Two decades ago, the U.S. FDA backed the pharmaceutical industry’s use of process analytical technology (PAT) through new guidance. The PAT framework was founded on process understanding with the goal of encouraging industry to modernize manufacturing.
The guidance defined process understanding as: “A process is generally considered to be well understood when (i) all critical sources of variability are identified and explained, (ii) variability is managed by the process, and (iii) product quality attributes can be accurately and reliably predicted over the design space established for the materials used, process parameters, manufacturing, environmental and other conditions.”
Hence, the objective identification and pedigree of batched chemical or pharmaceutical materials is fundamental to batch process understanding* (BPU), allowing insight into the identity and history of reactants, intermediates, and product materials. Natural-abundance stable isotopes (non-radioactive mass variants of common elements) are inherent in and compose virtually all materials and thereby act as tracers of product identity as well as process history.
In straightforward manufacturing processes, it is generally expected that one batch of reactant will lead to one batch of product. However, a recently presented first-of-its-kind study found otherwise, revealing the utility of natural stable isotopes in understanding the identity and sources of materials in the manufacturing and marketing supply chains.
*BPU derives from two of MIT LLC’s leading patents on isotopic product authentication and process understanding, the latter of which is still active.
Stable-isotopic analysis
As a complementary study to a process patent infringement case, we requested from our pharmaceutical client twenty matched pairs of reactants and products to test the batch (reactant-to-product) continuity in this study. We chose sulfur as the tracer of interest since it was transmitted within the molecular structure from the initial reactant to the final product without any chemical transformation. It was thus a conservative tracer of the manufacturing process.
If, as all had assumed, one batch of reactant yielded one batch of product, there should have been a simple 1:1 correspondence of the tracer between the reactants and products. Surprisingly, there was not a 1:1 relationship (see Figures 1-2); instead, we found two major anomalies: (i) The batches of reactants and products were shifted relative to each other by two batches in the batch records (batch shifting); and (ii) every batch of product was apparently a mixture that contained ~30% carryover of reactant from the previous batch (batch mixing).
These findings exposed that the contract manufacturer was apparently partially re-using batches of reactants — something which is not typically contractually allowed, i.e., all batches are supposed to be discrete or ‘neat’. In summary, natural stable-isotopic analysis of the sequential batches revealed the story of what actually occurred during production. Stable-isotopic analyses are fast, inexpensive and quite precise.
Implications of BPU
This story of batch process understanding is rather straightforward, and, as the FDA has favored in collaborative studies, there is nothing added to the system: It is all natural. We do not seek to point out error, but only to reveal the strength of natural stable isotopes in understanding the identity and sources of materials in the manufacturing and marketing supply chains. In the present study, an examination of batch processing dynamics revealed outcomes that deviate from the anticipated 1:1 relationship, challenging prevailing conceptions. Application of isotope-enabled BPU revealed an objective assessment of what actually happened.
Our findings do not necessarily invalidate established premises of batch understanding — but rather, they unveil unexpected and potentially insightful results regarding potential batch-record shifting and batch mixing. Such revelations underscore the complexity inherent in batch processing dynamics and emphasize the need for nuanced assessments.
We view the application of isotopic BPU as a promising avenue for further exploration and we hope that the industry sees the value of this approach and wants to pursue its full potential implications. We advocate for continued research endeavors in this direction, hoping that the industry recognizes the value of such approaches.
Currently, the main implications of BPU appear to be that (i) we have an objective means to trace the identity of reactants into products as opposed to written records, and (ii) we are able to trace potential mixing of reactant batches into final products — whether intended or not — to understand the flow of bulk materials as well as trace components (organic impurities, trace elements etc.).
