Development and Scale-Up of a Tablet-in-Tablet Formulation for Diclofenac Sodium and Misoprostol Using a Quality by Design Approach

Oral solid dosage forms represent the most widely used mode for drug administration, valued for their manufacturing efficiency, physicochemical stability, and patient convenience¹. Within this class, fixed-dose combinations (FDCs) have become an increasingly vital therapeutic strategy, offering the potential to simplify complex medication regimens, improve clinical outcomes through synergistic action, safety and enhance patient compliance.²

Diclofenac sodium, a potent non-steroidal anti-inflammatory drug (NSAID), is a principal agent in managing chronic inflammatory conditions. Its therapeutic effect is mediated through the non-selective inhibition of cyclooxygenase (COX-1 and COX-2) enzymes³. However, this non-selectivity is also the primary reason for side effects. While inhibition of COX-2 mediates the desired anti-inflammatory effect, the concurrent inhibition of the constitutively expressed COX-1 isoform critically impairs the synthesis of cytoprotective prostaglandins (e.g., PGE? and PGI?) in the gastric mucosa.? This loss of mucosal protection, coupled with direct topical irritation, frequently leads to significant gastrointestinal complications, including peptic ulcers and haemorrhage, which can limit the clinical utility of Diclofenac sodium, particularly in long-term therapy.?

A well-established clinical strategy to mitigate this dose-limiting gastrointestinal toxicity is the co-administration of a mucosal protective agent. Misoprostol, a synthetic analogue of prostaglandin E?, serves this function by directly addressing the mechanism of NSAID-induced gastric injury. It acts as a prostaglandin replacement, binding to epithelial prostaglandin receptors to inhibit basal acid secretion and crucially stimulate the production of protective mucus and bicarbonate?,?.  While the co-prescription of separate tablets is effective, it introduces significant risks of patient non-adherence, which can lead to suboptimal gastro-protection. Consequently, an FDC that ensures concurrent delivery and compliance is the preferred therapeutic option?.

Developing such a combination product, however, presents technical challenges. The two APIs are fundamentally incompatible. Misoprostol is hygroscopic in nature and extremely unstable; it is highly susceptible to chemical degradation, particularly through dehydration to Misoprostol A and epimerization, pathways that are accelerated in the presence of acidic agents like Diclofenac sodium?,¹? and acidic enteric coating material. This inherent instability makes its incorporation into a stable, enteric-coated dosage form having Diclofenac sodium a complex design. This creates a clear need for an alternative formulation platform that can ensure the stability of Misoprostol while achieving a varied drug release profile. To address this technological challenge, a unique design of a novel cup shaped tablet-in-tablet (TiT) dosage form, a platform well-suited for separating incompatible APIs and achieving complex, varied release profiles is developed¹¹,¹²

The primary objective of this research work was to scientifically design, optimize, proactively identify and eliminate potential risks of complex tablet design by using a Quality by Design (QbD) approach.¹³ This quality approach moves beyond traditional, empirical "trial-and-error" formulation development by building quality, robustness, and  understanding the product design strategy.¹? The formulation must meet two potentially conflicting sets of Critical Quality Attributes (CQAs): the varied drug release requirements and the critical protection of the enteric coated tablet layer. The inner enteric coated tablets must be sufficiently robust to withstand the compression forces of the placebo layer by tablet in tablet compression technique without fracturing the functional enteric coat.¹? Central to our QbD strategy was the use of Design of Experiments (DoE), a powerful statistical methodology.¹? Specifically, we selected a 2-level factorial design to efficiently map the complex and potentially interactive relationships between formulation variables (Critical Material Attributes, CMAs) and the resulting dissolution profile (a CQA). This multifactorial approach is essential for uncovering critical interactions between excipients—such as the interplay between a binder's cohesive force and a disintegrants swelling properties, which would be missed by a conventional one-factor-at-a-Time (OFAT) analysis. The ultimate goal was to develop an innovative tablet design with good product understanding to define a robust design space, ensuring consistent product quality, mechanical integrity of enteric coating layer and reliable performance upon scale-up.

MATERIALS AND METHODS

MATERIALS

Diclofenac sodium (USP), Misoprostol (USP), Lactose Monohydrate, Povidone K-30, Croscarmellose Sodium, Magnesium Stearate, Eudragit L30 D-55, Sodium Hydroxide, Triethyl Citrate, Talc, Silicified Microcrystalline Cellulose (Prosolv HD 90 and SMCC 50), and Sodium Stearyl Fumarate were gifted by Dr. Reddy's Laboratories Ltd. (Hyderabad, India). All solvents such as Methanol and Ethyl Acetate were industrial grade and reagents used for analysis were of HPLC or analytical grade.

METHODS

Core Tablet Optimization (DoE)

A 2³ factorial design (11 runs, 3 centre points) was employed to optimize the Diclofenac sodium core.¹? The independent variables (factors) were Croscarmellose Sodium (A), Povidone K-30 (B), and Magnesium Stearate (C), as detailed in Table 1. The dependent variables (responses) were the cumulative percentage of drug released at 5, 15, and 30 minutes.

Table 1: Independent Variables for the 23 Factorial Design

Core Tablet Preparation (Lab Scale)

Diclofenac sodium, Lactose, and intra-granular Croscarmellose Sodium (sifted, #14 mesh) were dry-mixed. A 15.5% w/w Povidone K-30 aqueous binder solution was sprayed onto the blend in a Fluid Bed Processor (FBP) (product bed temp 40 ± 5 °C). Granules were dried in the FBP (LOD NMT 4.0% w/w) and sized (#30 mesh), milling any over size granules. The extra-granular portion of Croscarmellose Sodium (#40 mesh) was admixed, followed by a 5-minute lubrication step with Magnesium Stearate (#60 mesh). The final blend was compressed into 135 mg tablets (7.0 mm round, standard concave tooling).

Core Tablet Coating

Sub-coating: A 8% w/w Hypromellose solution (IPA:water, 30:70) was applied in an automated coater to a 1.5% w/w gain.

Enteric Coating: A gastro-resistant Eudragit L30 D-55 suspension (Plasticizer: Triethyl Citrate, 10%; anti-tack: Talc-50% and sodium hydroxide -1% ratio with respect polymer content) was applied to the pre-warmed (50 ± 10 °C) sub-coated tablets to achieve a 5% w/w gain.

Final Tablet-in-Tablet (TiT) compression:

Silicified MCC, Povidone, and Croscarmellose Sodium (#40 mesh) were mixed (10 min), then lubricated with Sodium Stearyl Fumarate (#60 mesh, 5 min). A Cadmach press coat machine (13.0 mm cup-shaped tooling) was used²⁶.  The process involved partial die filling with the placebo blend, centering the enteric-coated core and final filling/compression.

Final Film Coating: TiT forms were coated (5% w/w gain) with a non-functional aqueous Hypromellose / Titanium dioxide solution to prevent enteric coat/ subsequent drug solution interaction.

Misoprostol Filling: A ~0.33% Misoprostol solution (API + Povidone K-30 in 75:25 Methanol/Ethyl Acetate) was dispensed (~58 µL) into the tablets cup-cavity using a Gilson 215 Liquid Handler and Filled tablets were tray-dried for 3 hours at 40 °C to evaporate solvents.

The entire manufacturing process was depicted below for ease of understanding (Figure1).

Physicochemical Characterization and Analytical Evaluation

Drug-Excipient Compatibility:

1:1 (w/w) binary mixtures were stored at 40°C/75% RH for 30 days¹?.Physical interaction was assessed by FTIR¹? and chemical stability (related substances) was monitored by a validated stability-indicating HPLC method (USP <621>)¹?.

Granule/Tablet Evaluation: Blends were tested for Bulk/Tapped Density (BD/TD) and Carr's Index. Tablets were evaluated per Weight Variation (USP <905> ), Disintegration Time (<701>), hardness, and friability per USP general chapters²?.

Assay (Diclofenac sodium):

UV-Vis spectrophotometry at 275 nm (in pH 6.8 buffer). In-Vitro Dissolution (DoE Cores): USP Type II (paddle), 50 RPM, 900 mL, pH 6.8 buffer (USP <711>)²¹.

In-Vitro Dissolution of Diclofenac sodium (Final TiT):

 A two-stage dissolution protocol (USP <711>) was used (USP II, 100 RPM).²¹ Stage 1: 2 hrs in 750 mL 0.1 N HCl. Stage 2: Adjusted to 1000 mL, pH 6.8. Analysis by UV-Vis (275 nm).

Misoprostol: Release was tested as per USP <711> (USP Type II, 50 RPM) in 500 mL, pH 3.0 buffer²¹.  Quantification of diclofenac sodium was carried out by HPLC (UV 200 nm).

Assessment of TiT Compression Force

A comparative dissolution study was conducted to evaluate the impact of mechanical stress from press-coating. The two-stage dissolution profile (USP <711>) of stand-alone enteric-coated cores (pre-compression) was compared against the final TiT dosage form (post-compression) to ensure the functional film's integrity was not compromised²¹.

Data Analysis and Similarity Factor (f2)

Dissolution profiles were compared using the model-independent similarity factor, f2 (value 50-100 indicates similarity)²².

Scale-Up Batch Manufacturing

The process was scaled up. Top spray granulation utilized a Glatt FBP with defined parameters: inlet air (45-65°C), product temp (25-35°C), spray rate (100-300 g/min), and atomization (1.0-2.5 bar); final LOD was 2.68%. Compression was performed on a Cadmach rotary press (14-16 kP) and challenged at 20, 30, and 40 RPM. Coating (seal 1.5%, enteric 5%) was executed in a 30kg Ganscoater. A curing study²⁷ (1, 2, and 3 hrs) was performed on the enteric-coated tablets. TiT assembly (Cadmach press, 12-18 RPM) and Misoprostol filling (Gilson 215) proceeded as in the lab-scale process.

RESULTS

Drug-Excipient Compatibility Assessment

Potential interactions between Diclofenac sodium and the key excipients were evaluated. FTIR spectroscopy showed no significant shifts in the characteristic absorption bands of Diclofenac Sodium in physical mixtures stored for one month at 40 ± 2 °C / 75 ± 5% RH, indicating physical compatibility with the selected excipients (Figures 2 to 5). Chemical compatibility was confirmed by HPLC analysis (Table 3 &4). No significant increase in the specified degradant (Impurity A), maximum unknown impurity or total impurities was observed for Diclofenac sodium in the presence of any tested excipients when compared to the pure API control under accelerated conditions

Table 3: Excipient compatibility study – Analytical data, 40 °C, 1 month.

Table 4: Excipient compatibility study – Analytical data, 40 °C / 75% RH, 1 month.

All 11 experimental DoE blends demonstrated good flow and compressibility characteristics. The resulting core tablets (Table 6) met all in-house specifications for weight variation, hardness, thickness, friability, and disintegration time. All batches exhibited negligible friability (NMT1%w/w) and high hardness suitable for subsequent processing (12.5–16.8kP). Tablet thickness was also consistent across batches (3.00-3.18 mm).

 Table 5: Pre-compression Properties of Granules for DoE Batches

The 23 factorial design explored the impact of the three factors on the CQA (drug release). The resulting dissolution profiles for the 11 DoE batches are shown in Table 7.

Table 7: Dissolution Profile Results for DoE Batches (% Drug Released)

 Statistical Analysis of In-Vitro Dissolution Data

Analysis of variance (ANOVA) was performed on the dissolution data (Table 8). The statistical models were found to be significant at all three time points (p < 0.05), and the Lack of Fit was non-significant (p > 0.05). The analysis identified Factor A (Croscarmellose sodium) and Factor B (Povidone K-30) as statistically significant CMAs. Factor A was significant at all-time points, with its impact (F-value) increasing over time (from 11.17 to 30.08). Factor B was significant at 5 and 15 minutes but not at 30 minutes. Factor C (Magnesium Stearate) and all interaction terms were found to be non-significant (p > 0.05).

Enteric coating percentage optimization on the diclofenac tablets

Different percentage of enteric coating was applied on the optimized core tablets, % coating ranging from 4 to 6% evaluated and details presented below.

The mechanical integrity of the enteric coat after the press-coating stage was evaluated. The dissolution profile of the stand-alone enteric-coated cores was compared directly to the profile of the final tablet-in-tablet (Table 10). A high degree of similarity was observed, with an f2 value of 74.13.

Table 10: Enteric Coated Vs Tablet-in-Tablet (TiT)

Figure 7: Dissolution profile EC tabs Vs TiT tabs

In-Vitro Drug Release from the Final Optimized Dosage Form

Based on the statistical model, an optimized core formulation (6.8mg Croscarmellose Sodium, 7.2mg Povidone K-30, 1.35 mg Magnesium Stearate) was manufactured as lab-scale batch D004B and processed into the final TiT dosage form. In the 2-hour acid stage, the release of Diclofenac sodium was 1%. The subsequent release profiles for both APIs were compared to the marketed product-Arthrotec. The Diclofenac Sodium release profile in pH 6.8 buffer (Table 11) yielded an f2 similarity factor of 72.32.

Table 11: Test Vs Marketed product (DS)

Figure 8: Dissolution profile of Test Vs Marketed

Table 12: Test Vs Marketed product (Miso)

Figure 9: Dissolution profile of Test Vs Marketed (Miso)

Scale-Up Batch Performance

The optimized process was successfully applied to the scale up batch. The granulated blend achieved the target LOD of 2.68%. The final lubricated blend exhibited a bulk density of 0.442 g/mL and a tapped density of 0.589 g/mL.

Table 13: Particle Size Distribution (PSD) of Scale-Up batch (Lubricated Blend)

Core tablets compressed at various speeds (20, 30, and 40 RPM) consistently met weight and hardness targets (14-16 kP). Content uniformity was excellent across all speeds, with individual values ranging from 95.6% to 100.6% and RSDs remaining below 2.0% (Table 14).

Table 14: Content Uniformity of Scale-Up Core Tablets at Different Compression Speeds

The enteric coating curing study (Table 15) demonstrated that 3 hours of curing was optimal to achieve the desired release profile.

Table 15: Effect of Enteric Coat Curing Time on Scale-Up Batch Dissolution

The final scale-up batch demonstrated excellent gastro-resistance in 0.1 N HCl (Individual values: 1.5, 1.5, 0, 1.5, 0.5, 0 %; Average: 0.83%(~1%)). The final dissolution profiles (Tables 16 and 17) confirmed statistical similarity to the Marketed product.

Table 16: Scale-up Vs Marketed product (DS) 

Final product analysis of the scale-up batch met all specifications. Assay values were 98.5% for Diclofenac sodium and 104.7% for Misoprostol. Water content (by KF) was 5.92%. Misoprostol Uniformity of Dosage Units testing yielded an average of 102.3%, SD 4.074, and RSD 4.0%. Residual solvent levels after Misoprostol filling and dried for 3 hrs. Related substances for both APIs were well within acceptable limits (Table 18).

Table 17: Scale-up Vs Marketed product

Figure11: Dissolution profile of Test Vs Marketed Product

Table 18: Related Substances Profile of Final Scale-Up Batch

DISCUSSION

The objective was to ensure the development of a mechanically robust core tablet formulation suitable for TiT and a reproducible, scalable manufacturing process capable of consistently achieving the predefined critical quality attributes (CQAs).

Drug-Excipient Compatibility Assessment:

A fundamental aspect of preformulation is the assessment of potential drug-excipient interactions to ensure the physicochemical stability of the active pharmaceutical ingredient (API) within the intended dosage form matrix. This preliminary compatibility screening helps prevent formulation instability or performance issues during development.

FTIR Spectroscopic Analysis:

Fourier-transform infrared spectroscopy was utilized to probe for potential molecular-level interactions between Diclofenac sodium and the selected excipients. The FTIR spectrum of the initial physical blend was compared against the spectrum of pure Diclofenac sodium. Key characteristic absorption bands corresponding to the API's functional groups (e.g., carboxyl C=O stretch, secondary amine N-H bend, aromatic C-H stretches) were observed to be retained in the blend, without significant shifts in wavenumber or changes in relative peak intensity. The spectrum of the blend appeared consistent with an additive overlay of the individual component spectra (API and placebo blend), lacking evidence of new peak formation or disappearance of characteristic API bands that would signify chemical interaction. This indicated the absence of strong covalent or significant hydrogen bonding interactions upon initial mixing.

Furthermore, samples subjected to accelerated storage conditions (40 ± 2 °C / 75 ± 5% RH for one month) were analysed. Comparison of the aged blend spectrum with the initial blend and an aged placebo blend showed that the principal peaks attributed to Diclofenac Sodium remained distinct and largely unaffected. This spectral consistency under elevated temperature and humidity conditions suggests the absence of detectable solid-state chemical reactions or significant physical transformations within the study period, supporting physical and chemical compatibility.

HPLC Analysis of Related Substances:

A validated, stability-indicating HPLC method provided a quantitative assessment of chemical stability by monitoring the formation of degradation products under defined storage conditions.

Storage at elevated temperature (40 °C): Samples stored at 40 °C showed (Table 3) minimal degradation attributable to thermal effects alone. Compared to the Diclofenac sodium API control (0.048% total impurities after 1 month), mixtures with most core excipients, such as PVP K30 (0.050%), HPMC (0.057%), and Lactose (0.063%), exhibited comparable total impurity levels. This suggests good intrinsic thermal stability of the API in the presence of these excipients under dry heat conditions. A minor elevation was noted with Triethyl Citrate (0.100%), perhaps suggesting a slight interaction, though the overall impurity level remained low.

Accelerated Stability Conditions (40 °C / 75% RH): The assessment under accelerated conditions (Table 4), simulating more challenging storage environments, is particularly relevant. The Diclofenac sodium API control demonstrated good stability (0.044% total impurities). Importantly, the presence of the selected formulation excipients did not substantially accelerate degradation. While minor increases in total impurities were seen with Microcrystalline Cellulose (0.119%) and Magnesium Stearate (0.097%), these levels do not signify a critical incompatibility. Mixtures containing Lactose (0.080%), PVP K30 (0.071%), Croscarmellose Sodium (0.079%), and the enteric polymer Eudragit L30D55 (0.081%) yielded impurity profiles closely matching or only slightly higher than the API control.

Collectively, the FTIR spectral analysis and the quantitative HPLC impurity profiling provide converging evidence supporting the physicochemical compatibility of Diclofenac sodium with the chosen excipients. This preliminary compatibility screening confirmed the suitability of the excipient selection with respect to the API's intrinsic chemical stability profile, allowing the development efforts to concentrate on optimizing the formulation's critical quality attributes and addressing the mechanical challenges associated with the tablet-in-tablet manufacturing process.

Core Tablet Properties: Establishing a Mechanically Robust Design Space

The physical characterization of the core tablets manufactured across the 11 experimental runs of the DoE (Table 6) was essential not only to confirm basic manufacturability but also to evaluate the formulation's suitability for the subsequent, mechanically demanding stages of the TiT manufacturing process. The data demonstrated that acceptable tablets could be produced across the entire design space explored. All formulations yielded tablets meeting standard pharmaceutical quality attributes, including acceptable weight variation and disintegration times.

Of particular significance were the friability and hardness results. The observation of negligible friability ("Nil") across all batches indicated that the granules compressed well, forming tablets with sufficient surface strength to resist attrition during handling and, importantly, during the coating processes. The uniformly high hardness values achieved (all batches > 12 kP, with many exceeding 14 kP) were a critical finding. For a TiT dosage form, the inner core's mechanical strength is not merely a standard quality parameter but a critical quality attribute (CQA). The core must possess sufficient tensile strength and resistance to deformation to withstand the high compressive forces applied during the placebo layer compression without fracturing itself or, more critically, the applied functional enteric film. A physically weak or brittle core would be prone to failure during this stage, leading to loss of the intended delayed-release profile. The consistent achievement of high hardness across the design space confirmed that the formulation strategy was capable of producing cores with the requisite mechanical integrity. Furthermore, the observed disintegration times (approximately 4–6 minutes) demonstrated that the formulation maintained adequate disintegration characteristics even at these high hardness levels, indicating the effectiveness of the incorporated disintegrant concentration (Croscarmellose Sodium). This established that the mechanical robustness required for TiT manufacturing did not compromise the core's fundamental disintegration behaviour.

Statistical Analysis and Formulation Performance Insights from DoE

The statistical evaluation of the dissolution data derived from the DoE matrix (Tables 7 & 8) provides the quantitative basis for understanding the formulation's drug release behaviour and developing a science-driven control strategy. The high statistical significance of the generated models across the dissolution profile (e.g., p < 0.005 at 30 min) indicates that the selected formulation variables showed a predictable influence on drug release profile. Furthermore, the non-significant Lack of Fit (p > 0.05) affirms that the model adequately describes the data within the experimental space, reinforcing confidence in its predictive capability. This statistical framework allows for a detailed interpretation of how the identified Critical Material Attributes (CMAs) influence the dissolution profiles from the core tablet.

Factor A (Croscarmellose Sodium): The analysis identified the concentration of this superdisintegrant as the primary formulation factor influencing the rate and extent of Diclofenac sodium release within the studied range of 5.44 mg to 8.16 mg. The substantial increase in its statistical impact over the dissolution time course (F-value increasing from 11.17 at 5 minutes to 30.08 at 30 minutes) suggests its critical function throughout the release process. Initially, it promotes rapid tablet disintegration via swelling and capillary action. Its sustained, dominant influence at later times indicates that it is essential for the effective deaggregation of the granules, thereby increasing the drug's effective surface area for dissolution and ensuring complete release.

Factor B (Povidone K-30): The polymeric binder, evaluated between 5.76 mg and 8.64 mg, demonstrated a significant effect predominantly during the initial phase of dissolution. Its statistical significance at 5 minutes (p=0.016) and 15 minutes (p=0.008) highlights its role in modulating the early drug release rate. Povidone K-30 contributes inter-granular cohesive strength, controlling the initial tablet breakup and preventing excessively rapid release. As dissolution progresses, the disruptive forces of the disintegrant appear to overcome this cohesion, reflected in the binder's lack of statistical significance at the 30-minute point. The balance between the binder's cohesive function and the disintegrant's disruptive action within their respective concentration ranges is therefore key to achieving the desired release profile.

Factor C (Magnesium Stearate): Within the investigated range of 1.08 mg to 1.62 mg, the concentration of the lubricant did not show a statistically significant impact on dissolution (p > 0.05 at all-time points). From a formulation robustness standpoint, this is a positive finding. It indicates that the drug release profile is tolerant to minor variations in lubricant level within this range, simplifying process control and reducing potential risks associated with variability in lubrication efficiency.

Interaction Effects: The statistical analysis also examined potential interactions between the factors. The two-factor interactions evaluated, AB (Croscarmellose Na * Povidone K-30) and AC (Croscarmellose Na * Magnesium Stearate), were found to be statistically non-significant (AB: p=0.257 at 5 min, p=0.197 at 15 min, p=0.309 at 30 min; AC: p=0.309 at 30 min; other AC interactions not significant). This lack of significant interactions suggests that, within the studied ranges, the effect of the disintegrant, binder, and lubricant on dissolution are largely independent. This simplifies the understanding of the formulation, as the impact of changing one component can be predicted without needing to account for complex synergistic or antagonistic effects with the other components, leading to a more robust and readily controllable formulation system.

This comprehensive understanding of the individual factor effects and their lack of interaction, derived directly from the statistical analysis of the DoE across the defined concentration ranges, informed the selection of the optimized core formulation. It also provides the scientific rationale for the Control Strategy (Table 19), which integrates material attribute controls (based on the DoE findings) with critical process parameters identified during development, ensuring that the CQAs of the drug product are consistently met.

Enteric Coating Performance and Process Integrity Validation

The successful functioning of this targeted drug delivery system is fundamentally dependent upon the robust performance and integrity of the enteric coating applied to the inner Diclofenac sodium core tablet. The results obtained validate both the effectiveness of the coating formulation and application process, as well as the mechanical resilience of the coated core during the demanding tablet-in-tablet compression stage.

Gastro-Resistance Performance: The primary critical quality attribute (CQA) related to the enteric coat is its ability to prevent drug release in the acidic environment of the stomach. The dissolution data confirm the achievement of this objective. The negligible release of Diclofenac sodium observed during the 2-hour acid stage for the optimized lab batch D004B(~1%) and an even lower average of 0.83% for the final scale-up batch demonstrates the formation of a competent, continuous gastro-resistant film using the Eudragit L30 D-55-based formulation at a 5% weight gain. This ensures that Diclofenac sodium protected from premature release, which could lead to gastric irritation and reduced intestinal absorption.

Manufacturing Process Robustness – Film Integrity Post-Compression: A significant potential failure mode for TiT dosage forms incorporating functionally coated core is the risk of mechanical damage to the coating during the high-force press-coating compression required to form the placebo layer tablet. Cracks or fractures in the enteric film would compromise its integrity, leading to premature drug release in acidic media. The comparative dissolution study, directly assessing the release profiles of the enteric-coated core before versus after their incorporation into the final TiT structure (Table -10 and fig-7) was specifically designed to evaluate this critical process risk. The resulting high degree of similarity between the two profiles, quantified by an f2 value of 74.13, provides compelling evidence that the enteric film's functional integrity was maintained. This finding validates that the Tablet in tablet process, did not induce significant damage to the gastro-resistant layer of the tablet. Successfully mitigating this critical manufacturing risk is paramount for the viability and robustness of the TiT platform for delivering functionally coated core tablet

In -Vitro Release Performance and Manufacturing Process Confirmation

The in vitro dissolution profile serves as a Critical Quality Attribute (CQA), providing a quantitative measure of drug release that reflects the combined influence of formulation composition and manufacturing process execution. Evaluating this CQA for both the optimized lab-scale batch and the scale-up batch allows for verification of the target formulation attributes and confirmation of process transferability and reproducibility.

Confirmation of Optimized Formulation (Lab Scale - D004B):

The performance of batch D004B demonstrated the appropriateness of the formulation components and ratios identified through the DoE, which established the link between Critical Material Attributes (CMAs) and the dissolution (CQA). The rapid Misoprostol release at initial 10mins (~77%) achieved the desired immediate-release profile and showed dissolution similarity to the marketed product (f2 = 71.61). Simultaneously, the Diclofenac core conferred the necessary gastro-resistance (1% release in acid), followed by dissolution in pH 6.8 buffer that was statistically similar to the marketed (f2 = 72.32). These data confirmed that the statistically determined optimal levels of the CMAs (Croscarmellose Sodium and Povidone K-30) produced the intended sequential drug release pattern in vitro at the laboratory scale along with optimised enteric coating 5%w/w build-up on core tablets.

Confirmation of Process Consistency and Scalability:

The performance of the scale-up batch was essential not only for confirming batch-to-batch consistency but also for verifying the operational control over identified Critical Process Parameters (CPPs) during manufacturing.

Impact of Enteric Coat Curing: The comparative dissolution study examining the effect of curing time (1, 2, and 3 hours) provided valuable data concerning the impact of this process step on the dissolution CQA. The determination that 3 hours of curing was required to achieve the desired dissolution profile, matching the marketed product, identifies this duration as a CPP. Precise control over the curing duration is therefore essential to ensure consistent gastro-resistance and attain the target delayed-release, directly linking a key process parameter to a critical quality attribute, consistent with QbD principles.

Integrity of Enteric Film Post-TiT Compression: A significant process-associated risk specific to TiT compression involves potential mechanical compromise of the functional film on the inner tablet. The comparative dissolution analysis between the stand-alone enteric-coated cores and the final TiT units (Section 3.6) yielding a high similarity factor (f2 = 74.13), provides robust quantitative data indicating successful mitigation of this process risk

Consistency of Release Profiles upon Scale-Up: The final dissolution results from scale-up batch verified the successful transfer of the optimized formulation composition and manufacturing procedure to the pilot scale. The Misoprostol immediate release profile remained consistent and similar to the marketed (f2 = 74.30). The Diclofenac sodium component, from the optimized 3-hour curing confirmed film integrity, displayed good gastro-resistance (1% release in acid) and achieved dissolution similarity (meeting the CQA target) to the marketed in pH 6.8 buffer (f2 = 60.85). The ability to consistently meet the target CQAs, particularly dissolution similarity, when transitioning from laboratory to pilot scale provides strong support for the effectiveness of the QbD approach. It indicates that the understanding of CMAs derived from the DoE, combined with the identification and control of CPPs such as core hardness and curing time (elements of the Control Strategy), resulted in a well-defined and transferable manufacturing process capable of consistently delivering the target in vitro drug release performance CQA.

Confirmation of Process Consistency and Scalability

A fundamental objective within the Quality by Design (QbD) framework is the development of pharmaceutical manufacturing processes that are not only well-characterized and controlled but also demonstrate consistency across scales and operational ranges. The successful transfer of the process from laboratory to pilot scale serves as a critical verification of the developed Control Strategy (Table 19) and provides confidence in the process's suitability for larger-scale production. The performance of the scale-up batch substantiates the consistency of the developed formulation and manufacturing process.

The scale-up involved utilizing equipment representative of larger production capabilities, including a Glatt Fluid Bed Processor, a 20-station Cadmach rotary tablet press, and a 30kg Ganscoater. The successful implementation of the process parameters on this equipment, yielding a batch that met all predefined quality attributes, confirms the suitability of the initial process design established during laboratory development. The Critical Quality Attributes (CQAs) of the scale-up batch, particularly gastro-resistance (~1% release) and the in vitro dissolution profiles for both Misoprostol (f2 = 74.30) and Diclofenac Sodium (f2 = 60.85), were consistent with those observed for the optimized lab-scale batch (D004B) and met the acceptance criteria. This consistency provides evidence that the process effectively controls the factors influencing these CQAs.

Furthermore, specific process parameters were deliberately varied during the scale-up run to assess process latitude and consistency. The evaluation of core tablet content uniformity across different compression speeds (20, 30, and 40 RPM) on the rotary press yielded results demonstrating excellent uniformity, with Relative Standard Deviations (RSDs) well below 2.0%. This finding is particularly relevant as it indicates that the granule blend possesses acceptable flow and compaction properties, and crucially, that the formulation performance is insensitive to variations in this studied Process Parameter within the evaluated range. Demonstrating such consistency against variations in operational parameters like compression speed is a key aspect of confirming process understanding and control, suggesting a well-defined operating space for this unit operation.

In summary, the successful process transfer to pilot scale, the consistent attainment of CQAs, and the demonstrated consistency against variation in a key process parameter (compression speed) collectively indicate that the Design Space, informed by the understanding of CMAs and process parameters, is appropriate. The established Control Strategy proved adequate for maintaining product quality during scale-up. This provides a strong scientific basis supporting the scalability and reproducibility of the manufacturing process for this complex tablet-in-tablet dosage form.

 Table 19: Control strategy and design space