The Right Tool for the Right Job

While I promised (sort of) to move away from supply chain matters, one trend seems to be M&A (mergers and acquisitions) to alleviate delays. This makes me smile, since I have been around long enough to remember when we made our own API (in-house), generated the clinical, scale-up, and production samples (in-house), packaged and stored the finished products (in-house), and delivered to hospitals and pharmacies from our own warehouses. Then came outsourcing…and the pandemic…and now comes “Back to the Future.” Look for more and more self-sufficient producers to re-build their former business models.

This trend was seen in lab instruments, too. In the 1970s, HPLC (light on the “H”) units were integral units (solvent supply, pump, injector, and detector). In the 1980s, companies like Waters trended to discreet pumps, injectors, columns, and detectors, supposedly for ease of repair and replacement of components. When this concept became clunky and a space-killer, the companies re-engineered their units into single, space saving units. So, the self-contained concept works at all levels, it would seem.

Since the pandemic exposed the weaknesses in our current mode of operations, we have had to re-examine all the steps of production from formulation through final product delivery. With shortages and actual inaccessibility of some key ingredients. As a consequence, simply “following SOPs” has become almost impossible, in many cases. The ability to freely substitute excipients and sources of API, for example, would be quite difficult under older cGMP batch-type manufacturing.

However, under PAT/QbD, the chemical and physical parameters of all materials is understood and, as a consequence, substitutes may be used to allow production to stay on schedule. To that end, I will discuss some technologies that would aid all sized companies in the formulation and production of small molecule drugs. I’ll also mention some bio-applications, later.

Until quite recently, the techniques/technologies I will cover were strictly used for R&D, formulation, and scale-up operations. With the increase in PAT/QbD and Continuous Manufacturing operations (and the snarky supply chain problems), production personnel need to know the physical/chemical properties of the raw materials they have on hand, not just rely on the previously specified parameters from R&D. There is an increasingly good chance that some substitution of the raw materials has occurred; not out of malice, but a necessity for suppliers to meet delivery guidelines. After all, abundans cautela non nocet (abundant caution does no harm), so some changes are needed for quality assurance of our products.

Some of these in situ tests are straightforward: NIRS or Raman for correct polymorphs; % moisture by NIRS; proper particle size,1 weight, etc. using spectroscopy (Raman or NIRS) and Principal Component Analyses. As an example, hydroxy methyl propyl cellulose (HPMC), when scanned by NIRS, produces a spectrum that is not only representative of the “basic” chemistry, but includes the effects of cross-linking, particle size, moisture, proper optical isomer (d or l; R or S,2 etc. Without using a bank of analytical methods, simply applying Principal Components Analysis will show which of incoming lots were “the same” or “different” from previously acceptable lots. Displaying the lots versus past lots in a 2-D display generates a “cluster analysis” (see Figure 1).

If a company runs compendial tests on incoming lots, and sorts them into categories such as “wrong particle size (larger or small),” “too high moisture,” wrong cross-linking or chain-length,” they can be made into “known” clusters and used to flag OOS lots and suggest which tests to be run. This qualitative scheme can be used for every incoming raw material (excipients and APIs). It is, in essence, an upgrade of what we instituted at Sandoz in 1985. This approach, alone, can screen out unacceptable lots, precluding OOS batches.

This approach is also good for final dosage forms (see Figure 2):

1. Clinical trials, determining that each dose is what the paperwork states. A tablet/capsule can be scanned through the blister and, non-destructively analyzed for ID and strength.

2. Granulations, especially direct blend vs. granulated, can be checked for not only homogeneity, but the levels of multiple APIs (and major excipients) are present in the proper ranges. The levels may also be measured throughout the process, an important test if there is a chance of “de-mixing” or stratification during the production process (vibrations are sometimes strong enough to change the blend uniformity status).

In 1986, we used Mahalanobis distances to show the proper ranges of aspirin, caffeine, butalbital, and codeine in a blend for a prescription analgesic tablet.3 While “semi-quantitative, it was an excellent screening technique—25 years before PAT was instituted.

3. Packaging rooms, to assure the correct product goes into the proper container. Yes, this happens far too often. Recalls are the result, assuming an injured patient doesn’t sue first.


Alternate/updated technologies
Back in the 1980s, when we were not only exploring NIRS for raw materials, but looking to update more mundane compendial tests such as particle size and melting range. Fortunately, we had a forward-thinking QA director who worked with us to design “alternate” methods: for melting range, we substituted thermal analysis for the side-arm tube, filled with mineral oil and thermometer and we substituted LASER-light scattering for sieve analysis. We began using both techniques and, in the year-end report, QA simply stated we have begun using the alternate methods. Were there to be a questionable result, we could always repeat the assay with the compendial method, of course.

Working with university groups who are heavily into PAT and modern methods of analysis is highly recommended. Schools like Rutgers and Duquesne have a good combination of academics and former industrial personnel and, for good measure, often have underwriting from major Pharma companies. Without naming names, I will simply say that Rutgers is located in New Brunswick, NJ (connect the dots).

One nice technique I observed at my alma mater was a device to show relative “slippage” of powders, both granulations and pure excipients. It consisted of a clear cylinder about 1.5 meters long and 0.5 meters in diameter. It was placed on set of rollers and rotated at 1-5 rpm. At one end was a series of lights and at the other, a series of detectors. When the tube was filled to approximately 1/3 with the powder to be checked, the lights were lit and the rotation started.

As the powder is “pulled up” the side of the cylinder by friction, it reaches a level, depending on its physical parameters, then tumbles back to the base. As the rotation continues, a profile of the slippage develops. This profile may then be compared with previous good and bad materials. It is a fast and simple way to determine whether the new (different vendor’s) material will perform correctly. Knowing the “relative slip-ability” could save a blend from cavitation in a hopper during production. If you are desperate to get the product out, knowing this data could allow you to place a tapping device to obviate cavitation.

Dissolution apparatus exists that allow in-vessel measurements (using fiber optic probes) at one second intervals. That is, in the specified 10-minute time for 80% dissolution of an immediate release dosage form, you can record 600 data points. What does that buy? Graphically, both freshly tableted and stability samples would generate dissolution profiles. The shapes of these graphs would allow the QC department to ascertain whether the substitute raw material both gives initial and longer-term comparisons of “new” versus “evaluation” materials in rather dramatic fashion. This approach is also useful for comparing the long-term physical stability for various formulations for a given product.

I know. Most generic and contract manufacturers do not have trained personnel to design and implement these proposed tests. Since the cost savings of no more OOS batches to destroy or recall, these savings could be used to hire trained chemists and purchase/build the needed instrumentation. Where are these scientists? Well, the aforementioned universities are turning out the very people you could use. While not “seasoned” in GMP methods, that would be a good thing since they will not need to be “deprogrammed” or forced to think “outside the box.” They already are firmly in the 21st century and ready to begin “new and improved” testing of your newly sourced and purchased raw materials. And, don’t forget that there are great catches out there as victims of the M&A frenzy after COVID shut-downs. A number of really good people are ready to step in and help.

It might turn out that the lemons of the supply chain slowdown (shut-down?) could wind up helping modernize your process and allow the newer, closer suppliers of your materials to save even more, now that you aren’t paying for transcontinental shipping.  And, as an added bonus, the better understanding of your products and processes will greatly help your product quality. Ready for some lemonade?

References

1. “Determination of Particle Size of Pharmaceutical Raw Materials Using Near-Infrared Reflectance Spectroscopy”, Spectroscopy 1 (7), 36 (1986).  Co-Authors: R.P. Torlini, M.P. Demkowitz
2. “Use of NIR in Thin Layer Chromatography”, Applied Spectroscopy Reviews 27 (2), 125 (1992).  Co-author: D.M. Mustillo.  (Invited)
3. “Identification of Actives in Multi-Component Pharmaceutical Dosage Forms via Near-IR Reflectance Analysis”, Spectroscopy 1 (1), 36 (1986). Co-Author: T.A. Maldacker.

---

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.