For the better part of the year, I have been commenting on ways in which smaller generic houses and contract manufacturing organizations (CMOs) can best weather the storms of supply chain difficulties. These mere shortages and delays have now been exacerbated by inflation and soaring energy costs. The fixes and improvements that this column have covered for supply chain mitigation are all still valid (e.g., using PAT/QbD parameters to rapidly qualify new/replacement vendors for raw materials), but, recent legislation should give further financial incentives to upgrade from batch-type GMP production to PAT/QbD and even continuous manufacturing (CM).

The latest bill working its way through Congress—approved by the Senate, in the House as this is being written—contains an amendment that would allow Medicare to negotiate drug prices for ALL the states (plus capping the cost of insulin for Medicare patients), versus the current patchwork approach of “every-man-for-himself,” favored by the Pharma companies. This should be law by the time this column is released.

This lack of competition explains why drug prices in the U.S. are often multiple times as expensive as in, say, Canada, for the same drugs produced by the same companies! Now, more than ever, companies need to maximize profits while rapidly lowering the cost of goods manufactured. And, as a side benefit, produce higher quality products for consumers.

For reference, I’ll reiterate some of the weak points of batch-type GMP production of solid dosage forms:

1. Raw materials arrive and are placed in quarantine, sampled, tested (usually by USP or EMA approved wet lab methods), passed, labels affixed and moved to the general population where they are taken to production. (Can take weeks.)
2. Raw materials are brought to blender, mixed, sampled and, when content uniformity is affirmed (necessity for generic houses), the material is dumped into containers and transported to a granulation room. (Several hours?)
3. The mix is “wet” with the proper solvent, granulated and dried. Samples are taken and analyzed for residual moisture—in a lab, no doubt. (Half a day, unless a rush is placed on the material.)
4. The dried granulation, often lubricated in the granulator, is then transported to the compression room (tableting machine) for tableting into cores. (Half a day?)
5. The compressed cores are then collected and transported to the coating room and loaded into the coating pans, or whatever instrument is used, and coated. Samples are regularly taken and weighed until it appears to be properly coated. (Adding several hours to the coating process.)

All these steps can lead to a 2, 3 or 4-month duty cycle for a single lot of a product, doing nothing to enhance sales or the company’s image.

It needs to pointed out that there is a loss of material at each of the first three steps, often as high as 10% (Ka-Ching! Profit already down 10%!). That, plus the costs involved in transporting and analyzing the intermediate products, plus the costs of cleaning and cleaning validation (and the time the equipment is own until cleaning validation results return from the lab), add up to more operating overhead and less profits.

What are some ways to speed up the process and minimize losses, you may ask? Well, drawing upon some of the previous columns, we can see that even minimal application of PAT/QbD technologies can bring large bonuses. Here are a few changes that can be made almost immediately:

1. Raw Materials
Since 1984, raw materials have been qualified with near-infrared instrumentation, resulting in immediate savings? Prior, sampling all the containers, or a statistically determined fraction, in the quarantine section of the warehouse, bagging/labeling the samples, bringing them to the QC lab, waiting for results, and assuming a passing grade, relabeling the containers and moving them to the “active” section of the warehouse, takes a minimum of a couple weeks between the delivery and the use of materials, assuming all things go smoothly, of course.

Using NIRS or Raman on incoming raw materials, allows them to be immediately passed or failed. Passed drums would go to the active section, while failing drums are simply put back on the truck and sent back to the supplier. Thus, we go from weeks to minutes.

2. Blending
Currently, under GMP rules, a blender is run for the time specified on the MMF (master manufacturing formula) then stopped, whether or not it is blended, under-blended, or over-blended. Samples are taken and sent to the QC lab to affirm its “blended-ness,” to coin a term. Many generic companies knowingly specify longer blend times than needed, then proceed to the next steps, assuming a blended mass, not waiting for lab results.

The dangers of over-blending have been shown: particle sizes can be reduced by friction, making a larger fraction of dust, which may be lost; the heat generated can cause degradation of the active pharmaceutical ingredient (API); the heat generated can cause polymorphic changes in both excipients and API. Some newer drugs are poorly soluble and need to remain in an amorphous state—heating can cause it to morph into a higher crystalline state.

And the two most obvious reasons to blend-until-blended? The blend is known to be well-mixed and, in another bonus find, time is saved—even if we only pretend to wait for lab results before proceeding, merely going through the act of sampling takes time.

Without crossing over to a full-blown PAT/QbD or continuous manufacturing mode, the tools used in these paradigms are available for piecemeal applications. The raw materials reference is the simplest to implement, with hand held NIR and Raman units available for numerous suppliers. In the aforementioned blending application, the wireless NIR units have been available since the early 1990s (Pfizer was a pioneer here) and are relatively inexpensive and simple to use.

With the advent of dazzlingly rapid NIR instruments, it has become increasingly possible to monitor (i.e., assay: measure weight, CU, assay and hardness) of tablets in real time. Figure 1 shows a unit that allows cores to be assessed for API (total and distribution) and hardness in 5 msec. The fastest tablet presses may be analyzed as the doses are formed. The unit also has the ability, using a directed air jet, to remove any tablet not within specs. Figure 2 shows how the tablet core is “lit up” by the incident light. With such a fast-measuring device, several snapshots of each core may be taken, further assuring good CU measurements. The hardness may also be correlated with dissolution prediction.

The instrument also comes with a microwave resonance attachment, allowing the weights to be measured, also in real time, and outliers removed, again, with an air jet. So, all tablets that move to the coating pans, if there is a coating step, have been screened for weight and API content. Figure 3 shows a schematic of the two modes used to measure the cores.

Lyophilized dosage forms may also be tested (100%) using similar technology. Figure 4 and Figure 5 show the methodology and how the entire vial is analyzed. The instrumentation also discards out-of-spec vials, assuring 100% of vials sent to labeling are in spec. These hints and techniques should be a good start on your path along the yellow brick road to better products, made in shorter times, with higher yields.

You’re welcome.

---

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.