CA1258427A - Process for the preparation of an aqueous insulin formulation which has been stabilized towards mechanical stresses, and its use - Google Patents

Abstract

Abstract of the disclosure:

The invention relates to a process for the prepara-tion of an aqueous insulin formulation which is stabilized towards mechanical stresses, has an insulin concentration of more than 1 I.U./ml and has a viscosity of at least 1.75 mPa.s at 4°C, which comprises adding a physiologic-ally acceptable thickener, by itself or additionally with a physiologically acceptable surface-active substance, to an aqueous insulin formulation. The invention furthermore relates to the use of such an agent as a medicine in the treatment of diabetes mellitus, in particular for use in devices for continuous release of insulin or as a medicine with a delayed action, the use of a physiologically acceptable thickener, by itself or together with a physio-logically acceptable surface-active substance, for stabilizing the abovementioned aqueous insulin formula-tions, and the use of a thickener as a depot auxiliary in aqueous insulin formulations.

Description

~2S~34~7 The invent;on relates to the preparation of novel insulin formulations with a delayed action~ the prepara-tion of stab;lized insulin formulations for use in auto-ma~;c metering apparatuses, and the use thereof, in parti-cular for the treatment of diabetes mellitus.
It is ~enerally known that particular requirementsare imposed on parenteral replacement therapy w;th insu-lin. These include, in part;cular, the question of de-layed pharmacok;netics, ~hich enable the diabet;c to be 10 stabilized with one or a few injections per day. A fe~
clinically proven principles exist to achieve such depot effects, these principles including the use of zinc or protamine sulfate as depot auxiliaries.
These kno~n depot principles are based on the 15 physical effect of the slow redissolving of an insulin form ~hich is spar;ngly soluble at the physiological pH, for example the 2-zinc crystal form~ If the product al-ready has a neutral pH, ~hich is advantageous from the point of chemical stab;lity and long storage, these pro-20 ducts are suspensions uhich must be very carefully homoge-n;zed by shak;ng before metering, in order to avoid ;n-correct dosages.
The insulin depot products previously kno~n fur-ther~ore have very specific act;on profiles ~hich can be 25 varied only w;thin certain limits by adding dissolved in-sulin. There are al~ays patients for ~hom alternat;ve action profiles~ for example those with a some~hat less rapid onset of an action of about the same length, are desirable. ~f the physician has such products available, 30 he can suit the specific habits and charac~eris~ics of the diabetic. In contrast, if the patient ~ere to be required to change his habits, ~his ~ould lead ~o the problem of patient "compliance"~ uhich in the end effect substan-tially influences the therapeutic result.
Kraegen et al. report, in ~rit~ Medu J~ 1975, 3 464-466D a stabilizing effect of adding up to 3.5X of ;12Sl34~

Haemaccel to very dilute insulin solutions (0.0~ I.U./ml~, by ~hich means, inter alia, the adsorption of the insulin in the stock vessel and tube system in infusion sys tems can be prevented.
It has no~ been found that aqueous insulin formu-lations which have a viscosity of at least 1.75 mPa.s at 4C and an insulin concentration of more than 1 I.U./ml surprisingly exhibit an improved physical stability and other i0pro~ed properties, for example ;n respect of the act;on profile. According to one embodiment of the inven-tion, the insulin formulations can also additionally con-ta;n a phys;olog;cally acceptable surface-actîve sub-stance, ~h;ch means that the stab;l;ty, ;n particular ;n peristalt;c pumps, can be further ;ncreased. These formu-lat;ons are particularly h;ghly stable to~ards mechan;calstress, in particular under elevated temperature, for example shak;ng and pump;ng movements.
The quest;on of the stability of insulin formula-tions has already previously been a ser;ous problem Thus, ;t ;s kno~n that dissolved proteins such as insulin are adsorbed at interfaces (including also the aqueous solutionta;r interface) (C~No Cumber and A.E. Alexander Trans. Faraday Soc. 46, 235 (1950)). Various secondary reactions are observed as a result of this adsorption at interfaces, and these are generally referred to by the term "denaturing". A change in the shape of the protein molecules adsorbed occurs tchange in the tertiary and/or secondary structure~. In add;t;on~ aggregation of ad-sorbed molecules to soluble or ;nsoluble polymer;c forms may also occur. The ~urbulence ~hich occurs ~hen ;nsulin solut;ons pass through narro~ channels also appears to promote insul;n denaturing.
The tendency of insulin to precipitate out of com-mercially ava;lable solut;ons and thereby to block mechan;cal components and supply lines have proved to be the main obstacle ;n further development and clinical use of continuous infus;on devices. There ;s also the ten-dency to reduce the size of these devices in order thus to ob~ain systems ~hich can be implanted, ~hich resul~s ~58 ~Z~

in a need for highly concentrated stable insul;n solu-tions~ ~hich in turn renders the above problems even more serious.
The question of the physical stability of insulin solutions has been under particular discussion since the development of aueomatic metering apparatuses. Tt is generally kno~n that specially stabilized insulins must be used in sùch apparatuses. In connection ~ith the in-adequate physical stab;lity of insulins~ not only is the 10 reduced biological activity under discussion, but recently also a process for the formation of amyloid A protein in the serum, bhich proceeds by stimulation of the macro-phages and can lead to a~yloidosis ;n various organs ~Brownlee et al., Lancet t1984), 411-413).
A number of proposals have already been made to solve these problems:
Ger~an Patent A-2,917,535 discloses aqueous solu-tions of insulin ~hich contain~ for protection from denatur-ing a surface-active substance of the general formula R
RbO~ CH2-CH-O-)p--RC ~I) in which Ra denotes hydrogen, methyl or ethyl, p denotes a number from 2 to 80, preferably 8 to 45, and Rb and Rc are identical or d;fferent and denote hydrogen, alkyl alcohol radicals ~ith 1 20 carbon atoms, carbo~ylic acid 25 radicals with 2-20 carbon ato~s, alkylphenol radicals with an alkyl chain of 1-10 carbon atoms or alkylamine radicals ~ith 1-20 carbon atoms, as a homopolymer~ block polymer or copolymer in a concentration o~ 2 to 200 mg/l.
European Patent A-18~09 describes aqueous solu-30 tions of insulin and a large number of other proteins ~hich are stable to~ards denatur;ng and contain a surface-active substance with a chain-like basic structure, the ~embers of ~hich contain weakLy hydrophobic and ~eakly hydrophilic regions in alternating arrangement~
~0-A-83tO0288 descr;bes stable aqueous insulin formulations ~h;ch are ~or use in insulin-metering devi-ces~ have a pH of 6.5 to 9 and contain up to 1,000 ppm of ~258~

a polyoxyethylenealkyl ether of the formula R7-o-~CH2-CH2-o~m-H (II), in uhich R7 denotes a saturated or unsaturated (C8-C15)-alkyl group and m denotes an integer from 2 to 25.
Finally, physically stabilized insulin solut;ons are kno~n from German ~atent A-3,240,177, ~h;ch contain stabilizing amounts of a phosphol;pid of the formula III
H-CH-oR4 H- CH- o_ p~OR t I I I ) OH
;n ~hich R4 and R5, ~hich can be identical or differ-ent, represent hydrogen, alkylcarbonyl, alkenylcarbonyl, alkanedienylcarbonyl, alkanetrienylcarbonyl or alkane-tetraenylcarbonyl~ with the proviso that R4 and R5 are not s;multaneously hydrogen, and in ~hich R6 represents a hydroph;lic group.
These surface-active stabil;zers are extremely effect;ve, in that they significantly increase the sta-b;l;ty of ;nsulin solutions to shaking. Shaking is ~ith-out doubt a substantial adverse influence on the insulin in metering apparatuses.
It has now been found, ho~ever, that, especially in peristaltic pumps, the squeezing of the elastomer tube of the pump and/or the shearing effects, such as occur ;n many pump pr;nc;ples~ additionally impair insulin stabil-ity. Precipitation of insulin in pump tubes or catheters may in this way occur, ;n spi~e of addition of the various stabil;zers kno~n h;therto.
Here and in the follo~;ng text, "insulins" are understood as s;ngle products or mixtures of several insu-lins~ and in particular no~ only human insulin and insu-l;ns of an;mal origin, such as mam~alian insulins (forexample from cattle or pigs)O the term also includes insu-lins in the broader sense, i~e. mod;fied insulins, such as de-PheB1-;nsulins tcf., for example~ German Patent

2,005,658 and European Patent A-46~979~ or insulins ~h;ch are modif;ed by basic substituents on the C terminus of 8 ~2~7 the B-chain ~such as insulin-B31-Arg-OH or insulin-B31-~rg-Arg-OH, proposed in German Patent Applications P 33 ~6 472.4, P 33 27 709.5, P 33 33 640.7 and P 33 34 407.8), and human proinsulins or other proinsulins or pro-insulin analogs (cf., for example, German PatentA-3~232,036), and alkali metal and ammonium salts. It is also possible for several of these insulins to be present in a mixture. Depending on the solubility, the insulin concentration can be up to about 1,500 I.U./ml, and is preferably between 5 and about 1,000 I.U./ml. In depot forms, any desired proportion of one or more insulins can be present in each case in the dissolved, amorphous and/or crystalline form, independently of one another.
Possible thickeners talso called gelling agents) are physiologically acceptable polymers, such as collagen and secondary products thereof, such as gelatin, hydroxy-polygelatin (Gelifundol(R)), modified liquid gelatin (Physiogel(R)), gelatin partial hydrolysates, ~hich can also be crosslinked, for example with diisocyanates, tPolygeline and Haemaccel~R)) or polysaccharides and derivatives, for example dextrans, levans and hydroxy-ethyl-starches, or polyvinylpyrrolidone. Some of the substances mentioned are also used in colloidal plasma substitutes.
The result of the thickeners is that the insulin formulation is slightly viscous to viscous or in the form of a hydrogel at low temperatures, for example 4C. The formuLations according to the invention preferably have a viscosity, measured at ~C, of at least 2 and in particular of at least 2.5 mPa.s. The gels already partly liquefy at room temperature or at temperatures close to body temperature The formulations according to the invention, ~hich have a high physical stability~ preferably contain more than 1% by weight, in particular bet~een ~ and 20% by ~eight, of ~hickeners. ~o~ever, hydrogels can aLready be formed when smaLler amounts of thickener are added; for example, the addit;on of about ~.2% of agar or 0.6~ of gelatin is sufficient for this purpose. The upper limit ~ZS84Z>~

of the content of gell;ng agent can be 30% or more, de-pending on its nature.
If the content of thickener is sufficiently high, all the formulations have the common property that they are in solid ~orm as a hydrogel under storage conditions.
This is an advantage in respect of the physical stability, since oligoner formation and denaturing are kno~n to be greatly accelerated by movement, that is to say in prin-ciple cannot be reliably excluded when liquid insulin formulations are handled. Storage in gel form can also be advantageous because this form renains considerably more homogeneous on storage than a solution or suspension.
In the case of sedimented crystal suspensions, for example, it is entirely conceivable that relatively stable crystal associates are formed on prolonged storage, ~hich can then be shaken up less easily to give a homo-~eneous suspension, so that metering errors may occur.
In contrast, if the crystal suspension is homogeneously "frozen" in a gel~ the insulin molecules can diffuse only slo~ly at the vessel uall or at the liquid/air interface, and such effects are to be excluded. Movements and turbu-lences within the gels during handling are also consider-ably reduced. The formulations, in particular those con taining surface-active substances~ thus have a particu-larly good storage stability.
Before use, the gels, as is usual with conven-t;onal insulins, are brought to room temperature to body temperature, ~hereupon they liquefy~ but nevertheless still have a greater viscosity than conventional insulin formulations. Suspensions do not then usually sediment immediately, so that inhomogeneities and metering errors can occur less easily, an inhomogeneity problem does not of course exist uith clear gel formulations. The custo-mary injection apparatuses can be used in all cases.
The increased physical stability of the gels a~cording to the invention is also present to a certain degree at body temperature, ;.e. in the liquid state. If such solutions are exposed to thermo~echanical stress in a ~otation experiment at 37~, a relative stability of ~5~ 2~7 about 3 to 5 is to be observed in comparison with conven-tional insulin solut;ons conta;ning no thickener.
Of the ;nsul;n formulat;ons conta;ning surface-active substances, preferred formulat;ons are those ~h;ch conta;n a surface-active substance, kno~n from European Patent A-18,609, ~h;ch has a chain-like basic structure, the members of ~h;ch contain ~eakly hydrophob;c and ~eakly hydroph;l;c re~ions ;n alternat;ng arrangement, in par-ticular those wh;ch conta;n a polymer, and in particular 10 homopolymer, copoly~er or block polymer, of the formula R2Y-Xn-R3 (IV) in which Xn is a chain of n members of the formulae -CH~R1)-CH(R1)-0- tV) or -CH(R1)-0- (Vl) in any desired sequence, and n is 2-80, preferably 8-45, Y ;s 15 -O- or -NH- and R1 ;s H, ~CH3 or -C2H5, ;t be;ng poss;ble for the radicals R1 to be ;dentical or differ-ent, but -CH3 or -C2H5 occurr;ng in at least hal~
of the cha;n members X, and ;n wh;ch R2 and R3 ;ndepen-dently of one another are H or an organ;c rad;cal. R2 20 and R3 pre~erably each denote alkyl ~;th 1-20 carbon atoms9 carboxyalkyl with 2-20 carbon atoms or alkylphenyl ~ith 1-10 alkyl-carbon atoms, but, ;f Y denotes -NH-, R2 is only alkyl ~;th 1-20 carbon atoms. R2 and/or R3 can also be polyvalent and bonded ~;th three or more poly-25 alkoxy cha;ns ~Xn~ to give branched products. These com-pounds are effect;ve in concentrations of 2-200 mg/l.
Other surface-active substances uh;ch may be used are the phospholipids of thP abovementioned formula III
kno~n frQn German Patent A-3,240,177, in ~hich R4 and 30 R5 have the meaning g;ven there;n and in earh case con-tain 8 to 22, preferably about 12-22~ carbon atoms. The conc~ntration of these compounds is in general 1 to 20X
by ~e;ght~ preferably 1 ~o 10 and in partisular 2.5 to 7.5% by ~eight. Examples of such hydrophilic groups are - 35 2-(trimethylammonium)e~hyl, 2-aminoethyl, 2-carboxy-2-aminoethyl, 2,3-dihydroxypropyl or 2,3,4,5,6~pentahydroxy-cyclohexyl.
Preferred compounds are those ;n which R4 and RS each represent alkylcarbonyl. Other preferred ~2S842~
_ 9 _ compounds are those in ~hich R6 represents 2-(trimethyl-ammonium)ethyl, such compounds being kno~n as lecithins, and those conpounds in ~hich R4 and R5 each represent alkylcarbonyl bith about 8 to 16 carbon atoms or ~ith about 12 to 16 carbon atoms and in which R6 represents 2-Strimethylammonium)ethyl, in particular those compounds in ~hich R4 and R5 each represent octanoyl.
Other possible surface-active substances are the polyoxyalkylene compounds kno~n from German Patent A-2,917,535 and from WO-A-83/00~88, for example compounds of the abovementioned formula II, in ~hich R7 represents alkyl with 8 to 15 carbon atoms or a correspond~ng ole-finic group and m denotes an integer from 2 to 25. These compounds are in general added ;n an amount of 2 to 200 15 mg/l. R7 is preferably (C12 or C13)-alkyl~ and m is preferably 4 to 23, in particular 6 to 15.
All the formulations according to the invention can in principle be prepared from dissolved or From amor-phous or crystalline insulin (clear or cloudy gels). They ZO have, in general, a pH bet~een 2D5 and 8.5, but in par-ticular between 6 and 8, and preferably contain a suitable isotonicity agent, a suitable preservative and, if approp-riate, a suitable buffer substance, for example those men-tioned below. The active substance is preferably present 25 in dissolved form.
The formulations according to the invention uith-out surface-active substances display a significantly de-layed action in animal experiments in contrast to corres-ponding ~omparison formulations w;thout the addition of 30 th;ckeners, the delaying effect increasing as the content of thickener ;ncreases (Figure 3). This is extremely sur-prising~ espec;ally for those ~ormulations investigated which are clear liquids at body temperature, i.e. ~hich do not display the sparing solubility effect (crystals or 35 amorphous suspensions~ ~hich exists ~ith other depot forms.
It is also possible to ;ntensify the depot effect by comb;nation uith customary auxil;aries ~ith the delay-ing action, such as, ~or example, by addition of sui~able ~584Z~
- ~o -amounts of z;nc, Surfen, globin or protam;nesulfate. The amount of zinc added can be up to 100 ~9 of Zn2+/100 insu~
lin units; it is preferably more than 35 and usuaLly less than 50 ~9 of Zn2+/100 insuLin units. The amount of protamine can be, for example, be~ween 0~28 mg and 0.6 mg per 10û units (based on protaminesulfate). Products which were hitherto inaccess;ble and have a part;cularly long-last;ng action can be prepared in this manner, the use of these products being of interest because recent knouledge 10 from ~herapy with insulin-metering apparatuses sho~s that prec;sely a basal amount of ;nsul;n seems to be thera-peutically advantageous.
A su;table phys;olog;cally acceptable carr;er med-;um ~h;ch is compat;ble w;th the ;nsul;ns is a ster;le 15 aqueous solution which has been rendered isotonic ~ith blood in the customary manner, for example by glycerol, sod;um chlor;de or glucose, and ~h;ch also conta;ns one or more of the usual preserYatives~ for example phenol, m-cresol, benzyl alcohol or p-hydroxybenzoic acid esters.
20 The carrier medium can addit;onally contain a buffer sub-stance, for example sodium acetate, sod;um citrate, sod;um phosphate or tr;s-(hydro~ymethyl)-am;nomethane. Dilute ac;ds ~typically HCl) or alkal;s (typically NaOH) are used to establ;sh the pH.
The ;nvent;on also relates to a process for the preparat;on of aqueous ;nsul;n formula~ions ~hich are stab;l;zed to~ards mechan;cal stress, which comprises adding a physiologically acceptable gelling agent and, if appropr;ate, a physiologically acceptable surface~act;ve 30 substance to an aqueous insulin formulation. The formu-lations are distinguished by par~icular stability and -in the case of subcutaneous or ;ntramuscular adm;nistra-t;on - by a delayed act;on.
The ~nvent;on furthermore relates to the use of 35 these ;nsul;n formulat;ons in the treatment of diabetes mell;tus, in particular by means of dev;ces for cont;nuous release of insul;n, and to the use of the formulat;ons for avoiding adsorption or denaturing of insulin on surfaces and other phase interfaces, in particular during ~Z58~27 purification by chromatography or crystallization, storage and therapeutic use.
The insulin formulations accord;ng to the inven-tion can be admin;stered parenterally, i.e. intravenously, subcu~aneousLy or intramuscularly, for the treatment of diabetes mellitus. The depot effect of the formulations ~ith or ~ithout surface-active substances manifests itself most markedly in the subcutaneous mode of administration, but also manifests itself clearly on intramuscular injec tion. ~hen administered intravasally, the formulations according ~o the invention, in clear solution, have a rapid action, in the same manner as the known dissolved insulinsO They are therefore outstandingly suitable for use in automatic metering apparatuses, such as pumps, in ~hich the infused insulin must be immediately effective, since only in this ~ay is rapid control, for example in accordance ~;th the blood glucose level, possible.
~ ith some metering principles~ it is necessary, or at least advantageous, to fill the reservoir ~ith de-gassed ;nsulin solution. As has been frequently demon-strated~ a;r~ together with contact uith the materials of the equipment, is the most adverse environment for insu-lin. This is a practical problem with conventional solu-tions, since solutions ~hich may have been degassed by the manufacturer in the end effect dissolve air again by move-ment (for example transportation) and diffusion. In con-trast, a solidified gel, which has been degassed as a liquid, is far less susceptible to this influence;
troublesome degassing (and the associated risk of non-sterility~ directly before use in the pu~p can then pos-sibly be dispensed with.
Another practical advantage of the fornulat;ons according to the invention can be that, if the produc~s are gels ~hich have solidified at the storage te~perature, direct contact w;th the stopper is avoided. In particu-lar, the usual stoppers tend, for exa~ple~ to absorb the stabilizers, ~hich can present problems in vie~ of the small amounts added in some cases~
The following Examples serve for further ~LZ589~2~

illustration, ~ithout restricting the invention to these.
The formulations prepared therein all have a viscosity of more than 1.75 mPa.s at 4~. The figures g;ven for dextran, for example 60, multiplied by 103, indicate the molecular weight. The distilled water was in each case p.i~ of pH 7.3. Polygeline is a product from Behringwerke AG, Marburg.
Examples 1 to 3) - Insul;n formulations containing 20Z of Polygeline According to Example 1, in each case a sterile solution of a) 250 9 of Polygeline, lyophilized, in distilled water, made up to 1 l, and b) 4.464 9 of human insulin t28 I.U./mg), 21.25 9 of glycerol, 7.50 9 of tr;s-(hydroxymethyl)-aminomethane,

3.375 9 of phenol, an amount of anhydrous zinc chloride such that the total zinc content ~as 0.035 9 and 0.0125 9 of polypropylene glycol, onto which in each case about 5 of polyethylene glycol had been polymerized on both sides (average molecular ~eight of 1,80û), in distilled water, made up to ~50 ml, were combined under sterile conditions.
According to Example 2, Example 1 was repeated, with the modification that no polypropylene glycol had been added to solution b).
According to Example 3, in each case a sterile solution of a) 200 g of Polygeline, in distilled water, made up to 800 ml, and of 30 b) 1.~29 9 of human insulin t28 I.U.lmg), 1.50 9 of m-cresol, 1000 9 of phenol, 17.00 9 of glycerol, an amount of anhydrous zinc chloride such that the total zinc content was 0.028 9 and 0.030 9 of lecithin~ in distilled water, made up to 200 ml, uere combined under sterile conditions.
~he solutions prepared according to Examples 1 to 3 were introduced into small glass bottles in the usual manner. At about 15C, they solidified as clear gels containing 1~0 I.U./ml.

According to ~ ~ 5~ in each case a sterile solution of a) 180 9 of dextran 60 or 160 9 of dextran 60, in d;stilled ~ater~ made up to 800 ml, and of b) 3.571 9 of human insulin (28 I~U./mg), 6.00 9 of tris-thydroxymethyl~-aminomethane, 2.00 g of m~cresol, 1.00 9 of phenol, 17.00 9 of glycerol, an amount of anhydrous zinc chloride such that the total zinc content was 0.028 9 and 0.010 g of polypropylene glycol ~see Example 1), in distilled water, made up to 200 ml, were combined under sterile condit;ons.
These formulations ~ere introduced into small glass bottles in the usual manner.
According to Exam~les 6 to 8, insulin formulations ~ere prepared analogously to Example 4, but ~ith 10 and 20X of dextran of different molecular ueight ~see TabLe 1).
9) Proinsulin formulation containina 20X of PolYgeline In each case a sterile solu~ion of a) 20 9 of Polygeline~ in distilled ~ater, made up to 80 ml, and of b) 100 mg of proinsulin from pigs, 0.21 8 of NaH2P04.2HzO, 0~30 9 of n-cresol, an amount of anhydrous zinc chloride such that the total zinc content ~as 0.0012 9 and 0.010 9 of polyoxyethylene-23 lauryl ether of molecular ~eight 1,200, in distilled uater, made up to 20 ml, ~ere combined under sterile cond;tions.
10 ~ 11) Insulin formulation containin~ 8X of Polygeline In each case a sterile solution of a) 1 l of a 10Z strength solution of Polygeline and b) according to Example 10, 1.786 9 of pig insulin, or according to Example 11, 1.786 9 of human insulin (in each case 28 I.U./mg), and 4.25 9 of glycerol~ 1.50 9 of tris-~hydroxymethyl)-aminomethane, 2.50 9 of phenol and an amount of anhydrous zinc chloride such that the ~otal zinc content ~as 0.014 9, in distilled ~ater, made up to 250 ml~
~ere combined under sterile condit;onsu Accord;ng to the biological test, the formulations thus prepared contained 40 I.U./ml.

~Z584Z~

Physical stability of the formulations according to Examples 1 and 4 to 8 The following stability data were obtained in a standardized pump experiment using a peristaltic pump (37C, movement, pump;ng rate of 12 I.U./d):
Table 1 Solution Time before first Relative clouding occurs stability H-Insulin Hoechst( 3 days Formulation b) according to Example 1, made up to 1.25 l with distilled water45 days 15 Formulation accord;ng to Example 1 >80 days >27 Formulation containing 8%
of dextran (Example ~) >60 days >20 Formulation contain;ng 16%
of dextran (Example 5) >80 days >27 Formulation containing 10%
of dextran 32 (Example 6) >60 days >20 Formulation containing 10%
of dextran 100 (Example 7) >bO days >20 Formulation containing 20%
of dextran 100 (Example 8) >80 days >27 Physical stability of the formulat;ons according to Examples_2 and lD _ In each case ~ small bottles of a formulation accord;ng to Example 2 and 10 were tested ;n a standard-ized rotation experiment at 37C, 1 Hz. A standard insul;n was investigated ~or comparison.

1'~584Z7 Tab(e 2 Solution Time before ~irst Relative clouding occurs stability H-Insulin Hoechst(R) 2 days Formulation b) according to Example 10, made up to 1.25 l with distilled water4 days 2 Formulation according to Example 2 15 days 7.5 Formulation according to Example 10 22 days 11 Phys;cal stability o~ the formuLat;ons according to Examples 3 and 9 _ _ _ The solutions prepared were investigated for their physical stability in a standardized circulatory pumping experiment using a peristaltic pump (37C, move~ent, recycling pumping at a rate of 5 ml/h = 500 I.U./h = 12,00û
I.U./d).
T~ble 3 Solution Time before first Relative clouding occurs stability H-Insulin Hoechst( 20 h Formulation containing 20 of polygeline (Example 3) > 7 days >~
Formulation according to Example 3, solution b)*2 days 2.4 Formulation containing 20%
of polygeline (Example 9) > 7 days >8 Formulation accord;ng to Example 9, solution b)*30 h 1.5 * in each case without polygline, but made up to 1 l (Example 3) or 100 ml (Exmple 9) with distilled water.
12 ~ 13) Crystalline insulin formulation containin~ 1D or 20% of Pol eline _ _ Y Q _ _ _ _ In earh case a ster;le solution of a) 125 or 250 9 of Polygeline, lyophilized! in distilled water~ made up to 1 l, ~Z58427 b) 2.10 g of NaH2P04.2H20, 16.00 9 of glycerol, 0.60 9 of phenol and 1.50 9 of cresol, ;n distilled hater, made up to 100 ml and c) a sterile crystal suspension of 1.786 9 of human insu-lin (28 I.U./mg), 0.159 9 of protamine sulfate, 0.525 9 ofNaH2P04.2H20, 4.00 9 of glycerol, 0.15 9 of phenol, 0.375 9 of m-cresol and anhydrous zinc chloride in an amount such that the total zinc content was 0.0108 9, in dis~illed ~ater, made up to 150 ml, ~ere combined under sterile conditions.
These suspensions ~ere introduced into small commercially available glass bottles. Directly after bottling, they were allo~ed to solidify at 4C, ~hereupon homogeneously cloudy gels containing 40 I.U./ml and 10 or 20X of Polygel;ne were formed.
Sedimentation experiment on the suspension according to Examples 12_and 13 Small bottles of the suspensions ~ere incubated at 37C~ ~ithout movement, and a cannula ~as inserted through the septum do~n to the bottom of the vessel~ and a second cannula uas inserted through the septum to just below the meniscus. Without moving the bottle, an ali-quot portion ~as taken at certain intervals o~ time and the insulin content ~as determined by means of high per-formance liquid chromatography. The follo~ing values,calculated in I.U./ml, were determined.
Table 4 WithdrawaL10X of poLygeLine20% of poLygeline after(Example 12) (Example 13) Bottom Men;scus ~ottom Meniscus of vessel of vesseL
0 min 40 I.U./ml 40 I.U./ml 40 I.U./ml 40 I.U./ml 5 min 41 39 41 40 15 min 41 40 40 41 30 min 40 40 40 41 60 min 46 35 40 40 120 min 66 14 46 35 ~2Sl~3L~127 14 ~ 15) Insul;n formulations containing 8 or 16X of dextran 60 ~ith a dela ed action Y
In each case a sterile solution of a) 80 or 1S0 9 of dextran 60, in distilled ~ater, made up to 800 ml, and of b) 1.455 9 of de-Phe-(B1)-pig insulin t27.5 I.U./mg), 17.00 9 of glycerol~ 6.00 9 of tris-(hydroxymethyL)-amino-methane, 1.50 9 of m-cresol and 1~00 9 of phenol, in dis-tilled water, ~ade up to 200 ml, were combined under sterile conditions.
These solut;ons, ~hich ~ere viscous at 4C, ~ere introduced into small commerciaLly available bottles.
According to the biological testO the activity was in each case ~0 I~U./ml~ In a dosage of 0.2 I.U./kg in rabbits, these insulin products have a delayed effect, on s.c.
administrat;on, uhich is as great as or greater than that of a commercially available neutral protamine Hagedorn delayed action product ~Figure 4~
16) Insulin formulation contain;ny 20X of Polygeline A sterile solution of a) 200 ~ of Polygeline, dissolved in d;st;lled ~3ter, made up to 800 ml, and of b) 3.571 9 of human insulin (28 I.U./mg), Z.10 9 of NaH2P04.2H20, 2.00 9 of m cresol and 1.00 9 sf phenol, dissolved in distilled water, made up to 200 ml, ~ere com-bined under sterile conditions.
The product ~as introduced into small commercially available bottles. The biological test showed an activ;ty of 100 I.U.lml. The delayed action ~as confirmed on rabbits.
17 - 19) Insulin formulations containing 20X of dextran of different molecular weights and the delayed action thereof In each case a sterile solution of a) 20 9 of dextran 32 (Example 17) or dextran 60 ~Exa~ple 18~ or dextran 100 (Example 19), in distilled ~ater~ made up to 80 ml, and of b) 0.0071 g of human insulin ~28 I.U./mg)~ 0~60 9 of tris-thydroxymethyl)-aminomethane, 0.20 9 of m-cresol, 0~10 9 of phenol, 17.00 9 of glycerol and anhydrous zinc chloride ;n an amount such that the tota~ zinc content ~as 0.0028 9, in dist;lled uater, made up to 20 ml, ~ere combined under sterile conditions.
On s.c. administration to rabbits in a dosage of 0~2 I.U./kg, these solutions exhibited a greatly delayed action tFigure 5). The duration of the action on rabbits uas about the same length as that of Basal-H-Insulin Hoechst(R).
Action profile of insulin formulations according to Examples 1~_2, 11, 14, 15 and 17 to 19 A) i.v. administration to dogs and rabbits In each case 0.2 I.U.1kg of body weight of the for-mulation according to Example 1 ~as administered into the ear vein. Human insulin Hoechst(~) was used as the - comparison insulin (II). Figures 1 and 2 show the time course of the blood glucose level tx ~ SEM). The values ~ere determined from measurements on in each case 5 2~ ani~als~ The formulation according to Example 1 tI) had an action ~hich was as rapid or even somewhat more rapid than the comparison unmodified insulin (II) both in dogs (Figure 1~ and in rabbits tFigure 2~ r B) S~Cn administration to rabbits The formulations ~ere diluted to in each case 1D
I.U /ml ~ith a corresponding placebo solution and were then administered subcutaneously in a dosage of D.2 I.U./kg to in each case 5 rabbits. The blood glucose values tx+SEM) ~ere measured after 0, 0.5, 1, 2, 3, 5 and 7 hours.
Figure 3 shows the values obtained on administration of a solution of human insulin analogous to Example 11 tcurve I), a solution analo0ous to Example 2 tcurve II) and a solution analogous to Fxample 2 but uithout the addition of Polygeline tcurve III). Figure 4 sho~s the values obtained on administration of a solution of de-Phe-tB1)-pig insulin accord;ng to Example 14 tcurve I) ~nd Example 15 tcurve II) and a commerc;ally available suspension of a human insulin delayed action product t8asal-Insulin 1~2584Z7 Hoechst(R) (curve III). Figure 5 shows the same values for formulations according to Examples 17 (curve I), 18 (curve II) and 19 (curve III) and for the human insulin delayed action product mentioned (curve IV). Figure 6 shows the same values for a formulation according to Example 1 (curve I) and for the human insulin delayed action product mentioned (curve II~.
In Figures 1 to 6, the length of the strokes for the individual measurement values indicates the standard deviation of the mean value (SEM) in the customary manner.

Claims (11)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A process for the production of an aqueous insulin preparation being stabilized against mechanical stress ha-ving an insulin concentration of more than 1 I.U./ml and at 4°C a viscosity of at least 1.75 mPa?s, which comprises in-corporating into an aqueous insulin solution a physiologi-cally acceptable thickening agent and a physiologically acceptable surfactant.

2. A process as claimed in claim 1, characterized by at least one of the features that a) the insulin concentration is up to about 1500 I.U./ml, b) the insulin is human insulin, a mammalian insulin, a modified insulin or a human proinsulin, c) there is added more than 1 % by weight of the thickening agent, d) the thickening agent added is collagen or a conversion product thereof, a polysaccharide or a derivative of po-lyvinyl pyrrolidone, and e) the pH-value is adjusted to a value in the range from 2.5 to 8.5.

3. A process as claimed in claim 1, wherein there is also added at least one of the agents a) an appropriate isotonic agent, b) an approrpiate preserving agent, c) an appropriate buffering substance, and d) zinc in an amount of up to 100 µg zinc ions per 100 I.U..

4. A process as claimed in claim 2, charaterized by at least one of the features that a) the insulin concentration is in the range from 5 to 1000 I.U./ml, b) there are added from 2 to 20 % by weight of the thicke-ning agent and e) the pH-value is adjusted to a value in the range from 6 to 8.

5. A process as claimed in claim 1 or 2 or 3, wherein the insulin preparation has a viscosity of at least 2 mPa?s.

6. A process as claimed in claim 1 or 2 or 3, wherein the insulin preparation has a viscosity of at least 2.5 mPa?s.

7. A process as claimed in claim 1 or 2 or 3,wherein the insulin preparation produced is present as a hydrogel at a temperature of 4°C.

8. A process as claimed in claim 1 or 2 or 3 wherein there is also added an auxiliary agent with a retarding effect.

9. An aqueous insulin preparation having an insulin concentration of more than 1 I.U./ml and at 4°C a viscosity of at least 1.75 mPa?s, containing a physiologically accep-table thickening agent and a physiologically acceptable surfactant.

10. An aqueous insulin preparation having an insulin concentration in the range from 5 to 1000 I.U./ml and at 4°C a viscosity of at least 2.5 mPa?s, having a pH-value in the range from 6 to 8 and containing from 2 to 20 % by weight of a physiologically acceptable thickening agent and a physiologically acceptable surfactant.

11. An aqueous insulin preparation having an insulin concentration in the range from 5 to 1000 I.U./ml and being at 4 °C present as a hydrogel, having a pH-value in the range from 6 to 8 and containing from 2 to 20 % by weight of a physiologically acceptable thickening agent and a physiologically acceptable surfactant.