Anyone who even casually glances at this column knows that I always address potential return on investment (ROI) for any expenditure of technology upgrades. I had my fiscal enlightenment while we were stocking/expanding the QC lab at Sandoz, NJ early in the 1980s with new hardware. The company had been “encouraged” by the FDA to expand and modernize their facilities, expanding from thirty-five to eighty-five analysts and supervisors, growing from four to twenty-four HPLC’s, near-infrared for raw materials qualification, and so forth. In less than nine months, we managed to spend several million dollars. The experience, however, did teach me an awful lot about how to approach obtaining analytical instrumentation.
Like many of my contemporaries I tended to buy the instrument I needed, based on scientific ideals, not monetary considerations, which often led to buying units we already had experience with, assuming the slow morphing of instruments would proceed much as it had in the 50s and 60s. We simply purchased the unit that gave us the best (meaning sometimes “slowest”) piece of information. During the massive lab expansion, I began to think in terms of volume of samples assayed and speed of reporting, while not ignoring accuracy and precision. Not an easy thing for a “traditional” lab person to wrap his head around, but I had no choice. One perfect example was the twenty HPLCs I had just added to the instrumentation lab. They were virtually new from Waters Corp. with state-of-the-art auto-samplers, at that time, each capable of holding forty-eight samples—essentially, four batches and bracketing reference samples per unit—and worked quite well. But, quickly we needed more sample throughput and we did not have the option of doubling our lab space, equipment, and personnel twice in less than a year.
What I did was bite the bullet and began writing capital requests for new(er) auto-samplers that could hold ninety-six samples—twice the number samples run per HPLC in one single overnight run. While it took some effort on the part of the analytical staff, which now numbered 85, the larger number of samples were prepared and the potential throughput of the lab doubled with no additional HPLCs being needed. I was concerned about negative blowback from management for the added expenditures, but, one day the vice president of production saw me in the hall, called me over and told me he liked the way I wrote capital requests. When I asked why, he answered, “All the other scientists go on and on about better resolution, sharper peaks, etc. You simply explain how many more samples a night the device can analyze.”
Well, today, the trend of outsourcing after mergers and acquisitions has been occurring for years and has stretched both the supply chain and created a boatload of new CROs and CMOs. The unfortunate part of these new enterprises is that they tend to carry over many or all the habits of the bigger, older companies, both good and bad. These CMOs for the most part diligently follow cGMPs and mimic the purchase, quarantine, assay and dispensing of APIs and raw materials. Then again, in the production and analysis of the products, they follow the manner of the client companies. That is, slow and steady wins the race? Hence, all my previous suggestions of moving to PAT to QbD to continuous processing to modernize, not just copy the traditional ways of the “elders.”
This is where my ideas of updating lab facilities come in. Yes, it will cost money but so does doing nothing. For a simple example, the updating of a simple HPLC to a UPLC is a minimal technology change, often requiring a mere transfer of the identical mobile phase or gradient and column chemistry. This is followed by injecting the same set of samples and standards used on the regular units in parallel with a normal set-up. Sounds like a minor, simple change, but in my experience, a working UPLC can easily do the work of four or five normal units. One company, for whom I consulted, sold several of their conventional HPLC units and used the money to pay for a new NIR instrument.
The result of the transaction? All the LC assays were done faster and they were now the proud owner of an entirely new technology, essentially for free. The operators, previously dedicated to simply making samples, maintaining, and running the older HPLCs can now be rotated between the new UPLC and the new NIR, getting cross-trained and proficient on both. More work could be performed on more lots for more clients, all with the same number of less bored and better trained analysts.
After this simple switch from a slower to a faster technology, you may well ask, “What now?” To begin with, I would suggest “cutting your teeth” on the newly acquired NIR for incoming raw materials and APIs. This was the first practical pharma application, dating back to the early 1980s, so there are a massive number of references, plus every supplier can assist in setting up your RMID program. With experience, you could check crystallinity, polymorphic form, moisture, average particle size, and many more parameters with a single scan. With further experience, you can easily start performing qualitative and quantitative analyses of dosage forms. This experience, alone, would enable you to package and distribute clinical supplies with the ability to check the identity of 100% of the contents of every blister pack in just about an hour. Et voilà, you are able to expand business with minimum effort.
The multi-faceted benefits of becoming a better-equipped and modern lab begin to show. The benefits are:
- First and foremost, even if nothing changes in your business model, you will begin to see the many benefits of the faster units including faster release of lots, lower solvent costs and more bench space.
- If and when you begin to move to PAT and QbD, you will have the laboratory capacity to run calibration and validation samples to calibrate and validate your in-process monitors without needing more analysts or hardware.
- Never underestimate the ability to attract and retain better analysts with modern equipment and interesting, varied analyses. The idea of cross training ensures that someone will always be able to run a particular instrument. As I always point out, every analyst eventually is absent—short- or long-term—from his post due to sickness, vacation, paid family leave, promotion/job change, or simply leaving the company.
- If you expand to become a contract lab, you have the capabilities without having to layout capital expenses. You will have the capacity with your new hardware to add samples without having to add personnel or hardware. Merely pay for the extra solvents and columns folded into the fees you charge, and you are in business.
- You will have better trained and motivated analysts, capable of performing more and varied analyses on short notice. While this sounds grueling, it is exhilarating too, and makes the job more interesting.
- As a means to an end, the added income and experience may be used to expand production capabilities. As mentioned in previous columns, many companies will be looking for CMOs and CROs capable of accepting PAT/QbD-based processes. In short order, they will be looking for continuous processing capacities, as well. With the FDA “encouraging” contract manufacturing, more and more companies, even smaller organizations, will be headed in that direction. The small footprint, lower COGS, speed, etc. are like an addictive substance. Once they get their first dose, it will be a slippery slope.
What I want to emphasize is that a superior analytical lab stocked with enough good people is first an asset to a normal, well-run manufacturing organization. But, it may also aid in upgrading the process space and could well become an asset to outside companies, more than paying its own overhead, while actually becoming when well-planned and well-run, a “cash cow” for the organization.
Emil W. Ciurczak
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.