Production-scale lyophilization of pemetrexed disodium heptahydrate begins with the pH-dependent solubility of the disodium salt and the seven lattice water molecules present in the crystalline hydrate. The sterile drug product is prepared as a lyophilized cake containing mannitol as a bulking agent; the hydrate is equivalent to either 100 mg or 500 mg of pemetrexed per vial. Reconstitution is performed only with 0.9% sodium chloride injection, using 4.2 mL for the 100 mg vial and 20 mL for the 500 mg vial to yield a nominal concentration of 25 mg/mL. The pH after reconstitution is controlled to 6.6–7.8. The molecule contains multiple carboxylic acid groups, and the FDA-approved labeling excludes diluents containing calcium, including lactated Ringer’s injection and Ringer’s injection, because poorly soluble calcium complexes may form. A commercial freeze-drying cycle for this type of mannitol-based formulation usually includes an annealing step above −40°C to crystallize mannitol while the product remains frozen, followed by primary drying at shelf temperatures below the collapse temperature and chamber pressure between 50 mTorr and 150 mTorr. Comparative pressure measurement with a Pirani gauge and a capacitance manometer is used to determine the end of primary drying. The lyophilized cake is tested for sterility by membrane filtration under USP <71>, bacterial endotoxins under USP <85>, subvisible particulate matter under USP <788>, and visible particulates under USP <790>. Residual moisture is controlled to not more than 2.0% w/w because excess water can accelerate hydrolysis of the polyglutamated antifolate and reduce shelf-life consistency. The product contains no antimicrobial preservative, so after reconstitution or dilution the solution is stored at 25°C and discarded within 24 h. Aseptic processing must comply with FDA 21 CFR Part 211 and EU GMP Annex 1, including media fill qualification and container closure integrity testing.

Scale-up from laboratory to production freeze dryers changes the heat transfer coefficient from the shelf to the vial because the edge-vial ratio and shelf fluid dynamics differ. Fully loaded production shelves require a conservative ramp to avoid excessively high product temperature; product temperature probes placed in center and edge vials are used to maintain a temperature difference not greater than 3°C across the shelf. The nitrogen backfill pressure at stoppering is controlled to approximately 0.8 atm partial pressure to reduce cake collapse and oxygen-mediated degradation. Lyophilized product batches that fail visual cake morphology are not automatically rejected if all release tests are met, but trend data on cake height and reconstitution time are tracked as process consistency indicators. The drug substance is released against USP <467> for residual solvents, USP <232> and USP <233> for elemental impurities, and X-ray powder diffraction to confirm the heptahydrate form before sterile formulation. The vial stopper is selected for low moisture vapor transmission, and the elastomeric closure must meet extractables limits under USP <381>. Stoppers are dried before loading to avoid moisture back-transfer to the freeze-dried cake.

Why Does Dry Granulation Limit Hydrate Shifting in Low-Dose Tablets?

In low-dose tablet development, direct compression of pemetrexed disodium heptahydrate is constrained by the cohesive nature of the fine API and by the hydrate stoichiometry. In tablet cores containing less than 50 mg of pemetrexed, direct compression with microcrystalline cellulose and mannitol often produces borderline blend uniformity because the API adheres to stainless steel and electrostatic charges develop at relative humidity below 20%. Dry granulation by roller compaction is therefore used to densify the pre-blend and fix the API onto carrier particles. The roller compactor is operated at a roll gap of 1–2 mm and a roll force of 20–50 kN, with the target ribbon density between 0.9 g/cm³ and 1.1 g/cm³. Ribbon density below this range produces friable granules, while higher density can create dissolution lag. The granulate is micro-milled and lubricated with sodium stearyl fumarate at 0.5–1.0% w/w. Magnesium stearate is generally avoided above 0.5% w/w because its hydrophobic film can delay wetting of an already dissolution-sensitive molecule. Compression on a rotary press is performed at 8–15 kN compression force, producing tablet hardness of 60–100 N and friability below 1.0% per USP <1216>. Blend uniformity is tested by USP <905> with acceptance value AV<15. Dissolution is measured by USP <711> Apparatus 2 at 50 rpm in 900 mL of media at 37°C, with media pH varied from 1.2 to 6.8 to characterize the pH-dependent release. The processing suite is held at 30–40% RH because the heptahydrate can exchange water with the environment above 60% RH. When punch filming occurs, the lubricant level is increased stepwise and the compression force is lowered to 6–10 kN. Published data for this specific configuration is limited, but the hydrate-shift risk is inferred from the crystalline water stoichiometry and standard pharmaceutical process behavior.

Within capsule-filling suites, the low bulk density of unmilled pemetrexed disodium heptahydrate and its tendency to segregate from free-flowing excipients drive the choice of an encapsulated granulate rather than a direct powder fill. A roller-compacted or dry-mixed blend is filled into size 0 or size 1 hard capsules on an intermittent-motion capsule filler. Commercial machines in this class run at 30,000–100,000 capsules/h, depending on dosing disc configuration and powder flow. The fill material is tested for bulk density, tapped density, and Carr’s index before filling; a Carr’s index below 25 is generally required to maintain weight variability within ±3%. The capsule shell material matters because the hydrate API can transfer water to gelatin, making the shell brittle at relative humidity below 40%. Hydroxypropyl methylcellulose capsules with a shell water content of 3–6% are often selected when clinical packaging will be stored in dry conditions. Fill weight and content uniformity are verified by USP <905>, and dissolution is performed by USP <711> Apparatus 1 at 100 rpm or Apparatus 2 at 50 rpm, depending on capsule density. Lubrication is held at 0.5–1.0% w/w; overlubrication above 1.5% w/w sodium stearyl fumarate causes delayed disintegration and a reduced dissolution rate at pH 6.8. The suite is maintained at 20–25°C and 35–45% RH, and in-process monitoring includes loss on drying of the filled blend not more than 2.0% w/w.

When Fluidized-Bed Granulation Is Used for Sachet-Stable Oral Granules

A top-spray fluidized-bed granulation process is used for sachet-stable oral granules when rapid dispersibility and dose flexibility are required. A binder solution of hypromellose or povidone in purified water is sprayed onto a fluidized pre-blend of pemetrexed disodium heptahydrate, mannitol, and crospovidone. Inlet air temperature is set between 50°C and 70°C, product temperature is held at 30–40°C, spray rate is adjusted to 10–30 g/min per kilogram of batch, and atomizing air pressure is maintained at 1.0–2.0 bar. The outlet air relative humidity is kept below 25% because the hydrate can release lattice water during wet massing, and overdrying can strip the water of crystallization and alter crystallinity. Granules are screened through an 850 µm sieve and filled into sachets on a vertical form-fill-seal machine using a PET/aluminum/LDPE laminate with a moisture vapor transmission rate below 0.5 g/m²/day. The finished granulate is tested for loss on drying between 1.0% and 2.0% w/w, pour density, and particle size distribution. Microbial attributes follow USP <61> and USP <62> for nonsterile oral preparations. Suspended granules should disperse in 100 mL of water at 25°C within 2 min without foaming or floating. Processing at larger scale requires attention to bag filter loading and electrostatic accumulation; if fines adhere to the filter surface, spray rate is reduced and inlet air humidity is increased slightly to dissipate charge. The fluidized-bed process is preferred over wet massing in a high-shear granulator because the heat and moisture exposure is shorter, reducing the probability of hydrate transition and acid-base degradation.

Cisplatin Co-Administration Sequence and Infusion Bag Contact Materials

When pemetrexed disodium heptahydrate is reconstituted for the approved malignant pleural mesothelioma regimen, the dose is 500 mg/m² infused intravenously over 10 min, and cisplatin 75 mg/m² is infused over 2 h beginning 30 min after the pemetrexed infusion completes. The cycle repeats every 21 days. The infusion solution is prepared only in 0.9% sodium chloride injection and is compatible with polyvinyl chloride and polyolefin infusion bags and sets. The diluted solution is stored at 25°C for no more than 24 h because the product contains no preservative. Patients receive dexamethasone 4 mg orally twice daily on the day before, the day of, and the day after pemetrexed administration to counteract skin toxicity. Folic acid 400–1000 µg orally once daily is started 7 days before the first dose and continued for 21 days after the last dose. Vitamin B12 1000 µg intramuscularly is given 1 week before the first dose and every 3 cycles thereafter. Renal function is evaluated before each cycle; the drug is contraindicated when creatinine clearance is below 45 mL/min, and dose adjustment is required based on nadir blood counts prior to the next cycle. The infusion line must not be mixed with other drugs or with calcium-containing solutions. The pemetrexed solution is visually inspected and must be free of particulate matter; any vial with discoloration or particles is discarded under USP <790> criteria. The dose withdrawal from a 500 mg vial is adjusted to the calculated body surface area and is not based on body weight alone; an independent two-person check of the calculated dose and pump settings is typical in oncology pharmacy practice.

Because pemetrexed is classified as a hazardous drug, pharmacy compounding is performed in a biological safety cabinet with closed-system transfer devices used to limit aerosol release. The compounded sterile preparation is assigned a beyond-use date not exceeding 24 h at room temperature under USP <797> and USP <800>. The final infusion bag is labeled with the patient-specific dose, the date and time of preparation, and the administration route. Infusion pumps used for the 10 min pemetrexed infusion are programmed with a volume limit and a line-flush sequence using 0.9% sodium chloride injection to ensure the full dose is delivered without mixing with subsequent chemotherapy infusions. The package insert defines this two-drug sequence as a single-day regimen; any deviation in timing is recorded as a dose administration variance in the oncology pharmacy quality system.