Recombinant and Urinary Gonadotropins: Comparison of Efficacy, Safety and Consistency
presented by Scott C. Chappel, PhD
print this page
access the printable interview
transcript or curriculum vitae
Gonadotropins have been the cornerstone of infertility treatment for quite a while Ė 40 years with the urinary versions of gonadotropins, and then the last ten years with recombinants.
What I will be speaking about are these different FSH preparations, the urinary and recombinant and evaluate with you their efficacy, their safety in terms of the way in which they are manufactured, and their consistency.
First, Iíd like to consider efficacy.
Urinary gonadotropins have been available through Serono, Organon, and Ferring for the last 40 years. Clearly, with that kind of life span theyíve been considered to quite efficacious. Iím proud to have been part of a team that was involved in the development of recombinant
gonadotropins, and we used the urinary gonadotropins to judge how effective we were in producing recombinants. Indeed, we were pretty effective, and about ten years ago we launched recombinant gonadotropins. They too have been shown to be quite effective in their ability to stimulate follicular development.
During the last ten years there have been quite a few of publications on comparing and contrasting urinary and recombinant versions of FSH. This has been on the preclinical side looking at the biochemistry of these molecules; their in vitro and in vivo activity, and their pharmacokinetics in non-human primates. Further, weíve looked at the clinical performance of urinary and recombinant FSH. Thatís what weíre going to do today, look at selected publications in the literature. Obviously there are many more than we can present in this short time period, but I have selected some to present to you.
The way in which Iím doing this is down at the bottom of the slide is the actual reference. Up at the top is the topic that Iíd like to address. This is the title of the presentation, and this is my interpretation of the key findings directed out of the abstract or the discussion.
So, the first is a publication that came from the Organon group when recombinants and purified urinary versions of FSH were first available. Looking at the biochemistry of these molecules, and you can see that from this publication they determined that the isoform distribution profile of recombinant and urinary gonadotropins, were surprisingly similar.
The second presentation comes from the Serono group, again looking at carbohydrates and biological activity. They conclude that both recombinant and urinary versions of FSH have similar oligosaccharide, the carbohydrate moedies, on the alpha and beta sub unit. And similarity in vitro and in vivo biological activity for the recombinant and the urinary.
This is another publication from the Organon group; again looking at the comparison between recombinant and urinary, and now neutralizing, or normalizing, the amount of FSH activity thatís being measured in in vitro and in vivo bio-assays. And showing that when equivalent amounts of immunoreactivity of FSH were tested, the recombinant, urinary and pituitary FSH displayed comparable values in radioreceptor assays, which means binding to the FSH receptor on granulosis cells or sertoey cells and the in vitro biologic activity.
Moving onto pharmacokinetics, this is a paper, again by Serono, looking at the plasma half-life clearance and distribution half-life of recombinant, versus urinary, FSH in the non-human primate, the monkey. Their conclusions were that the clearance, the distribution half-life and terminal half-lives, for recombinant and urinary FSH, were similar.
So in all of these preclinical studies it seems that by and large, although there are some publications that show some differences, for instance in isoelectric focusing, that in terms of carbohydrate structure, isoelectric focusing, and in vitro and in vivo activity, preclinically, these compounds were quite similar.
Moving on to clinical performance, again there is a huge body of literature testing the effects of recombinant and urinary FSH. This is one of the first publications from the Serono group looking at their recombinant FSH, and comparing it to highly purified urinary versions of FSH, and seeing that the clinical pregnancy rates, when measured by two different ways, were not significantly different between urinary and recombinant.
This is supporting an earlier publication from the same group, thatís Serono, again, showing that the recombinant FSH was as safe and efficacious as urinary versions of FSH and stimulating follicular maturation.
Not all publications show similarities between recombinant and urinary. This is a publication from the Organon group looking at their version of recombinant FSH, and showing that when compared with purified urinary versions of FSH, their recombinant showed greater pregnancy rates.
A more recent study looking at Bravelle, sponsored by Ferring, showed that when compared to the Organon, Puregon and Bravelle, they found comparability in efficacy and safety between recombinant FSH and urinary versions of FSH.
In an attempt to try and collate all of this clinical data, meta analyses have been performed and this is a study that was prepared by Dr. Daya and his colleagues in 2002. It looked at the overall performance of recombinant and urinary versions of FSH, in terms of their clinical performance, and concluded that recombinant versions of FSH showed higher pregnancy rates, and required lower dose.
Another group had also performed a meta analyses looking at the differences between recombinant and urinary, and concluded that there was no evidence of clinical superiority between these two groups.
I present these reports in the literature to suggest to you that although there have been some differences that have been reported, there clearly is no consensus on superiority of one, or the other, of the compounds in terms of their in vitro ability to exhibit FSH-like activities. Or their clinical performance either in oocyte number, embryo quality, clinical pregnancy rates. Or on the safety side, rates of ovarian hyper-stimulation. So the conclusion to be drawn from this is that both preparations are equally effective.
Next, Iíd like to talk about the safety. And Iím not talking now about the clinical safety. As I showed you, several reports talked about the fact that recombinant and urinary versions of FSH had similar safety profiles clinically. I now want to discuss the production of urinary or recombinant versions of FSH, and safety associated with the manufacture of those preparations.
When we talk about the production process, either for the urinary or recombinant, weíre talking about two general phases. One is the way in which bulk material is collected, and the second is how that bulk material is prepared and analyzed to assure safety, reproducibility and efficacy.
Clearly, bulk material is collected in two very different ways between recombinant and urinary versions. However, if we look at the way in which the crude material is processed, I think youíll see that they are processed in quite a similar fashion.
First letís talk about the urinary version of FSH, and how bulk material is collected.
Bulk material for the urinary FSH preparation is collected from thousands of liters of post-menopausal urine. Clearly this immediately elicits some consistent and rational concerns about safety. Those concerns can be due to the fact that products obtained from human tissue could be infected with viruses or prions. There are certainly human proteins that are contaminating in the urine, present in human urinary proteins. The urine can be contaminated with bacteria, which would release endotoxins, and need to be removed, and many other adventitious agents that we need to be concerned about. This requires constant surveillance of the manufacturing process to oversee individual batches and ensure quality and safety.
To that end, as far as the Ferring product is concerned, the collection of raw material occurs only in safe countries. The idea of safe countries would mean countries that have no incidence of prion disease, such as Jacob Creutzfeld Disease and a low HIV rate. That material is collected, sent to a central processing facility where a resin is added, a manufacturing resin is added. All of the gonadotropins and other proteins are adsorbed onto that resin. The resin is washed. Itís eluded in a high salt buffer, the gonadotropins and other proteins are eluded off from the resin. The eluate is then dialyzed to remove the high salt and itís tested for its initial FSH activity.
Two percent of the total protein in the bulk material is FSH. An eight-step process takes place using conventional chromatographic methods with viral removal steps, to remove viral load, as well as viral inactivation steps, and then of course, removal of other human urinary proteins. The final product, which is greater than 95% pure, is then tested for its purity, absence of contaminating proteins, absence of detectable levels of viruses, presence of degradation products, characterization of the FSH in terms of amino acid content, carbohydrate isoelectric focusing, and of course, biologic activity.
The FDA has reviewed this process of collection of bulk and processing and purification of the final product, and found it to be safe, effective and reproducible.
Letís talk now about the way in which recombinant gonadotropins are produced. Obviously this is now looking at the way in which bulk material is produced by cell banks that have been genetically engineered to express human recombinant FSH.
So before we talk about the safety of the bulk, we need to talk about how the bulk material is actually generated, and therefore how a cell line is produced. And how that cell line is selected and expanded. And how we create a Master Cell Bank, a Working Cell Bank, and then the production process itself.
First the genes need to be cloned that encode the alpha and beta sub unit, and then those genes are engineered for expression in a mammalian cell line. That cell line is a Chinese hamster ovarian cell line. In the case of Serono we amplified the genes for the alpha and beta sub unit. Thatís inserting one copy of the gene into the Chinese hamster ovarian cell, but then using molecular biology processes to create hundreds of copies of the gene integrated within the host cell genome.
And this is a schematic of how thatís done. We have the expression vectors for the alpha and beta sub units of FSH, and they are inserted into the Chinese hamster ovarian cell. They find their way to the nucleus where they integrate into the genome, and into a particular chromosomal site in the Chinese hamster ovarian cell, which regulates then the expression of a transcription of messenger RNA. This in turn goes down in to the cytoplasm to direct protein production of the alpha protein and beta protein, which heterodimerize and then are glycosylated, and then secreted into the culture media in the form of the mature FSH molecule.
This is a chromosomal spread of Chinese hamster ovarian cells, looking at fluorescence in situ hybridization showing that thereís not just one copy in this case of the alpha sub unit gene, but hundreds of copies of the alpha sub unit gene thatís been incorporated into the host cell genome.
Following amplification of the gene for the alpha and beta sub unit, we identify the highest expressing cell line and then expand that cell line to create a clonal population of CHO cells that are expressing human recombinant FSH.
Here are some pictures of how that process occurs. We start off with, in this case, a 96 well plate looking at very small numbers of cells that express FSH. We collect those cells and put them into a T25 flask where they are growing in culture on the inside of the plastic dish. We are looking at expansion of those cells, the replication of those cells.
Following achievement of confluence in the T75 flask, we move to roller bottles, and in this case we look atÖthese are a bank of roller bottles, that are turning at 37ļ.
The CHO cells are growing on the inside of the plastic of the bottles, and again, weíre looking to expand this population of clonal FSH secreting cells.
When we reach the number of cells that we want, then we create a Master Cell Bank and a Working Cell Bank. The Master Cell Bank is frozen down and it becomes the lifetime supply of the cells that will be used in all subsequent bio-reactor runs. The Working Cell Bank is taken from the Master Cell Bank and that becomes the actual cells that will be used for the production of FSH.
The Master Cell Bank is obtained from the roller bottles. We treat the roller bottle cells with certain enzymes to release them from the plastic and to create single-cell suspensions. They are somewhat expanded, they are fed and then they are suspended in dimethyl sulfoxide, glycerine, and fetal bovine serum. A slow, constant freezing process takes place to get them frozen, ultimately to -70ļ C. And, as I said, this is the bank upon which all subsequent production runs will occur.
So, when weíre filling by bioassay, when weíre using the same bioassay to fill urinary or recombinant products, and weíre using the same international standard to quantify the amount of activity by definition, when we look at 75 IUs of one product of another, they should be equivalent. In fact the only variation can be the variation in the bioassay itself from lab to lab, and believe me there is quite a bit of variation between labs, or even within a lab. There have been studies showing that analyzing the same recombinant or urinary version of FSH in the same lab, using the same bioassay, and doing it 50 times, shows that there is obviously error in measuring. This is a biologic assay; the error can be as great as 20%.
However, when we talk about variability in gonadotropins and compare that to the whole process that you all are involved with from wanting to have a baby and bringing home a baby, and all the steps in between, 20% variation in the amount of FSH, in a vial, in my opinion, is relatively insignificant. When we look at patient to patient variation, or cycle to cycle variation within a particular patient; ways in which different laboratories handle eggs, sperm, embryos, and the synchronization of the readiness of the embryo to implant, and the receptivity of the endometrium, the conclusion, as far as Iím concerned, is that there has been no report in the literature that has compared consistency between urinary and recombinant FSH, and shown that some difference affects clinical outcome.
Again, this is a picture of a freezer box and pulling out a vial or two of the Master Cell Bank, and then initiating a new culture as we have done to create the Master Cell Bank.
Transfer of those cells to beads. The beads are put in a bioreactor and a production run occurs, usually a continuous harvest. When we then collect the condition media, we need to filter the condition media to remove debris and dead cells that have leeched off the beads during the production run.
This is a photomicrograph of the beads with Chinese hamster cells growing on the surface of the beads.
Those beads are inserted into, in this case, a stainless steel bioreactor. This bioreactor is about four feet high. On the top we see ports for entry of gasses and substrates that are required to feed the cells, and this is a port for the retrieval of the condition media.
So now weíve gone through the process from engineering a cell to the production of the bulk material. And again, similar to the way in which urinary gonadotropins are produced, there are certain safety concerns that need to be addressed within this process that Iíve just described.
And that includes looking at the safety of the host cell, as in any recombinant process, especially ones that use mammalian cells as hosts. We need to be worried about the fact that these mammalian cells can be targets for infections by bovine, porcine or rodent viruses, as well as retro-viruses. And as well, other adventitious agents including prions. These cell lines need to be tested and qualified beforehand to make sure that they are not harboring any viral disease.
Iíve mentioned to you that weíve amplified the cells. We amplified the genes in the cells. Whenever weíve made an amplification process where youíre creating hundreds of copies of genes within a cell, weíre worried about mutagenesis. That can affect genetic stability of the cell, which could affect metabolism of the cell, the performance of the cell, and we want to make sure that the performance remains at a high level so as not to affect the authenticity of the recombinant FSH thatís being directed for manufacture in that cell type.
In the process of creating the Master Cell Bank and then using the Master Cell Bank to produce FSH, weíre worried because of the fact that this is a sterile process and infectious agents are always a concern. Each step along the way we need to make sure that we maintain sterility and not have any contamination. Every step along the way these cells have been exposed to fetal bovine serum, and theyíve been exposed to enzymes, such as trypsin and collagenates that have been obtained from animal material. We need to make sure that the production runs occur in a very constant fashion. If we run the bioreactors slightly differently than we did the previous time, a change in temperature, a change in the way gases or substrates are entering into the cell, we can affect the performance of those cells, which ultimately could affect the product.
Finally, weíre concerned about the handling of the production bulk and as I mentioned to you before, in the process of production, some of the cells die and they release their cellular contents including lysosomal enzymes. Those enzymes can chew on proteins in the condition media. And one of those proteins is FSH. So we need to handle the bulk material in the same way every time to make sure that we do not have digestion of the expressed protein.
These too are constant and rational concerns in the production process, and they are addressed by well controlled processes under standard operating procedures, SOPs, each step along the way. All of the products that are exposed to the cell in terms of animal products are sourced from safe countries, and all of the tests that are used at each step along the way are highly validated to ensure safety, quality and reproducibility.
The purification of FSH as with the urinary version starts with about 2%, hopefully achieving greater than 95% purity, in the case of recombinant gonadotropins, routinely at least 97%. Itís a six-step process including an immunoaffinity step, and that means that a murine monoclonal antibody column is made to grab onto the human recombinant FSH as part of the purification process. In that process we have viral inactivation and removal steps, host cell proteins, in this case Chinese hamster ovary cells. Those proteins need to be removed, as well as those proteins present in bovine serum, and any murine antibodies that have leeched off the chromatography column. They all need to be removed.
Furthermore we need to have assays to show that host cell DNA is removed. That the Chinese hamster DNA is removed. And similar to other processes that need to be sterile, we want to make sure there are no endotoxins or pyrogens. The protein is purified and checked for purity, as well as any degradation products that may have occurred in the processing. And then the final product is characterized in terms of its amino acid content, carbohydrate because carbohydrates are required for biologic activity, isoelectric focusing profile, and of course, biologic activity.
The FDA has evaluated this process from production of the bulk, and purification of FSH, and found that it too is safe, effective and reproducible.
So, in terms of safety, weíre looking at urinary versus recombinant. We see that the same concerns appear for both in terms of making sure that the product is free of viral contaminants, free of prions, and that all the host cell proteins are removed, either human or hamster. Host cell proteins are removed and the challenge for the scientists in the laboratory are to purify this material from 2% starting, to greater than 95% final product.
Both products are considered to be produced under very established, and very highly controlled, processes. Safety concerns for both; we went over both of those. The urinary concerns, the starting material coming from human recombinant aspects of the production process. Rigorous purification methods for both. In-process controls. Batch testing to assure quality and safety. The conclusion is that both of these manufacturing processes are safe.
Finally, letís talk about consistency. This has been a long-standing concern in the world of gonadotropins.
Concern about batch to batch variation, and this is a rational concern due to the fact that urinary gonadotropins are produced from donor pools, and those donor pools are constantly changing. Furthermore, from purification run to purification run there can be differences in the performance of the promedigraphic procedures. And recombinants, I mentioned to you before, bioreactor runs could change from run to run, which also affects consistency, and the performance of the purification process may not be optimal from run to run. That needs to be a concern.
So there have been attempts in the past to assess consistency within a product, or between products. In my opinion this is a very difficult, or impossible goal to achieve. And thatís because up until now, all of these gonadotropin products are filled by bioassay. What does that mean? It negates the difference to see a difference because of the way in which the vial is filled; itís filled until an activity is reached. What do I mean?
I mentioned to you that when bulk material is prepared, either urinary or recombinant, it is measured for activity. And itís measured for activity using the Steelman Pohley assay, thatís an in vivo rat bioassay looking for ovarian weight gain. Itís filled according to a bioactivity. What that means is, once a bioactivity estimate is obtained, the operator gets the relationship between the volume of FSH, and the final product and the activity, and just adds a volume until he or she achieves 75 IU. So theoretically, if a less potent molecule had been generated by the process, one would just add more volume until you achieve 75 IU of activity.
So, when we are filling by bioassay, when weíre using the same bioassay to fill urinary or recombinant products, and weíre using the same international standard to quantify the amount of activity, by definition when we look at 75 IUs of one product or another, they should be equivalent.
In fact the only variation can be the variation in the bioassay itself, from lab to lab, and believe me there is quite a bit of variation between labs, or even within a lab and there have been studies showing that analyzing the same recombinant or urinary version FSH in the same lab, using the same bioassay and doing it 50 times, shows that there is obviously error in measuring. This is a biologic assay, the error can be as great as 20%.
However, when we talk about variability in gonadotropins, and compare that to the whole process that you all are involved with from wanting to have a baby, and bringing home a baby, and all the steps in between, 20% variation in the amount of FSH in a vial, in my opinion, is relatively insignificant. When we look at patient to patient variation, or cycle to cycle variation within a particular patient, ways in which different laboratories handle eggs, sperm, embryos, and the synchronization of the readiness of the embryo to implant, and the receptivity of the endometrium, the conclusion, as far as Iím concerned, is that there has been no report in the literature that has compared consistency between urinary and recombinant FSH and shown that some difference affects clinical outcome.
More recently, several studies have been reported trying to get away from the Steelman Pohley assay, the in vivo bioassay, and fill by mass, which has definite advantages to get away from the spread of variability in the in vivo bioassay. This is a paper that was published two years ago. Only comparing gonadotropins hasnít been done with urinary gonadotropins, this is recombinant FSH that had been filled by mass, as shown in this column, and filled by the Steelman Pohley bioassay, and then clinical evaluation takes place looking at all these clinical end points, including overall pregnancy rate. As you can see, there is no significant difference in the performance when we fill by mass, or fill by bioassay. And in my opinion, that supports the suggestion that differences in the laboratory or in the patient, are much greater than any differences that may occur in the filling of the vials of recombinant or urinary FSH.
So, in summary, what I presented to you today were concepts about efficacy, safety and consistency of urinary FSH, and recombinant FSH. In terms of efficacy, we talked about the fact that while there are some differences, in general I think most would conclude that there is no consensus of superiority, therefore these compounds are equivalent. In terms of safety, there are real and rational concerns about safety of both of these compounds, and they are addressed by the companies that manufacture and produce these compounds for your use. In terms of consistency, in my opinion, I have not been able to find a report in the literature that shows a difference in consistency between urinary versions of FSH, and recombinant versions of FSH, which affect the clinical outcome. That suggests to me that in terms of efficacy, safety and consistency, these preparations are equivalent.
Thank you very much.