Ian Haley. Senior Technology & Application Consultant at Mettler Toledo.

Ian Haley

 Ian Haley. Senior Technology and Application Consultant at Mettler Toledo.In wet granulation, many companies used to rely on empirical data and quality control, but processes and technologies have improved in a variety of ways in recent years. The key differentiator is an increase in process knowledge and understanding, which has enabled robust and reliable processes to be designed from the outset, and existing processes to be improved with concomitant improvement in quality.

Improved understanding doesn’t just happen; it must be acquired through observation and analysis. The industry’s adoption of PAT has been a major enabler in this area. Of course, PAT is nothing new — the fine chemical and food industries have been using it for many years with demonstrable successes, but the pharma industry has lagged behind. In 2003, however, the FDA released its Draft Guidance on PAT, which served as an industry‑wide wake‑up call and began to seriously push PAT into pharma manufacturing. PAT is certainly an enabler, but it is a means to an end rather than a solution in its own right.

Demand for process measurement has led to equipment providers developing a number of instruments, such as NIR spectrometers and in‑line particle size analysers that are suitable for working in the hostile environment inside a granulator. In turn, this has also led to an increase in the quantity and quality of the measurement data available from a granulation process, enabling the concept of Quality by Design (QbD) to be adopted. By having good process knowledge and the data to prove the validity of that knowledge, you can understand what process variables and parameters drive product quality. In my mind, QbD is the single most important advancement. It sounds simple; if you know what makes your process tick, you can design a process from the outset to give a consistent high quality product with the minimum of out-of-specification material. It’s just taken the pharma industry a while to get there!

Interestingly, when it comes to the granulation equipment itself, I think there has been only modest advance. With the exception of a touch screen and some nice graphics, a process operator of 15‑years ago would have little difficulty recognising or using a modern granulator. The only exception is the development of ‘one‑pot’ granulators, where granulation and drying steps are combined.

In-process particle size distribution technology

In-process particle size technology is exactly as it sounds; a measurement of particle size and/or particle number is made in real time on the particles and granules during a granulation batch. Data is obtained continuously and rapidly, enabling the evolution of granules to be tracked throughout the batch.

It is not pleasant for a particle size measurement instrument to be on the inside of a high shear wet granulator. The particles are travelling at high speed, there are solids and liquid droplets present, and the particles quickly become wet and sticky, with a tendency to coat anything with which they collide. 

One approach is to avoid going inside the granulator all together and to capture images of the granules via a suitable window. Computational image analysis then yields the required particle size distribution data. Modern computers can certainly handle the number-crunching required, but such imaging techniques require the use of powerful lasers and are very tricky to set up to obtain reliable data. To date, such equipment has not escaped the confines of academic research groups and remains non-commercial.

By far, the most widely-implemented and commercially successful approach has been the use of probe‑based optical particle size techniques using the principle of light reflectance or light transmission to measure particle dimension and number. Such devices measure particles as they move past a window in the probe, and the challenge has been to overcome the problem of granules sticking to the window and thus preventing the measurement of good quality data. One solution (Mettle Toledo’s FBRM C35 technology) uses a mechanical wiper blade to physically remove stuck particles from the probe window.

The use of real‑time, in‑process particle size instrumentation allows users to track changes to particle size and number in actual time, and quantify the effect that key process variables such as impeller speed, water quantity and addition rate, formulation and batch size have on average granule size and size distribution. This, in turn, ensures the rapid and effective development of a granulation process that can consistently produce granules of the required size and distribution to give an oral dosage form of the desired performance characteristics

Probably the most important new capability that in-process particle size distribution technology gives to pharma companies is reliable and consistent real-time endpoint determination. A typical wet granulation batch process is quite short (approximately 10–15 min). Towards the end of the batch, significant granule growth can occur in a very short space of time. Only a real‑time measurement gives you the necessary speed of response to stop the process at precisely the right moment.

Historically, endpoint was often determined by monitoring torque or power consumption, but this is a hit-and-miss approach. As a granulator ages, mechanical wear on the gearbox and motor changes the power consumption anyway. Furthermore, these parameters only tell you how much effort is required to push the impeller blades round. While the granule size does influence that to an extent, many other factors are also involved.

Although not a new activity as such, another key benefit is the ability to shorten process development and scale-up times. In‑process particle size technology enables a huge amount of critical information to be obtained, with minimum experimentation and test batches that can help bring a product to market faster. 

The technology can also be used for continuous processing. I’m tempted to say that the technology enables continuous processing to happen, but that is probably pushing it a bit too far. Certainly it presents a possible solution to a rather thorny issue affecting the adoption of continuous processing in the pharmaceutical industry; namely, how do you ensure (and prove) sustained product quality to the satisfaction of regulatory bodies? In‑process particle size technology allows you to monitor the output of your continuous process all the time. With traditional offline analysis from a batch process, you assume that the results from the few grams of granules you extract are representative of the rest of the 50 kg batch. With a continuous process, however, you can directly measure a significantly higher proportion of your output, thus giving much greater confidence in the measure of product quality.

Real-time measurement also makes real-time control possible. With sufficient understanding of the ‘performance envelope’ of your process, it is possible to identify when the process is drifting towards an out-of-spec situation and make a corrective action before unacceptable material is produced. It’s not exactly rocket science for someone running an oil refinery or an ice cream factory, but it’s a significant change for pharmaceutical secondary manufacturing!

Despite the advantages, there are still two key areas where the technology can be improved upon. The first is in the interpretation of data. With in-process instrumentation it is possible to acquire a colossal amount of data in a very short space of time. Data, per se, however, has little value until converted to meaningful information that enhances knowledge or understanding. It is this data-to-information process that can be improved upon by the development of intelligent software, and improved training and support from technology suppliers. 

The second area is the provision of physical information, in addition to granule size distribution. While granule size distribution is a critically important parameter, other granule properties such as shape, density and compressibility also play a part. An in‑process instrument that can provide a real‑time measure of granule density or compressibility is probably a long way off, but granule shape measurement is a much more realistic goal.

All of the major pharmaceutical companies are fully aware of the potential benefits of QbD and PAT. However, some have been quicker to adopt new technologies than others. The use of in-process NIR spectroscopy for moisture content or blend uniformity is widely used and accepted, but in-process particle characterisation lags a little way behind.

Fear of the unknown is only one of the factors. Clearly, a poor experience with one PAT method will make companies wary of reviewing alternative methods, even though the technologies may be investigating totally different unit operations and be unrelated to each other. Sometimes, there is also concern that putting a probe into a granulator will in itself change the granule size. This concern is understandable, particularly for an existing process where there are regulatory barriers to changing the process. Accumulating evidence, however, suggests that the presence of a probe has no significant effect on granule size.

Sometimes it is simply logistical reasons that act as a barrier. Some companies have pilot plants where every granulator is used to make batches for clinical trials and, thus, subject to strict change control rules. There is a significant bureaucratic and cost penalty to making equipment modifications. Even adding an extra port to the lid of a granulator to install an in‑process particle size analyser to facilitate better process development, which is an easy process, still involves a huge cost. Of course, all of these obstacles can be overcome and I am sure that we will see much more widespread use of in‑process particle characterisation in the coming years.

This article is part of a special feature on granulation that was published in the March issue of PTE Digital, available at 

Articles

In wet granulation, many companies used to rely on empirical data and quality control, but processes and technologies have improved in a variety of ways in recent years.