GnRH agonists such as Synarel, Lupron and Zoladex are the heirs to danazol. They seek to mimic the hypoestrogenic menopausal state by inducing a profound suppression of pituitary gonadotropins. However, since this approach shares danazol's historical basis of origin, one would naturally predict that GnRH agonists would replicate danazol's shortcomings, although with different side-effects.
An early report on the efficacy of one GnRH agonist, buserelin, assessed response of the endometrium rather than endometriosis, assessed the visual appearance of endometriosis at the conclusion of therapy when errors of visual identification are common, and spoke of disappearance and resorption of lesions without biopsy proof. The definite impression was left that GnRH agonist therapy resulted in physical eradication of disease.
The large multicenter randomized, danazol-controlled, study of nafarelin-treated patients which helped gain its FDA approval, although impressive on the surface, repeats some of the errors found in earlier studies of medical therapy. The study was based on the undefined and unreferenced statement that "endometriosis resolves after ovariectomy and menopause." (emphasis added) Of 236 patients begun on therapy with either of two doses of nafarelin or one dose of danazol, almost 10% were immediately censored from the study for various reasons. Visualization at laparoscopy before and at the end (when ovarian suppression can give the false impression that disease has been eradicated) of Synarel therapy was used to assess response of the disease.
The undefined concept of "resolution" was the endpoint of therapy. "Resolution" may mean many things to many people, including such disparate results as physical destruction or eradication ("cure"), partial eradication, visual improvement which indicates physical eradication, or visual improvement alone with no implication as to the physical eradication of disease. Biopsy control was not used to ensure that all patients had endometriosis or that all lesions visualized were indeed endometriosis.
It is not known whether all of the eighteen investigators in the study had the same degree of expertise in diagnosing endometriosis in all of its subtle manifestations. Other peritoneal changes such as chronic inflammation or microcalcification can sometimes mimic endometriosis, and without biopsy, there is no way to be certain what is or is not endometriosis. Patients treated with Synarel were compared to danazol, but not to placebo. The decision not to compare Synarel to placebo was made on ethical grounds. Although the ethical concerns were not detailed, presumably it was because an earlier study had shown pain relief during therapy.
Concern has been voiced that medical (or surgical) studies of infertility which do not include a placebo group may over-represent the fertility results, since the background pregnancy rate with untreated endometriosis is up to 57%. Detailed analysis of this study indicates it is actually not a study of infertility patients, however. Nonetheless, the crude pregnancy rate among the unstated number of patients treated with Synarel who later attempted pregnancy was between 30-52% by 12 months after treatment ended. Since the group attempting pregnancy presumably included patients with normal but untested fertility, this is not an impressive pregnancy rate following therapy.
Long-term follow-up was not done; patients were followed only until they became pregnant, or until 12 months after treatment ended. Follow-up was apparently done primarily for patients attempting pregnancy. By 6 months after the end of treatment, 15% of the original 236 patients entered in the study had been censored or apparently lost to follow-up.
The choice of symptoms and methodology of study of symptom response during treatment was somewhat illogical. Ironically, dysmenorrhea was clustered with dyspareunia and pelvic pain not related to menstruation, and a total symptom score was developed to assess symptom response during treatment. Since nafarelin in high dose is a potent inhibitor of ovarian function, most women can be expected to become amenorrheic while on therapy. Since dysmenorrhea is usually thought of as pain or cramps due to the menstrual flow, cessation of menses for any reason will obviously result in improvement of dysmenorrhea.
Since this improvement is assured by amenorrhea, this may skew the results of symptom response in an artificially favorable direction by lumping a symptom which will always respond to amenorrhea (dysmenorrhea) with symptoms which may not respond to amenorrhea (dyspareunia, pain not associated with the menstrual flow). It would have been more helpful to assess the individual response of each symptom. Since it is known that other causes of pelvic pain are also estrogen-responsive (such as fibroids or adenomyosis), improvement of symptoms during therapy cannot be ascribed with absolute certainty to improvement of symptoms caused by endometriosis. Improvement of symptoms is therefore based on the assumption that symptoms were all due to endometriosis.
While endometriosis is a potent cause of pain, and while it may be true that much of the pain relief experienced during treatment may truly have been caused by endometriosis, we simply do not know for certain that the observed improvement represented a favorable response of symptoms due to endometriosis. This is a reprise of Kistner's postulate of an "assumed fact." This is why it is much safer to study biopsy-proven evidence of persistence or recurrence of endometriosis after medical or surgical treatment, since this is the only measurement which is absolutely indicative of the true response of the disease to treatment. Anything else is just a matter of opinion, not fact. This study was recently sent to all gynecologists by Syntex after FDA approval of Synarel as a promotion of the drug.
Summary of generic flaws of studies of medical therapy
As noted above, lack of biopsy-controlled studies has led to ambiguous results which are made all the more so by the now well-known concern over accurate identification of endometriosis. Have all clinicians participating in medical studies had the same expertise in the diagnosis of disease?
Lack of long-term follow-up makes it difficult to know where to put medical therapy in the context of treating endometriosis. What happens to these patients later? How many require surgery later? How many have persistent endometriosis? These are important common-sense questions that patients and practicing gynecologists need to have answered.
The assumption has been made in medical therapy studies that the symptoms of pain and infertility are caused by endometriosis, and that resolution of either symptom meant that the disease was responding to treatment, perhaps even being physically destroyed by that treatment. In fact, many pain symptoms may not be due to endometriosis, yet may respond during cessation of ovarian function by medical therapy. Dysmenorrhea will always improve with cessation of menses, and inclusion of this symptom in a global score for pain is misleading. Medical therapy for endometriosis is non-specific and will always seem to be better than it really is.
The lack of untreated control groups in studies of infertility patients is disturbing in light of recent concern that patients with Stages I and II endometriosis (most patients who have endometriosis are in these two stages) conceive without treatment at rates equal to or better than the rates of patients undergoing medical therapy. There is increasing recognition that this means endometriosis may not be the potent cause of infertility it was once thought to be. Actually, many studies have found that fertile women predominate in populations of patients with endometriosis.
One of the most important, fundamental errors of the past is the notion that infertility is the most common symptom of the disease. As a result, success in treatment of the disease is still measured primarily in terms of pregnancy rates. Even though pain seems to be a more common symptom of endometriosis than is infertility, the patient with pain has been largely ignored, both by gynecologists in search of patients for study and by the current measure of successful treatment, i.e. pregnancy. These patients have also been slighted by the AFS Classification System of Endometriosis, which exists primarily for comparing infertility patients.
These problems continue to be magnified today. Thus, many endometriosis patients targeted for study are recruited from the ranks of the infertile rather than the ranks of pain sufferers. Many papers on endometriosis are produced from departments of infertility/reproductive endocrinology at metropolitan medical schools rather than from clinicians in general practice. Once this infertile minority is selected out, they are subjected to medical treatments based on scientifically unproven clinical impressions (that menopause or pregnancy makes endometriosis go away or get better). Many modern studies have shown these medical therapies to be ineffective in eradicating endometriosis and to have no apparent beneficial effect on pregnancy rates. Therapy based on these errant findings is then offered to the majority of patients with pain as a primary symptom.
A final difficulty facing medical treatment is the finding that many recent studies have shown that individual endometriosis lesions have widely varying concentrations of estrogen and progesterone receptors, and that these hormones are usually present in amounts lower than the native endometrium. For this reason, endometriosis does not bleed predictably or reliably with the menstrual cycle 46 as generations of gynecologists have been taught. (This is one of the explanations for atypical, subtle or non-hemorrhagic forms of the disease.)
For pregnancy, birth control pills, danazol, GnRH , or menopause to have a beneficial hormonal effect on endometriosis, the lesions would have to respond to the presence of estrogen and progesterone (birth control pills or pregnancy), or lack of estrogen and progesterone, (danazol, GnRH, menopause.) The first problem is that these therapeutic endpoints are diametrically opposed and irreconcilable. You can't have it both ways. Progesterone and estrogen cannot have both a positive and a negative impact on the disease (many would argue that they have no impact at all). In addition, for any of these hormonal states to have a therapeutic effect, the estrogen and/or progesterone would have to have a fairly uniform effect on all populations of endometriosis cells.
The low and varying population of estrogen and progesterone receptors in endometriosis cells in the same patient predicts that any hormonal effect based on our current knowledge would be erratic, inconstant, ineffective and therefore irrational. If we have been putting most of our eggs in the infertility basket, and we now realize the basket has holes and the eggs have cracks, why do we persist in using obsolete thought processes which run down our legs like yolk? Why do we still try to cram the square peg of endometriosis into the round hole of infertility? Why is this happening? Who is in control?
One argument is often offered in support of medical therapy. One traditional theory of origin proposes that endometriosis is caused by monthly regurgitation of viable endometrial cells in a reverse flow through the fallopian tubes. These cells are thought to fall like seeds on fertile soil, implant and grow. Following this theory, if the monthly regurgitation could be stopped or reduced, it might reduce the likelihood of future implants and perhaps allow resorption of implants already there.
For starters, this theory of reflux menstruation and implantation is just one of at least 10 theories of origin and it has not been proven. The real problem, however, is that physicians continue to prescribe ineffective and experimental medicines in a relative vacuum populated only by words such as "perhaps," "might," and " is thought to." Greater insight is needed into the basics of this disease before rational medical treatment can be offered. Actually, it will be impossible ever to study the effect of medical therapy on endometriosis for a simple reason: you can't biopsy the same cell twice to study it.
Back to "What to Ask"