For the manufacture of lyophilized cytarabine for injection, the API is dissolved in Water for Injection at a concentration matched to the final vial presentation, commonly 10 mg/mL or 20 mg/mL before freeze-drying, with 0.9% w/v sodium chloride added as an isotonicity-adjusting agent where required by the approved dossier. The solution pH is adjusted with dilute hydrochloric acid or sodium hydroxide to the range established for the finished product, and the bulk solution is passed through a 0.45 µm prefilter followed by a 0.22 µm sterilizing-grade polyethersulfone or polyvinylidene fluoride membrane. Filtration is executed under positive-pressure nitrogen in a Grade A laminar airflow zone adjacent to a Grade B cleanroom according to EudraLex Volume 4 Annex 1 and FDA 21 CFR 211.42. The filtered solution is filled into Type I borosilicate glass vials meeting USP <660> surface hydrolytic resistance requirements, partially stoppered with butyl rubber closures, and transferred to a freeze-dryer. The lyophilization cycle includes a freezing ramp to −40 °C or below, an annealing step between −20 °C and −10 °C to reduce inter-vial heterogeneity in pore structure, primary drying at shelf temperatures between −25 °C and −15 °C with chamber pressure in the 50 mTorr to 150 mTorr range, and secondary drying at shelf temperatures from 20 °C to 30 °C. Endpoint of secondary drying is determined by comparative pressure measurement or a predefined residual moisture specification, typically verified by coulometric Karl Fischer titration using USP <921>. The resulting lyophilized cake is a white to off-white porous solid that must reconstitute rapidly in Water for Injection or sterile saline to a clear solution meeting USP <790> visible particulate requirements and USP <788> subvisible particulate limits. In the finished vial, the assay of cytarabine and the profile of related substances, including uracil arabinoside, are controlled by the high-performance liquid chromatographic procedures described in the USP Cytarabine Injection monograph. Sterility is confirmed by membrane filtration under USP <71>, and bacterial endotoxins are controlled by the Limulus amebocyte lysate method under USP <85> with a limit calculated from the maximum bolus dose and route of administration. Process-scale batch records repeatedly identify three variables as critical to cake appearance and reconstitution time: the temperature at which the bulk solution is held before filtration, the subcooling rate during the freezing ramp, and the extent of annealing. Holding the bulk solution above 25 °C for extended periods increases the rate of hydrolytic degradation to uracil arabinoside and must be challenged with a manufacturing-scale hold-time study. Freezing too slowly produces large ice crystals that leave a brittle cake with high surface area, while freezing too quickly without annealing can trap residual water and extend reconstitution time. Equipment qualification of the freeze-dryer includes shelf temperature mapping across all shelves, condenser capacity verification, and filter integrity testing by bubble point or diffusive flow per ASTM F838-20 for the 0.22 µm membrane.

What Limits Ready-to-Use Infusion Stability below 2°C?

Ready-to-use cytarabine infusion containers are prepared by aseptic dilution of the reconstituted solution or concentrated injection into 0.9% w/v sodium chloride injection or 5% w/v dextrose injection in polyolefin or ethylene vinyl acetate containers. Polyvinyl chloride containers are generally not used for extended holding of cytotoxic infusions unless prospectively validated because of potential plasticizer leaching and sorption. The pH of the admixture is maintained in the neutral window in which the arabinosyl glycosidic bond of cytarabine is most resistant to acid- and base-catalyzed hydrolysis. At lower pH values, acid-catalyzed conversion to uracil arabinoside accelerates; at higher pH values, base-catalyzed degradation becomes significant and can produce additional polar degradants resolved by stability-indicating HPLC. Refrigerated storage at 2 °C to 8 °C is used to slow molecular mobility and hydrolysis, but the working practice must account for the risk of precipitation in concentrated admixtures and for condensation-driven dilution at the container closure interface. Visual inspection before administration uses USP <790> criteria; any turbidity, crystal formation, or particulate matter requires rejection because subvisible particles are not reliably detected by visual inspection alone. The maximum validated holding period is not an intrinsic property of the API and must be derived for each final concentration, diluent, container polymer, and temperature condition. Published compatibility data for this specific configuration are limited beyond the commonly used 0.9% sodium chloride and 5% dextrose vehicles, and the available labels should be treated as diluent-specific rather than universally transferable. In process terms, the limiting factors are the rate of formation of uracil arabinoside at the pH of the admixture, the oxygen permeability of the container, and the degree of light exposure during storage. A stability program for a ready-to-use infusion line therefore includes pH at time zero and at the final hold time using USP <791>, assay and related compounds by the HPLC method of the USP Cytarabine Injection monograph, subvisible particulate matter by USP <788>, and container closure integrity by a dye intrusion or vacuum decay method validated under USP <1207>. Light exposure studies are relevant because cytarabine in solution can undergo photodegradation; however, the published kinetic constants and wavelength dependence for photodegradation are limited, and product-specific light protection is normally established through forced degradation studies according to ICH Q1B. The formulation and packaging must therefore be treated as a combined system rather than as an API attribute alone.

At 10 mg/mL, Intrathecal Liposomal Suspension Must Be Manufactured Aseptically

A distinct downstream application is the multivesicular liposomal depot used for intrathecal administration, in which cytarabine is encapsulated within a lipid-particle suspension to prolong cerebrospinal fluid residence time after lumbar puncture. The terminal product is typically a 10 mg/mL cytarabine suspension in a 5 mL vial providing 50 mg per vial, although regional approval dossiers may differ in volume. The lipid matrix of this depot product includes diarachidonoylphosphatidylcholine, dipalmitoylphosphatidylglycerol, cholesterol, and a triacylglycerol such as triolein; the precise lipid molar ratios and the ratio of aqueous cytarabine solution to lipid phase are defined by the approved formulation and are not interchangeable between manufacturers. The manufacturing process uses a water-in-oil-in-water double emulsion or an equivalent solvent evaporation protocol in which cytarabine in the internal aqueous phase is entrapped within multivesicular liposomes. Particle size distribution is subsequently controlled by extrusion or high-pressure homogenization and measured by laser diffraction according to ISO 13320:2020 or by dynamic light scattering under ISO 22412:2017 for the submicron fraction. Because the lipid vesicles cannot be terminally sterilized by moist heat without disrupting the lamellar structure and releasing free cytarabine, the entire manufacturing train operates as an aseptic process under EudraLex Volume 4 Annex 1 and FDA 21 CFR 211.42. The cytarabine solution is filtered through a 0.22 µm membrane before the lipid-loading step, whereas the lipid phase is separately sterile-filtered through a validated sterilizing-grade compatible membrane; sterile components are then assembled in a closed system. Release testing includes assay of free and encapsulated cytarabine after disrupting the vesicles, related substances including uracil arabinoside, pH under USP <791>, osmolality by freezing point depression, particle size, sterility under USP <71>, and bacterial endotoxins under USP <85>. The clinical route imposes additional limits on foreign particulate matter because the injection enters the cerebrospinal fluid; subvisible particle counts are therefore measured by USP <788>, and the visible appearance is assessed against USP <790>. Process-scale failure modes observed in aseptic manufacturing of this dosage form include vesicle disruption during transfer through peristaltic pumps, aggregation during storage at 2 °C to 8 °C, and elevated free cytarabine levels when the double-emulsion mixing time exceeds the validated range. Filtration of the finished suspension is not possible because it would remove the active liposomal particles; therefore, sterility assurance depends entirely on component preparation, environmental control, and media fills. The free-to-encapsulated ratio is release-critical and is determined by ultrafiltration or ultracentrifugation followed by HPLC assay. Published data for this specific formulation are product-specific, and generic substitution of lipid excipients without a new bioavailability or clinical equivalence study is not valid.

Assessing Oral Tablet and Capsule Feasibility Against Intestinal Cytidine Deaminase

Cytarabine is a substrate for cytidine deaminase in the intestinal mucosa and liver, leading to rapid first-pass conversion to uracil arabinoside and low systemic exposure after oral administration. The oral route is not the basis for currently approved cytarabine products in most jurisdictions, and development activities in this segment are largely confined to investigator-initiated trials, bioavailability studies, or combination regimens in which intestinal deaminase inhibitors such as tetrahydrouridine are co-administered. If a tablet or capsule is formulated, the API is handled as a potent cytotoxic compound in an isolator or closed containment system meeting 21 CFR 211.42 and industrial hygiene requirements. Direct compression and dry granulation are preferred over wet granulation because an aqueous granulation step can dissolve a portion of cytarabine and create a sticky mass that is difficult to dry without generating high levels of uracil arabinoside. A typical development blend may contain a diluent such as mannitol or microcrystalline cellulose, a disintegrant such as croscarmellose sodium, and a lubricant such as magnesium stearate, with the API ratio determined by the intended dose and the need to maintain acceptable content uniformity under USP <905>. Blending is often performed in a bin blender at low shear, and the powder is compressed or encapsulated using periodic checks of weight, hardness, and disintegration time according to USP <701>. Dissolution method development for oral cytarabine is not defined by a harmonized pharmacopoeial monograph because no official oral tablet monograph exists; laboratory studies must use a validated in-house method under USP <711> with pH and apparatus selected from biorelevant media and sink-condition data. The central technical constraint is not formulation processability but the pharmacokinetic limitation imposed by first-pass deaminase activity. Published human oral bioavailability data for free cytarabine are limited and indicate low systemic exposure; therefore, oral feasibility cannot be assumed from in vitro dissolution alone. Any oral product would require a substantial clinical pharmacology program, and specific bioavailability comparisons must be generated under 21 CFR 320.21 or the corresponding national requirement. Section 3.2.P of a Common Technical Document for such a product would be expected to justify the salt form, particle size distribution, polymorphic identity by X-ray powder diffraction, and polymorph stability under ICH storage conditions.

When Dry Granulation Is Selected for Capsule Fill

In capsules and powder-filled sachets intended for oral or nasogastric administration, dry granulation is used to increase bulk density, reduce dust generation, and improve flow of high-potency cytarabine blends while avoiding the hydrolytic risk of water-based granulation. The API is first passed through a 0.5 mm or 1.0 mm screen to deagglomerate; it is then blended with a portion of the filler and disintegrant, lightly lubricated, and compacted on a roller compactor with controlled roll speed, roll gap, and hydraulic pressure. Ribbon density and thickness are monitored because overcompaction reduces tablet disintegration and undercompaction generates a high proportion of fines that segregate in the hopper. The compacted ribbons are milled through a screen of typically 1.0 mm to 1.6 mm, and the resulting granules are sized and blended with extragranular disintegrant and magnesium stearate in a V-blender or bin blender. The final blend is sampled for bulk and tapped density, flow index, moisture by USP <921>, and content uniformity per USP <905> prior to encapsulation. Capsule fill is performed on an automatic capsule machine using tamper-evident hard gelatin or hypromellose capsules depending on the target patient population and dietary restrictions; gelatin capsules are not preferred if the formulation is moisture-sensitive or if the product is to be stored in high humidity beyond the barrier properties of the primary container. Finished capsules are tested for weight variation, disintegration under USP <701>, and assay by HPLC. Because no harmonized official monograph governs oral cytarabine capsules, the acceptance criteria for related substances must be based on safety qualification of degradants under ICH Q3A(R2) and ICH M7(R2) if the degradants are mutagenic impurities. The granular intermediate itself is also a downstream article when supplied to clinical manufacturing contractors; in this case, the granule is characterized by particle size distribution, moisture, tapped density, and endotoxins when intended for oral or enteral use. A significant process boundary is the lubrication time: magnesium stearate above 1.0% w/w or over-lubrication in a high-shear tumbler can coat the granules and delay disintegration, but too little lubrication can cause powder sticking to the encapsulation tooling. This trade-off is resolved by a design-of-experiments approach in which disintegration time and the dissolution profile are the primary response variables. The decision between direct compression and dry granulation is based on the API particle size and flow properties; direct compression is acceptable only if the blend meets mass uniformity limits in the target capsule size.

A separate sterile injectable application is the co-encapsulation of cytarabine and daunorubicin in a liposomal formulation for intravenous infusion. In this product class, the fixed molar ratio of 5:1 cytarabine to daunorubicin is maintained because the two active molecules interact with the liposomal bilayer and internal aqueous compartments in different ways; daunorubicin is loaded via an ion gradient while cytarabine is encapsulated in the aqueous core. The manufacturing process is aseptic and does not include terminal filtration of the final liposomal suspension because the particle size, typically in the nanometer to micrometer range, is part of the product’s disposition profile. The intravenous infusion is prepared by aseptic dilution of the liposomal dispersion into 0.9% w/v sodium chloride or 5% w/v dextrose injection according to the approved label; the final admixture is administered through an infusion set without an in-line filter smaller than 15 µm, because filters that are too fine can rupture or retain the liposomes. Particle size distribution is controlled by laser diffraction under ISO 13320:2020, zeta potential by electrophoretic light scattering under ISO 13099-1:2012, and the encapsulated fraction by ultrafiltration or ultracentrifugation. Release testing includes assay of both actives, related substances, free drug levels, sterility under USP <71>, bacterial endotoxins under USP <85>, pH under USP <791>, and particulate matter under USP <788>. The terminal product is not interchangeable with conventional cytarabine solution because the pharmacokinetic and organ-disposition profiles are dominated by the liposomal carrier. Batch-to-batch variance in the extrusion or homogenization step has been identified as a critical source of variation in mean particle diameter and the ratio of free-to-liposomal drug; therefore, in-process particle size and free daunorubicin are measured after each homogenization pass. The infusion container and administration set must also be qualified for compatibility with the liposomal dispersion, and the extended holding time after dilution is more restrictive than for a simple aqueous solution because lipid membrane integrity and pH drift affect drug retention. Published data for this specific configuration are confined to the approved product dossier and peer-reviewed clinical studies; direct extrapolation to other cytarabine-to-daunorubicin ratios or lipid compositions requires new bioequivalence, pharmacokinetic, and stability data.