- Email: [email protected]
Techniques to Form a Suitable Lipiodol-Epirubicin Emulsion by Using 3-Way Stopcock Methods in Transarterial Chemoembolization for Liver Tumor
LABORATORY INVESTIGATION
Techniques to Form a Suitable LipiodolEpirubicin Emulsion by Using 3-Way Stopcock Methods in Transarterial Chemoembolization for Liver Tumor Tetsuya Masada, MD, Toshihiro Tanaka, MD, PhD, Hideyuki Nishiofuku, MD, PhD, Yasushi Fukuoka, MS, Takeshi Sato, MD, Shota Tatsumoto, MD, Nagaaki Marugami, MD, PhD, and Kimihiko Kichikawa, MD, PhD ABSTRACT Purpose: To compare physicochemical properties of emulsions of ethiodized oil (Lipiodol; Guerbet, Villepinte, France) and epirubicin prepared using different techniques for conventional transarterial chemoembolization. Materials and Methods: Lipiodol was mixed with epirubicin solution (8.33 mg/mL) by using a 3-way stopcock. The following technical parameters were compared: ratio of epirubicin solution to Lipiodol (1:2 vs 1:1), number of pumping exchanges through the stopcock (20 exchanges vs 10 exchanges), pumping speed (1 s/push vs 2 s/push), and first push syringe (epirubicin solution vs Lipiodol). Results: The mean percentage of water-in-oil was 70.45 ± 1.51 in the 1:2 epirubicin-Lipiodol ratio and 16.03 ± 2.95 in the 1:1 ratio (P < .001). The first push syringe did not influence emulsion type. Median droplet sizes were significantly larger in the slower pumping speed (52.0 μm in 2 s vs 33.7 μm in 1 s; P < .001), whereas there was no significant difference in number of pumping exchanges. Droplet sizes enlarged during 30 minutes after pumping. Viscosity was lower in the 1:1 ratio and the slower pumping speed. Viscosity decreased during 30 minutes after pumping. Conclusions: The ratio of epirubicin to Lipiodol is a significant factor to form water-in-oil emulsions with higher viscosity. The percentage of water-in-oil is limited to 70% using current pumping techniques. The pumping speed strongly influences droplet size and viscosity.
ABBREVIATIONS O/W ¼ oil-in-water, W/O ¼ water-in-oil
Conventional transarterial chemoembolization using ethiodized oil (Lipiodol; Guerbet, Villepinte, France) has been performed worldwide for inoperable hepatocellular carcinoma since the 1980s (1,2). Forming Lipiodol–cytotoxic
From the Department of Radiology, Nara Medical University, 840 Shijo-cho, Kashihara 634-8522, Japan. Received December 1, 2016; final revision received and accepted March 22, 2017. Address correspondence to T.T.; E-mail: [email protected] None of the authors have identified a conflict of interest. Video 1 is available online at www.jvir.org. © SIR, 2017 J Vasc Interv Radiol 2017; ▪:1–6 http://dx.doi.org/10.1016/j.jvir.2017.03.032
drug emulsions using a 3-way stopcock is the standard technique of conventional transarterial chemoembolization (3). It is widely known that the physicochemical characteristics of the emulsion can differ depending on the technical parameters. Previous research studies focused on the type of emulsion (water-in-oil [W/O] or oil-in-water [O/W]) and demonstrated the advantages of W/O in conventional transarterial chemoembolization for hepatocellular carcinoma. W/O emulsion has been associated with a high embolic effect (4,5); a high viscosity, which can increase its tumor retention (4,6); a high drug carriage capacity; and a longer drug release time (7–9). A technical recommendation of conventional transarterial chemoembolization published in 2015 suggested that (i) the volume of drug aqueous solution should be lower than the volume of Lipiodol, (ii) contrast medium should be used for
2 ▪ Lipiodol-Epirubicin Emulsions for Chemoembolization
preparation of doxorubicin aqueous solution, (iii) the contents of the syringe loaded with the drug should first be pushed toward the syringe containing Lipiodol, and (iv) at least 20 pumping exchanges through the stopcock are needed (3). Although these recommendations were based on the results of previous basic studies or theories, several questions remain unanswered: Which technical parameters could be the key to form an ideal emulsion? What percentage of W/O could be created by the standard pumping methods using a 3-way stopcock? In addition, more detailed technical information is needed, such as pumping speed. In 2013, a questionnaire survey of Japanese radiologists reported the use of different techniques in the preparation of emulsion for conventional transarterial chemoembolization (10). Although previous literature described that the type of emulsion (W/O or O/W) may not be reproducible when a 1:1 ratio is used to mix the epirubicin solution and Lipiodol (8), 15 of the 19 institutions in Japan (79%) that were part of the Japan-Korea Cooperative Study (JIVROSG 0604) on conventional transarterial chemoembolization presently use a 1:1 ratio (11). The present ex vivo study was conducted to clarify and update emulsion preparation techniques and report key techniques that could optimize the therapeutic results of conventional transarterial chemoembolization.
MATERIALS AND METHODS Lipiodol was mixed with epirubicin solution by using a 3-way stopcock. Five types of emulsions (control emulsion, emulsions A, B, C, and D) were created. In the control emulsion, the following technical parameters were used: (a) ratio of epirubicin solution to Lipiodol, 1:2; (b) number of pumping exchanges through the 3-way stopcock, 20 times; (c) pumping speed, 1 second per 1 syringe push; and (d) the syringe containing epirubicin solution was first pushed toward the syringe containing Lipiodol. To examine the effects of the parameters, each parameter was changed and compared with the control emulsion: in emulsion A, a 1:1 ratio was used to mix the epirubicin solution and Lipiodol; in emulsion B, the number of pumping exchanges through the 3-way stopcock was 10; in emulsion C, the pumping speed was 2 seconds per 1 syringe push; and in emulsion D, the syringe containing Lipiodol was pushed first. Emulsions A, B, C, and D were compared with the control emulsion as follows: percentages of W/O in the emulsion, droplet sizes of the dispersion phase, viscosities of the emulsion, and microscopic findings. To conduct these evaluations, 50 emulsions were produced: 20 for percentages of W/O (4 for each of 5 different emulsions), 20 for droplet sizes (4 for each emulsion) and microscopic examination, and 10 for viscosities (2 for each emulsion).
Preparation of Epirubicin Solution and Pumping Epirubicin hydrochloride (Epirubicin; Nippon Kayaku Co, Ltd, Tokyo, Japan) 50 mg was dissolved in 6 mL of contrast
Masada et al ▪ JVIR
material (250 mgI/mL). To create the 250 mgI/mL contrast material, 5 mL of iomeprol (Iomeron; Eisai Co, Ltd, Tokyo, Japan) 300 mgI/mL with specific gravity of 1.32 was mixed with 1 mL of saline with specific gravity of 1.06. The specific gravity of the mixture should be 1.27, which is equal to the specific gravity of Lipiodol (9). Two 10-mL syringes (SS-10LZ; Terumo, Tokyo, Japan) were filled with 2.5 mL or 5 mL of the epirubicin solution and 5 mL Lipiodol, respectively. Then, with a 3-way stopcock (Discofix C 3SC; B. Braun AG, Melsungen, Germany), using the push and back method, the syringes mixed the epirubicin solution with the Lipiodol. The pumping exchanges were performed 20 or 10 times. The push and back (1 full exchange) equaled 1 count.
Measurement of Percentage of W/O To measure the amount of W/O in each emulsion, a dilution test was performed (12). The whole quantity of the produced emulsion was gently poured into a solution of sodium dodecyl sulfate (0.5%), which is a hydrophilic surfactant. The W/O part precipitated at the bottom of the beaker, whereas the O/W part dispersed in the water. Then the water with dispersed O/W was immediately extracted using a dropper, and the precipitated W/O was collected in a graduated cylinder to measure the volume. The percentage of the W/O (%W/O) was calculated using the following formula: %W=O ¼ Volume of the precipitated W=O ðmLÞ= total volume of the poured emulsion ðmLÞ 100 The %W/O was measured at the following time points: immediately, 3 minutes, 15 minutes, and 30 minutes after emulsion preparation in the control emulsion and emulsions A, B, C, and D. One emulsion sample at each time point in each emulsion was used.
Measurement of Droplet Size in W/O The entire amount of the emulsion was gently poured into a kerosene solution with a hydrophobic surfactant of polyglycerol polyricinoleate (0.5%). The W/O part dispersed in the solution, whereas the O/W part precipitated at the bottom of the beaker. Approximately 0.5 mL of the dispersed solution was then collected with a dropper, and the droplet size was measured using a laser diffraction particle analyzer (Wing SALD 2000; Shimadzu Corp, Kyoto, Japan). A container in the particle analyzer was filled with a kerosene solution containing polyglycerol polyricinoleate. A droplet of the dispersion phase in the W/O part was detected, and the size distribution was automatically measured. The droplet size was measured at the following time points: immediately, 3 minutes, 15 minutes, and 30 minutes in the control emulsion and emulsions A, B, C, and D. The measurements were performed 3 times in each emulsion.
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017
3
Measurement of Viscosity A glass container was filled with 2 mL of the produced emulsion and set in a tuning fork vibro viscometer (SV-1A; A&D Co, Ltd, Tokyo, Japan). The viscometer simultaneously showed the viscosity. The viscosity was measured at immediately and 30 minutes after pumping in the control emulsion and emulsions A, B, C, and D. One emulsion sample at each time point in each emulsion was used.
Microscopic Examination The same emulsion produced for the measurement of the droplet size was used for microscopic examination. An emulsion volume of 0.02 mL was dropped on a glass slide and immediately observed by an optical microscope without a cover glass. The approximate ratio of the W/O part and the O/W part was visually estimated in a low-power field (magnification 40–100). The shape and approximate sizes of the droplets were observed in a high-power field (magnification 400). Microscopic examinations were performed immediately, 3 minutes, 15 minutes, and 30 minutes after pumping.
Figure 1. Percentages of W/O. The %W/O of emulsion A (1:1 ratio) was significantly lower than the %W/O of the control emulsion (1:2 ratio), and the %W/O of emulsion C was significantly higher than the control emulsion.
Statistical Analysis The goal of this study was to compare emulsions A, B, C, and D with the control emulsion. Pairwise comparisons of
Table. Technical Parameters and Results W/O (%)*
Droplet Size (μm)†
Viscosity (cP)‡
70.45 ± 1.51
33.7 ± 2.65
152.7
16.03 ± 2.95
22.6 ± 0.58
53.2
70.73 ± 4.49
38.5 ± 3.31
107.4
Emulsion C
78.55 ± 2.26
52.0 ± 4.38
87.3
2 s/push Emulsion D
69.03 ± 3.21
33.1 ± 2.52
165.2
Emulsions Control emulsion Epirubicin-toLipiodol ratio 1:2 20 exchanges 1 s/push Epirubicin solution first push Emulsion A Epirubicin-toLipiodol ratio 1:1 Emulsion B 10 exchanges
Lipiodol first push W/O ¼ water-in-oil. *Mean percentages for 30 min. Emulsion A showed a significantly lower %W/O than the control emulsion (P < .001). † Mean values of the median sizes obtained immediately after pumping. Emulsion C was significantly larger than the control emulsion (P < .001). Emulsion A was significantly smaller than the control emulsion (P ¼ .002). ‡ Viscosity of each emulsion was obtained immediately after pumping.
Figure 2. Droplet size distribution curves in the W/O part obtained immediately after pumping. Emulsion C (2 s/1 push) had significantly larger droplet sizes than the control emulsion (1 s/1 push), and emulsion A (1:1 ratio) had significantly smaller droplet sizes than the control emulsion (1:2 ratio).
means of variables between control and experimental emulsions were performed with the Student t test. Values of P < .05 were considered significant. All analyses were performed using IBM SPSS Statistics for Windows, Version 21.0 (IBM Corp, Armonk, New York).
RESULTS Percentage of W/O The Table summarizes the results. The %W/O in each emulsion is shown in Figure 1. Emulsion A showed a significantly lower %W/O than the control emulsion (P < .001) (Video 1). Emulsion C showed a significantly higher %W/O than the control emulsion (P ¼ .004).
Droplet Sizes The size distributions of the droplet in the dispersion phase in W/O immediately after pumping are shown in Figure 2.
4 ▪ Lipiodol-Epirubicin Emulsions for Chemoembolization
Masada et al ▪ JVIR
emulsion A was significantly smaller than in the control emulsion (P ¼ .002). The changes in the droplet sizes are shown in Figure 3. In the control emulsion and emulsions B, C, and D, the droplet sizes enlarged after 30 minutes, whereas the droplet sizes in emulsion A remained stable.
Viscosity All emulsions except emulsion D showed lower initial viscosity than the control emulsion, and all emulsion viscosities decreased after 30 minutes (Fig 4).
Figure 3. Changes of the median sizes of droplets during 30 min after pumping. Droplets in all emulsions except emulsion A gradually enlarged.
Microscopic Findings In all the emulsions, both W/O and O/W areas were mixed. In the control emulsion and emulsions B, C, and D, > 50% of the areas were W/O, whereas > 50% of the area in emulsion A was O/W. These findings were consistent with the results of the W/O percentage measurements. Although all the emulsions contained various droplet sizes in the W/O area, the mean sizes of the droplets visually corresponded to the results of the particle analyzer (Fig 5a, b).
DISCUSSION
Figure 4. Viscosities of each emulsion immediately after and 30 min after pumping. Emulsions A, B, and C had lower viscosities than the control emulsion and emulsion D. In all emulsions, the viscosities decreased 30 min after pumping.
The median size of droplets in emulsion C was significantly larger than the median size of droplets in the control emulsion (P < .001). Droplet size in
To date, there is no widely accepted evidence-based prescription for the preparation of Lipiodol-drug emulsion for conventional transarterial chemoembolization. Although advantages of W/O emulsion were reported (4–9), the percentage of W/O produced using current pumping techniques is unclear. Previous reports simply defined a mixture of a lower volume of drug solution with a higher volume of Lipiodol as W/O and a higher volume of drug solution with a lower volume of Lipiodol as O/W, although all emulsions contained both W/O and O/W. The present study evaluated the actual percentage of W/O using a previously reported dilution test (12). The results of this study confirmed that a 1:2 ratio of epirubicin solution to Lipiodol is preferable to a 1:1 ratio to
Figure 5. (a) Both W/O and O/W areas are present in this microscopic field (magnification 100). In the O/W area, Lipiodol droplets are floating in the continuous phase of epirubicin solution, whereas in the W/O area, epirubicin solution droplets are floating in the continuous phase of Lipiodol. (b) W/O area (magnification 400). The droplets of epirubicin solution in the oil are seen.
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017
produce W/O emulsion, similar to previous reports (6). Although a lower ratio of drug to Lipiodol is technically recommended for conventional transarterial chemoembolization (3), many institutions in Japan continue to adopt the 1:1 ratio (10,11). The produced W/O ratio was only 17% when a 1:1 ratio was used in the present study. According to the dilution test results, emulsion C showed a higher percentage of W/O. This could be explained by the large droplet sizes, which is discussed subsequently. Although the technical recommendation described that the contents of the syringe loaded with the drug should first be pushed toward the syringe containing Lipiodol (3), the advantage of this technique was not shown in the present study. The results of this study suggest that the pumping speed seems to influence the droplet size compared with the number of pumping exchange times. Specifically, pumping speed should be faster (1 s/1 push) to produce smaller droplets. Stirring speed has been previously related to droplet size in W/O emulsions (13). However, the methodology in this study may not be identical, particularly regarding pumping speeds. In contrast, only a small difference in the droplet size was found between the 10 times versus 20 times complete cycle. Despite this, the technical recommendation includes 20 pumping exchanges through the stopcock (3), although the description of a single exchange or cycle remains unclear. Droplet size is related to emulsion stability; emulsions with smaller droplets of dispersion phase were more stable compared with emulsions with larger droplets (13). All emulsions except emulsion A were unstable in size. The droplet sizes enlarged during 30 minutes after the pumping. This is a limitation of the current pumping technique using the 3-way stopcock without any surfactants (14,15). Emulsion A showed the lowest viscosity, which is consistent with the result of a previous report showing W/O had a higher viscosity than O/W. Emulsions C and D also had lower viscosity owing to larger droplet sizes (13). Although emulsion C showed higher %W/O as mentioned previously, the low viscosity could be unfavorable for conventional transarterial chemoembolization; the viscosity of emulsion is not related to the embolic effect, but rather to its tumor retention (4,6). The viscosities in all the emulsions decreased after 30 minutes, which can be explained by the coalesced droplets. This study has some limitations. First, epirubicin was used in this study, although doxorubicin is used for conventional transarterial chemoembolization worldwide. Epirubicin is most commonly used for conventional transarterial chemoembolization in Japan because of lower cardiac toxicity associated with epirubicin compared with doxorubicin (16). Epirubicin is an anthracycline anticancer agent, which is a stereoisomer of doxorubicin. The hydroxyl group of epirubicin is inverted to the position 40 . The molecular structures of epirubicin and doxorubicin are quite similar (17), which should not influence the emulsion type, and thus the results of this study may address doxorubicin
5
emulsion as well. Second, only limited technical parameters were assessed; epirubicin-to-Lipiodol ratios of 1:1 and 1:2 were examined, although lower volumes of epirubicin solution might produce higher percentages of W/O emulsion. Third, these results could not directly lead to efficacy in chemoembolization for liver cancer patients because this was an ex vivo study, although previous basic studies discussed ideal emulsions for conventional transarterial chemoembolization. In conclusion, the pumping speed strongly influenced droplet size and viscosity, whereas the syringe first pushed did not influence the emulsion characteristics. The ratio of epirubicin solution and Lipiodol of 1:2 was an optimal parameter to produce W/O emulsion. This result may prompt physicians who are using a 1:1 ratio to use a 1:2 ratio. However, the percentage of W/O is limited to 70%–80%, and the droplet size and viscosity are unstable, which are critical limitations of the current pumping technique. Development of better techniques may provide more optimal emulsions.
ACKNOWLEDGMENTS We thank Mr. Mitsuteru Fujihara and Mr. Daigo Maeda for assistance in the ex vivo study. We thank Ms. Marian Pahud for assistance in submitting the article.
REFERENCES 1. Ohishi H, Uchida H, Yoshimura H, et al. Hepatocellular carcinoma detected by iodized oil. Use of anticancer agents. Radiology 1985; 154: 15–19. 2. Nakamura H, Hashimoto T, Oi H, et al. Transcatheter oily chemoembolization of hepatocellular carcinoma. Radiology 1989; 170:783–786. 3. de Baere T, Arai Y, Lencioni R, et al. Treatment of liver tumors with Lipiodol TACE: technical recommendations from experts opinion. Cardiovasc Interv Radiol 2016; 39:334–343. 4. de Baere T, Dufaux J, Roche A, et al. Circulatory alterations induced by intra-arterial injection of iodized oil and emulsions of iodized oil and doxorubicin: experimental study. Radiology 1995; 194:165–170. 5. Demachi H, Matsui O, Abo H, Tatsu H. Simulation model based on nonnewtonian fluid mechanics applied to the evaluation of the embolic effect of emulsions of iodized oil and anticancer drug. Cardiovasc Interv Radiol 2000; 23:285–290. 6. Becker S, Lepareur N, Cadeillan V, et al. Optimization of hepatocarcinoma uptake with radiolabeled Lipiodol: development of new Lipiodol formulations with increased viscosity. Cancer Biother Radiopharm 2012; 27:149–155. 7. Kan Z, Wright K, Wallace S. Ethiodized oil emulsions in hepatic microcirculation: in vivo microscopy in animal models. Acad Radiol 1997; 4: 275–282. e JM, Guiu B. Use of Lipiodol as a drug-delivery system for trans8. Ide catheter arterial chemoembolization of hepatocellular carcinoma: a review. Crit Rev Oncol Hematol 2013; 88:530–549. 9. Sakaguchi H, Uchida H, Nishimura Y, et al. Pharmacokinetic study of adriamycin in the emulsion mixed with lipiodol-difference resulting from composition and methods of preparation, and behavior after mesenteric arterial injection in rat [in Japanese]. Gan To Kagaku Ryoho 1991; 18: 1349–1355. 10. Arai Y, Ikeda M. Techniques to prepare epirubicin-Lipiodol emulsion for transcatheter arterial chemoembolization of hepatocellular carcinoma— additional survey of the Japanese Institutions in a Cooperative Study between Korea and Japan. Jpn J Intervent Radiol 2013; 28:304–307. 11. Ikeda M, Arai Y, Park SJ. Prospective study of transcatheter arterial chemoembolization for unresectable hepatocellular carcinoma: an Asian
6 ▪ Lipiodol-Epirubicin Emulsions for Chemoembolization
cooperative study between Japan and Korea. J Vasc Interv Radiol 2013; 24:490–500. 12. Ortenzio LF, Opalsky CD, Stuart LS. Factors affecting the activity of phenolic disinfectants. Appl Microbiol 1961; 9:562–566. 13. Anisa ANI, Nour AH. Affect of viscosity and droplet diameter on water-in-oil (w/o) emulsion: an experimental study. WASET 2010; 4: 213–216. 14. Fukushima S, Nishida M, Nakano M. Preparation of and drug release from W/O/W type double emulsions containing anticancer agents using an oily lymphographic agent as an oil phase. Chem Pharm Bull 1987; 35: 3375–3381.
Masada et al ▪ JVIR
15. Hayashi K, Umakoshi H, Shimanouchi T. Comparison of the interfacial properties of Span 80 vesicle, W/O emulsion and liposomes. Solv Ext Res Dev 2014; 21:191–199. 16. Untch M, Muscholl M, Tjulandin S, et al. First-line trastuzumab plus epirubicin and cyclophosphamide therapy in patients with human epidermal growth factor receptor 2-positive metastatic breast cancer: cardiac safety and efficacy data from the Herceptin, Cyclophosphamide, and Epirubicin (HERCULES) trial. J Clin Oncol 2010; 28:1473–1480. 17. Ganzina F. 4’-Epi-doxorubicin, a new analogue of doxorubicin: a preliminary overview of preclinical and clinical data. Cancer Treat Rev 1983; 10:1–22.
Techniques to Form a Suitable LipiodolEpirubicin Emulsion by Using 3-Way Stopcock Methods in Transarterial Chemoembolization for Liver Tumor Tetsuya Masada, MD, Toshihiro Tanaka, MD, PhD, Hideyuki Nishiofuku, MD, PhD, Yasushi Fukuoka, MS, Takeshi Sato, MD, Shota Tatsumoto, MD, Nagaaki Marugami, MD, PhD, and Kimihiko Kichikawa, MD, PhD ABSTRACT Purpose: To compare physicochemical properties of emulsions of ethiodized oil (Lipiodol; Guerbet, Villepinte, France) and epirubicin prepared using different techniques for conventional transarterial chemoembolization. Materials and Methods: Lipiodol was mixed with epirubicin solution (8.33 mg/mL) by using a 3-way stopcock. The following technical parameters were compared: ratio of epirubicin solution to Lipiodol (1:2 vs 1:1), number of pumping exchanges through the stopcock (20 exchanges vs 10 exchanges), pumping speed (1 s/push vs 2 s/push), and first push syringe (epirubicin solution vs Lipiodol). Results: The mean percentage of water-in-oil was 70.45 ± 1.51 in the 1:2 epirubicin-Lipiodol ratio and 16.03 ± 2.95 in the 1:1 ratio (P < .001). The first push syringe did not influence emulsion type. Median droplet sizes were significantly larger in the slower pumping speed (52.0 μm in 2 s vs 33.7 μm in 1 s; P < .001), whereas there was no significant difference in number of pumping exchanges. Droplet sizes enlarged during 30 minutes after pumping. Viscosity was lower in the 1:1 ratio and the slower pumping speed. Viscosity decreased during 30 minutes after pumping. Conclusions: The ratio of epirubicin to Lipiodol is a significant factor to form water-in-oil emulsions with higher viscosity. The percentage of water-in-oil is limited to 70% using current pumping techniques. The pumping speed strongly influences droplet size and viscosity.
ABBREVIATIONS O/W ¼ oil-in-water, W/O ¼ water-in-oil
Conventional transarterial chemoembolization using ethiodized oil (Lipiodol; Guerbet, Villepinte, France) has been performed worldwide for inoperable hepatocellular carcinoma since the 1980s (1,2). Forming Lipiodol–cytotoxic
From the Department of Radiology, Nara Medical University, 840 Shijo-cho, Kashihara 634-8522, Japan. Received December 1, 2016; final revision received and accepted March 22, 2017. Address correspondence to T.T.; E-mail: [email protected] None of the authors have identified a conflict of interest. Video 1 is available online at www.jvir.org. © SIR, 2017 J Vasc Interv Radiol 2017; ▪:1–6 http://dx.doi.org/10.1016/j.jvir.2017.03.032
drug emulsions using a 3-way stopcock is the standard technique of conventional transarterial chemoembolization (3). It is widely known that the physicochemical characteristics of the emulsion can differ depending on the technical parameters. Previous research studies focused on the type of emulsion (water-in-oil [W/O] or oil-in-water [O/W]) and demonstrated the advantages of W/O in conventional transarterial chemoembolization for hepatocellular carcinoma. W/O emulsion has been associated with a high embolic effect (4,5); a high viscosity, which can increase its tumor retention (4,6); a high drug carriage capacity; and a longer drug release time (7–9). A technical recommendation of conventional transarterial chemoembolization published in 2015 suggested that (i) the volume of drug aqueous solution should be lower than the volume of Lipiodol, (ii) contrast medium should be used for
2 ▪ Lipiodol-Epirubicin Emulsions for Chemoembolization
preparation of doxorubicin aqueous solution, (iii) the contents of the syringe loaded with the drug should first be pushed toward the syringe containing Lipiodol, and (iv) at least 20 pumping exchanges through the stopcock are needed (3). Although these recommendations were based on the results of previous basic studies or theories, several questions remain unanswered: Which technical parameters could be the key to form an ideal emulsion? What percentage of W/O could be created by the standard pumping methods using a 3-way stopcock? In addition, more detailed technical information is needed, such as pumping speed. In 2013, a questionnaire survey of Japanese radiologists reported the use of different techniques in the preparation of emulsion for conventional transarterial chemoembolization (10). Although previous literature described that the type of emulsion (W/O or O/W) may not be reproducible when a 1:1 ratio is used to mix the epirubicin solution and Lipiodol (8), 15 of the 19 institutions in Japan (79%) that were part of the Japan-Korea Cooperative Study (JIVROSG 0604) on conventional transarterial chemoembolization presently use a 1:1 ratio (11). The present ex vivo study was conducted to clarify and update emulsion preparation techniques and report key techniques that could optimize the therapeutic results of conventional transarterial chemoembolization.
MATERIALS AND METHODS Lipiodol was mixed with epirubicin solution by using a 3-way stopcock. Five types of emulsions (control emulsion, emulsions A, B, C, and D) were created. In the control emulsion, the following technical parameters were used: (a) ratio of epirubicin solution to Lipiodol, 1:2; (b) number of pumping exchanges through the 3-way stopcock, 20 times; (c) pumping speed, 1 second per 1 syringe push; and (d) the syringe containing epirubicin solution was first pushed toward the syringe containing Lipiodol. To examine the effects of the parameters, each parameter was changed and compared with the control emulsion: in emulsion A, a 1:1 ratio was used to mix the epirubicin solution and Lipiodol; in emulsion B, the number of pumping exchanges through the 3-way stopcock was 10; in emulsion C, the pumping speed was 2 seconds per 1 syringe push; and in emulsion D, the syringe containing Lipiodol was pushed first. Emulsions A, B, C, and D were compared with the control emulsion as follows: percentages of W/O in the emulsion, droplet sizes of the dispersion phase, viscosities of the emulsion, and microscopic findings. To conduct these evaluations, 50 emulsions were produced: 20 for percentages of W/O (4 for each of 5 different emulsions), 20 for droplet sizes (4 for each emulsion) and microscopic examination, and 10 for viscosities (2 for each emulsion).
Preparation of Epirubicin Solution and Pumping Epirubicin hydrochloride (Epirubicin; Nippon Kayaku Co, Ltd, Tokyo, Japan) 50 mg was dissolved in 6 mL of contrast
Masada et al ▪ JVIR
material (250 mgI/mL). To create the 250 mgI/mL contrast material, 5 mL of iomeprol (Iomeron; Eisai Co, Ltd, Tokyo, Japan) 300 mgI/mL with specific gravity of 1.32 was mixed with 1 mL of saline with specific gravity of 1.06. The specific gravity of the mixture should be 1.27, which is equal to the specific gravity of Lipiodol (9). Two 10-mL syringes (SS-10LZ; Terumo, Tokyo, Japan) were filled with 2.5 mL or 5 mL of the epirubicin solution and 5 mL Lipiodol, respectively. Then, with a 3-way stopcock (Discofix C 3SC; B. Braun AG, Melsungen, Germany), using the push and back method, the syringes mixed the epirubicin solution with the Lipiodol. The pumping exchanges were performed 20 or 10 times. The push and back (1 full exchange) equaled 1 count.
Measurement of Percentage of W/O To measure the amount of W/O in each emulsion, a dilution test was performed (12). The whole quantity of the produced emulsion was gently poured into a solution of sodium dodecyl sulfate (0.5%), which is a hydrophilic surfactant. The W/O part precipitated at the bottom of the beaker, whereas the O/W part dispersed in the water. Then the water with dispersed O/W was immediately extracted using a dropper, and the precipitated W/O was collected in a graduated cylinder to measure the volume. The percentage of the W/O (%W/O) was calculated using the following formula: %W=O ¼ Volume of the precipitated W=O ðmLÞ= total volume of the poured emulsion ðmLÞ 100 The %W/O was measured at the following time points: immediately, 3 minutes, 15 minutes, and 30 minutes after emulsion preparation in the control emulsion and emulsions A, B, C, and D. One emulsion sample at each time point in each emulsion was used.
Measurement of Droplet Size in W/O The entire amount of the emulsion was gently poured into a kerosene solution with a hydrophobic surfactant of polyglycerol polyricinoleate (0.5%). The W/O part dispersed in the solution, whereas the O/W part precipitated at the bottom of the beaker. Approximately 0.5 mL of the dispersed solution was then collected with a dropper, and the droplet size was measured using a laser diffraction particle analyzer (Wing SALD 2000; Shimadzu Corp, Kyoto, Japan). A container in the particle analyzer was filled with a kerosene solution containing polyglycerol polyricinoleate. A droplet of the dispersion phase in the W/O part was detected, and the size distribution was automatically measured. The droplet size was measured at the following time points: immediately, 3 minutes, 15 minutes, and 30 minutes in the control emulsion and emulsions A, B, C, and D. The measurements were performed 3 times in each emulsion.
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017
3
Measurement of Viscosity A glass container was filled with 2 mL of the produced emulsion and set in a tuning fork vibro viscometer (SV-1A; A&D Co, Ltd, Tokyo, Japan). The viscometer simultaneously showed the viscosity. The viscosity was measured at immediately and 30 minutes after pumping in the control emulsion and emulsions A, B, C, and D. One emulsion sample at each time point in each emulsion was used.
Microscopic Examination The same emulsion produced for the measurement of the droplet size was used for microscopic examination. An emulsion volume of 0.02 mL was dropped on a glass slide and immediately observed by an optical microscope without a cover glass. The approximate ratio of the W/O part and the O/W part was visually estimated in a low-power field (magnification 40–100). The shape and approximate sizes of the droplets were observed in a high-power field (magnification 400). Microscopic examinations were performed immediately, 3 minutes, 15 minutes, and 30 minutes after pumping.
Figure 1. Percentages of W/O. The %W/O of emulsion A (1:1 ratio) was significantly lower than the %W/O of the control emulsion (1:2 ratio), and the %W/O of emulsion C was significantly higher than the control emulsion.
Statistical Analysis The goal of this study was to compare emulsions A, B, C, and D with the control emulsion. Pairwise comparisons of
Table. Technical Parameters and Results W/O (%)*
Droplet Size (μm)†
Viscosity (cP)‡
70.45 ± 1.51
33.7 ± 2.65
152.7
16.03 ± 2.95
22.6 ± 0.58
53.2
70.73 ± 4.49
38.5 ± 3.31
107.4
Emulsion C
78.55 ± 2.26
52.0 ± 4.38
87.3
2 s/push Emulsion D
69.03 ± 3.21
33.1 ± 2.52
165.2
Emulsions Control emulsion Epirubicin-toLipiodol ratio 1:2 20 exchanges 1 s/push Epirubicin solution first push Emulsion A Epirubicin-toLipiodol ratio 1:1 Emulsion B 10 exchanges
Lipiodol first push W/O ¼ water-in-oil. *Mean percentages for 30 min. Emulsion A showed a significantly lower %W/O than the control emulsion (P < .001). † Mean values of the median sizes obtained immediately after pumping. Emulsion C was significantly larger than the control emulsion (P < .001). Emulsion A was significantly smaller than the control emulsion (P ¼ .002). ‡ Viscosity of each emulsion was obtained immediately after pumping.
Figure 2. Droplet size distribution curves in the W/O part obtained immediately after pumping. Emulsion C (2 s/1 push) had significantly larger droplet sizes than the control emulsion (1 s/1 push), and emulsion A (1:1 ratio) had significantly smaller droplet sizes than the control emulsion (1:2 ratio).
means of variables between control and experimental emulsions were performed with the Student t test. Values of P < .05 were considered significant. All analyses were performed using IBM SPSS Statistics for Windows, Version 21.0 (IBM Corp, Armonk, New York).
RESULTS Percentage of W/O The Table summarizes the results. The %W/O in each emulsion is shown in Figure 1. Emulsion A showed a significantly lower %W/O than the control emulsion (P < .001) (Video 1). Emulsion C showed a significantly higher %W/O than the control emulsion (P ¼ .004).
Droplet Sizes The size distributions of the droplet in the dispersion phase in W/O immediately after pumping are shown in Figure 2.
4 ▪ Lipiodol-Epirubicin Emulsions for Chemoembolization
Masada et al ▪ JVIR
emulsion A was significantly smaller than in the control emulsion (P ¼ .002). The changes in the droplet sizes are shown in Figure 3. In the control emulsion and emulsions B, C, and D, the droplet sizes enlarged after 30 minutes, whereas the droplet sizes in emulsion A remained stable.
Viscosity All emulsions except emulsion D showed lower initial viscosity than the control emulsion, and all emulsion viscosities decreased after 30 minutes (Fig 4).
Figure 3. Changes of the median sizes of droplets during 30 min after pumping. Droplets in all emulsions except emulsion A gradually enlarged.
Microscopic Findings In all the emulsions, both W/O and O/W areas were mixed. In the control emulsion and emulsions B, C, and D, > 50% of the areas were W/O, whereas > 50% of the area in emulsion A was O/W. These findings were consistent with the results of the W/O percentage measurements. Although all the emulsions contained various droplet sizes in the W/O area, the mean sizes of the droplets visually corresponded to the results of the particle analyzer (Fig 5a, b).
DISCUSSION
Figure 4. Viscosities of each emulsion immediately after and 30 min after pumping. Emulsions A, B, and C had lower viscosities than the control emulsion and emulsion D. In all emulsions, the viscosities decreased 30 min after pumping.
The median size of droplets in emulsion C was significantly larger than the median size of droplets in the control emulsion (P < .001). Droplet size in
To date, there is no widely accepted evidence-based prescription for the preparation of Lipiodol-drug emulsion for conventional transarterial chemoembolization. Although advantages of W/O emulsion were reported (4–9), the percentage of W/O produced using current pumping techniques is unclear. Previous reports simply defined a mixture of a lower volume of drug solution with a higher volume of Lipiodol as W/O and a higher volume of drug solution with a lower volume of Lipiodol as O/W, although all emulsions contained both W/O and O/W. The present study evaluated the actual percentage of W/O using a previously reported dilution test (12). The results of this study confirmed that a 1:2 ratio of epirubicin solution to Lipiodol is preferable to a 1:1 ratio to
Figure 5. (a) Both W/O and O/W areas are present in this microscopic field (magnification 100). In the O/W area, Lipiodol droplets are floating in the continuous phase of epirubicin solution, whereas in the W/O area, epirubicin solution droplets are floating in the continuous phase of Lipiodol. (b) W/O area (magnification 400). The droplets of epirubicin solution in the oil are seen.
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017
produce W/O emulsion, similar to previous reports (6). Although a lower ratio of drug to Lipiodol is technically recommended for conventional transarterial chemoembolization (3), many institutions in Japan continue to adopt the 1:1 ratio (10,11). The produced W/O ratio was only 17% when a 1:1 ratio was used in the present study. According to the dilution test results, emulsion C showed a higher percentage of W/O. This could be explained by the large droplet sizes, which is discussed subsequently. Although the technical recommendation described that the contents of the syringe loaded with the drug should first be pushed toward the syringe containing Lipiodol (3), the advantage of this technique was not shown in the present study. The results of this study suggest that the pumping speed seems to influence the droplet size compared with the number of pumping exchange times. Specifically, pumping speed should be faster (1 s/1 push) to produce smaller droplets. Stirring speed has been previously related to droplet size in W/O emulsions (13). However, the methodology in this study may not be identical, particularly regarding pumping speeds. In contrast, only a small difference in the droplet size was found between the 10 times versus 20 times complete cycle. Despite this, the technical recommendation includes 20 pumping exchanges through the stopcock (3), although the description of a single exchange or cycle remains unclear. Droplet size is related to emulsion stability; emulsions with smaller droplets of dispersion phase were more stable compared with emulsions with larger droplets (13). All emulsions except emulsion A were unstable in size. The droplet sizes enlarged during 30 minutes after the pumping. This is a limitation of the current pumping technique using the 3-way stopcock without any surfactants (14,15). Emulsion A showed the lowest viscosity, which is consistent with the result of a previous report showing W/O had a higher viscosity than O/W. Emulsions C and D also had lower viscosity owing to larger droplet sizes (13). Although emulsion C showed higher %W/O as mentioned previously, the low viscosity could be unfavorable for conventional transarterial chemoembolization; the viscosity of emulsion is not related to the embolic effect, but rather to its tumor retention (4,6). The viscosities in all the emulsions decreased after 30 minutes, which can be explained by the coalesced droplets. This study has some limitations. First, epirubicin was used in this study, although doxorubicin is used for conventional transarterial chemoembolization worldwide. Epirubicin is most commonly used for conventional transarterial chemoembolization in Japan because of lower cardiac toxicity associated with epirubicin compared with doxorubicin (16). Epirubicin is an anthracycline anticancer agent, which is a stereoisomer of doxorubicin. The hydroxyl group of epirubicin is inverted to the position 40 . The molecular structures of epirubicin and doxorubicin are quite similar (17), which should not influence the emulsion type, and thus the results of this study may address doxorubicin
5
emulsion as well. Second, only limited technical parameters were assessed; epirubicin-to-Lipiodol ratios of 1:1 and 1:2 were examined, although lower volumes of epirubicin solution might produce higher percentages of W/O emulsion. Third, these results could not directly lead to efficacy in chemoembolization for liver cancer patients because this was an ex vivo study, although previous basic studies discussed ideal emulsions for conventional transarterial chemoembolization. In conclusion, the pumping speed strongly influenced droplet size and viscosity, whereas the syringe first pushed did not influence the emulsion characteristics. The ratio of epirubicin solution and Lipiodol of 1:2 was an optimal parameter to produce W/O emulsion. This result may prompt physicians who are using a 1:1 ratio to use a 1:2 ratio. However, the percentage of W/O is limited to 70%–80%, and the droplet size and viscosity are unstable, which are critical limitations of the current pumping technique. Development of better techniques may provide more optimal emulsions.
ACKNOWLEDGMENTS We thank Mr. Mitsuteru Fujihara and Mr. Daigo Maeda for assistance in the ex vivo study. We thank Ms. Marian Pahud for assistance in submitting the article.
REFERENCES 1. Ohishi H, Uchida H, Yoshimura H, et al. Hepatocellular carcinoma detected by iodized oil. Use of anticancer agents. Radiology 1985; 154: 15–19. 2. Nakamura H, Hashimoto T, Oi H, et al. Transcatheter oily chemoembolization of hepatocellular carcinoma. Radiology 1989; 170:783–786. 3. de Baere T, Arai Y, Lencioni R, et al. Treatment of liver tumors with Lipiodol TACE: technical recommendations from experts opinion. Cardiovasc Interv Radiol 2016; 39:334–343. 4. de Baere T, Dufaux J, Roche A, et al. Circulatory alterations induced by intra-arterial injection of iodized oil and emulsions of iodized oil and doxorubicin: experimental study. Radiology 1995; 194:165–170. 5. Demachi H, Matsui O, Abo H, Tatsu H. Simulation model based on nonnewtonian fluid mechanics applied to the evaluation of the embolic effect of emulsions of iodized oil and anticancer drug. Cardiovasc Interv Radiol 2000; 23:285–290. 6. Becker S, Lepareur N, Cadeillan V, et al. Optimization of hepatocarcinoma uptake with radiolabeled Lipiodol: development of new Lipiodol formulations with increased viscosity. Cancer Biother Radiopharm 2012; 27:149–155. 7. Kan Z, Wright K, Wallace S. Ethiodized oil emulsions in hepatic microcirculation: in vivo microscopy in animal models. Acad Radiol 1997; 4: 275–282. e JM, Guiu B. Use of Lipiodol as a drug-delivery system for trans8. Ide catheter arterial chemoembolization of hepatocellular carcinoma: a review. Crit Rev Oncol Hematol 2013; 88:530–549. 9. Sakaguchi H, Uchida H, Nishimura Y, et al. Pharmacokinetic study of adriamycin in the emulsion mixed with lipiodol-difference resulting from composition and methods of preparation, and behavior after mesenteric arterial injection in rat [in Japanese]. Gan To Kagaku Ryoho 1991; 18: 1349–1355. 10. Arai Y, Ikeda M. Techniques to prepare epirubicin-Lipiodol emulsion for transcatheter arterial chemoembolization of hepatocellular carcinoma— additional survey of the Japanese Institutions in a Cooperative Study between Korea and Japan. Jpn J Intervent Radiol 2013; 28:304–307. 11. Ikeda M, Arai Y, Park SJ. Prospective study of transcatheter arterial chemoembolization for unresectable hepatocellular carcinoma: an Asian
6 ▪ Lipiodol-Epirubicin Emulsions for Chemoembolization
cooperative study between Japan and Korea. J Vasc Interv Radiol 2013; 24:490–500. 12. Ortenzio LF, Opalsky CD, Stuart LS. Factors affecting the activity of phenolic disinfectants. Appl Microbiol 1961; 9:562–566. 13. Anisa ANI, Nour AH. Affect of viscosity and droplet diameter on water-in-oil (w/o) emulsion: an experimental study. WASET 2010; 4: 213–216. 14. Fukushima S, Nishida M, Nakano M. Preparation of and drug release from W/O/W type double emulsions containing anticancer agents using an oily lymphographic agent as an oil phase. Chem Pharm Bull 1987; 35: 3375–3381.
Masada et al ▪ JVIR
15. Hayashi K, Umakoshi H, Shimanouchi T. Comparison of the interfacial properties of Span 80 vesicle, W/O emulsion and liposomes. Solv Ext Res Dev 2014; 21:191–199. 16. Untch M, Muscholl M, Tjulandin S, et al. First-line trastuzumab plus epirubicin and cyclophosphamide therapy in patients with human epidermal growth factor receptor 2-positive metastatic breast cancer: cardiac safety and efficacy data from the Herceptin, Cyclophosphamide, and Epirubicin (HERCULES) trial. J Clin Oncol 2010; 28:1473–1480. 17. Ganzina F. 4’-Epi-doxorubicin, a new analogue of doxorubicin: a preliminary overview of preclinical and clinical data. Cancer Treat Rev 1983; 10:1–22.