When I was a youngster my dad gave me some good advice: “If you don’t want to do something, you’ll always find a reason not to do it. If you WANT to do something, you’ll FIND a WAY to do it.” This applies to converting from same-old, same-old GMP batch processing—and testing a miniscule portion of the final product—to real-time testing and a de facto examination, if not actual analysis of the entire lot of product.

I have already on multiple occasions discussed the cost savings due to a speedier time to completion, less warehousing needed for intermediates, better quality control leading to fewer customer complaints and FDA/EMA mandated recalls, and so forth. The reasons were logical and based on quality, as well as financial return. However, I may have overlooked the familiarity of the personnel at smaller companies (CMOs, CROs, and smaller generic companies) with the latest hardware and software for doing PAT/QbD. No, I am not going to expound upon the wonders and rewards of continuous manufacturing or merely point out that the revolution, begun almost two decades ago, is now far easier than ever to join.

I understand that the professionals who work at these smaller companies are as bright and educated as those at the “Big Boys.” What is different is the “loose change” available for them to attend professional meetings and seminars. As much as one can learn from websites and Google, actually attending seminars, meetings and conferences allows professionals to see the equipment and meet with their peers and exchange ideas on how to utilize the newer instruments, what they disclose, and how to use the information to control and improve a product’s manufacture.

Certainly, a column this size cannot replace hand-on experience or even a full-sized primer; but I will highlight several pieces of new and relatively new hardware and software that can allow even the novice to dip his/her toe in the (PAT) water. Spoiler alert: you install a PAT program much as you would eat an elephant—one bite at a time.

When the PAT (draft) Guidance was unveiled in 2004, the author (Ajaz Husaain, CDER, FDA) envisioned a company shutting down an entire production line, re-fitting all the equipment, calibrating and validating the line, training the operators and, after a year or so off-line, restarting the line under PAT. At the time, I was very, very cynical about a mass conversion to the process. Unfortunately, I was correct. However, I had a slightly different idea of how to convert the masses.

Unknown to me at the time, I was taking the first steps into the PAT-world in 1983, when I instituted a 100% qualification of all incoming raw material containers. Looking at existing wet and instrumental methods in the USP, BP, etc., I knew that even the major Pharma company that employed me (Sandoz) would be hard-pressed to expand the QC department, buy enough equipment for the process, and absorb the costs.

Happily, I was able to work with the foremost NIR supplier in the world at that time to develop qualitative software, generate equations, validate them, and institute NIRS for 100% of all batches of all materials entering our facility. One simple example of the time and cost savings would be a single lot of Lactose, USP. We generally received about 220 bags of the material per load—sometimes, several times a month. To check all the bags would entail:

1. Quarantine the lot and place red/orange stickers on each bag;
2. Open each bag;
3. Using a sample thief, sample each bag;
4. Label a whirl-pack and insert the sample;
5. Bring the samples to the QC lab;
6. Log the samples;
7. Assign the samples (220, on the average) to an analyst;
8. Run the wet tests and the Infrared/UV comparisons with a standard;
9. Report the results;
10. Have a supervisor check the results;
11. Label each bag with an appropriate sticker (green = OK to use); and
12. Move the bags to the part of the warehouse where approved materials reside.

While steps 2-5 existed until portable analyzers were developed, the other steps were neatly eliminated. The materials could be sampled where they were delivered and, as we found through experience, a typical Lactose, USP lot could be sampled, bagged, brought to the lab, scanned, and approved or rejected, all before lunch. Yes, the new method, performed with newly written software on a 1982 instrument, could analyze more than 200 samples in four hours! The wet and instrumental methods, aside from requiring an entire shift to be hired, would have taken up to two weeks to analyze the same samples While not a true PAT step—this is eight years before the PAT Guidance was even written—it was the first step towards accessing information needed for a QbD program (avg. particle size, moisture, polymorphic form, crystallinity, etc.).

These parameters were formerly measured by compendial methods, if only for checking that the correct materials were delivered. I discovered micronized API where it should have been 100-120 mesh by NIR and several bags of lactose were above the moisture limit, so 100% testing worked well. When this ability is extended to a formal SOP, we have the first under-pinning of PAT: characterization of raw materials so that we can control the process better.

Now, we can attach a wireless spectrometer (NIR for 2% => and LIF [light-induced fluorescence] for less than 2% API) to our blender. You say you’ve never done this before? No problem, the technique and technology has been around since approximately 1990, so the instrument suppliers have the wherewithal to assist you in choosing, installing, and validating their blend uniformity equipment. Now you have step one (raw material qualification) and step two (blending), so we move on to step three (drying/granulating).

Most major suppliers of fluid bed driers routinely supply extra ports, besides the traditional sampling port, to accommodate fiber optic probes from NIR or Raman instruments. Literature is available and the suppliers of both the driers/granulators and spectroscopic instruments are able and willing to assist with your timed conversion, then analyzed to monitor and dry/granulate to an appropriate endpoint. Since the PAT Guidance asserts that, in lieu of the minimum three “demonstration lots,” then fix your blend/granulation time via SOP, each batch is a validation batch. That is, the process continues, not to a fixed time point, but to completion.

So, through steps three/four, depending on whether you count drying as separate from granulation, we have now finally met the legal requirement for GMP (as seen in the CFR): that “meaningful” in-process tests must be performed. Considering many of the “traditional” tests for solid dosage forms (hardness, friability, and disintegration, along with loss on drying, all off-line), with these monitors we would have control and the required measurements. That leaves the penultimate, or final, step: compression, or encapsulation.

If the previous steps were not meaningfully checked, then the final dosage forms really, really miss the stated requirement: “a statistically significant number of final dosage forms must be tested.” Now, I’m just a simple, country chemist, but I don’t believe that analyzing 20-30 tablets from a batch of 1,000,000 to 5,000,000 units is in any way “statistically significant.” Here’s where several vendors come to the rescue. Several commercial instruments are available, and for far less than the cost of one rejected or recalled batch of product, that are capable of fully analyzing up to 100,000 tablets or capsules per hour, per line. According to military standards of testing, a batch of over 500,000 should have at least 25,000 of its units checked. Ask yourself: “which is cheaper and faster: HPLC or NIRS?”

For those of you that coat your tablets, the prior step may be used to assure the core integrity, and a separate spectroscopic unit (NIR, Raman) may be used to measure the coating levels, in real time, assuring the proper time to completion. So, it seems that PAT is not only allowed under GMP, but aside from the money and time saved over current methods, aligns your process more closely with cGMP rules. And, as I mentioned at the start, all these need not be done in one bite, which is expensive and time-consuming. Taking one “nibble at a time” is a better approach.

In truth, other than raw materials analysis, whichever of the other steps you update does not matter. You need not go in order. In fact, see which step is giving your operators the most trouble and start there. 

Author’s note: specific suppliers may be obtained by contacting me via email. The column is not meant to be a commercial for specific suppliers.

---

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.