Custom peptide APIs supplied for oral solid dose use are released against three particle-level attributes that govern downstream blend homogeneity: desalted peptide content on the anhydrous free-base equivalent basis, residual counterion identity and content, and particle size distribution after final drying. A lyophilized peptide acetate with a tapped bulk density of 0.15–0.35 g/cm³ and an angle of repose above 40° will not sustain feed-frame flow on a rotary tablet press unless the API is either spray-dried with mannitol or pre-blended by geometric dilution in a bin blender. Milling through a 0.5 mm conical screen at 1,000–2,000 rpm reduces agglomerates but also increases amorphous surface area; if the peptide is hygroscopic, water uptake above 2.0% w/w during milling shifts the glass transition below ambient and causes sticking during subsequent slugging or capsule tamping. Release testing for solids includes assay by reversed-phase UPLC with peptide content expressed as free-base equivalent, individual related impurities evaluated under ICH Q3A(R2) thresholds, residual water by USP <921>, residual counterion by ion chromatography, and microbiological limits for oral products per USP <61> and USP <62>. The API should be received with a certificate of analysis that includes peptide assay 95.0–105.0% on anhydrous basis, water content ≤3.0% w/w, and endotoxin less than 0.25 EU/mg for oral solid applications if the same material may later enter parenteral development; endotoxin data are not release parameters for oral products but avoid re-qualification when the site is audited under EU GMP Annex 1 principles. Blend development then uses a 1:10 pre-blend of API with mannitol 200 SD or spray-dried lactose monohydrate because direct addition of a low-dose peptide below 1.0% w/w nominal content creates content uniformity failure risk under USP <905>. Sodium stearyl fumarate at 0.5–1.5% w/w is preferred over magnesium stearate when peptide acetates are sensitive to magnesium-catalyzed aspartic acid isomerization; the lubricant is added last and mixed for 3–5 min to avoid overlubrication, which prolongs disintegration beyond the upper limit of 30 min for uncoated tablets under USP <701>.

What Limits Direct Compression of Lyophilized Peptide APIs Without Cryogenic Co-Processing?

Direct compression of a lyophilized peptide API is limited by poor flow, low bulk density, and viscoelastic recovery after compaction. Feasibility batches on an instrumented rotary press with 8 mm round concave tooling show that the peptide fraction in a direct-compression blend is often held below 5–10% w/w because higher loads generate capping after ejection. The compaction event is monitored with compression force and displacement data; out-of-die Heckel analysis identifies a yield pressure that shifts when the peptide is pre-compacted by dry granulation. Lubricant concentration is a critical threshold: 0.5% w/w magnesium stearate provides sufficient anti-adhesion, but 1.5% w/w reduces tablet hardness by more than 20% and pushes disintegration beyond 15 min in the same formula. The release criterion for uncoated cores is friability <1.0% after 100 drops per USP <1216>, hardness 40–80 N, and assay 95.0–105.0% of label claim. Because peptide acetates are frequently amorphous with a glass transition between 30 °C and 70 °C, tableting speed is limited to 20–50 rpm and die temperature measured by infrared thermography should remain below 40 °C; higher temperatures induce sticking to upper punches and formation of cleaved peptide fragments. If the API is lyophilized and then milled, residual moisture above 2.5% w/w acts as plasticizer and lowers the glass transition, worsening sticking and increasing related impurities after compression. Direct compression is therefore reserved for formulations where the peptide content is low, the filler is compressible mannitol or cellactose, and the target tablet weight is above 150 mg.

Wurster Fluid-Bed Layering and Enteric pH Differential

Granule manufacturing for acid-labile peptide APIs uses Wurster fluid-bed layering because the peptide is exposed to lower shear than high-shear granulation and the coating step can be completed in the same chamber. The peptide is dissolved in purified water with hydroxypropyl methylcellulose 3 mPa·s at binder solids 2–5% w/w of total solution weight; the solution is sprayed onto microcrystalline cellulose spheres 250–355 µm at an inlet air temperature of 50–60 °C and product temperature 35–42 °C. The spray rate is adjusted so bed humidity remains below the deliquescence point of the peptide salt; a pressure drop across the distribution plate of 4–8 mbar is used to maintain a stable fountain. After peptide layering, an enteric coat of methacrylic acid-ethyl acrylate copolymer Eudragit L30 D-55 is applied as an aqueous dispersion at 20–30% dry polymer weight gain with triethyl citrate 10–15% w/w of dry polymer as plasticizer and talc 30–50% w/w of dry polymer as anti-tacking agent. The enteric coating is cured at 30–35 °C for 2–4 h to allow film coalescence; incomplete curing yields pinhole defects and premature peptide release in 0.1 N HCl during acid-stage dissolution testing. Release testing per USP <711> uses a two-stage pH shift: acid stage 0.1 N HCl for 2 h with no more than 10% peptide released, followed by phosphate buffer pH 6.8 with at least 80% released within 45 min. Sieve analysis of the final granules is controlled at D10 > 200 µm and D90 < 1,000 µm to ensure uniform capsule filling or sachet dosing; bulk density is typically 0.45–0.70 g/cm³. Terminal products may be filled into hard capsules or sealed into aluminum sachets, and water activity below 0.60 is maintained to avoid peptide hydrolysis during shelf life.

When Low-Fill Capsule Formulations Require Desiccant Protection and Tamping Pin Control

When a custom peptide API is filled into two-piece hard gelatin or HPMC capsules at fill weights below 100 mg, the dominant process risk is moisture transfer from the shell to a hygroscopic peptide powder, which can cause local deliquescence and capsule embrittlement. Gelatin shells are most stable at 35–50% RH; exposure below 30% RH increases brittleness and above 60% RH softens the shell. The powder blend is therefore dried to water activity 0.30–0.50 and filled in an environment maintained at 25 ± 2 °C and 35 ± 5% RH. Dosator and tamping pin geometries are selected according to the flow function coefficient of the peptide-excipient blend; blends with a flow function coefficient below 5 have poor capsule fill weight control and require colloidal silicon dioxide at 0.2–0.5% w/w or glidant-free re-granulation. Fill weight uniformity is monitored at 15 min intervals and must meet USP <905> content uniformity at stage 1. A desiccant canister of silica gel or molecular sieve is inserted when the packaged configuration is HDPE with a polypropylene cap; the desiccant quantity is calculated from moisture permeation of the bottle at 40 °C/75% RH per ICH Q1A(R2) accelerated conditions. For blister packaging, cold-form aluminum laminate provides moisture vapor transmission below 0.05 g/m²/day, which is lower than Aclar and PVC/PVDC, and is used when the peptide salt shows deliquescence below 60% RH. The filled capsule is tested for dissolution in USP apparatus II at 50 rpm with media selected according to peptide solubility; disintegrated capsule shells must not form crosslinked films that delay release beyond 60 min.

For sterile injectable liquid formulations, the aqueous solubility of the peptide acetate is mapped over a pH range of 3.0–7.5 before buffer selection, because ionic strength and pH affect both solubility and aggregation. A target concentration of 1–10 mg/mL is formulated in 10–50 mM acetate or histidine buffer; phosphate is avoided for peptides containing divalent cation-binding motifs or when calcium-dependent degradation is observed. Tonicity is adjusted with mannitol or trehalose to 280–320 mOsm/kg; for acidic peptides, sodium chloride is used only if the peptide has no salt-induced aggregation at 0.9% w/v. The solution is filtered through a 0.22 µm PVDF or PES membrane; filter compatibility testing includes peptide adsorption, which can exceed 10% on unmodified nylon membranes. Sterile filtration under aseptic conditions follows EU GMP Annex 1; terminal sterilization by moist heat at 121 °C for 15 min is rarely used because peptide deamidation and aggregation increase, so filtration is the default when the peptide has a molecular weight below 10 kDa and can pass through the membrane without shear-induced aggregation. Subvisible particles are controlled by USP <787>: for small-volume parenterals the acceptance criterion is no more than 6,000 particles per container for ≥10 µm and no more than 600 per container for ≥25 µm. The filled solution is inspected by automated light obscuration for visible particles per USP <790>, and the container closure system is tested by USP <1207> with dye ingress or vacuum decay. A pH specification of ±0.2 units around the optimum is applied because pH drift promotes aspartic acid isomerization and glutamine deamidation; for long-term storage at 2–8 °C, the solution formulation may be limited to 24 months unless freeze-dried.

Lyophilization Cycle Design Maps Collapse Temperature Before Primary Drying

Collapse temperature mapping is performed by freeze-drying microscopy before primary drying is designed for a peptide injection vial. The formulated solution containing a peptide at 1–20 mg/mL and a bulking agent such as mannitol, glycine, or trehalose at 2–5% w/v is frozen at 0.5–1.0 °C/min; the collapse temperature typically falls between -15 °C and -35 °C depending on the amorphous phase composition. Annealing at -20 °C for 2–4 h crystallizes mannitol and fills the void volume, which reduces vial-to-vial cake height variability. Primary drying is then conducted at a shelf temperature 10–20 °C below the collapse temperature, often -30 °C to -10 °C, with chamber pressure 50–150 mTorr. The endpoint is determined by comparative pressure measurement between a Pirani gauge and a capacitance manometer; a pressure ratio approaching 1.0 indicates near-complete sublimation. Secondary drying is run at 25–40 °C for 4–12 h to reduce residual moisture below 1.0% w/w by USP <921>. The resulting cake should be uniform with no shrinkage, meltback, or collapsed edges; reconstitution time is controlled to <2 min with 1.0 mL of water for injection. Residual moisture is critical because levels above 2.0% w/w accelerate solid-state deamidation and peptide aggregation after storage at 25 °C/60% RH. Vials are stoppered under nitrogen overlay with oxygen headspace below 5%; container closure integrity is verified by USP <1207> dye ingress or laser-based headspace analysis. Batches that fail collapse temperature mapping cannot be routinely freeze-dried without reformulation with higher ratios of trehalose to mannitol or a lower peptide concentration. Process-scale lyophilizers with shelf temperature uniformity of ±1 °C across all shelves are required; poor uniformity at the front and rear shelves causes non-uniform residual moisture and variable reconstitution times.

Prefilled syringe and pen injector formats require higher peptide concentrations than vial-based injections because dose volumes are fixed below 1.5 mL. A peptide acetate is concentrated to 10–50 mg/mL by tangential flow filtration or lyophilization-reconstitution, and viscosity is measured at 20 °C with a cone-and-plate viscometer at 100 s⁻¹. Silicone oil from the glass barrel can induce aggregation in a shear-sensitive peptide; siliconized barrels are evaluated by subvisible particle counts after 24 h under horizontal stress. The formulation requires a plunger glide force below 20 N at 300 mm/min injection speed to ensure patient usability, while maintaining bubble-free filling with a peristaltic pump or ceramic rotary piston pump. Filling accuracy is controlled to ±2% of target volume; for a 0.5 mL dose, this corresponds to ±10 µL. The needle gauge, typically 29 G or 31 G, raises injection force and may require citric acid or sodium chloride tuning. Terminal sterilization is not compatible with many peptides; aseptic filling in an ISO 14644-1 Class 5 environment is used, and container closure integrity is retested after shipping per USP <1207>. Stability at 2–8 °C is the primary limit; agitation during transport can produce air-liquid interface aggregation, so polysorbate 20 at 0.001–0.01% w/v or trehalose at 1–5% w/v is included only if aggregation data support it.

Controlling Burst Release in PLGA Depot Microspheres

In PLGA depot manufacturing, the peptide API is incorporated into poly(lactic-co-glycolic acid) microspheres by a water-in-oil-in-water double emulsion. The inner aqueous phase contains the peptide at 10–50 mg/mL with stabilizers such as gelatin or cyclodextrin; the oil phase contains PLGA 50:50 or 75:25 at 5–15% w/v in dichloromethane. The primary homogenization at 10,000–15,000 rpm forms a W/O emulsion, which is then poured into an aqueous phase containing polyvinyl alcohol at 0.5–2.0% w/v at 2–8 °C with stirring at 500–1,000 rpm. Solvent extraction is performed by dilution into water for 2–4 h; the rate of solvent removal determines the porosity of the polymer matrix and the initial burst. Microspheres are collected through 25–125 µm sieves and lyophilized to residual moisture <1.0% w/w. Residual dichloromethane is analyzed by gas chromatography per USP <467> and must not exceed the 600 ppm concentration limit assigned to dichloromethane under ICH Q3C. In vitro release is conducted in 100 mM phosphate-buffered saline at pH 7.4 and 37 °C; an initial burst of <15% peptide released within 24 h is typical for a depot intended for weekly to monthly administration, but formulations with porous surfaces may release more than 30%, which requires annealing or a second polymer coat. Sterile manufacture uses aseptic filtration of the external phase only; microspheres cannot be sterile-filtered and are produced under ISO 14644-1 Class 5 conditions with terminal gamma irradiation at 25–40 kGy if the peptide is stable to radiation. Dose uniformity is evaluated by peptide content per vial after reconstitution with 1.5–2.0 mL diluent; the suspension must pass syringeability through a 21 G needle without plugging. This dosage form is limited by microsphere batch-to-batch particle size variation, which can exceed 10% RSD if solvent extraction temperature and stirring speed are not held within narrow ranges; for a custom peptide with unknown radiation stability, published data for the specific configuration is limited and electron-beam sterilization may require validation.