During direct compression of Bb-G90 lyophilizate for tablets, the API is characterized by laser diffraction, water activity, and anaerobic plate count before blending. The powder is typically milled through a 600 µm stainless steel sieve and blended in a bin blender at 10 rpm for 15 min with microcrystalline cellulose PH-102 and anhydrous lactose; anhydrous lactose is selected because its low moisture uptake assists in holding water activity below 0.20. Magnesium stearate is screened with a portion of the blend and added at 0.5–1.0% w/w, with lubricant blending limited to 3–5 min to avoid segregation and localized shear. Direct compression on a rotary tablet press fitted with 10 mm round concave B-tooling and a gravity feed frame is operated with pre-compression at 2.5 kN and main compaction force below 12 kN; higher force is associated with loss of CFU because compression shear disrupts the bacterial cell envelope, and viability is confirmed by enumeration on MRS agar supplemented with 0.05% w/v L-cysteine hydrochloride under anaerobic incubation at 37 °C for 72 h. Tablet hardness is maintained at 40–70 N, friability is below 0.8% w/w per USP <1216>, and disintegration time is tested according to USP <701>. Tablets intended for ileo-colonic release are coated with a methacrylic acid–ethyl acrylate copolymer dispersion; coating weight gain is held at 6–10% w/w, the coating pan inlet air temperature is maintained at 30–35 °C, and bed relative humidity below 45% RH to avoid premature water uptake. The delayed-release performance is tested according to USP <711> with 0.1 N hydrochloric acid for 2 h followed by pH 6.8 phosphate buffer. Final tablet moisture is controlled at ≤ 5.0% w/w and water activity at ≤ 0.20; if bulk powder water activity exceeds 0.25, the dry blend is adjusted with anhydrous excipients rather than subjected to heat drying. Published data for Bb-G90 at production scale are limited, so viability loss must be re-qualified on each rotary press and coating configuration instead of extrapolating from laboratory values.

What Changes When Bb-G90 Enters a Capsule Filling Line at 55% RH?

Encapsulation of Bb-G90 powder in low-moisture hydroxypropyl methylcellulose capsules requires stricter environmental control than conventional immediate-release powders. Gelatin shells are generally avoided because shell moisture transfers to the fill and raises water activity above the probiotic stability threshold; HPMC capsule shells with residual moisture 4.0–7.0% w/w are conditioned at 20–25 °C and ≤ 35% RH for 48 h before filling. Tamping-pin capsule machines generate frictional heat at the pin and ring stations; pin compression force is adjusted to keep localized temperatures below 30 °C, and machine speed is reduced if the capsule body temperature measured by an infrared sensor exceeds that limit. Dosator fillers are suitable for granulated Bb-G90 because axial slug compression is brief, but the tapped density of the fill should not exceed 0.65 g/mL to avoid crushing the lyophilized particles. Static charging of the powder blend is controlled by maintaining floor humidity at 45–55% RH only outside direct product contact zones and by grounding all contact parts; high static charge causes powder sticking to the dosator bore and excessive fill weight variation. Filled capsules are band-sealed or liquid-sealed with HPMC-based polymer, and the headspace is purged with nitrogen to achieve residual oxygen below 5.0% v/v before sealing. Capsule weight variation is checked per USP <905> after sealing. Microbial enumeration is repeated after encapsulation according to USP <2021> or Ph. Eur. 2.6.36; a CFU reduction greater than 0.5 log from the blend to the finished capsule triggers an investigation. Aluminium/aluminium blister packaging with desiccant can hold water activity below 0.20 for a 24-month shelf life under ICH Q1A conditions at 30 °C/75% RH, but Bb-G90 strain-specific stability data remain the controlling validation source.

Wet Granulation of Bb-G90 Is Replaced by Dry Granulation Once Granulate Moisture Exceeds 3.0%

In sachet and stick-pack production, fluid-bed wet granulation with aqueous binders is not the first-line process for Bb-G90 because added water plus dryer heat can reduce viable counts by more than 1.0 log CFU/g. Roller compaction is applied when granulation is required for flow and low-dust filling. The roller compactor is configured with a roll gap of 1.0–1.5 mm, roll force 10–20 kN/cm, and granulator sieve aperture 800 µm; resultant granules have loose bulk density 0.45–0.55 g/mL, which is sufficient for volumetric sachet filling on a stick-pack machine. Fines below 150 µm are recycled into the feed hopper at no more than 20% w/w to maintain consistent powder flow without over-densifying the granulate. If wet granulation is unavoidable for taste masking or prebiotic incorporation, the granulation solvent system is restricted to an aqueous ethanol mixture, and the wet mass is vacuum-dried at a shelf temperature not exceeding 25 °C; oven drying above 35 °C is not used for viable bifidobacterial preparations. Sachet fill weight variation is controlled per Ph. Eur. 2.9.5 or USP <905>, and powder flow is measured by Ph. Eur. 2.9.36 with a Hausner ratio below 1.25 as the acceptance limit. Post-fill moisture ingress is managed with high-barrier laminate containing aluminium foil of at least 9 µm thickness; desiccant is added to the sachet to hold headspace relative humidity below 30% RH during use. The granulate is tested for water activity by USP <922> with a release limit of ≤ 0.20, rather than relying on moisture percentage alone.

For oral or enteral administration, reconstitution of Bb-G90 lyophilized powder is performed immediately before dosing because rehydrated cells lose viability rapidly in water. A suspending vehicle containing 2.0% w/v sucrose, 0.5% w/v sodium chloride, and 0.1% w/v xanthan gum is prepared and cooled to 4–8 °C before the powder is dispersed with gentle stirring. Reconstitution viscosity is maintained below 50 mPa·s at 25 °C to permit passage through a 10 Fr nasogastric tube without cell sedimentation. The reconstituted suspension is not filtered; instead the API powder is pre-screened through a 315 µm stainless steel sieve to remove lumps before dosing. Potable water used for reconstitution is boiled and cooled, or sterile water is used, because chlorine residuals above 0.5 ppm can reduce viable counts. The vehicle pH is adjusted to 5.5–6.0 with citrate buffer; pH below 4.0 accelerates loss of membrane integrity and shortens in-use hold time. Administration through a closed enteral syringe is completed within 30 min of reconstitution, and any unused portion is discarded. The oral liquid dosage form is labelled with an in-use stability of 2 h at 2–8 °C until site-specific validation demonstrates longer stability. Published data for Bb-G90 in oral liquid form are limited, so in-use stability protocols should use the strain manufacturer’s CFU assay as the decision criterion.

When Bb-G90 Lyophilizate Is Prepared as an Injectable Parenteral Suspension

Although Bb-G90 is supplied as a pharma-grade API for injection, the parenteral administration of live bifidobacterial cells is not established in general monographs and requires a specific live biotherapeutic product licence. Terminal sterilization by autoclaving at 121 °C for 15 min or gamma irradiation at 25 kGy is not compatible with viability; therefore, the entire process is aseptic under EU GMP Annex 1 and FDA 21 CFR Parts 210/211. The filling suite is maintained at ISO 5 under Grade A conditions with unidirectional airflow; settle plates and active air sampling are monitored throughout the campaign. The lyophilized API is reconstituted aseptically with a vehicle composed of 0.9% w/v sodium chloride, 2.0–5.0% w/v sucrose or trehalose, and 0.02% w/v polysorbate 80, adjusted to pH 6.0–6.5. The diluent is prepared and filtered through a 0.2 µm membrane; whole bacterial cells cannot pass through a sterilizing filter, so the final suspension is not terminal-filtered. Osmolality is controlled at 280–320 mOsm/kg to reduce osmotic shock; the reconstitution step is delayed if the diluent temperature exceeds 8 °C. For injectable dosage forms, bacterial endotoxin and pyrogen testing are required per Ph. Eur. 2.6.14 or USP <85>; however, cell-wall components of Bifidobacterium may interfere with the limulus amoebocyte lysate assay, so method suitability includes inhibition/enhancement controls per Ph. Eur. 2.6.14. Injectable Bb-G90 is limited to the intended local or investigational route specified in the clinical protocol, and intravenous administration of live bacteria is generally contraindicated unless a validated product-specific reference supports it.

Aseptic Lyophilization Cycle Design and Vial Closure Integrity for Bb-G90 Parenteral Preparations

Within the freeze-drying chamber, vials containing 1.0–2.0 mL of formulated Bb-G90 suspension are loaded aseptically in 2R or 4R glass vials. The shelf is ramped from 4 °C to −40 °C at 0.5 °C/min and held for 4 h; an annealing step at −20 °C for 2 h may be introduced to improve cake structure and reduce primary drying time. Primary drying is conducted at 0.2 mbar chamber pressure with a shelf temperature of −20 °C for 48–72 h, followed by secondary drying at 25 °C for 6–10 h. Product temperature must remain below the collapse temperature of the sucrose-mannitol matrix; if collapse occurs, the cake becomes dense and reconstitution time exceeds 10 min with foaming. The lyophilized cake is stoppered under nitrogen with headspace oxygen below 5.0% v/v; residual moisture is determined by Karl Fischer titration per Ph. Eur. 2.5.12. Container closure integrity is verified by vacuum decay or helium leak testing per USP <1207> or Ph. Eur. 3.2.9. The vial is labelled with a reconstitution diluent volume that yields a cell concentration of not less than 1 × 10⁹ CFU/mL; Bb-G90-specific recovery after lyophilization is confirmed by plate count on reinforced clostridial medium or MRS agar supplemented with 0.05% w/v L-cysteine hydrochloride under anaerobic incubation at 37 °C for 72 h. Batch-to-batch variance in lyophilization survival remains a known production bottleneck; cycle robustness is assessed by comparing pre-lyophilization and post-lyophilization CFU counts, and a viability loss greater than 1.0 log CFU/vial triggers a deviation investigation.