Sourcing of active pharmaceutical ingredients

DURING EARLY formulation development many pharmaceutical companies use single-sourced active APIs and typically only a small number of API batches. However, ICH Q8(R2)¹ indicates that, “at a minimum, those aspects of drug substances that are critical to product quality should be determined and control strategies justified”.

Additionally, APIs are often being developed in parallel to the drug products and consequently their critical quality attributes (CQAs) can change with time. Unfortunately, there is often too much focus on the chemical purity of the API; forgetting that changes in particle size, shape and bulk density can be equally important to the resultant drug product. The habit of an API can also be impacted by polymorphism; for example, orthorhombic paracetamol shows improved compression behaviour compared to the monoclinic form.²

Leane, et al.³ recently developed a four‑box model for drug products, termed the manufacturing classification system (MCS), for oral solid dosage forms. MCS classifies products according to their manufacturing processes; ie, direct compression (DC – MCS 1), dry granulation (DG – MCS 2), wet granulation (WG – MCS 3) and other technologies (MCS 4). DC is cheaper, faster and much less complicated than DG or WG; however, DC is also likely to be the process that is most reliant on the initial API particle properties. The API mean particle size and distribution, together with particle shape, are particularly critical for DC as there are no additional processing steps to mitigate those unfavourable API properties. The dosage strength and API loading are important considerations. At low drug doses (ie, ≤1mg) or drug loadings (<2.0 percent) DC strategies can be challenging, particularly from the perspective of blend and content uniformity.³ As such, it is important to understand the consistency of physical attributes of the API as there can be large batch-to-batch variability.⁴ Some API physical properties can also change with time. Freshly milled metformin HCl contains appreciable levels of surface crystal defects, which are not present in aged samples. These in turn can cause differences in flowability of DC powder blends during drug product processing.⁵ Many companies often introduce “quarantine” periods post-milling or micronising to address such issues.⁶

Unfortunately, those factors that influence API particle size and shape are not well understood and can be influenced by reactor size, shape, agitation speed and dryer type and configurations. They can also be vulnerable to scale up following the initial transfer into production.^7,8 In parallel, additional product demand and/or the need for supply chain flexibility often drives companies to assess dual-sourcing of their API. Ferreira, et al.⁹ recently highlighted an issue where a dryer at a second API site produced “denser, less porous material that leads to processability issues” in the drug product. Lastly, as the product enters the latter stages of its product lifecycle there can be economic considerations that underpin outsourcing the manufacture of the API and/or drug product to a third-party vendor. Here, the situation is often exacerbated as “institutional knowledge” underpinning the processing rationale for the product can typically be lost. Additionally, none of the major pharmacopoeias address physical properties as part of their API monographs. As such, it is not uncommon to encounter issues with the product/ process that can be attributed to subtle changes in the API affecting the drug product. Chan¹⁰ reported capping of metformin DC tablets arising from the introduction of a second source of API, which was attributed to different particle morphologies, ie, acicular versus plates. These potential issues highlight that the transfer and/or buyer specifications for second site sourcing of the API needs to focus on more than cost and chemical criteria; it should also encompass pharmacopoeial monographs and physical properties.

References

1. ICH Q8(R2). Pharmaceutical Development. Current Step 4 version dated August 2009.
2. Joiris E, et al. Compression behavior of orthorhombic paracetamol. Pharm Res. 1998; 15(7): 1122-1130.
3. Leane M, Pitt K, Reynolds G. A proposal for a drug product Manufacturing Classification System (MCS) for oral solid dosage forms. Pharm Dev Technol. 2015 Jan; 20(1): 12-21.
4. Gamble JF, et al. Application of image-based particle size and shape characterization systems in the development of small molecule pharmaceuticals. J Pharm Sci. 2015; 104(5): 1563-1574.
5. Vippagunta RR, et al. Investigation of Metformin HCl Lot-to-Lot Variation on Flowability Differences Exhibited during Drug Product Processing. J Pharm Sci. 2010; 99(12): 5030-5039.
6. Depasquale R, et al. The Influence of Secondary Processing on the Structural Relaxation Dynamics of Fluticasone Propionate. AAPS PharmSciTech. 2015; 16(3): 589-600.
7. Variankaval N, et al. From form to function: Crystallization of active pharmaceutical ingredients. AIChE J. 2008; 54(7): 1682-1688.
8. Hadjittofis E, et al. Influences of Crystal Anisotropy in Pharmaceutical Process Development. Pharm Res. 2018; 35; 2374-2379.
9. Ferreira AP, et al. Enhanced Understanding of Pharmaceutical Materials Through Advanced Characterisation and Analysis. AAPS PharmSciTech, 2018; 19(8): 3462-3480.
10. Chan M. The Effect of a Source Change for an Active Pharmaceutical Ingredient (API) or Excipient on the Finished Drug Product. A thesis presented to the University of Waterloo in fulfilment of the thesis requirement for the degree of Master of Science in Pharmacy. 2016.