Introduction

Peritoneal metastases (PM) are a frequent occurrence in patients with colorectal cancer, affecting about 10–15% of patients and severely diminishing their prognosis.1,2,3,–4 These patients can be treated with cytoreductive surgery with or without hyperthermic intraperitoneal chemotherapy (CRS and HIPEC), or with palliative systemic chemotherapy. Intraperitoneal therapies have been developed since most patients do not qualify for CRS and HIPEC, and palliative systemic chemotherapy is considered less effective against PM as compared to liver or pulmonary metastases.1,5,6

Pressurized intraperitoneal aerosol chemotherapy (PIPAC) is a recently developed intraperitoneal therapy. With this technique, a low dose of chemotherapeutic agent is aerosolized into the abdominal cavity by a laparoscopic approach, theoretically leading to: (1) high intraperitoneal concentrations, but low systemic concentrations and low systemic toxicity, (2) a homogeneous distribution, and (3) deep local tissue penetration.7,8,9,10,11,12,13,14,–15 The addition of electrostatic precipitation to the aerosol could result in a greater tissue penetration of the chemotherapeutic agent (ePIPAC).16,17,–18

During (e)PIPAC for colorectal cancer, current protocols typically use oxaliplatin at a dosage of 92 mg/m2 with a simultaneous intravenous bolus of 5-fluorouracil (400 mg/m2) with leucovorin (20 mg/m2), which is considerably lower than the oxaliplatin dosage used during HIPEC (460 mg/m2).19,20,21,–22

HIPEC with oxaliplatin results in high concentrations of oxaliplatin in peritoneal fluid and tumor tissue, with relatively limited systemic concentrations.20,23,24,25,–26 Nevertheless, the pharmacokinetics of oxaliplatin administered by PIPAC have only been investigated in one animal study, and not yet in humans.15

The CRC-PIPAC trial was performed to investigate the feasibility, safety, efficacy, systemic pharmacokinetics, quality of life, and costs of repetitive ePIPAC with oxaliplatin in patients with PM from colorectal cancer.27 The current manuscript aimed to analyze the systemic pharmacokinetics of oxaliplatin in plasma, plasma ultrafiltrate, peritoneal fluid, and urine after intraperitoneal administration by ePIPAC.

Materials and Methods

A prospective, single-arm, open-label, phase II study was performed in two Dutch HIPEC centers. The Central Medical Ethics Committee (MEC-U, Nieuwegein, Netherlands, R17.038) and institutional review boards of the participating institutions approved of the study, and the principles of the Declaration of Helsinki were followed. The study protocol has recently been published.27 Therefore, we will only provide a brief summary.

Patient Population

Patients were eligible to participate if they were older than 18 years with a WHO performance status of 0–1, and were diagnosed with unresectable colorectal/appendiceal PM without systemic metastases, did not present with symptoms of obstruction, had adequate organ functions, did not have any contra-indications for laparoscopy or chemotherapy, and no previous (e)PIPACs. All patients provided written informed consent.

ePIPAC Procedure

Patients were treated with repetitive ePIPAC under general anesthesia with oxaliplatin (92 mg/m2) and a simultaneous intravenous bolus of leucovorin (20 mg/m2 in 10 min) and 5-fluorouracil (400 mg/m2 in 15 min) every 6 weeks, and did not receive systemic therapy in between subsequent ePIPAC procedures. The intravenous bolus of leucovorin and 5-fluorouracil is infused during laparoscopic surgery, after which the intraperitoneal injection of oxaliplatin is subsequently started. Oxaliplatin was prepared in a total volume of 150 mL dextrose solution and injected through the nebulizer (CapnoPen, Capnomed GmbH, Villingendorf, Germany) in 5 min, after which the Ultravision generator (Ultravision, Alesi Surgical, Cardiff, UK) administered electrostatic precipitation to the aerosol. The electrostatic field and the 12 mm Hg capnoperitoneum were then maintained at 37 °C for another 25 min. A detailed description of the general PIPAC procedure has also been reported by Solass et al.7 and Giger-Pabst et al.28

Trial treatment could have been terminated due to disease progression, toxicity, at the patient’s request, or at a physician’s discretion. For a complete description of the ePIPAC procedure and evaluation schedule, we refer to the study protocol.27

Sample Collection and Analysis

We collected samples of whole blood, urine, and peritoneal fluid at multiple moments after each ePIPAC. Sampling was performed at each procedure to study any potential accumulation and intra-individual variation. All samples were stored at − 80 °C until analysis.

Whole blood was collected at t = 0, 5, 10, 20, 30, 60, 120, 240, 360, and 1080 min after the start of injection of oxaliplatin. Whole blood samples were immediately cooled on ice and centrifuged for 5 min at 2000×g at 4 °C. One ml from the obtained plasma was loaded on an Ultrafree Millipore membrane (Merck Millipore Ltd., Tullagreen, Carrigtwohill, Co. Cork, Ireland) and ultrafiltrated for 20 min at 2000×g at 4 °C to obtain plasma ultrafiltrate (free fraction of oxaliplatin).

Single urine samples were collected pre-operatively and at t = 1, 3, 5, and 7 days after the injection of oxaliplatin; peritoneal fluid was collected at t = 30 min after the injection of oxaliplatin.

Oxaliplatin concentrations were measured using atomic absorption spectrometry on a Thermo Fisher Solaar ICE 3500 graphite-furnace spectrophotometer with Zeeman correction (Thermo Fisher Scientific, Bremen, Germany).

Pharmacokinetic Analysis

Pharmacokinetic parameters were estimated using two differently developed population pharmacokinetic models based on the oxaliplatin plasma and plasma ultrafiltrate concentration–time data, respectively, by non-linear mixed-effects modeling using NONMEM (V7.4, Icon Development Solutions, Ellicot City, MD, USA). The Perl-speaks-NONMEM toolkit version 4.7.0 and Pirana version 2.9.7 were used as modeling environment. Results were plotted using R statistics (v3.4.4, Boston, MA, USA) and RStudio (v1.1.453). A first-order conditional estimation method with interaction was used throughout the analysis. One, two and three compartmental pharmacokinetic models with first order elimination were compared with the observed oxaliplatin plasma and plasma ultrafiltrate concentration–time data to find the optimal fit. Model selection was based on statistical significance, goodness of fit and stability. Throughout the model building process, an altered model was chosen over a pre-cursor model if the difference in the objective functions (− 2 log likelihood) was > 6.63 (P < 0.01, with 1 degree of freedom, assuming χ2 distribution). The final population pharmacokinetic model was validated by means of a prediction corrected visual predictive check (500 runs). Subsequently, exposure (AUC0–24 and AUC0–48) was calculated using the individual predicted values (concentration) and time data using the trapezoidal rule with R statistics.

Results

Patient Characteristics

Between October 2017 and April 2019, twenty patients were treated with repetitive ePIPAC: four (20%) patients only received one ePIPAC, three (15%) patients received two ePIPAC, and thirteen (65%) patients received three or more ePIPAC. All patients received the intended dose of chemotherapeutic agents. No dose reductions were required due to systemic toxicity, including peripheral neuropathy. Twelve (60%) patients were male and eight (40%) patients were female. The median age was 63.5 years. Most patients had synchronous peritoneal metastases (n = 15, 75%) from a right-sided (n = 14, 70%) colorectal adenocarcinoma that had not been resected (n = 14, 70%). Nine patients (45%) had been treated with palliative systemic chemotherapy before study inclusion, of whom eight patients received an oxaliplatin-containing regimen, and nine patients (45%) had undergone an explorative laparotomy with the intention of CRS and HIPEC, which all resulted in an open-close procedure. The baseline characteristics are provided in Table 1.

Table 1 Baseline characteristics
Full size table

Pharmacokinetics of Plasma and Plasma Ultrafiltrate

Oxaliplatin pharmacokinetics in plasma were best described by a 2-compartment model with first order absorption from the peritoneal compartment with first order elimination, as shown in supplementary Fig. 1. Oxaliplatin pharmacokinetics in the ultrafiltrate were best described by a 3-compartment model with first order absorption from the peritoneal compartment with first order elimination with allometric scaling. Random effect parameters for inter-individual variability in absorption, clearance, distribution volume of the central compartment and the peripheral compartment, and intercompartmental clearance were identified. The concentration–time profiles of oxaliplatin in plasma and plasma ultrafiltrate are provided in Figs. 1 and 2, respectively. Both models performed well, although the plasma model proved to be slightly more accurate than the plasma UF model due to denser sampling (goodness of fit plots as supplementary Figs. 2 and 3).

Fig. 1
figure 1

Prediction corrected visual predictive check for the concentration of oxaliplatin in plasma. The red line represents the median and the blue lines represent the 10th and 90th observation percentiles. The observed oxaliplatin concentrations are shown as open circles. The shaded areas represent the 95% confidence interval around each of the prediction percentiles

Full size image
Fig. 2
figure 2

Prediction corrected visual predictive check for the concentration of oxaliplatin in plasma ultrafiltrate. The red line represents the median and the blue lines represent the 10th and 90th percentiles. The observed oxaliplatin concentrations are shown as open circles. The shaded areas represent the 95% confidence interval around each of the prediction percentiles

Full size image
Fig. 3
figure 3

Concentration of oxaliplatin (µg/mL) in urine during the first week after injection of oxaliplatin

Full size image

Table 2 provides the calculated estimates of the pharmacokinetic parameters of oxaliplatin in plasma and plasma ultrafiltrate (UF) after the first, second and third ePIPAC. The maximum concentration of oxaliplatin (Cmax) in plasma UF and in plasma varied between 1.36 and 1.90 µg/mL and between 2.67 and 3.28 µg/mL, respectively. The maximum concentrations of oxaliplatin in plasma UF and in plasma were reached after approximately 30 and 90 min (Tmax), respectively. The median plasma UF AUC0–24 h varied between 9.6 and 11.7 µg/mL * h. The median plasma AUC0–24 h varied between 49.0 and 59.5 µg/mL * h, and the median plasma AUC0–48 h varied between 95.4 and 114.9 µg/mL * h.

Table 2 Pharmacokinetics of oxaliplatin in plasma and plasma ultrafiltrate
Full size table

The Cmax, the AUC0–24 h, and the AUC0–48 h of oxaliplatin in both plasma UF and plasma increased significantly from the first to the second and third ePIPAC.

Finally, the absorption rate constant (Ka) was 1.13 (1.04–1.30) L/h.

Pharmacokinetics of Urine and Peritoneal Fluid

Figure 3 shows the concentration of oxaliplatin in urine during the first week after each ePIPAC. Concentrations were highest on the first day after ePIPAC after which a rapid decrease was observed. At day 3, 90% of the urine samples were below 8.70 µg/mL; at day 5, 90% of the samples were below 4.90 µg/mL; at day 7, 90% of the samples were below 3.60 µg/mL. With the applied method of analysis (AAS), no accumulation of oxaliplatin could be detected between the various ePIPAC procedures.

The concentrations of oxaliplatin in peritoneal fluid 30 min after the start of the injection are shown in Fig. 4. The highest concentration was found after the first ePIPAC, although the differences in concentrations after the second and third ePIPAC were not statistically significant (p = 0.38).

Fig. 4
figure 4

Concentration of oxaliplatin (µg/mL) in peritoneal fluid, 30 min after injection of oxaliplatin

Full size image

Discussion

This study aimed to describe the systemic pharmacokinetics of oxaliplatin administered by ePIPAC, and was the first to do so in humans. ePIPAC with oxaliplatin (92 mg/m2) resulted in significant systemic concentrations. After 7 days, the excretion of oxaliplatin in urine had decreased to less than 3.60 µg/mL in 90% of the samples. No accumulation of oxaliplatin was observed between the ePIPAC-procedures.

Remarkably, the Cmax and AUC of oxaliplatin of plasma and plasma UF increased significantly during the second and the third ePIPAC. This increase was an unexpected finding, and might potentially be due to the fact that the extent of peritoneal disease decreased during treatment, thereby obtaining a greater “healthy” surface area available for systemic drug absorption, subsequently leading to a higher AUC. However, this is only a hypothesis, and this finding of increased systemic exposure with an increasing number of PIPAC will have to be confirmed in another pharmacokinetic study.

Oxaliplatin has been part of systemic chemotherapy protocols and has been used as a HIPEC drug for several decades. The pharmacokinetics of oxaliplatin after systemic administration or intraperitoneal administration by HIPEC have been described by several research groups.

With systemic chemotherapy, oxaliplatin is administered in a dose of 130 mg/m2 or of 85 mg/m2, depending on the combination with either capecitabine (CAPOX/XELOX) in a 3-weekly schedule or 5-fluorouracil/leucovorin (FOLFOX) in a 2-weekly schedule, respectively. Since our ePIPAC regimen with oxaliplatin (92 mg/m2) and an intravenous bolus of 5-fluorouracil/leucovorin (400/20 mg/m2) resembles the oxaliplatin component of a systemic cycle of FOLFOX, we compared our pharmacokinetic results with the pharmacokinetics of systemic chemotherapy with FOLFOX as described by other studies.29,30 After intravenous administration of 85 mg/m2 oxaliplatin, these studies described a Cmax of 1.61–1.92 µg/mL in plasma and of 0.38–0.68 µg/mL in plasma UF, which is slightly lower than the Cmax of oxaliplatin after ePIPAC found in our study (2.67–3.28 µg/mL and 1.36–1.90 µg/mL, respectively). The most likely explanation for the lower Cmax of oxaliplatin after intravenous administration is that it is administered over a much longer time, up to 2–6 h, while the complete dose of oxaliplatin during ePIPAC is administered in less than 5 min. Another explanation for this difference could be the slightly higher dose of oxaliplatin used during ePIPAC, and the application of pressure and electrostatic precipitation during ePIPAC, possibly increasing absorption.9

Remarkably, the plasma AUC of oxaliplatin was comparable after intravenous administration and after ePIPAC (51.4–118.0 µg/mL vs. 49.0–59.5 µg/mL, respectively), whereas the plasma UF AUC of oxaliplatin was lower after intravenous administration, compared with other literature (4.3–4.8 µg/mL vs. 9.6–11.7 µg/mL, respectively).29,30

A much higher dose of oxaliplatin is used during HIPEC, varying from 360 to 460 mg/m2.24,26,31 The Cmax in plasma and plasma UF following administration by HIPEC varied from 10.7 to 14.0 µg/mL and 7.0 to 8.5 µg/mL, respectively, which are roughly fivefold higher than the maximum concentrations of oxaliplatin after ePIPAC in our study.24,26,31 This could be explained by the fivefold lower dose of oxaliplatin of 92 mg/m2 that was used during ePIPAC.

In these studies, maximum concentrations of both plasma and plasma UF are reached after approximately 30 min (Tmax). In our study, the plasma UF Tmax was 30 min as well, although the plasma Tmax was 90 min. This could partially be explained by the removal of oxaliplatin after 30 min of HIPEC, whereas oxaliplatin remains in the abdomen after ePIPAC, allowing further absorption.

In contrast to the Cmax, the AUC of plasma UF after ePIPAC is almost equal to the AUC of plasma UF after HIPEC (9.6–11.7 µg/mL * h and 13.7–14.8 µg/mL * h, respectively).24,26,31 Although not the goal of ePIPAC, this suggests that equal levels of systemic exposure can be achieved with ePIPAC, and further justifies the fivefold lower dose used during ePIPAC as compared to HIPEC. Possible explanations are the application of pressure and electrostatic precipitation during ePIPAC, and the removal of chemotherapy from the abdominal cavity after 30 min of HIPEC, whereas oxaliplatin remains in the abdominal cavity after ePIPAC.

The absorption rate constant (Ka) during ePIPAC was lower than during HIPEC (1.13/h vs. 1.40–1.42/h, respectively), which may be explained by the lower absolute dose of oxaliplatin used during ePIPAC as rate of absorption is dose-dependent with 1st order pharmacokinetics.24,26

Eveno et al. presented in-animal results of the pharmacokinetics of oxaliplatin after intravenous administration (5 mg/mL) and after PIPAC (0.028 mg/mL), resulting in a systemic concentration of 0.089 µg/mL and 0.019 µg/mL 7 days after administration of oxaliplatin, respectively.15 They found higher concentrations of oxaliplatin after intravenous administration than after PIPAC (p = 0.008). Unfortunately, we were not able to compare our results with Eveno et al., because their study was conducted in mice and therefore used different dosages of oxaliplatin. Also, in their study, the oxaliplatin concentration was only measured on one occasion after 7 days, which does not allow a complete pharmacokinetic analysis as was performed in our study.

Limitations

This study has several limitations. First of all, it is limited by a small sample size and a non-controlled design, disabling direct comparison with the systemic pharmacokinetics of oxaliplatin after intravenous administration and after HIPEC. In addition to that, we were not able to accurately calculate the clearance of oxaliplatin from plasma due to too short sampling. The last sampling occasion was 18 h post-ePIPAC, after which the absorption and distribution phase proved to be not totally finished yet. This does not allow a proper estimation of the elimination phase—a final sample after, e.g., 48 h would have been required. Nonetheless, clearance and elimination rates are not expected to be different following administration by ePIPAC, HIPEC or intravenous infusion. As shown in the goodness of fit plots, the model for plasma is more accurate than the model for plasma UF, which is probably due to fewer plasma UF samples—we were not able to obtain sufficient ultrafiltrated plasma from some samples.

The sampling of peritoneal fluid was only performed once, 30 min after the start of the injection of oxaliplatin. Although this single measurement confirmed that oxaliplatin remained in the abdomen after the procedure and was thus still able to exert its anti-tumor effects, this measurement alone did not allow us to describe the AUC ratio of intraperitoneal and plasma oxaliplatin as has been described for various other drugs administered by HIPEC.32

Future research could elaborate on the direct comparison of the systemic absorption of oxaliplatin administered by ePIPAC and by systemic chemotherapy, allowing determination of the absolute bioavailability.

Conclusion

To our knowledge, this is the first study to investigate the systemic pharmacokinetics of oxaliplatin administered by ePIPAC in humans. In conclusion, the systemic absorption of oxaliplatin administered by ePIPAC was higher than expected, reaching concentrations equal to those reached with systemic chemotherapy, as reported in other studies. This finding does not support one of the proclaimed benefits of ePIPAC: a decreased systemic exposure and potentially less toxic profile compared with systemic administration. Additional sampling at 24 and/or 48 h after administration of ePIPAC could have increased the accuracy of the oxaliplatin plasma AUC and clearance. Oxaliplatin was mainly excreted during the first 7 days and no accumulation was observed. This allows repeated treatment with ePIPAC at 6-week intervals and potentially also more frequently without further increasing toxicity.