`microtubule structures (18). However, it has little effect on ornithine
`decarboxylase activity, and this fact could be why it is ineffective in the
`inhibition of neuroblastoma growth.
`Although bromoacetylcholine binds to the nicotinic receptor at the
`neuromuscular junction irreversibly (19), as does a-bungarotoxin (20),
`the latter compound produced little effect on neuroblastoma growth,
`indicating that bromoacetylcholine inhibits neuroblastoma growth
`through an action mechanism that differs from a-bungarotoxin, similar
`to action at cholinergic receptors. Neuroblastoma cells possess adrenergic,
`cholinergic, and nonspecific receptors but very few serotonergic receptors.
`It thus is understandable that neuroblastoma cell growth was not in-
`hibited by a serotonergic neuron degenerator, 5,6-dihydroxytrypt-
`amine.
`Among the routes of administration studied, intratumor administra-
`tion for bromoacetylcholine, bromoacetate, and 1,3-diaminopropane and
`intraperitoneal administration for cyclophosphamide were the best.
`In summary, it seems that drugs capable of inhibiting ornithine de-
`carboxylase can suppress the cell growth of neuroblastoma. A more potent
`ornithine decarboxylase inhibitor that produces few side effects may be
`developed as an effective weapon to treat neuroblastoma.
`
`REFERENCES
`
`(1) C. Pochedly, “Neuroblastoma,” Publ. Sci. Group, Inc., Acton,
`Mass., 1976.
`(2) J. Z. Finklestein, E. Arima, P. E. Byfield, J. E. Byfield, and E. W.
`Fonkalsrud, Cancer Chemother. Rep., 57,405 (1973).
`(3) C. Y. Chiou, J . Pharm. Sci., 64,469 (1975).
`(4) Ibid., 66,837 (1977).
`
`(5) Ibid., 67,331 (1978).
`(6) S. K. Chapman, M. K. Martin, M. S. Hoover, and C. Y. Chiou,
`Biochem. Pharmacol., 27,717 (1978).
`(7) C. Y. Chiou and B. V. R. Sastry, ibid., 17,805 (1968).
`(8) C. Y. Chiou, N. E. Liddell, M. K. Martin, and C. J. Chu, Arch. Int.
`Pharmacodyn. Ther., 233,235 (1978).
`(9) A. Bjorklund, H. G. Baumgarten, and A. Nobin, Adu. Biochem.
`Psychopharmacol., 10,13 (1974).
`(10) K. Fuxe and G. .Jonsson, ibid., 10,l (1974).
`(11) C. Y. Chiou, C. J. Chu, and N. E. Liddell, Arch. Int. Pharmaco-
`dyn. Ther., 235,35 (1978).
`(12) U. Bachrach, Proc. Natl. Acad. Sci. U S A , 72,3087 (1975).
`(13) J. L. Clark and P. Duffy, Arch. Biochem. Biophys., 172, 551
`(1976).
`(14) Z. N. Canellakis and T. C. Theoharides, J. Biol. Chem., 251,4436
`(1978).
`, - . . -, .
`(15) T. C. Theoharides and Z. N. Canellakis, Nature (London), 255,
`733 (1975).
`(16) P. P. McCann, C. Tardiff, P. S. Mamont, and F. Schuber, Bio-
`chern. Biophys. Res. Commun., 64,336 (1975).
`(17) K. J. Lembeck, Biochim. Biophys. Acta, 354,88 (1974).
`(18) P. Calabres and R. E. Parks, in “The Pharmacological Basis of
`Therapeutics,” L. S. Goodman and A. Gilman, Eds., Macmillan, New
`York,N.Y., 1 9 7 5 , ~ . 1284.
`(19) C. Y. Chiou, Eur. J . Pharmacol., 26,268 (1974).
`(20) C. C. Chang, J . Formosan Med. Assoc., 59,315 (1960).
`
`ACKNOWLEDGMENTS
`
`Supported in part by American Cancer Society Grant CH-81
`
`Effect of pH, Chlorobutanol, Cysteine Hydrochloride,
`Ethylenediaminetetraacetic Acid, Propylene Glycol,
`Sodium Metabisulfite, and Sodium Sulfite on Furosemide
`Stability in Aqueous Solutions
`KIRIT A. SHAH *, V. DAS GUPTA”, and KENNETH R. STEWART *
`
`Received November 2,1979, from the College of Pharmacy, University of Houston, Houston, T X 77004.
`*Present address: College of Pharmacy, Texas Southern University, Houston, TX 77004.
`20, 1979.
`Houston, TX 77030.
`
`Accepted for publication December
`$Ben Taub Hospital Pharmacy,
`
`Abstract 0 A study was conducted to determine the effects of pH, two
`antioxidants, a chelating agent, a preservative, and propylene glycol on
`t‘urosemide stability. Aqueous solutions of furosemide containing 10%
`alcohol (v/v) were prepared in phosphate buffers with various pH values
`(5,6, and 9) whose ionic strength was adjusted to 0.1 M with potassium
`chloride. Some solutions contained chlorobutanol, ethylenediamine-
`tetraacetic acid, or sodium metabisulfite. Another set of aqueous solutions
`contained phosphate buffer (0.1 M ) , alcohol (10% v/v), and propylene
`glycol (409 v/v) with or without cysteine hydrochloride, ethylenedia -
`minetetraacetic acid, and sodium sulfite. Thesolutionswere divided into
`two parts, stored at 24 and SOo, and assayed frequently using a previously
`developed high pressure liquid chromatographic procedure. At the lowest
`pH value (pH 5). furosemide appeared to be very unstable. Cysteine
`
`hydrochloride, ethylenediaminetetraacetic acid, and sodium sulfite failed
`to improve the stability of furosemide. Chlorobutanol and sodium met-
`abisulfite had an adverse effect on the stability, probably due to the fact
`that they decreased the pH of the solution. The pH value appears to be
`the only critical factor for the stability of furosemide. Buffered solutions
`containing propylene glycol were very stable at both temperatures for
`170 days, and they tasted good.
`in aqueous solutions, effect of
`Keyphrases 0 Furosemide-stability
`formulation factors 0 Diuretics-furosemide,
`stability in aqueous so-
`lutions, effect of formulation factors 0 Stability-furosemide
`in aqueous
`solutions, effect of formulation factors
`
`Furosemide (I) is a widely used diuretic, but little in-
`formation is available concerning the stability of this drug
`in dosage forms. Rowbotham et al. (1) reported that
`aqueous furosemide solutions undergo hydrolysis and
`photochemical degradation. Quantification of photo-
`chemical degradation products of furosemide by the USP
`XIX (2) UV assay procedures was not successful. A sta-
`bility-indicating assay for furosemide using high-pressure
`
`liquid chromatography (HPLC) was developed by
`Ghanekar et al. (3). It also was reported that an aqueous
`furosemide solution containing sorbitol and 10% alcohol
`(v/v) had limited stability. The pH of the solution was
`adjusted to -8.5. However, it was difficult to maintain the
`pH value of the solution, which caused rapid decomposi-
`tion.
`The objectives of the present investigation were to study
`
`594 I Journal of Pharmaceutical Sciences
`Vol. 69, No. 5, May 1980
`
`0022-35491 801 0500-0594$0 1.0010
`@ 1980, American Pharmaceutical Association
`
`MYLAN INST. EXHIBIT 1120 PAGE 1
`
`MYLAN INST. EXHIBIT 1120 PAGE 1
`
`
`
`Table I-Prepared Solutions
`
`(0.5 mg of Furosemide/ml)
`
`Table 111-Assay Results at 50"
`
`PHI
`(fO.l)
`5.2
`3.3
`5.2
`3.0
`
`Percent of Label Claim
`170 Days
`90 Days
`-
`93.5
`-
`51.8
`88.7
`-
`-
`-
`
`240 Days
`Solution 35 Days
`94.7
`94.5
`1
`2
`80.9
`3.5c
`3
`94.8
`79.3
`4
`15.3"
`-
`5-12b
`-
`-
`99.5
`8.4
`100.1
`13
`-
`-
`14
`8.2
`99.2
`100.1c
`-
`-
`99.4
`8.4
`15
`99.4
`-
`-
`100.0
`8.3
`16
`100.7
`-
`-
`101.1
`17
`100.5
`8.0
`a Crystals were found in the solution, so it was not reassayed. Not studied at
`Color had changed to
`50' since they were not stable even at room temperature.
`light yellow.
`
`ml with water and are reported in Table I. All of the solutions prepared
`are listed in Table I.
`The solutions were divided into two portions, stored at 24 and 50", and
`assayed frequently using the HPLC method reported previously (3),
`except that the absorbance unit for full-scale deflection was 0.1 instead
`of 0.16 and the injection volume was 20.0 instead of 40.0 pl. The results,
`which were calculated as previously reported (31, are presented in Tables
`I1 and 111. Sample chromatograms are presented in Figs. 1 and 2.
`
`RESULTS AND DISCUSSION
`For convenience, the results from Solutions 1-12 (Table I), which did
`not contain any propylene glycol, will be discussed first, followed by the
`results from Solutions 13-17 (Table I), which contained 40% (v/v) pro-
`pylene glycol.
`
`I
`
`w
`B
`v) 2
`
`v) w
`II:
`a
`0
`t;
`x w
`I-
`
`A
`
`B
`
`I
`
`*
`
`D d
`
`k-
`0
`W
`
`7 z
`
`Buffer
`Solution
`(fO.l) pH
`
`Solution
`
`Other Ingredients
`
`None
`~ 5
`1
`2
`O.Z%-Chlorobutanol
`5
`0.05% I11
`5
`3
`0.1,0.2, and 0.4% V, respectively
`4-6
`5
`0.1,0.2, and 0.4% V, respectively
`6
`7-9
`0.1,0.2, and 0.4% V, respectively
`9
`10-12
`n.4
`None
`13
`~-
`0.02% I1
`8.2
`14
`0.05% IV
`8.4
`15
`0.05% I1 and 0.05% IV
`8.3
`16
`0.05% II.0.05% IV, and 0.05% 111
`8.0
`17
`(I All solutions contained 10% alcohol (vlv) since the stock solution was made in
`alcohol; Solutions 13-17 contained propylene glycol, and the pH values of the final
`solutions were those obtained after diluting 1:lO with water.
`
`Table 11-Assay Results at Room Temperature
`
`PHf
`Percent of Label Claim
`Solution 35 Days 90 Days 152 Days 170 Days 240 Days
`( f O . l )
`-
`95.9
`5.2
`1
`100.0
`-
`97.7
`96.9
`5.1
`2
`-
`3
`96.4
`93.3
`5.2
`-
`5.2"
`3.3
`4
`-
`-
`-
`5
`15.1a
`2.9
`-
`-
`6
`44.8"
`2.8
`-
`-
`7
`91.8
`5.0
`-
`-
`8
`94.7
`4.5
`-
`9
`98.3
`5.4
`-
`82.1
`10
`98.1
`7.5
`-
`-
`95.2
`6.9
`11
`-
`-
`-
`91.7
`5.7
`12
`-
`1.3
`100.6
`99.7
`8.4
`-
`i4
`100.7
`8.2
`100.8
`-
`15
`99.5
`99.5
`8.4
`-
`16
`100.8
`100.4
`8.3
`-
`17
`100.9
`100.2
`8.0
`0 Crystals were found in the solution, so it was not reassayed Color had changed
`to light yellow (from colorless). Fungus.
`
`the effect of pH, chlorobutanol, cysteine hydrochloride
`monohydrate (II), ethylenediaminetetraacetic acid (111),
`sodium sulfite (IV), and sodium metabisulfite (V) on the
`stability of furosemide in aqueous solutions and to develop
`an appropriate buffering system to maintain the adjusted
`pH value.
`
`EXPERIMENTAL
`
`chemicals and reagents were USP,
`Chemicals and Reagents-All
`NF, or ACS grade and were used without further purification. Furo-
`semide' powder was used as received.
`Apparatus-The
`apparatus was the same as that reported previously
`(3), except that a variable-wavelength detector2 set at 254 nm was used.
`The column also was the same as that reported previously (3).
`Preparation of Solutions-A stock solution of furosemide was pre-
`pared by dissolving 1.25 g of furosemide in enough alcohol to make 250
`ml. The solutions for stability studies were prepared by diluting an ap-
`propriate quantity of the stock solution with a buffer of an appropriate
`pH value (0.05 M phosphate buffers of pH 5,6, and 9 whose ionic strength
`was adjusted to 0.1 M with potassium chloride). The phosphate buffers
`were prepared according to the USP procedure (4). Before the solutions
`were diluted to volume, additional ingredients, if any, were added.
`Another set of solutions was prepared by dissolving an appropriate
`quantity of furosemide in 50.0 ml of 0.2 M aqueous dibasic potassium
`phosphate solution and then adding 10 ml of ethanol and bringing the
`solution to volume (100.0 ml) with propylene glycol. Additional ingre-
`dients, if any, were dissolved before bringing the solution to volume. The
`pH values of these solutions were determined after diluting 5.0 ml to 50.0
`' Generously supplied by Hoechst-Roussel Pharmaceuticals, Somerville, N.J.
`* Spectroflow monitor 770, Schoeffel Instrument Corp.. Westwood, N.J.
`
`dl
`3
`0
`
`3
`
`
`
`0
`
`
`
`L-
`5
`Figure 1-Sample chromatograms where peak 1 is from furosemide.
`Key: A, standard solution of I; 8, solution of I(0.057;) in pH 5 buffer
`after 240 days of storage at 50°: C , same as B except that it also con-
`tained 0.2% chlorobutanol; and D , same as B except that it also con-
`tained 0.05", III.
`
`~
`
`~~
`
`Journal of Pharmaceutical Sciences I 595
`Vol. 69, No. 5. May 1980
`
`MYLAN INST. EXHIBIT 1120 PAGE 2
`
`MYLAN INST. EXHIBIT 1120 PAGE 2
`
`
`
`all of the oxygen was consumed, protected furosemide from the oxidation
`effect of the sulfate ion. This effect also has been reported (5).
`The described effects were not found in Solutions 10-12, probably due
`to their high pH values (Table 11). In these solutions, the effect of pH
`appears to be more prominent. Moreover, in solutions with basic pH
`values, furosemide does not appear to be susceptible to oxidation. The
`same may be true of solutions with pH values above 5.5.
`Crystals of furosemide were found in several solutions after 35 days,
`especially in those kept at room temperature (Table 11, Solutions 4-6).
`Furosemide is known to be poorly soluble in aqueous systems, especially
`acidic ones.
`Several other solutions (Solutions 7,9,11, and 12) became discolored
`after 152 days (Table 11). The cause of this discoloration was not deter-
`mined.
`It was not possible to treat the data mathematically because of very
`little decomposition (Solutions 1-3) or changes in pH values (Solutions
`4-12).
`Solutions 13-17 (with Propylene Glycol)-Solutions 13-17 proved
`to be very stable at both 24 and 50’. Considering experimental errors,
`the results of both temperatures were almost 1 W o for all solutions at 170
`days (Tables I1 and 111). Furthermore, the pH values of the solutions did
`not change (Tables I1 and 111). However, the addition of cysteine hy-
`drochloride monohydrate (II), ethylenediaminetetraacetic acid, sodium
`sulfite, or their combinations did not improve furosemide stability. Al-
`though the solution containing I1 became discolored, the loss in furo-
`semide potency was negligible. Compound I1 apparently was oxidized
`in this solution. This discoloration did not occur when I1 was present in
`combination with sodium sulfite due to the protection provided by the
`sodium sulfite. Nevertheless, the addition of these ingredients is not
`desirable. In weakly basic solutions, furosemide does not appear to be
`susceptible to oxidation.
`The only critical factor in furosemide stability is the pH of the solution,
`which must he slightly basic and not change on storage. A stable liquid
`dosage form of furosemide can be prepared by dissolving an appropriate
`quantity of furosemide powder in 0.2 M aqueous dibasic potassium
`phosphate solution (50% of the total volume), adding alcohol (10% v/v),
`and bringing the solution to volume with propylene glycol. Since the
`solution did not lose potency at 50” for -6 months, it should be stable
`for a t least 2 years at room temperature. A panel of three persons ap-
`proved the taste of the product.
`
`REFERENCES
`(1) P. C. Rowbotham, J. B. Standrod, and J. K. Sugden, Pharm. Acta
`Helu., 51,305 (1976).
`(2) “The United States Pharmacopeia,” 19th rev., Mack Publishing
`Co., Easton, Pa., 1975, p. 213.
`(3) A. G. Ghanekar, V. D. Gupta, and C . W. Gibbs, J. Pharm. Sci., 67,
`808 (1978).
`(4) “The United States Pharmacopeia,” 19th rev., Mack Publishing
`Co., Easton, Pa., 1975, pp. 653,654.
`(5) R. Bonevski, J. M. Culjat, and L. Balint, J . Pharm. Sci., 67, 1474
`(1978).
`
`A
`
`B
`
`C
`
`d
`
`I- u
`
`z w, -
`
`I
`
` -
`
`3
`
`0
`
`5
`MINUTES
`Figure 2-Sample chromatograms where peak 1 is from furosernide.
`Key: A, solution of I (0.0570) in pH 5 buffer with 0.1 % sodium meta-
`bisulfite after 170 days of storage at room temperature; B, same as A
`except that pH 6 buffw was used; and C, same as A except that pH 9
`buffer was used.
`Solutions 1-12 (without Propylene Glycol)-The
`results indicate
`that Solutions 1-12 were not very stable for 170 days either at room
`temperature or at 50’ (Tables I1 and 111). Furthermore, the additional
`ingredients, ethylenediaminetetraacetic acid, chlorobutanol, or sodium
`metabisulfite, did not improve furosemide stability. In fact, chlorobutanol
`and sodium metabisulfite had an adverse effect on its stability, especially
`when added with pH 5 buffers. This effect may be due to a decrease in
`the pH of the solutions (Table 111). Furosemide is known to be unstable
`at lower pH values (3). The pH values of most solutions changed after
`storage (Tables I1 and 111).
`The effect of sodium metabisulfite (as the concentration was raised
`from 0.1 to 0.410) on furosemide stability is interesting. In Solutions 4-6
`and 7-9 (Table II), the stability of furosemide improved as the concen-
`tration of sodium metabisulfite was increased, perhaps due to initial
`oxidation of metabisulfite to sulfate by the oxygen present. The sulfate
`ion increased the rate of decomposition (oxidation) of fwsemide. A
`similar scheme using epinephrine has been reported (5).
`Furthermore, the additional amount of sodium metabisulfite, after
`
`506 I Journal of Pharmaceutical Sciences
`Vol. 69, No. 5, May 1980
`
`MYLAN INST. EXHIBIT 1120 PAGE 3
`
`MYLAN INST. EXHIBIT 1120 PAGE 3
`
`