EP0739352A1

EP0739352A1 - New composition of glycoprotein isoforms having follicle stimulating activity

Info

Publication number: EP0739352A1
Application number: EP95907574A
Authority: EP
Inventors: Renato De Leeuw; Ferdinand Rombout
Original assignee: Akzo Nobel NV
Current assignee: Akzo Nobel NV
Priority date: 1994-01-20
Filing date: 1995-01-20
Publication date: 1996-10-30
Also published as: HUT76660A; HU9601975D0; CA2181711A1; WO1995019991A1; FI962923A0; BR9506557A; JPH09508114A; CN1138864A; FI962923A; AU1575195A

Abstract

The invention is concerned with a mixture of isoforms of a glycoprotein with follicle stimulating activity, a pharmaceutical composition comprising such isoforms as well as a method to induce follicular growth and maturation. Better results were obtained if a pharmaceutical composition with a relatively high proportion of basic isoforms were administered in controlled hyperstimulation. More oocytes could be retrieved and a lower dosage as well as a shorter treatment period sufficed.

Description

NEW COMPOSITION OF GLYCOPROTEIN ISOFORMS HAVING FOLLICLE STIMULATING ACTIVITY

The present invention is concerned with a pharmaceutical composition comprising a mixture of isoforms of a glycoprotein with follicle stimulating activity as well as an improvement in a method to induce follicular growth and maturation.

Follicle stimulating hormone (FSH) produced by the anterior pituitary plays a major role in female and male reproduction by stimulating gonadal differentiation and maturation via its regulatory action on the Sertoli cell in the testes and granulosa cell in the ovary.

FSH is produced and secreted by the pituitary in different molecular forms (isohormones or isoforms), which vary in overall charge, receptor binding affinity, biological activity, and plasma residence time. This micro-heterogeneity is due to differences in the amount and/or composition of the carbohydrate residues, in particular sialic acid. Multiple forms of gonadotropins have been isolated and characterised from anterior pituitary glands, serum and urine of several non-mammalian and mammalian species, including man. Relatively acidic FSH isoforms, which are more heavily sialylated, exhibit lower receptor affinity and in vitro biological activities than more basic isoformes. However, due to their longer plasma residence time these more acidic forms have greater in vivo biological activities (UUoa- Aguirre et al., 1988, Hum.Reprod., 3, 491-501).

FSH is used for ovulation induction and controlled ovarian stimulation in in vitro fertization (INF). The aim of controlled superovulation is to increase the number of retrievable mature oocytes for INF and subsequent embryo transfer (ET). Generally, up to three embryos are replaced per transfer. As usually more than one treatment is necessary, in most infertility clinics spare embryos or fertilized oocytes are frozen and transferred in subsequent cycles. Assuming normal fertilization, the more oocytes retrieved the higher the number of possible transfers and thus the higher the chance of a woman to become pregnant after one treatment cycle. In case of male infertility, the chances of establishing fertilization, and thus pregnancy, also increases with the number of oocytes recovered.

Previously, it was shown that the majority of circulating FSH isoforms during the follicular and luteal phase of the normal menstrual cycle, have pl values < 4.8, whereas during midcycle, or after treatment with oestrogens considerably more alkaline FSH isoforms were present (Padmanabhan et al., 1988, J.Clin.Endocrinol.Metab.,

67, 465-473). This indicates that the more acidic FSH isoforms play a role in the growth and recruitment of ovarian follicles during the follicular phase.

Hitherto for clinical purposes FSH preparations have been used which have been isolated from natural sources. Isohormone distribution profiles of commercially available preparations have been reported (e.g. Harlin et al, 1986, Fert. Ster., 46, 1055-1061). It appears that these FSH compositions consist of relatively acidic isohormone fractions.

Surprisingly, it now has been found that a mixture of glycoprotein isoforms with isoelectric points in between and extending from the range 4.8 to 4.2 , having follicle stimulating activity which consists for more than 15% of isoforms with isoelectric points above 4.8 and for less than 30% of isoforms with isoelectric points below 4.2 , when used in the same clinical settings, exert a better effect than the known glycoprotein compositions.

Clinical studies have been performed with a composition of glycoprotein isohormones according to the invention and obtained by recombinant DNA methods. A separation of FSH isohormones according to their isoelectric point (pi) revealed that the preparation according to the invention has relatively basic isoforms. The isoelectric points as described herein were determined by means of chromato- focusing (see example 1). The commonly known glycoprotein isohormone compositions (of which Metrodin^ used in this clinical study is an example) have a more acidic profile. The fact that a more basic isohormone profile gives better results in superovulation is surprising because it has been thought in the literature that the in vivo bioavailability is the most important factor for the follicle stimulating effect of the glycoproteins (European patent application EP 388 223). The more basic isohormones have a shorter half-life than the more acidic isoforms which would make the basic isoforms less suited for use in IVF schemes and thus teaches away from the present invention.

It is preferred that the glycoprotein isohormone mixture according to the invention contains a relatively high proportion of relatively basic isoforms. This can be established by having glycoprotein isoform compositions having more than 15% of the isoforms with isoelectric points above 4.8 and less than 30% below 4.2 , also preferably, isoforms mixtures can be used wherein more than 15% of the isoforms have isoelectric points above 4.8 and less than 25% below 4.2 , also preferably more than 25% of the isoforms have isoelectric points above 4.8 and less than 25% below 4.2 ,also preferably more than 25% of the isoforms have isoelectric points above 4.8 and less than 20% below 4.2

The percentage of the isoforms as used herein is defined as the relative amount of immunoreactive isoforms recovered after chromatofocussing. Alternatively, also protein contents can be determined by colorimetric assays.

In a clinical study, as described herein, it has been found that in controlled superovulation the number of oocytes retrieved is different in the two FSH preparations used i.e. ' the preparation according to the invention and Metrodin^ with respect to the number of oocytes retrieved. More oocytes could be retrieved from the group treated by a preparation according to the invention. In addition, a significantly lower FSH dosage and a shorter treatment period sufficed.

The mixture of glycoproteins according to this invention may be derived from urinary origin. For natural FSH preparations, the number and relative amount of each isoform species depends on the source

(pituitary, serum or urine), the age and the endocrine status of the donor, and the isolation procedure applied for its purification (Wide, 1981, J.Clin.Endocrinol.Metab., 55, 682-688; Wide and Hobson, 1983, J.Clin.Endocrinol.Metab., 56, 371-375).

Alternatively, the glycoprotein might also be a recombinant glycoprotein.

For recombinant FSH (recFSH) the charge heterogeneity is determined by the host cell-line chosen for its production as well as cell culture conditions. For glycoproteins, such as FSH, a Chinese Hamster Ovary (CHO) cell line is an obvious choice since these cells are known to produce glycoproteins with oligosaccharides identical or closely related to those found in man (Sasaki et al.,J. Biol.Chem.,1987, 262, 12095-12076). However, also other host cell lines can be used.

A recombinant FSH preparation according to this invention can be produced by a CHO cell line stably transfected with a plasmid containing the two subunit genes encoding human FSH (hFSH).

Such a CHO cell line produces intact, glycosylated hFSH that is secreted into the culture medium, which is the source for further purification. Batches like this can be prepared as described (van Wezenbeek et al, 1990, in: From Clone to Clinic, Kluwer Academic Publishers, 245-251). Previous preclinical studies demonstrated that the receptor binding affinity and in vitro and in vivo efficacy of recFSH are compatible to those of FSH isolated from natural sources (Mannaerts et al., 1991, Endocrinology, 129, 2623-2630).

FSH preparations according to the invention can be selected based upon their chromatofocusing profile. In addition, the profile can be influenced by a particular host cell line choice or by adaptation of culture conditions.

As an alternative, also glycoprotein isoform mixtures according to the invention can be isolated from cell lines the glycoprotein gene expression of which has been turned on by site specific targeting of a regulatory sequence to the glycoprotein gene. (PCT application WO

92/19255).

In addition to wild-type cell lines, basic isohormones can be obtained by expression of recombinant FSH in cell lines which are impaired in glycosylation. Such a cell line might be e.g. a cell line deficient in the enzyme N-acetylglucosamine transferase or in sialic acid transport into the Golgi (Galway et al., 1990, Endocrinology, 127, 93-100).

Other embodiments of this invention are basic glycoprotein isoforms mixtures obtained by enzymatic or chemical modification. With such a treatment parts of the carbohydrate chains can be removed without affecting the amino acid sequence. Glycoprotein batches can e.g. be treated with HF (Chen et al, 1982, J.Biol.Chem., 257, 14446- 14452). Partial desialylation can be performed by enzymatic hydrolysis with neuraminidase (Vaitukaitis and Ross, 1971, J.Clin.Endocrinol.- Metab., 55, 308-311).

Also subject of this invention are basic glycoprotein hormone mixtures having biological i.e. follicle stimulating activity which are prepared by removing one or more N-linked oligosaccharide chains from either one of the subunits. These chains can be removed by site- directed mutagenesis of the gene encoding the glycoprotein. In this way a single amino acid change can be obtained in such a way that the carbohydrate attachment site is no longer present. Consequently, one or several carbohydrate chains cannot be attached on the a or β subunit (Matzuk and Boime, 1988, J.Cell Biol., 106, 1049-1059). The mixture of relatively basic isohormone forms can be used according to the invention for the manufacture of a medicament to be used e.g. in the treatment of ovulation induction or controlled ovarian stimulation.

Thus, the invention relates to a pharmaceutical composition comprising these glycoprotein isoforms admixed with pharmaceutically acceptible auxiliairies. Methods for making preparations and admixtures are disclosed in Remingtons 's Pharmaceutical Sciences, pp. 1463-1497 (16th ed. 1980, Mack Publ. Co of Easton, Pa, USA). For example, ampoules containing the pharmaceutical composition according to the invention may contain 1 to 1000 μg of the glycoprotein mixture (e.g. 75 IU is considered a therapeutic amount).

The invention also relates to an improved method to induce follicular growth and maturation wherein the pharmaceutical composition containing the mixture of basic glycoprotein isoforms as described above is used. According to the present invention there is no need to include in the treatment schedule the administration of more acidic isoform mixtures. Another improvement of the method consists of a repeated administration of pharmaceutical compositions containing a mixture of isoforms of a glycoprotein with follicle stimulating activity, each composition having identical, i.e. having the same isoform profile. Thus, during a treatment the administration of only relatively basic isoforms is sufficient.

Such mixtures can also be prepared by isolation of only basic isoforms e.g. by preparative chromatofocussing. Preferably such isoforms have an isoelectric point above 4.2 . It will be clear that fractionation of isoform mixtures can be performed on glycoprotein batches obtained from different origins such as preparations isolated from urine or recombinant DNA cell lines which may or may not be chemically or enzymatically modified.

Pharmaceutical compositions according to this invention can be used in clinical treatments in combination with e.g. GnRH antagonists or agonists and/or LH activity e.g. HCG or LH to induce superovulation.

Example 1

Comparison of the FSH immunoreactive isohormone profile of recombinant human FSH (Org 32489) and urinary FSH (Metrodin^R)

In the present study the isohormone profile of several batches of Org 32489 (recombinant human FSH, N.V. Organon), derived from Chinese Hamster Ovary cells transfected with the genes encoding for hFSH, was examined and compared with that of commercially available batches of urinary hFSH, i.e. Metrodin^.

Purified (>99%) Org 32489 bulk and final product preparations were obtained from Diosynth and Organon (The Netherlands), respectively. Urinary hFSH, i.e. Metrodin^R (Serono, Rome, Italy; 75

IU/ampoule declared in vivo bioactivity relative to IS 70/45), was used as reference. In total twelf Org 32489 bulk batches, five Org 32489 clinical preparations (CP's), and twelve Metrodin^R batches were used for isohormone comparison. Metrodin^R batches indicated with an in- house CP-code have been used in a trial with Org 32489 CP batches to compare the clinical efficacy (see Example 2).

By employing chromatofocusing, the different isohormones of hFSH were separated based on their isoelectric point (pi). The distribution of hFSH was determined after quantification by an enzyme immunoassay employing antibodies that showed comparable activities towards all isoforms.

Chromatofocusing was performed in the range of pH 6-3 on a fast protein liquid chromatography (FPLC) column HR 5/20 (Pharmacia, Woerden, The Netherlands) packed with polybuffer exchanger 94

(PBE-94; Pharmacia) and equilibrated with 0.034 mol/1 L-histidine (Aldrich Chemie, Steinheim, FRG) adjusted to pH 6.2 with HC1. Before use each column was eluted with 53 ml polybuffer 74 (Pharmacia) diluted 1 :11 (v/v) with distilled water and adjusted to pH 3.0 with HC1 (elution buffer) until a pH curve with a constant decline from pH 6 to 3 was obtained. Before each experimental run the column was eluted with elution buffer and 2 ml 0.034 mol/1 L-histidine containing 100 μg/ml human serum albumin (HSA; Behring, Marburg, FRG).

Of each hormone preparation, approximately 225 IU FSH in terms of in vivo bioactivity was dissolved in 3 ml equilibration buffer (stock solution) and 2 ml of this solution (150 IU) was applied to the column.

Subsequently, the column was eluted with the elution buffer. Fractions of 1 ml were collected at a flow rate of 1 ml/min. After 53 fractions 2 ml of a 2 mol/1 NaCl solution was applied to the column and 7 additional fractions were collected. For the quantification of FSH, fraction 1 , 2 and 3 and fraction 4, 5 and 6 had to be pooled. All other fractions were adjusted up to 2.5 ml with a 1:1 mixture of HAM F12 and DMEM (Gibco, Grands Islands, NY, USA) supplemented with 1 g/1 bovine serum albumin (BSA; Sigma, St. Louis, MO, USA) (medium ⁺). In addition, 0.3 ml of the FSH stock solution was adjusted to 2.5 ml with 2.2 ml medium⁺. All FSH fractions were desalted by applying each fraction to a PD-10 column (Pharmacia), equilibrated with 12 ml medium ⁺. After elution with 3.5 ml medium ⁺ , the collected samples were stored at -20 ^°C until determination of FSH immunoreactivity.

FSH immunoreactivity was measured in a two-site sandwich enzyme immunoassay, using a β-directed capturing antibody (monoclonal antibody 4B) and an α-directed HRP-labeled detection antibody (monoclonal antibody 116B) as described previously (Mannaerts et al., 1991, Endocrinology, 129, 2623-2630). This assay recognises only intact dimers and was found to detect all FSH isoforms equally well. The assay sensitivity in terms of IS 70/45 was 0.4 IU/1 and the intra- and interassay coefficients of variations were 7% and 8%, respectively. The cross-reactivity with hLH and hCG was < 0.01 % and < 0.01 % , respectively

In order to quantify the isohormone composition of Org 32489 and Metrodin^R, the distribution of FSH immunoreactivity after chromatofocusing was divided in the following six pH ranges: > 5.3,

5.30 - 4.81,

4.80 - 4.20, 4.19 - 3.60, 3.59 - 3.00, < 3.0 (salt fraction).

These pH ranges were selected to obtain a symmetrical dividing of the bell-shaped distribution of FSH immunoreactivity after chromatofocusing of Org 32489. The amount of FSH immunoreactivity which eluted within each pH range was expressed as percentage of total eluted FSH immunoreactivity. In addition, recovery, i.e. total eluted FSH immunoreactivity expressed as percentage of FSH immunoreactivity applied to the column, pH range, and pH top were determined. Recoveries below 75 % and above 125 % were considered not-acceptable resulting in rejection of the chromatofocusing run.

The recovery, pH range, pH top and the distribution of FSH immunoreactivity after chromatofocusing of the Org 32489 bulk and Metrodin^R batches are shown in Table 1. Representative chromatofocusing profiles of Org 32489 are shown in Figure 1. The recovery of FSH immunoreactivity after chromatofocusing of Org 32489 and Metrodin^R was comparable, being 93.7 (SD 8.9) % and 89.1 (SD 6.1) %, respectively. Org 32489 exhibited a bell-shaped distribution of FSH immunoreactivity in a pH range of 5.68 (SD 0.11)

- 3.18 (SD 0.09) with a top at pH 4.53 (SD 0.08) (n= 12). The FSH immunoreactive profile of Metrodin^R ranged from 5.58 (SD 0.18) to 3.07 (SD 0.05) with a top at pH 4.28 (SD 0.13) (n=12).

The distribution of FSH immunoreactivity over the six pH ranges showed that in comparison with Metrodin^R, Org 32489 contained an approximately 2-fold higher percentage of relative basic isoformes with pl > 4.8 (20.9 % vs 12.7 %) and an approximately 2-fold lower percentage of relative acidic isoforms with pi < 4.2 (21.1 % vs 41.7 %) (Table 1 and 2). In addition, all Metrodin^R batches contained more FSH immunoreactivity in the salt fraction (3.3 % vs 0.8 % for Org

32489), representing FSH isoforms with a pi of less than 3.0.

In addition to the above described Org 32489 bulk batches, the chromatofocusing profile of five clinical preparations was investigated. The mean FSH distribution of these CP's was very comparable to that found for the bulk batches (Table 1). Like the twelve batches Org

32489 and Metrodin used for isohormone comparison, the Org 32489 and Metrodin clinical preparations used to compare the clinical efficacy (see Example 2) showed the same difference in the relative contribution of basic and acidic isoforms, i.e. the Org 32489 CP's contained more basic isoforms than the Metrodin CP's (17.120.2 % vs 12.7 % if pi > 4.8).

This comparative study comprising chromatofocusing profiles of Org 32489 and urinary hFSH, i.e. Metrodin^R, revealed that Org 32489 containes more basic isoforms with pi > 4.8 and less acidic isoforms with pi < 4.2 than Metrodin.

Example 2

Comparative efficacy study of an Org32489 preparation in patients undergoing IVF

Study design

The study was designed as a multicentre, randomized, assessor- blind, group-comparative study, in which safety and efficacy of Org 32489 and Metrodin^R were compared in infertile pituitary-suppressed subjects undergoing in vitro fertilisation (IVF) and embryo transfer (ET). Approximately one thousand subjects were included in this study with a ratio between subjects treated with Org 32489 and with Metrodin^R of 3:2. The study period covered no more than 3 treatment cycles. Analysis of efficacy included first treatment cycles only.

Inclusion of subjects was based on the following entrance criteria:

Inclusion criteria - At least 18 and at most 39 years of age at the time of screening;

- Cause of infertility of the subject potentially solvable by IVF;

- Maximum of three previous IVF, gamete intra-fallopian transfer (GIFT) or zygote intra-fallopian transfer (ZIFT) attempts, in which oocytes were collected at least once; - Normal ovulatory cycles with a mean length of between 24 and

35 days and an intra-individual variation of plus or minus 3 days (but never outside the 24-35 days range);

- Good physical and mental health;

- Body weight between 80 and 130% of the ideal body weight; and

- Willing to give written informed consent.

Exclusion criteria

- Infertility caused by endocrine abnormalities such as hyperprolactinaemia, polycystic ovary syndrome, and absence of ovarian function; - Male infertility defined according to the following criteria: < 10x10^ sperms per mL and/or < 40% normal morphology and/or 40% normal motility;

- Contra-indications for the use of GnRH analogues, FSH, hMG, and/or hCG;

- Any ovarian and/or abdominal abnormality that would interfere with adequate ultrasound investigation;

- Hypertension (sitting diastolic blood pressure > 90 mm Hg and/or systolic blood pressure > 150 mm Hg); - Chronic cardiovascular, hepatic, renal, or pulmonary disease;

- History (within 12 months) or current abuse of alcohol or drugs; and

- Administration of investigational drugs within three months prior to screening.

Treatment schedule

Buserelin treatment was started on the first day of the menstrual cycle. The first day of the menstrual cycle was defined as the first day the subject waked up with menstrual bleeding. Treatment with Org 32489 (75 IU/ampoule) or Metrodin^R (75 IU/ampoule) was started 14 to 18 days after the onset of the buserelin intake when this treatment had resulted in an hypogonadotropic state (i.e. , serum E2 < 50 pg/mL). In case this state had not been achieved, Org 32489 or Metrodin treatment was postponed and the dose of buserelin was increased to 4 x 300 μg/day. This increased buserelin dose was sustained during the whole further treatment period up to the moment FSH treatment was stopped. Pretreatment with buserelin was not allowed to exceed a period of five weeks. If after complete down- regulation on the first FSH treatment day a follicular cyst > 20 mm had developed, this cyst was punctured before the first FSH injection.

In each treatment cycle dosing was performed according to the schedule below: Buserelin Dose 4x150 μg/day

Route intranasal

Org 32489/Metrodin^R Dose Treatment day 1-4: 150-225 IU/day. from day 5 on: dose was adjusted per subject

Route i.m.

HCG (10.000 IU) was administered when at least three follicles > 17 mm were present. No more than three embryos were replaced and frozen embryos were replaced in natural cycles or stimulated cycles. A cycle was considered cancelled if no embryo transfer had taken place. The luteal phase was monitored by progesterone assessments. Luteal phase support (e.g. three injections of 1500 IU hCG or at least 50 mg progesterone i.m. per day or 400 mg progesterone intravaginally per day) was given for at least two weeks after the injection of 10.000 IU hCG.

Statistical evaluation

The total number of oocytes recovered in the first treatment cycle was a primary efficacy variable. Other analysed variables in the first treatment cycle were the number of FSH ampoules administered for ovarian stimulation and the duration of FSH treatment. The outcome of these parameters were analysed by means of Cochran's method of combining individual centre results. In addition, the Wilcoxon rank sum test and an analysis of variance (ANOVA) were applied to consolidate the outcome of Cochran's method.

Study outcome of first treatment cycles

Altogether 981 subjects started FSH treatment in 18 study-centres. In total 907 subjects had a puncture, 546 in the Org 32489 group and 361 in the Metrodin group. With this number of punctured subjects it was possible with probability 0.8 (the so-called power of the study) to detect at least a difference of 1.2 oocytes between the two treatment groups.

The number of oocytes recovered in each treatment group per centre is presented in Table 3. In all centres the Org 32489 group had a larger mean number of oocytes recovered than the Metrodin group. The overall mean number of oocytes (weighted over centres) was 10.84 oocytes in the Org 32489 group and 8.95 oocytes in the Metrodin group, resulting in a treatment difference of 1.89 oocytes in favour of Org 32489. The difference of 1.89 oocytes is more than 5 times its standard error and highly significant (P < 0.0001). The resulting 95% confidence interval indicated that on the average subjects treated with Org 32489 end up with at least 1.2 and at most 2.6 oocytes more than those treated with Metrodin^R. Exclusion of immature oocytes from the total number of recovered oocytes did not influence the treatment effect (see Table 4).

The total number of ampoules administered to punctured subjects is presented per centre in Table 5. The total dosage of Org 32489 administered was in 14 out of the 18 centres lower than the total dosage of Metrodin . The overall mean FSH dosage (weighted over centres) was 28.5 ampoules in the Org 32489 group and 31.8 ampoules in the Metrodin group, resulting in a treatment difference of 3.3 ampoules which is highly statistically significant (P< 0.0001). The resulting 95% confidence interval indicated that on the average subjects treated with Org 32489 receive minimally 2.1 and maximally 4.5 ampoules less than those treated with Metrodin^R.

The duration of FSH treatment of punctured subjects is presented per centre in Table 6. The treatment duration was in 13 out of 18 centres shorter in the Org 32489 group than in the Metrodin group. The overall mean treatment duration (weighted over centres) was 10.7 days in the Org 32489 group and 11.3 days in the Metrodin group, resulting in a treatment difference of 0.6 days which is highly statistically significant (P<0.0001). The resulting 95% confidence interval indicated that on the average for subjects treated with Org 32489 the treatment period is minimally 0.3 day and maximally 0.9 day shorter than for those treated with Metrodin^R.

The outcome of the Wilcoxon rank sum test and the analysis of variance confirmed all of the above findings.

Thus, it can be concluded that in this study controlled superovulation by means of Org 32489 treatment demonstrated superiority over Metrodin with respect to the number of oocytes retrieved with a significantly lower FSH dosage and shorter treatment period.

Legends

Figure 1

Distribution of FSH immunoreactivity after chromatofocusing of a specific Org 32489 batch. The profile of one batch is shown as example. Upper and lower panel represent separate runs, (sf = salt fraction). Total recovery 91.9 and 91.3 %, respectively.

Table 1

Mean recovery, pH range, pH top and distribution of FSH immunoreactivity over pH ranges after chromatofocusing of twelve Org 32489 bulk batches, five Org 32489 final products, i.e. clinical preparations (CP's) and twelve Metrodin^R batches.

Org 32489 Metrodin

Bulk material CP's

n 12 5 12

Recovery 93.7 ± 8.9 % 84.2 ± 3.6 % 89.1 ± 6.1% pH range 5.68 - 3.18 5.60 - 3.13 5.58 - 3.07 pH top 4.53 4.52 4.28

FSH distribution: pH range > 5.3 2.0 % 2.2 % 1.3 %

5.30- 4.81 18.9 % 18.0 % 11.4 %

4.80- 4.20 58.2 % 55.5 % 45.6 %

4.19- 3.60 18.6 % 21.7 % 33.8 %

3.59- 3.00 1.7 % 2.0 % 4.6 %

< 3.0 0.8 % 0.8 % 3.3 %

Table 2

Percentage basic isoforms, defined as FSH immunoreactivity at pi > 4.8, and percentage acidic isoforms, defined as FSH immunoreactivity at pi < 4.2 of the Org 32489 and Metrodin^R CP's used to compare the clinical efficacy (see Example 2).

Preparation Isohormone distribution

% basic % acidic

(pl>4.8) (pl <4.2)

Org 32489: bulk batches n= 12 20.9 21.1

CP's comparative efficacy study 17.1 23.9

(n=2)

Metrodin^R:

Batches n= 12 12.7 41.7

CP's comparative efficacy study 11.9 44.4

(n=3)

Table 3

Statistical analysis (Cochran's approach) of the total number of oocytes recovered Including all punctured subjects.

Total number of oocytes recovered Group Org 32489 minus Metrodin

Cycle 1 Org 32489 Metrodin Estimate of Standard 95V confidence treatment error of interval difference estimate (p-value)

Centres n 546 361 combined Mean 10.84 8.95 1.89 0.37 1.2 to 2.6 j adjusted for centre UO.OOOl)

Test for interaction (0.88) |

Centre

1 n 67 41 Mean 12.54 11.80 0.73 1.33

2 n 12 7 Mean 12.67 8.71 3.95 3.64

3 n 4 5 Mean 9.50 4.00 5.50 1.74

4 n 11 7 Mean 7.91 5.43 2.48 1.89

5 n 13 8 Mean 15.46 10.38 5.09 3.20

6 n 30 17 Mean 12.60 9.53 3.07 2.04

7 n 44 24 Mean 7.23 5.21 2.02 1.01

8 n 19 15 Mean 7.68 6.60 1.08 1.92

9 n 27 19 Mean 10.78 9.79 0.99 1.58

10 n 41 25 Mean 11.12 9.08 2.04 0.96

11 n 24 14 Mean 9.50 8.71 0.79 1.56

12 n 16 13 Mean 9.38 8.62 0.76 1.85

13 n 33 Mean 11.70 1 10.²4²5 1.24 1.47

14 n 64 Mean 12.80 ! 12.⁴1⁶7 0.62 1.29

15 n 83 Mean 12.48 ! ⁵⁹ 9.81 2.67 1.13

16 n 27 j 17 Mean 13.70 12.76 0.94 2.40

17 i n 17 ! ^ι3 j Mean 13.29 11.23 2.06 ! 1.93

18 n 14 Mean 16.43 1i 11.6⁹7 4.76 j 3.77 Table 4

Statistical analysis (Cochran's approach) of the number of mature oocytes recovered Including all punctured subjects.

Number of mature oocytes recovered Group Org 32489 minus Metrodin

Cycle 1 Org 32489 Metrodin Estimate of Standard 95% confidence treatment error of interval difference estimate (p-value)

Centres n 544 361 combined Mean 8.55 6.76 1.79 0.33 1.1 to 2.4 adjusted for centre (<0.0001)

Test for interaction (0.89)

Centre

1 n 67 41 Mean 12.48 11.29 1.31

12 7

Mean 10.92 7.43 3.49 3.10

4 5

5.75 1.60 4.15 1.24

10 7 7.50 5.43 2.07 2.02

13 8 13.23 8.25 4.98 3.20 n 30 17 Mean 10.53 8.35 2.18 1.81 n 44 24 Mean 7.23 5.21 2.02 1.01

19 15 4.37 4.27 0.10 1.30

27 19 10.59 9.74 0.86 1.59

41 25

Mean 6.93 5.40 1.53 0.99

11 ⁿ 24 14

Mean 8.17 7.21 0.95 1.27

12 j ⁿ 16 13

; Mean 6.81 7.15 -0.34 1.98

13 33 22

! Mean 10.94 9.82 1.12 1.48

14 i ⁿ 64 46

! Mean 12.59 11.35 1.25 1.33

15 ! ⁿ 83 59

Mean 7.39 4.85 2.54 0.79

16 ! ⁿ 27 17

Mean 12.56 11.24 1.32 2.33

17 13 13.29 11.23 2.06 1.93

13 9 10.31 8.11 991

21

Table 5

Statistical analysis (Cochran's approach) of the total FSH dose administered Including all punctured subjects.

Total number of FSH vials/ampoules Group Org 32489 minus Metrodin administered

Cycle 1 Org 32489 Metrodin Estimate of Standard 95% confidence treatment error of interval dif erence estimate (p-value)

Centres n 546 361 combined Mean 28.49 31.82 -3.33 0.62 -4.5 to -2.1 adjusted for centre O.OOOl)

Test for interaction (0.06) centre

1 67 41 29.29 38.13 -8.84 3.44 n 12 7 Mean 28.08 27.43 0.65 2.80 n 4 5 Mean 39.50 42.60 -3.10 8.10 n 11 7 Mean 36.36 37.86 -1.49 5.49 n 13 8 Mean 30.54 42.00 -11.46 4.86 n 30 17 Mean 26.87 28.76 -1.90 2.95 n 44 24 Mean 27.07 33.83 -6.77 2.21 n 19 15 Mean 49.26 48.77 0.50 5.66

27 19

Mean 35 04 32 79 2.25 3.09

10 n 41 25 Mean 35 70 37 88 -2.18 3.59

24 14

Mean 30 75 38 07 -7.32 2.78

12 16 13

Mean 32 38 32 00 0.38 6.01

13 n 33 22 Mean 27 61 29 00 -1.39 2.15

14 n 64 46 Mean 30 59 31 22 -0.62 1.29

15 n 83 59 Mean 23 67 28.19 -4.52 1.44

16 n 27 17 Mean 23.26 28.47 -5.21 2.65

17 n 17 13 Mean 25.79 33.23 -7.44 3.64

18 n 14 9 Mean 26.93 35.56 -8.63 3.33 Table 6

Statistical analysis (Cochran's approach) of the duration of the FSH treatment Including all punctured subjects.

Duration of FSH treatment (day) Group Org 32489 minus Metrodin

Centres n 546 361 coablned Mean 10.66 11.27 -0.60 0.13 -0.9 to -0.3 adjusted for centre (<0.0001)

Test for interaction (0.02)

Centre

1 n 67 41 Mean 11.99 13.73 -1.7S 0.65

2 n 12 7 Mean 9.08 9.00 0.08 0.66

3 n 4 5 Mean 13.75 14.00 -0.25 1.41

4 n 11 7 Mean 11.64 11.57 0.06 1.15

5 n 13 8 Mean 9.92 11.38 -1.45 0.69

6 n 30 17 Mean 10.83 11.24 -0.40 0.50

7 n 44 24 Mean 12.41 13.46 -1.0S 0.55

8 n 19 15 Mean 13.11 13.53 -0.43 0.96

9 n 27 19 Mean 11.15 10.26 0.88 0.59

10 n 41 25 Mean 12.24 12.72 -0.48 0.43

11 n 24 14 Mean 11.33 12.93 -1.60 0.57

12 n 16 13 Mean 12.25 11.23 1.02 1.08

13 n 33 22 Mean 8.55 8.45 0.09 0.35

14 n 64 46 Mean 10.09 10.33 -0.23 0.35

15 n 83 59 Mean 10.81 12.02 -1.21 0.37

16 n 27 sr Mean 10.96 11.82 -0.86 0.48

17 n 17 13 Mean 10.06 11.77 -1.71 0.67

18 n 14 9 Mean 9.79 11.33 -1.55 0.B4

Claims

1.

A pharmaceutical composition comprising a mixture of isoforms of a glycoprotein hormone with isoelectric points in between and extending from the range 4.8 to 4.2 having follicle stimulating activity, in admixture with pharmaceutically acceptable auxiliaries, characterized in that more than 15% of the isoforms have isoelectric points above 4.8 and less than 30% below 4.2 .

2.

The composition according to claim 1 characterized in that more than 15% of the isoforms have isoelectric points above 4.8 and less than 25% below 4.2 .

3.

The composition according to claim 2 characterized in that more than 25% of the isoforms have isoelectric points above 4.8 and less than 25% below 4.2 .

4.

The composition according to claim 3 characterized in that more than 25% of the isoforms have isoelectric points above 4.8 and less than 20% below 4.2 .

5.

The composition according to claims 1-4 characterized in that the glycoprotein is recombinant follicle stimulating hormone.

6.

The composition according to claims 1-5 characterized in that the glycoprotein is modified by enzymatic or chemical treatment.

7. The composition according to claim 5 characterized in that the recombinant follicle stimulating hormone is modified by eliminating at least one of the N-linked oligosaccharide chains.

8.

Method to induce follicular growth and maturation in females by administering a pharmaceutical composition according to claims 1-7.

9.

Method to induce follicular growth and maturation by administering during the ovulatory cycle identical pharmaceutical compositions of a mixture of isoforms of a glycoprotein with follicle stimulating activity characterized in that said isoforms have an isoelectric point above 4.2 .