Exploration of Daptomycin Adsorption by Polymethylmethacrylate Hemofilter In Vitro

Summary

Blood purification therapy with cytokine-adsorbing hemofilters has been used to treat sepsis-associated hypercytokinemia. Polymethylmethacrylate (PMMA) hemofilters are frequently used for this purpose; however, adsorption and removal of teicoplanin, a therapeutic agent, have been reported. Similar concerns have been shared regarding daptomycin because its structure resembles that of teicoplanin; nevertheless, there have been no reported effects associated with daptomycin in this context. We studied the adsorption of daptomycin onto a PMMA hemofilter in vitro and investigated its adsorption onto hollow fiber membranes by adding cut PMMA membranes to a daptomycin solution. Additionally, the daptomycin solution was circulated in a dialysis circuit connected to a PMMA hemofilter, and changes in daptomycin content were examined. The daptomycin content decreased immediately after adding the hollow fiber membranes, similar to that observed for teicoplanin. The daptomycin content was lower than that of the standard reagent in the dialysis circuit model, reaching values below the measurement limit after 20 min. These results suggested that daptomycin was adsorbed and removed by the PMMA hemofilter. Encountering this effect during clinical use is plausible; therefore, daptomycin administration via a PMMA hemofilter should be avoided during blood purification therapy.

INTRODUCTION

Since the publication of Sepsis-3 in 2016, treating sepsis and septic shock has become important not only to control infection with antimicrobial agents but also to improve hypercytokinemia, which is believed to cause organ damage.¹⁾ Hirasawa²⁾ emphasized the importance of addressing hypercytokinemia and reported the effectiveness of cytokine removal through blood purification therapy, such as continuous hemodiafiltration (CHDF). They also recommended using cytokine-adsorbing hemofilters consisting of AN69ST or polymethylmethacrylate (PMMA) membranes, which have excellent cytokine adsorption.³⁾ The frequent utilization of the PMMA hemofilter stems from its availability on the market for continuous renal replacement therapy since 1991, supported by numerous reports highlighting its clinical usefulness.^4,5)

PMMA membranes exhibit hydrophobic properties owing to the absence of polyvinylpyrrolidone, a conventional hydrophilic agent used to enhance biocompatibility.^6,7) The pores of PMMA membranes possess a uniform symmetrical structure with sizes conducive to effectively trapping cytokines. These structural characteristics facilitate the adsorption and subsequent removal of cytokines.

However, performing CHDF using a PMMA hemofilter (PMMA–CHDF) has inherent risks. One such concern is the possibility of removing the therapeutic agents during dialysis. Unlike conventional hemofilters that rely on membranes for diffusion and filtration and for predicting drug behavior based on molecular weight, it is challenging to anticipate the adsorptive properties of PMMA hemofilters. An in vitro study reported that vancomycin hydrochloride (VCM) and teicoplanin (TEIC), which are both anti-methicillin-resistant Staphylococcus aureus (anti-MRSA) agents, possess similar molecular weights; however, TEIC exhibits a significantly higher adsorption rate onto PMMA membranes than VCM.⁸⁾ Furthermore, it has been reported that the blood concentration of TEIC decreases due to its adsorption on PMMA membranes during clinical application.⁹⁾

Monitoring TEIC levels is essential. Hence, dosage adjustments can be made by periodically measuring the blood concentrations during PMMA–CHDF. However, the blood concentrations of several drugs were not monitored, posing challenges in adjusting their dosages during the PMMA–CHDF administration. Daptomycin (DAP), an anti-MRSA agent, belongs to this category. DAP shares several similarities with TEIC in terms of molecular weight, structure, and protein binding rate. Hence, there are concerns regarding the adsorption of DAP onto PMMA membranes; however, to date, there have been no reported effects.

This study aims to confirm the effects of DAP adsorption on PMMA hemofilters. Furthermore, we investigated the effects of DAP administration during blood purification therapy by using a PMMA hemofilter.

MATERIALS AND METHODS

Materials

The target drug utilized was the pharmaceutical DAP (CUBICIN^® IV 350 mg; MSD Co., Ltd., Tokyo; molecular weight: 1620.67). Comparator drugs included TEIC (TEICOPLANIN, intravenous for drip use; 200 mg; Fuji Pharma Co., Ltd., Tokyo; molecular weight: 1564.25–1893.68) and VCM (VANCOMYCIN HYDROCHLORIDE for IV Infusion [MEIJI]; 0.5 g; Meiji Seika Pharma Co., Ltd., Tokyo; molecular weight: 1487.71). Cyanocobalamin (CN-Cbl; FUJIFILM Wako Pure Chemical Co., Osaka; molecular weight: 1355.37) was used as the standard reagent for experiments using a dialysis circuit model. During preliminary tests, no adsorption onto the PMMA membranes was observed. Despite its structural disparity with DAP, CN-Cbl has a similar molecular weight. The PMMA hemofilter employed was a Hemofeel^® CH-1.8 W (Toray Medical Co., Ltd., Tokyo). The hollow fiber membranes were extracted by disassembling the hemofilter housings. A blood purification system (AcuFil Multi 55XII^®; Toray Medical Co., Ltd.) was used for dialysis circuit model experiments.

Experiment Using Hollow Fiber Membranes

Each drug was diluted with saline to a molar concentration of 60–80 µmol/L. DAP and TEIC were prepared at a concentration of 120 µg/mL, while VCM was prepared at a concentration of 100 µg/mL. The PMMA membranes (approximately 0.08 m² in surface area) were added to 200 mL of each drug solution and agitated at 37°C [Fig. 1(A)].¹⁰⁾ Drug solution samples were collected before and at intervals of 1, 5, 10, 15, 20, 30, and 45 min after introduction of the PMMA membrane. Absorbance readings were obtained at the respective absorption wavelengths of each drug using a spectrophotometer (UVmini-1240, Shimadzu Corporation, Kyoto). Control measurements were conducted over time in the absence of a hollow fiber membrane. The measurement wavelengths of TEIC and VCM were set to 279 nm¹¹⁾ and 280 nm,¹²⁾ respectively, whereas that of DAP was set to 261 nm by measuring the absorption spectrum based on the disclosed information.¹³⁾ The drug content was assessed using concentrations derived from the measured absorbance. The adsorption rates were calculated from drug concentrations at specific time points and compared. The experiment was performed four times for each drug, and the values are expressed as the mean±standard deviation.

Fig. 1. Devices Used in the Experiment

(A) Device used in the hollow fiber membrane experiment. The hollow fiber membranes were cut into polymethyl methacrylate (PMMA) membranes. (B) Device for dialysis circuit model experiment. Subsequently, a PMMA hemofilter was connected. Black arrows indicate the flow of the solution. Black stars indicate sampling points.

Experiment Using the Dialysis Circuit Model

Priming was performed using heparin-containing saline (5000 U/L). Drug concentrations were prepared with DAP at 120 µg/mL and the standard reagent CN-Cbl at 100 µg/mL to match the molar concentrations. The drug solutions (1000 mL) were circulated at a solution flow rate (Q_B) of 80 mL/min and maintained at 37°C. Saline was used as dialysate. The dialysate (Q_D) and effluent flow rates (Q_E) were set at 500 mL/h to eliminate the effects of ultrafiltration [Fig. 1(B)].¹⁰⁾ The drug solutions were collected at 0, 5, 10, 20, 30, 60, and 120 min after circulation using a dialysis circuit model, and the absorbance was measured. The measurement wavelengths of DAP and CN-Cbl were set to 261 nm¹³⁾ and 361 nm,¹⁴⁾ respectively. The drug concentrations were determined based on the measured absorbance. The beginning of the experiment was set to the time the drug solution replaced the priming solution in the circuit, and 0 min was set after a 150 s delay. Furthermore, the drug solutions were collected at 5 and 60 min at the inlet and outlet of the hemofilter, and their concentrations were measured to calculate DAP clearance.

Calculations

Drug concentrations were calculated based on absorbance measured using a calibration curve. The calibration curves showed linearity for all drugs (r²≧0.9997) and were reproducible with our previous study.¹⁰⁾

The drug content was calculated using the following Eq. (1).

  

(1)

The drug adsorption rates were calculated using the following Eq. (2).

  

(2)

where C₀ is the initial concentration of the drug solution, and C_t is the concentration of the drug solution collected over time.

The drug clearance was calculated using the following Eq. (3).

  

(3)

where C_in and C_out are the concentrations of the drug solutions collected at the inlet and outlet of the hemofilter, respectively, and Q_B is the flow rate of the solution.

Statistical Analysis

To evaluate the adsorption onto the PMMA membrane, Tukey’s test was performed using a multiple comparison method for each drug. Statistical analysis was performed using Rcmdr Plugin within EZR (Easy R).¹⁵⁾ p<0.01 denoted statistical significance.

RESULTS

Experiment Using Hollow Fiber Membranes

The change in drug content after adding the PMMA membrane is shown in Fig. 2(A). The DAP content decreased rapidly and was 55.4±1.7% at 1 min. It continued to decrease and reached 21.0±1.6% at 5 min and 4.6±6.0% at 10 min. It finally decreased below the measurement limit after 15 min. The TEIC content also decreased rapidly to 26.0±1.0% at 10 min. Thereafter, the decrease slowed and was only 17.8±0.9% at 45 min. In contrast, the VCM content remained nearly unchanged at 97.5±1.6% at 45 min. The control doses for each drug without adding hollow fiber membranes showed no decrease in drug content.

Fig. 2. Representative Example of the PMMA Membrane Experiments

(A) Changes in anti-MRSA agent content after PMMA membrane exposure. ●: Daptomycin; ■: teicoplanin; ▲: vancomycin; ○: daptomycin without membrane; □: teicoplanin without membrane; △: vancomycin without membrane. ∗: Unable to measure because the level was below the measurement limit. The measurement limit for DAP concentration was determined to be approximately 3 µg/mL from the noise level of the spectrophotometer. All data are expressed as mean±S.D. (n=4). (B) The rates of adsorption of anti-MRSA agents onto PMMA membranes after a 10-min period were evaluated. A comparison was made between the drugs, and the results were analyzed using Tukey’s test.

The adsorption rate of each drug at 10 min is illustrated in Fig. 2(B). The adsorption rates of DAP, TEIC, and VCM were 95.4±6.0%, 74.0±1.0%, and 3.1±1.3%, respectively. Tukey’s test for each drug showed that DAP and TEIC adsorption rates were significantly different from those of VCM (p<0.01). The adsorption rate of DAP was also significantly different from that of TEIC (p<0.01).

Experiment Using the Dialysis Circuit Model

The changes in the content of each drug after circulation in the dialysis circuit model are shown in Fig. 3. The DAP and CN-Cbl contents decreased by approximately 20% at 0 min owing to the dilution of the priming solution in the circuit. DAP content decreased rapidly, reaching 56.2% at 5 min, 38.9% at 10 min, and below the measurement limit after 20 min. In contrast, CN-Cbl slowly decreased to 35.6% after 120 min.

Fig. 3. Changes in the Content of Daptomycin in the in vitro Dialysis Circuit Model

●: Daptomycin; ○: cyanocobalamin (standard reagent). ∗: Unable to measure because the level was below the measurement limit. The measurement limit for DAP concentration was determined to be approximately 3 µg/mL from the noise level of the spectrophotometer. The time of commencement (0 min) was set after a 150-s delay to allow the circuit to be filled with the drug solution.

The clearance of DAP in the dialysis circuit model is presented in Table 1. DAP clearance was established at 80 mL/min at 5 min because the outlet concentration at 5 min, and the inlet and outlet concentrations at 60 min fell below the measurement limits. The clearance of CN-Cbl was 11.28 mL/min at 5 min and 4.87 mL/min at 60 min, which were lower than those of DAP.

Table 1. Clearance of Daptomycin in the in vitro Dialysis Circuit Model Using PMMA Hemofilter

Drugs	Clearance (mL/min)
	5 min	60 min
Daptomycin	80.00	N.D.
Cyanocobalamin	11.28	4.87

The tentative determination of DAP clearance was set at 5 min, because DAP concentrations were below the measurement limits at the inlet at both 5 and 60 min. The assessment at the 60-min mark could not be performed because of the same rationale. The measurement limit for DAP concentration was determined to be approximately 3 µg/mL from the noise level of the spectrophotometer. Cyanocobalamin was used as the standard reagent.

DISCUSSION

In this study, we examined the in vitro adsorption of DAP, an anti-MRSA drug, onto PMMA hemofilters. Initially, we verified the adsorption of DAP onto the hollow fiber membranes. Our findings indicated an immediate decrease in both DAP and TEIC content upon the addition of the PMMA membrane, which was notably lower than that of VCM. Sawada et al.¹⁶⁾ reported that the TEIC concentration decreased owing to adsorption immediately after the addition of PMMA membranes to the TEIC solution containing human serum albumin. The reduction in the TEIC concentration observed in our study could be attributed to adsorption onto the PMMA membranes. This suggests that DAP, which showed a similar decrease in concentration to TEIC, was also adsorbed onto the PMMA membranes. Additionally, we inferred that DAP was more readily adsorbed onto the PMMA membranes than TEIC because the DAP concentration was significantly lower than that of TEIC.

Next, we investigated the effect of DAP adsorption using a dialysis circuit model connected to a PMMA hemofilter. The principles of dialysis include diffusion, filtration, and adsorption¹⁷⁾; however, this study primarily addressed adsorption and omitted the effects of ultrafiltration. Therefore, we hypothesized that the decrease in the concentration of CN-Cbl that was not adsorbed on the PMMA membrane occurred due to removal via diffusion and that DAP was removed via adsorption when its concentration decreased below that of CN-Cbl. Based on this, we compared the change in DAP content, revealing that the DAP content significantly decreased compared to that of CN-Cbl after treatment initiation. These findings imply that both diffusion and adsorption contribute to the removal of DAP from the PMMA hemofilter. Yamamoto et al.¹⁸⁾ reported that when the dialysate flow rate is significantly lower than the blood flow rate, it can be equated to the theoretical diffusion clearance. In our study, the dialysate flow rate was 500 mL/h (8.3 mL/min), which is markedly lower than the blood flow rate (80 mL/min). Therefore, the theoretical diffusion clearance was determined to be 8.3 mL/min. By applying this value to the clearance of DAP at 5 min, the ratio of adsorption to diffusion was calculated, revealing an adsorption clearance approximately 8.6 times greater than diffusion clearance. Consequently, we inferred that the elimination of DAP by the PMMA hemofilter primarily depends on adsorption rather than diffusion.

The mechanism of daptomycin adsorption onto PMMA hemofilter remains to be clarified. However, it is known that cytokine adsorption occurs as a function of hydrophobic interaction and entrapment within the pores.⁶⁾ The pore size of the PMMA membrane is designed to target interleukin-6.⁶⁾ Moriyama et al.¹⁹⁾ reported that tumor necrosis factor-α, interleukin-6, and interleukin-8 with molecular weights of 17000, 21000, and 8000, respectively, circulated in a closed circulation system connected to the PMMA hemofilter, and all of them showed higher clearance than the filtration flow rate. This indicates that the PMMA hemofilter can adsorb and remove cytokines with a wide range of molecular weights owing to the strong adsorption effect of hydrophobic bonding. DAP, a cyclic lipopeptide anti-MRSA drug, has a molecular weight of approximately 1600 and contains 13 amino acid residues (Fig. 4). These include several hydrophobic amino acid residues. Accordingly, we assumed that DAP was adsorbed and removed from the PMMA hemofilter because the hydrophobic amino acid residues were bound hydrophobically to the hydrophobic PMMA membrane.

Fig. 4. Structural Formula of Daptomycin

Finally, we evaluated the adsorption capacity of DAP on PMMA membranes. Results from the hollow fiber membrane experiments indicated an adsorption capacity of at least 24 mg of DAP per 0.08 m² of PMMA membrane. Using this value, the adsorption capacity of DAP on a PMMA hemofilter (1.8 m²) was estimated to be approximately 540 mg. Conversely, in the dialysis circuit experiment, the adsorption capacity of DAP on the PMMA hemofilter was approximately 172 mg, calculated after 20 min, when it fell below the measurement limit. Discrepancies in the calculated adsorption capacity of DAP between experiments arose. Several possible factors were that DAP remained below the measurement limit in all experiments, and the entire hollow fiber membrane was exposed to DAP in the hollow fiber membrane experiments. Despite the challenges in precisely evaluating the adsorption capacity of DAP on PMMA membranes in this study, it was evident that a significant amount of DAP was adsorbed onto the PMMA membrane, which is crucial for the appropriate use of DAP.

Despite these key findings, this study has several limitations. First, more than 90% of DAP is protein bound in the blood.²⁰⁾ However, this study did not include albumin or other substances, and protein-binding was not assumed. Therefore, the DAP adsorption rate may fluctuate in real-world clinical practice. TEIC, which also exhibits high protein binding, has been reported to undergo adsorption in the clinical setting. Therefore, the greater adsorption of DAP compared with TEIC observed in this study should be considered in clinical practice. Second, the sample size was small, with only one sample in the dialysis circuit model. Further studies with larger sample sizes are required to confirm the reproducibility of our results. Further investigation is warranted to explore these issues and their implications in clinical practice.

Regarding DAP pharmacokinetics during continuous renal replacement therapy, Taguchi et al.²¹⁾ reported that DAP clearance did not increase with polysulfone or cellulose triacetate membranes. However, in this study involving PMMA membranes, DAP clearance was significantly enhanced by the adsorption and removal of DAP from PMMA membranes, albeit in vitro. Moreover, these effects are likely to be manifested in clinical applications. Consequently, we advocate avoiding the administration of DAP using PMMA membranes during blood purification therapy. Instead, measures such as transitioning to alternative anti-MRSA drugs or using membranes that do not adsorb DAP are warranted.

Acknowledgements

We sincerely thank Tomoaki Hashida, Director of the Emergency Medical Center, and Ryota Sakamoto, Department of Clinical Engineering, Eastern Chiba Medical Center, for their cooperation. This study was supported by JSPS KAKENHI (Grant Number 21K06686).

Conflict of Interest

The authors declare no conflict of interest.

REFERENCES

• 1) Hashida T., Oshima T., Nakada T., J. Jpn. Soc. Blood Purif. Crit. Care, 12, 45–51 (2021).
• 2) Hirasawa H., J. Jpn. Soc. Blood Purif. Crit. Care, 2, 143–151 (2011).
• 3) Hirasawa H., Oda S., Nakamura M., Watanabe E., Shiga H., Matsuda K., Blood Purif., 34, 164–170 (2012).
• 4) Nakada T.-A., Oda S., Matsuda K.-I., Sadahiro T., Nakamura M., Abe R., Hirasawa H., Mol. Med., 14, 257–263 (2008).
• 5) Okazaki T., Ikeda T., Senno Y., Suzuki T., Kibayashi T., Niibe T., Nishinata J., Ishikawa A., Miyashita M., Nakanouchi T., J. Jpn. Soc. Blood Purif. Crit. Care, 6, 140–144 (2015).
• 6) Kozawa H., Jinko Zoki, 43, 228–232 (2014).
• 7) Ueno Y., Takanashi H., Sugaya H., J. Jpn. Soc. Blood Purif. Crit. Care, 3, 14–19 (2012).
• 8) Shiraishi Y., Okajima M., Sai Y., Miyamoto K., Inaba H., Anaesth. Intensive Care, 40, 442–449 (2012).
• 9) Nakamura A., Takasu O., Morita T., Nabeta M., Yamaga O., Mori S., Yamashita N., Sakamoto T., J. Jpn. Soc. Blood Purif. Crit. Care, 6, 115–118 (2015).
• 10) Inano Y., Tsuchiya K., Kumano R., Miura G., Nakasa H., Ren. Replace. Ther., 10, 8 (2024).
• 11) Sanofi K. K., “Interview form TARGOCID^® for i.v. infusion.”: 〈https://www.e-mr.sanofi.co.jp/dam/jcr:0c41d3f9-6780-4ab2-bd97-644b713858f7/targocid.pdf〉, cited 10 May, 2024.
• 12) Meiji Seika Pharma Co, Ltd. “Interview form VANCOMYCIN HYDROCHLORIDE for I.V. Infusion [MEIJI].”: 〈https://www.meiji-seika-pharma.co.jp/medical/product_med/item/000505/upload/revision/interview/000505_ITV.pdf〉, cited 10 May, 2024.
• 13) SIELC Technologies. “By Compound Daptomycin.”: 〈https://sielc.com/daptomycin〉, cited 10 May, 2024.
• 14) The Ministry of Health, Labour and Welfare, Japan. “Ultraviolet-visible Reference Spectra to The Japanese Pharmacopoeia Eighteenth Edition.”: 〈https://www.mhlw.go.jp/content/11120000/000912393.pdf〉, cited 10 May, 2024.
• 15) Kanda Y., Bone Marrow Transplant., 48, 452–458 (2013).
• 16) Sawada M., Uchida M., Yamazaki S., Suzuki T., Suzuki T., Ishii I., Artif. Organs, 47, 667–679 (2023).
• 17) Nishida O., J. Jpn. Soc. Blood Purif. Crit. Care., 2, 52–60 (2011).
• 18) Yamamoto T., Hisaka A., Suzuki H., Jpn. J. Nephrol. Pharmacother., 3, 3–19 (2014).
• 19) Moriyama K., Kato Y., Hasegawa D., Kurimoto Y., Kawaji T., Shimomura Y., Nishida O., J. Jpn. Soc. Blood Purif. Crit. Care, 10, 5–9 (2019).
• 20) Benvenuto M., Benziger D. P., Yankelev S., Vigliani G., Antimicrob. Agents Chemother., 50, 3245–3249 (2006).
• 21) Taguchi M., Ide T., Takesue Y., Ikawa K., Morikawa N., Nishi S., J. Jpn. Soc. Blood Purif. Crit. Care, 11, 124–127 (2020).