Direct compression of methylcobalamin (mecobalamin) API at label claims between 250 µg and 1,500 µg per unit is governed less by compressibility than by segregation potential and photolytic loss from the corrin macrocycle. A pre-blend is prepared by passing the API through a 60-mesh screen (250 µm) with mannitol 200 SD or pregelatinized starch at a ratio of 1:10; this first cut is then diluted geometrically to 1:100 before addition to the main tablet blend. Blending in a 300 L to 1,200 L bin blender at 6 rpm to 10 rpm for 18 min to 25 min produces blend uniformity RSD values below 3.0% when sampled at 10 locations. However, higher-speed tumble cycles above 12 rpm have been observed on production lines to induce static charge and cause methylcobalamin to adhere to polished stainless steel surfaces, reducing assay by 0.5% to 1.5% before compression. Magnesium stearate is added at 0.5% to 1.0% w/w after the API pre-blend has achieved target uniformity; lubrication time is limited to 3 min to 5 min because the hydrophobic film generated by extended mixing suppresses dissolution of low-dose methylcobalamin particles embedded in the lubricant layer. Tablets are compressed on a rotary tablet press equipped with 6 mm to 8 mm round tooling at 5 kN to 15 kN compression force, with average hardness of 5 kp to 10 kp and friability not exceeding 1.0% when tested per USP <1216>. The finished tablet blend is held at 25°C and 30% RH to 35% RH; cumulative exposure to visible light above 1,000 lux during compression and film coating is limited to less than 4 h because the red methylcobalamin surface layer can bleach to hydroxocobalamin, producing a measurable increase in related substances above 0.5%. Content uniformity after compression is evaluated per USP <905>; an acceptance value ≤ 15.0 is required, but low-dose methylcobalamin tablets with D90 below 74 µm and geometric pre-blending typically achieve AV values of 3.0 to 8.0. Dissolution is tested per USP <711> Apparatus 2 at 50 rpm in 900 mL degassed water at 37°C; a validated Q value of 75% at 45 min is commonly registered, with measured variability of 4% to 8% RSD for direct compression formulations. Where a regulatory monograph does not specify a methylcobalamin dissolution method, the limit is established under ICH Q2(R2) and supported by accelerated stability data.

What Changes When Methylcobalamin Moves From Tablets to Hard Capsules?

Capsule formulations are generally filled at 0.33% to 1.0% w/w API load when the target claim is 0.5 mg per size 3 or 4 capsule. The critical operation is the first dilution, not the filling speed. A ratio of 1:5 API to lactose monohydrate followed by a 1:10 second dilution is frequently insufficient for dosator-style capsule machines, because the tailings from the dosator chamber can contain API-depleted fines that exceed 10% relative standard deviation if the pre-blend is not de-lumped through a 40-mesh screen. 0.5% to 1.0% sodium stearyl fumarate may replace magnesium stearate when hard gelatin capsule shells are sensitive to alkaline lubricant residues, although published data for ternary methylcobalamin–lactose–sodium stearyl fumarate blends is limited and requires accelerated stability confirmation. The filled capsules are packaged in aluminum/aluminum cold-form blister strips; PVC/PVDC blisters are avoided for tropical markets because methylcobalamin-loaded powder can absorb moisture above 60% RH and show related substance drift of approximately 0.3% over six months at 40°C/75% RH. The terminal capsule product is tested for disintegration per USP <701> with a limit of not more than 30 min, and for assay and content uniformity per USP <905> to confirm that the low-dose filling operation has not created super-potent or sub-potent units in the batch.

Roller-Compacted Granules for Light-Protected Sachet Lines Require Dry Binder Adjustments

Roller compaction is selected over aqueous wet granulation because the contact time between solubilized methylcobalamin and granulating water is zero. The API is first blended with microcrystalline cellulose at 20% to 35% w/w, mannitol at 40% to 60% w/w, and crospovidone at 2% to 5% w/w in a low-shear V-blender at 10 rpm for 15 min. The resulting blend is passed through a roller compactor with 1.0 mm roll gap and 20 kN/cm to 40 kN/cm specific roll force to produce ribbon density of 1.05 g/cm³ to 1.20 g/cm³. Ribbons are milled through a 1.0 mm screen, and the retained fraction above 20-mesh is remilled; granules below 200-mesh are minimized because low-dose methylcobalamin tends to segregate into fine fractions under vibration during sachet filling. Single-dose sachet fill weight is maintained at 500 mg to 2,000 mg depending on pediatric versus adult presentation; fill weight variation is held to ±5% by mass, and content uniformity per USP <905> is confirmed by sampling 3 sachets from beginning, middle, and end of the filling run. When the granule is reconstituted in 20 mL purified water at 25°C, methylcobalamin release is immediate and no insoluble residue is observed. The oral liquid prepared from granules is administered within 30 min under reduced light; prolonged standing under fluorescent light causes a color shift from red to purple-brown, indicating conversion to hydroxocobalamin.

Aqueous injectable solutions of methylcobalamin are compounded under light exclusion and inert headspace because the dissolved molecule undergoes photoaquation to hydroxocobalamin within minutes under 5,000 lux white light. The formulation is typically buffered with 0.1% to 0.5% sodium acetate or sodium citrate at pH 5.5 to 7.0, and the vehicle is Water for Injection purged with pharmaceutical-grade nitrogen to maintain dissolved oxygen below 0.5 ppm. Methylcobalamin injection is filled as 0.5 mg/mL or 1.0 mg/mL nominal strength into 1 mL to 2 mL amber Type I borosilicate glass ampoules or vials. Terminal sterilization by steam at 121°C is not applicable because the corrin ring undergoes thermal decomposition; therefore the solution is sterile-filtered through a 0.22 µm PVDF or PES membrane under EU GMP Annex 1, and the viable process hold time in bulk is limited. The container-closure system must exclude UV wavelengths below 400 nm; amber glass meeting USP <660> is commonly specified, and secondary cartons are required because even amber ampoules transmit some visible light above 450 nm. Particulate matter is controlled to not more than 6,000 particles ≥ 10 µm and 600 particles ≥ 25 µm per container under USP <788>, adapted for small-volume parenterals. Compounding with dextrose or alkaline bicarbonate solutions is not recommended because pH shifts above 8.0 accelerate corrin ligand substitution; the solution may be diluted with 0.9% sodium chloride injection immediately before use, and light-shielded infusion sets are required when the diluted product is administered over more than 15 min. Transition-metal ion contamination is controlled to 1 ppm or below for iron and copper because these ions catalyze free radical oxidation of the reduced cobalt center when dissolved oxygen is not rigorously excluded; ICH Q3D limits for parenteral products apply per the product’s permitted daily exposure. Admixture with ascorbic acid at acidic pH is not recommended because the reduced corrin can be oxidized to aquocobalamin; published stability data for this specific binary combination is limited.

When Lyophilization Is Selected Over Terminal Sterilization, Cake Structure Dictates Reconstitution Time

Lyophilization is used when aqueous methylcobalamin stability does not support a shelf-life acceptable for marketed ampoule solution. The freeze-dried cake is prepared from a sterile-filtered solution containing 0.5 mg/mL or 1.0 mg/mL methylcobalamin, 4% to 6% w/v mannitol or trehalose, and a pH 5.5 to 6.5 acetate buffer. The vial is filled to 1 mL or 2 mL, partially stoppered, and loaded into a lyophilizer with shelf temperature ramped from 5°C to -45°C at 0.5°C/min to 1.0°C/min. Primary drying is executed at a shelf temperature of -25°C to -15°C and chamber pressure of 0.15 mbar to 0.25 mbar for 24 h to 48 h; collapsing the cake by exceeding the collapse temperature produces a shrunken glossy mass that fails visual inspection and reconstitutes beyond 3 min. Secondary drying at 25°C to 35°C for 6 h to 12 h reduces Karl Fischer moisture to ≤ 2.0% w/w. The lyophilized product is reconstituted with 1 mL to 2 mL of Water for Injection, 0.9% sodium chloride, or 5% dextrose; the resulting solution is used within 1 h under light protection. For this specific lyophilized configuration, published thermomechanical data on the maximally freeze-concentrated glass transition for methylcobalamin–mannitol systems is limited, so process development must confirm the collapse boundary experimentally rather than by extrapolation from non-cobalamin bulking agents.

Orodispersible Tablet Formulation Limits and Superdisintegrant Loading

Orodispersible methylcobalamin tablets for buccal or sublingual delivery are compressed with a superdisintegrant loading that balances disintegration time against low-dose content uniformity. Crospovidone at 2% to 5% w/w and fine mannitol at 60% to 80% w/w produce tablets with hardness 2.5 kp to 4.5 kp and friability 0.8% to 2.0%; these tablets disintegrate in 15 s to 30 s in 2 mL water at 37°C per USP <701>. The main process failure is lamination during high-speed compression because the formulation is intentionally weak; tablet press turret speed is limited to 30 rpm to 50 rpm on a 16-station to 36-station rotary press, and pre-compression force is set to 1.5 kN to 3.0 kN before main compression at 4 kN to 6 kN. The API is pre-dispersed into mannitol at 1:20 to avoid segregation in the shallow die. The terminal product is packaged in cold-form aluminum blister with 10 g/m² to 15 g/m² aluminum foil to block water vapor and light; a desiccant pouch may be added in humidity zones where ambient RH exceeds 50% during storage.