Why Low-Dose Wet Granulation Requires In-Process Redox Control Beyond ICH Q3D Limits

Wet granulation of the thiol-containing API in low-dose tablets is selected when direct compression cannot achieve acceptable content uniformity at API loadings below 10.0 wt%. The granulating fluid is deaerated purified water prepared with nitrogen purging to dissolved oxygen below 0.5 mg/L; dissolved oxygen is the primary process variable driving disulfide-related impurity formation. A representative formulation contains 5.0 wt% API, 52.0 wt% lactose monohydrate, 24.0 wt% pregelatinized starch, 12.38 wt% mannitol, 4.0 wt% povidone K30, 2.0 wt% croscarmellose sodium, 0.1 wt% sodium metabisulfite, 0.02 wt% disodium edetate, and 0.5 wt% magnesium stearate. The high-shear granulation cycle uses an impeller speed of 250 rpm and chopper speed of 1500 rpm; granulating fluid is added to a final water content of 12–16% w/w over 2 min, followed by 3 min wet massing. The wet mass is discharged through a 4 mm screen, dried in a fluid-bed dryer at inlet air temperature 48 °C and product temperature 32–36 °C to loss on drying 1.5–2.5%, then milled through a 0.8 mm screen to granule D50 180–250 µm. Batch-to-batch variance is commonly traced to dissolved oxygen in process water or trace iron from unpassivated 316L stainless-steel vessel surfaces; the disodium edetate chelates leached metal ions, while the metabisulfite provides a sacrificial reducing environment. The disulfide impurity is monitored by HPLC with UV detection at 254 nm; its specification is linked to the ICH Q3A identification threshold of 0.2% for a 5 mg daily dose. Elemental impurity release relies on USP <232> and USP <233> rather than ICH Q3D alone because redox-active elements can alter degradation kinetics before they exceed permitted daily exposure. Finished tablets are tested for content uniformity per Ph.Eur. 2.9.40, disintegration per Ph.Eur. 2.9.1, and dissolution per USP <711>. Equipment cleaning must avoid strong oxidizing disinfectants such as sodium hypochlorite or iodophor because residual oxidant reacts directly with the thiol sulfur and forms disulfide dimers.

When Dry Granulation Replaces Aqueous Binder Addition in Thiol-Sensitive Tablet Manufacture

Roller-compacted formulations intended for moisture-sensitive thiol APIs eliminate aqueous binder exposure and thereby remove hydrolytic stress and dissolved oxygen introduced during wet massing. The pre-blend contains 15.0 wt% API, 58.0 wt% spray-dried mannitol, 21.0 wt% microcrystalline cellulose, 4.0 wt% croscarmellose sodium, 0.5 wt% colloidal silicon dioxide, and 1.5 wt% sodium stearyl fumarate. Blending is performed in a 600 L bin blender at 10 rpm for 20 min. Roller compaction is run on an Alexanderwerk WP120 or equivalent with smooth rolls at hydraulic pressure 40–70 kN, roll speed 4–8 rpm, and gap width 2.0–2.5 mm. The target ribbon density is 1.15–1.25 g/cm³, monitored by Ph.Eur. 2.9.34. Ribbons are milled through a 1.0 mm screen followed by a 0.5 mm screen to final granule D50 200–300 µm. Roll surface temperature is maintained below 35 °C with chilled water to prevent local thiol oxidation at compaction points. Final compression uses 9 mm concave punches at 8–14 kN to tablet hardness 70–110 N. Dry granulation is preferred when the API has low ductility and cannot tolerate even short aqueous contact; however, compactability may be lower than direct compression because the thiol particle surface undergoes plastic deformation during ribbon formation. Friability is controlled at ≤ 0.6% per Ph.Eur. 2.9.7, and content uniformity follows USP <905>. The finished tablet is a moisture-protective, film-coated immediate-release product with measured bulk and tapped density data retained per USP <616>.

Low-shear encapsulation of the API in hard gelatin capsules is selected when dose flexibility and rapid disintegration are required without exposure to aqueous binder systems. A size 3 capsule is filled with 200 mg of blend containing 10 mg API, 143 mg lactose monohydrate, 36 mg pregelatinized starch, 5 mg talc, 4 mg sodium starch glycolate, and 2 mg magnesium stearate. The API and diluents are blended in a low-shear tumble blender at 12 rpm for 20 min; magnesium stearate is added and mixed for 3 min to limit over-lubrication. Capsule filling is run on an MG2 Planeta or Bosch GKF 1500 with dosator or tamping-pin configuration; fill weight RSD is controlled ≤ 2.0%. The gelatin shell moisture content is 13–16% w/w, and the finished capsule is packaged with a 1 g silica gel desiccant because moisture migration from the shell into the thiol-containing blend accelerates disulfide formation. Disintegration testing follows USP <701> with a limit of not more than 15 min in water at 37 °C. Capsule formulation compatibility with gelatin is verified per USP <2040> and ICH Q1A. The terminal product is a hard gelatin capsule, not a coated tablet, and it relies on the capsule shell to mask the API taste during oral administration.

Aqueous Injectable Stabilization Through pH Shift and Nitrogen Blanketing in Type I Glass Vials

Aqueous injectable presentation of 2-mercaptobenzimidazole requires an oxygen-controlled formulation and filling train because the thiol sulfur oxidizes rapidly in neutral to weakly alkaline solution. The formulation uses 5 mg/mL API in 10 mM sodium phosphate buffer, 8.0 mg/mL sodium chloride to target osmolality of 290 mOsmol/kg, and 0.02% w/v disodium edetate. The pH is adjusted to 7.0 ± 0.2 with 1 M sodium hydroxide or 1 M hydrochloric acid; pH excursion below 5.0 or above 9.0 is avoided because it promotes metal-soluble species and thiolate oxidation. The formulation vessel is sparged with filter-sterilized nitrogen until dissolved oxygen is below 0.5 mg/L, and the solution is filtered through a 0.22 µm PVDF membrane under nitrogen pressure. Aseptic filling is performed in a Grade A unidirectional airflow zone at 0.45 m/s ± 20% into 10 mL Type I borosilicate glass vials depyrogenated at 250 °C for 45 min. Stoppers are low-extractable chlorobutyl coated with silicone, selected to minimize peroxide and trace zinc migration. Terminal sterilization by autoclaving at 121 °C for 15–20 min is not assigned as a primary limit because published data for this specific configuration is limited; if terminal sterilization is required, the disulfide impurity must remain below the ICH Q3A qualification threshold after the cycle. For lyophilized presentation, the same solution is filled at 5 mL into 10 mL vials, shelf-frozen to −40 °C, primary-dried at −20 °C and 100 mTorr for 24 h, then secondary-dried at 25 °C for 6 h to residual moisture < 1.0%. Stoppering is performed under nitrogen headspace at 700 mbar. Injectable release includes particulate matter per USP <788>, visible particulates per USP <790>, bacterial endotoxins per USP <85>, sterility per USP <71>, and subvisible particle testing per Ph.Eur. 2.9.19. Manufacturing compliance follows EU GMP Annex 1 for aseptic processing and FDA 21 CFR 211.67 for equipment cleaning. Contact surfaces in holding tanks and filling lines are borosilicate glass or PTFE-lined 316L stainless steel; bare stainless steel is avoided for aqueous holding times beyond 6 h because trace iron leaching catalyzes oxidative degradation.

Fluid-bed drug layering onto inert core spheres produces the oral granule dosage form with a lower surface-area-to-mass ratio than direct compression blends and allows dose-specific sachet filling. A spray suspension contains 10.0 wt% API, 4.0 wt% povidone K30, 1.0 wt% talc, and 0.05 wt% sodium metabisulfite in deaerated purified water. The suspension is sprayed onto 500–600 µm sucrose or mannitol starter spheres in a Glatt GPCG 5.1 Wurster column at inlet air temperature 55 °C, product temperature 38–42 °C, spray rate 2–4 g/min, and atomizing pressure 1.5 bar. Inlet air is nitrogen-enriched to limit oxidative load. The resulting granules have D50 600–800 µm, loss on drying ≤ 2.0% per Ph.Eur. 2.2.32, and are filled into PET/Al/PE sachets under nitrogen at fill weight 500 mg with fill weight variation ≤ 5%. Disintegration of granules follows Ph.Eur. 2.9.1; dissolution is tested per USP <711> in pH 6.8 buffer. This oral granule presentation is distinct from tablet dry granulation because the API is deposited as a thin layer rather than compacted into a ribbon, which reduces localized thermal stress but increases sensitivity to fluid-bed humidity. Residual solvent testing follows ICH Q3C, and the final sachet product is a single-dose oral granule intended for reconstitution or direct administration.