents lists available at ScienceDirect
Gynecologic Oncology j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y g y n o
Which factors predict bowel complications in patients with recurrent epithelial ovarian cancer being treated with bevacizumab? D.L. Richardson, F.J. Backes, J.D. Hurt, L.G. Seamon, L.J. Copeland, J.M. Fowler, D.E. Cohn, D.M. O'Malley ⁎ Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, The Ohio State University College of Medicine, M-210 Starling Loving Hall, 320 West 10th Avenue, Columbus, OH 43210, USA
a r t i c l e
i n f o
Article history: Received 9 September 2009 Available online 10 April 2010 Keywords: Bevacizumab Bowel perforation Fistula Risk factors Epithelial ovarian cancer
a b s t r a c t Background. Increased rates of bowel perforation in patients with recurrent epithelial ovarian cancer (EOC) treated with bevacizumab have been reported, but the risk factors for this association are uncertain. We sought to identify factors associated with bowel perforation and fistula formation in recurrent EOC patients treated with bevacizumab. Methods. A chart review of all patients treated with bevacizumab for recurrent EOC at a single institution was performed. Pertinent patient characteristics and treatment information were collected. Univariate logistic regression was performed to analyze multiple variables. Results. One hundred twelve patients who were treated with 160 different bevacizumab regimens were identified. The median age was 60 years (range, 29–78 years). Patients had received a median of 4 prior chemotherapy regimens (range, 1–10). The median number of cycles was 4 (range, 0.5–31). Ten patients (9%) were diagnosed with bowel perforations, and another 2 patients (1.8%) were diagnosed with fistulas. The 30-day mortality following perforation was 50%, with 30% of patients dying within 1 week. Patients with rectovaginal nodularity were more likely to develop a bowel perforation or fistula than those who did not have this finding, OR = 3.64 (95% CI = 1.1 to 12.1, p = 0.04). None of the other variables were significantly associated with bowel perforations or fistula formation. Conclusions. Rectovaginal nodularity is associated with an increased risk of bowel perforation or fistula formation for patients with recurrent EOC treated with bevacizumab. Careful consideration should be given prior to initiating bevacizumab treatment in EOC patients with rectovaginal nodularity since the mortality rate with bevacizumab associated bowel perforations is 50%. © 2010 Elsevier Inc. All rights reserved.
Introduction Ovarian cancer is the leading cause of gynecologic cancer deaths in the United States [1]. While the majority of patients respond to front line chemotherapy with platinum and paclitaxel, most advanced stage patients will recur. Although there are many cytotoxic agents with activity in the recurrent setting, ultimately the disease becomes chemotherapy resistant. Therefore, drugs with novel mechanisms of action designed to target specific pathways, for example angiogenesis, are being evaluated for efficacy and safety in both the front line and recurrent setting. Bevacizumab (Genentech, San Francisco, CA, USA), a monoclonal antibody against vascular endothelial growth factor, is increasingly being utilized for the treatment of recurrent ovarian cancer. It has modest activity as a single agent, with reported response rates ranging from 10% to 21% and median progression-free survival (PFS)
⁎ Corresponding author. Fax: +1 614 366 7942. E-mail address: David.O'[email protected] (D.M. O'Malley). 0090-8258/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2010.01.011
of 4–5 months [2-5]. Combinations of bevacizumab and other novel compounds have also been studied. A phase II trial of bevacizumab combined with erlotinib (an EGFR tyrosine kinase inhibitor) demonstrated no improvement in response rates to single agent bevacizumab in the first phase of accrual with a concerning rate of bowel perforation. Therefore, enrollment was halted after 13 patients were treated with this combination [6]. In contrast, response rates of 24– 78% have been reported when bevacizumab has been combined with cytotoxic chemotherapy [7-10]. The large range of response rates may be attributable to different patient populations (platinum sensitivity and number of previous regimens), different cytotoxic drugs used, and different definitions for response (measurable disease versus CA125 levels). Serious bowel complications, including bowel perforations and fistulae, have been reported more commonly in ovarian cancer patients treated with bevacizumab than in most other types of cancer patients treated with this drug. Bowel perforation and fistula have been reported to occur in 0–11% of patients treated with single agent bevacizumab for recurrent ovarian cancer and in 0–9% of patients treated with bevacizumab in combination with cytotoxic
48
D.L. Richardson et al. / Gynecologic Oncology 118 (2010) 47–51
chemotherapy [2-4, 7-10]. The highest bowel perforation rate, 15%, was reported in recurrent ovarian cancer patients treated with combination bevacizumab and erlotinib who had previously received 2 or less prior regimens [6]. The risk factors for bowel perforation and fistula formation in recurrent ovarian cancer patients treated with bevacizumab are unclear. Possible risk factors include number of previous chemotherapy regimens, bowel wall thickening on CT, and bowel obstruction [3,10]. It is also unknown whether careful patient selection can avoid this complication. Although Simpkins et al. [11] suggested that bowel perforation can be avoided by restricting bevacizumab to patients without bowel obstruction, bowel involvement on CT scan, and no evidence of rectosigmoid involvement on pelvic exam, this has not been prospectively evaluated. In the present study, we sought to identify the risk factors for increased risk of gastrointestinal complications (bowel perforation and fistula formation) in consecutive patients with recurrent ovarian cancer treated with bevacizumab. Our secondary aim was to report the associated 30-day mortality rate of bowel perforation in this group of patients. Materials and methods A retrospective chart review was performed of all patients who were treated with bevacizumab with or without cytotoxic chemotherapy for recurrent epithelial ovarian, primary peritoneal, or fallopian tube cancer at a single institution from April 2005 (the first time a patient was treated with bevacizumab for recurrent ovarian cancer at our institution) through August 2008. Institutional review board approval was obtained. Patients were identified by reviewing chemotherapy log books. Patients who received at least one dose of bevacizumab were included. Multiple patients received more than one regimen of bevacizumab. Data from all regimens were abstracted, not only from the first regimen, as the patient remained at risk for bowel complications. Charts were abstracted for pertinent data, including patient characteristics, treatment, response, gastrointestinal complications (bowel perforation, fistula), and management and outcome of gastrointestinal complications. Bowel perforation was diagnosed by either free air on an abdominal radiograph or CT scan, evidence of contrast extravasation on a CT scan, or bowel contents in the abdomen on surgical exploration. Fistula was defined as either clinical or radiographic evidence of an abnormal communication between two epithelium-lined organs. Response rates were calculated for patients who received at least 2 cycles of bevacizumab, with or without cytotoxic chemotherapy. Measurable disease was defined as at least one lesion that was greater than 1 cm on spiral CT, as defined by Response to Treatment in Solid Tumors (RECIST) criteria [12]. In the absence of measurable disease, CA-125 levels were used to assess response per modified Rustin criteria [13]. Complete response was disappearance of gross evidence of disease, resolution of measurable disease on CT scan, or a normalization of CA-125 levels from an elevated level. Partial response was a 30% or greater reduction in measurement of longest dimension of each lesion or N50% decrease in CA-125 levels. Progressive disease was a 20% or greater increase in measurement of longest dimension of each lesion or the appearance of any new lesion or a doubling of CA-125 levels within 8 weeks of the start of chemotherapy. Stable disease was any condition which did not meet the above criteria. Best response was reported. CA-125 levels were routinely drawn with the standard pre-chemotherapy labs. CT scans were obtained as appropriate to follow measurable disease and at the physician's discretion. Simple descriptive statistics were used to describe response rates, including mean, median, and standard deviation. Logistic regression was used to analyze the risk of gastrointestinal complications for multiple variables p b 0.15 from the univariate analysis. Continuous
variables included total dose of bevacizumab (mg), total number of cycles of bevacizumab, age, and body mass index. The number of previous chemotherapy regimens was dichotomized at the median (N4 vs. ≤4), and also at 2 (N2 vs. ≤2) based on previous literature [3]. Binary variables included history of bowel surgery, previous bevacizumab, platinum resistance at time of treatment with bevacizumab, carcinomatosis, history of small bowel obstruction, rectovaginal nodularity, and bevacizumab treatment dose (15 vs. 10 mg/kg). Because of the number of different cytotoxics combined with bevacizumab and the small numbers of patients treated with some of these combinations, the type of cytotoxic combined with bevacizumab was not analyzed as a risk factor for perforation. The final model contained only variables which were significant at the p b 0.05 level. Stata release 10 (College Station, Texas) was used to perform statistical calculations. Results One hundred and twelve patients who were treated with 160 bevacizumab regimens were identified, and all patients were included in the analysis. Thirty-one percent (35/112) of the patients received more than one bevacizumab regimen. The median age was 60 years (range, 29–78 years), and the majority of patients were Caucasian. Sixty-four percent (103/160) were platinum resistant (primary or secondary) at the time of treatment. The median body mass index at the start of treatment was 26.6 kg/m2 (range, 14.9–45.6 kg/m2). The Table 1 Baseline demographics of the study population. Baseline demographics (N = 112) Median age (range), years Mean age (SD) Race Caucasian African-American Stage at diagnosis I II III IV Not staged Histology Serous Endometrioid Mucinous Mixed epithelial Clear cell Transitional Adenocarcinoma NOSb Unknown Grade at diagnosis 1 2 3 Unknown Neoadjuvant chemotherapy Residual disease at primary surgery No gross residual Optimal (b1 cm) Suboptimal (N 1 cm) No surgery Unknown Platinum response Sensitive Resistant/refractory Secondary resistant Median number of previous chemotherapeutic regimens Received more than one regimen of bevacizumab a b c d
Percentages may not add up to 100% due to rounding. NOS, not otherwise specified. Includes 5 patients who received neoadjuvant chemotherapy. Includes 2 patients who received neoadjuvant chemotherapy.
Number (%)a 60 (29–78) 59 (9.5) 107 (96) 5 (4) 4 (4) 7 (6) 89 (79) 11 (10) 1 (1) 83 (74) 6 (5) 5 (4) 5 (4) 2 (2) 1 (1) 7 (6) 3 (3) 4 (4) 14 (13) 85 (76) 10 (9) 36c (32) 53d (47) 17 (15) 1 (1) 5 (4) 75 (67) 37 (33) 30 4 (1–10) 35 (31)
D.L. Richardson et al. / Gynecologic Oncology 118 (2010) 47–51
patients were relatively heavily pre-treated, with a median of 4 prior chemotherapy regimens (range, 1–10) (Table 1). The most common regimen was weekly paclitaxel 60–70 mg/m2 and bevacizumab day 1 and day 15 every 28 days, with 55 patients receiving this regimen. Please refer to Table 2 for a listing of all the bevacizumab combinations. The dose of bevacizumab ranged from 5 to 15 mg/kg in all regimens. A quarter of the regimens (39/160) used bevacizumab 15 mg/kg. The median number of cycles of each regimen was 4 (range, 0.5–31). The median number of cycles of all regimens containing bevacizumab was 6 (range, 0.5–36). The median total dose of bevacizumab was 8313 mg (range, 430–54653 mg). Twenty-one patients (13%) were still undergoing treatment with bevacizumab at the completion of this study. Eighteen patients (16%) had a history of a small bowel obstruction prior to treatment with bevacizumab. Four of these patients were treated with 2 bevacizumab regimens. Twenty-nine percent (31/107) of the patients had rectovaginal nodularity prior to initiation of bevacizumab, 5 of whom were treated with more than one bevacizumab regimen . Rectovaginal exam was not performed or documented in 5 patients. CT scans were obtained about 50% (78/ 160) of the time prior to starting bevacizumab. Five patients had PET scans, and 2 patients had MRIs performed prior to starting treatment. In addition, 2 patients had undergone secondary cytoreduction prior to starting treatment with bevacizumab and had no gross residual disease. Bowel involvement was noted on 10 (13%) of the CT scans, and carcinomatosis was noted on radiologic imaging in 21/85 (25%) cases. In total, 56/160 (35%) regimens of bevacizumab would have been eliminated if bevacizumab was withheld from patients with either rectovaginal nodularity, bowel involvement on CT, or history of small bowel obstruction. Response rates were not evaluable for 16/160 (10%) because the subjects received less than 2 cycles. The response rate for all the evaluable regimens was 47% (67/144); there were 32 (22%) complete responses and 35 partial responses (24%). An additional 32% (46/144) had stable disease, and 22% (31/144) had progressive disease. There were 7 complete responses and 9 partial responses in patients with rectovaginal nodularity, for a response rate of 52% (16/31). Six patients who had rectovaginal nodularity were not evaluable for response, as they received less than 2 cycles of treatment. There was a 42% (8/19) response rate for patients with history of small bowel obstruction prior to the bevacizumab regimen, 2 complete responses and 6 partial responses. Three patients were not evaluable for response. The response rate was 50% (4/8) for patients who had bowel involvement on CT prior to treatment with bevacizumab, while 2 patients were not evaluable for response. Ten patients (9%) were diagnosed with bowel perforations, and 2 patients (1.8%) were diagnosed with fistulas (1 enterocutaneous, 1 fistula in ano). Considering all regimens of bevacizumab, the bowel perforation rate was 6% (10/160). Bowel perforations were diagnosed Table 2 Cytotoxic drugs combined with bevacizumab in this study. The most common combination was weekly paclitaxel and bevacizumab. Bevacizumab combinations (N = 160)
Number (%)
Weekly paclitaxel Gemcitabine, cisplatin Gemcitabine, carboplatin Gemcitabine Topotecan Cyclophosphamide Liposomal doxorubicin Docetaxel Carboplatin, paclitaxel Single agent bevacizumab Carboplatin, docetaxel Carboplatin Other
55 (34) 27 (17) 23 (14) 10 (6) 11 (7) 9 (6) 6 (4) 5 (3) 4 (3) 4 (3) 2 (1) 2 (1) 2 (1)
49
2 to 25 days after treatment with bevacizumab. Bowel perforations occurred after 0.5 to 5 cycles of the patient's current bevacizumab regimen; however, 3 patients had received 6, 10, and 15 previous cycles of bevacizumab during different bevacizumab regimens, respectively. The 30-day mortality following perforation was 50%, with 30% of patients dying within 7 days. Four patients were managed surgically, and 75% of them survived N30 days. Four of the patients opted for best supportive care given their disease status, previous treatment history, and condition at time of presentation, and all succumbed within 30 days. The 2 patients who were managed conservatively with intravenous antibiotics and percutaneous drain placement survived. However, one of these patients developed an enterocutaneous fistula 9 months after bowel perforation at the site of the perforation and required a right hemicolectomy. The other patient developed a large bowel obstruction in the region of previous perforation 4 months after perforation, requiring a sigmoid colectomy with side to side reanastomosis (Table 3). Both patients had biopsy proven recurrence. Patients with rectovaginal nodularity were more likely to develop a bowel perforation or fistula than those who did not have this finding, OR = 3.64 (95% CI = 1.1 to 12.1, p = 0.04). None of the other variables were significantly associated with bowel perforations or fistula formation (Table 4). None of the patients who had bowel perforation had bowel involvement on CT scan, so this was dropped from the model. Nineteen out of the 37 (51%) patients with rectovaginal nodularity had a CT scan prior to treatment with bevacizumab. Only 1 of these patients was noted to have rectal involvement on CT scan. No multivariate analysis was performed because the only variable which met inclusion criteria was rectovaginal nodularity. Rates of bowel perforation for different cytotoxic combinations were as follows: carboplatin, docetaxel, bevacizumab 50% (1/2), docetaxel, bevacizumab 20% (1/5), topotecan, bevacizumab 18% (2/11), gemcitabine, carboplatin, bevacizumab 3/23 (13%), paclitaxel, bevacizumab 5% (3/55). Discussion Bevacizumab is used to treat a variety of solid tumors, both as a single agent and in combination with cytotoxic chemotherapy. The rate of bowel perforation for patients treated with bevacizumab varies with primary disease site. Bowel perforations have been reported to occur in 1–2% of metastatic colorectal cancer patients treated with bevacizumab [14-16]. It has been recognized that ovarian cancer patients have a higher risk of bevacizumab associated bowel perforation than colorectal patients, with studies reporting rates of 0–15% [2-10]. The bowel complication rate of 9% in this study is consistent with these previous reports. Interestingly, pancreatic cancer patients also have a higher rate of bowel perforation (5–6%) than colorectal patients [17,18]. Badgwell et al. [19] reported on bowel complication rates (bowel perforation and fistula formation) for over 1400 patients treated with bevacizumab. The overall rate of perforation was 1.4% but varied by disease site as follows: colorectal 1.3%, renal cell 1.5%, lung 1.5%, unknown primary 1.7%, pancreatic 5%, and gastroesophageal 5.3%. Ovarian cancer patients had the highest perforation rate (6%) in their series. While the bowel perforation rate is similar in colorectal patients whether bevacizumab is used in the up front or recurrent setting (b2% versus 1.1%), this may not be the case in ovarian cancer, as Micha et al. [20] had no bowel perforations in 20 patients treated with carboplatin, paclitaxel, and bevacizumab as first line treatment. Ongoing studies such as GOG 218 (carboplatin, paclitaxel, with or without bevacizumab-NCT00483782) will help to better clarify the risk of bowel perforation in previously untreated ovarian cancer patients receiving bevacizumab [21]. The reason for this higher bowel perforation rate in ovarian cancer patients is not well understood. Cannistra et al. [3] noted that patients
50
D.L. Richardson et al. / Gynecologic Oncology 118 (2010) 47–51
Table 3 Characteristics of patients who experienced bowel perforation while receiving treatment with bevacizumab. Patient
Cytotoxic agent combined with bevacizumab
Number of previous regimens
Number of previous cycles of bevacizumab
Number of cycles completed on current regimen
Management
Died within 30 days of bowel perforation
1 2 3 4 5 6 7 8 9 10
Docetaxel Gemcitabine, carboplatin Paclitaxel Topotecan Gemcitabine, carboplatin Paclitaxel Gemcitabine, carboplatin Topotecan Carboplatin, docetaxel Paclitaxel
3 1 5 3 1 3 2 6 1 2
0 0 15 10 0 6 0 0 0 0
3 2 1 0.5 2 0.5 5 2 2 2
Surgical Surgical Surgical Supportive care Conservative Surgical Conservative Supportive care Supportive care Supportive care
Yes No No Yes No No No Yes Yes Yes
treated with 3 prior chemotherapy regimens compared to 2 was significantly associated with bowel perforation, and other authors have suggested that the risk of bowel perforation increases in heavily pre-treated patients [10]. However, perforations have been noted in patients receiving less than 3 prior regimens [8,9]. In the current study, we found no association between number of previous regimens and risk of bowel perforation. Cannistra et al. [3] also found a trend toward increased risk of bowel perforation with bowel wall thickening or bowel obstruction on CT scan (p b 0.1). Simpkins et al. [11] suggested that bowel perforation can be avoided by restricting bevacizumab to patients without bowel obstruction, bowel involvement on CT scan, and no evidence of rectosigmoid involvement on pelvic exam. However, we previously reported on 2 patients treated with combination gemcitabine, platinum, and bevacizumab with bowel perforation who had none of these potential risk factors [9]. In the current study, rectovaginal nodularity was the only variable associated with bowel perforation or fistula formation. Of note, rectovaginal nodularity is routinely assessed in patients with recurrent epithelial ovarian cancer receiving treatment at our institution. The 30-day mortality rate from bowel perforation in this study was 50%. This high mortality rate is partially attributable to the decision for best supportive care by 40% of the patients. Previous reports demonstrate a 30-day mortality rate of 0–100% [3,6,8-10]. This large range is probably due to a variety of factors, including patient characteristics, condition of patient at diagnosis of bowel perforation, and response to chemotherapy. There are several limitations to this study. It is retrospective and therefore subject to the biases associated with retrospective studies. Table 4 Results of univariate logistic regression analyzing the association of possible risk factors with bowel perforations and fistula formation in patients with recurrent EOC treated with bevacizumab. No multivariate analysis was done as only one factor (rectovaginal nodularity) reached statistical significance. Variable
Numbera
OR
95% CI
p value
Previous bowel surgery (yes or no) Number of previous regimens (N2 or ≤2) Number of previous regimens (N4 or ≤4) Previous bevacizumab (yes or no) Platinum resistant at time of treatment (yes or no) Total number cycles bevacizumab Total dose of bevacizumab (mg) Bevacizumab treatment dose (15 vs. ≤10 mg/kg) Body mass index (kg/m2) Age (years) History of small bowel obstruction (yes or no) Carcinomatosis (yes or no) Rectovaginal nodularity (yes or no)
44/159 108 59 47 103
1.35 0.66 0.85 1.22 0.95
0.39–4.7 0.2–2.2 0.25–3 0.35–4.3 0.27–3.39
0.64 0.51 0.8 0.76 0.93
6b 8313b 39
0.94 1 0.61
0.83–1.06 1–1 0.13–2.9
0.25 0.82 0.51
26.6b 60b 22
0.99 0.97 1.28
0.9–1.08 0.91–1.03 0.26–6.27
0.78 0.35 0.76
21/90 37/155
1.37 3.64
0.25–7.6 1.1–12.1
0.73 0.04
a b
Denominator equals 160 except where listed. Median—the variable entered into the model as a continuous variable.
To reduce the risk of selection bias, all patients who received bevacizumab for recurrent ovarian cancer were included. Because of the relatively few events, this study is underpowered to detect more than 1 risk factor, so it is possible that the failure to detect other risk factors is a type II error. In conclusion, rectovaginal nodularity is a risk factor for bowel perforation and fistula formation in patients with recurrent ovarian cancer. Because of the high 30-day mortality rate associated with bowel perforation in this population, we recommend using bevacizumab in patients with rectovaginal nodularity only after exhausting other treatment options or with comprehensive counseling regarding the risks and benefits of bevacizumab containing regimens.
Conflict of interest statement Dr. David O'Malley and Dr. Larry Copeland have received research support from Genentech. Some of the patients included in this study are part of an ongoing prospective trial of gemcitabine, carboplatin, and bevacizumab, supported by Lilly and Genentech.
References [1] Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, et al. Cancer Statistics, 2008. CA Cancer J Clin 2008;58:71–96. [2] Burger RA, Sill MW, Monk BJ, Greer BE, Sorosky JI. Phase II trial of bevacizumab in persistent or recurrent epithelial ovarian cancer or primary peritoneal cancer: a Gynecologic Oncology Group Study. J Clin Oncol 2007;25:5165–71. [3] Cannistra SA, Matulonis UA, Penson RT, Hambleton J, Dupont J, Mackey H, et al. Phase II study of bevacizumab in patients with platinum-resistant ovarian cancer or peritoneal serous cancer. J Clin Oncol 2007;25:5180–6. [4] Monk BJ, Han E, Josephs-Cowan CA, Pugmire G, Burger RA. Salvage bevacizumab (rhuMAB VEGF)-based therapy after multiple prior cytotoxic regimens in advanced refractory epithelial ovarian cancer. Gynecol Oncol 2006;102:140–4. [5] Wright JD, Secord AA, Numnum TM, Rocconi RP, Powell MA, Berchuck A, et al. A multi-institutional evaluation of factors predictive of toxicity and efficacy of bevacizumab for recurrent ovarian cancer. Int J Gynecol Cancer Jul 2007;21:1–7. [6] Nimeiri HS, Oza AM, Morgan RJ, Friberg G, Kasza K, Faoro L, et al. Efficacy and safety of bevacizumab plus erlotinib for patients with recurrent ovarian, primary peritoneal, and fallopian tube cancer: a trial of the Chicago, PMH, and California Phase II consortia. Gynecol Oncol 2008;110:49–55. [7] Chura JC, Van Iseghem K, Downs LS, Carson LF, Judson PL. Bevacizumab plus cyclophosphamide in heavily pretreated patients with recurrent ovarian cancer. Gynecol Oncol 2007;107:326–30. [8] Garcia AA, Hirte H, Fleming G, Yang D, Tsao-Wei DD, Roman L, et al. Phase II clinical trial of bevacizumab and low-dose metronomic oral cyclophosphamide in recurrent ovarian cancer: a trial of the California, Chicago, and Princess Margaret Hospital Phase II consortia. J Clin Oncol 2008;26:76–82. [9] Richardson DL, Backes FJ, Seamon LG, Zanagnolo V, O'Malley DM, Cohn DE, et al. Combination gemcitabine, platinum, and bevacizumab for the treatment of recurrent ovarian cancer. Gynecol Oncol 2008;111:461–6. [10] Wright JD, Hagemann A, Rader JS, Viviano D, Gibb RK, Norris L, et al. Bevacizumab combination therapy in recurrent, platinum-refractory, epithelial ovarian carcinoma. Cancer 2006;107:83–9. [11] Simpkins F, Belinson JL, Rose PG. Avoiding bevacizumab related gastrointestinal toxicity for recurrent ovarian cancer by careful patient screening. Gynecol Oncol 2007;107:118–23. [12] Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000;92:205–16. [13] Guppy AE, Rustin GJ. CA125 response: can it replace the traditional response criteria in ovarian cancer? Oncologist 2002;7:437–43.
D.L. Richardson et al. / Gynecologic Oncology 118 (2010) 47–51 [14] Giantonio BJ, Catalano PJ, Meropol NJ, O'Dwyer PJ, Mitchell EP, Alberts SR, et al. Bevacizumab in combination with oxaliplatin, fluorouracil, and leucovorin (FOLFOX4) for previously treated metastatic colorectal cancer: results from the Eastern Cooperative Oncology Group study E3200. J Clin Oncol 2007;25:1539–44. [15] Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004;350:2335–42. [16] Kabbinavar FF, Schulz J, McCleod M, Patel T, Hamm JT, Hecht R, et al. Addition of bevacizumab to bolus fluorouracil and leucovorin in first-line metastatic colorectal cancer: results of a randomized phase II trial. J Clin Oncol 2005;23: 3697–705. [17] Kindler HL, Friberg G, Singh DA, et al. Phase II trial of bevacizumab plus gemcitabine in patients with advanced pancreatic cancer. J Clin Oncol 2005;23: 8033–40.
51
[18] Ko AH, Dito E, Schillinger B, et al. A phase II study evaluating bevacizumab in combination with fixed-dose rate gemcitabine and low-dose cisplatin for metastatic pancreatic cancer: is an anti-VEGF strategy still applicable? Invest New Drugs 2008;26:463–71. [19] Badgwell BD, Camp ER, Feig B, et al. Management of bevacizumab-associated bowel perforation: a case series and review of the literature. Ann Oncol 2008;19: 577–82. [20] Micha JP, Goldstein BH, Rettenmaier MA, et al. A phase II study of outpatient firstline paclitaxel, carboplatin, and bevacizumab for advanced-stage epithelial ovarian, peritoneal, and fallopian tube cancer. Int J Gynecol Cancer 2007;17: 771–6. [21] U.S. National Institutes of Health. Clinical trials information. Available from URL http://clinicaltrials.gov/ct2/show/NCT00483782?term = ovarian+cancer +AND+bevacizumab and rank = 11 [accessed June 5, 2009].