Direct compression of 7-ethyl-10-hydroxycamptothecin (SN-38) as neat API is not pursued on production-scale rotary presses because the crystalline lactone form shows aqueous solubility below 50 µg/mL at pH 7.4 and the equilibrium shifts to the ring-opened carboxylate at pH 6.0 and above. Spray-dried amorphous solid dispersion is the preferred upstream step. A binary dispersion with hypromellose acetate succinate (HPMCAS M grade) is prepared at a drug:polymer ratio between 1:3 and 1:9 w/w using a solvent system of methanol and dichloromethane or acetone and water depending on residual solvent compliance under ICH Q3C. In a pilot-scale spray dryer with a two-fluid nozzle, inlet temperature is maintained at 90–110 °C and outlet temperature at 40–50 °C to minimize SN-38 thermal degradation while achieving moisture content below 2.0% w/w. The resulting dispersion is passed through a 0.5 mm cone mill to break agglomerates before blending.

The compression blend combines the milled dispersion equivalent to 5.0 mg SN-38 per unit, microcrystalline cellulose q.s. to 100 mg core weight, croscarmellose sodium 3.0% w/w, colloidal silicon dioxide 0.5% w/w, and magnesium stearate 1.0% w/w. On an instrumented rotary press, compression force is held at 8–12 kN, producing tablet hardness of 5–8 kp and friability not more than 1.0% by USP <1216>. Content uniformity testing follows USP <905> with an acceptance value of L1 ≤ 15.0. Dissolution is evaluated with USP <711> Apparatus II at 75 rpm in 900 mL of 0.1 N HCl at pH 1.2 maintained at 37 °C; the acidic medium preserves the lactone form during the 60 min run. Disintegration testing is conducted per USP <701> with a limit of not more than 15 min in water at 37 °C. Processing humidity above 60% RH is not recommended because HPMCAS dispersions take up moisture and undergo amorphous recrystallization that depresses dissolution area under the curve.

Capsule Fill Composition and Precipitation Boundaries

Liquid-filled hardshell capsules become relevant when the API is pre-solubilized in a lipid-surfactant matrix and filled into size 1 or size 0 capsules under controlled nitrogen overlay. A representative screening formulation contains 10.0 mg SN-38 equivalent, 300 mg medium-chain triglycerides, 200 mg PEG-40 hydrogenated castor oil, and 100 mg diethylene glycol monoethyl ether per size 1 capsule. The fill mass is limited to 700 mg to avoid sealing defects on an automatic capsule banding line. Mixing is performed in a jacketed vessel at 40–45 °C under low-shear impeller agitation until visual clarity is obtained, then passed through a 0.45 µm filter before filling. The fill nozzle temperature is maintained at 35 °C to avoid lipid solidification at the nozzle tip; cooling below 30 °C in the fill mass leads to viscosity rise and capsule weight variation beyond ±5%. In-process pH of the fill is not adjusted above 4.0; a higher pH accelerates SN-38 carboxylate formation even in nonaqueous surfactant systems. Capsule shells are sealed with a hydroalcoholic banding solution and dried at 25 °C for 2 h. The terminal capsule is tested for dissolution by USP <711> using 900 mL of 0.01 N HCl with 0.5% w/v polysorbate 80 to maintain sink conditions. Residual solvent analysis follows ICH Q3C; diethylene glycol monoethyl ether is treated as a Class 2 residual solvent and monitored by headspace gas chromatography.

Fill mass above 700 mg or storage below 15 °C produces gelatin cross-linking or lipid precipitation; both failure modes delay disintegration beyond 30 min and require batch rejection under USP <701>.

For granule-based oral reconstitution, the formulation objective is delivery of SN-38 in an acidified vehicle immediately after suspension because the lactone pharmacophore is stable below pH 4.0 but rapidly hydrolyzes at neutral pH. A high-shear granulation process loads 2.5 mg SN-38 per sachet onto mannitol and microcrystalline cellulose carriers, with citric acid monohydrate at 100 mg per sachet as the lactone-stabilizing agent. The granulating fluid is purified water or an aqueous ethanol solution; the wet mass is granulated at impeller speed 300 rpm and chopper speed 1500 rpm for 3–5 min, then dried in a fluid bed at inlet air temperature not exceeding 60 °C to prevent thermal stress on SN-38. Final granule loss on drying is kept between 1.0% and 2.5% w/w, because lower moisture causes brittle granules and higher moisture accelerates hydrolysis during storage. The sachet content is dispersed in 10 mL water before administration; the resulting suspension pH is below 4.0 and should be consumed within 30 min. Microbial quality is tested per USP <61> and USP <62>, with total aerobic microbial count not more than 10² CFU/g and absence of Escherichia coli in 1 g. Assay and uniformity are controlled by HPLC using a C18 column and fluorescence detection at 380 nm excitation and 550 nm emission.

What Freeze-Drying Cycle Boundaries Govern SN-38 Injectable Powder?

Lyophilized injectable powder manufacturing for SN-38 must resolve a narrow pH window that simultaneously keeps the API in the active lactone form and avoids acid-catalyzed degradation of the lyoprotectant matrix. The API is dissolved in a vehicle containing 20% w/v hydroxypropyl-beta-cyclodextrin and 50 mM lactate buffer at pH 3.5, then filtered through a 0.22 µm polyvinylidene fluoride membrane. Filling is performed aseptically into 10 mL Type I borosilicate vials at a nominal fill volume of 5.0 mL. Freeze-dry microscopy is used to identify the collapse temperature; the primary drying shelf temperature is set 5–10 °C below that value, commonly starting at -25 °C with a ramp rate of 0.5 °C/min. Chamber pressure is maintained at 100 mTorr during primary drying and reduced to 50 mTorr during secondary drying. Secondary drying shelf temperature is raised to 35 °C over 6 h and held until product temperature reaches 30 °C for 4 h. The resulting cake is tested for residual moisture by USP <921> Karl Fischer titration with a limit of not more than 3.0% w/w. Reconstitution time is measured with 5.0 mL sterile water for injection; a fully dissolved solution should be achieved within 2 min without visible particles. Subvisible particulate matter is evaluated by USP <788> Method 1; for containers ≤ 100 mL, the limits are 6000 particles/container ≥10 µm and 600 particles/container ≥25 µm. Bacterial endotoxin is controlled per USP <85> with the limit calculated from the maximum administered dose using K = 5 EU/kg; for a 5 mg dose, the calculated limit is 1.0 EU/mg if the maximum dose is 1 mg/kg. Sterility testing follows USP <71>. Terminal steam sterilization is not applied because SN-38 degrades on prolonged exposure to moist heat; aseptic filtration is mandatory. The pH after reconstitution must remain between 3.0 and 4.0; at pH above 6.0, the lactone fraction falls below 50% within 1 h at 37 °C in neutral buffer.

Residual moisture above 3.0% w/w accelerates both hydrolytic ring opening and cake collapse during storage; vials with chamber pressure excursions above 150 mTorr during primary drying produce collapsed cakes and are rejected by visual inspection. The lyophilized product should be stored at 2–8 °C; reconstituted solution should be used within 6 h when held at 25 °C and protected from light.

Liposomal Remote Loading with Ammonium Sulfate Gradients

Liposomal encapsulation of SN-38 relies on a pH or ammonium sulfate gradient that converts neutral SN-38 into a poorly permeable precipitate inside the aqueous core. The lipid phase is typically composed of hydrogenated soy phosphatidylcholine, cholesterol, and DSPE-PEG₂₀₀₀ with cholesterol content held at 30–40 mol% to reduce bilayer permeability. The lipids are dissolved in chloroform, dried to a thin film under reduced pressure at 40 °C, and hydrated with 250 mM ammonium sulfate at 60 °C. The resulting multilamellar vesicles are homogenized with a high-pressure homogenizer at 10,000–15,000 psi for 5–8 discrete passes to obtain a Z-average diameter of 80–120 nm by dynamic light scattering. Extrusion through 0.1 µm polycarbonate membranes may be used to tighten the size distribution to a polydispersity index below 0.1. The external ammonium sulfate is removed by diafiltration or size-exclusion chromatography against sucrose-histidine buffer. SN-38 is then added at a drug:lipid molar ratio between 0.1 and 0.3 and loaded at 60 °C for 30–60 min. Encapsulation efficiency is measured by reversed-phase HPLC after disruption with methanol; values above 90% are typical when the internal sulfate concentration is maintained. Terminal sterilization is not applied; the product is aseptically filtered through a 0.22 µm membrane into depyrogenated vials. In vitro release is tested in phosphate-buffered saline at pH 7.4 and 37 °C with dialysis; release should not exceed 20% in 2 h to confirm bilayer integrity. Subvisible particles are controlled by USP <788>; liposome size stability is monitored by dynamic light scattering after 24 h at 2–8 °C. Heating above 65 °C during remote loading is avoided because it hydrolyzes phosphatidylcholine to lysolipid and increases the risk of complement activation-related pseudoallergy.

Published data for specific liposomal SN-38 configurations are limited; therefore, the above operational ranges are treated as platform parameters that require formulation-specific verification under ICH Q2(R1) for the particle size and encapsulation efficiency methods. The terminal product is a liposomal suspension for injection, not a solution; the label must include the measured mean particle size, polydispersity index, and free SN-38 limit to avoid infusion variability.

When SN-38 Serves as the Payload in Antibody-Drug Conjugates

In antibody-drug conjugate manufacturing, SN-38 is conjugated through a cleavable linker because free SN-38 is highly protein-bound and has a short plasma half-life, while the conjugate permits tumor-directed delivery and prolonged exposure. The approved ADC sacituzumab govitecan uses SN-38 attached via the CL2A linker to a humanized anti-Trop-2 monoclonal antibody at a drug-to-antibody ratio of approximately 7.6. Conjugation begins with partial reduction of interchain disulfide bonds using tris(2-carboxyethyl)phosphine at 25 °C for 2 h, followed by pH adjustment to 6.5–7.0 and addition of the maleimide-terminated CL2A-SN-38 intermediate. Unreacted species are removed by tangential flow filtration using a 30 kDa regenerated cellulose membrane; free SN-38 content is monitored by reversed-phase HPLC and controlled below 1.0% of total SN-38. The conjugate is formulated in a histidine-lactate buffer containing trehalose dihydrate and polysorbate 20, filled aseptically into glass vials, and lyophilized. Release testing includes sterility per USP <71>, bacterial endotoxin per USP <85>, subvisible particulate matter per USP <788>, drug-to-antibody ratio by hydrophobic interaction chromatography, and free payload by HPLC. Residual organic solvent from the linker conjugation step is monitored under ICH Q3C; the final product is tested for elemental impurities under ICH Q3D. The terminal product is reconstituted with sterile water for injection and diluted into infusion bags; in-use stability is limited to 4 h at room temperature because free carboxylate SN-38 can form above pH 6.0 in infusion solutions.

The conjugation reaction is quenched with N-acetylcysteine to cap unreacted maleimide groups; the quench time must be controlled within 15 min because excess thiol exposure reduces the drug-to-antibody ratio below 7.0 and produces an inconsistent potency profile. Processing temperature above 30 °C during conjugation increases aggregate levels beyond 2.0% by size-exclusion chromatography, triggering re-purification.

Multi-arm PEG-SN-38 conjugates are produced by covalent conjugation of SN-38-20-O-glycinate to a 40 kDa four-arm polyethylene glycol activated with N-hydroxysuccinimide esters. Published clinical formulations of this type have an SN-38 loading of approximately 3.7% w/w, which corresponds to roughly 4 SN-38 molecules per four-arm PEG molecule. The conjugation reaction is carried out in anhydrous dimethylformamide or dimethylacetamide at 25 °C for 2–4 h, quenched with glycine, and purified by diafiltration against sterile phosphate-buffered saline at pH 5.5 to keep the lactone ring closed. The product is freeze-dried with mannitol as a bulking agent. The resulting cake is reconstituted in water for injection to a concentration of 5 mg/mL SN-38 equivalent and administered by intravenous infusion over 60–90 min. Critical release parameters include free SN-38 content below 1.0%, residual dimethylformamide below 880 ppm per ICH Q3C, endotoxin per USP <85>, and sterility per USP <71>. The formulation is not heat-sterilized; sterile filtration through a 0.22 µm membrane is used after reconstitution because the conjugate solution viscosity remains below 10 cP at 25 °C. In-use stability at room temperature is limited by gradual lactone hydrolysis; the formulation pH is maintained at 5.5 because neutral infusion diluents accelerate carboxylate formation and reduce the active lactone fraction.

Particle formation in the reconstituted solution is monitored by visual inspection and light obscuration per USP <788>; aggregate formation above 1000 particles/mL ≥10 µm requires discarding the infusion bag.

Thermal and pH thresholds shift sharply during polymeric micelle processing.

Polymeric micelle formulations of SN-38 use amphiphilic block copolymers such as polyethylene glycol–poly(glutamic acid) with SN-38 covalently bound to the carboxylate side chains. The block copolymer is synthesized by amidation of the poly(glutamic acid) block with an SN-38-glycine derivative, with a target SN-38 mass content of 20–30% w/w in the polymer. Self-assembly is induced by dissolving the conjugate in dimethyl sulfoxide and dialyzing against 10 mM citrate buffer at pH 5.0, producing micelles with a Z-average diameter near 20 nm. Sterile filtration is performed through a 0.22 µm polyethersulfone membrane; filtration pressure is kept below 10 psi to avoid shear-induced aggregation. The filtered micelle solution is freeze-dried with sucrose at a sucrose:polymer mass ratio of 2:1 to protect the core during lyophilization. Critical process limits include temperature not exceeding 40 °C during solvent evaporation and pH not rising above 6.0 during dialysis. The terminal product is tested for particle size by dynamic light scattering with polydispersity index below 0.1, free SN-38 content below 1.0%, sterility per USP <71>, and endotoxin per USP <85>. Dilution stability is confirmed in 5% w/v dextrose at 25 °C for 6 h; if the Z-average diameter increases by more than 10 nm or the polydispersity index exceeds 0.2, the infusion is discarded. pH above 6.0 during storage accelerates both SN-38 lactone ring opening and micelle swelling; storage at 2–8 °C in the lyophilized state is required to maintain physical stability for the assigned shelf life.