Formulation Design,Optimization,and In Vitro Evaluation of Palbociclib-Loaded Polymeric Nanoparticles

Abstract

Palbociclib (PLB), an advanced breast cancer drug, shows limited success in clinical scenarios due to toxicity and resistant antitumor activities. The article aimed at preparing folic acid-modified chitosan-based polymeric nanoparticles (PNS) for controlled and targeted release of PLB. PLB-loaded nanoparticles were prepared via solvent evaporation. Central composite design was adopted in optimizing the formulation, with concentration levels of chitosan and polyvinyl alcohol (PVA) making responses of size and zeta potential (ZP) its independent variables. Mean particle size (MPS) ranged from 376 to 412 nm, and ZP ranged from 29.87 to 36.76 mV. Quadratic models were significant for both responses (MPS: F = 9.81, p = 0.0128; ZP: F = 5.24, p = 0.0466), with a nonsignificant lack of fit, confirming model adequacy. Particle size was significantly influenced by chitosan and PVA quadratic terms, while ZP was primarily affected by PVA. Optimization predicted a formulation with 63.35 mg of chitosan, 20 mg of PVA, an MPS of 389.1 nm, and a ZP of 31.2 mV, closely matching experimental observations (MPS of 237.8 ± 1.76 nm and ZP of 32.09 ± 3.38 mV). Entrapment efficiency and drug loading were 81.21 ± 1.80% and 40.79 ± 1.98%, respectively. PLB-PNS exhibited sustained, pH-responsive release, releasing 70.48% at pH 5.4 and 50.77% at pH 7.4 at 12 h, while free drug released 97.20% and 91.52%, respectively, confirming controlled and tumor-microenvironment-responsive drug delivery. Drug release followed Korsmeyer–Peppas kinetics, indicating diffusion-controlled release. Scanning electron microscope analysis revealed smooth, spherical nanoparticles. These developed PLB-loaded nanoparticles were indicated as an attractive tumor microenvironment-responsive drug-delivery carrier system capable of achieving enhanced therapeutic outcomes in breast cancer therapy.

Keywords

chitosan nanoparticle palbociclib pH-responsive drug release tumor-microenvironment surface analysis

INTRODUCTION

Breast cancer is one of the most common cancers in women that is common in the United States and stands as the very second chief cause of female mortality. Approximately 250,000 new cases of breast cancer are reported among women each year in the United States.^1,2 Globally, 1 in 20 women develop breast tumors, with breast cancer being the most common in high-income countries, affecting 1 in 8 women.³ Breast cancer biomarkers include estrogen receptor (ER), progesterone receptor, and human epidermal growth factor receptor 2, leading to classifications such as triple-positive (all overexpressed) or triple-negative (none). Most patients (83%) have hormone receptor-positive tumors that respond to hormonal therapy to inhibit tumor growth.⁴ Approximately 80% of breast cancers are ER-positive, with ERs overexpressed and widely distributed on nuclear and plasma membrane sites, compared to less than 10% in normal breast cells.⁵ Postchemotherapy, tumor removal often involves surgery, followed by hormone, radiation, or advanced therapies such as immunotherapy and gene therapy to prevent recurrence.⁶ Hence, it is absolutely imperative to design a combination program with two or more strategies so as to overcome their inadequacies while producing synergistic effects.⁷ It is very unfortunate that neurotoxicity and hypersensitivity are two of many severe side effects resulting from chemotherapy. The primary reason for this is the non-specific nature of chemotherapy, which results in damage to both healthy and cancerous cells.^8,9

To treat advanced breast cancer, Pfizer developed a chemotherapeutic drug named palbociclib (PLB) that has the desired effect of silencing cyclin-dependent kinases 4 and 6.¹⁰ PLB’s well-known side effects include neutropenia, anemia, and fatigue, with additional toxicity to the liver, lungs, kidneys, and reproductive organs due to off-target effects; its efficacy is further limited by p-glycoprotein-mediated drug efflux causing tumor drug resistance.^11–13 Researchers are indeed excited about different kinds of pharmaceutical nanotechnology because of their application in the treatment of cancer. The basic elements are, but are not limited to, improving the effectiveness of drug solubility or bioavailability, lowering normal-cell toxicity, and delivering drugs to specific locations in the body.¹⁴ Nanomedicine techniques such as polymer, SSLN, micelles, and silica nanoparticles have improved anticancer drug delivery, highlighting the need for innovative systems to target resistant and metastatic breast cancer cells.¹⁵ Kommineni et al. developed stealth liposomes with PLB, enhancing pharmacokinetics (1.45-fold area under the curve) and anticancer efficacy (1.63× cytotoxicity) against aggressive 4T1 breast cancer cells.^16,17 Hypoxic tumors have been reported to be resistant to therapy and less sensitive to radiation than normoxic tumors by threefold.¹⁷ So, a few suggestions in favor of therapy against advanced-stage breast tumors were offered. PLB-loaded polymeric nanoparticles (PNS) are also suggested in the therapy of refractory advanced-stage breast malignancies. The technique of formulation and characterization is imperative to the performance of nanoparticle-based drug delivery systems owing to the fact that it does influence physicochemical properties of the drug delivery system that indirectly do affect therapy outcomes and drug delivery effectiveness. Assessing the size, the shape, the surface charge, and related particularities of the nanoparticles is critical in achieving maximum effectiveness of nanoparticles that is exercised in the concept of drugs delivery. The process ensures that the nanoparticles are tailored to meet specific therapeutic needs which in turn makes them more efficient in delivering medicines where they are needed. Future investigations shall be focused on extended stability studies, in vitro cellular uptake, and cytocompatibility evaluations to better predict biological performance under physiological conditions. Mechanistic studies examining polymer–drug interactions at the molecular level will be particularly important, as advanced biophysical approaches such as in-cell or intracellular interaction analyses have been shown to provide critical insight into native bimolecular behavior. Such studies could strengthen understanding of formulation behavior within the cellular environment and support rational optimization.¹⁸ In vivo evaluation and translational feasibility should be addressed in future work to assess therapeutic efficacy, biodistribution, and safety. Scaling-up feasibility and process reproducibility must also be explored to establish formulation robustness for clinical translation. In this context, emerging genome and RNA-based therapeutic strategies demonstrate that sustained and targeted molecular interventions can be achieved when delivery systems are well-characterized and biologically compatible. Integrating these considerations will enhance the translational potential of the developed formulation and support its progression toward clinical application.¹⁹

MATERIALS AND METHODS

Natco Pharma provided PLB as Loba Chemie Pvt. Ltd. supplied chitosan and ethyl cellulose. For the natural solvents such as acetone, ethanol, and polyvinyl alcohol (PVA), we went the extra mile and bought them in packaged containers supplied by Himedia Laboratories Pvt. Ltd. with codes assigned to the PVA containers.

Method

Drug–excipients interactions by Fourier-transform infrared spectroscopy

Fourier-transform infrared spectroscopy (FTIR) was used to analyze PLB, chitosan, PVA, their mixtures, and lyophilized formulations (with and without drug) in a 1:1 ratio over 4,000–400 cm⁻¹ using a Broker Lab India Alpha 11 Spectrometer.²⁰

Experimental design

Formulation optimization was performed using response surface methodology (RSM) with a randomized central composite design (CCD) in Design-Expert^® software (version 11.1.2.0). Chitosan (20–100 mg) and PVA (10–20 mg) were selected as critical variables and studied at coded levels from −1 to +1, with formulation composition detailed in Table 1 . The design comprised 11 experimental runs without blocking and applied a quadratic polynomial model to evaluate linear, interaction, and curvature effects. Mean particle size (MPS, nm) and zeta potential (ZP, mV) were selected as the primary responses for nanoparticle characterization.²¹

Table 1.

Central Composite Design Matrix Showing Formulation Composition, Independent Variables, and Observed Responses for Palbociclib-Loaded Nanoparticles

Run	Factor A: Chitosan (mg)	Factor C: PVA (mg)	Folic acid (mg)	Acetone (mL)	Acetic acid (mL)	Distilled H₂O (mL)	Response 1: MPS (nm)	Response 2: ZP (mV)
F1	100	15	10	5	10	Qs	402	34.5
F2	60	15	10	5	10	Qs	376	33.47
F3	20	10	10	5	10	Qs	412	36.76
F4	20	15	10	5	10	Qs	398	35.65
F5	60	15	10	5	10	Qs	389	34.67
F6	60	15	10	5	10	Qs	382	36.65
F7	20	20	10	5	10	Qs	411	32.41
F8	60	10	10	5	10	Qs	401	31.2
F9	100	10	10	5	10	Qs	412	29.87
F10	60	20	10	5	10	Qs	389	30.94
F11	100	15	10	5	10	Qs	399	32.12

Independent variables	Unit	Low (−1)	High (+1)	Dependent variables	Unit	Goal
Chitosan	mg	20	100	Mean particle size	nm	Minimum
Polyvinyl alcohol	mg	10	20	Particle charge	mV	Minimum

MPS, mean particle size; PVA, polyvinyl alcohol; ZP, zeta potential.

Formulation of PLB Polymer-Based Nanoparticles

All 11 experimental batches (F1–F11) that were produced using a modified solvent evaporation approach were accompanied by ionic gelatin. The procedure was followed in a uniform manner, and the amounts of chitosan and PVA were the only differences made as per the experimental design.

In other words, the necessary amount of chitosan (20–100 mg, depending on the specific batch) was dissolved in 10 mL of 1% v/v acetic acid under magnetic stirring at 800 rpm and 25 ± 2°C until a clear solution was obtained. PVA (10–20 mg) was dissolved in distilled water first at 70°C and then cooled to room temperature; this was done as it acted as a stabilizer. Afterward, 10 mg of folic acid was put into the chitosan solution and stirred at 600 rpm for 2 h for the purpose of the uniform hydration of the polymer. PLB (75 mg) was dissolved in 5 mL of acetone and poured with the help of a funnel into the water phase with polymer, while at the same time, a magnetic bar was continuously mixing the systems at 1,000 rpm, thus enabling the formation of an oil-in-water emulsion. For solvent evaporation and nanoparticle formation to occur completely, the emulsion was further stirred at 800 rpm for 4 h at 25°C. The next step was the introduction of ionic gelatin through sodium tripolyphosphate solution (0.5 mL) addition in gelatin being carried out drop by drop and with the system still being agitated at 500 rpm from another direction, thus leading to electrostatic cross-linking of chitosan and the stabilization of the nanomatrix over a period of time.

The volume of the final dispersions was corrected by the addition of distilled water and left to mix for 30 min, and then it was subjected to centrifugation at 15,000 rpm for 30 min at 4°C. The nanoparticles were thus collected, washed, dispersed, and stored at 4°C for later processing. The job was done in a way that maintained the consistency of the product and allowed a systematic evaluation of run-time factors for all 11 runs.²¹

Evaluation of PLB-PNS

Analysis of particle size charge

The particle size and ZP of PLB-loaded nanoparticles (PNs) were analyzed using a Zetasizer Nano-ZS (Malvern, UK) with a 633 nm He-Ne laser. The measurements were performed in phosphate buffer saline (PBS; 1 mg/mL at pH 7.2–7.4), averaging 10 and 15 readings, respectively, in triplicate. Mean values were calculated using automated analysis based on electrophoretic mobility and the Smoluchowski model.

Drug loading capacity

To determine drug loading, the PNP formulation was centrifuged to separate the dispersed phase from the continuous phase. The supernatant was collected, and the released drug was assayed spectrophotometrically at 261 nm. The precipitate particles were filtered using Whatman filter paper, washed with water, and then accurately weighed.²¹ The percentage of drug loading was calculated using the following equation:

% Drug loading = \frac{weight o f drugs i n NPs}{weight o f NPs} \times 100

Determination of Percentage Entrapment Efficiency of PNS

Entrapment efficiency (EE) was determined by an indirect method. The nanoparticle suspension was centrifuged at 15,000 rpm for 30 min at 4°C, and the free drug in the supernatant was filtered (0.22 µm) and quantified by ultraviolet–visible spectrophotometry at 342 nm using a calibration curve. The nanoparticle pellet was washed, lyophilized, and weighed to obtain drug-loaded nanoparticles.²¹ Analysis of free drug content was utilized to determine EE using the following equation:

% EE = \frac{(Total drug i n formula - free drug)}{Total drug i n formula} \times 100

Drug Release In Vitro

This experiment evaluated in vitro drug release of PLB from nanoparticles in PBS pH 7.4 and K acetate buffer pH 5.4 with PLB release analysis through dialysis bag diffusion. Nanoparticles containing PB, equal to that in 15 mg of PB, were put into a dialysis bag (1 kDa, LA387), which was then specifically and then placed in a graduated bottle and surrounded by 100 mL of releaser medium. The complete system was used at the temperature of 37 ± 5°C with a stirrer rate of 100 rpm to maintain the uniformity of the medium. After each sampling time point that was set beforehand from the receiver compartment, the same volume of fresh releaser medium was replaced.²²

In Vitro Drug Release Kinetics Model Fitting

Several kinetic models were used to predict drug release processes, including zero-order, first-order, Hixson–Crowell, Korsmeyer–Peppas (K–P), Higuchi, and Hixson–Crowell models by fitting the cumulative drug release data into DD Solver software.²³

Scanning Electron Microscope

Scanning electron microscopy (ESEM EDAX XL-30) was used to characterize PNS. Freeze-dried samples were stored in a gold-coated graphite sample container. The chamber was maintained at a pressure of 0.6 mm Hg and a voltage of 20 kV. Scanning electron photomicrographs were taken at various magnifications.²⁴

RESULT AND DISCUSSION

Construction of the Calibration Curve of PLB

A calibration curve was constructed by plotting absorbance on the abscissa against concentration on the ordinate in the range of 5–40 µg/mL. The plot showed a linear relationship with a 0.997 regression coefficient (R²), and thus confirmed the compound obeying Beer–Lambert law, as shown in Figure 1 .

Fig. 1.

Graph plot of a standard curve of palbociclib.

Drug–Excipients Interaction

FTIR analysis confirmed the structural stability and compatibility of PLB with formulation excipients. Pure PLB Figure 2A showed characteristic N–H stretching (3,400–3,300 cm⁻¹), aromatic and aliphatic C–H stretching (3,060–2,850 cm⁻¹), and prominent C = O/C = N stretching bands (1,650–1,600 cm⁻¹), along with aromatic C = C and C–N vibrations confirming drug identity. The PLB–excipients spectrum Figure 2B retained all major drug peaks, with slight band broadening at 3,500–3,200 cm⁻¹ due to overlapping O–H and N–H stretches of chitosan and PVA. Bands at 1,650–1,550 cm⁻¹ reflected combined drug carbonyl and chitosan amide vibrations, while polymer-specific C–O–C and C–O stretches appeared at ∼1,150 and 1,050 cm⁻¹, respectively. The absence of new peaks or significant shifts indicates no chemical interaction, confirming drug stability and excipient compatibility.

Fig. 2.

Drug–excipient interaction study by FTIR spectroscopy: (A) pure palbociclib and (B) palbociclib with excipients, showing retention of characteristic drug peaks mixture. FTIR, Fourier-transform infrared spectroscopy.

Design of Experiments

Statistical analysis of particle size and particle charge results from the designed experiment

Particle size (Y1) ranged from 376 to 412 nm, and ZP (Y2) ranged from 29.87 to 36.76 mV. A quadratic regression model with analysis of variance (ANOVA) analysis (via Design-Expert software) showed the highest F-values for both responses, confirming it as the best-fit model. Results are shown in Table 2 .

Table 2.

Analysis of Variance for the Quadratic Models Showing the Effects of Chitosan (A) and Polyvinyl Alcohol (B) on Mean Particle Size and Zeta Potential of Palbociclib-Loaded Nanoparticles

Source	MPS (nm) sum of squares	Df	Mean square	F-value	p-Value	Significance	ZP (mV) sum of squares	Df	Mean square	F-value	p-Value	Significance
Model	1347.25	5	269.45	9.81	0.0128	Significant	46.73	5	9.35	5.24	0.0466	Significant
A—chitosan	10.67	1	10.67	0.388	0.5604	NS	11.56	1	11.56	6.48	0.0515	NS
B—PVA	112.67	1	112.67	4.10	0.0987	NS	0.93	1	0.93	0.52	0.5029	NS
AB	36.00	1	36.00	1.31	0.3040	NS	10.89	1	10.89	6.11	0.0565	NS
A²	606.57	1	606.57	22.09	0.0053	Significant	2.40	1	2.40	1.35	0.2981	NS
B²	277.90	1	277.90	10.12	0.0245	Significant	23.27	1	23.27	13.05	0.0153	Significant
Residual	137.30	5	27.46	—	—	—	8.92	5	1.78	—	—	—
Lack of fit	52.63	3	17.54	0.41	0.7627	Not significant	3.76	3	1.25	0.49	0.7263	Not significant
Pure error	84.67	2	42.33	—	—	—	5.16	2	2.58	—	—	—
Cor total	1484.55	10	—	—	—	—	55.65	10	—	—	—	—

Effect of Particle Size (Y1)

A quadratic model best described the effect of formulation variables on MPS and was statistically significant (F = 9.81, p = 0.0128), showing strong agreement between experimental and predicted values (R² = 0.9075). The lack of fit was nonsignificant (p = 0.76), confirming model adequacy. While linear and interaction effects were insignificant, the quadratic terms A² (chitosan, p = 0.0053) and B² (PVA, p = 0.0245) significantly influenced particle size, indicating a pronounced concentration-dependent curvature effect. The contour plot ( Fig. 3A ) illustrates the combined influence of chitosan and PVA on MPS, while the 3D surface plot ( Fig. 3B ) clearly depicts the nonlinear response behavior. Model fitting statistics are summarized in Table 3 , confirming the reliability of the optimization process.

MPS = 383.211 + - 1.33333 * A + - 4.33333 * B + - 3 * AB + 15.4737 * A^{2} + 1 0. 4737 * B^{2}

Fig. 3.

Response surface analysis of mean particle size (MPS): (A) contour plot and (B) 3D surface plot showing the combined quadratic effect of chitosan and PVA concentrations on particle size, demonstrating a significant nonlinear response.

Table 3.

Statistical Model Fit Parameters for Mean Particle Size and Zeta Potential

Fit statistic	Response 1: MPS (nm)	Response 2: ZP (mV)
Standard deviation	5.24	1.34
Mean	397.36	33.48
C.V. (%)	1.32	3.99
R²	0.9075	0.8398
Adjusted R²	0.8150	0.6795
Predicted R²	0.5476	0.2222
Adeq precision	8.2548	7.1627

Adeq Precision, adequate signal-to-noise ratio; R², regression coefficient.

Effect of Particle Charge (Y2)

A quadratic model adequately explained the influence of formulation variables on ZP and was statistically significant (F = 5.24, p = 0.0466), with a nonsignificant lack of fit (p = 0.73), confirming model adequacy. ZP was predominantly influenced by the quadratic term of PVA concentration (B², p = 0.0153), indicating a strong nonlinear dependence on stabilizer level, while linear and interaction effects were insignificant. The contour plot ( Fig. 4A ) illustrates the combined effect of chitosan and PVA on ZP, highlighting regions of increased surface charge, whereas the 3D surface plot ( Fig. 4B ) depicts the predicted curvature of the response. The adequate signal-to-noise ratio (Adeq Precision = 7.16) supports the suitability of the model for navigating the design space. Model fitting details are summarized in Table 3.

ZP = 34.5984 + - 1.38833 * A + - 0. 393333 * B + 1.65 * AB + 0. 973947 * A^{2} + - 3 .0 3105 * B^{2}

Fig. 4.

Response surface analysis of zeta potential (ZP): (A) contour plot and (B) 3D surface plot showing the combined quadratic effect of chitosan and PVA concentrations on particle surface charge, demonstrating a significant nonlinear response. PVA, polyvinyl alcohol.

Optimization by Desirability Function

Formulation optimization was performed using a desirability-based approach with the explicit goal of minimizing MPS and ZP while maintaining chitosan (20–100 mg) and PVA (10–20 mg) within practical formulation ranges. Equal weights and importance were assigned to all variables to ensure balanced optimization. MPS was minimized within the range of 376–412 nm to obtain nanoscale particles suitable for tumor penetration, while ZP was minimized within 29.87–36.76 mV to ensure electrostatic stability without excessive surface charge. The desirability ( Fig. 5A ) and overlay plots ( Fig. 5B ) identified the optimal design space satisfying all constraints and guided formulation selection. Six feasible optimized solutions were generated with desirability values ranging from 0.557 to 0.716 ( Table 4 ). Solution 1 was selected as the optimized formulation due to its highest desirability (0.716), comprising 63.35 mg of chitosan and 20.00 mg of PVA, and predicting an MPS of 389.10 nm and ZP of 31.20 mV. Solutions 2 and 3 exhibited similar desirability values, indicating a stable optimization region with minimal sensitivity to small compositional variations. Lower desirability observed in solutions with reduced chitosan or PVA content was associated with increased particle size and surface charge, highlighting the critical stabilizing role of PVA. Overall, the desirability-based optimization identified a robust formulation region enabling effective control of nanoparticle size and surface charge, supporting further experimental validation.

Fig. 5.

Optimization by desirability function: (A) desirability plot and (B) overlay plot showing the optimal chitosan and PVA concentrations for simultaneous minimization of mean particle size and zeta potential.

Table 4.

Optimization Criteria, Constraints, and Desirability-Based Optimized Formulations for Palbociclib-Loaded Polymeric Nanoparticles

Category	Solution No.	Goal/value	Lower limit	Upper limit	Weight (L–U)	Importance	Chitosan (mg)	PVA (mg)	MPS (nm)	ZP (mV)	Desirability	Remark
Optimization Criteria	A: Chitosan	In range	20	100	1–1	3	—	—	—	—	—	—
	B: PVA	In range	10	20	1–1	3	—	—	—	—	—	—
	MPS	Minimize	376	412	1–1	3	—	—	—	—	—	—
	ZP	Minimize	29.87	36.76	1–1	3	—	—	—	—	—	—
Optimized Solutions	1	—	—	—	—	—	63.352	20.000	389.096	31.203	0.716	Selected
	2	—	—	—	—	—	62.931	20.000	389.116	31.198	0.716	—
	3	—	—	—	—	—	62.382	20.000	389.148	31.193	0.716	—
	4	—	—	—	—	—	60.000	20.000	389.351	31.174	0.714	—
	5	—	—	—	—	—	57.020	20.000	389.760	31.160	0.709	—
	6	—	—	—	—	—	71.015	11.379	393.247	32.656	0.557	—

Comparison of Predicted and Observed Values

The optimized nanoparticle formulation (chitosan 63.35 mg, PVA 20 mg) predicted an MPS of 389.1 nm and ZP of 31.2 mV with a desirability of 0.716, while the observed values were 237.8 ± 1.76 nm and 32.09 ± 3.38 mV, respectively. All key physicochemical evaluations, including particle size and ZP, were performed in triplicate (n = 3 independent experiments), and the results were presented as mean ± standard deviation, ensuring the reproducibility and reliability of the experimental outcomes. The surface charge closely matched the prediction, whereas particle size decreased significantly, likely due to slower solvent evaporation, less efficient stirring, and reduced agglomeration. Dynamic light scattering (intensity mode; Fig. 6 ) confirmed the nanoscale particle distribution, and ZP measurement ( Fig. 7 ) verified stable surface charge, demonstrating that the desirability function effectively guides the formulation of robust and reproducible PNS.

Fig. 6.

Dynamic light scattering (intensity mode) profile of the optimized palbociclib-loaded nanoparticles confirming nanoscale particle size distribution.

Fig. 7.

Zeta potential of the optimized nanoparticle formulation demonstrating stable surface charge and formulation robustness.

Drug Loading and EE

The formulated PLB-PNS had high drug incorporation rates. It was determined that the drug loading and EE results showed ( Table 5 ) good encapsulation of PLB into the polymeric matrix.

Table 5.

Result of Drug Loading and Entrapment Efficiency

Entrapment efficiency (%)				Drug loading (%)
Reading I	Reading II	Reading III	Mean ± SD	Reading I	Reading II	Reading III	Mean ± SD
81.28	79.37	82.98	81.21 ± 1.80	42.96	40.38	39.05	40.79 ± 1.98

In Vitro Drug Release Study

The in vitro release study showed a distinct pH-dependent characteristic of PNs. PLB-loaded PNS exhibit pH-dependent drug release, with a faster release at acidic pH 5.4 (simulating tumor environments) and a slower release at physiological pH 7.4. At pH 5.4, the drug releases rapidly due to polymer swelling and increased diffusion pathways, reaching nearly 97% after 26 h, while at pH 7.4, the release is more controlled, reaching about 90% in the same period. This behavior is driven by physicochemical responses of the polymer matrix, primarily diffusion and swelling, enabling targeted drug delivery to tumors with minimal release in normal tissues. All experiments were conducted in triplicate, confirming the reproducibility of these findings (Fig. 8 ).

Fig. 8.

In vitro drug release profile of palbociclib-loaded nanoparticles showing pH-dependent controlled release with accelerated drug release under acidic conditions compared to physiological pH.

In Vitro Drug Release Kinetics Model

The in vitro drug-release kinetics of PNs was analyzed using DD-Solver ( Table 6 and Fig. 9 ). Among the models tested, the K–P model best described the release (R² = 0.9977, R²_adj = 0.9974, MSE = 2.5282, MSC = 5.4899), indicating diffusion-controlled release from the polymer matrix. Zero-order and first-order models also correlated well (R² = 0.9848) but were less precise, while Higuchi (R² = 0.9391) and Hixson–Crowell (R² = 0.9944) models were less predictive. These results confirm that the nanoparticles provide continuous and controlled PLB release, supporting their potential as an effective therapeutic system.

Table 6.

Goodness-of-Fit Parameters for In Vitro Drug Release Kinetic Models of Palbociclib-Loaded Nanoparticles

Goodness of fit
Parameter	Zero-order	First-order	Higuchi	Korsmeyer–Peppas	Hixson–Crowell
N_observed	11	11	11	11	11
DF	10	10	10	9	10
R_obs-pre	0.9933	0.9933	0.9848	0.9988	0.9973
Rsqr	0.9848	0.9848	0.9391	0.9977	0.9944
Rsqr_adj	0.9848	0.9848	0.9391	0.9974	0.9944
MSE	14.6871	14.6871	58.9573	2.5282	5.4651
MSE_root	3.8324	3.8324	7.6784	1.5900	2.3378
Weighting	1	1	1	1	1
SS	146.8707	146.8707	589.5727	22.7537	54.6514
WSS	146.8707	146.8707	589.5727	22.7537	54.6514
AIC	56.8851	56.8851	72.1734	38.3720	46.0107
MSC	3.8069	3.8069	2.4171	5.4899	4.7955

Fig. 9.

In vitro drug-release kinetics of palbociclib nanoparticles by DD-Solver, showing diffusion-controlled release (Korsmeyer–Peppas model).

Morphological Characterization of Optimized Batch

The surface morphology of the optimized PNs was analyzed by SEM ( Fig. 10 ), showing smooth, spherical particles. This uniform shape supports effective drug release, stability, and uptake, confirming the success of the formulation method.

Fig. 10.

SEM image of optimized palbociclib-loaded nanoparticles showing smooth, spherical, and uniform morphology. SEM, scanning electron microscope.

DISCUSSION

A PNs formulation was optimized using RSM with CCD, analyzing the effects of chitosan and PVA on particle size and ZP, and resulting in statistically significant quadratic models for predicting optimal formulation parameters.

ANOVA indicated significant quadratic effects (A² and B²) on particle size, showing a nonlinear relationship between polymer concentrations and nanoparticle size, consistent with Khan and Sandhya’s findings,²⁵ who reported solid–lipid nanoparticles (SLNs) containing PLB-loaded particles ranging from 103 to 239 nm and PS values between 15 and 29 mV. However, in the current investigation, the optimized chitosan–PVA nanoparticles presented validated particle sizes at 237.8 ± 1.764 nm, which is a broader boundary in the size range reported before. Cationic chitosan could offer better electrostatic stability, as larger ZPs of 32.09 ± 3.38 mV showed an increase concerning the above.²⁵

In the study by Mehata, the chemical changes during conjugation were identified: Egen showed characteristic peaks, which shifted after carboxylation to form Egen-COOH, and further modifications in CS-g-Egen indicated amide bond formation, evidenced by new NH bending peaks. The present FTIR data showed small displacements—all around C = O (1600.59→1601.79 cm⁻¹), N–H (3699.56→3506.30 cm⁻¹), and C≡N (2947.63→2922.80 cm⁻¹)—indicating primarily physical interactions through hydrogen bonding. The lack of any new ester or amide peaks suggests that there was no chemical incompatibility, supporting the stability of the PLB–excipients system.²⁶

Khan and Sandhya²³ described the formulation of liquisolid tablets containing PLB. They ranged from 78.63% to 93.32% in drug encapsulation, wherein tricapric and Cremophor RH40 also have a significant effect on PLB encapsulation. Interestingly, PNS looked promising in terms of drug entrapment, with 81.21 ± 1.80% efficiency, and drug loading together with encapsulation was 40.79 ± 1.98%. These findings support the encapsulation of PLB into the polymeric matrix to a great extent, comporting with the performance range observed in a previous SLN-introduction study.²⁵

Khan and Sandhya reported a 12-h cumulative release of 78.37%–98.87% from PLB-loaded SLNs, indicating rapid diffusion from the lipid matrix. In the present study, the PLB-PNS formulation exhibited a more sustained, pH-responsive release profile. The nanoparticles released 70.48% at pH 5.4 and 50.77% at pH 7.4 at 12 h, demonstrating slower and controlled diffusion compared to SLNs. The accelerated release in acidic medium (pH 5.4) confirms tumor-microenvironment–responsive behavior, which is advantageous for anticancer delivery. In contrast, the free drug showed rapid and nearly complete release (97.20% at pH 5.4 and 91.52% at pH 7.4 within 12 h), confirming its inability to provide controlled release. Further analysis of the release kinetics determined that the drug release from the PNS was best described by the K–P model. The high correlation coefficient, along with the associated model selection criteria, suggests that the primary release mechanism is anomalous transport, a combination of drug diffusion and polymer chain relaxation. This supports the observed sustained release profile and the pH-responsive swelling of the chitosan-based matrix.²⁵

Transport at the super case-II category owing to high values (0.949–0.961) for the R² and an exponent greater than 1.0 (n) was achieved using PLB-loaded SLN, as concluded by Khan and Sandhya in 2022.²⁵ After assessing the data for the kinetic models, the best fit was established with the K-P’s model because R_obs-pre was 0.9988 and R² was 0.9977. The adjusted R² was the highest at 0.9974, while the least and second-least MSE and AIC values were 2.5282 and 38.37, comparatively. It emphasizes the occurrence of polymerically controlled diffusion-controlled drug release. In contrast, at a surface-area-dissolution-depletion model, the R²-value of the Higgs–Crowell model exhibited a further agreement of 0.9944. Yet an additional R²-value was predicted as 0.9391 in the Higuchi model, which demonstrated the most minimal interaction.

CONCLUSIONS

The present study of PLB-loaded chitosan PNS was successfully optimized using a CCD and desirability function. The concentrations of chitosan and PVA significantly influenced particle size and ZP, as demonstrated by quadratic modeling and ANOVA. The optimized formulation (63.35 mg of chitosan and 20 mg of PVA) exhibited an MPS of 237.8 ± 1.76 nm and a ZP of 32.09 ± 3.38 mV, indicating good colloidal stability and formulation reproducibility. FTIR analysis confirmed drug–polymer compatibility, while SEM images revealed smooth and spherical nanoparticle morphology. In vitro drug release followed diffusion-controlled kinetics and was best described by the K–P model (R² = 0.9977), reflecting sustained release from the polymeric matrix. Additionally, the formulation showed high EE (81.21 ± 1.80%) and drug loading (40.79 ± 1.98%), demonstrating efficient drug incorporation and controlled release performance under in vitro conditions.

Despite these favorable outcomes, the study is limited to in vitro physicochemical characterization and release evaluation. No biological interaction studies or long-term stability assessments under varied storage conditions were included, which may influence nanoparticle performance. Future investigations should focus on extended stability studies, in vitro cellular uptake and cytocompatibility assessments, and mechanistic evaluation of polymer–drug interactions under physiological conditions. Additionally, scaling-up feasibility and process reproducibility should be explored to further establish the formulation’s robustness.

AUTHORS’ CONTRIBUTIONS

N.H.: Conceptualization, formal analysis, methodology, investigation, resources, visualization, and writing—original draft. A.C.: Methodology, funding acquisition, project administration, software, supervision, and writing—review and editing.

Footnotes

ACKNOWLEDGMENT

Assam down town University supported its institutional resources to make this study possible. The authors gratefully acknowledge the university’s contribution to facilitating the successful completion of this work. The authors express their gratitude to the faculty and staff of the Faculty of Pharmaceutical Science who provided professional advice and technical contribution.

DISCLOSURE STATEMENT

The authors declare that there are no conflicts of interest.

FUNDING INFORMATION

This research was financially supported by Assam down town University through a Seed Money Grant (Sanction Order No. AdtU/DRA-II/2023-24/191, dated 09/10/2023).

Abbreviations Used

References

1. Hoque

, Choudhury

, Baishya

, et al. Comprehensive review of breast cancer risk factors, diagnosis, screening, and treatment methods. J Appl Pharm Res 2024;12(5):16–27.

2. Akhilesh , Singh

, Tiwari

. TRPA1 siRNA-loaded nanoformulation ameliorates chemotherapy-induced peripheral neuropathy. ACS Chem Neurosci 2025;16(11):2060–2075.

3. Dhamija

, Mehata

, Tamang

, et al. Redox-sensitive poly(lactic-co-glycolic acid) nanoparticles of palbociclib: Development, ultrasound/photoacoustic imaging, and smart breast cancer therapy. Mol Pharm 2024;21(6):2713–2726.

4. Bahreyni

, Mohamud

, Ashraf Nouhegar

, et al. Synergistic viro-chemoimmunotherapy in breast cancer enabled by bioengineered immunostimulatory exosomes and dual-targeted coxsackievirus B3. ACS Nano 2024;18(5):4241–4255.

5. Jiang

, Cao

, Yao

, et al. Chitosan and hyaluronic acid in breast cancer treatment: Anticancer efficacy and nanoparticle and hydrogel development. Int J Biol Macromol 2025;301:140144.

6. Maughan

, Lutterbie

, Ham

. Treatment of breast cancer. Am Fam Physician 2010;81(11):1339–1346.

7. Burande

, Viswanadh

, Jha

, et al. Correction to: EGFR targeted paclitaxel and piperine co-loaded liposomes for the treatment of triple negative breast cancer. AAPS PharmSciTech 2020;21(6):207.

8. Sahu

, Mehata

, Viswanadh

, et al. Dual-receptor-targeted nanomedicines: Emerging trends and advances in lung cancer therapeutics. Nanomedicine (Lond) 2022;17(19):1375–1395.

9. Xue

, Qu

, Li

. Stimuli‐responsive crosslinked nanomedicine for cancer treatment. Exploration 2022;2(6); doi: 10.1002/EXP.20210134 Epub ahead of print 21 December

10.

10. Dhillon

. Palbociclib: First global approval. Drugs 2015;75(5):543–551.

11.

11. Felip

, Llobera

, Perez-Mañá

, et al. New drugs, old toxicities: Pneumonitis related to Palbociclib – A case report. Breast Care (Basel) 2020;15(5):548–552.

12.

12. Farhat

, Tarabaih

, Kanj

, et al. Palbociclib safety and efficacy beyond Ribociclib-induced liver toxicity in metastatic hormone-receptors positive breast cancer patient. Anticancer Drugs 2020;31(1):85–89.

13.

13. Catlin

, Bowman

, Engel

, et al. Reproductive and developmental toxicity assessment of palbociclib, a CDK4/6 inhibitor, in Sprague-Dawley rats and New Zealand White rabbits. Reprod Toxicol 2019;88:76–84.

14.

14. Muthu

, Kutty

, Luo

, et al. Theranostic vitamin E TPGS micelles of transferrin conjugation for targeted co-delivery of docetaxel and ultra bright gold nanoclusters. Biomaterials 2015;39:234–248.

15.

15. Guo

, Ma

, Zhang

, et al. Nanotechnology-assisted immunogenic cell death for effective cancer immunotherapy. Vaccines (Basel) 2023;11(9):1440; doi: 10.3390/vaccines11091440 Epub ahead of print 31 August

16.

16. Agarwal

, Sau

, Iyer

, et al. Multiple strategies for the treatment of invasive breast carcinoma: A comprehensive prospective. Drug Discov Today 2022;27(2):585–611.

17.

17. Zhang

, Zhang

, Wang

, et al. Hypoxia in breast cancer—Scientific translation to therapeutic and diagnostic clinical applications. Front Oncol 2021;11:652266; doi: 10.3389/fonc.2021.652266 Epub ahead of print 11 March

18.

18. Eladl

. In-cell NMR reveals the first direct observation of endogenous interaction between HIV Tat protein and Tat RNA aptamer in human cells. Sci Rep 2025;15(1):29373.

19.

19. Eladl

. CRISPR interference-mediated silencing of pro-inflammatory genes in autoimmune models: Functional and biophysical evaluation with sustained effects in primary human PBMCs. Int Immunopharmacol 2025;163:115257.

20.

20. Krishnakumar

, Sulfikkarali

, RajendraPrasad

, et al. Enhanced anticancer activity of naringenin-loaded nanoparticles in human cervical (HeLa) cancer cells. Biomedicine & Preventive Nutrition 2011;1(4):223–231.

21.

21. Sartaj

, Qamar

, Qizilbash

, et al. Polymeric nanoparticles: Exploring the current drug development and therapeutic insight of breast cancer treatment and recommendations. Polymers (Basel) 2021;13(24):4400.

22.

22. Yang

C-R

, Zhao

X-L

, Hu

H-Y

, et al. Preparation, optimization and characteristic of Huperzine A loaded nanostructured lipid carriers. Chem Pharm Bull (Tokyo) 2010;58(5):656–661.

23.

23. Khan

, Saeed

, Khan

. Nanoparticles: Properties, applications and toxicities. Arab J Chem 2019;12(7):908–931.

24.

24. Yuan

, Mao

, Du

, et al. Surface charge switchable nanoparticles based on Zwitterionic polymer for enhanced drug delivery to tumor. Adv Mater 2012;24(40):5476–5480.

25.

25. Khan

and Sandhya

. Palbociclib loaded solid lipid nanoparticles by Box-Behnken design: Solubility enhancement. Int J Pharm Sci Res 2022;13:1241–1250. Career Point University, Kota - 325003, Rajasthan.

26.

26. Mehata

, Singh

, et al. Chitosan-g-estrone nanoparticles of Palbociclib vanished hypoxic breast tumor after targeted delivery: Development and ultrasound/photoacoustic imaging. ACS Appl Mater Interfaces 2023;15(29):34343–34359.