Annals of Surgical Oncology

, Volume 15, Issue 4, pp 1183–1192

Cytoreductive Surgery and Intraoperative Hyperthermic Intraperitoneal Chemotherapy with Paclitaxel: A Clinical and Pharmacokinetic Study


    • Department of Surgical OncologyUniversity Hospital, Crete University Medical School
  • Hilde Rosing
    • Department of Pharmacy and PharmacologySlotervaart Hospital/The Netherlands Cancer Institute
  • Dimitris Filis
    • Department of Surgical OncologyUniversity Hospital, Crete University Medical School
  • John Romanos
    • Department of Surgical OncologyUniversity Hospital, Crete University Medical School
  • Maria Melisssourgaki
    • Department of Surgical OncologyUniversity Hospital, Crete University Medical School
  • Markos Daskalakis
    • Department of Surgical OncologyUniversity Hospital, Crete University Medical School
  • Maria Pilatou
    • Department of Surgical OncologyUniversity Hospital, Crete University Medical School
  • Elias Sanidas
    • Department of Surgical OncologyUniversity Hospital, Crete University Medical School
  • Panagiotis Taflampas
    • Department of Surgical OncologyUniversity Hospital, Crete University Medical School
  • Konstantinos Kalbakis
    • Department of Medical OncologyUniversity Hospital, Crete University Medical School
  • Jos H. Beijnen
    • Department of Pharmacy and PharmacologySlotervaart Hospital/The Netherlands Cancer Institute
  • Dimitris D. Tsiftsis
    • Department of Surgical OncologyUniversity Hospital, Crete University Medical School
Gynecologic Oncology

DOI: 10.1245/s10434-007-9792-y

Cite this article as:
de Bree, E., Rosing, H., Filis, D. et al. Ann Surg Oncol (2008) 15: 1183. doi:10.1245/s10434-007-9792-y



Intraperitoneal chemotherapy has been recommended as a treatment option for ovarian cancer with peritoneal dissemination. Although its treatment duration is significantly shorter, intraoperative hyperthermic intraperitoneal perfusion chemotherapy (HIPEC) has several advantages over simple intraperitoneal instillation chemotherapy. While platinum compounds have usually been used, only a few have administered paclitaxel during HIPEC. Its large molecular weight suggests a much more favorable pharmacokinetic profile than that of platinum compounds. The pharmacokinetics of paclitaxel during and after HIPEC have not been studied before.


Thirteen women, mainly with ovarian cancer, underwent cytoreductive surgery and HIPEC with 175 mg/m2 paclitaxel for 2 h. Morbidity was noted. Peritoneal fluid samples and blood samples were harvested during and until 5 days after HIPEC for pharmacokinetic study in ten patients.


No treatment-related mortality was noted. Overall morbidity was 38% (two wound infections, one deep venous thrombosis, two grade 1 thrombopenia, one grade 2 neutropenia, and one grade 3 pancytopenia). Mean maximal intraperitoneal paclitaxel concentration was 101 mg/L, which was an average of 1178 times higher than the peak plasma levels. The peritoneal fluid versus plasma AUC ratio was 1462 for the 2-h HIPEC duration and 366 for the total 5-day study period. Cytotoxic drug concentrations were detected in peritoneal fluid for a mean period of 2.7 days, despite drainage of the drug solution after 2 h of treatment.


HIPEC with paclitaxel following cytoreductive surgery is feasible, relatively safe, and associated with a highly favorable pharmacokinetic profile, despite its short treatment duration. Larger studies with a more homogenous patient cohort and adequate follow-up should be performed to demonstrate its efficacy.

Ovarian cancer is the fifth most frequent cancer among females with approximately two-thirds of the patients presenting with advanced disease and is responsible for 6% of all cancer deaths in females.1 Progress in the treatment of this disease is evident, with an improvement in survival rate during the last 30 years. In the 1960s, the 5-year survival rate for ovarian cancer was 30%, whereas recent statistics indicate an increase to 44%.1 Improvement of the quality of cytoreductive surgery as well as development of novel drugs and new chemotherapy regimens are the main contributors to this improvement. Prior to 1993, chemotherapy of choice for advanced ovarian cancer was cisplatin or carboplatin in combination with a classic alkylating agent such as cyclophosphamide. Since the mid-1990s cytoreductive surgery and systemic combination chemotherapy with a platinum compound and a taxane have become the standard of care for this disease.2,3 Despite this progress in systemic chemotherapy, most patients will ultimately die from their disease. The addition of intraperitoneal chemotherapy to the management of advanced ovarian cancer has been demonstrated to be beneficial and results in further improvement of survival.3 In the past, two large randomized trials (GOG 104 and 114)4,5 had demonstrated a clear benefit in small residual ovarian cancer. However, an old systemic chemotherapy regimen and addition of intravenous carboplatin administration only in the experimental arm were important criticisms of these studies. While many drugs have been administered intraperitoneally in the past, cisplatin has been most frequently used. During the last 15 years, intraperitoneal chemotherapy with paclitaxel has been used for the treatment of primary and recurrent advanced ovarian cancer in different clinical studies.6 Most recently, significantly improved survival was noted for the use of intraperitoneal chemotherapy with paxlitaxel and cisplatin in a large multicentric randomized trial (GOG 172).7 This study revealed an improvement in median progression-free survival from 18.3 to 23.8 months by intraperitoneal chemotherapy and a relative recurrence risk of 0.80 (P = 0.05) in favor of intraperitoneal treatment when compared with conventional intravenous chemotherapy. Overall survival data followed a similar trend favoring intraperitoneal chemotherapy with a median overall survival of 65.6 versus 49.7 months and a relative death risk of 0.75 (P = 0.03). This is one of the largest benefits ever observed for a new therapy in gynecological oncology. Based on the results of this and other trials, the National Cancer Institute has issued a clinical announcement recommending that women with stage III ovarian cancer who undergo optimal surgical cytoreduction should be considered for intraperitoneal chemotherapy.8

In primary and secondary peritoneal surface malignancies, a higher drug exposure to the tumor cells can be achieved with intraperitoneal drug administration. Limited and delayed drug absorption from the peritoneal cavity and the first-pass effect from the liver result in high locoregional with low systemic drug concentrations and exposure with consequently potentially higher therapeutic efficacy and limited systemic side-effects.9 Ovarian cancer is theoretically an attractive malignancy for this regional treatment, because the disease remains largely confined to the peritoneal cavity. Patients with microscopic or small volume residual disease seem to be most suitable for this approach. In more bulky disease, intraperitoneal chemotherapy is unlikely to be beneficial because drug penetration is limited and hence adequate surgical cytoreduction has to precede application of intraperitoneal chemotherapy.10,11 In addition, comprehensive adhesiolysis, mobilization of the bowel, and intraoperative peritoneal cavity expansion may overcome the problem of incomplete exposure of the seroperitoneal surface to the drugs, as observed during simple instillation chemotherapy.9,11 Thus, to optimize drug distribution, intraperitoneal chemotherapy has also been applied intraoperatively immediately after cytoreductive surgery. Another advantage of intraoperative use is that intraperitoneal chemotherapy can be administered under hyperthermic conditions, which are poorly tolerated by a patient who is awake. Hyperthermia is directly cytotoxic, and enhances the efficacy and penetration depth of many drugs.8,11 Finally, intraperitoneal administration of some agents, including cisplatin and paclitaxel, may cause severe abdominal pain, which is better tolerated intraoperatively.7,1317 A theoretical disadvantage of intraoperative intraperitoneal chemotherapy may be the short duration of treatment, when compared with instillation intraperitoneal chemotherapy (1–2 h versus 24 h).9 However, experimental studies indicate that even short-time exposure of tumor cells to high drug concentrations, as established in intraperitoneal chemotherapy, is very effective.1820

As will be discussed later the use of taxanes for intraperitoneal chemotherapy has many potential advantages.6 Previously, we have reported on the highly favorable pharmacokinetics of docetaxel during intraoperative hyperthermic intraperitoneal chemotherapy (HIPEC).21 In our institute we have been using docetaxel for HIPEC in ovarian cancer patients with promising results.22 Unfortunately, although at least as effective as paclitaxel in systemic chemotherapy, docetaxel has not been approved for treatment of ovarian cancer in Greece and many other countries. Hence, we focused our interest on the administration of paclitaxel during HIPEC and examined the feasibility of this procedure. Pharmacokinetic data of paclitaxel during HIPEC have not been previously reported. We report on such a pharmacokinetic study and compare the results with data from the previously reported pharmacokinetic study of docetaxel,21 which was similarly designed, and of those concerning intraperitoneal instillation chemotherapy with paclitaxel. Furthermore, morbidity was discussed.


Patient Eligibility

Patients with primary and secondary peritoneal tumors expected to be sensitive to paclitaxel were treated by cytoreductive surgery and HIPEC with paclitaxel. There should be no evidence of extra-abdominal or parenchymal metastases. The patient’s WHO performance status had to be ≤ 2. Blood cell count and biochemical liver and renal function tests had to be within the normal range. The protocol of this feasibility and pharmacokinetic study was approved by the ethics committee and the scientific committee of our institution, while written informed consent was obtained from all patients.

Cytoreductive Surgery and Intraperitoneal Chemotherapy

The technique of cytoreductive surgery and HIPEC used in our institution has been previously described in detail elsewhere.23 In short, the abdomen was approached through a median xyphoid-pubic incision. Comprehensive adhesiolysis was performed. The primary tumor was excised, if still present, and all visceral or parietal peritoneal surface tumor deposits were removed as completely as possible. If a deposit was infiltrating deeply into an organ and it was impossible to peel the malignancy from its surface, the involved organ or a segment of it was excised. The objective of cytoreductive surgery was to leave no macroscopic disease or, when this was impossible, tumor deposits of less than 0.5 cm in diameter behind. Anastomoses were performed arbitrarily before or after HIPEC. Subsequently, after closing the skin of the laparotomy wound only and placement of in- and outflow catheters, the peritoneal cavity was perfused using a closed perfusion model with a roller pump and a heat exchanger. The system was filled with normal saline, the pump system started, and the circulating perfusate heated. Fluid was added until an optimal abdominal expansion was achieved to facilitate adequate exposure of the entire seroperitoneal surface. At that time, an intra-abdominal pressure of at least 13 mmHg was measured. When the intraperitoneal temperature, measured by a probe situated at the mesenteric root, reaches 41°C, a dose of 175 mg/m2 paclitaxel dissolved in 4.2% Cremophor EL (Taxol®, Bristol-Myers Squibb) was administered. This dose is similar to that used for systemic chemotherapy and thus, even if all the paclitaxel were to be absorbed, a definitely acceptable systemic toxicity was to be expected. This possibly limited dose was chosen conservatively because its administration was combined with major surgery. The intraperitoneal temperature was allowed to fluctuate between 41°C and 43°C and the duration of the perfusion was 2 h. To facilitate uniform temperature and drug distribution throughout the peritoneal cavity vigorous agitation of the abdomen was performed continuously. The core temperature was kept below 38.5°C. After completion of the HIPEC, the abdomen was reopened and the canulae and temperature and pressure probes were removed. The excess fluid was drained from the abdominal cavity, but no attempt was made to dry the abdominal cavity completely. The recovered volume had not been exactly measured, but generally the difference between administered and recovered volume was estimated to be approximately 0.5 L on average. The deficit was due to absorption of the carrier solution from the peritoneal cavity as well as retention of fluid in the abdomen after HIPEC. One drain in Douglas’s pouch was left in place for postoperative drainage. The abdominal wall was closed in layers in the usual way.


Toxicity and other complications were recorded in all HIPEC procedures in which paclitaxel was administered. Hematological, cardiac, renal and other drug-related toxicity was evaluated according to WHO criteria. Anemia during the first two postoperative days was not considered hematological toxicity, but due to extensive surgery. However, hemoglobulin decline after these first two days was regarded as systemic toxicity. All surgical complications were noted.

Pharmacokinetic Study

Sample Collection

Blood and perfusate samples were obtained for measurement of paclitaxel levels. Blood samples were collected during and after HIPEC at 0.5, 1, 1.5, 2, 2.5, 3, 4, 5, and 6 h after administration of paclitaxel. Subsequently blood samples were obtained at 3-hour intervals until 24 h after drug administration, at 6-h intervals during the following 2 days and at 12-h intervals during the last 2 days. The last of the 27 blood samples was collected 5 days after paclitaxel administration. During HIPEC perfusate samples were obtained just after paclitaxel administration (at 5 min) and at the end of the perfusion (at 2 h). After 5 min the drug was considered to be equivocally distributed in the perfusate and the sample at this moment represents the maximal drug concentration in the perfusate. Because during this short period a quite linear concentration decrease is expected due to slow absorption of paclitaxel and fluid from the peritoneal cavity, samples were harvested only at these two time points, while serum levels were measured more frequently during HIPEC because of the dynamic interaction between absorption and elimination.

Postoperatively, peritoneal fluid samples were obtained from the abdominal drain which was left in place in Douglas’s pouch 24 h after the start of the HIPEC and at four consecutive days. The samples were centrifuged (10 min at 2,500 g) and the plasma layer was removed and stored at −30°C.

Paclitaxel Measurement

After completion of entry of patients in this study, drug concentrations were determined by means of a high-performance liquid chromatography (HPLC) method. A HPLC-mass spectrometer (MS)/MS method has been validated for the quantification of paclitaxel in human heparinized plasma using 13C6-paclitaxel as internal standard. The plasma sample clean-up procedure was performed by liquid–liquid extraction using tert-butyl methyl ether. After mixing and centrifuging, the aqueous layer was frozen instantaneously in a dry ice–ethanol mixture and the organic solvent was decanted into a clean tube. After evaporation of the solvent the residue was reconstituted and 25 μL aliquots were injected onto the analytical column. The analytical column was a Zorbax Extend C18 column (150 × 2.1 mm I.D., 5 μm particle size). High-performance liquid chromatography coupled with tandem mass spectrometry (MS/MS) was used for the determination of paclitaxel concentrations in human plasma. A mixture of 10 mM ammonium hydroxide–methanol (30:70, v/v) was used as eluent. With an eluent flow of 0.2 mL/min the run time was around 9 min. Positively charged ions were created at atmospheric pressure and were transferred to an API 3000 triple quadrupole MS (Sciex, Thornhill, Canada). The transitions for paclitaxel were selected from m/z 854 to 509 and for the internal standard from m/z 860 to 515. The validated concentration range is from 0.25 to 1,000 μg/L using 200 μL plasma.24 Peritoneal fluid was diluted in control human plasma and the same sample pretreatment and drug measurement method as for plasma was used. Intraperitoneal drug concentrations above the upper limit of quantification were diluted in control human plasma and reassayed.

The area under the concentration versus time curve (AUC) was calculated by the trapezoidal rule in peritoneal fluid and in plasma (WinNonlin 5.0, Pharsight Corporation, Mountain View, CA, USA), both for the HIPEC duration (i.e., 2 h) and until the drug was below the lower limit of quantification (i.e., 0.25 μg/L). When after five days drug concentrations were still measured above this lower limit of quantification, data from the first five postoperative days were extrapolated.


Patient Characteristics

Thirteen patients were consecutively enrolled in this study and underwent cytoreductive surgery and HIPEC with paclitaxel. All patients were females, with an age varying from 28 to 73 years old (mean 51 years, median 61 years). Ten patients had Fédération Internationale de Gynécologie et d’Obstétrique (FIGO) stage III ovarian cancer, while cystic mesothelioma, mixed Müllerian tumor of the uterus and pseudomyxoma peritonei were considered the primary tumor each in one case. In the last patient, pseudomyxoma peritonei was considered to be of ovarian rather than appendiceal origin and hence to be sensitive to paclitaxel. Eight patients were previously treated with paclitaxel containing chemotherapy regimens; four patients exhibited a complete response and the other four a partial response. The last chemotherapy had been administered averaged 2.5 months (range 1–4 months, median 3 months) before the HIPEC procedure. In six patients with ovarian cancer, six 3-weekly cycles of paclitaxel–carboplatin had been given, while in one, five cycles of paclitaxel–cisplatin had been administered. The patient with stage IV mixed Müllerian uterine tumor had demonstrated complete regression of systemic disease and partial response of peritoneal involvement with six 3-weekly cycles of farmorubicin–cisplatin and three of paclitaxel–carboplatin. The remaining five patients were chemotherapy naïve. In three cases cytoreductive surgery and HIPEC was part of the primary treatment. Other indications for this procedure were persistent disease (four patients), recurrent diseases (two patients) and positive second-look operation (four patients). Eleven patients had been previously operated for their tumor.

Treatment and Complications

The surgical procedures performed during cytoreductive surgery included, besides a dedicated attempt to meticulous excision of all peritoneal noduli and involved peritoneum, hysterectomy with bilateral ovariectomy (in eight patients), omentectomy (eight), cholecystectomy (nine), appendectomy (six), colectomy (two), and resection of a large local recurrence of cystic mesothelioma (one). In ten patients surgical cytoreduction was considered optimal, while in the three remaining patients tumor deposits of more than 0.5 cm in diameter had to be left behind (suboptimal cytoreduction). Subsequently, 3.5–7 L of normal saline (mean 5.0 L, median 4.5 L) were added to the perfusion system. A dosage of 175 mg/m2 paclitaxel resulted in total doses varying from 210 to 360 mg (mean 297 mg, median 300 mg).

No death was noted peroperatively and during the first 30 postoperative days. Treatment-related complications were seen in five patients, resulting in a morbidity of 38%. Surgery-related complications included wound infections in two patients and deep venous thrombosis in another case. Chemotherapy-related toxicity was observed in four patients, in two cases grade 1 thrombopenia, in one patient grade 2 neutropenia, and in another grade 3 pancytopenia. The latter patients had received systemic chemotherapy until, respectively, 4 and 3 months before HIPEC. No other systemic toxicity, including hemoglobulin decline after the first two postoperative days, was observed.

Postoperatively, seven of the ten patients with ovarian cancer and the patient with a mixed Müllerian uterine tumor received systemic chemotherapy, while the three remaining ovarian cancer patients and the patients with pseudomyxoma peritonei and cystic mesothelioma did not undergo further treatment until eventual recurrence.

Pharmacokinetic Study

From ten consecutive patients, samples were harvested for pharmacokinetic analysis. Data are summarized in Tables 1 and 2. Despite the large interindividual variation in pharmacokinetics, high intraperitoneal concentrations were obtained for a long period of time in all patients. The maximal intraperitoneal concentration was averaged 1178 times higher than the maximal plasma concentration, which was reached after 2.5 h (± 1.3 h, standard deviation). Concentrations above the cytotoxic threshold level of 0.1 μmol/L25 ( = 85 μg/L) paclitaxel were detected in peritoneal fluid for a mean period of 2.7 days (range 1–4 days, median 2.5 days) after HIPEC. In one patient they were detectable until the measure point of 24 h, in four patients for the first 48 h, in two patients for 3 days and in another three patients for 4 days. In Fig. 1, drug concentrations over time in peritoneal fluid and plasma of patient 9 are displayed as example.
Table 1.

Maximal paclitaxel concentrations in intraperitoneal fluid and plasma




Ci.p., 5 min/Cplasma,max

C5 min (mg/L)

C2 h (mg/L)

Cmax (mg/L)

Tmax (h)









































































C5 min, concentration at 5 min after drug administration (= Ci.p., max for intraperitoneal fluid); C2 h, concentration at 2 h, Cmax, maximal concentration; i.p., intraperitoneal; SD, standard deviation.

Table 2.

Intraperitoneal and systemic paclitaxel exposure


During HIPEC (0–2 h)

Total sampling period (0 –120 h)

Total period (0–infinity)*

AUC i.p (h × mg/L)

AUCplasma (h × mg/L)

AUCi.p./AUC plasma

AUCi.p (h × mg/L)

AUCplasma (h × mg/L)

AUCi.p./AUC plasma

AUCi.p (h × mg/L)

AUCplasma (h × mg/L)






















AUC, area under drug concentration versus time curve; i.p., intraperitoneal; *with extrapolation of data from 0–120 h.
Fig. 1.

Paclitaxel concentrations in peritoneal fluid (■) and plasma (▴) in patient 9 during and after intraoperative hyperthermic intraperitoneal chemotherapy are displayed as example. Vertical log axis: paclitaxel concentrations in μg/L. Horizontal axis: time after drug administration in hours. The dashed line represents the lowest cytotoxic concentration of paclitaxel, which is 0.1 μmol/L (= 85 μg/L). In this patient, it appears that cytotoxic drug levels are present in the peritoneal cavity for more than 4 days after treatment.

During HIPEC, the exposure to the drug (AUC) was an average of 1462 (± 831, 574–3154) times higher in peritoneal fluid than in plasma (Table 2). When AUCs were calculated for the entire sample collection period (i.e., 5 days), thus also after drainage of the drug solution from the peritoneal cavity, the mean AUC ratio was, although much smaller, still significantly high (average 366 ± 278, 140–1088). When the data were extrapolated to calculate AUCs for infinity, until no drug should have been detectable anymore, the mean AUC ratio was quite similar (387 ± 260, 159–1057). Averaged extrapolations of only 2.81% (0.93–4.61%) and 0.05% (0.03–0.09%) were needed for plasma and peritoneal fluid AUCs, respectively, demonstrating that the total duration of sample collection was adequate.

When the circulating perfusate volume is considered to be constant during 2 h of HIPEC, 59–89% (mean 78%) of the total amount of initially administered paclitaxel was still present in the perfusate at completion of perfusion.


Due to the fact that paclitaxel has a high molecular weight (853.9), nonvesicant properties, and hepatic metabolism, high intraperitoneal to systemic drug concentration and exposure ratios might be expected after intraperitoneal delivery. Since the response to paclitaxel seems to be dose dependent in clinical studies,2629 increased efficacy is to be expected during intraperitoneal chemotherapy with this agent. Moreover, its penetration depth in tumor noduli seems adequate.3032 Prolonged exposure due to slow absorption from the peritoneal cavity seems to result in enhanced tumor penetration. In an in vitro model, paclitaxel penetrated approximately 40 cell layers in 4 h and more than 80 cell layers in 24 h.33 Hence, paclitaxel is theoretically an attractive agent for intraperitoneal chemotherapy, not only because of its observed high efficacy after systemic administration in primary and recurrent ovarian cancer,2,3 but also due to its expected favorable pharmacokinetics after intraperitoneal administration.

Several clinical studies have demonstrated the favorable pharmacokinetics of paclitaxel during intraperitoneal instillation chemotherapy (Table 3).1417,34 Peak intraperitoneal drug levels are in the micromolar range rather than in the nanomolar range as for plasma levels after intravenous administration,1417,33 while cytotoxic drug levels are generally maintained in the peritoneal cavity for several days.13,17 The intraperitoneal chemotherapy exposure advantage is probably best expressed by the maximal concentration ratio and AUC ratio of the drugs from the peritoneal cavity to the peripheral blood. For paclitaxel, the average maximal concentration and AUC ratios are approximately 800–1000 and 550–2000, respectively.1416,34 Probable explanations for this significant variation in mean values are the differences in techniques, drug dose, surfactant vehicles, kind and volume of carrier solution, treatment frequency, patient population, and method of drug concentration measurement. Despite these variations, in all studies this ratio is much higher than that for intraperitoneal administration of other drugs that are frequently used in ovarian cancer such as cisplatin and carboplatin; for these platinum compounds ratios of approximately 5–20 has been calculated.35
Table 3.

Results of clinical pharmacokinetic studies of paclitaxel during intraperitoneal instillation chemotherapy and intraoperative hyperthermic intraperitoneal chemotherapy (HIPEC)



Instillation i.p chemotherapy or HIPEC


Cmax i.p (μM)a

Cmaxi.p./Cmax plasmaa

AUCi.p./AUC plasmaa

Markman et al.14


Instillation i.p. chemotherapy

25–175 mg/m2/3–4 wks




Francis et al.17


Instillation i.p. chemotherapy

20–75 mg/m2/wk



Fushida et al.15


Instillation i.p. chemotherapy

120–180 mg



Hofstra et al.16


Instillation i.p. chemotherapy

75 mg/m2 D1 + 8/4 wk



Mohamed et al.33


Instillation i.p. chemotherapy

20 mg/m2/day for 5 days




Present study



175 mg/m2



1462 / 387f

i.p., intraperitoneal; HIPEC, intraoperative hyperthermic intraperitoneal perfusion chemotherapy; D, day of chemotherapy cycle; a mean values; b for 125–175 mg/m2; c range, not mean values; d equivalent to 101.1 mg/L (Table 1); e for the duration of intraperitoneal instillation chemotherapy (24 h); f the first number is the ratio for the duration of HIPEC (2 h) and the second number the ratio for the entire period until the drug is not detectable anymore.

In the present HIPEC study, the pharmacokinetic data were advantageous, similarly to those of intraperitoneal instillation chemotherapy with paclitaxel, despite its significantly shorter treatment duration (Table 3).1417,34 The maximal intraperitoneal drug concentrations were slightly lower than in some intraperitoneal instillation chemotherapy studies,14,17 probably because of the higher carrier solution volume used in our study to achieve optimal drug exposure through the entire abdominal cavity. A considerable variation in maximal intraperitoneal drug concentrations was observed, which can be partially explained by significant interindividual variation in carrier solution volume and total drug dose. Other contributing factors might have been drug absorption on tubes and reservoirs of the perfusion system and suboptimal sampling time. The peritoneal fluid versus plasma maximal drug concentration ratio as well as AUC ratio during HIPEC were similar to those during 24 h of intraperitoneal instillation chemotherapy (Table 3).1416,34 In the latter studies, AUC was not calculated for the total period in which drug concentrations were detectable. In the present study, the AUC ratio was still considerably high for a total period of 5 days after drug administration. Cytotoxic drug concentrations (>0.1 μmol/L25) were present in the abdominal cavity for many days, despite drainage of the drug solution at the end of the 2-h HIPEC procedure. This is in accordance with observations in two intraperitoneal instillation chemotherapy studies.13,17

Pharmacokinetic data of paclitaxel seem to be more favorable than those in two clinical studies of intraperitoneal instillation chemotherapy with docetaxel and in our previously published HIPEC study.6,36,37 Using the same technique but with a much lower dose (75 mg/m2 docetaxel versus 175 mg/m2 paclitaxel),6 the mean peritoneal fluid versus plasma maximal concentration ratio was 45 for docetaxel versus 1462 for paclitaxel in the present study. The mean peritoneal fluid versus plasma AUC ratios during HIPEC were 207 and 1178, respectively. A smaller difference in mean AUC ratios was observed for the total sample period (207 versus 366). These differences might be mainly caused by the surfactant vehicle used. Taxanes need to be dissolved in these vehicles to overcome their low solubility. The solvent vehicle of paclitaxel is traditionally 4.2% Cremophor EL, while 1.5% Polysorbate-80 is usually used for docetaxel. In a rat model, the absorption rate of taxanes after peritoneal administration was strongly influenced, in a concentration-dependent manner, by the surfactant vehicle used.38 The AUC ratio was three times higher for paclitaxel in comparison to docetaxel, when the conventional vehicles were used. AUC ratios similar to that of paclitaxel were obtained, when docetaxel was dissolved in 4.2% Cremophor EL or 7.5% Polysorbate-80. An experimental clinical study39 confirmed the importance of the surfactant vehicles and demonstrated that Cremophor EL is largely responsible for the pharmacokinetic advantage of intraperitoneal in regards to intravenous administration of paclitaxel. At high local concentrations, paclitaxel is entrapped in Cremophor EL micelles, leading to prolonged intraperitoneal activity. In accordance with these observations, the drug absorption during HIPEC in our studies was significantly lower for paclitaxel than for docetaxel (22% versus 79%).6

The present data are also highly favorable, when compared with pharmacokinetic data of 1- or 3-h intravenous infusion chemotherapy with the same dose of paclitaxel (175 mg/m2).4043 The maximal intraperitoneal drug concentration during HIPEC is 12–30 times higher than the maximal plasma concentration after intravenous chemotherapy. The peritoneal fluid AUC during HIPEC and the total peritoneal fluid AUC are, respectively, 10–18 and 49–59 times higher than the total plasma AUC after such an intravenous administration of paclitaxel. The maximal peritoneal fluid concentration appeared to be approximately 500 times higher than the maximal ascitic fluid concentration after a 6-h intravenous infusion of the same amount of paclitaxel.25

In our institution we advocate HIPEC rather than instillation intraperitoneal chemotherapy for the reasons discussed previously. The duration of HIPEC is generally arbitrarily and varies among centers from 1 to 2 h, while that of instillation intraperitoneal chemotherapy is approximately 24 h. In our institution we perform HIPEC for 2 h. Taking into account pharmacokinetic and thermal enhancement studies, a perfusion duration of 90–120 min seems most adequate.9 Arguments which may be used against the application of HIPEC instead of instillation intraperitoneal chemotherapy is the shorter tumor exposure time. However, in experimental studies we have demonstrated that even short-time (2 h) exposure of tumor cells to high (micromolar) concentrations paclitaxel, as during HIPEC, is sufficient to induce extended cell growth arrest and cell death by necrosis.19,20 Moreover, HIPEC with paclitaxel does not exclude postoperative application of intraperitoneal instillation chemotherapy. Another argument against the use of paclitaxel during HIPEC is the conflicting results that are reported regarding the interaction of heat and paclitaxel. Critics may argue that there has been concern for an antagonistic effect of hyperthermia on the cytotoxic effect of paclitaxel in two in vitro studies.44,45 However, in these as well in two other in vitro studies,46,47 in which a lack of thermal enhancement for paclitaxel was observed, drug concentrations (5 nM–1 μM) and duration of exposure to the drug (4–24 h) and to heat (30–60 min) more closely resembled the conditions of systemic chemotherapy with a period of external heating of the target area. We have previously studied in vitro the cytotoxic effect of drug concentrations of 10–20 μM paclitaxel under mild hyperthermia (41.5°C and 43°C) for 2 h, conditions similar to those during HIPEC.17,18 We as well as others who used similar drug concentrations in vitro48 observed thermal enhancement in some cell lines. Definitely no impairment of paclitaxel cytotoxicity by heat was observed in these in vitro experiments. Furthermore, in three of the four in vivo studies, cytotoxicity of paclitaxel was increased by 60 min of mild hyperthermia.4951 Notably, in the fourth study that demonstrated a lack of thermal enhancement, the duration of hyperthermia was only 30 min.52 In conclusion, it seems that at a high paclitaxel concentration a prolonged period of hyperthermia is associated with enhancement and certainly not with impairment of cytotoxicity. Hence, because of its favorable pharmacokinetic profile and the probable thermal enhancement of its cytotoxicity, paclitaxel seems to be a rather attractive agent for HIPEC.

Various investigators have reported encouraging results regarding cytoreductive surgery and HIPEC for ovarian cancer.23,5361 Currently, a multicentric randomized controlled study is underway to assess its exact contribution in the treatment of persistent and recurrent ovarian cancer.58 Most clinicians have used platinum compounds, with or without doxorubicin, mitoxantrone or mitomycin C, while only most recently have two centers60,61 reported on the administration of paclitaxel. Rufian et al.60 treated 33 patients, 19 as primary treatment and 14 with recurrent disease, with a much lower dose (60 mg/m2) for 60 min, while Bae et al.61 administered 175 mg/m2 paclitaxel for a 90-min HIPEC procedure as secondary treatment in 22 patients. In these series, postoperative surgery-related morbidity was 32–36%. Both groups reported neither grade 3 or 4 hematological toxicity nor peroperative or postoperative mortality. In the present series, data on adverse effects are very similar. Because of the limited systemic toxicity observed in this study, a higher paclitaxel dose might be used in the future. In both studies, HIPEC with paclitaxel seemed to be beneficial concerning survival. In the relatively small retrospective comparative study of Bae et al.,61 HIPEC with paclitaxel appeared to be probably more efficient than HIPEC with carboplatin and was definitely more effective than systemic chemotherapy only. In the present study, the group of patients was too heterogeneous and the follow-up period too short for any adequate conclusion regarding outcome.

In conclusion, the use of paclitaxel for HIPEC following cytoreductive surgery seems feasible and relatively safe, with acceptable morbidity, in patients with primary and secondary peritoneal tumors. Its administration is associated with a highly favorable pharmacokinetic profile, despite HIPEC’s short treatment duration. Very high local drug exposure, which is approximately 50–60 times higher than achieved after intravenous administration, and low systemic drug levels make paclitaxel a very attractive drug for HIPEC. Larger studies with a more homogeneous patient cohort and prolonged follow-up should be performed to demonstrate its definite efficacy.

Copyright information

© Society of Surgical Oncology 2008