Comparison of Downstream Processing of Nanocrystalline Solid Dispersion and Nanosuspension of Diclofenac Acid to Develop Solid Oral Dosage Form

The conventional “top-down”, “bottom-up” and “combination” approaches of generating drug nanocrystals produce a “nanosuspension” (NS). It requires significant downstream processing for drying the liquid by suitable means followed by its granulation to develop an oral solid dosage form (OSD). In this paper, we used a novel, spray drying-based NanoCrySP technology for the generation of drug nanocrystals in the form of nanocrystalline solid dispersion (NCSD). We hypothesized that the NCSD would require minimal downstream processing since the nanocrystals are obtained in powder form during spray drying. We further compared downstream processing of NS and NCSD of diclofenac acid (DCF) prepared by wet media milling and NanoCrySP technology, respectively. The NS and NCSD were characterized for crystallinity, crystal size, assay and dissolution. The NCSD was physically mixed with 0.3% Aerosil® 200, 1.76% croscarmellose sodium (CCS) and 0.4% sodium stearyl fumarate (SSF) and filled into size 0 hard gelatin capsules. The NS was first wet granulated using Pearlitol® SD 200 (G1 granules) and Celphere® 203 (G2 granules) in a fluidized bed processor, and the resulting granules were mixed using the same extra granular excipients as NCSD and filled into capsules. A discriminatory dissolution method was developed to monitor changes in dissolution behavior due to crystal growth during processing. Cost analysis and comparison of process efficiency was performed using an innovation radar tool. The NS and NCSD were successfully fabricated with a crystal size of 363 ± 21.87 and 361.61 ± 11.78, respectively. In comparison to NCSD-based capsules (65.13%), the G1 and G2 granules showed crystal growth and decrease in dissolution to 52.68% and 48.37%, respectively, in 120 min. The overall cost for downstream processing of NCSD was up to 80% lower than that of NS. An innovation radar tool also concluded that the one-step NanoCrySP technology was more efficient and required less downstream processing than the two-step wet media milling approach for conversion of nanocrystals to OSD.


Acetone
Alone --up to 300 mg is soluble in 7 mL (7 mL acetone: 3 mL water) Ethanol alone --up to 100 mg is soluble in 7 mL (7 mL ethanol: 3 mL water)  Table S1. A difference of less than 7 MPa 1/2 indicates miscibility between APIs and MAN while compounds with δt difference of more than 7 MPa 1/2 are likely to be immiscible [1]. The difference in δt values of DCF and MAN was found to be 14.4 MPa 1/2 which indicated immiscibility between them.

Investigation for plasticization or heterogeneous nucleation using mDSC
The mechanism of generation of DCF nanocrystals in the presence of crystallization inducing excipient i.e., plasticization or heterogeneous nucleation or both, was investigated using modulated differential scanning calorimetry (mDSC). mDSC analysis of physical mixtures of DCF and MAN in varying ratios were carried out in duplicates. The plasticization effect can be confirmed if there is a decrease in Tg of DCF with an increasing concentration of MAN while a decrease in the heat capacity (ΔCp) of amorphous form of DCF with an increment of MAN in the physical mixture supports heterogeneous nucleation mechanism [1].
Heat-cool-heat (HCH) protocol was used wherein both DCF and MAN were first melted by heating sample up to 190 °C at a heating rate of 20 °C/min followed by rapid cooling in the second cycle. In the third cycle, ± 0.80 °C modulation amplitude for every 60 s applied and then the sample was heated up to 190 °C at a heating rate of 2 °C/min to observe glass transition temperature (Tg) of generated amorphous form of DCF. Representative heating curves of the third cycle were reported in Figure S1. Table S3 showed that with an increase in the proportion of MAN, ΔCp of amorphous form of DCF has been decreased significantly and change in Tg was subtle as compared to the 100% amorphous DCF. This proved that heterogeneous nucleation was responsible for the generation of DCF nanocrystals embedded in the ternary NCSDs. Also, the percent amorphous DCF form was decreased as MAN concentration increased.

Solubility study of DCF
Apparent solubility of DCF was determined at different pH conditions--1.2 (HCl buffer), 4.5 (acetate buffer), 5.5 (citrate buffer), and 6.8 (phosphate buffer). The excess amount of DCF was dispersed into each flask containing 10 mL of media in triplicates. The stirring of dispersion was maintained for 72 h using a mechanical shaking bath at 60 rpm and 37 ± 0.5 °C . One millilitre sample was collected from each flask at specified time intervals of 24, 48, and 72 h and filtered through a nylon membrane syringe filter of pore size 0.1 µ m. Using acetonitrile in 1:1 (v/v) ratio, 500 µ L of the filtrate was diluted immediately and samples were analysed using the developed HPLC method. Solubility studies conferred that citrate buffer (pH: 5.5) with 28 µ g/mL solubility of DCF, would provide non-sink conditions for dissolution as compared to phosphate buffer (pH: 6.8) having higher solubility of DCF i.e., 320 µ g/mL.

Screening of discriminatory dissolution medium
This study aimed to develop a discriminatory dissolution medium that showed differentiable dissolution with variable DCF particle size which was essential to detect any change in DCF nanocrystals size during downstream processing. Nanosuspensions with variable D90 values of DCF nanocrystals i.e., NS2: 234 nm, NS3:750 nm and NS4: 1289 nm were prepared according to the procedure mentioned in the Section 2.2.4. A comparison of dissolution profiles of these three nanosuspensions was helpful to develop better discriminatory medium. Table S6 has listed out various dissolution conditions screened for this purpose.
Also, F2 (similarity) factor values calculated by comparing NS2 and NS3 dissolution profiles in the respective medium were mentioned in Table S9. Comparative dissolution profiles of nanosuspensions in different media are given in Figure S2. Initially, 900 mL water with 0.1% of SLS was used as the medium along with 75 rpm, wherein dissolution profiles of nanosuspensions were overlapped with each other and no discrimination was observed. Solubility studies showed that DCF was more soluble in citrate buffer (pH: 5.5) and phosphate buffer (pH: 6.8) as compared to other media. Thus, these two media were further evaluated to develop size based discriminatory dissolution medium. First, 900 mL phosphate buffer (pH: 6.8) with 75 rpm was tested which showed higher %DCF dissolved with time but was found to be incapable of discriminating dissolution profiles of nanosuspensions. After this, 900 mL of citrate buffer (pH: 5.5) with 75 rpm was evaluated and it had shown slightly variable dissolution profiles of nanosuspension but these profiles were not size based discriminatory. The addition of 0.1% Sodium Lauryl Sulphate (SLS) in this medium resulted in variable dissolution with time of nanosuspensions but did not produce discriminatory dissolution profiles. Thus, SLS did not help to improve discrimination of dissolution profiles based on DCF particle size. The discriminatory dissolution profiles of nanosuspensions were observed when 1000 mL of citrate buffer (pH: 5.5) with sink index 0.8 and 50 rpm was chosen as the discriminatory dissolution medium. With F2 value of 44.77, this medium helped to show DCF particle size based discrimination in the above-mentioned nanosuspensions.

Influence of granulating substrate on the dissolution profile
Dissolution with time of G1 capsules (52.68% after 120 min) in discriminatory dissolution medium was higher than that of G2 capsules (48.37% after 120 min). This difference in % DCF dissolved with time signified the impact of excipient properties on drug release. Celphere® 203 which is a grade of MCC doesn't dissolve in water but rather swells to absorb water unlike mannitol based Pearlitol® SD 200 which is water soluble. For drug dissolution, it is necessary to have osmotic pressure difference between the outer and inner part of the granules and this pressure difference may be responsible for the release of the active ingredient which can be explained by capillary pressure theory [2]. In the case of the G1 capsules, this pressure difference was achieved and the active ingredient was released from granules. In contrast, % drug dissolved with time from G2 capsules was slow and less as compared to G1 capsules as the required pressure difference did not build up. As sugar based excipients have higher osmotic activity than MCC, they drove more water into the system which further increased the dissolution from G1 capsules [3,4]. Another probable reason for lesser % drug dissolution with time of G2 capsules is attributed to crystalline gel model followed by MCC. During wet granulation when MCC comes in contact with water, water destroys crystalline structure of it and MCC converts into a gel. The further drying process results in autohesion to form a stable solid matrix which does not disintegrate easily after coming in contact with water [5].

Effect of solid content of NCSD on cost of NanoCrySP technology
Cost analysis for NCSD with a solid content of 2% w/v and 5% w/v is mentioned in Table S13 to understand the impact of solid content of NCSD on phase 1 and phase 2 cost. When solid content of NCSD increased from 0.5 to 2 or 5% w/v then Phase 1 cost of NanoCrySP technology reduced by 43% or 52%, respectively. With an increment in solid content of NCSD, significant decrease in manpower and electricity cost further reduced the overall phase 1 cost for NanoCrySP technology. However, Phase 2 cost for NCSD with 5% w/v solid content did not decrease significantly as compared to 0.5% solid content. This showed that increment in solid content of NCSD did not have a profound impact on downstream processing cost for NanoCrySP technology.  Yes due to organic solvent usage Less as no involvement of organic solvents  100,000-300,000 Less than 100,000