28 Analyst, January, 1968, Vol. 93, $9. 28-33 Analytical Evaluation of Gestogens in Oral Contraceptives BY G. R. KEAY (City Laboratories Service, Shortley Road, Coventry) Oral contraceptive preparations are now manufactured and prescribed on a large scale, and the public analyst has a duty to ensure that a chemically satisfactory product is supplied. A scheme has been devised that will enable all of the ingredients, both progestogenic and oestrogenic, of the preparations a t present available in this country to be identified and assayed. The active ingredients are separated from the tablet excipients by solvent extraction, the solution is evaporated and the residue identified by using infrared spectrophotometry and thin-layer chromatography. Assay of the ingredients involves the use of ultraviolet spectrophotometry.A few excep- tions were found to this general scheme and these have been successfully overcome by suitable modifications. THE group of sex hormones referred to as gestogens consists of those that constitute the steroid fraction of the oral contraceptive tablet. This steroid content is composed of a member of the oestrogenic or follicle-stimulating hormones, together with a much larger concentration of progestogen or corpus luteum secreted material. The most potent of all oral oestrogens is ethinyloestradiol which, with its 3-methyl ether, mestranol, may be regarded as a semi-synthetic steroid, containing the benzenoid nucleus in the structure. Either of these two compounds may be found as the powerful oestrogenic component in the variety of proprietary oral contraceptives.The earliest known synthetic progestogen was the 17-ethinyltestosterone, known as ethisterone, but this has been largely replaced by more potent compounds in recent years. Djerassi, Miramontes, Rosenkranz and Sondheimerl investigated derivatives of 19-nor- testosterone, which itself proved to be the least hazardous of the compounds examined, and these derivatives are now widely used. Among common oral progestogens are norethisterone (17-ethinyl-19-nortestosterone) and its acetate, and norethynodrel, which differs only in the position of a double bond. In modern proprietary oral contraceptives, progestogens are usually present at dose levels between 1 and 5 mg, whereas the ethinyloestradiol or mestranol component rarely exceeds 0.1 mg, and is frequently present at half this level.Some proprietary products are so compounded that the oestrogenic hormone is present alone in several of the tablets, the remaining month’s supply containing both oestrogen and progestogen. Analytically the problem presented to the public analyst is fundamentally one of ensuring that the dispensed medicament contains the labelled ingredients and thus conforms with the quantitative declaration. Because of the extremely close relationship to each other of members within each class of compound, the method of quantitative assessment must, if possible, be also one of identification. In those tablets containing both oestrogen and progestogen, the method applied must enable the minor oestrogenic constituent to be accurately determined.Lastly, the sample size submitted for analysis is severely limited, so that any method that is devised must, if possible, be such that little more than about two tablets are sufficient for an assay. The available analytical literature dealing specifically with this class of compound is scarce and, while it is appreciated that individual manufacturers may be fully conversant with methods of assay for their particular material, the public analyst is somewhat at a disadvantage in having the responsibility on behalf of the public of ensuring that such proprietary products from any manufacturers are of the correct composition. 0 SAC and the author.KEAY 29 The following three basic analytical techniques were used in evolving a scheme of analysis that would satisfy the above requirements.Ultraviolet spectrophotometric examination. Thin-layer chromatography. Infrared spectrophotometry. The scheme evolved consists of solvent extraction of the active ingredients, followed by identification by using thin-layer chromatography or infrared spectrophotometry. The active ingredients in all instances but one are then quantitatively determined by ultraviolet spectrophotometry . Only two oestrogens, mestranol and ethinyloestradiol, occurred in all of the tablets examined, and these differ from one another only by the substituent on the 3 carbon atom. The similarity between these oestrogens requires their previous identification by thin-layer chromatography before determination. Wavelength, rnp Fig.1. Ultraviolet absorption curve of mestranol in methanol Both give similar ultraviolet spectra with a main peak at about 280mp, followed by another peak on the shoulder at about 288mp (Fig. 1). As can be seen from Fig. 1, the slope of the main peak absorption is re-joined at about 290 mp. The main peak was subject to some interference from the progestogens present so that the shoulder peak was used for the determination of the oestrogen, except when it was known that no progestogen was present. The height of the shoulder peak was found to be proportional to concentration, which enabled it to be used for determining the oestrogen content. EXPERIMENTAL REAGENTS- ,411 reagents should be of analytical-reagent grade whenever possible. A ntimony trichhride . Carbon tetrachloride.Chloroform. Cy clo hexane . Ethyl acetate. Hydrochloric acid, 20 per cent. VIV. Methanol. Potassium bromide. Silica gel G-Obtainable from E. Merck and Co. Inc.30 KEAY: ANA4LYTICAL EVALUATION OF GESTOGENS IN ORAL CONTRACEPTIVES [Afia&St, VOl. 93 APPARATUS- Calibrated jhsks, 10 and 100 ml-These were made of low-actinic glass. Thin-layer plates, 20 x 5 em. Thin-layer spot remover-See Fig. 2. Ultraviolet light source-A mercury vapour lamp was used. PREPARATION OF THIN-LAYER PLATES- Prepare a slurry of 30 g of silica gel G mixed with 60 ml of water, and coat plates to give an absorbent thickness of 300 p. Activate the plates by heating a t 110" C for 30 minutes. PROCEDURE- Weigh an amount of the powdered tablets equivalent to about 6 mg of progestogen and transfer it into a 100-ml calibrated flask with 50ml of chloroform.Shake the mixture continuously for 30 minutes. Make up to volume with chloroform, mix and filter. Use this solution for identification and assay of the steroids. IDENTIFICATION- Thin-layer chromatography-Evaporate an aliquot of the chloroform solution to dryness and dissolve the residue in 1 ml of a solution of chloroform - methanol (1 + 1 v/v). Place a spot of this solution, at least 25 ply on the thin-layer plate at a distance of 2 cm from the base of the plate. A solution containing authentic steroids, in the proportion encountered in the tablet under examination, was also spotted on the same plate. Develop the plate in the solvent system cyclohexane - ethyl acetate (1 + 1 v/v) until the solvent front is within 1 cm of the top of the adsorbent, thus giving a development run of about 16 cm.Dry the plate at 100" C for 10 minutes and spray, while still warm, with a saturated solution of antimony trichloride in chloroform. Observe the spots produced both in daylight and under ultra- violet light. It was found that although the colours faded fairly rapidly in daylight, they remained fairly stable under ultraviolet light for at least 24 hours. The method is essentially that of Golab and Layne.2 The results obtained are shown in Table I. TABLE I DETECTION OF GESTOGENS AFTER SEPARATION BY THIN-LAYER CHROMATOGRAPHY Oestrogens- Progestogens- Ethinyloestradiol . . Mestranol . . .. Chlormadinone acetate Norethisterone . . Norethynodrel . . Ethynodiol diacetate Megestrol acetate .. Lynoestrenol .. RF .. 0.54 .. 0.68 .. 0.42 .. 0.47 .. 0-56 .. 0-70 I . 0.74 .. 0.40 Spot under Spot in daylight ultraviolet light Purple Violet Yellow Orange Pale yellow Yellow Violet Red Purple Purple Red Pink Red - violet but turns Lime Virtually colourless Pale yellow to blue - grey INFRARED SPECTRA- Progestogens-Evaporate an aliquot of the chloroform solution directly on to powdered potassium bromide. Compound the resultant homogeneous mass into a disc of suitable thickness, with a pressure of 15 tons on the ram and a vacuum of less than 2 cm of mercury. Record the spectrum and compare against a standard spectrum. Oestrogens-Because of the small amount of oestrogen compared with progestogen, the oestrogen must first be separated and concentrated by means of thin-layer chromatography.Evaporate an aliquot of the chloroform solution to dryness and dissolve the residue in 1 ml of chloroform - methanol (1 + 1 v/v). Place duplicate spots of solution, at least 50 pl, on thin-layer plates and develop, as before, with cyclohexane - ethyl acetate (1 + 1 v/v). Dry at 100" C for 10 minutes. Mask off a portion of plate containing one of the spots and spray the exposed portion, while still warm, with a saturated solution of antimony trichloride in chloroform. Remove the located oestrogen spot from the unsprayed portion of plate byJanuary, 19681 KEAY: ANALYTICAL EVALUATION OF GESTOGENS IN ORAL CONTRACEPTIVES 31 means of the extractor (see Fig. 2). Extract with chloroform and evaporate the chloroform solution directly on to powdered potassium bromide.Compound into a disc, as before, and record the spectrum. The infrared spectra obtained exhibit maxima only at the same wavelengths as authentic specimens of gestogens prepared in a similar manner. A = Suction head B = Filter tube, with porosity C = Receiver tube 3 sinter Fig. 2. Apparatus for removal of spot from thin-layer plate ULTRAVIOLET ASSAY- Progestogens-Transfer by pipette a 20-ml aliquot of the chloroform solution into a 100-ml calibrated flask. Evaporate it to dryness by using a current of air. Dissolve the residue in methanol and make up to 100 ml. Measure the extinction of the solution at the maximum, with l-cm silica cells and methanol as blank. If norethynodrel is present it must be isomerised to norethisterone.Add 1 ml of 20 per cent. v/v hydrochloric acid to the residue dissolved in 40 ml of methanol. Allow the solution to stand for 1 hour at room temperature and make up to 100 ml. Measure the extinction of the solution at the maximum, with l-cm silica cell and methanol as blank. Table I1 gives determined values for progestogens. TABLE I1 DETERMINED VALUES FOR PROGESTOGENS Progestogen Wavelength of maximum extinction, mp E,l:m Lynoestrenol . . . . . . .. 200 Norethynodrel (isomerised) . . .. 240 Norethisterone . . .. .. .. 240 Norethisterone acetate . . .. .. 240 Chlormadinone acetate . . .. .. 285 Megestrol acetate.. .. .. .. 289 300 670 560 506 510 650 Oestrogens present alone or with a Progestogen showi.pzg a maximztm at lower than 250 mp- Evaporate a 50-ml aliquot of the chloroform solution, containing about 200 pg of oestrogen, to dryness in a current of air.Dissolve the residue in methanol and make up to 10 ml in a calibrated flask. Measure the extinction of the resulting solution in a 1-cm silica cell at the following wavelengths : mestranol, 285, 287.5 and 290 mp; and ethinyloestradiol, 286, 288.5 and 291 mp. The content of mestranol in milligrams per tablet is given by the expression- and the content of ethinyloestradiol in milligrams per tablet by- where Y is the average tablet weight in grams and W is the weight in grams of sample taken. Oestrogens firesent with a pyogestogen showing a maximum in the region 280 to 290 mp- Evaporate 50 ml of the chloroform solution to dryness and dissolve the residue in 1 ml of chloroform - methanol (1 + 1 v/v).Place 0.5 ml of this solution on to a thin-layer plate at a distance of 2 cm from the base of the plate. Place a spot of standard oestrogen alongside32 KEAY: ANALYTICAL EVALUATION OF GESTOGENS IN ORAL CONTRACEPTIVES [Analyst, Vol. 93 and develop, as before, but with the solvent cyclohexane - ethyl acetate - carbon tetrachloride (2 + 2 + 1 v/v). Remove the plate, dry and heat at 100" C for 10 minutes. Mask off the portion of the plate containing the sample. Locate the standard spot by spraying with a saturated solution of antimony trichloride in chloroform. Remove the area from the plate containing the untreated sample spot with the extractor and extract oestrogen from the silica gel with methanol.Make the extract up to 10 ml with methanol. Prepare a blank by extracting a band of similar size from the plate containing no steroids. Measure the extinction of the solution in a l-cm silica cell at the same wavelengths as given above. The calculation of oestrogen content is as previously outlined, but the result must be multiplied by 2. After thin-layer separation, or when the tablet is composed of oestrogen alone, the concentration is found by measuring the extinction of the methanol solution at the maximum, at about 280mp. Ethinyloestradiol is soluble in 0.1 N sodium hydroxide and the wavelength of maximum extinction shifts from 280 to 300 mp. The measurement of the extinction of 0.1 N sodium hydroxide solution, at the maximum, provides an alternative means of assay of ethinyl- oestradiol, and also differentiates it from mestranol.Determined values for oestrogens are shown in Table 111. Oestrogen TABLE I11 DETERMINED VALUES FOR OESTROGENS Wavelength of maximum Solvent extinction, m p E;%& Mestranol (main peak) . . .. .. Methanol 279 Mestranol (shoulder peak) . . . . Methanol 287.6 Ethinyloestradiol (main peak) . . . . Methanol 281.5 Ethinyloestradiol (shoulder peak) . . Methanol 288.5 Ethinyloestradiol .. .. .. 0.1 "aOH 298.5 82 14-4 89 11.4 81 THIN-LAYER ASSAY- A solution of ethynodiol diacetate in methanol showed no peaks in the ultraviolet and gave complete absorption below 210 mp. This substance was therefore determined by thin- layer chromatography as follows. Evaporate an aliquot of the chloroform solution to dryness and dissolve the residue in a known volume of methanol - chloroform (1 + 1 v/v).Spot on to a thin-layer chromato- graphic plate, at a distance of 2 cm from the base of the plate, an aliquot of the sample solution containing about 10 pg of ethynodiol diacetate. Alongside this spot place standard spots in the range 8 to 12 pg. Develop the plate with cyclohexane - ethyl acetate (1 + 1 v/v) until the solvent front is 1 cm from the top of the plate. Dry the plate at 100" C for 10 minutes. While still warm, spray with a saturated solution of antimony trichloride in chloroforni. From the intensity of the spot colours, calculate the ethynodiol diacetate content of the sample. RESULTS Table IV summarises the results obtained by using the method described to determine mestranol recovery from a base made up of purified talc, magnesium stearate, lactose and starch.TABLE IV RECOVERY OF MESTRANOL FOR TABLET BASES Mestranol added, pg Recovered, pg 0 125 250 500 1000 - 126 248 486 1006 In Tables V and VI the results obtained by using the methods described to determine progestogens and oestrogens in proprietary preparations are summarised.January, 19681 KEAY: ANALYTICAL EVALUATION OF GESTOGENS IN ORAL CONTRACEPTIVES 33 TABLE V RESULTS FOR THE DETERMINATION OF PROGESTOGENS IN COMPOUNDED TABLETS Norethisterone . . .. Norethisterone acetate . . Norethynodrel . . .. Chlormadinone acetate . . Megestrol acetate .. Ethynodiol diacetate . . Lynoestrenol .. .. Nominal content, mg per tablet .. 2.0 2.0 .. 2.6 3-0 4.0 .. 2.6 2.6 2.5 5-0 6.0 ..2.0 .. 4-0 1.0 .. 1.0 1.0 2.0 .. 2.5 6.0 TABLE VI Found, mg per tablet 2-19 2.18 2-60 3.10 3.80 2.30 2.33 2-56 4.70 4.83 2-05 3.70 1.01 1-00 1.00 2.00 2-70 6-36 RESULTS FOR THE DETERMINATION OF OESTROGENS IN COMPOUNDED TABLETS Nominal content, Found, Oestrogens pg per tablet pg per tablet Mestranol . . .. .. .. 76 77 75 72 75 70 80 78 80 81 100 98 100 100 100 97 100 94 100 100 100 91 1 00 95 100 97 100 99 160 144 Ethinyloestradiol .. .. 50 60 50 46 50 48 60 46 100 108 100 108 DISCUSSION Examination of samples of sixteen proprietary oral contraceptives gave analytical results that were all within 10 per cent. of the declared ingredient value. This was considered to be a reasonable tolerance for the techniques used, although a 10 per cent. deficiency would be regarded with somewhat more concern than a 10 per cent. excess. A deficiency of this magnitude was not found, and recovery experiments and standard additions showed the accuracy of the methods used to be better than 5 per cent. The amount of sample used for the complete analysis was four tablets, and this was found to be sufficient for all stages of the scheme. I express my thanks to the pharmaceutical companies that have supplied authentic substances for reference purposes, and to Coventry Corporation, in whose laboratories the work was carried out, for permission to publish this paper. REFERENCE 1. 2. Djerassi, C., Miramontes, L., Rosenkranz, G., and Sondheimer, F., J . Amer. Chem. Soc., 1964, Golab, T., and Layne, D. S., J . Chrontat., 1982, 9, 321. Received February 2nd, 1967 76, 4092.