Pharmaceuticals and fermentation: from molecule to medicine: Formulation: turning powder into a dose

3.1 Dosage forms and the job of excipients

The dose of most drugs is small: tens or hundreds of milligrams. That is too little to handle, so the API is diluted with excipients. Kent explains that oral drugs, the most common and preferred route, are usually tablets or capsules in which a measured dose is mixed evenly with an inert filler chosen to control how fast the drug dissolves. Some medicines, such as children's antibiotics, are solutions, as are injections.

Kent's formulation table, written for pesticide tablets but describing the same physical product, lists the usual tablet ingredients: the active ingredient, a binder that holds particles together, a lubricant that stops the mixture sticking to the machine, and a disintegrant that makes the tablet break up in water. Pollak adds that excipients confer bulk, stability, colour and taste. These are the materials in the "Pharmaceutical excipients" tile, and many come from the "Starch and cellulose derivatives" tile.

The choice of dosage form starts with solubility. A liquid or injection needs the drug dissolved, preferably in water. Kent notes that molecules with a basic amine or an acidic group can usually be converted into water-soluble salts, while neutral molecules are harder. When a drug dissolves or crosses membranes poorly, chemists may make a prodrug: a derivative carrying a detachable group that improves solubility or absorption and is cleaved in the body to release the active drug.

Penicillin shows how one API becomes several products through its salt. Dryden lists sodium or potassium salts for fast-acting injections, potassium, calcium or aluminium salts for oral tablets, and procaine salts for slow-release oil-based injections. Biologics, Kent adds, are large molecules that so far must be injected because they cannot survive digestion.

• Kent (2012), Ch 11 Chemistry in the Pharmaceutical Industry · pdf pp. 405, 422
• Kent (2012), Ch 17 Agrochemicals (formulation types) · pdf pp. 671, 673 (Table 17.8)
• Pollak (2011), Fine Chemicals · pdf p. 29
• Dryden (1997), IVH Pharmaceutical Industry · pdf p. 574
• Shreve (1984), Ch 40 Pharmaceutical Industry · pdf p. 814

3.2 Tablet manufacture: blend, granulate, compress, coat

A tablet is, in Kent's definition, a pre-formed solid of uniform shape and size, usually round with flat or convex faces. Making millions of identical ones is a problem in powder handling, and the sources describe its stages rather than machine settings.

First the API is made physically suitable. Kent stresses that particle size, density and flowability are critical, which is why milling is part of API finishing. The API and excipients are weighed and blended so that every tablet gets the same dose; with a few milligrams of drug in a much larger tablet, uneven mixing would mean some tablets with too much and some with too little.

Fine powders flow badly and separate, so the blend is usually granulated: wetted with a binder solution and worked into denser, free-flowing granules. Kent's table of powder technologies, written for enzyme products, describes the equipment classes: high-shear drum granulators such as Lödige mixers, which give tough granules at high capacity; extruders and spheronisers; and fluid beds, in which powder is suspended in warm air while liquid is sprayed on. The granules are dried and blended with lubricant.

The granules are then formed into tablets; the sources give no press settings. Many tablets are finally coated, to mask taste, protect the drug from moisture or control release. Kent describes fluid-bed coaters, which spray a coating onto particles suspended in air in a single contained vessel, and notes that coatings can delay release or protect an active through harsh conditions.

Each stage has quality checks, and Kent's list of cGMP topics includes packaging and labelling as well as production. Capsules hold the blend or granules in a shell instead; Shreve lists pharmaceutical capsules among the uses of gelatin.

• Kent (2012), Ch 17 Agrochemicals (formulation types) · pdf pp. 671, 673
• Kent (2012), Ch 30 Industrial Biotechnology: Discovery to Delivery · pdf p. 1151
• Kent (2012), Ch 11 Chemistry in the Pharmaceutical Industry · pdf pp. 405, 425-426
• Pollak (2011), Fine Chemicals · pdf pp. 186-187, 239

3.3 Sterile products: injections and the war on contamination

An injection bypasses the body's defences, so it must be sterile, meaning free of live organisms, and pyrogen-free. Pyrogens are fever-causing substances, chiefly endotoxins: fragments of bacterial cell walls that survive even after the bacteria are killed. Sterile manufacture is therefore about two separate problems: killing or removing organisms, and keeping out their debris.

The sources show how this is done for penicillin, which cannot be heat-sterilised because it breaks down. Shreve's process ends by passing the purified solution through Seitz filters, pads fine enough to hold back bacteria, then precipitating the procaine or potassium salt and drying it under sterile conditions. Dryden describes reverting the penicillin into pyrogen-free distilled water, separating it in a supercentrifuge, forcing it through a biological filter to remove the last bacteria and pyrogens, and then concentrating it by freeze drying or vacuum spray drying. Freeze drying removes water by sublimation from the frozen solution, so the drug is never heated.

The same logic governs the fermentation plants that make such products. Dryden's design rules for sterile operation include no direct connection between sterile and unsterile parts, all-welded smooth interiors, diaphragm valves, sterile areas held at positive pressure so unsterile air cannot leak in, separate sterilisation of each part, and steam on vulnerable zones.

Biologics push the limits furthest. Kent's chapter on therapeutic proteins gives the impurity levels generally accepted in a final protein product: no cell debris, host-cell protein below 0.1 to 10 mg/L, DNA at 10 ng per dose, and endotoxin at 0.25 endotoxin units per mL. Meeting these is a large part of why biologics cost so much more per kilogram than small molecules.

• Shreve (1984), Ch 40 Pharmaceutical Industry · pdf pp. 818, 840
• Dryden (1997), IVH Pharmaceutical Industry · pdf pp. 339, 576
• Kent (2012), Ch 32 Industrial Production of Therapeutic Proteins · pdf pp. 1243-1245

See it in the table

Check your understanding

Why is penicillin for injection sterilised by filtration rather than by heating?

In a tablet, what is the job of the disintegrant?

What led to the 1938 US law requiring proof of safety before a drug is marketed?