Studies of PLGA Microspheres

Controlled release of a medicament, particularly one that needs to be administered frequently in high doses and can result in severe adverse drug events (ADEs), has been a field of intense research. Of the various drug delivery systems that have been studied, particulate drug delivery systems, particularly microspheres, have yielded the most promising results.^1,2 Polymers such as albumin and chitosan have been used to prepare microspheres in different sizes to deliver drugs in a controlled fashion to the target organ.^3,4

Poly(DL-lactide-co-glycolide) (PLGA) has been used to prepare microspheres that possess tremendous potential to release a drug in a controlled manner.^{5, 6} Previously, several attempts have been made to gain more understanding about the effect of changing different process parameters on the characteristics of PLGA microspheres.^7-9 In this article, the authors have determined the effect of two process parameters (drug and polymer concentration) on selected pharmaceutical characteristics of PLGA microspheres. Further, the parameters and their effects have been correlated to define their collective effect through an appropriate mathematical model. Drug and polymer concentrations have been selected because they are among the most important factors affecting microsphere properties.

The study used methotrexate as a water insoluble drug. Methotrexate belongs to the group of medicines known as antimetabolites. It is used to treat cancer of the breast, head and neck, lung, blood, bone and lymph, and tumours in the uterus. It may also be used to treat other kinds of cancer. The drug needs frequent administration, which may lead to abrupt bone marrow depression and leucopaenia, thrombocytopaenia and anaemia.¹⁰ Several attempts have previously been made to control these complications by encapsulation of methotrexate using biodegradable polymers.¹¹

Although it is important for a dosage form to possess all of the desired qualities to give a premeditated therapeutic result, it is equally important that it remains stable until its administration. Whereas microspheres used for targeting entrapped/encapsulated medicaments are usually administered in suspension to the target site, they are also administered orally after filling in capsules or compressed in tablets. Microspheres, if stored in suspension, are prone to show some undesirable changes such as swelling, leaching of the drug to suspension media and a change in morphological attributes. Therefore, microspheres may be expected to be more stable in dry form. However, there may also be some changes in the microspheres when stored in dry conditions because of susceptibility of the polymer and/or drug to atmospheric conditions.

The authors have determined the stability of the PLGA microspheres containing methotrexate in solid state at 2-8 °C; higher temperatures cause polymer instability. The stability of the microspheres was evaluated by characterizing the microspheres for both physical and chemical parameters. Differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) have been widely used to detect physical changes, for example, transition temperature (T_g), polymorphism, melting, decomposition and crystallization. Changes in crystallinity can be further confirmed by X-ray diffraction. High performance liquid chromatography (HPLC) is a standard pharmacopoeial method for the analysis of methotrexate; the authors used HPLC as a qualitative tool to determine the presence of impurities produced by drug degradation during stability studies.

Materials and methods

Materials

PLGA (Resomer RG 502H; intrinsic viscosity 0.16-0.24 dL/g) was provided by Boehringer Ingelheim (Germany). Methotrexate (Biochem Pharmaceutical Industries, India) was of pharmacopoeial grade. Other reagents used were polyvinyl alcohol ([PVA] analytical reagent grade; SD Fine Chemicals Ltd, India) and dichloromethane ([DCM] analytical reagent grade; Allied Chemical Corporation, India).

Experimental design

Drug and polymer concentrations were chosen (Table I). Full factorial design of 2

²

was used as a research methodology that required preparation of four batches (Table II). The sequence of these experiments was randomized. The full factorial experimental design was used to conduct a comprehensive study of the effect of the process parameters and their interaction using a suitable statistical tool. All batches were prepared in triplicate.

Preparation of PLGA microspheres

PLGA microspheres containing methotrexate were prepared by the method described by Parikh et al. with slight modification.

¹²

A drug solution (0.75 mL) containing a different concentration of methotrexate (prepared in pH 2 phosphate buffer), was added to 5 mL of polymer solution, which was prepared by dissolving a different concentration of polymer (PLGA) in DCM. The mixture was sonicated for 5 min with a 250 W ultrasonicator (Magnapak-250 Libra ultrasonicator; Kolkata, India) to prepare a primary water/oil (w/o) emulsion. The emulsion was added to 50 mL of aqueous PVA solution with continuous stirring at maximum speed on a magnetic stirrer (MS-500; Remi Equipments, India). The water/oil/water (w/o/w) emulsion obtained was stirred for 1.5 h to allow solvent evaporation and microsphere formation.

Later, the suspension containing microspheres in PVA solution was kept in a vacuum chamber for 45 min. The microspheres were separated by centrifugation at 500 rpm for 5 min. The microspheres collected were washed with 15 mL of distilled water three times. The final product was dried under vacuum at a controlled temperature for 15 h. The dried microspheres were placed in a desiccator and stored in a refrigerator.

Determination of particle size and particle size distribution

The microspheres obtained as a suspension in aqueous PVA solution were subjected to particle size analysis using low angle laser light scattering (Sympatec GmbH, Germany). The sample was dispersed in deionized water before counting. An He-Ne laser tube of 632.8 nm with a minimum power of 5 MW was used as the light source. The particle size distribution was determined in terms of a SPAN factor expressed as:

SPAN5(D_90%2D_10%)/(D_50%)

where D_90%, D_10% and D_50% are the diameter sizes and the given percentage value is the percentage of particles smaller than that size. A high SPAN value indicates a wide size distribution and a high polydispersity.

Determination of percentage drug entrapment ( PDE )

PLGA microspheres (25 mg) were dissolved in DCM to prepare a 10 mL solution. The drug was extracted three times from DCM using 25 mL of phosphate buffer (pH 2). The concentration of the drug extracted was determined by UV spectrophotometry (UV-1601; Shimadzu Corp., Japan) at a predetermined l

_max

(308 nm).

In vitro release study

Drug loaded PLGA microspheres for each drug were subjected to an in vitro release test under static conditions as per the method used by Nam and Park with slight modification.

¹³

In brief, microspheres (75 mg) were dispersed in 30 mL of n-saline phosphate buffer (pH 7.4) in a stoppered conical flask. The system was kept in an incubator at 37 °C. After defined time intervals, 3 mL of the dispersion medium was drawn and filtered using filter paper. The residue was returned to the suspension and the clear filtrate was subjected to UV spectrophotometry (after dilution, whenever required) for determination of drug content released from the microspheres.

Figure 1 Effect plot for (a) X1 (drug concentration) and (b) X2 (polymer concentration) on PDEE of the microspheres.

Morphology

The morphological attributes of the microspheres were studied by scanning electron micrography (SEM). A small amount of microsphere solution was spread on an aluminium stub, which was placed in an SEM chamber (Philips, India). Photographs were taken at an acceleration voltage of 30 KV and a chamber pressure of 9.7 mmHg.

X - ray diffraction study

Sample preparation. Vacuum grease was applied over a glass slide. Sample (~100 mg) was sprinkled onto the slide to form a layer with a thickness of approximately 0.5 mm.

Instrument operation. All the experiments were performed on a Philips X'Pert X-ray diffractometer. The slide was placed vertically at an angle of 0° in the sample chamber. An X-ray beam of 2 KW crossed the sample. As the slide moved at an angle of 20°, the detector detected diffracted X - rays at an angle of 20°.

Stability study

One batch (batch 3, Table II) of PLGA microspheres containing methotrexate was selected for stability studies. The sample containing microspheres was placed in a 20 mL borosilicate glass ampoule. The mouth of the ampoule was closed tightly with aluminium foil to prevent air coming into contact with the microspheres. Six samples (placed in a dessicator) were stored in a refrigerator. One ampoule was drawn at the end of each week for 6 weeks. The microspheres in the drawn sample were subjected to the following studies:

• Particle size analysis.

• Drug content.

• DSC and TGA. DSC was performed (Pyris-1; Perkin Elmer, USA) by heating a weighed amount of sample from 0-380 °C at a rate of 10 °C/min. The sample was maintained at 40 °C for 1 min. TGA was performed with a thermal gravimetric analyser (TGA-7; Perkin Elmer) by heating a weighed amount of sample from 40-350 °C at a rate of 10 °C/min. The sample was maintained at 40°C for 1 min.

• Chemical analysis. Qualitative assay of methotrexate was done as per the Indian pharmacopoeial method. Microspheres containing methotrexate were added to the 250 mL of mobile phase. Chromatography was performed using a stainless steel column (10 cm36 mm) packed with stationary phase LC1 (5 µm) in acetonitrile; a mixture of 92 volumes of phosphate buffer pH 6.0 and 8 volumes of acetonitrile as the mobile phase with a flow rate of 1.4 mL/min; a detection wavelength of approximately 303 nm; and a fixed-volume loop injector.

• An SEM morphological study.

Data analysis

The effect of the individual variable process parameters and their interaction on the pharmaceutical characteristics was measured by fitting the data to a mathematical model using a non-linear least square regression (SPSS 9.05; SPSS, Inc., USA). The kinetics of the drug release were determined by simple regression.

Results and discussion PDEE

Percentage drug entrapment efficiency (PDEE) in the microspheres was found to increase with an increase in drug concentration and polymer concentration (Table II). The effect can be summarized by Equation 1 that was obtained by the non-linear least square regression of the data:

Table III shows the coefficient and probability of the effect of individual parameters as well as the interaction effect between the two process parameters. From the coefficient values and probability of significance (p), there is little interaction; that is, the influence of a process parameter on the PDEE is independent of the level of the other parameter. Figure 1 further illustrates the interaction between the two parameters. As the two curves are parallel to each other, there is little possibility of any interaction.

Figure 2 Effect plot of (a) X1 (drug concentration) and (b) X2 (polymer concentration) on microsphere particle size.

From the coefficients of the two parameters, it can be inferred that the effect of a change in drug concentration on PDEE is greater than a change in polymer concentration. An increase in PDEE resulting from an increase in the drug concentration is directly related to an increase in the drug content in the internal aqueous phase that gets entrapped by PLGA. As methotrexate is practically insoluble in water, there is very little loss of the drug from leaching.

Figure 3 SEMs of PLGA microspheres containing methotrexate. (a) Methotrexate loaded PLGA microspheres before exposure to stability conditions. (b) Methotrexate loaded PLGA microspheres after exposure to stability conditions.

An increase in the polymer concentration has been found to have an insignificant effect on PDEE. It has been reported that an increase in polymer concentration leads to an increase in PDEE.¹⁴ This has been ascribed to the decreased loss of the drug to the external aqueous phase that resulted from factors such as the increase in the viscosity of the oily phase separating the two aqueous layers. However, these factors will impart the effect only when the drug has a tendency to leach out to the external aqueous phase. This is possible if the drug has appreciable aqueous solubility. However, in the present case, the drug is practically insoluble in the external aqueous or oily phases. As the entire drug remains in the internal phase, an increase in the polymer concentration has no effect.

Mean particle size

The effect of drug concentration on particle size (D

_10%

) has been shown in Table II. Particle size is increased by both an increase in polymer concentration as well as an increase in drug concentration. The effect can be summarized by Equation 2:

From the values of the coefficient, it appears that the two factors affect particle size negatively. However, this is contrary to the present observation as particle size increased with an increase in one or both factors. This may be because of interaction between the two parameters (from the t-value obtained by fitting the data to a generalized additive model), rather than the individual process parameters (Figure 2). The sharp increase in particle size at higher values of drug and polymer concentrations may be a result of the combined effect of the two parameters.

Effect on polydispersity

The change in polydispersity of the microspheres (represented by SPAN in Table II), because of a change in one process parameter, was found to be dependent upon the level of the other parameter. An increase in any of the two parameters resulted in an increase in the polydispersity (at a lower level of the other parameter) or a decrease in the polydispersity (at a higher level of the other parameter).

Surface morphology

PLGA microspheres containing methotrexate were spherical with an irregular surface (Figure 3a). This may be because of a sudden precipitation of the drug present on the interphase of inner aqueous phase and the oily phase. The sudden precipitation results in the formation of particles that become embedded in the matrix of the PLGA microspheres. These particles give an irregular and rough appearance to the surface of the microsphere (Figure 3).

In vitro release

The release of methotrexate from the PLGA microspheres is shown in Figure 4. More than 50% of the drug was released during the first hour. Greater release of methotrexate was expected because of its higher aqueous solubility. To determine the mode of release of the drug from the microspheres, three graphs were plotted:

• drug remaining in the matrix versus time

• log ( drug remaining in the matrix ) versus time

• percentage drug released versus square root of time.

The curves obtained were regressed. The values of R² for the three plots were 0.91, 0.97 and 0.93, respectively, which infers that the release of the drug from the microsphere matrix follows first order kinetics. Low R² for percentage drug released versus square root of time indicates the absence of diffusion-mediated drug release. This fits well in relation to the solubility of the polymer and the drug.

Microsphere stability

Figure 5 shows the DSC curves of sample before and after the stability test. Similarly, Figure 6 shows the TGA curve of the sample before and after the stability test. The DSC curve of initial sample shows different endotherms between 270-340 °C. The endothermic presence in this region may be because of drug melting; dehydration of the residual solvents in the microspheres; or degradation of the polymer. Although the melting point of methotrexate (182-183 °C) is lower than the initiation temperature for the observed peak, the shift may be because of the existence of the drug in amorphous form or the presence of other constituents.

Figure 4 Line fit plot showing first order in vitro release of methotrexate from PLGA microspheres.

The thermal curves of the microspheres after the stability study show an endotherm in the same temperature range as the initial sample. However, the peak was broader with a higher ³H. An increase in ³H should be because of aging of the microspheres. A decrease in the weight of the sample (Figure 7) started at the same temperature as the initial sample, but the process came to an end at a lower temperature (~260 °C).

Figure 5 DSC curve of PLGA microspheres containing methotrexate (a) before exposure to stability conditions and (b) after exposure.

This means that the thermal events in the sample were very much similar before and after the stability test. However, the same reaction occurred vigorously after the exposure to stability conditions. It may be interesting to further investigate this observation.

The X-ray diffraction pattern of the microspheres before and after the stability test showed no peak for crystallinity (Figure 7). Thus, it can be deduced that there is no change in the amorphous state of the constituents of the microspheres. This is expected as the molecular mobility of the drug or the polymer at refrigerated temperatures must be too low for spontaneous crystal-lization. The microsphere system retains its initial amorphous state.

Figure 6 TGA curve of PLGA microspheres containing methotrexate (a) before exposure to stability conditions and (b) after exposure.

No prominent change in the particle size (D_10%) was found (Figure 8). This is further confirmed by the SEM of the microspheres after the completion of the stability test (Figure 3b). The particle size (D_10%) of the microspheres at the beginning of the stability study was a little higher than the particle size of the batch obtained during the optimization study, which is because of the different state of the samples when determining particle size. A chromatogram of the microsphere extract did not show any additional peak or prominent shift in the position of peak, indicating that the drug did not undergo any degradation during the stability test.

Figure 7 X-ray diffraction patterns of (a) drug, (b) PLGA microspheres containing methotrexate before stability test, (c) after stability test and (d) polymer.

Conclusion

This work studied the factorial effect of two selected process parameters on the pharmaceutical characteristics of PLGA microspheres containing methotrexate as a water insoluble drug. It was found that the PDEE and mean particle size of microspheres increased with a rise in polymer and drug concentrations.

The size distribution of the microspheres was dependent on the level of the parameters. At a low level of one parameter, an increase in the other parameter resulted in an increase in particle size distribution. However, at a higher level of one of the parameters, particle size distribution decreased with an increase in the other parameter. Other observations included:

• Microspheres were found to be spherical with an irregular surface.

• The release of the water insoluble drug from PLGA microspheres followed first order kinetics.

• Storage of microspheres at refrigerated temperatures led to an increase in T_g (ageing effect) that resulted in a broader endotherm in the thermal curve obtained by DSC.

No other significant changes were noted in the microspheres. Thus, it can be concluded that the microspheres can be stored at refrigerated temperatures without any significant change in their pharmaceutical characteristics.

Acknowledgements

The authors are thankful to Boehringer Ingelheim (Germany) and Biochem Pharma. Ltd (India) for providing the polymer and drug samples. The authors further express their gratitude to the All India Council of Technical Education and the Government of India for the grant awarded to R.H. Parikh for this work. Finally, the authors wish to thank all the members of SICART, CVM for their active participation in completing the analytical part of this project.

Figure 8 Change in particle size of PLGA microspheres during stability studies.

References

1. P.K. Gupta, C.T. Hung and F.C. Lam, "Applications of Particulate Carriers in Intratumoral Drug Delivery," in Pharmaceutical Particulate Carriers: Therapeutic Applications (Marcel Dekker, Inc., New York, New York, USA, 1999) pp 135-164.

2. T. Kato, "Encapsulated Drugs in Targeted Cancer Therapy," in S.D. Bruck, Ed., Controlled Drug Delivery Vol. II: Clinical Applications (CRC Press, Boca Raton, Florida, USA, 1983) pp 189-240.

3. P.K. Gupta and C.T. Hung, "Albumin Microspheres II: Applications in Drug Delivery," J. Microencapsulation 6(4), 463-472 (1989).

4. A.K. Singla and M. Chawla, "Chitosan: Some Pharmaceutical and Biological Aspects - An Update," J. Pharm. Pharmacol. 53(8), 1047-1067 (2001).

5. H. Okada, "One- and Three-month Release Injectable Microspheres of the LH-RH Superagonist Leuprorelin Acetate," Adv. Drug Deliv. Rev. 28(1), 43-70 (1997).

6. B. Hutchinson and B.J.A. Furr, "Biodegradable Polymer Systems for the Sustained Release of Polypeptides," J. Control. Rel. 13(2-3), 279-294 (1990).

7. C. Thomasin, H.P. Merkle and B. Gander, "Drug Microencapsulation by PLA/PLGA Coacervation in the Light of Thermodynamics. II. Parameters Determining Microspheres Formation," J. Pharm. Sci. 87(3), 269-275 (1998).

8. R. Jeyanthi et al., "Effect of Processing Parameters on the Properties of Peptide Containing PLGA Microspheres," J. Microencapsulation 14(2), 163-174 (1997).

9. F. Gabor et al., "Ketoprofen-poly(D, L-lactic-co-glycolic acid) Microspheres: Influence of Manufacturing Parameters and Type of Polymer on the Release Characteristics," J. Microencapsulation 16(1), 1-12 (1999).

10. B.A. Chabner et al., "Antineoplastic Agents," in J.G. Hardman et al., Eds., Goodman & Gilman's The Pharmacological Basis of Therapeutics, Ninth Edition (McGraw-Hill, Columbus, Ohio, USA, 1955) pp 1233-1288.

11. E.B. Denkbas, M. Seyyal and E. Piskin, "5-fluorouracil Loaded Chitosan Microspheres for Chemoembolization," J. Microencapsulation 16(6), 741-749 (1999).

12. R.H. Parikh et al., "Poly(D,L-lactide-co-glycolide) Microspheres Containing 5-FU: Optimization of Process Parameters," AAPS PharmSciTech. 4(2), Article 13 (2003).

13. Y.S. Nam and T.G. Park, "Protein Loaded Biodegradable Microspheres Based on PLGA-protein Biconjugates," J. Microencapsulation 16(5), 625-637 (1999).

14. M. Boisdron-Celle, P.H. Menei and P. Benoit, "Preparation and Characterization of 5-FU-loaded Microparticles as Biodegradable Anticancer Drug Carriers," J. Pharm. Pharmacol. 47, 108-114 (1995).