Ifosfamide, paclitaxel, and carboplatin, a novel triplet regimen for advanced, recurrent, or persistent carcinoma of the cervix: A phase II trial

Methods Eligible patients had histologically proven stage IVB, recurrent, or persistent carcinoma of the cervix not amenable to curative treatment with surgery and/or radiation therapy. Chemotherapy was given on day 1 of a 28-day cycle: mesna (600 mg/m 2 ) prior to ifosfamide (2 g/m 2 ), paclitaxel (175 mg/m 2 ), carboplatin (AUC 5). Response rates were determined according to RECIST criteria. Toxicity was graded according the National Cancer Institute's common toxicity criteria. Quality of life measurements were obtained using the FACT-Cx. Results Twenty-eight patients participated in this study, with 21 evaluable for response rate. Overall, 7 patients (33%) had a demonstrated objective response (4 complete responses, 3 partial responses). Stable disease was documented in 3 patients. The overall median survival for all patients was 10 months. Median progression free survival for evaluable patients was 5.0 months. Bone marrow suppression was the most common toxicity. There were no negative effects of this treatment regimen on quality of life assessments. Conclusion Ifosfamide, paclitaxel, and carboplatin is an effective regimen in treating advanced or recurrent carcinoma of the cervix and has an acceptable toxicity profile. Research Highlights ►Ifosfamide, paclitaxel, and carboplatin demonstrated a 33% response rate in patients with advanced or recurrent cervical cancer. ►Neutropenia (grade 3, 4) was the most common adverse event occurring in 72% of patients. ►The overall median survival for all patients enrolled was 10.0 months. Keywords Cervical cancer Carboplatin Paclitaxel Ifosfamide Phase II trial Introduction In the United States advanced, recurrent, or persistent carcinoma of the cervix will account for an estimated 4210 deaths in 2010 [1] . This translates into more than 11 deaths per day. Worldwide, the problem is more staggering, with cervical cancer being the second most common cause of death due to cancer in women [2] . In 2000, the worldwide estimate of deaths due to cervical cancer was 233,000 (over 630 deaths per day) [3] . Survival is directly related to the stage of disease at the time of diagnosis. In North America, approximately 40% of cases of cervical cancer are diagnosed when the disease has spread beyond the cervix [4] . The 5-year survival in this group of women is dramatically less than that of patients diagnosed with stage I disease. The estimated 5-year survival for stage III disease is 30–50%, and this decreases to 5–15% for stage IV disease [4] . For patients with recurrent or persistent disease, or distant metastasis upon presentation, systemic chemotherapy is an important treatment option. While cisplatin has been the cornerstone of systemic chemotherapy in cervical cancer, over 50 other agents have been studied [5] . Recent cooperative studies of the Gynecologic Oncology Group (GOG) and the Hellenic Cooperative Oncology Group (HeCOG) have demonstrated increased response rates, and in some cases improved survival, for combination regimens [6–8] . Specifically, the GOG reported that the combination of cisplatin and topotecan provided superior overall and progression free survival when compared to cisplatin alone for the treatment of recurrent or advanced cervical cancer [6] . Additionally, a randomized phase II study of cisplatin and ifosfamide with or without paclitaxel demonstrated increased activity with the triple drug combination [8] . As a salvage regimen for persistent or recurrent squamous cervical carcinoma, cisplatin, ifosfamide, and paclitaxel was previously reported to achieve an objective response rate of 67% (30 of 45 patients) [10] . Ninety-one percent of patients experienced grade 3–4 granulocytopenia. Grade 3–4 nausea-emesis was experienced by 67% of patients, and peripheral neuropathy occurred in 24% of patients. Thus, the combination regimen of cisplatin, ifosfamide, and paclitaxel has significant activity against cervical cancer. The regimen also has significant toxicity. The more recent HeCOG study reported a slightly lower but still robust response rate for the triple drug regimen of cisplatin, ifosfamide, and paclitaxel at 59%. Additionally, the triple drug arm had improved progression free (7.9 months versus 6.3 months, p = 0.023) and overall survival (15.4 months versus 13.2 months, p = 0.048) when compared to the cisplatin–ifosfamide doublet. As anticipated, the rate of neurotoxicity was significantly higher for the patients receiving the paclitaxel containing regimen, 43% versus 11% ( p < 0.001). The combination of cisplatin, ifosfamide, and paclitaxel has also been studied in squamous cell cancer of the head and neck. In the first phase II trial of this three drug regimen in head and neck cancer, an objective response rate of 58% (30 of 52 patients) was noted [11] . This was a two-fold increase over previously published response rates for alternate regimens in head and neck cancer. Neutropenia (grade 3–4) was the most common toxicity, experienced by 90% of patients. Moderate to severe peripheral neuropathy was cumulative, with three patients experiencing grade 3 neuropathy. To decrease the toxicity associated with cisplatin, ifosfamide, and paclitaxel, these same investigators then substituted cisplatin with carboplatin. Thus, the subsequent study was a triple drug combination of carboplatin, ifosfamide, and paclitaxel for squamous head and neck cancer. Of the 54 patients evaluable for tumor response, a 59% objective response rate was noted [12] . In this second study, the substitution of cisplatin with carboplatin did not appear to decrease the efficacy of the regimen. Regarding toxicity, replacing cisplatin with carboplatin resulted in less peripheral neuropathy. Patients also experienced less fatigue on this regimen. These studies served as the basis for investigating the combination regimen of carboplatin, ifosfamide, and paclitaxel in cervical cancer. Our present protocol was based upon the demonstrated activities of platinum, ifosfamide, and paclitaxel against cervical cancer [7–10] . Because this regimen had not been studied in patients with cervical cancer, a site specific phase I/II trial was undertaken. The phase I component was essential given that this regimen had not been studied in patients who had received pelvic irradiation, a factor that might increase the toxicity of this treatment. Phase I data revealed the maximum tolerated dose of this outpatient regimen to be: ifosfamide (2 g/m 2 ) given over 2 hours, paclitaxel (175 mg/m 2 ) given over 1 h, and carboplatin (AUC of 5) given over 45 min on day 1 of a 28-day cycle [13] . The objectives of this phase II study were: (1) to determine the response rate of advanced, recurrent, or persistent carcinoma of the cervix to ifosfamide, paclitaxel, and carboplatin; (2) to determine the progression free interval and survival rate in patients treated with this regimen; (3) to describe the toxicities associated with this regimen; and (4) to evaluate the quality of life of patients during therapy with a triple drug regimen. Materials and methods Patients were treated according to the University of Minnesota Women's Cancer Center Protocol 28R. The Institutional Review Board and Cancer Protocol Review Committee of the University of Minnesota reviewed and approved this protocol. All patients gave informed consent prior to enrollment. Eligible patients had histologically proven stage IVB, recurrent, or persistent carcinoma of the cervix not amenable to curative treatment with surgery and/or radiation therapy. Patients were required to have measurable disease in two dimensions by either physical examination, CT scan, or MRI. GOG performance status was 0, 1, or 2 for all participants. Patients had recovered from the effects of recent surgery, radiation, or chemoradiation. Patients were free of clinically significant infection. Patients were also required to have adequate hematologic (ANC ≥ 1500/μL, platelets ≥ 100,000/μL), renal (creatinine ≤ 1.5 mg/dL), and hepatic function (bilirubin ≤ 1.5× normal, SGOT and alkaline phosphatase ≤ 3×normal, albumin ≥ 3.0 g/dL). The only prior chemotherapy received was agents that served as radiation sensitizers. Patients were excluded from study participation if they had bilateral hydronephrosis not amenable to decompression by either ureteral stents or percutaneous drainage. Patients with craniospinal metastasis were excluded as were those with documented hypersensitivity to paclitaxel, ifosfamide, or carboplatin. Prior to each course of chemotherapy, patients were evaluated by history, physical exam, and laboratory studies, including a pre-treatment albumin. Additional laboratory measurements were obtained at the midpoint of each 28-day cycle to assess for toxicity. After the third and sixth cycles of treatment, measurable disease was assessed according to RECIST criteria [14] . Toxicity was assessed using the National Cancer Institute's common toxicity criteria. Participants remained in the study for six cycles of chemotherapy. They were withdrawn from study participation if any of the following occurred: progression of disease, significant toxicity as determined by study protocol, or voluntary withdrawal. Chemotherapy was given on day 1 of a 28-day cycle. Patients were premedicated with dexamethasone (20 mg) orally, 12 and 6 h prior to treatment, and then 10–20 mg intravenously, 30–60 min prior to infusion. Diphenhydramine (50 mg) intravenously and cimetidine (300 mg) intravenously were also given 30 min prior to treatment. For nausea, patients were treated with ondansetron (32 mg) intravenously or granisetron (10 μg/kg) intravenously. Mesna (600 mg/m 2 ) was infused 15 min prior to ifosfamide (2 g/m 2 ) given over 2 h. Paclitaxel (175 mg/m 2 ) over 1 h was followed by carboplatin (AUC 5) over 45 min. This was followed by mesna (600 mg/m 2 ) intravenously, 4 h after the initial dose of ifosfamide, and then mesna (1200 mg/m 2 ) orally, 4 h later (total 3 doses of mesna). The third dose of mesna could also be given intravenously at 600 mg/m 2 . No treatment was given unless day 1 laboratory studies were adequate (ANC ≥ 1500/μL and platelets ≥ 100,000/μL). If the platelet nadir was less than 50,000/μL, the doses of ifosfamide and carboplatin were reduced one dose level. The dose of ifosfamide was reduced to 1.5 g/m 2 and the dose of carboplatin was reduced to AUC of 4. The dose of mesna was also reduced accordingly. For febrile neutropenia, treatment was delayed 1 week. The use of granulocyte colony stimulating factors was allowed. For febrile neutropenia that occurred despite the use of G-CSF, ifosfamide and carboplatin were reduced 1 dose level as described above. For grade 3 or 4 renal toxicity, treatment was held for a creatinine > 2 mg/dL and a creatinine clearance < 50 cm 3 /min. Therapy was resumed after the creatinine clearance improved to > 50 cm 3 /min. For a serum albumin less than 3.0 g/dL, all treatment was held. Patients were removed from study if no treatment was received for 6 weeks due to hypoalbuminemia. For grade 2 ifosfamide-related neurologic adverse events, the dose was reduced by 25%. If symptoms recurred despite this dose reduction, patients were removed from study. For grade 3 or 4 neurologic symptoms related to ifosfamide, patients were removed from study. The Functional Assessment of Cancer Therapy-Cervix (FACT-Cervix) was employed to evaluate the effects of treatment on quality of life. The FACT-Cervix (version 3) contains 56 items of quality of life measurements. It consists of the FACT-General, a 34 item assessment tool previously validated in cancer patients plus an additional cervix cancer specific subset of questions [15] . Quality of life assessments were obtained at baseline, and then after the third and sixth cycles of treatment. The reason for a missed assessment was to be documented when known. Assessments were deemed incomplete if less than 80% of the items were completed. All patients who received at least one cycle of chemotherapy were assessed for toxicity. Patients who received a minimum three cycles were evaluable for response to treatment. A complete response (CR) was defined as the disappearance of all target lesions. A partial response (PR) was described as at least a 30% decrease in the sum of the longest diameters of target lesions, using the baseline studies as a reference. Progressive disease (PD) consisted of a 20% increase in the sum of the longest diameters of target lesions. Stable disease (SD) was defined by neither sufficient decrease in lesions to qualify as a partial response nor sufficient increase in lesions to qualify as progressive disease. To be defined a response (CR or PR), repeat assessments were obtained to confirm tumor measurements. The progression free interval was determined from the time of study entry until disease progression or date of last contact for patients free of disease. The overall survival was defined from the time of enrollment until death from any cause. Patients alive at the time of analysis were censored at the time of their last contact. The Kaplan–Meier method was used to estimate overall and progression free survival. Results Twenty-eight participants were enrolled onto the study protocol from September 2001 to October 2004. All 28 patients were assessed for toxicity. Twenty-one patients (75%) were eligible for assessment of response to treatment. Of the 7 patients not assessable for response, 6 received less than 3 cycles of chemotherapy and 1 had no measurable disease at enrollment (surgically resected). The 6 patients who received less than three cycles of chemotherapy are accounted as follows: 2 experienced rapid, clinical deterioration which included pulmonary embolism, partial small bowel obstruction, deep vein thrombosis and pneumonia; two relocated and two withdrew. The baseline characteristics of the patients are listed in Table 1 . The median patient age was 52 years with a range of 19–79 years. Twenty (71%) of the cancers were squamous histology, 5 (17%) were adenocarcinomas, and 1 each was glassy cell, spindle cell, and adenosquamous. The average number of cycles of chemotherapy given was 4.1 per patient with a range of 1–6 cycles. Three patients required dose reductions: one patient required dose reductions of all three agents, and two patients required dose reductions of carboplatin alone. No dose-delays were recorded. Two patients received granulocyte colony stimulating factor (G-CSF) support, seven patients received erythropoietin stimulating agents (ESA), and one patient required both treatments during therapy. The overall response rate was 33% (4 complete responses, 3 partial responses). Stable disease occurred in 3 (14%) patients. Therapeutic benefit (CR + PR + SD) was observed in 10 patients (47%). All patients who had partial responses had distant disease alone. Three of the 4 patients who experienced complete responses also had distant lesions only. Thus, 6 of the 7 patients (86%) who responded to treatment had extra-pelvic disease only. Fifteen of the 21 patients (71%) evaluable for response had received cisplatin as a radiation sensitizer. Among this subset there were 2 complete responses, 2 partial responses, and 3 stable disease, for an overall response rate of 27% and therapeutic benefit achieved for 47%. Of the four patients who achieved a complete response, 2 had previous treatment with cisplatin; of the three patients with a partial response, 2 had previous treatment with cisplatin; all three patients with stable disease had prior cisplatin as a radiation sensitizer. The duration of the four complete responses was 7.0, 10.0, 12.0, and 13.0 months, respectively. These durations were based upon follow-up, confirmatory studies. Grade 3 and 4 adverse events are described in Table 2 . There was no grade 3 or 4 peripheral neuropathy. One patient experienced grade 3 neurologic toxicity (ifosfamide related) after her third cycle of treatment. She was removed from study per treatment protocol. An additional 13 (41%) patients experienced grade 1 or 2 neurologic toxicity, with the majority (62%) being grade 1. Four (14%) patients experienced grade 3 anemia. A total of 11 patients (34%) received blood transfusions, two for episodes of acute bleeding and the remaining 9 for chronic anemia (grades 2 and 3). Quality of life measurements were completed by 24 patients (86%) prior to initiating treatment ( Table 3 ). The median score was 76.5 ± 13.2 (range, 55–110). Two patients who did not complete surveys received their treatments at an outside facility, and this is the presumed reason for the survey not being completed. These patients did not complete any QOL instruments, although one received 3 cycles of treatment and the other received 6 cycles. For the other 3 patients who did not complete baseline QOL questionnaires, the reason is not known. After the third cycle of chemotherapy, 19 patients were eligible to complete the QOL measurement. A total of 15 (79%) patients did complete the assessment. The median score was 77.5 ± 7.1 (range, 68–89). The reasons for non-compliance are not known. Thirteen patients received 6 cycles of treatment. Of these, 12 patients (92%) completed surveys with a median score of 80.5 ± 11.5 (range, 52–87). The median scores for the subscales (physical well-being, functional well-being, emotional well-being, social/family well-being, relationship with doctor, and cervix subset) were also similar over the three time-points of assessment. None of the surveys were deemed incomplete (more than 80% of the questions were answered on each assessment). The overall median survival for all patients was 10.0 months (95% C.I., 8.6, 11.3) ( Fig. 1 ). Median progression free survival for evaluable patients was 5.0 months (95% C.I., 3.7, 6.2) ( Fig. 2 ). Discussion Our overall response rate of 33% is comparable to that seen with recent GOG studies of combination regimens. In GOG protocol 204, cisplatin plus paclitaxel had a response rate of 29%, while cisplatin plus topotecan had a 23% response rate. In a prior study, the response rate of cisplatin plus topotecan was reported as 27% [6] . Progression free survival was similar to that seen with cisplatin plus topotecan, 5 months versus 4.6 months, respectively. Overall survival of 10 months was also similar to that reported for cisplatin plus topotecan (9.4 months). This triple drug regimen had a significantly lower response rate, however, when compared to the 1999 study of Zanetta et al. (67%) and the more recent HeCOG study (59%). We had extrapolated from data of head and neck cancer that substituting cisplatin with carboplatin would not significantly alter overall response rates. While cross-study comparisons do help to provide benchmarks, such comparisons are not necessarily valid. The lower response rate of our carboplatin containing regimen compared to the cisplatin based regimen warrants further investigation. This observation is especially intriguing given that 71% of the patients evaluable for treatment response had received cisplatin as a radiation sensitizer. One important difference to note when comparing our current study to the results of the HeCOG study and GOG 204 is that we did not analyze for response based upon intention to treat principles. Since this was a phase II study, it was decided a priori that only patients receiving a minimum of three cycles would be evaluable for treatment response. The assessment points for determining response were at enrollment, 3 and 6 months. Thus, if all patients were included in the analysis of treatment response, our reported rate would be lower using an intention to treat analysis. Toxicities were predictable based upon the regimen, with neutropenia being the most common adverse event. The use of myeloid growth factors should be considered in patients receiving this regimen given the significant number of patients who experienced grade 3 or 4 neutropenia and the 11% incidence of febrile neutropenia. Neurologic adverse events due to ifosfamide were common and usually mild, as only one patient was removed from the treatment protocol because of this. Only 2 patients experienced peripheral neuropathy (grade 1) with the current study protocol. Quality of life scores remained stable for patients throughout the study. One concern with combination therapy compared to single agent therapy is that combination regimens tend to be more toxic. Although adverse events occurred in a substantial number of patients on this study, this did not produce a significant reduction in QOL measurements. This finding is especially important when treating patients in a palliative setting. One caveat in interpreting these data is that the reasons for non-compliance with the QOL assessments are not well documented, and thus, patient selection may bias the findings. Multiple combinations of chemotherapy have been studied as treatment of advanced, recurrent, and persistent carcinoma of the cervix [16–19] . The underlying hypothesis was that the improved response rates seen with combination therapy would translate into improved survival data. GOG 179 was the first randomized trial to demonstrate a survival advantage for combination therapy (cisplatin plus topotecan) over single agent cisplatin [6] . Overall survival for the combination arm was 9.4 months compared to 6.5 months in the cisplatin arm ( p = .017). In this study of 294 patients, 57% had received prior treatment with cisplatin as a radiation sensitizer. This number was equally distributed in both arms of the study. The overall response rate for patients in the cisplatin group was 13% compared with 27% in the cisplatin plus topotecan group. One hypothesis that accounts for the benefit seen with combination therapy is that single agent cisplatin may have reduced activity in patients who received cisplatin as a radiation sensitizer. In this current study, cisplatin was replaced with carboplatin based upon an improved side-effect profile without an apparent decrease in efficacy [12] . Paclitaxel is known to cause peripheral neuropathy in a dose-dependent fashion. When combined with cisplatin, the peripheral neuropathy associated with paclitaxel is reported to be more frequent and more severe [20] . As noted, peripheral neuropathy was not problematic for our patients. This study demonstrates that combination therapy with ifosfamide, paclitaxel, and carboplatin is reasonably effective in treating patients with advanced and recurrent carcinoma of the cervix. This regimen is especially appealing for patients with significant renal dysfunction, a contraindication to cisplatin therapy. Furthermore, this regimen (day 1 of a 28-day cycle) appears to be more convenient than current protocols such as cisplatin plus topotecan which is given over 3 days on a 21-day cycle or cisplatin plus paclitaxel which is a 2-day regimen. Based upon the overall response rate and survival data, the combination of ifosfamide, paclitaxel, and carboplatin should be considered for inclusion as an experimental arm in future randomized trials of combination therapy for advanced or recurrent cervical cancer. Conflict of interest statement The authors have no conflicts of interest to report. References [1] American Cancer Society Cancer Facts & Figures 2010 Accessed August 17, 2010 via: http://www.cancer.org/acs/groups/content/@nho/documents/document/acspc-024113.pdf [2] E.L. Franco N.F. Schlecht D. Saslow The epidemiology of cervical cancer Cancer J 9 2003 348 359 [3] E. Duarte-Franco E.L. Franco Cancer of the uterine cervix BMC Women's Health 4 suppl 1 2004 S13 [4] S.E. Waggoner Cervical cancer Lancet 361 2003 2217 2225 [5] J.D. Bloss J.A. Blessing B.C. Behrens Randomized trial of cisplatin and ifosfamide with or without bleomycin in squamous carcinoma of the cervix: a Gynecologic Oncology Group Study J Clin Oncol 20 2002 1832 1837 [6] H.J. Long B.N. Bundy E.C. Grendys Randomized phase III trial of cisplatin with or without topotecan in carcinoma of the uterine cervix: a Gynecologic Oncology Group Study J Clin Oncol 23 2005 4626 4633 [7] B.J. Monk M.W. Sill D.S. McMeekin Phase III trial of four cisplatin-containing doublet combinations in stage IVB, recurrent, or persistent cervical carcinoma: a Gynecologic Oncology Group Study J Clin Oncol 27 2009 4649 4655 [8] G. Mountzios M.A. Dimopoulos A. Bamias Randomized multicenter phase II trial of cisplatin and ifosfamide with or without paclitaxel in recurrent or metastatic carcinoma of the uterine cervix: a Hellenic Cooperative Oncology Group (HeCOG) study Ann Oncol 20 2009 1362 1368 [9] E.J. Buxton C.A. Meanwell C. Hilton Combination bleomycin, ifosfamide, and cisplatin chemotherapy in cervical cancer J Natl Cancer Inst 81 1989 359 361 [10] G. Zanetta F. Fei G. Parma Paclitaxel, ifosfamide and cisplatin (TIP) chemotherapy for recurrent or persistent squamous cell cervical cancer Ann Oncol 10 1999 1171 1174 [11] D.M. Shin B.S. Glisson F.R. Khuri Phase II trial of paclitaxel, ifosfamide, and cisplatin in patients with recurrent head and neck squamous cell carcinoma J Clin Oncol 16 1998 1325 1330 [12] D.M. Shin F.R. Khuri B.S. Glisson Phase II study of paclitaxel, ifosfamide, and carboplatin in patients with recurrent or metastatic head and neck squamous cell carcinoma Cancer 91 2001 1316 1323 [13] L.S. Downs P.L. Judson P.A. Argenta A phase I study of ifosfamide, paclitaxel, and carboplatin in advanced and recurrent cervical cancer Gynecol Oncol 95 2004 347 351 [14] P. Therasse S.G. Arbuck E.A. Eisenhauer New guidelines to evaluate the response to treatment in solid tumors J Natl Cancer Inst 92 2000 205 216 [15] D.F. Cella D.S. Tulsky G. Gray The functional assessment of cancer therapy scale: development and validation of the general measure J Clin Oncol 11 1993 570 579 [16] G.A. Omura J.A. Blessing L. Vaccarello Randomized trial of cisplatin versus cisplatin plus mitolactol versus cisplatin plus ifosfamide in advanced squamous carcinoma of the cervix: a Gynecologic Oncology Group Study J Clin Oncol 15 1997 165 171 [17] J.D. Bloss J.A. Blessing B.C. Behrens Randomized trial of cisplatin and ifosfamide with or without bleomycin in squamous carcinoma of the cervix: a Gynecologic Oncology Group study J Clin Oncol 20 2002 1832 1837 [18] P.G. Rose J.A. Blessing D.M. Gershenson R. McGehee Paclitaxel and cisplatin as first-line therapy in recurrent or advanced squamous cell carcinoma of the cervix: a Gynecologic Oncology Group Study J Clin Oncol 17 1999 2676 [19] D.H. Moore J.A. Blessing R.P. McQuellon Phase III study of cisplatin with or without paclitaxel in stage IVB, recurrent, or persistent squamous cell carcinoma of the cervix: a Gynecologic Oncology Group study J Clin Oncol 22 2004 3113 3119 [20] C. Wasserheit A. Frazein R. Oratz Phase II Trial of paclitaxel and cisplatin in women with advanced breast cancer: an active regimen with limiting neurotoxicity J Clin Oncol 14 1996 1993 1999