I just attended my first Contract Pharma Conference and I was introduced to a new world. Although I have been in the Pharma industry since 1970, it was almost all with larger companies. When I began my career (Ciba, Summit, NJ), the active pharmaceutical ingredient (API) was usually synthesized in-house, the product made in-house, packaged in-house, stored in our own warehouses and shipped directly to hospitals and pharmacies. And then, blockbusters happened (Valium, Librium, Viagra, and yes, OxyContin) and the companies grew quickly to accommodate the demand. At one point in the early 1990s I believe Pfizer had over 80 sites running, and the other biggies were into M&A, too.

Then, a funny thing happened; well several, actually:

1. The major pharma companies became truly international—not just selling to European and U.S. markets—and expanded the supply chain across the globe.

2. With the spread of markets, several things happened:
  a. First, packaging and distribution were decentralized (ceding record-keeping to contractors)
  b. Then manufacturing and clinical studies were outsourced.
  c. Along the way, manufacturing sites were opened in developing countries.

This trend away from centralized “doing everything” has generated an entire industry. Contract organizations, and supporting industries, have multiplied exponentially over the last few decades. This is a good thing on many levels:

1. In a large organization, especially with fewer lots made due to generic competition, there is a fair amount of “down-time” for much of the equipment. With a much higher overhead (wages, HVAC, taxes), it becomes attractive to have routine production of existing products manufactured by a contract manufacturing organization (CMO).

2. Each specialized CxO can focus on an individual portion of the R&D through sales of a drug product:
  a. One can concentrate on properly synthesizing APIs.
  b. One can focus on clinical studies (smaller batches, heavy paperwork, statistics of patient responses, etc.)
  c. CMO’s have been in existence for years, concentrating on one thing: making the product as well as possible, in a timely manner.

3. A “cottage industry” of parts/components suppliers has grown around the outsourcing trend. These smaller manufacturers can focus on making better “widgets,” such as bottles, packaging materials, labels, package inserts, and such. This is important because, unlike a smaller number of manufacturing sites (previously only in-house for the larger companies) making larger batches of drugs, we now have many CMOs and generic houses making smaller batches of the same drugs at many locations. This very “smallness” can make them more agile and faster to change over from one manufacturer of product X to another, also making X under a different brand name.

4. Contract organizations may be located closer to where their output (clinical supplies, API, final products) is needed, lowering supply-chain and shipping charges and complexity. This makes “just-in-time” manufacturing and product deliveries possible.

From a quality and safety point of view, the contract organization would have a simpler range for, say, the lab personnel. A manufacturer’s lab is simply responsible for analyzing raw materials, in-process and final products, and stability samples. All are usually compendial or modified compendial methods, derived from USP, ASTM, BP, JP, etc. This assures that the products throughout the industry are, at least, consistently analyzed.

Having said all these nice things, there are some negatives to this diaspora of responsibilities. The pharmaceutical industry, in general, is the most conservative in the world. Of course, it is the most regulated and inspected in the world, so that is one obvious reason for glacier-like pace of change. There are a few other practical reasons for not applying newer technologies:

1. When a company has almost more orders for a product than it can fill, it spends more time producing lots than finding newer methods of analyses. It has long been seen that the industry uses amazing technology to develop and new API, analyze the blood of clinical patients, and develop a new dosage form.

At this point, much of the company’s patent life has been used, leaving a relatively short time to make R&D costs back and see a profit. The chance of an NDA (new drug application) being delayed by using “new” analytical methods causes them to continue to use USP-based methods, many dating back decades. This means a natural resistance to PAT/QbD (Process Analytical Technology/Quality by Design) programs.

2. If the originator of a drug product uses compendial methods, with all the resources available, there is not much chance of a smaller contract organization adopting newer, initially expensive technologies. In addition, adding a methodology after the fact is harder and the CMO seldom has the personnel or time (or permission or desire to submit an ANDA) to develop a new method for a contracted product.

3. Should a contracted organization desire to apply a newer, non-traditional technology (i.e., Near-Infrared, Raman, or LIF [LASER-induced fluorescence]), it would be most difficult since product clinical studies, final dosage form development, pilot plant scale-up, and production/stability can easily be spread over four locations in four different states, or countries for that matter. Clearly, the technology involved would have to be duplicated over all four locations—mandating validation in all locations—and would be far more expensive, time-consuming, and complex than simply applying it at a single company performing all these functions.

Another difficulty for a CMO to develop alternative methods is often the sheer number of products a typical site could be producing in a given timeframe. A generic house, for example might file ANDAs for multiple drugs each quarter. Where an initiator company generally only files a few NDAs per years, allowing for sophisticated (alternative) analytical methods (lab and on-line) to be developed. The smaller company, burdened with smaller budgets AND many more products would be less likely to be the point of modernization. So, if the new technology isn’t handed down to a CMO, it is not likely to be adapted by them at a later date.

So, the creation of this major industry by the growth of outsourcing has numerous good, some not so good, and some potential downsides. I mentioned at the beginning of this column that all work was performed in-house in the “good old days.” What should also be explained is that all excipients and some APIs were, to use an organic chef’s line, “Locally Sourced.” That is, the talc came from a mine in Alabama, starch was from the Midwest, lactose from our dairy states, and so forth. The USP, developed for neighborhood pharmacies to identify and check basic safety of materials, was dependent on American (and sometimes Canadian) suppliers to basically be honest and subject to U.S. law.

Now take a look at the heparin “problem” that happened a few years back. Most of our heparin is imported from China, which does not (did not, at least) have the same levels of oversight as here from the U.S. FDA. For reference, there was a drought in China, causing a loss of pigs; heparin is derived from the stomach lining of slaughtered pigs, collected by farmers, NOT pharmacists or chemists. The shortfall of actual heparin was made up by adding over-sulfated chondroitin, which is used for joint pain in humans and animals. Heparin is a linear polysaccharide made up of α-l-iduronic acid, β-d-glucuronic acid and a-d-glucosamine repeat units. Chondroitin sulfate is a sulfated glycosaminoglycan (GAG) composed of a chain of alternating sugars (N-acetylgalactosamine and glucuronic acid). Both are seen as equal in older “wet” tests. The older USP ID tests did not differentiate between the two, so a 50/50 mix could not be determined, leading to a number of deaths. [Spoiler alert: easily distinguished by NIRS.]

The tests, designed as a double-check for local pharmacists, are not optimal for a multinational supply chain. As newer technologies were introduced (e.g., NIRS for raw materials qualification, a staple since 1985), many of these problems, due to inexperience or lack of oversight abroad, can be obviated by moving to NIR or Raman spectrometers for all incoming materials.

It is expensive and complex when the single-company, even with a multi-site company, moves production to several independent contract locations. What might have been one or two units in a warehouse, now need units in several locations, all validated and operated by different personnel. This carries over to dosage form analysis, and so forth.

So, in short:
1. Multiple sites for multiple functions: lower overhead, since larger companies have higher costs; this is spread over the cost for each batch, making them more expensive to produce, when compared with a contracted company. Increased logistics of control, simply from having to interface with multiple sites in different states or countries.

2. Updating technology of analysis: Initially, analyses may be dispersed, since it is often based on compendial methods. This assures that parallel labs would have parallel methods. The downside is less chance of new technologies introduced into the analysis scheme.

A potential difficulty is variations in raw materials in differing locales (Pfizer ships materials to its overseas production centers), but, in time, all problems can be overcome, making outsourcing a good trend.

---

Emil W. Ciurczak
DoraMaxx Consulting

Emil W. Ciurczak has worked in the pharmaceutical industry since 1970 for companies that include Ciba-Geigy, Sandoz, Berlex, Merck, and Purdue Pharma, where he specialized in performing method development on most types of analytical equipment. In 1983, he introduced NIR spectroscopy to pharmaceutical applications, and is generally credited as one of the first to use process analytical technologies (PAT) in drug manufacturing and development.