Is Lyophilization Deceptive and Devious at Times?

/Home/Lyophilization Articles & Reports

Is Lyophilization Deceptive and Devious at Times?
Thomas A. Jennings, Ph.D.
December 2003
INSIGHT
Recently I was in correspondence with a colleague whom I respect a great deal. I share with them my thoughts on the above question and to my surprise found that there was quite a difference of opinion. It could have been my blunt choice of words or that I seem to have given a technology a rather ugly personality that aroused such a reaction. But then statements from one who considers himself a Novice in this field may tend to cause such a reaction from someone who has devoted much of their life to studying this field of science.
So it is the intent of this INSIGHT that I would like to examine why I tend to feel that the lyophilization process can at times appear to have a personality that is deceptive and devious in nature. The objective of the INSIGHT is not to cast the process in a negative light but alert the unwary that what may appear to be a simple process is one which is highly complex and why there is such a need for more research in this field to guarantee what we did today we can repeat tomorrow.
Deceptive: The word deceptive implies that one is mislead by the process. If one has a process that has worked a numerous number of times without any problem then why would it be viewed as being deceptive? I would have to agree that if one knew why the process was successful each time and also knew under what operating conditions it would not be successful, then I would be the first to say that the lyophilization process is not deceptive for no one would be mislead. So it would be quite natural for one to ask the question, how can the lyophilization process be deceptive? Let me answer that question with a hypothetical example of when the process could be deceptive.
An aqueous formulation is developed in which a large molecular weight protein is found to be stable at 4 8 oC for several weeks. One hundred vials of the formulation were placed in a research and development freeze dryer. A 1.0 mL fill volume of the formulation was frozen to -40 oC and allowed to remain at that temperature for 2 hours. The condenser surfaces were then chilled to -60 oC and the drying chamber evacuated to 13.3 Pa (100 mTorr). The refrigeration to the shelves was turned off and the temperature of the shelves was warmed heat generated by the circulating pump. When the pressure in the chamber was equal or lower than 6.6 Pa (50 mTorr) and the product temperature had maintained 22 C +/- 2 C for no less than 4 hours, the vials were then stoppered and thechamber was backfilled to one atmosphere. The product was removed and the activity of the protein was found to be within acceptable limits. Results of a stress test at 40 C and relative humidity of 50% for 30 days showed that the loss in protein activity was not significant.
Similar results were found when the product was produced with the same process in a larger pilot dryer. The process was repeated using a production dryer and once again the product was found to be acceptable. Real time stability studies showed that the product was stable for more than 2 years at 25 oC +/- 2 oC.
So how could such a simple process that appears to work in just about any type of freeze dryer be devious or misleading in nature? It can be devious and misleading because there is no understanding or rational for the lyophilization process. Especially, when processes ran in the research dryer, the pilot dryer and production dryers were all significantly different during the freezing process and drying process.
Now let us assume that in order to increase production, the older production dryer is replaced with an even larger freeze dryer. As you might have expected the process in this new dryer will also differ from any of the previous processes; however, the product produced is found to be not acceptable. Now what had seemed to be an easy lyophilization process has now turned into a nightmare and one has no rational understanding as to how to go about correcting the problem. Had the process failed during the research and development stage, one would be forced to take a scientific approach in solving the problem such as understanding the thermal properties of the formulation [1]. Instead, one was ensnared into a false sense of security in thinking that the process was transferable to any freeze dryer only to find that such a premise was wrong. The main point that I wish to make is that the process will appear to be deceptive so long as we continue to look for an easy means to manage a complex process. So as long as we fall victim to the question ?Why don?t you give it a try?? then the lyophilization process will appear at times to be deceptive by giving us a false sense of security.
Devious: While the latter example led us to make an erroneous premise regarding the process we finally did recognize the error of our ways; however, it is when the lyophilization process proves to be devious in nature that one pays the real price for having used trial and error instead of science and technology in its development. To illustrate an example of its devious nature, and there are many examples, consider the following hypothetical example.
Let us take a formulation that consists of a very small amount of protein, 10 mM of sucrose, 10 mM of a pH modifier and 2.0% of a bulking agent. The pH of the final solution was adjusted using either 1.0 N HCl or 1.0 N NaOH to 8.5 +/- 0.1. The formulation was prepared in accordance with a protocol.
Unlike the first example, the lyophilization process for this formulation was carefully documented so that it could be transferred from dryer to dryer. The shelf-surface temperature was used instead of the fluid temperature (see INSIGHT Vol. 1 No. 7), the chamber pressure was carefully controlled and measured by a capacitance manometer gauge (see INSIGHT Vol. 1 No. 6) and each step of the process was controlled so that product temperature was always within well defined limits. As a result, there was close agreement between the drying processes obtained in the research and development unit, the pilot dryer and the production dryers. Everyone was pleased that the product produced in the production dryer was indistinguishable to that obtained during the development of the product.
Needless to say everything was going along just fine and then there was a batch where the entire final product showed obvious signs of collapse. The dryer was checked and so was the process controller and data collection system. Every component was found to be working within its specifications. The next batch was successful and so was the next batch. In fact there may be five or six batches processed and the next one would fail.
Management wanted to know what was going on and to find the problem and fix it. It was only after a comparison of the thermal properties of the reconstituted acceptable product was compared to that of the failed product did they discover the problem. The collapse temperature of the good product was about -35 oC while for the rejected batch it was about -52 oC. The question was do they need to perform a thermal analysis of the formulation prior to starting the manufacturing process? That analysis would take too much time and violate the production protocol. And if they did find the formulation with the lower collapse temperature would they discard the formulation and waste valuable protein material and fall behind in their production schedule?
It was only after a review of the protocol for the preparation of the formulation did they discover the real source of the problem, i.e. the pH adjustment. The addition of NaOH did not have an affect on the composition of the formulation; however, when using HCl one produced a small quantity of NaCl which lead to the reduction in the collapse temperature to -52 oC. Unfortunately, the protocol only called for the adjustment of the pH to 8.5 +/- 0.1 and no one thought to record the quantity of NaOH or HCl that was added. For even from that alone they would have immediately realized that the use of HCl was affecting the lyophilization process.
So what is the answer to the question - is lyophilization deceptive and devious at times? Or is it really our hurry to get things done without taking the necessary time to fully understand the complete process which must include the preparation of the formulation? In my opinion it is a bit of both. If the lyophilization process forced us to use science and technology in order to use the process then in a sense it would lead to greater confidence and less frustration.
Reference:
1. T. A. Jennings, Lyophilization - Introduction and Basic Principles, Interpharm Press,
Buffalo Grove, IL 1999.