an outpatient intraperitoneal chemotherapy regimen for advanced ovarian cancer
TRANSCRIPT
Gynecologic Oncology 113 (2009) 63–67
Contents lists available at ScienceDirect
Gynecologic Oncology
j ourna l homepage: www.e lsev ie r.com/ locate /ygyno
An outpatient intraperitoneal chemotherapy regimen for advanced ovarian canceri
Emily Berry a,⁎, Kellie S. Matthews c, Diljeet K. Singh b, Barbara M. Buttin b, John R. Lurain b,Ronald D. Alvarez c, Julian Schink b
a Division of Gynecologic Oncology, Department of Obstetrics andGynecology, OregonHealth and Science University, 3181 SWSam Jackson Park Road,Mail Code L-466, Portland, OR 97239, USAb Northwestern University Feinberg School of Medicine, Chicago, IL 60611, USAc University of Alabama at Birmingham, Birmingham, AL 35249, USA
i Preliminary findings of this study were presenGynecologic Oncologists 39th Annual Meeting, Tampa, F* Corresponding author. Fax: +1 503 494 1835.
E-mail address: [email protected] (E. Berry).
0090-8258/$ – see front matter © 2009 Elsevier Inc. Adoi:10.1016/j.ygyno.2008.12.035
a b s t r a c t
a r t i c l e i n f oArticle history:
Objectives. To assess the Received 2 September 2008Available online 7 February 2009Keywords:Intraperitoneal chemotherapyOvarian adenocarcinomaOutpatientAntineoplastic combined chemotherapyCisplatinPaclitaxel
feasibility, associated toxicities, and reasons for early cessation of an outpatientintraperitoneal (IP) chemotherapy regimen for treatment of advanced ovarian cancer following optimalcytoreductive surgery.
Methods. Between January 2006 and December 2007, 42 patients with stages IIC–IV epithelial ovarian,tubal, or primary peritoneal cancer who had residual disease b1 cm after cytoreductive surgery were treatedwith an outpatient IP chemotherapy protocol. Patients received intravenous (IV) docetaxel 75 mg/m2 and IPcisplatin 75–100 mg/m2 on day 1, followed by IP paclitaxel 60 mg/m2 on day 8, with the intent to treatpatients every 21 days for 6 cycles of chemotherapy. Charts were abstracted for demographic, chemotherapy,and toxicity-related data.
Results. The median age of the 42 patients was 59 years (range 33–70) and the majority of patients hadepithelial ovarian cancer (80k), FIGO stage IIIC (83k), and papillary serous histology (74k). Of an intended252 IP chemotherapy cycles, 172 (68k) were administered. Twenty-nine patients (69k) completed 4 cyclesand 12 (29k) received all 6 IP cycles. Common grade 3/4 toxicities by patient included neutropenia (43k),infection (21.5k), and gastrointestinal effects (14k). There was one treatment-related death. Reasons fordiscontinuation were largely chemotherapy (43k) or port (37k) related.
Conclusions. With supportive measures, such as scheduled hydration and granulocyte colony-stimulatingfactors, outpatient administration of IP chemotherapy was feasible. This regimen resulted in fewhospitalizations or treatment delays and demonstrated less toxicity than previously reported IP chemotherapyregimens. Port-related complications were a leading cause of IP chemotherapy discontinuation.
© 2009 Elsevier Inc. All rights reserved.
Introduction
In the United States, epithelial ovarian cancer continues to be theleading cause of death among women with gynecologic malignancies[1]. Asmost patients presentwith advanceddisease, chemotherapy is acrucial adjunct to surgical cytoreduction. The combination of aplatinum and taxane agent has proven effective in improving theoverall survival in these patients [2], and the intravenous (IV) regimenof carboplatin and paclitaxel is currently the most frequently usedadjuvant treatment.With itsmanageable side-effect profile and ease ofadministration in the out-patient setting, this regimenhas experiencedwidespread use since demonstrating equal efficacy and less toxicitywhen compared with a combination of IV cisplatin and paclitaxel [3].
Besides the intravenous route, chemotherapy can be instilleddirectly into the peritoneal cavity, the principal site of disease spread
ted orally at the Society ofL, March 10, 2008.
ll rights reserved.
in ovarian cancer. Intraperitoneal (IP) chemotherapy was firstproposed in the 1970s as a way to maximize drug delivery to thetumor while avoiding systemic toxicities associated with IV admin-istration of the same agents [4–6]. Owing to the unique properties ofthe peritoneum, IP chemotherapy affords the opportunity to usehigher concentrations of drugs for prolonged periods of time todirectly bathe resected tumor beds, lymph node basins, and residualtumor nodules. Many chemotherapeutic agents used in the treatmentof ovarian cancer, including cisplatin and paclitaxel, have been foundto be effective and safe for intraperitoneal administration [4,7].
In the last 10 years, three large prospective phase III clinical trialshave shown survival advantages for patients receiving IP versus IVchemotherapy for the treatment of optimally cytoreduced, advancedovarian cancer [8–10]. Despite this demonstrated survival advantage,IP chemotherapy continues to await universal acceptance as first-linetreatment for advanced epithelial ovarian cancer. Recognized barriersto its widespread use include increased cost, inconvenience of in-patient administration, and the potential for increased toxicities andcatheter complications. With the intent to diminish some of thesebarriers, an outpatient IP chemotherapy protocol was developed and
Table 2Patient and disease characteristics (N=42)
Characteristic Number (k)
Median age (range) 59 years (33–70 years)Gravity, parityNulliparous 10 (24k)Multiparous 32 (76k)
RaceCaucasian 40 (95k)Asian 2 (5k)
FIGO stageIIC 2 (5k)IIIA 1 (2k)IIIB 3 (7k)IIIC 35 (83k)IV 1 (2k)
Disease siteOvary 36 (86k)Fallopian tube 2 (5k)Peritoneal 4 (9k)
HistologyPapillary serous 31 (74k)Mixed 4 (9k)Endometrioid 2 (5k)Clear cell 2 (5k)MMMT 2 (5k)Poorly differentiated 1 (2k)
Where stagedNU/UAB* 38 (91k)Outside hospital 4 (9k)
Radical surgical proceduresRadical hysterectomy 4 (9k)Peritonectomy 3 (7k)Rectosigmoid resection with reanastamosis 5 (12k)Ureteral diversion 1 (2k)
⁎ NU=Northwestern University, Chicago, IL USA. UAB=University of Alabama -Birmingham, Birmingham, AL USA.
64 E. Berry et al. / Gynecologic Oncology 113 (2009) 63–67
piloted at the Northwestern University (NU) Lurie ComprehensiveCancer Center, commencing January 2006, using the GynecologicOncology Group (GOG) Study #172 [8] as a template.
The primary objectives of this retrospective review of our pilotprogram include: 1) to examine the feasibility of administering IPchemotherapy on an outpatient basis, 2) to evaluate the toxicitiesassociated with this regimen and, 3) to assess the reasons for earlycessation of IP chemotherapy.
Materials and methods
Utilizing clinic registries, patients undergoing IP chemotherapy atNorthwestern University (NU) and the University of Alabama atBirmingham (UAB) between January 1, 2006 and December 31, 2007were identified. Forty-two patients were deemed evaluable for thisretrospective chart review, (NU=30, UAB=12). At both institutions,Institutional Review Board approval was obtained.
Patients eligible for evaluation included those with epithelialovarian, fallopian tube, and peritoneal adenocarcinoma, FIGO stagesIIC–IV, who had undergone primary surgical cytoreduction. Allpatients were left with minimal (lesion size 1 cm) to no gross residualdisease, had a GOG performance status of 0–1, and had adequate bonemarrow, renal, and hepatic functions.
Patient records were abstracted for demographic data, surgicalprocedures performed, and histologic tumor diagnosis. Detailspertaining to the timing of IP port placement and its anatomicallocation were noted. Charts were reviewed for type and dose ofchemotherapeutic agents administered and associated adverse effectsrecorded at the routine clinic visits preceding each treatment cycle.Toxicities were graded using the Common Terminology Criteria forAdverse Events (CTCAE), version 3.0. Treatment delays 1 week andrelated hospitalizations were detailed, and reasons for discontinuingIP therapy prematurely were recorded.
The outpatient IP regimen consisted of treatment on days 1 and 8of a 21-day cycle for an intended 6 cycles, and is detailed in Table 1. Ofnote, patients at NU received 100 mg/m2 of IP cisplatin on day 1,preceded by an amifostine infusion. Patients at UAB received 75 mg/m2 of IP cisplatin on day 1, without amifostine. Dose reductions ofchemotherapeutic agents were made when a patient experienced anyof the following: febrile neutropenia, recurrent delays in therapy,grade 2 abdominal pain, and grade 2 neuropathy. When IPchemotherapy was discontinued prematurely, patients were offeredcontinuationwith intravenous carboplatin and paclitaxel or docetaxel,
Table 1Outpatient intraperitoneal chemotherapy protocol
Day 1:Intravenous pre-hydration of 1 L 0.9 NS with simultaneous intra-peritoneal infusion of1.5 L 0.9 NS, administered over 2 hPre-chemotherapy medications including decadron, lorazapam, aprepitant andpalonosetronIntravenous Docetaxol 75 mg/m2, administered over 1 hAmifostine 910 mg/m2 IV over 10 minIntraperitoneal Cisplatin 75–100 mg/m2, administered over 1 hPost-chemotherapy hydration of 1 L 0.9 NS over 1 h
Day 2–7:Filgrastim 300 mcg subcutaneously once dailyHome hydration with electrolyte replacement as needed on days 3 and 5
Day 8:Pre-chemotherapy medications including decadron, diphenhydramine, cimetidine, andondansetronIntraperitoneal Paclitaxel 60 mg/m2
Post-chemotherapy hydration of 1.5 L 0.9 NS IP over 1 h
Day 9–10:Pegfilgrastim 6 mg subcutaneously once on day 9Home hydration and electrolyte replacement as needed on day 10
administered once every 3 weeks until a total of 6 cycles ofchemotherapy had been completed or progression of disease noted.
Intraperitoneal ports were placed intra-operatively (NU) or post-operatively by interventional radiology (NU), or postoperatively usinga laparoscopic approach (UAB). In all cases, a BardR (BardPort, BardAccess Systems, Salt Lake City, UT, product # 0602270) titanium, high-profile port with a pre-attached silicone 9.6 Fr single lumen catheterwas used. When placed at the time of surgery, a separate incision wascreated (lower abdominal quadrant or inferior to the costal margin).The subcutaneous tissue was dissected until the underlying fascia wasidentified. The port was anchored to the fascia with 2-0 Ticron suturesin 4 locations after the catheter had been tunneled into the peritonealcavity. Catheters were cut to allow 20–30 cm of length within theabdomen. After placement, the port was flushed with 10 mL heparin(100 units/mL) and the overlying skin was closed. For laparoscopicplacement, a 10 mm laparoscopewas typically placed in the left upperquadrant of the abdomen. Then, a 5 cm incisionwas created in the leftlower abdominal quadrant, superior to the left iliac crest, forplacement of the port. The subcutaneous tissue was dissected to thelevel of the fascia, the port positioned, and the catheter tunneledcephalad and inserted into the peritoneal cavity. As mentioned above,the port was anchored to the fascia, the patency tested, and theoverlying skin incision closed.
Results
Patient and disease characteristics are listed in Table 2. Themajority of the 42 patients had epithelial ovarian cancer (86k), FIGO
Fig. 1. Number of cycles of IP chemotherapy completed per patient.
Table 4Frequency of adverse events by patient
NU/UAB study(N=42)
GOG 172 IP arm(N=201)
HematologicGrade 2 neutropenia 2 (5k)Grade 3/4 neutropenia 18 (43k) 152 (76k)Grade 3/4 thrombocytopenia 2 (5k) 24 (12k)
Neurologica
Grade 1 10 (24k)Grade 2 9 (21k)Grade 3 0 39 (19k)
Infectious (port-related)Grade 2 3 (7k)Grade 3 5 (17k)
Infectious (other)b
Grade 2 1 (2k)Grade 3/4 4 (9.5k) 33 (16k)
GastrointestinalGrade 2 abdominal pain 1 (2k)Grade 2 nausea/vomiting/diarrhea 4 (9.5k)Grade 3 nausea/vomiting/diarrhea 6 (14k) 92 (46k)
ConstitutionalGrade 2 fatigue 1 (2k)Grade 3 fatigue 4 (9.5k) 36 (18k)
Metabolicc
Grade 2 2 (5k)Grade 3 1 (2k) 55 (27k)
RenalGrade 1 acute renal failure 2 (5k)Grade 3 acute renal failure 1 (2k) 14 (7k)
VascularGrade 3 pulmonary embolus 2 (5k)
DermatologicGrade 2 extravasation injury 1 (2k)
AuditoryGrade 2 tinnitus 1 (2k)
The following notations refer to the NU/UAB data only:a Refers to peripheral sensory neuropathy.b The grade 2 infection was a vaginal cuff cellulitis treated with oral antibiotics; the
grade 3 infections included three patients with infected lymphocysts, all of whomrequired hospital admission, drain placement, and intravenous antibiotics; the grade 4infection occurred in a neutropenic, febrile patient found to be growing MRSA from herchest port-a-cathR. This patient became septic and required a twoweek hospitalization.
c Includes the conditions of hypokalemia and hypomagnesemia.
65E. Berry et al. / Gynecologic Oncology 113 (2009) 63–67
stage IIIC (83k), and papillary serous histology (74k). Four patients(9k) underwent surgical staging at an institution other than wherechemotherapy was administered. To achieve optimal cytoreduction,radical surgical procedures were performed in 13 patients (31k). IPports were placed either at the time of staging surgery (60k), as aseparate procedure (26k), or by Interventional Radiology (14k). TheCA-125 tumormarker was elevated in 91k of patients prior to surgicalstaging.
All 42 patients received their intraperitoneal chemotherapy on anoutpatient basis as outlined in the protocol. Of the intended 252 cyclesof IP chemotherapy, 172 (68k) were administered. The remainingcycles were given intravenously (79 cycles, 31k) or omitted (1 cycle).Twenty-nine patients (69k) completed 4 cycles and 12 patients(29k) completed all 6 cycles of IP chemotherapy (Fig. 1). Six patientsreceived intravenous chemotherapy (carboplatin and paclitaxel) fortheir first treatment cycle before commencing IP chemotherapy, 3because of colon resection and re-anastamosis, 2 because they wereawaiting port placement after staging at an outside hospital, and 1because her port could not be accessed at the time of cycle one (portplaced in radiology between cycles 1 and 2).
During chemotherapy, 9 patients (21k) underwent thirteentreatment-related hospitalizations, with themedian length of hospitalstay being 4 days (range 1–19 days). Reasons for admission were bothchemotherapy (4 patients) and port-related (5 patients). Treatmentdelays in administration of IP chemotherapy of 1 week or greateroccurred among 9 (5k) of 172 IP chemotherapy cycles. Reasons for thetreatment delays not necessitating hospitalization were attributed toport-related complications (3), grade 3 neutropenia (3), leukocytosis(2), and post-operative abdominal wound infection (1). One treat-ment-related death was attributed to erosion of the intraperitonealcatheter into the small bowel with resulting perforation; exploratorylaparotomy and jejunostomy were performed, but the patientsuccumbed to sepsis 1 month after the operation.
Ultimately, 30 of the 42 patients (71k) discontinued intraper-itoneal chemotherapy prior to the completion of 6 cycles of treatment.
Table 3Reasons for discontinuation of intraperitoneal chemotherapy (N=30)
Chemotherapy-related 13 (43k)Grade 1 acute renal failure 1Grade 3 nausea/vomiting 3Grade 3 fatigue 4Grades 1–2 peripheral neuropathy 5
Port-related 11 (37k)Bowel perforation 1Peritoneal–vaginal fistula 1Port leaking 3Infection/cellulitis 6
Other 6 (20k)
For the majority of patients (43k), chemotherapy-related toxicitieswere the inciting reason (Table 3), with grades 1 and 2 peripheralneuropathy cited most frequently (5 patients). Port-related complica-tions contributed to 11 patients' decision to discontinue IP che-motherapy (37k), including most commonly port infection/cellulitis(6 patients) and leaking (3 patients). Other reasons for suspendingtherapy included infected lymphocyst (1), vascular access catheterinfection with resulting sepsis (1), disease progression/platinumresistance (1), and intolerable quality of life while on IP therapy (3).
Toxicities were recorded and graded prior to each chemotherapyadministration. As seen in Table 4, the most frequent grade 3 or 4toxicities noted per patient over the course of treatment includedneutropenia (43k), infection (port-related or otherwise; 21.5k),gastrointestinal (14k), and fatigue (9.5k). There were no grade 3 or 4neurologic toxicities noted.
Port-related complications are reported in Table 5. Not allport and catheter-related complications led to discontinuation of IP
Table 5Port-related complications
Category Number (k) N=42
Infection/cellulitis 7 (17k)Leaking 5 (12k)Access 2 (5k)Bowel perforation 1 (2k)Peritoneal–vaginal fistula 1 (2k)Abdominal pain 1 (2k)
66 E. Berry et al. / Gynecologic Oncology 113 (2009) 63–67
chemotherapy. Infection/cellulitis was the most common complica-tion observed (17k), followed by leaking (12k). The one patientwith a peritoneal–vaginal fistula was able to resume IP treatmentafter 1 cycle of IV chemotherapy. Four of the seven patients with portaccess difficulties or port leaking ultimately required a new port atsome point during their treatment, all placed by InterventionalRadiology. Of these 4, two received a cycle of IV chemotherapy whilethe port was replaced, the other two stayed on schedule to receivethe next cycle on the IP regimen.
Discussion
Our study details an outpatient IP chemotherapy regimen designedas a modification of the GOG #172 protocol [8] with an aim to reduceassociated toxicities. This outpatient regimen is feasible, with 69k ofpatients completing 4 or more cycles of IP therapy. Grade 3 and 4toxicities were most frequently hematologic, infectious, or gastro-intestinal, and port-complications led to discontinuation of IP therapyalmost as frequently as did chemotherapy side-effects (37k versus43k).
This early experiencewith an outpatient IP chemotherapy regimenmet with both achievements and obstacles. We were able tosuccessfully administer all IP cycles on an outpatient basis, facilitatedby scheduled home hydration and electrolyte replacement. Inaddition, the frequency of grade 3 and 4 toxicities was notablymitigated by things such as granulocyte colony stimulating factors, anaggressive anti-emetic regimen, the substitution of IV docetaxel for IVpaclitaxel, and potentially also by the addition of amifostine. Thisstudy contains a small number of patients and is retrospective innature, with no long-term outcomes or survival data available torecommend it as a viable alternative to the existing IP chemotherapyregimen. Twelve patients (29k) were able to complete all 6 cycles(versus 42k in GOG 172), a smaller percentage than was hoped forwith this modified protocol, with many patients discontinuing IPchemotherapy for moderate (grade 2) toxicities. Among all grade 3infections noted, the majority were due to IP ports. We believe thatsome of the reported port infections were instead soft tissue chemicalcellulitis secondary to paclitaxel extravasation/port leaking becausewhile all cases in this category had fever and erythema around theport site, not all had other objective evidence of infection such aselevation of the white blood cell count, purulent discharge from theport site, or culture-proven infection from samples taken from thesubcutaneous pocket where the port was located. The number andtype of port and catheter-related complications were high, andaccounted for many patients discontinuing IP therapy. While thesmall sample size makes correlation with things like the timing andtechnique of catheter placement difficult, further investigation intothis complication rate is warranted.
The extent of chemotherapy-related toxicities experienced bypatients on the GOG #172 regimen influenced the design of ourprotocol. In particular, we focused on reducing the incidence of grades3 and 4 neutropenia/leukopenia, as well as gastrointestinal (nausea/vomiting), metabolic, and neurologic toxicities. As severe leukopeniawas noted among 76k of patients receiving IP chemotherapy on GOG#172, and because our protocol substituted docetaxel for the 24-hourpaclitaxel infusion on day 1, all patients received scheduled granulo-
cyte-colony stimulating factors, resulting in an incidence of grade 3/4neutropenia of 43k, including no hospitalizations for febrile neu-tropenia. The rationale for use of filgrastim and pegfilgrastim followedthe 2006 American Society of Clinical Oncology recommendations aswe anticipated our regimen to carry a greater than 20k risk of febrileneutropenia [9]. Gastrointestinal and metabolic toxicities were notedamong 46k and 27k of all patients receiving IP chemotherapy on theGOG #172 protocol, respectively. Acknowledging the emetogenicpotential of the cisplatin dose of 100 mg/m2, an anti-emetic regimenincluding aprepitant, palonosetron, dexamethasone and lorazapamwas employed [10], resulting in only 12k of patients experiencingsevere nausea/vomiting. Furthermore, using a home healthcareagency to provide scheduled hydration and electrolyte repletion,severe metabolic events were experienced by only 2k of patients.Efforts to reduce neurotoxicitiy, specifically peripheral neuropathy,were two-fold. Docetaxel was substituted for the 24-hour paclitaxelinfusion on day 1 of the regimen, with the rationale for this changebased on the 2004 study by Vasey et al. [11], where patients withovarian cancer receiving the combination of docetaxel and carboplatinshowed similar survival rates and significantly less grade 2 or higherneurosensory toxicity than those receiving paclitaxel and carboplatin(11k versus 30k, respectively). The dose of cisplatin used in both theGOG #172 and our regimen by the NU patients, 100 mg/m2, wasundoubtedly another cause of neurotoxicitiy. To counteract this,amifostine, a cytoprotective agent shown to reduce cisplatin-relatedneurotoxicity, was administered to the NU patients without anytoxicities noted in those who received it. In the largest of severalstudies examining the cytoprotective abilities of amifostine againstcombinations of paclitaxel and platinum [14–17], Kemp et al. [12]demonstrated that in ovarian cancer patients treatedwith six cycles ofIV cyclophosphamide 1000 mg/m2 and cisplatin 100 mg/m2,pretreatment with amifostine significantly reduced the severity ofperipheral neuropathy (P=.029). Though not employed in this study,the substitution of IP carboplatin for cisplatin as a toxicity-reducingmeasure is a valid area of interest and could be considered. Apreliminary toxicity analysis of IP carboplatin in combination with IVpaclitaxel by Fujiwara et al. [13] has shown that the dose-limitingtoxicity associated with IP carboplatin is thrombocytopenia andrecommended an area under the curve (AUC) dose of 6.0–7.0. Withfurther study, carboplatin may emerge once again as an equallyeffective and less toxic alternative to cisplatin.
As mentioned earlier, the patients treated at NorthwesternUniversity (N=30) all received an IP cisplatin dose of 100 mg/m2,as opposed to the 12 patients from UAB who received an IP cisplatindose of 75 mg/m2. This difference allowed for further analysis of thetwo groups in respects to common cisplatin-related toxicities. Withregards to renal, neurologic, and metabolic toxicities, these eventswere so few that no difference was seen between the two patientpopulations. In terms of grade 3/4 GI events (nausea and vomitingspecifically), 2 of the 30 NU patients versus 3 of the 12 UAB patientsexperienced this complication. When looking at grade 3/4 myelo-suppression (neutropenia, thrombocytopenia), we observed 12 of 30NU patients with this toxicity as opposed to 8 of 12 UAB patients.Based on this very small data set, one cannot obviously conclude thatthe higher cisplatin dose received by NU patients accounted forexcessive toxicity. The role of amifostine in reducing neurotoxicity isless clear. The NU cohort did receive amifostine, the UAB group didnot. Neither cohort had grade 3/4 neurotoxicity. Interestingly, morepatients in the UAB group (9) finished all 6 cycles of IP chemotherapyas opposed to the NU group (3). When stopping IP chemotherapy dueto chemotherapy-related toxicities, there was no difference betweenthe number of patients stopping for grade 1/2 toxicities (6 of 13) asopposed to those stopping for grade 3/4 toxicities (7 of 13). Whetheror not receiving all 6 cycles of adjuvant chemotherapy intraperitone-ally translates into a survival difference is still unknown. Until then, IPchemotherapy regimens such as ours continue to require scrutiny in
67E. Berry et al. / Gynecologic Oncology 113 (2009) 63–67
all aspects of the protocol in order to ultimately achieve a greaterpercentage of patients completing all 6 cycles of treatment via the IProute.
Administration of intraperitoneal chemotherapy is not without itscomplications. Indeed, 37k of patients reviewed in this studydiscontinued IP chemotherapy secondary to port and cathetercomplications, many of which contributed to treatment-relatedhospitalizations and delays. Over the last 20 years, the MemorialSloan-Kettering Cancer Center has reviewed their experience withfenestrated intraperitoneal catheters on three occasions. In the firstseries, Davidson et al. [14] reported a complication rate of 17.6k (40 of227 patients), and found that of the 20 (8.8k) patients who developedcatheter-related infections, 8 (3.5k) were associated with bowelperforation. A trend toward increased infectious complications withplacement of the catheters at the time of large bowel surgery wasobserved but was not statistically significant. In a follow-up study,Makhija et al. [15] found a complication rate of 10k among 301patients, with no bowel perforations reported. An impressive 93k ofpatients completed the planned therapy and the authors found thatboth avoidance of catheter placement at the time of large bowelsurgery and placement at the time of second-look laparoscopyresulted in fewer inflow and infectious complications. Among thevariety of chemotherapeutic agents administered, mitoxantrone wasassociated with the highest complication rate (21.3k). In the largestand most recent series, Black et al. [16] found only 3k of patientsdiscontinuing chemotherapy due to catheter complications. In thisstudy the majority of catheters were placed laparoscopically (67k),for consolidation therapy, and no patient received IP mitoxantrone.Patients with catheter placement at the time of large bowel surgeryhad similar completion rates to those who did not undergo bowelsurgery. Walker et al. [17] reviewed the 119 patients discontinuing IPchemotherapy in the GOG #172 trial and found that 40 patients (34k)discontinued treatment for catheter-related reasons, mainly infection(n=21). Most patients underwent port placement as a separateprocedure (n=133). Important findings included failure to initiate IPchemotherapy in 16k of patients who had versus 5k of patients whodid not have left colon or rectosigmoid colon resection and a lack ofassociation between timing of catheter insertion and failure tocomplete IP therapy. Finally, in a recent single institution review ofport complications experienced among 83 patients undergoing IPchemotherapy with a platinum and taxane agent, Landrum et al. [18]found a 13k port complication rate, with the majority of patients(67k) having the port placed at a second procedure, including allpatients undergoing small and large bowel resection. Infection wasthe most common reason for catheter-related discontinuation of IPtherapy (4 of 11 patients). The 6 cycle completion rate was notaffected by the timing of port placement (63k at initial surgery versus59k at mini-laparotomy, P=.73). The compilation of these reportssuggests that port and catheter-related complications are reduced inthe hands of an experienced center and when port placement isdelayed in the setting of large bowel surgery.
In conclusion, with this small study we have demonstrated thefeasibility of an outpatient IP chemotherapy regimen designed as amodification of the GOG #172 protocol. The use of granulocyte-colonystimulating factors, multiple anti-emetics, scheduled home hydration
and IV docetaxel reduced toxicities associated with our protocol. Alarge number of patients experienced port and catheter complicationsdespite our experience with these devices, and this deserves furtherattention as it affected the number of patients ultimately able tocomplete 6 cycles of IP therapy.
Conflict of interestThe authors declare that there are no conflicts of interest.
References
[1] Jemal A, Siegel R, Ward E, Murray T, Xu J, Thun MJ. Cancer Statistics, 2007. CACancer J Clin 2007;57:43–66.
[2] McGuire WP, Hoskins WJ, Brady MF, Kucera PR, Partridge EE, Look KY, et al.Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin inpatients with stage III and stage IV ovarian cancer. N Engl J Med 1996;334:1–6.
[3] Ozols RF, Bundy BN, Greer BE, Fowler JM, Clarke-Pearson D, Burger RA, et al. PhaseIII trial of carboplatin and paclitaxel compared with cisplatin and paclitaxel inpatients with optimally resected stage III ovarian cancer: a Gynecologic OncologyGroup study. J Clin Oncol 2003;21:3194–200.
[4] Jones RB, Myers CE, Guarino AM, Dedrick RL, Hubbard SM, DeVita VT. High volumeintraperitoneal chemotherapy (bbelly bathQ) for ovarian cancer. Pharmacologicbasis and early results. Cancer Chemother Pharmacol 1978;1:161–6.
[5] Dedrick RL, Myers CE, Bungay PM, DeVita Jr VT. Pharmacokinetic rationale forperitoneal drug administration in the treatment of ovarian cancer. Cancer TreatRep 1978;62:1–11.
[6] Dedrick RL, Flessner MF. Pharmacokinetic problems in peritoneal drug adminis-tration: tissue penetration and surface exposure. J Natl Cancer Inst 1997;89:480–7.
[7] Markman M, Rowinsky E, Hakes T, Reichman B, Jones W, Lewis Jr JL, et al. Phase Itrial of intraperitoneal taxol: a Gynecoloic Oncology Group study. J Clin Oncol1992;10:1485–91.
[8] Armstrong DK, Bundy B, Wenzel L, Huang HQ, Baergen R, Lele S, et al.Intraperitoneal cisplatin and paclitaxel in ovarian cancer. N Engl J Med 2006;354:34–43.
[9] Smith TJ, Khatcheressian J, Lyman GH, Ozer H, Armitage JO, Balducci L, et al. 2006update of recommendations for the use of white blood cell growth factors: anevidence-based clinical practice guideline. J Clin Oncol 2006;24:3187–205.
[10] Hesketh PJ, Grunberg SM, Gralla RJ, Warr DG, Roila F, de Wit R, et al. The oralneurokinin-1 antagonist aprepitant for the prevention of chemotherapy-inducednausea and vomiting: a multinational, randomized, double-blind, placebo-controlled trial in patients receiving high-dose cisplatin—the Aprepitant Protocol052 Study Group. J Clin Oncol 2003;21:4112–9.
[11] Vasey PA, Jayson GC, Gordon A, Gabra H, Coleman R, Atkinson R, et al. Phase IIIrandomized trial of docetaxel-carboplatin versus paclitaxel-carboplatin as first-line chemotherapy for ovarian carcinoma. J Natl Cancer Inst 2004;96:1682–91.
[12] Kemp G, Rose P, Lurain J, Berman M, Manetta A, Roullet B, et al. Amifostinepretreatment for protection against cyclophosphamide-induced and cisplatin-induced toxicities: results of a randomized control trial in patients with advancedovarian cancer. J Clin Oncol 1996;14:2101–12.
[13] Fujiwara K, Suzuki S, Ishikawa H, Oda T, Aotani E, Kohno I. Preliminary toxicityanalysis of intraperitoneal carboplatin in combination with intravenous paclitaxelchemotherapy for patients with carcinoma of the ovary, peritoneum, or fallopiantube. Int J Gynecol Cancer 2005;15:426–31.
[14] Davidson SA, Rubin SC, Markman M, Jones WB, Hakes TB, Reichman B, et al.Intraperitoneal chemotherapy: analysis of complications with an implantedsubcutaneous port and catheter system. Gynecol Oncol 1991;41:101–6.
[15] Makhija S, Leitao M, Sabbatini P, Bellin N, Almadrones L, Leon L, et al.Complications associated with intraperitoneal chemotherapy catheters. GynecolOncol 2001;81:77–81.
[16] Black D, Levine DA, Nicoll L, Chou JF, Iasonos A, Brown CL, et al. Low risk ofcomplications associated with the fenestrated peritoneal catheter used forintraperitoneal chemotherapy in ovarian cancer. Gynecol Oncol 2008;109:39–42.
[17] Walker JL, Armstrong DK, Huang HQ, Fowler J, Webster K, Burger RA, et al.Intraperitoneal catheter outcomes in a phase III trial of intravenous versusintraperitoneal chemotherapy in optimal stage III ovarian and primary peritonealcancer: a Gynecologic Oncology Group Study. Gynecol Oncol 2006;100:27–32.
[18] Landrum LM, Gold MA, Moore KN, Myers TK, McMeekin DS, Walker JL.Intraperitoneal chemotherapy for patientswith advancedepithelial ovarian cancer:a review of complications and completion rates. Gynecol Oncol 2008;108:342–7.