|
Human Menopausal Gonadotropin Supplied
Luteinizing Hormone Activity - Helpful, Harmful or
Necessary?
presented by Richard
P. Marrs, MD
print this page
access the printable video
introduction transcript or curriculum vitae
Dr. Chappel really showed you the evolution of not only the urinary gonadotropins and purification process, but also the probably more complex process of recombinant gonadotropin production. All of us in this room have been faced with choices over the last ten years, more so than the years prior to that, and what I am going to be talking about today is how we make some choices in clinical situations. As you know, a lot of you in this room were around, twenty years ago, when we started talking about stimulation cycles for ART.
In those days, the early ‘80s, we really had only single-agent therapy. In this country, it was the clomiphene group and the hMG group and both of those types of stimulation were adequate for what we expected at that time, which was not very much. In the late ‘80s, the GnRH agonists came about which made all of our lives easier. We stopped chasing the elusive LH surge, either in the urine or in the serum, we were able to time our oocyte recovery easier and it made life better in general. We will talk about how you never get anything for free.
The agonists did make life easier, but they added some issues and questions that we are still trying to answer today. About the same time, the initial purifications of the gonadotropins started giving us more purified products and FSH-enriched products which then turned into what Dr. Chappel was talking about as far as recombinant FSH in the early ‘90s. About the same time, the purification of the urinary products continued and improved so that we had choices between urinary and recombinant FSH. At the turn of the century, we saw the use of antagonists and today probably all of you are using GnRH antagonists maybe more than you thought you would. Again, it adds a perspective to our stimulation choices and a difference to the endocrine milieu in which we are stimulating.
These two products, whether you talk about agonists or antagonists, I think it changed how we have to think about follicular growth and development in our clinical situations. I am going to give you some information that is relatively new on even more highly-purified hMG products that have recently been reported from some European studies and how that may impact on our stimulation choices in the future. Again, we all grew up in our training years and thereafter with this type of concept as far as follicle development in the human.
As I was saying, we began our training years by looking at multiple follicle development based on some theories. Gary Hodgins was one of the main proponents as far as identifying how single follicle development occurred in the human. For the most part, we looked at the menstrual cycle and the FSH pulsitility and amplitude of FSH as the determinant of follicle selection and ultimate follicle growth for single follicle selection in the ovary. As dominance occurred, FSH pulsitility or amplitude of pulsitility went down and the single follicle was matured and the remainder of the cohort went through atresia. In this process, LH pulsitility was really not looked at except for the LH surge as a final maturation event. We jumped on to the single follicle theory and began our use of gonadotropins as a way to override and produce multiple follicles. Again, with the idea that FSH being the primary gonadotropin that was necessary for multiple follicle growth and oocyte maturity.
I think that is still consistent today. The question is: is there a place or is there a need for LH during this process? That is mainly what we are going to be looking at for the rest of this talk. I just want to mention for a minute, expanding our thought process, before the onset of menses, those of you who have had a fair amount of experience with antagonists cycles now may be experiencing issues of follicle synchronization, follicle growth, multiple follicle development in patients who may be having changes in FSH pulsitility prior to the onset of menses which decreases your ability to produce multiple follicles or override the dominant follicle issue. I think there is more and more data coming into the literature that is looking at ways to affect the FSH pulsitility prior to the onset of menses and prior to the onset of stimulation for GnRH antagonist cycles. This is an area that, again, I think is of interest.
We are going to focus on whether LH is necessary. Is there a need or is there a reason for LH activity that is either added in the form of LH or LH activity that comes in the form of hMG during controlled ovarian stimulation.
Some of the facts that we can all agree upon were developed and published. Hillier and his group clearly showed in the early ‘90s that follicle growth is an FSH-dependent process. We got away to some extent from the two-cell theory in the early ‘90s thinking that we could drive follicle growth easily with FSH and the more purified FSHs we were seeing, the recombinant FSHs, they certainly pushed our follicles into growth and maturity. That same group showed that the follicle really does not produce or have active LH receptors prior to the mid-follicular phase. If you are looking at follicles that are somewhere between 10mm and 12mm diameter on ultrasound, that is when we expect or that is when they showed in both animal and human models that LH receptor activity was present.
Chappel and Howles at about the same time were showing that endogenous LH levels, even in the face of GnRH agonist down-regulation activity still remained, usually within a range they found to be consistent with supporting follicle growth, theca cell activity, estradiol production and so on. As long as LH activity was in that range of somewhere between 1.5 and 10 IU per litre, successful growth and estradiol patterns were normal. However, as we have seen in the past few years, prolonged suppression with either GnRH agonists or more recently, suppression with GnRH antagonists below that 1.5 IU per ml, may affect the ability of follicle growth and estradiol pattern production and so on.
Fleming in the late ‘90s was one of the first people to show that with GnRH in down-regulated cycles, LH activity may be necessary or could change the estradiol production pattern. It is interesting that as we are looking at LH this morning, I think we have to recognize some interesting data that was published two years ago for the first time by Filicori and his group which looked at hCG addition for follicle growth and maturity. In fact, not only did he show that the addition of hCG on a daily basis to an FSH stimulation protocol did produce follicle growth, maybe we can agree that LH activity is probably something we want to have around during controlled ovarian stimulation.
The question then becomes, when is it really needed? Should it begin with the start of stimulation, should it start at some time during stimulation? The next question is, how much do we really need to make a follicle and oocyte competent?
This is Fleming’s data trying to answer the question of ‘when LH is needed.’ It is one of the studies that actually showed that in patients who were down-regulated with the gonadotropin-releasing hormone agonist, Lupron, excuse me, these are Buserlin cycles, stimulation occurred with either hMG alone, standard hMG product or with purified FSH with less than one IU per litre of LH activity which is the yellow line, and FSH purified gonadotropin FSH urinary gonadotropin with greater than one IU per litre of LH activity. As you can see from day one of stimulation, the estradiol pattern, which you see in this graph, was not really significantly different up until about the fifth day of stimulation. At that time, there was a divergence of the estradiol production curve in the hMG stimulated patients.
These were not patients who were profoundly suppressed, they were down-regulated and initiated their stimulation right after down-regulation occurred. That significant difference continued on until the time of hCG administration. It is interesting when you look at the fifth day of stimulation. If you think about what you are seeing, if you did an ultrasound on day five of stimulation of your patient, you are going to be seeing follicles, usually around the 10mm to 12mm size as far as follicle diameter, which is consistent, it goes back to the basic science data we received from a number of people – Ericson, Chappel, and those folks – about when LH activity or LH receptor activity started in the developing follicle.
It would seem from this information that maybe the fifth day of stimulation would certainly be early enough if you wanted to add LH activity at that time.
This is a more recent study that was published earlier this year from, again, a prospective randomized clinical trial with patients who were down-regulated with GnRH agonists. Stimulation was performed with three different ratios of urinary FSH to LH in these three patient groups. The triangle or yellow bar was a standard 1-to-1 FSH to LH or standard hMG-type FSH / LH activity. The squares or light blue bar was the add-on protocol which meant the first 5 days of stimulation were with urinary FSH only, then beginning on the fifth day the hMG was added on with a 3 to 1 ratio of FSH to LH. The circles or magenta bar were a constant low does of hMG starting with a 3 to 1 ratio of urinary FSH to LH; after the fifth day, the FSH could be increased but the hMG remained constant.
Again, what we see is similar to what was shown to you on the prior slide, on the fifth day there began to be a significant difference seen in 17 beta estradiol production in these three groups. The one with the balanced FSH / LH ratio had a significant increase in estradiol production just about on a day-to-day basis up until the time of hCG administration. Again, these patients were not profoundly suppressed with agonists. They were down-regulated and then went directly into their stimulation protocol. So this is all suggesting that certainly by the fifth day of follicle growth or when LH receptor activity is present, when there is an LH presence in the stimulation, there may be a difference in estradiol production patterns in those patients compared to patients who have FSH alone or very low amounts of LH activity.
That is important, most people might believe that the estradiol pattern will influence the endometrial receptivity, the endometrial synchrony and so on. But there is also some recent information in several studies in the literature now that are suggesting that the 17 beta estradiol production in multiple follicle development may actually have a direct influence on oocyte capabilities as far as functioning of the oocyte as far as genomic imprinting and so on. So there may even be a correlation between 17 beta estradiol production, the quality of eggs and the quality of the embryo.
Going back to the very beginning of the use of the antagonist, this information to me is quite interesting in the conversation we are having today. This is the Ganirelix dose-finding study which was published in 1998 and it is interesting because of what was seen in the dose-finding study. The use of the GnRH antagonist, the low-dose up to the highest dose used, showed a direct correlation with increasing dose of antagonist with a decreasing level of LH seen in the serum in these patients who were stimulated with recombinant FSH alone. So antagonists used beginning on the sixth day of stimulation which was the protocol followed in this study showed a profound suppression of LH above the .25mg dosage of the antagonist.
Consistent with the decline in LH activity was a significant and profound decline in implantation rates, as well as on-going pregnancy rates compared to the patients who received .25mg or less. Is this an effect on the endometrium from the GnRH antagonist? That was the hypothesis and conclusion they took from this data. My question is, what happened with the estradiol pattern? Maybe there is a combination effect here as far as decreasing LH in the serum, during stimulation not only on the endometrium but maybe on the oocyte itself.
If we need to support LH, it appears that starting somewhere at least by the fifth day of stimulation as can be extrapolated by the data I am presenting, the question then becomes, how much do we really need?
This is an attempt to answer. This is a study that was published by Gordon and her group several years ago. Again, this is a prospective randomized trial. These were all patients who were down-regulated with gonadotropin-releasing hormone agonist. They then were randomized to either recombinant FSH alone which had no LH activity or a highly-purified urinary FSH with one IU per litre of LH activity compared with a purified hMG which was basically a 75 IU of FSH and 25 IU of LH activity and then a standard hMG 75 / 75 urinary product. You can see there is no difference in any of the four categories as far as follicle numbers produced, oocytes produced. Interestingly enough, when they compared implantation rates, the implantation rate with the highest amount of LH activity group had a significantly higher implantation rate compared with the other three groups. The live births per cycle did not reach significance based on these group sizes.
This alludes to the fact that, again, there may be a minimum amount of LH activity that is necessary. This definitely does not answer the question but it certainly leaves us to think that there is perhaps a minimum and the question then becomes, what is the true minimum and is there a maximum of LH activity we can possibly add that would then not be productive. I do not have the answer to that question today.
Another interesting finding, again from a study that was published by Bill Schoolcraft and his group in 1999, and the main reason for this publication was to look at blastocyst rate and blastocyst production in their IVF cycles. This data was a secondary finding that I thought was quite interesting. This was a fairly young group of patients they reported. This was not a prospective study, this was a retrospective analysis of cycles where they were looking at blast rate and blast implantation rates in these patients. When you break down their study group and look at outcome based on stimulation, they used pure FSH, again these were GnRH agonist down-regulated patients and they were either stimulated with urinary FSH, highly purified FSH or FSH and hMG in combination.
Similarities in age, number of eggs, fertilization. The blastocyst percentage formation was no different between the two stimulation groups and they transferred an equal number of embryos in both groups. However, when implantation rates were compared, the pure FSH group had a significantly lower implantation rate than the FSH / hMG combination. That also showed a significant difference in on-going delivered pregnancy rates. Retrospective data is certainly difficult to use as a very strong point, but when you look at this in relation to what we saw from the Ganirelix dose-finding study and other pieces of information in literature that I do not have time to show, again it is showing either an impact on the egg, the embryo or the endometrium or all of the above. The question is, where is LH activity affecting or is it affecting one more than the other?
This is the last bit of data we are going to look at and this is from a very large study performed in Europe and Israel looking at a highly-purified hMG product. We do not have this product in the US so I cannot tell you that I have had any experience with it. The highly-purified hMG is basically similar to, from my understanding, highly-purified urinary FSH. In other words, hMG has been purified to the point that the extraneous or extra proteins that are found commonly in urinary hMG products have been removed, so this is a highly-purified FSH / LH protein combination they were using. They did a prospective clinical trial on a very large number of patients, comparing the highly-purified hMG with standard recombinant FSH. The initial publication was a couple years ago and they really found no differences in the number of oocytes, fertilization rates and so on, and on-going pregnancy rates were similar in the two groups and they concluded from this study that at least there was an efficacy and equivalent between this highly-purified hMG and recombinant FSH in this large prospective comparison.
Of interest from this study was what came two years later. This same data set was re-reviewed or re-analyzed by Platteau and his group. They took the same information from the prospective trial and then broke it down into the two groups as far as the type of ART procedure the patients had, whether they underwent a standard IVF procedure or they had inter-cytoplasmic sperm injection (ICSI). This is looking at the data broken down as to the type of stimulation and ICSI outcome. Again, they found no differences in number of eggs, fertilization, bottom line implantation and on-going pregnancy rates in the ICSI population was no different from what they saw in the whole study population combined.
However, when they looked at the IVF patients, there were some differences that were seen. First of all, in the recombinant FSH stimulation, more oocytes per patient cycle were collected in the recombinant FSH compared to the hMG. Fertilization was still the same. They kept consistent embryo transfer numbers. There was a trend toward significance in their implantation rates now with their highly-purified hMG compared to the recombinant FSH. They did find a significant difference in on-going delivered pregnancy rates with the highly-purified hMG compared to FSH.
They then looked at LH activity. This is day six of stimulation. The yellow bars are the hMG product, the blue are recombinant FSH and no matter how they broke down their serum LH levels on day six, they found no differences in on-going pregnancy rates between the two groups, whether they looked at the entire group in general or they looked at the ICSI patients or the IVF patients separately. So serum LH levels did not distinguish anything between the comparisons.
Interestingly enough, when they measured hCG levels in the serum, they did find some differences. This is the serum hCG levels in the IVF group and the ICSI group. They looked at hCG of less than one IU per litre and then one to less then two and then greater than two. In the IVF population, there were significant differences among these three levels. As hCG increased in serum on day six of stimulation, they found a statistically increasing rate of on-going and delivered pregnancies. The ICSI population showed differences between the lowest and the highest but it was not significant in the mid-range group. They concluded that maybe the hCG activity present and found in the hMG product was having an effect either again on maturation on the endometrium which then led to the improved implantation rate with IVF and so on. It again raised the question of LH activity in stimulation and how it affects outcome.
I think what we can safely assume at this point, based on what I have shown you and a whole lot more literature available but which I have not shown you this morning, is that GnRH antagonists certainly have been shown to have a direct dose-dependent effect on LH levels and LH levels during stimulation do have an impact on cycle outcome. The use of FSH-only stimulation in cycles where there may be a profound suppression of LH has been demonstrated to have a negative impact on ART outcome. It is interesting in this last bit of information that in the patients receiving the highly-purified hMG there was a statistical correlation with rising hCG levels and implantation rates and on-going pregnancy rates. This goes back and I think adds more information to what Filicori initially showed in his studies with hCG alone. There is always the question, is hCG good? Is hCG bad? Well, it is the same question as is LH good, is LH bad? The answer to that is probably a little bit is good and too much may be too bad. The question really is, what is enough? We still do not have that answer to any great extent.
The bottom line is, I think today the use of LH or products that give us LH activity in agonist and antagonist cycles are going to improve or maintain a higher pregnancy rate in those cycles. There is no evidence I can find to say that what we are adding in the form of hMG or hCG during the stimulation has any negative effect on stimulation. When to add it? It is variable. You can add it from the beginning or you can add it on the fifth day of follicle development or stimulation. There is no evidence to show that adding it any particular time has made a lot of difference. I think the question is, in some cases, we really do need to add LH, those cases where we have had a prolonged down-regulation or were using agonists or antagonists for a long period of time. LH activity may be much more important that those types of cycles.
Thank you very much.
http://www.gonadotropin.org
|