Influence of PLGA and PLGA-PEG on the dissolution profile of oxaliplatin

Oxaliplatin was inserted into polymeric matrices aiming to study the interaction of this drug with these polymers and its capability to diffuse to the environment. Tested polymers were: (1) polyethylene glycol (PEG), (2) poly(lactic-co-glycolic acid) (PLGA), and (3) a copolymer of them (PLGA-PEG). The latter two were synthesized by us using polycondensation in bulk. Oxaliplatin was included in the matrices by the melt mixing process followed by casting. Fourier tran sform infrared spectroscopy (FTIR), proton nuclear magnetic resonance (1H-NMR) and X-ray diffraction (DRX) studies of the polymers were performed proving the obtaining of the desired materials. In addition, the interaction between drug and matrices and the release profile of the oxaliplatin from these matrices were analyzed. Among them, PEG did not control the oxaliplatin release. In turn, PLGA and PLGA-PEG present drug release profiles quite similar. Oxaliplatin was completely released from PLGA and PLGA-PEG in 5 hours, by a relaxation mechanism. There was no evidence of oxaliplatin interaction with the different polymers. In addition, as the PEG improves the biocompatibility and biomasking, obtained results prove the obtaining of a drug release system, which allowed the total use of the drug improving the cancer treatment and even the welfare of the patients.


Introduction
More than 10 million people are diagnosed with cancer annually, making this disease a leading cause of death.Furthermore, new projections allow predicting that by 2020, there will be 15 million new cases of cancer every year [1,2] .Nowadays, the third most common cancer worldwide is the colorectal cancer and this is also the fourth most common cause of death [3][4][5][6] .
Oxaliplatin is a third-generation platinum-based chemotherapeutic agent, which presents a remarkable success on the cancer treatment, as the other platinum-based drugs, such as cisplatin and carboplatin [7] .Specifically, oxaliplatin is a organoplatinum compound with significant activity against advanced or metastatic digestive tumors, mainly colorectal ones [8] .Therefore, oxaliplatin is considered one of the first-line drugs in the treatment of advanced colorectal cancer [9][10][11][12] .In addition, oxaliplatin does not induce nephrotoxicity to the same extent as these other platinum-based drugs [13] .On the other hand, recent reports criticize the oxaliplatin use due to some cases of peripheral neuropathy caused by cumulative doses of this drug [13,14] .Drugs entrapped into biodegradable and biocompatible polymeric matrices have been successfully employed for the controlled release of drugs [15][16][17][18][19] .Potentially, these systems are able to maintain drug concentrations within therapeutic ranges even along days.Consequently, it may diminish side effects caused by high concentrations and repeated administrations, and improve patient compliance as compared to conventional regimens [20] .Poly(lactic-co-glycolic acid) (PLGA) copolymers [21] have been extensively used in drug delivery systems (DDS) as alternative to improve conventional formulations [21][22][23] .These DDSs allow extended and controlled releases of drugs dispersed into the polymer [24] .However, PLGA hydrophobicity may be a problem for some drugs.Incomplete release is usual if drugs have interactions with them [25] .The attaching of PEG chains to PLGA may solve these situations [26,27] .The inclusion of PEG into material could also be convenient for improving biocompatibility of the systems [28] .However, high solubility of PEG could accelerate the drug release rate which in some cases it is an undesirable effect.
In this paper, the effect of the hydrophilicity of the polymeric matrix on the release profiles of oxaliplatin was studied.In addition, possible interactions between the drug and the matrices were investigated.As expected, PEG did not act as a release controlling matrix for oxaliplatin.However it is possible to exploit the biocompatibility provided by the PEG as part of PLGA-PEG copolymer because its presence does not change the drug release profile.Oxaliplatin was completely released from both PLGA and PLGA-PEG in five hours.

Materials
PEG (Mn 6000) was purchased from SIGMA-ALDRICH, Brazil.The oxaliplatin was gently provided by Center for Research and Development of Medicine, Havana, Cuba.All others chemical used were purchased from VETEC, Brazil.All chemical materials were used as received.

Synthesis of block copolymers
The PLGA block copolymer was synthesized by polycondensation in bulk from lactic acid (50 ml) and glycolic acid (62 ml) in equimolar ratio (1:1).Sulfuric acid (98% v/v, 0.25mL) was added as catalyst, in a closed system under nitrogen atmosphere and slight vacuum.Temperature was kept at 140°C under magnetic stirring for 10 h [29] .The PLGA-PEG block copolymer was synthesized following the same procedure, with addition of commercial PEG (5g) [27] .

Preparation of polymer/drug systems
The three polymers (PEG; PLGA; PLGA-PEG) were molten separately at 120°C.This temperature was chosen since it is considerably lower than the degradation temperature of the drug which is 400°C according to the results of the TGA analysis.Oxaliplatin was incorporated in the three polymers in a 1% w/w concentration by melting at 120°C inside a 50mL reactor using mechanical stirring and dry N 2(g) flow.After the drug incorporation, melted mixture was casted, producing disks (diameter: 1.0 cm; height: 0.5 cm).The DDS prepared were named as PEGO, PLGAO and PLGA-PEGO, respectively.

Materials characterization
Hydrogen nuclear magnetic resonance ( 1 H-NMR) and Fourier transform infrared (FT-IR) were used to study the composition of the copolymers.Samples were dissolved in CDCl 3 and their 1 H NMR spectra were recorded on a Varian  equipment model Oxford 300.FTIR spectra were obtained from neat films cast from the chloroform sample solutions on KBr tablets, with a Varian equipment model 3100 FTIR Excalibur Series, using a resolution of 4 cm -1 and 20 scans from 4000 to 400cm -1 .X-Ray Diffraction (XRD) was also used to characterize the copolymers, the samples prepared and the drug.The equipment used was a Rigaku Miniflex X-ray diffractometer in a 2θ range from 2° to 80° by the method FT (fixed time).The steps were equal to 0.05°/s, using a tube voltage and current equal to 30 kV and 15 mA, respectively.The radiation used was CuKα = 1.5418Å.

Presence of oxaliplatin in matrices
The technique of least squares was used to compare the FTIR data obtained from the DDS containing oxaliplatin with the one of matrices without the drug.Specifically, the coefficient of determination (R 2 ) and the root mean square error (RMSE) were used as indicatives of the presence of oxaliplatin into the matrices.This test was reported by our group elsewhere [30] .

Dissolution test of oxaliplatin
For each test, one gram of PEGO, PLGAO and PLGA-PEGO was studied using the USP Apparatus I at 75 rpm.First, the released amount of the drug was analyzed in HCl 0.1 M (900 ml) by one hour.Soon afterwards, this solution was substituted by the same volume of phosphate buffer pH 7.8 and drug elution was determined by other 4 hours.Samples were collected at 0.25h; 0.5h; 1h; 1.25h; 1.5h; 2h; 3h; 4h and 5h.All these tests were performed in triplicate.
Ultraviolet-visible spectrophotometry (UV-Vis) was used to quantify the oxaliplatin in the samples.The measurements were performed at 260 nm using a monochromator Biospectro spectrophotometer model SP-220.The spectra of the solutions with known concentrations of oxaliplatin were collected and used to set up an analytical curve.

Results and Discussion
Figure 1 shows the 1 H-NMR spectrum of the PLGA copolymer.The peak at 1.5 ppm is associated with the repetitive methyl groups of the lactic acid.The peaks at 5.2 ppm and 4.8 ppm are related to the CH-CH 3 from lactic acid and CH-H from glycolic acid, respectively, confirming the synthesis of the copolymer [27,29] .The molar amount of each comer of the copolymer was calculated using the relationship among the peak areas and the number of H atoms.The synthesized copolymer presented 46.6mol% of their molar mass related to the glycolate comer, while the lactate comer corresponded to 53.4mol% of the copolymer.
Figure 2 shows the 1 H-NMR spectrum of the PLGA-PEG copolymer.In this spectrum, same peaks of PLGA are present and the new peak at 3.64 ppm is assigned to the methylene groups present in PEG [27,[31][32][33] .Again, the peak areas were used to estimate the molar amount of PEG in the system PLGA-PEG.Calculated value is around 5.9mol%, while the molar mass related to the glycolate and lactate comers corresponds to 45.7mol% and 48.4mol%, respectively.
FTIR spectra of PEG, PLGA and PLGA-PEG are shown in Figure 3.The FTIR spectrum shown in Figure 3a corresponds to PEG.The smooth wide band around 3500 cm -1 is related to the oxygen atom able to form hydrogen bonds.The characteristic band at 2888 cm -1 is attributed to the stretching of CH, CH 2 and CH 3 groups.The characteristic bands at 1464 and 1343 cm -1 are associated with the C-H bending.The stretching of C-C-O group appears at 1280 and 1100 cm -1 .The later characteristic band is conjugated with the C-O-C group.The harmonic bands of C-C-O group appear around 950 and 840 cm -1 .The spectrum shown in Figure 3b corresponds to PLGA copolymer, again is possible to see a smooth wide band around 3500 cm -1 related to the oxygen capability to form hydrogen bonds.The small doublet around 3000 cm -1 is related to CH, CH 2 and CH 3 groups.The characteristic peak at 1752 cm -1 is attributed to the stretching of the C=O.While the characteristic band at 1182 cm -1 can be assigned to the ether group.The characteristic bands at 1130 cm -1 and 1452 cm -1 are attributed to C-O-C group and C-H bond of the methyl group, respectively [32,[34][35][36][37] .The Figure 3c shows the PLGA-PEG spectrum.The spectrum profile is very similar to the PLGA one, which can be explained by the small amount of PEG in the system, around 5.9mol%.As showed by the spectra, the PLGA characteristic bands are stronger than PEG, since the former corresponds to more than 90% of the block copolymer.Therefore, the PEG presence can be barely noticed by the small signal of the C-C-O harmonics, placed around 950 and 840 cm -1 .
The FTIR spectra of oxaliplatin and the samples PEGO, PLGAO, PLGA-PEGO are showed in Figure 4.In the spectrum of oxaliplatin (Figure 4a), the peaks at 3264 and 3509 cm -1 confirm the presence of an NH stretch, and peak at 812 cm -1 shows N-H bending; C=O stretch was observed at 1707 cm -1 .In spectra of matrices combined with oxaliplatin (Figure 4b, 4c and 4d), signals corresponding to drug were not observed, maybe due to its low concentration in the mixtures [38] .
Due to small differences among the FTIR spectra of the matrices in comparison to the ones containing oxaliplatin, the RMSE statistical tool was used to prove the presence of the drug in the materials.These differences can be expressed using appropriate statistical tools.Among them, the root mean squared error (RMSE) is very useful to study the misfit between experimental data and model [30,39] .Therefore, RMSE was calculated through linear regressions between absorbances of the matrix and analogous filled with oxaliplatin, using the least squares approach [40] .The obtained results are shown in Figure 5.As reference, absorbance of the PLA was plotted as a function of its own absorbance and RMSE is, obviously, equal to zero.In turn, absorbances of PEGOxPEG, PLGAOxPLGA and PLGA-PEGOxPLGA-PEG produced RMSE values equal to 4.79, 3.27 and 3.11, respectively.In spite observed differences, probably produced by the chemical differences among tested materials, all of these results were obtained with p<0.05, indicating that matrices were successfully loaded with oxaliplatin.
Figure 6 shows the XRD analysis for polymers and DDS.The diffractogram of PEG (Figure 6a) exhibits two peaks indicative of crystallinity: 19.2°(2θ) and 23.4° [ 41] .The diffractogram of pure oxaliplatin (see Figure 6d) exhibits many characteristic peaks due to its crystalline nature [38] .The diffractograms of PLGAO and PLGA-PEGO (see Figure 6f and g), respectively) show amorphous halos of the copolymer and the characteristic peaks of the drug.In turn, the XRD of the PLGA-PEG (Figure 6c) also shows only amorphous halos due the low proportion of the PEG in the copolymer (3.5%) and PLGA (Figure 6 (b)) presents only amorphous halos which are in agreement with the literature [34] .Dissolution tests of oxaliplatin-loaded DDS were monitored by UV-Vis analysis.Results are shown in Figure 7.Total drug release from PEG matrix (Figure 7a) took place in one hour due to its high hydrophilic character.The PLGA (Figure 7b) and PLGA-PEG (Figure 7c) matrices control effectively the release rate of oxaliplatin, which was completely delivered in five hours.At the end of study disks were completely degraded.The experimental data corresponding to these samples was fitted using the model presented in the Equation 1: In Equation 1, RAO is the released amount of oxaliplatin, A 1 is the lower RAO limit, A 2 is the higher RAO limit, x 0 is the inflexion (half amplitude) point and dx is the width.
Results are showed in Table 1.All the parameters are statistically equal, confirming that oxaliplatin is released from PLGA and PLGA-PEG matrices in the same way.
The knowledge about the drug release mechanism and release rates is very important for programming the characteristics of the systems.Consequently, mathematical modeling of drug release processes from polymeric matrices plays an important role to predict the behavior of the system and to determine the structural parameters of the polymer that affect and control the drug release profiles.The drug release can occur through a pure Fickian diffusion, by  processes of relaxation and sometimes a combination of these two mechanisms takes place [42] .When the semi-empirical equation of Ritger-Peppas, [43] , is used for fitting the experimental data, the diffusional exponent (n) indicates the mechanism for drug transport.An exponent value for cylindrical geometric n = 0.45 indicates Fickian diffusion, n within 0.45-0.89indicates the anomalous transport, n = 0.89 indicates case II transport and n more than 0.89 indicates super case II transport.An anomalous transport (0.45<n<0.89) is referred to a drug transport by diffusion and relaxation of the polymer chains [43] .
The polymeric materials that controlled the release of oxaliplatin (PLGA and PLGA-PEG) have a value for Tg lower than the temperature for running the dissolution tests [44] .In that sense it is reasonable to suppose that drug transport could occur by a combination of both: diffusion and relaxation.Thus, for obtaining a preliminary idea of the release mechanism for oxaliplatin included in these disks, experimental data was fitted to theoretical release profiles given by a Peppas-Sahlin equation (Equation 2), which combines the Fickian diffusion and non-Fickian movement of drug molecules through polymer chains [45] .
In Equation 2, k 1 and k 2 are Fickian and relaxational contribution respectively [46] .They allow calculating the approximate contributions of the diffusional and relaxational mechanisms.
Experimental modeling was performed using the data from drug released until one hour, which corresponded to 60% of the total drug contained in DDS.The exponent m was assumed equal to 0.44 due to geometry of the samples, which were disks.For PLGAO, k 1 was -0.77 ± 0.01 and k 2 was 0.91 ± 0.01; for PLGA-PEGO, k 1 was -0.93 ± 0.02 and k 2 was 1.05 ± 0.02 (R 2 in both cases was greater than 0.99).A negative value of k 1 was obtained in both cases, which can be considered as an insignificant effect of Fickian diffusion on drug release compared to the relaxation process.This result can be related to Tg of the polymers which are lower than 37ºC.In that sense when disks are placed into release fluid the combination of heating and the swelling in water could favor the polymer change from glassy to rubbery state.Consequently, in this case, relaxation of the polymer is the driving force for the release of oxaliplatin.

Conclusions
This work presented the release of oxaliplatin from PEG, PLGA and PLGA-PEG matrices.PLGA and PLGA-PEG were obtained by polycondensation and the synthesis of products was confirmed by 1H-NMR, FTIR and XRD.The release studies, monitored by ultraviolet analysis, showed that the PEG matrix cannot sustain the release for more than one hour, maybe due to the high solubility of this polymer.In contrast, the PLGA and PLGA-PEG copolymers release oxaliplatin during 5 hours in a similar way.So, the modification of PLGA using PEG did not modify oxaliplatin release.In addition, the characterization of materials by FTIR and the complete release of oxaliplatin from DDS allow believing that there are not interactions between drug and copolymers while the release mechanism seems to be due to relaxation of the polymer chains.Therefore, the PLGA-PEG copolymers could be preferred for designing controlled release systems for oxaliplatin with enhanced biomasking and biocompatibility.
Table 1.Parameters of the dissolution curves of oxaliplatin released from PLGA and PLGA-PEG matrices.