# Pharmaceuticals and fermentation: from molecule to medicine: Enzymes, biologics and the economics

Source page: https://www.materialinsights.org/learn/chemicals/pharmaceuticals-and-fermentation/p5/

[Understanding Chemicals](https://www.materialinsights.org/learn/chemicals/) / [Module 10: Pharmaceuticals and fermentation: from molecule to medicine](https://www.materialinsights.org/learn/chemicals/pharmaceuticals-and-fermentation/) Part 5 of 5

## Enzymes, biologics and the economics

Enzymes let manufacturers do single chemical steps with biology, often replacing harsh reagents. Biologics take biology all the way, using living cells to make protein drugs. The last section pulls the module together: where cost sits in a medicine, why batch still rules, what regulation costs, and how India and China came to supply much of the world's generic APIs.

### 5.1 Enzymes and biocatalysis: one clean step at a time

Enzymes are protein catalysts made by living cells. Kent stresses their attractions: they are highly specific, so they make few by-products; they work at ambient temperature and near-neutral pH; they can be made by fermentation from sugar; and they can be engineered. The world market for "Industrial enzymes" exceeded $3.5 billion in 2010, growing 5 to 8% a year. Over half of sales are proteases for detergents, dairy and leather; carbohydrases such as amylases are nearly 40%. Novozymes and Genencor (now DuPont) lead.

An enzyme is expensive and fragile in solution, so the key industrial trick is immobilisation: fixing it to a solid support so it can be reused. Shreve lists adsorption, covalent bonding and entrapment in a polymer. Kent's glucose isomerase example shows the payoff. Over 500 t of immobilised enzyme are made each year, mostly by binding whole cells into a clay-based granule. In a packed column above 60°C its half-life exceeds a year, and one kilogram makes 20,000 kg of high-fructose syrup.

Biocatalysis now does steps in drug manufacture. Penicillin G acylase, immobilised on a resin, snips the side chain off penicillin G to make 6-aminopenicillanic acid (6-APA), the nucleus from which ampicillin and amoxicillin are built. Kent gives productivities over 2,000 kg of product per kg of catalyst and reuse for over 1,000 cycles; the reaction is held at pH 7.5 to 8 against the phenylacetic acid it releases. Enzymes similarly make 7-aminocephalosporanic acid (7-ACA) from cephalosporin C. Worldwide output of semisynthetic penicillins is around 10,000 t a year. These are the "Antibiotic intermediates (6-APA, 7-ACA)" tile.

Data Glucose isomerase: batch versus continuous (normalised, Genencor)

| Parameter | Batch, soluble enzyme | Batch, immobilised | Continuous, immobilised |
|---|---|---|---|
| Reactor volume (m3) | 1,100 | 1,100 | 15 |
| Temperature (°C) | 65 | 65 | 60 |
| Relative reactor size | 75 | 75 | 1 |
| Relative enzyme consumption | 90 | 6 | 1 |
| Relative residence time | 40 | 40 | 1 |
| Relative energy use | 5 | 5 | 1 |
| Relative conversion cost | 100 | 6 | 1 |

Immobilisation plus continuous packed-bed operation cut conversion cost a hundredfold for this process.
 Source: Kent (2012), Ch 31 Industrial Enzymes and Biocatalysis, pdf p. 1213 (Table 31.11)  Case

#### Acrylamide and 7-ACA: the green dividend

Kent's flagship case is acrylamide from "Acrylonitrile". The enzyme nitrile hydratase runs at 10°C instead of 70°C for the copper-catalysed route, converts 100% of feed instead of 80%, and reaches 50% product concentration instead of 30%, the strength at which acrylamide is sold, so no recycle or evaporation is needed. For 7-ACA, the enzymatic route developed by Hoechst cut waste from 31 kg to 0.3 kg per kg of product. The enzymatic 6-APA route is only modestly cheaper, Kent says, but avoids hazardous reagents such as trimethylsilyl chloride.

- Kent (2012), Ch 31 Industrial Enzymes and Biocatalysis · pdf pp. 1184-1185, 1208, 1213, 1215-1218
- Shreve (1984), Ch 31 Fermentation Industries · pdf pp. 618-619

### 5.2 Biologics: proteins made by living cells

Biologics are protein drugs: insulin, growth hormone, erythropoietin, monoclonal antibodies. Kent's biotechnology chapter puts biopharmaceutical sales at about $45 billion, 10% of the pharmaceutical market, with 25% of new drugs now biologics. Kent's pharmaceutical chapter notes biologics went from none to eight of the top 20 US drugs between 2006 and 2009.

Small proteins can be made by bacteria. Kent describes human insulin made in E. coli as a precursor, proinsulin, in an 18 to 24 hour fermentation at 37°C. The protein piles up inside the cells as insoluble clumps called inclusion bodies, 10 to 30% of dry cell mass, which protect it and ease recovery. It is then refolded and trimmed with enzymes.

Larger proteins need mammalian cells, because bacteria cannot fold them correctly or attach the sugar chains (glycosylation) that many need to work. Kent states that 60 to 70% of recombinant protein drugs are made in cultured mammalian cells, mostly Chinese hamster ovary (CHO) cells. These are delicate: 37°C, pH 7.0 to 7.4, tight control of dissolved oxygen and osmolality, and protection from bubble damage. Batch culture runs 3 to 5 days and gives under 0.5 g/L. Fed-batch, the industrial norm, runs one to two weeks and gives over 1.5 g/L, with the best reaching about 4 g/L. Perfusion, a continuous mode that keeps cells in the vessel, gives roughly ten times the volumetric productivity. Production bioreactors hold 10,000 to 20,000 L.

Purification follows a sequence of capture, intermediate purification and polishing, mainly by chromatography. For antibodies, Protein A affinity resin captures the product in one step with purification of several thousandfold, then ion-exchange steps remove host-cell protein, DNA and viruses.

Data Modes of mammalian cell culture (Kent 2012)

| Mode | Duration | Cell density (cells/mL) | Product titre | Notes |
|---|---|---|---|---|
| Batch | 3-5 days | Under 5 million | Under 0.5 g/L | Simplest; some proteins degrade in medium |
| Fed-batch | 1-2 weeks | Over 10 million | Over 1.5 g/L (best about 4 g/L) | Feed concentrated nutrients when depleted |
| Perfusion | Many weeks | About 50 million | Lower than fed-batch | Cells retained; about 10x volumetric productivity; more effort |

 Source: Kent (2012), Ch 32 Industrial Production of Therapeutic Proteins, pdf pp. 1229, 1239-1240  In the data

#### What a cubic metre of reactor is worth

Pollak contrasts mammalian cell culture with chemical API plants (indicative figures). Investment per m3 of reactor: about $5 million against about $500,000. Output per m3 per year: several tens of kg against several thousand kg. Sales per m3 per year: $5-10 million against $250,000-500,000. Product concentration: 2-6 g/L against about 100 g/L. Reaction time: about 20 days against about 6 hours. Plant construction: 4-6 years against 2-3. Boehringer-Ingelheim's 90,000 L plant cost $320 million, $3.5 million per m3.

- Kent (2012), Ch 30 Industrial Biotechnology: Discovery to Delivery · pdf p. 1180
- Kent (2012), Ch 11 Chemistry in the Pharmaceutical Industry · pdf p. 422
- Kent (2012), Ch 32 Industrial Production of Therapeutic Proteins · pdf pp. 1229, 1239-1245
- Pollak (2011), Fine Chemicals · pdf pp. 66-67

### 5.3 Where the cost of a medicine sits

Pull the module together and the cost structure of a medicine looks nothing like that of a commodity chemical. Pollak, using data from European generic producers and IMS Health, splits the price of a drug product into four parts: cost of goods (API, formulation and packaging) about 20%, R&D 15 to 20%, marketing and administration 35 to 40%, and profit about 25%. The API alone is usually under 10% of the selling price. Within a fine chemical company the API's own price is about 35% raw materials and 55% conversion cost, the rest margin.

Two consequences follow. Originators price by the daily cost of treatment, not the cost of the molecule; Pollak notes 100 Lipitor tablets cost the same at 20, 40 or 80 mg. And after patent expiry the arithmetic reverses: generic producers chase every gram of yield.

Batch versus continuous follows from scale. Most APIs are needed in 10 to 1,000 t a year, so multipurpose batch plants win; Kent adds that batch suits cGMP because each lot is accepted or rejected on its own. Continuous operation pays where volumes are large and the catalyst is robust, as the glucose isomerase figures show.

Regulation is a cost at every stage: years of trials, validation of each process, batch release testing, and inspected plants. Pollak counts more than 2,000 producers of APIs for generics worldwide, but fewer than 10% meet the highest GMP and regulatory standards for the regulated markets. China has become the main source of large-volume generic APIs such as penicillins, and Pollak lists Indian firms such as Cipla, Divi's Laboratories, Dishman, Hikal and Ranbaxy among the fine chemical producers serving the industry, a position built on the process skills Dryden describes.

Data Where the money goes: drug product price structure (Pollak)

| Element | Share of drug product price | Comment |
|---|---|---|
| Cost of goods: API, formulation, packaging | About 20% | API itself usually under 10% |
| Research and development | 15-20% | Carries the cost of failed candidates |
| Marketing and general administration | 35-40% | Largest single element |
| Profit | About 25% | Before patent expiry |
| Generic discount after expiry | About 10% at first, up to 90% in competitive markets | Retail: 100 amlodipine 5 mg tablets $217 originator and $25 generic in the US, $3 in India |

 Source: Pollak (2011), Fine Chemicals, pdf pp. 185-187, 225, 239  Case

#### A $4 prescription and a $33 API

Pollak describes Wal-Mart's $4 generics programme: a month's supply of 300 common drugs for $4, against an average $43 pharmacy price for sixty 500 mg tablets of metformin. Sixty tablets hold 30 g of API, so $4 is $133 per kg of API in finished form. If the API is allowed 25% of that, the producer's target price is about $33/kg. Wal-Mart reached it by buying directly from the most competitive makers in China and India. At such prices step yields decide who survives.

- Pollak (2011), Fine Chemicals · pdf pp. 75, 185-187, 225, 239-240
- Kent (2012), Ch 11 Chemistry in the Pharmaceutical Industry · pdf p. 425
- Dryden (1997), IVH Pharmaceutical Industry · pdf pp. 566-567

### See it in the table
[Cu Copper](https://www.materialinsights.org/commodities/copper/)[Su Sugar](https://www.materialinsights.org/commodities/sugar/)[O2 Oxygen](https://www.materialinsights.org/chemicals/oxygen/)[ACN Acrylonitrile](https://www.materialinsights.org/chemicals/acrylonitrile/)[6APA Antibiotic intermediates (6-APA, 7-ACA)](https://www.materialinsights.org/chemicals/antibiotic_intermediates/)[Enz Industrial enzymes](https://www.materialinsights.org/chemicals/enzymes/)

### Check your understanding

Immobilised penicillin G acylase gives over 2,000 kg of 6-APA per kg of enzyme. Why does immobilisation matter so much for cost?
 Kent explains immobilised acylases can be reused many times, over 1,000 cycles in some cases, and are free of contaminating enzymes; spreading enzyme cost over thousands of kg of product makes the step cheap.

Which bioreactor mode is the industrial norm for mammalian cell culture, and why?
 Kent gives fed-batch titres above 1.5 g/L, three times batch; perfusion gives more productivity but needs more development and operating effort.

Pollak puts cost of goods at about 20% of a drug product's price. What is the largest single element?
 Marketing and general administration is 35-40%, against R&D 15-20%, profit about 25% and cost of goods about 20%.

[← Part 4: Fermentation: microbes as chemical plants](https://www.materialinsights.org/learn/chemicals/pharmaceuticals-and-fermentation/p4/)[Mark read · Wrap-up →](https://www.materialinsights.org/learn/chemicals/pharmaceuticals-and-fermentation/review/)

## Other sections

- [Module overview](https://www.materialinsights.org/learn/chemicals/pharmaceuticals-and-fermentation/index.md)
- [The pharmaceutical value chain](https://www.materialinsights.org/learn/chemicals/pharmaceuticals-and-fermentation/p1/index.md)
- [Making the API: multistep batch chemistry](https://www.materialinsights.org/learn/chemicals/pharmaceuticals-and-fermentation/p2/index.md)
- [Formulation: turning powder into a dose](https://www.materialinsights.org/learn/chemicals/pharmaceuticals-and-fermentation/p3/index.md)
- [Fermentation: microbes as chemical plants](https://www.materialinsights.org/learn/chemicals/pharmaceuticals-and-fermentation/p4/index.md)
- [Key terms, quiz and sources](https://www.materialinsights.org/learn/chemicals/pharmaceuticals-and-fermentation/review/index.md)