Source: E-mail dt. 4.4.2012


Uncertainities Management in Drug Substance Development


Mr. G. Karthik

Project Leader employed with MNC Pharmaceutical Company, Bangalore

Dr. K. Mohan

Professor & Head, Department of International Business, Pondicherry University, Pondicherry-14.



It is quite interesting to see same technical function behaves differently in different business environments. Though differences get highlighted and would always be seen with which one is correct and what is wrong, overall business environment drives the final verdict. The paper aims to capture how uncertainties management differ in drug substance development with respect to, when DS is in drug development stage (new chemical entities) and when DS goes off patent (generic active pharmaceutical ingredients). Difference is captured in such a way it starts with uncertainty definition, how uncertainties are managed in three different scenarios, before concluding it with what can be mutually exchanged between the two.


Full Paper:


There exist different names for same chemical. Different people call it in different period of time and they are drug substance (DS), new chemical entity (NCE), candidate drug (CD) and active pharmaceutical ingredient (API). Each of these names has specific purpose to call and normally used on particular context to which the chemical is referred. When different names are used to refer to one chemical under different circumstances, it become more interesting to see how uncertainties are managed in this kind of dynamic environment.


Different fields mean uncertainty is multiple ways. Literature meaning states it as doubt, skepticism and dubious. Quantum physics means it as simultaneous knowledge of position and momentum of particle is impossible. Medically it can be called as state of various situations and conditions. In summary, it can be defined as a task/activity/event that might have variety of results depending on various permutations and combinations. Taking a cue from various definitions of uncertainty would like to assess various scenarios involved in drug substance and API development and how in each scenario uncertainty is managed.


Scenario 1: What specification mean in NCE versus API development?


NCE: NCE is referred when drug is under development. This covers a period of drug development between preclinical to phase 3 clinical trials. Knowledge on NCE synthesis progresses as the drug progresses from various toxicology and clinical trials. It roughly takes 8 to 10 years for the drug to make this transition.


Specification also evolves during this period of drug development. Often development teams get into dubious situation on what quality of NCE that they would end up. This is because of fact that teams get exposed to NCEs for first time. We can definitely describe this situation as uncertainty.


And this uncertainty is managed by defining specification for NCEs as what can be achieved rather than what had to be achieved.  Quality of NCE’s during development would normally be made as “fit for purpose” as NCE’s find itself used for different purposes (toxicology, clinical and formulation development). “Fit for purpose” working model in turn helps to free up the resources, time and hence reduce doing redundant work.


API: Active pharmaceutical ingredient is called when the drug goes off patent and gets open to manufacture by either players. Generic companies typically carry out process development prior to large scale manufacture. Generic companies have an inherent advantage of knowing specification of the API before start of development work. But this comes with a challenge that development teams need to synthesize API that meets pre defined specification and can also be called as uncertainty. Uncertainty in this case draws analogy with medical domain definition of operating under various conditions. Uncertainty hence this case is managed by extensive understanding on process, risk assessments, strong documentation and well defined protocols.


This makes an interesting affair to note how same chemical development undergoes different approaches over period of time. May be can we attribute this reason also to change of people too. Nevertheless, it becomes evident that there exists a difference in the way uncertainty is managed.


Scenario 2: Regulatory environment for NCEs versus APIs


NCEs: Regulatory interactions with various agencies (US FDA, UK MHRA, EMEA, TGA, etc.,) take place during entire course of drug development for NCEs. Every clinical trial planned to conduct in human beings in specific country would require respective agency’s approval to initiate, conduct and conclude the trials. Respective agencies are thus kept appraised with progress made on the NCEs during the phase of development. NCE development teams need to interact with agencies right from phase 1 to 3 clinical trials subsequent to which marketing authorization is sought. Levels of information that need to be submitted during this period keep increasing in proportion to the NCE progress to next stage. This would mean that knowledge acquainted as on submission date is good enough to submit during that date’s application. So NCE development teams normally have an option of mentioning any specific information is currently not known and had been planned to develop in future. Agencies during this phase of development tend to assess the risk that NCEs would cause to patients and absorb few acceptable uncertainties appreciating the fact that it is under experimentation. But it comes with a rider that trial needs to be halted if NCE causes any adverse effects on patients. To sum it up how uncertainty is managed in this case is by moving forward with the plan until making predictions/theories come to real. The progression would be put on hold at a point where the situation causes an adverse impact.


APIs: Interactions with agencies for APIs happens post process validation in the intended manufacturing facility. Agency interaction is necessitated to market APIs in regulated markets. Situation here is different with respect to NCEs interactions given the fact that NCEs interact based on both acquired and to be acquired knowledge whereas APIs interact only with already acquired knowledge. This proves to be a crucial factor in deciding the level of information revealed during submissions. This makes the API submissions to operate under well defined boundaries and cannot take any kind of leverage of mentioning any specific information/data is planned to be developed. This again replicates uncertainty is managed with lot of preset conditions.


Scenario 3: Cost of development for NCEs versus APIs


NCEs: Cost is very vital for NCEs during development phase. The reason is quite obvious for the fact that to keep drug development cost under control. One more reason that can be attributed to keep costs under control is that NCEs during drug development are often produced at risks. Risks can be described as NCEs get used in toxicology or clinical trials where often outcome is unknown. So costs do takes a centre stage during drug development. But contrary to the statement above costs are taken from long term strategic perspective rather than short term. This can be explained by considering a hypothetical situation. Consider that an NCE needs to be produced which would be used in toxicology and clinical trials for first time. Cost to produce NCE during this situation can be flexible but needs to be produced on time to start the trials as per schedule. This is because toxicology and clinical trials schedule cannot be compromised as their costs are significantly higher compared to NCE’s cost.


NCE development teams try to keep the cost under control (by deploying additional resources in development either to find cheaper route or to improve throughput in existing route) during times where NCE deliveries for other trials are off critical path. NCE development teams devise a strategy to produce bare minimum quantity in order to keep the trials going and also to keep the cost under check. These are various means how uncertainty is managed to strike a balance between costs, time and quantity.


APIs:  Cost is more important in terms of API development when NCEs development. It is again very much understandable that to be very competitive in market. As APIs need to compete with lot of generic manufacturers, brand that commands superior quality and affordable pricing creates and sustain its market. Cost in APIs during development goes to an extent that all possible avenues like solvents recovery/recycle and multiple crops recovery in processing are exploited. With more and more payers demanding for cheaper and affordable drugs, generic makers are always kept in their toes to put extreme efforts in brand pricing. Development teams are put under constant pressure to manage the product life cycle with continuous improvement over pricing so that margins are kept intact.


Again this replicates how uncertainties are managed under different conditions and can be showcased as another example.


Commonality that can be used by both NCEs and API development teams: Risk assessment is one common thread that can run along both the development teams. It can be effectively applied in both during API and NCE development in order to maximize opportunity benefits and reduce/mitigate risk impacts. As both teams need to handle uncertainties at will, various risk assessment tools can be deployed to make first time right in all endeavors.  




Again would like to reinstate that how same chemical behaves differently under different conditions. Uncertainty itself is a separate interesting area to see how different people handle same uncertainty in different setup. Combining uncertainty and NCEs versus APIs might provide some insights which can be tried at either ends. It could also be an interesting affair to try some NCEs approaches in APIs and vice versa. There are few more situations that are worth mentioning which include platforms like documentation and skill set required in running NCEs and API development program. Finally author would like to conclude stating that some parallels and analogies which are present in NCEs and APIs can also be tried other segments of the business as well.




Author wishes to acknowledge senior management team (in organization where author is employed) for the encouragement and opportunity to pen down these thoughts. Also would like to thank Dr. Mohan, professor in Department of international business in Pondicherry University in helping author to make this work known in wider circles.




a)      http://www.managedluck.co.uk/basics/index.shtml

b)      ftp://cs.pitt.edu/chang/handbook/01UNb.pdf

c)      Author’s personal work experience




The contents expressed in this paper are purely based on author’s personal views and does not represent the organization with which they are associated with.