Capital expenditures scale up, then scale down. Regulatory scrutiny expands, then contracts. Mid-sized and large companies want to insource and build capacity to keep risks and control in-house, but then it becomes more fashionable to outsource and focus on core competencies. It’s the same type of effect as occurs with market volatility. There are so many multifactorial inputs that elucidating ‘the’ major trend driver is elusive, and too sensitive to other variables to be actually predictable. Any predictive accuracy in these cases either has an exceptionally short time horizon, or is simply a matter of chance. To claim that more is known is what is called a ‘ludic fallacy’ and is based on a misunderstanding of the actual level of precision of available data and how well those data explain an underlying trend. For example, in many outsourcing surveys, the margin of error is equal to (or in some cases higher!) than the mean. What this translates to in actual fact is:

Fictitious Analysis
Reported results: 50 companies responded; 30% of mid-sized to large pharmas outsource 30% of their API manufacture now; Intent reported is that the 30% will increase outsourcing next year by 10% (to average 40%)

What’s actually behind the statistics: With 50 companies responding, 30% represents 15 companies (see Figure 1).

The margin of error in the outsourcing reported would mean that if they had 10,000L of capacity each (on average)—than between almost zero and up to 6,000L of capacity may be outsourced currently, with a response of 30% selected in the survey (see Figure 3).

Then if the reported result for next year’s outsourcing isn’t a paired comparison, you could have a completely different subset of 15 out of the 50 companies from the first response (see Figure 2) suggesting that they may outsource by about 10% more next year. Again, this could draw from a different baseline capacity average and apply a different expected growth. These statistical errors have been found in some of the industry trend analyses, and lead to wildly-misplaced capital investments (and then capacity shortfalls or excesses).

What (We Think) We Know
The outsourcing of API and intermediates also differs widely depending on the slice of the industry you look at. For emerging biopharmas and virtual pharmas, a vast majority of their API and intermediates manufacturing is done by outsourcing providers. This is also largely the case with generics manufacturing; orchestration is done by the generic pharma company to have the API manufacture outsourced by a large proportion.

Within the top pharma and biopharma companies (again, getting different trends by different stratification of the business), there is still a substantial upward trend of intent to outsource, but a nearly equal proportion of manufacturing of APIs is done in-house. Much of this change in proportion in top pharmas relative to smaller, emerging, or virtual companies is due to excess capacity at the larger companies. But as the biosimilar landscape begins to get more clarified, and as pricing and cost-consciousness becomes more of a paradigm, there are likely to be changes in approach to pipeline development, and increased utilization of outsourcing, with the most rapid growth expected to occur beginning around 2017. Importantly, excess capacity (under-utilization) can indicate a strategic prioritization of a pipeline, and not necessarily an anemic pipeline. Also, higher corporate burn rates can be associated with drastically enhanced learning curves, leading to important growth over a period of a few years.

Risk Aversion and Pipelines Vis-à-Vis Outsourcing
Because risk aversion within drug pipelines has become more of a status quo over the past eight or so years, only the most promising drug candidates will survive the selection process. This has led to market impressions that the number of drugs in development has decreased. But although individual pipeline proportions have contracted, there are more smaller-scale pipelines at play worldwide. ‘Pipeline rationalization’ has become de rigueur, and developability of molecules has made it possible to apply predictive analytics to those treatments which offer the highest likelihood of success. Though in the short-run this may reduce a total pipeline’s volume, it also means that each individual molecule within the pipeline each have a higher probability of clinical success and marketability. In this way, to say that smaller pipelines ipso facto mean a worse pharma market is entirely untrue, and represents a logical fallacy of false equivalence.

Worldwide Outlook
The lower cost structures in India and China have led to more rapid and expansive growth in API manufacture there than anywhere else in the world when other variables are controlled for. Regulatory scrutiny of outsourcing providers—especially in Asia—remains very high and Quality problems continue to persist. Though whether this is due to objective differences from the rest of the world ex-Asia or an artifact of higher-resolution examination of the industry in Asia is a difficult question to answer substantively.
Because there has been increased use of API outsourcing providers in Asia for clinical development, this segment of business development remains a bit riskier than other aspects. Regulatory scrutiny has been aggressive and relatively conservative, but will that continue in the face of increased drug therapy pricing pressures worldwide and a desire to bring generics and biosimilars to underserved regions? The answers to these questions remain to be seen, and are more a function of popular opinion and social pressures, rather than an underlying objective set of truths.

However, an increase in willingness of innovator companies to use outsourced API and advanced intermediates is a tacit signal that there is more trust in outsourcing as a way to meet business goals. For example, a single-source supplier of API who incurs a warning letter due to insufficient or invalid practices can dissolve one full year (or more) from the forecasted targets of the innovator company contracted with them. These outsourcing decisions are therefore not taken lightly by the parent pharma companies, and because it has been more of a common practice to engage in these outsourcing relationships for advanced intermediates and APIs shows that it’s a risk more companies are willing to tolerate and take.

---

Ben Locwin
Healthcare Science Advisors

Ben Locwin, PhD, MBA, MS writes the Clinically Speaking column for Contract Pharma and is an author of a wide variety of scientific articles for books and magazines, as well as an acclaimed speaker. He also provides advisement to many organizations and boards for a range of healthcare, clinical, and patient concerns.