Effects of Drug Concentration, Rate of Infusion, and Flush Volume on G-CSF Drug Loss When Administered Intravenously

Abstract

Purpose: This study evaluated factors that affect granulocyte-colony stimulating factor (G-CSF) adsorption in the infusion tube by measuring the G-CSF concentration, rate of G-CSF infusion, and volume of flush solution. Methods: The concentrations of G-CSF in all samples were measured by an enzyme-linked immunosorbent assay (ELISA) using human G-CSF Quantikine^® ELISA kits. The concentration of G-CSF, the rate of administration, and the volume of flush solution were studied respectively. The concentration of G-CSF and the rate of administration that had a significantly lower G-CSF percent recovery after the infusion via the infusion set were used for further investigation in the study. All samples were diluted with 5% dextrose in water (D5W) to the final concentration within the standard concentration range. All experiments were performed in triplicate. Results: The concentration of G-CSF that was administered through the infusion tube at 20 µg/mL was a significantly higher G-CSF percent recovery compared with the G-CSF concentrations of 5, 10, and 15 µg/mL. The infusion rate of 15 and 20 mL/h percent recovery of G-CSF adsorption was significantly higher than the infusion rates of 30 and 40 mL/h. The concentration of G-CSF at 15 µg/mL and an infusion rate of 30 mL/h were selected to investigate the flush volumes of D5W on G-CSF adsorption. The D5W flush volume of 40 mL dramatically decreased the G-CSF adsorption with a recovery of 103 ± 1.73%. Conclusion: The G-CSF concentration of 20 µg/mL with an infusion rate of 20 mL/h, using a 40 mL D5W flush, was appropriate for intravenous G-CSF administration.

Introduction

Granulocyte-colony stimulating factor (G-CSF), or filgrastim, is the glycoprotein hormone that plays an important role in stimulating the proliferation and differentiation of neutropoietic progenitor cells to granulocytes and functionally activating the mature neutrophil in healthy people.

G-CSF is largely used in the treatment of hematologic disorder to improve myelosuppression, which might be a direct result of the disease or may be indirectly induced by a wide variety of chemotherapy agents. G-CSF can reduce the depth and duration of neutropenia in hemato-oncological patients experiencing bone marrow suppression after chemotherapy.¹ G-CSF is also used to stimulate stem cell production in bone marrow transplantation. Currently, G-CSF is used in patients receiving chemotherapy who have a risk of febrile neutropenia (FN) greater than 20% for primary prophylaxis or in patients receiving chemotherapy who have prior experience of FN in Thailand.

G-CSF is administered to patients by subcutaneous (SC) injection or intravenous (IV) injection.² Subcutaneous injection is the most common route of G-CSF administration due to evidence that demonstrates SC as superior to IV administration in reducing neutrophil recovery time.³ Based on previous study results, G-CSF was adsorbed by the infusion tube between 10% and 32% resulting in an inferior efficacy of IV administration.^4-6 Each previous study had different approaches to IV G-CSF administration, which included the G-CSF concentration, the volume of tube flushing, and the injection method. However, many hematologic malignancy patients developed thrombocytopenia after receiving chemotherapy. The patients with platelet levels less than 20 000 cells/µL had an increased risk of bleeding which required them to receive administration of G-CSF IV instead of SC.⁷

There are many methods to increase the effect of IV G-CSF administration such as shortening the infusion set, increasing the concentration of G-CSF, adding the polysorbate 80 solution in G-CSF, or adding human albumin when the G-CSF concentration is ⩽15 µg/mL.^4,6,8

This is a pilot study evaluating the rate of G-CSF infusion and the volume of flush solution needed to increase the percent recovery of G-CSF when administered by IV. This study is significant because information on G-CSF adsorption when administered IV and appropriate administration methods remains limited.

Materials

The same lot numbers of 300 µg in 0.5 mL single-dose prefilled syringe of G-CSF (filgrastim) was purchased from Amgen Inc (California). Normal saline 0.9% sodium chloride injection, 5% dextrose in water (D5W), and sterile water for injection in a polypropylene bottle were purchased from General Hospital Products (Bangkok, Thailand). G-CSF Quantikine^® ELISA kits were purchased from R&D Systems (Minneapolis). Terufusion^® (TS-PM270LA) administration set with volumetric chamber for infusion pump was purchased from Terumo Co, Ltd (Tokyo, Japan).

Methods

Quantitative Assay of G-CSF in Sample

The concentration of G-CSF in all samples was measured by an enzyme-linked immunosorbent assay (ELISA) using human G-CSF Quantikine^® ELISA kits according to the manufacturer’s instructions. In brief, 100 µL of the buffered protein-based assay diluent was added to each well. Next, 100 µL of the standard, control, or sample was added and incubated for 2 hours at room temperature. The well was aspirated and washed with 400 µL of buffer wash by using a Microplate Washer (Thermo Scientific, Massachusetts) for 3 washes. The 200 µL of human G-CSF conjugate was added to each well and incubated for 1 hour at room temperature. After 1 hour of incubation, the well was aspirated and washed in the same manner. The 200 µL of substrate solution was added and incubated for 20 minutes at room temperature and protected from light. Finally, 50 µL of stock solution was added. The color changed from blue to yellow and the optical density of each well was measured using a Multiskan GO Microplate Spectrophotometer (Thermo Scientific, Massachusetts). The optical density was measured at 450 nm with a correction wavelength of 540 nm. The human G-CSF standard solutions were prepared in the range of 78 to 1250 pg/mL. The control sample was the only reagent used in ELISA and the blank sample was D5W. All samples were diluted with D5W to make the final concentration within the standard concentration range. All experiments were performed in triplicate.

Influence of G-CSF Concentration on Infusion Set Adsorption

G-CSF (300 µg in 0.5 mL single-dose prefilled syringe) was mixed with D5W to make the final concentrations of 5, 10, 15, 20, and 25 µg/mL in the volumetric chamber of an administration set. The solutions were administered through an infusion tube with an infusion rate of 40 mL/h using the Terufusion^® TE-LM700 infusion pump (Terumo Co, Ltd, Tokyo, Japan). An additional 20 mL of D5W was added to the volumetric chamber to flush the deposited G-CSF solution in the administration set. The sample was collected in a 100 mL volumetric flask (Pyrex^®, Massachusetts). The volume was adjusted to 100 mL with D5W. The G-CSF concentration was measured using the ELISA technique. All experiments were performed in triplicate.

Influence of G-CSF Infusion Rate on Infusion Set Adsorption

The highest G-CSF concentration, which had a significantly lower G-CSF percent recovery on the infusion tube set, was selected from the previous section for this study. The G-CSF was mixed with D5W to the appropriate final concentrations in the volumetric chamber of an administration set. The samples were administered through an infusion tube set at the infusion rate of 15, 20, 30, and 40 mL/h. Then, an additional 20 mL of D5W was added in the volumetric chamber to flush the deposited G-CSF solution in the administration set. The sample was collected in a 100 mL volumetric flask and adjusted to 100 mL with D5W. The G-CSF concentration was measured using the ELISA technique. All samples were performed in triplicate.

Influence of Flush Volume of D5W on Infusion Set Adsorption

The lowest infusion rate of G-CSF, which had a significantly lower G-CSF percent recovery on the infusion tube set and G-CSF concentration in the previous section, was selected for this study. The G-CSF was mixed with D5W to the appropriate final concentrations in the volumetric chamber of an administration set. The samples were administered through an infusion tube set at the highest adsorbed infusion rate of G-CSF. D5W in volumes of 20, 30, 40, and 50 mL was added to the volumetric chamber to flush the deposited G-CSF solution in the administration set. The sample was collected in a 100 mL volumetric flask and adjusted to 100 mL with D5W. The G-CSF concentration was measured using the ELISA technique. All experiments were performed in triplicate.

The Best Condition of G-CSF Administration on Infusion Tube

The best condition of G-CSF concentration, infusion rate, and D5W flush volume of G-CSF administration was measured using the ELISA technique. All experiments were performed in triplicate.

Statistical Analysis

The amount of G-CSF remaining after administration into the infusion tube was reported as the percent recovery. The percent recovery in each experiment was carried out using the analysis of variance (ANOVA) with Tukey’s post hoc test. Significance was tested at the .05 level of probability. Statistical analysis was performed using the IBM SPSS software version 22.0 for Windows.

Results

Quantitative Assay of G-CSF in Sample

The calibration curve of human G-CSF standard solutions showed linearity at the concentrations 78, 156, 313, 625, and 1250 pg/mL and the correlation coefficient of the calibration curve was >0.999. The optical density of control and blank samples did not show any significant difference. The recovery of G-CSF in single-dose prefilled syringe (300 µg in 0.5 mL) was 98.54 ± 11.67%.

Influence of G-CSF Concentration on Infusion Set Adsorption

The recovery of G-CSF after administration of medication through the infusion tube at various concentrations is shown in Figure 1. The low concentrations of G-CSF at 5, 10, and 15 µg/mL showed a lower recovery compared with the higher concentrations of G-CSF at 20 and 25 µg/mL. The G-CSF content administered through the infusion tube at 20 µg/mL showed a higher significant percent recovery of G-CSF when compared with the G-CSF concentrations at 5, 10, and 15 µg/mL. Increasing the G-CSF concentration above 20 µg/mL revealed a steady recovery of G-CSF. Thus, G-CSF adsorption was decreased when the concentration was increased to ⩾20 µg/mL.

Figure 1.

Percent recovery of G-CSF after administration through the infusion tube at various concentrations.

Note. G-CSF = granulocyte-colony stimulating factor.

^aThe G-CSF concentrations at 20 and 25 µg/mL have a higher significant percent recovery of G-CSF when compared with the G-CSF concentrations at 5, 10, and 15 µg/mL (P < .02).

Influence of G-CSF Infusion Rate on Infusion Set Adsorption

The concentration of G-CSF at 15 µg/mL was selected to investigate the appropriate infusion rate of G-CSF adsorption. The recovery of G-CSF after administration through the infusion tube at various infusion rates is shown in Figure 2. The percent recovery of G-CSF significantly decreased when the infusion rate of G-CSF was above 20 mL/h. The infusion rate of 15 and 20 mL/h showed significantly higher recovery of G-CSF content than the infusion rate of 30 and 40 mL/h.

Figure 2.

Percent recovery of G-CSF at 15 µg/mL after administration through the infusion tube at various infusion rates.

Note. G-CSF = granulocyte-colony stimulating factor.

^aThe infusion rate of 15 and 20 mL/h showed significantly higher recovery in G-CSF content than the infusion rate of 30 and 40 mL/h (P < .03).

Influence of Flush Volume of D5W on Infusion Set Adsorption

The concentration of G-CSF at 15 µg/mL with the rate of infusion of 30 mL/h was selected for investigation of the flush volumes of D5W on G-CSF adsorption. The G-CSF recovery with various flush volumes of D5W is shown in Figure 3. The adsorption of G-CSF decreased when the D5W flush volume was increased. The flush volume plays an important role in reducing the G-CSF adsorption on the infusion set. Although, the G-CSF at a concentration of 15 µg/mL at the infusion rate of 30 mL/h had the highest G-CSF adsorption. The D5W flush volume of 40 mL showed a dramatic decrease of G-CSF adsorption with the recovery up to 103 ± 1.73%.

Figure 3.

Percent recovery of G-CSF at 15 µg/mL with an infusion rate of 30 mL/h after administration through the infusion tube at various flush volumes.

Note. G-CSF = granulocyte-colony stimulating factor; D5W = 5% dextrose in water.

^aThe D5W flush volume of 40 and 50 mL showed significantly higher recovery G-CSF content than the D5W flush volume of 20 and 30 mL (P < .01).

Moreover, the lowest adsorption condition of the G-CSF on the infusion set was selected for evaluation of the G-CSF content. The G-CSF at 20 µg/mL at an infusion rate of 20 mL/h using a D5W flush volume of 40 mL was administered through the infusion set. The recovery of G-CSF at the lowest adsorption condition was 107 ± 12.24% in contrast to the highest adsorption condition (15 µg/mL, infusion rate of 30 mL/h and D5W flush volume of 30 mL) which was 47.22 ± 8.15%.

Discussion

This is a pilot study to evaluate factors that affect G-CSF adsorption in the infusion tube by measuring the G-CSF concentration, rate of G-CSF infusion, and volume of flush solution.

The results of the study showed the G-CSF concentration ⩾20 µg/mL was decreased G-CSF adsorption. This result is consistent with previous studies, which reported that the adsorption rate of G-CSF increased at lower concentrations of G-CSF.⁶ There was a lack of information on the appropriate rate of G-CSF administration from previous studies. This study demonstrated that a higher rate of G-CSF administration results in a significant reduction in the amount of G-CSF after administration via the infusion set. The higher rate of G-CSF administration might increase the pressure of IV system and induce enough heat to activate G-CSF aggregation in the infusion tube similar to the protein medication insulin.⁹ For G-CSF ELISA test kit, the recently standard test kit could not detect the only free form of G-CSF. Based on the ELISA-based study, a specific assay for the free form of monoclonal antibody is required to provide accurate drug measurement in future studies.¹⁰ For these reasons, the loss of G-CSF when infused at the higher rate of administration might be caused by the aggregation and adsorption of G-CSF in the infusion system. However, the aggregated form of G-CSF did not have the clinical efficacy needed to stimulate the white blood cell count. For the effects of the flush solution, flush solution with a volume ⩾40 mL was the major important result in this study that had not been mentioned in previous studies.

The results of this study show that both the rate of drug administration and the volume of the flush solution also affect drug adsorption. Selection of an appropriate flush volume can increase G-CSF recovery by nearly 50%. The landmark study of G-CSF administration showed that the time to neutropenia resolution was longer with IV bolus administration of G-CSF compared with SC G-CSF.³ The study was prematurely discontinued in the second interim analysis because of increased mortality with IV G-CSF, which did not show a statistically significant difference. According to the results, the study preferred SC G-CSF administration in hospitalized hemato-oncological patients. The IV G-CSF administration in the study was given as a bolus injection through a central venous catheter followed by flushing the catheter without mention of the volume of the flush solution. The IV G-CSF administration in the study might increase G-CSF adsorption in the infusion tube because IV G-CSF was administrated at a high rate and inappropriate volume of flush solutions. The patients in the IV G-CSF group might have been taking a lower dose of G-CSF compared with the SC G-CSF group. These variances might have affected the outcome of the IV G-CSF group in the study.

However, this study was an in-vitro study concerning appropriate IV G-CSF administration. Further study of treatment for neutropenia in hemato-oncological patients might be needed to confirm the appropriate IV G-CSF compared with SC G-CSF. Adding polysorbate 80 solution or albumin to G-CSF might increase the effects of IV G-CSF administration which should be studied further.

The G-CSF administration through the infusion tube with a volumetric chamber is a routine clinical practice in our institution. However, other institutions might be using different methods of G-CSF administration, such as IV bag or syringe infusion pump. The other methods of G-CSF administration could differently impact the G-CSF adsorption. Therefore, the study of other methods of G-CSF administration should be performed in the future.

Conclusion

The best conditions for reducing G-CSF adsorption by the infusion tube were having a concentration of 20 µg/mL, infusion rate of 20 mL/h, and flush solution volume of 40 mL.