sø- & handelsrettens dombog

- alj udskrift af sø- & handelsrettens dombog ____________ dom afsagt den 18. februar 2016 t-4-13 1) astrazeneca ab 2) astrazeneca a/s (advokat peter-ulrik plesner og advokat kasper frahm for begge) mod krka sverige ab (advokat klaus ewald madsen) sagens baggrund og parternes påstande hovedspørgsmålet i denne sag, som er anlagt den 8. marts 2013, er, om krka sverige ab (herefter krka) ved markedsføring af lægemidlet esomeprazol ‛krka‛ i danmark i perioden fra den 8. marts 2010 til den 27. maj 2014 (patentudløb) har krænket astrazeneca ab’s og astrazeneca a/s’ (herefter tilsammen astrazeneca) rettigheder til patent dk/ep 1 020 461. -2- astrazeneca har nedlagt endelig påstand om, at krka til astrazeneca skal betale 67.000.000 kr. med tillæg af procesrente fra sagens anlæg. krka har nedlagt påstand om frifindelse, subsidiært frifindelse mod betaling af et mindre be- løb end påstået af astrazeneca. sagens oplysninger parterne astrazeneca-koncernen driver virksomhed globalt med forskning og udvikling af originale lægemidler. astrazeneca ab har stået for udviklingen af det originale lægemiddel, nexium, og var indtil patentets udløb indehaver af patent dk/ep 1 020 461 (herefter stridspatentet). nexium indeholder et magnesiumsalt af aktivstoffet ‛esomeprazol‛, jf. nærmere herom ne- denfor. nexium markedsføres i danmark af astrazeneca a/s. krka-koncernen er en global virksomhed, som beskæftiger sig med produktion og markeds- føring af generiske lægemidler, herunder det i sagen omhandlede generiske lægemiddel, esomeprazol ‛krka‛, som også indeholder et magnesiumsalt af esomeprazol. esomeprazol ‛krka‛ fremstilles af koncernens moderselskab, krka d.d., på koncernens hovedsæde i novo mesto, slovenien. datterselskabet, krka sverige ab, forestår markedsføringen af esomepra- zol ‛krka‛ i danmark. markedsføringen i danmark blev påbegyndt den 8. marts 2010. omeprazol og esomeprazol i 1980’erne udviklede astrazeneca aktivstoffet ‛omeprazol‛, som astrazeneca markedsførte som lægemiddel under varemærket losec®. dette aktivstof var efter det oplyste det første eksempel på en såkaldt ‛protonpumpeinhibitor‛, dvs. et stof, som hæmmer udviklingen af mavesyre i mavesækken. omeprazol anvendes således til behandling af mavesår, sure op- stød, m.v. omeprazol er et kiralt (evt. ‛chiralt‛) molekyle (også kaldet et ‛racemat‛ eller en ‛racemisk blanding‛). dette vil sige, at molekylet indeholder både en højredrejet og en venstredrejet -3- konfiguration (såkaldte ‛enantiomerer‛), bestående af lige store mængder af højre- og ven- stredrejede enantiomerer, som er spejlbilleder af hinanden. de højredrejede og venstredrejede enantiomerer kan benævnes henholdsvis (+)-omeprazol og (-)-omeprazol. (+)-omeprazol betegnes også som (r)-omeprazol eller r-enantiomeren af omeprazol, og (-)-omeprazol betegnes også som (s)-omeprazol, s-enantiomeren af omepra- zol eller esomeprazol. (-)-omeprazol, (s)-omeprazol, s-enantiomer og esomeprazol er således forskellige betegnelser for den samme kemiske forbindelse. jo højere overskud af den ene enantiomer, der forefindes i den kemiske forbindelse, i forhold til den anden enantiomer i forbindelsen, des højere er det ‛enantiomere overskud‛ (på en- gelsk: ‛enantiomeric excess‛, som forkortes ‛e.e.‛). det enantiomere overskud benævnes og- så i visse af sagens bilag som forbindelsens ‛optiske renhed‛. det enantiomere overskud / den optiske renhed illustreres af et af krka fremlagt hjælpebilag under sagen: ‛< enantiomert indhold af esomeprazol enantiomer s (%) r (%) overskud (% e.e.) racemat 50 50 0 75 25 50 95 5 90 99,4 0,6 98,8 99,5 0,5 99,0 99,6 0,4 99,2 99,7 0,3 99,4 99,8 0,2 99,6 særligt optisk rent esomeprazol 99,9 0,1 99,8 ifølge stridspatentet <‛ stridspatentet, herunder stridspatentets forhistorie stamansøgningen og stampatentet i 1990’erne forsøgte astrazeneca efter det oplyste at finde frem til en protonpumpeinhibitor, som havde bedre egenskaber end omeprazol (losec). to forskere hos astrazeneca fandt -4- frem til, at s-enantiomeren af omeprazol (esomeprazol) i høj optisk renhed i form af dets magnesiumsalt havde sådanne bedre egenskaber end omeprazol. den 27. maj 1994 ansøgte astrazeneca om patentbeskyttelse på bl.a. magnesiumsaltet af esomeprazol, jf. international patentansøgning wo 94/27988 (herefter stamansøgningen). om baggrunden for det ansøgte patent var det anført bl.a.: ‛< < it is desirable to obtain compounds with improved pharmacokinetic and metabol- ic properties which will give an improved therapeutic profile such as a lower degree of interindividual variation. the present invention provides such compounds, which are novel salts of single enantiomers of omeprazole. < the present invention in a further aspect provides a novel method for preparing the novel compounds of the invention in large scale. this novel method can also be used in large scale to obtain single enantiomers of omeprazole in neutral form. < < single enantiomers of omeprazole have hitherto only been obtained as syrups and not as crystalline products. by means of the novel specific method according to one aspect of the invention of preparing the single enantiomers of omeprazole, the salts defined by the present invention are easy to obtain. in addition, the salts, however not the neutral forms, are obtained as crystalline products. because it is possible to purify optically impure salts of the enantiomers of omeprazole by crystallisation, they can be obtained in very high optical purity, namely ≥99.8% enantiomeric excess (e.e.) even from an optically contaminated preparation. moreover, the optically pure salts are stable towards racemization both in neutral ph and basic ph, which was surprising since the known deprotonation at the carbon atom between the pyridine ring and the chiral sulphur atom was expected to cause racemization under alkaline conditions. this high stability towards racemization makes it possible to use a single enantiomeric salt of the invention in therapy. <‛ stamansøgningen førte til udstedelsen af patent ep 652 872 b1 (herefter stampatentet) den 8. november 2000, der indeholdt bl.a. følgende krav: ‛1. magnesiumsalt af (-)-5-methoxy-2-[[(4-methoxy-3,5-dimethyl-2-pyridinyl)- methyl]sulfinyl]-1h-benzimidazol (mg-salt af (-)-enantiomeren af omeprazol). 2. fremgangsmåde til fremstilling af mg-saltet af (-)-enantiomeren af omepra- zol k e n d e t e g n e t ved, at en diastereomer blanding af en ester med formel iii -5- [formel udeladt], hvor acyl betegner en chiral acylgruppe med enten r- eller s- konfiguration, separeres for at opnå de separerede diastereomerer, hvorefter diaste- reomeren omfattende acyloxymethylderivatet af (-)-enantiomeren af omeprazol oplø- ses i en alkalisk opløsning, hvor acyloxymethylgruppen borthydrolyseres til opnåelse af (-)-enantiomeren af omeprazol som omdannes til magnesiumsaltet. < 9. mg-salt af (-)-enantiomeren af omeprazol til anvendelse ved behandling. 10. anvendelse af mg-saltet af (-)-enantiomeren af omeprazol til fremstilling af en farmaceutisk formulering til behandling af mavesyre-relaterede sygdomme ved inhibition af mavesyresekretion. <‛ kravene i stampatentet var ikke begrænset til en bestemt optisk renhed af esomeprazol mag- nesiumsaltet. stampatentet blev den 19. december 2006 kendt ugyldigt af epo’s technical board of appeal (sag t 401/04) med henvisning til manglende opfindelseshøjde. om spørgsmålet om nyhed udtalte technical board of appeal i afgørelsen følgende: ‛3.1.1 novelty the novelty of the subject-matter of this claim was not under dispute and resides from the fact that neither document (d2) nor (d3) discloses the use for the manufac- ture of a pharmaceutical formulation for the treatment of reflux esophagitis.‛ den afledte patentansøgning og stridspatentet astrazeneca havde den 18. april 2000, dvs. godt et halvt år før udstedelsen af stampatentet, indgivet en afledt patentansøgning af stamansøgningen (herefter den afledte patentansøg- ning). den afledte patentansøgning indeholdt bl.a. følgende krav: ‛1. the use of an alkaline salt of the (-)-enantiomer of omeprazole for the manu- facture of a pharmaceutical preparation having improved pharmacokinetic and metabolic properties. <‛ der er fremlagt korrespondance mellem astrazenecas repræsentant og epo’s examining division i forbindelse med epo’s behandling af den afledte patentansøgning. -6- af brev af 7. november 2008 fra astrazenecas repræsentant til epo fremgår bl.a.: ‛< (

1. a)claim 1 has been limited to cover only the magnesium salt of the (-)-enantiomer of omeprazole with an enantiomeric excess of ≥ 99.8%. this is based on the appli- cation as filed at page 4, line 11, which states that the present application teaches enantiomers of omeprazole in very high optical purity, namely ≥ 99.8% e.e., and material of this purity is exemplified in example 5 at page 12 of the application as filed; < < the present claim set is not only clearly distinguished over the claims that were considered by the board in decision t 401/04, but also all of the prior art that was cited in those proceedings (and by the examining division in the present proceed- ings). in this respect, the limitation of the main claim (to a magnesium salt of the (-)- omeprazole with an enantiomeric excess (e.e.) of ≥ 99.8%) is not an arbitrary one made merely with the sole purpose of distinguishing the claims over those that were revoked in the earlier decision. rather, for the reasons explained in detail below, it serves to distinguish the invention over the cited prior art in a meaningful, technical and, above all, patentable way. < the enclosed evidence was not available at the time the board made its decision. had this evidence < been available to the board, there is no doubt that the board would have reached the opposite conclusion and upheld the patent in accordance with the interlocutory decision of the opposition division. < moreover, crystalline materials were capable of thereafter being re-crystallised with a view to further enhancing optical purity. in this respect < another unexpected finding associated with the present invention was the alkali and/or alkaline earth metal salts of the (-)-enantiomer were found to be more stable than corresponding salts of the racemate and therefore crystallized out first, if the (-)-enantiomer was in excess. the reverse was true for the neutral (base) form, where the racemate (omeprazole) was found to be more crystalline than cor- responding salts. < thus, the applicant now had a process that enabled it to make material of extremely high chemical and optical purity. < we fully appreciate that the board of appeal in t 401/04, in the context of the fact situation as they saw it, concluded that these clinical advantages did not make the then-claimed product inventive. however, in the light of the true fact situation that has since been elucidated, which shows that the currently-claimed subject matter is novel and non-obvious, the clinical advantages referred to above mean that the high- -7- ly optically-pure enantiomer of (-)-omeprazole offers genuine and significant benefits to patients compared to the racemic drug (omeprazole) and compared to other ppls. < if the skilled person were to turn instead to document d3 in the search for a solution to the problem of the production of optically-pure (e.g. ≥ 99.8% e.e.) (-)-omeprazole (or salts thereof), this document teaches only a partially-successful resolution of the enantiomers of omeprazole, with (-)-omeprazole being provided with a e.e. of only 91.2%. this in itself is not a teaching of the claimed invention. moreover, d3 does not provide any motivation to further improve the optical purity of (-)-omeprazole, given that, as discussed above, it teaches that (as expected) racemisation occurs and that (also as expected) there is no difference between the enantiomers in a biological con- text. <‛ af brev af 13. februar 2009 fra astrazenecas repræsentant til epo fremgår endvidere bl.a.: ‛< 2. the claims of the divisional application have been distinguished from those of the parent patent. the principal point of distinction, namely the excep- tionally high degree of enantiomeric purity, is a feature of great technical significance, as is explained in detail in our letter of 7 november 2008 on the divisional application. in brief, the prior art did not enable such a degree of purity to be achieved, and the optically pure compound has special proper- ties (including the fact that it does not racemise, whereas less optically pure preparations do racemise, preventing the production of crystalline salts).‛ af et yderligere brev af 13. februar 2009 fra astrazenecas repræsentant til epo fremgår bl.a.: ‛< an invention is defined by the claims of a patent application or patent. the board ex- amined different claims from those that are now the subject of this divisional applica- tion. in particular, the claims are now limited to preparations and uses thereof having at least a 99.8% enantiomeric excess of the active compound. the importance of this feature, in relation to what a person skilled in the art was able to do at the priority date and also in relation to the special properties of the compound when it is so high- ly optically pure, was discussed fully in our letter dated 7 november 2008. this claim limitation, and the technical evidence that we filed with that letter, were not pre- viously considered by the board. <‛ epo godkendte og udstedte den 22. juli 2009 stridspatentet, dk/ep 1 020 461, med gyldighed for bl.a. danmark. -8- stridspatentets krav 1 og krav 9 har i den danske oversættelse følgende ordlyd: ‛1. anvendelse af et magnesiumsalt af (-)-5-methoxy-2-[[(4-methoxy-3,5-dimethyl-2- pyridinyl)methyl]sulfinyl]-1h-benzimidazol ((-)-omeprazol) med en optisk renhed på ≥ 99,8 % enantiomert overskud (e.e.) til fremstilling af et medikament til inhibition af mavesyresekretion. < 9. magnesiumsalt af (-)-5-methoxy-2-[[(4-methoxy-3,5-dimethyl-2-pyridinyl)- methyl]sulfinyl]-1h-benzimidazol ((-)-omeprazol) med en optisk renhed på ≥ 99,8 % enantiomert overskud (e.e.). <‛ om opfindelsens område og baggrunden for opfindelsen anføres i stridspatentets beskrivel- se bl.a.: ‛opfindelsens område den foreliggende opfindelse angår nye forbindelser med en høj optisk renhed, deres anvendelse i medicin og deres anvendelse ved fremstilling af farmaceutiske præpara- ter. baggrund for opfindelsen forbindelsen 5-methoxy-2-[[(4-methoxy-3,5-dimethyl-2-pyridinyl)methyl]sulfinyl]- 1h-benzimidazol med det generiske navn omeprazol og terapeutisk acceptable alka- liske salte deraf er beskrevet i henholdsvis ep 5129 og ep 124 495. omeprazol og dets alkaliske salte er virkningsfulde mavesyresekretionsinhibitorer og er egnede som an- tiulcus-midler. forbindelserne, som er sulfoxider, har et asymmetrisk center i svovl- atomet, dvs. de findes som to optiske isomerer, (enantiomerer). det er ønskeligt at opnå forbindelser med forbedrede farmakokinetiske og metaboliske egenskaber, som vil give en forbedret terapeutisk profil, såsom en lavere grad af variation fra individ til individ. den foreliggende opfindelse tilvejebringer sådanne forbindelser, som er nye salte af rene enantiomerer af omeprazol. < den foreliggende opfindelse tilvejebringer i et yderligere aspekt en hidtil ukendt fremgangsmåde til fremstilling af den hidtil ukendte forbindelse ifølge opfindelsen i stor målestok. denne hidtil ukendte fremgangsmåde kan også anvendes i stor måle- stok til opnåelse af rene enantiomerer af omeprazol i neutral form. der er ingen eksempler inden for den kendte teknik på isolerede eller karakteriserede salte af optisk rent omeprazol, dvs. rene enantiomerer af omeprazol, og heller ingen eksempler inden for den kendte teknik på isolerede eller karakteriserede salte af no- gen optisk rene omeprazol-analoger. detaljeret beskrivelse af opfindelsen -9- den foreliggende opfindelse angår optisk rene magnesiumsalte af omeprazol i hen- hold til forbindelse iib [kemisk formel udeladt]. ifølge opfindelsen tilvejebringes et magnesiumsalt af (-)-omeprazol med en optisk renhed på ≥ 99,8 % enantiomert overskud (e.e., enantiomeric excess) og anvendelsen af et sådant salt til fremstilling af et medikament til inhibition af mavesyresekretion. rene enantiomerer af omeprazol er hidtil kun blevet opnået som sirupper og ikke som krystallinske produkter. ved hjælp af den nye specifikke fremgangsmåde ifølge et aspekt af opfindelsen, hvor de rene enantiomerer af omeprazol fremstilles, er de salte, der er defineret ved den foreliggende opfindelse, nemme at opnå. desuden er saltene, dog ikke de neutrale former, opnåelige som krystallinske produkter. efter- som det er muligt at oprense optisk urene salte af enantiomererne af omeprazol ved krystallisation, kan de opnås i meget høj optisk renhed, nemlig ≥ 99,8 % enantiomert overskud (ee, enantiomeric excess), selv fra et ikke optisk rent udgangsmateriale. desuden et det optisk rene salt stabilt over for racemisering både ved neutral ph og basisk ph, hvilket var overraskende, eftersom den kendte deprotonering ved carbo- natomet mellem pyridinringen og det chirale svovlatom var forventet at forårsage ra- cemisering under basiske betingelser. denne høje stabilitet over for racemisering gør det muligt at anvende et salt af en ren enantiomer ifølge opfindelsen i terapi. <‛ opfindelsen i henhold til stridspatentet er i beskrivelsen illustreret ved en række eksempler, herunder ved et eksempel 1b, i hvilket der opnås en optisk renhed på 99,8 %. stridspatentet blev efterfølgende ved afgørelse af 29. juli 2011 ophævet af epo’s ‛opposition division‛. astrazeneca appellerede denne afgørelse til epo’s technical board of appeal, som i november 2012 tilsidesatte afgørelsen fra ‛opposition division‛ og ved endelig afgø- relse bestemte, at stridspatentet skulle opretholdes. stridspatentet udløb den 27. maj 2014. forbudssag krka påbegyndte som nævnt markedsføring af esomeprazol ‛krka‛ i danmark den 8. marts 2010. parternes repræsentanter korresponderede herom i foråret og sommeren 2010, uden at der dog kunne opnås enighed om, hvorvidt krkas markedsføring udgjorde en krænkelse af stridspatentet. ved begæring af 9. juli 2010 indledte astrazeneca forbudssag mod krka ved københavns byret med påstand om forbud mod krkas markedsføring af esomeprazol - 10 - ‛krka‛ i danmark. af fogedrettens kendelse af 10. december 2010 fremgår bl.a. følgende om parternes forudgående korrespondance: ‛< astrazeneca har < modtaget prøver af den aktive ingrediens, og har anmodet om tilsendelse af den europæiske drug master file, dmf, og kopi af master batch re- cord. krka har ved brev af 2. april 2010 < sendt den < fremlagte dmf, hvor procesbetin- gelserne, så som anvendte temperaturer, mængder og tider, er slettet. krka har i bre- vet anført blandt andet, at krka ikke vil sende master batch record før end astraze- neca har underskrevet en fortrolighedserklæring. astrazeneca har afslået at under- skrive en sådan. astrazeneca har ved brev af 3. maj 2010 < anmodet om oplysning om krka’s frem- gangsmåde. krka har ved brev at 24. maj 2010 < og brev af 22. juni 2010 < oplyst at anvende ros1 *‛route of synthesis 1‛+ med brug af opløsningsmidlerne i-propylal- kohol (variant
2. c)og toluen (variant
3. b)ved fremstillingen af alle esomeprazol produk- ter. <‛ fra professor robert madsens forklaring under fogedsagen er bl.a. gengivet følgende i fo- gedrettens kendelse: ‛< hvis et 100% rent stof skulle iblandes en forurening, ville det kræve et yderligere procestrin med dertil hørende prøver. dette ville være omkostningsfuldt og ufordel- agtigt. <‛ ved kendelsen af 10. december 2010 nægtede fogedretten at fremme forbudsforretningen. af fogedrettens begrundelse fremgår bl.a.: ‛< det fremgår, at stridspatentet ikke omfatter en fremgangsmåde for tilvirkning af stof- fets kemiske forbindelse. det lægges som ubestridt til grund, at det af krka i danmark markedsførte lægemid- del ikke indeholder et magnesiumsalt af esomeprazol med en optisk renhed på ≥ 99,8% enantiomert overskud (e.e.). krka har oplyst, at det enantiomere overskud i krkas produkter i praksis ligger på omkring 99,24% [98,24%] og 98,92% e.e. efter stridspatentets oplyste formål og ordlyden af patentkrav 1 er der tale om et an- vendelsespatent, der retter sig mod benyttelse af et magnesiumsalt af esomeprazol - 11 - med en optisk renhed på ≥ 99,8% e.e., med det formål at fremstille et lægemiddel til inhibition af mavesyresekretion. hvorvidt stridspatentet udgør en patenteret fremgangsmåde, og dermed nyder indi- rekte produktbeskyttelse i medfør af patentlovens § 3, stk. 1, nr. 3, afhænger af, om opfindelsen definerer en proces til fremstilling af et produkt. stridspatentet indehol- der ikke en nærmere angivelse af, hvordan de farmaceutiske formuleringer skal fremstilles. da den kemiske forbindelses anvendelse er det procesbærende element ved fremstillingen af lægemidlet, finder fogedretten, at patentkrav 1 er et sådant fremgangsmådekrav, at kravet nyder indirekte produktbeskyttelse i medfør af pa- tentlovens § 3, stk. 1, nr. 3. patentlovens § 64 a indeholder en regel om omvendt bevisbyrde i tilfælde af kræn- kelse af fremgangsmådepatenter. herefter må den påståede krænker bevise, at han ikke har benyttet samme fremgangsmåde, som patentet dækker. beskyttelsen af stridspatentets omfang bestemmes af patentkravene, jf. patentlovens § 39. bestemmelsen fortolkes i overensstemmelse med den europæiske patentkon- ventions artikel 69, hvorefter fortolkningen skal forene en passende beskyttelse for patenthaveren, således at dennes beskyttelsesområde ikke indsnævres i urimelig grad, med en rimelig retssikkerhed for tredjemand. beskyttelsesomfanget fastlægges konkret for patentkravet. fogedretten finder ikke grundlag for at fortolke stridspatentet udvidende til også at omfatte kemiske forbindelser med en lavere renhedsgrad. der er herunder lagt vægt på, at astrazenecas patentagenter ved brev af 7. november 2008 til european patent office har oplyst, at baggrunden for de ændrede patentkrav, herunder begrænsnin- gen af kravet til et magnesiumsalt af esomeprazol med et enantiomert overskud på 99,8% (e.e.), ikke er en vilkårlig begrænsning, men snarere har til formål at adskille opfindelsen fra den nævnte kendte teknik på en meningsfyldt, teknisk og, frem for alt, patenterbar måde. efter vidnerne karl anker jørgensen og robert madsens forklaringer lægges det til grund, at den kemiske forbindelses renhedsgrad navnlig bestemmes af de under kry- stallisationstrinnet anvendte procesparametre, og at disse kan varieres, hvorfor kry- stallisationen ikke uundgåeligt vil medføre et enantiomert overskud på 99,8%. robert madsen har forklaret, at krka anvender andre procesparametre under krystallisatio- nen, end de procesparametre, som martin bohlin anvendte under sine forsøg på ef- tergørelse af krka’s proces. henset hertil og til, at det af krka i danmark markedsførte lægemiddel ikke indehol- der en aktiv ingrediens af samme høje renhed som stridspatentet, finder fogedretten, at krka på tilstrækkelig vis har dokumenteret at anvende en anden proces end den patenterede fremgangsmåde. fogedretten finder herefter ikke, at patentkrav 1 er krænket. astrazeneca har herefter ikke godtgjort eller sandsynliggjort, at de handlinger, der søges forbudt, strider mod rekvirentens ret, jf. retsplejelovens § 642, stk. 1, nr. 1 <‛ - 12 - fogedrettens kendelse af 10. december 2010 blev af astrazeneca kæret til østre landsret den 20. december 2010. astrazeneca hævede kæremålet ved brev af 21. september 2011, da epo’s ‛opposition division‛ i mellemtiden havde truffet afgørelse om, at stridspatentet ikke kunne opretholdes, jf. herom ovenfor. nærmere om esomeprazol ‛krka‛, herunder krkas fremstillingsproces bioækvivalens i forhold til nexium af ‛public assessment report‛ af 20. januar 2010 vedrørende godkendelsen af esomeprazol ‛krka‛ fremgår bl.a.: ‛< vi. overall conclusion, benefit/risk assessment and rec- ommendation esomeprazol ‚krka‛ 20 mg and 40 mg gastro-resistant capsules, hard has a proven chemical-pharmaceutical quality and is a generic form of nexium gastro-resistant tablets. < <‛ det er under sagen ubestridt, at krkas lægemiddel, esomeprazol ‛krka‛, er ‛bioækvivalent‛ med astrazenecas nexium. dokumenter omhandlende krkas proces for fremstilling af esomeprazol ”krka” fremlagt under for- budssagen under forbudssagen forelå bl.a. følgende dokumenter, som også indgår i nærværende sag:  uddrag af krkas drug master file (‛dmf‛) vedrørende esomeprazol ‛krka‛ (sagens bilag 14) det fremgår af dmf’en, at krkas fremstilling af esomeprazol ‛krka‛ overordnet består af to processer; krka anvender først en proces til dannelsen af et esomeprazol magne- sium ‛crude‛-materiale (dvs. et råmateriale). ‛crude‛-materialet er et magnesiumsalt, som ubestridt har et enantiomert overskud af s-enantiomeren på 98-99 % e.e. dette ef- - 13 - terfølges af en krystallationsproces, hvis formål er at omkrystallisere blandingen af omeprazol-enantiomererne i ‛crude‛-materialet, idet s-enantiomeren opkoncentreres i forhold til r-enantiomeren, hvorved der opnås en højere optisk renhed. opkoncentra- tionen sker ved i et opløsningsmiddel at udfælde s-enantiomeren, som går fra opløst form til en fast, krystallink form (et magnesium dihydrat-salt), der kan udskilles fra re- sten af suspensionen. af dmf’en fremgår nærmere om fremstillingsprocessen bl.a.: ‛< racemic omeprazole < is the key starting material in krka’s synthesis process of esomeprazole, s-enantiomer of omeprazole. esomeprazole magnesium is obtained by two optional synthetic routes: ros1 *‛route of synthesis 1‛+: diasteroisomeric crystallization of omeprazole < with (s)-binol < in ethanol, 2-propanol or toluene solvent to perform esomeprazole binol complex < isolated esomeprazole binol complex intermediate is further transformed over potassium salt of esomeprazole in the extraction step with isopropyl acetate and precipitated with magnesium chloride to magnesium salt of esomeprazole to get a crude product. ros2: [‛chiral hplc chromatography‛; en proces, der efter det oplyste ikke anven- des af krka] < the active ingredient esomeprazole, in the form of magnesium salt dihydrate is pre- pared by the crystallization of crude esomeprazole magnesium salt from the metha- nol, acetone and water mixture. <‛ det fremgår af dmf’en, at dokumentet er oprettet den 17. juli 2009 og senest opdateret den 1. december 2009. dokumentet blev fremsendt til astrazeneca ved brev af 2. april 2010 i anonymiseret form, bl.a. således at visse af de i dokumentet anførte procesbetin- gelser (f.eks. mængde- og temperaturangivelser) samt det anførte udbytte af krystalla- tionsprocessen var streget ud (også omtalt i fogedrettens kendelse, jf. ovenfor). - 14 -  krkas internationale patentansøgning, wo 2008/092939 a2, af 31. januar 2008 (bilag 17). denne patentansøgning indeholder bl.a. et eksempel 19 omhandlende ‛preparation of (s)-omeprazole magnesium dihydrate form a from racemate‛.  analyserapport af 24. juni 2010 udarbejdet af kemiker hos astrazeneca, martin bohlin (bilag 16) – martin bohlins 1. forsøg. martin bohlins forsøg (det første af i alt tre) havde til formål at afprøve procestrinene i krkas dmf (bilag 14). under procestrinene anvendte martin bohlin de procesbetingel- ser, som fremgår af eksempel 19 i krkas patentansøgning wo 2008/092939 (bilag 17), til at udfylde de procesbetingelser, som ikke fremgik af den til astrazeneca fremsend- te, anonymiserede dmf (bilag 14). martin bohlins opgave er i analyserapporten gengivet således: ‛< ‘mix esomeprazole potassium (about 7.8
4. g)with an enantiomeric purity of (or slightly above) 98% enantiomeric excess with water (150 ml). dilute the mixture with 50 ml methanol. add one-half equivalent of an aqueous solution of magnesium chlo- ride (mgcl2) dropwise while stirring. filter the crude esomeprazole magnesium product and wash with water. dry the solid overnight. analyze the crude esomepra- zole magnesium to determine enantiomeric composition.’ ‘dissolve the crude esomeprazole magnesium (about 3.3
5. g)in an amount of methanol and at a temperature (not to exceed 54°
6. c)sufficient to dissolve all the material. filter the solution and then evaporate at up to 54°c and reduced pressure, until an oily re- sidue is obtained. dissolve the oily residue in 6 ml methanol at room temperature. gradually add 27 ml of an acetone/water mixture (4:1, v/
7. v)to the methanol solution so that the temperature does not exceed 25°c. stir the resulting suspension for about 12-16 hours at 20°c (+/- 2°c), then for an additional 3 hours at 0-5°c. filter the sus- pension and wash with 3 ml acetone. dry the wet filter cake at reduced pressure and a temperature from 20 to 40°c for about 6-10 hours. continue drying at a tempera- ture between 60 to 70°c for an additional 1-2 hours. determine the enantiomeric composition of the esomeprazole magnesium product.’ <‛ - 15 - det fremgår af analyserapporten, at martin bohlin med udgangspunkt i en optisk ren- hed i ‛crude‛-materialet på 98,66 % e.e. efter udførelsen af krystallisationen opnåede et esomeprazol magnesium dihydrat-salt med en optisk renhed på 100 % e.e. udbyttet af krystallisationen var målt til 71,6 % (bilag 2 til rapporten).  erklæring af 9. juli 2010 fra professor karl anker jørgensen (bilag 18). professor karl anker jørgensen overværede martin bohlins første forsøg. karl anker jørgensen bekræfter i sin erklæring, at martin bohlin udførte forsøget som beskrevet i martin bohlins analyserapport (bilag 16), og at martin bohlin under forsøget benyttede de procesbetingelser, som er beskrevet i eksempel 19 i patentansøgning wo 2008/092939 (bilag 17).  erklæring af 23. september 2010 fra professor robert madsen (bilag 23). i denne erklæring fremkommer professor robert madsen med kommentarer til martin bohlins første forsøg (bilag 16). erklæringen er udarbejdet på grundlag af et besøg på krkas produktionsfacilitet i slovenien, hvor robert madsen overværede en krystallisa- tion fra esomeprazol magnesium ‛crude‛ til esomeprazol magnesium dihydrat. af erklæringen fremgår bl.a.: ‛< summary of the crystallization process at krka, d. d. the crystallization of esomeprazole magnesium at krka, d. d. took 3 days and gave a final product with an optical purity of 98.84% (
8. ee)from a crude product with an opti- cal purity of 98.57% (ee). the crystallization was not performed with the same ratio of esomeprazole magne- sium, methanol, acetone and water as in the experiment by martin bohlin. crystalli- zation temperature and time were also different. therefore, it is not unexpected that the outcome is different. the crystallization at krka, d. d. was carefully monitored and allowed to proceed until the desired amount of the r-isomer in the crystals had been achieved. - 16 - all the practical work was performed by operators and technicians working in 8 hour shifts. they all appeared to know what to do at the different stages of the crystalliza- tion process and did not require any significant instruction. during my presence i was followed most of the time by the engineer and head of the synthesis building. i asked several questions about his experience with the crystallization process. for ex- ample, the addition of the acetone – water mixture to the esomeprazole magnesium solution in methanol takes a while and the crystallization starts at some point during the addition. at the beginning of the addition i asked the engineer when he would expect the crystallization to start and he gave a very precise amount of acetone – wa- ter that had to be added. it was easy to see visually through a window in reactor 2 when the crystallization started and i was able to observe that his prediction was true. he also predicted the development in the r-isomer content during the crystalli- zation and this was also later confirmed by the measurements. it was my clear im- pression that everybody involved in the process were very familiar with the crystalli- zation and knew the intimate details very well. conclusion martin bohlin has performed one crystallization experiment with esomeprazole magnesium in the laboratory under certain process conditions (as disclosed in the in- ternational patent application, wo 2008/092939 a2) and achieved esomeprazole magnesium with an optical purity of 100%. i have witnessed the actual crystallization process for esomeprazole magnesium at krka, d. d. in novo mesto and observed that esomeprazole magnesium with an opti- cal purity of 98.84% (
9. ee)was isolated. the crystallization process at krka, d. d. em- ployed different solvent ratios, temperature and time than used in the experiment by martin bohlin, and a different outcome was therefore obtained. the result of the actual crystallization of esomeprazole magnesium that i explored was not surprising, as it is well known for a person skilled in the technical field that variations and combinations of various parameters can significantly affect the out- come of a crystallization process. the process conditions for the crystallization process can be designed in such a way that the desired amount of the r-isomer can be obtained. thus, i am able to firmly conclude that it is not inevitable that the crystallization of esomeprazole magnesium will always result in an optical purity of 100%. based on the above, it can be concluded that  the crystallization of esomeprazole magnesium at krka, d. d. is carried out under different conditions than the experiment described by martin bohlin at astrazeneca ab, i.e. example 19 in wo 2008/092939 a2 is not followed at krka, d. d.  the crystallization process at krka, d. d. is performed in accordance with the dmf, p. 16-17.  the process at krka, d. d. affords a final product with an optical purity < 99.8% (ee). - 17 -  the crystallization of esomeprazole magnesium does not inevitably lead to an enantiomerically pure material with an optical purity of 100%.‛  engelsk oversættelse af et udateret processkrift indleveret af generis farmacêutica s.a. og mer medicamentos lda i portugal (bilag 33). generis farmacêutica s.a. og mer medicamentos lda, som efter det oplyste er krkas grossister i portugal, har i en tilsvarende forbudssag anlagt af astrazeneca ved de por- tugisiske domstole indleveret et processkrift, hvoraf fremgår bl.a.: ‛< 114 it happens that krka which supplies the medicinal product to generis and mer has developed an innovative process of its own to also obtain esomeprazole and for which it submitted for registration patent wo2008/092939, whose copy of the appli- cation taken from the database of wipo (world intellectual property organization) is hereby submitted as doc. 3. <‛  krkas internationale patentansøgning, wo 2010/072759 a1, af 22. december 2009 (bilag 20). astrazeneca blev efter det oplyste opmærksom på denne patentansøgning, efter at martin bohlin havde udført sit første forsøg (bilag 16). ansøgningen angår også frem- stilling af esomeprazol eller salte af esomeprazol og indeholder bl.a. et eksempel 9, som anvender nogle lidt andre procesbetingelser end procesbetingelserne i eksempel 19 i krkas patentansøgning wo 2008/092939 (bilag 17).  martin bohlins analyserapport af 8. oktober 2010 (bilag 21) – martin bohlins 2. forsøg. dette forsøg havde også til formål at afprøve de afgørende krystallisationstrin i hen- hold til krkas dmf (bilag 14). under procestrinene brugte martin bohlin denne gang procesbetingelserne i eksempel 9 i krkas patentansøgning wo 2010/072759 (bilag 20) til at udfylde procesbetingelserne i krkas dmf. - 18 - det fremgår af analyserapporten, at martin bohlin med udgangspunkt i en optisk ren- hed i ‛crude‛-materialet på 98,6 % e.e efter udførelsen af krystallisationen opnåede et esomeprazol magnesiumsalt dihydrat-salt med en optisk renhed på 100 % e.e. udbyt- teresultatet var denne gang på 72,8 %.  erklæring af 7. oktober 2010 fra professor karl anker jørgensen (bilag 22). professor karl anker jørgensen overværede også denne gang martin bohlins forsøg. karl anker jørgensen bekræfter i sin erklæring, at forsøget resulterede i dannelsen af et esomeprazol magnesium dihydrat-salt med en optisk renhed på 100 % e.e.  erklæring af 21. oktober 2010 fra professor robert madsen (bilag h). i denne erklæring fremkommer professor robert madsen med kommentarer til martin bohlins 2. forsøg. af erklæringens danske oversættelse fremgår det bl.a.: ‛< den største forskel mellem det andet krystalliseringsforsøg og det første krystallise- ringsforsøg, som jeg kommenterede i min første erklæring, er størrelsesforholdet. det første krystalliseringsforsøg blev udført med 3,31 g ‛crude‛ esomeprazolmagnesium, mens det andet krystalliseringsforsøg blev udført med 50,0 g ‛crude‛ esomeprazol- magnesium, dvs. i en 15 gange større målestok. mængderne af methanol, acetone og vand i det andet krystalliseringsforsøg blev ligeledes forøget 15 gange. krystallise- ringstiden i det andet krystalliseringsforsøg var næsten den samme som i det første krystalliseringsforsøg, og temperaturen var præcis den samme. således var de væ- sentlige parametre for en krystallisering (koncentrationer af opløsningsmidlerne, krystalliseringstemperatur og -tid) næsten identiske i de to krystalliseringsforsøg. dermed er det andet krystalliseringsforsøg udført af martin bohlin blot en kopi af det første krystalliseringsforsøg i en 15 gange større målestok. derfor er det ikke overraskende for en fagmand på området, at et tilsvarende resultat opnås i de to krystalliseringsforsøg hvad angår produktets optiske renhed, eftersom den eneste store ændring er en øgning af målestoksforholdet for processen. krystalliseringen af esomeprazolmagnesium hos krka, d. d. udføres under andre be- tingelser end det andet krystalliseringsforsøg beskrevet af martin bohlin hos astra- zeneca ab, dvs. at eksempel 9 i wo 2010/072759 al ikke følges af krka, d. d.‛ - 19 -  to ‛certificates of analysis‛, begge dateret den 22. juli 2010, vedrørende batch k44323 og batch k44479. det fremgår af disse certifikater, at indholdet af r-enantiomeren af omeprazol i de to batches med aktivstof var henholdsvis 0,54 % og 0,38 %, dvs. at den optiske renhed i begge batches var på mindre end 99,8 % e.e.  uddrag af eg s.p.a.’s dmf af 10. december 2010 for fremstilling af esomeprazol mag- nesium dihydrat (bilag 29). af den generiske lægemiddelproducent eg s.p.a.’s åbne dmf af 10. december 2010 for fremstilling af esomeprazol magnesium dihydrat fremgår bl.a.: ‛< omeprazole magnesium ph.eur., is added to result in esomeprazole magnesium di- hydrate containing 0.11% to 0.2% of r-omeprazole. <‛ yderligere dokumenter omhandlende krkas proces for fremstilling af esomeprazol ”krka”, som ikke indgik i fogedrettens afgørelsesgrundlag der er fremlagt bl.a. følgende yderligere materiale:  martin bohlins analyserapport af 25. februar 2011 (bilag 30) – martin bohlins 3. forsøg. i dette forsøg gentog martin bohlin sit forsøg fra sin 2. analyserapport af 8. oktober 2010 (bilag 21). med henblik på at analysere tidens indflydelse på krystallisationen blev dette forsøg dog forlænget til 32,5 timer. det fremgår af analyserapportens bilag 1, at martin bohlin efter udførelsen af krystallisationen opnåede et esomeprazol magnesium dihydrat-salt med en optisk renhed på 100 % e.e og et udbytte af krystallisationen på ‛ca. 72 %‛. - 20 -  uddrag af krkas dmf vedrørende esomeprazol ‛krka‛, som krka har indleveret til de franske lægemiddelmyndigheder (bilag 35b / bilag n). det dokument, som astrazeneca har fremlagt som bilag 35b, er efter oplyste opnået af astrazeneca ved bevissikringssag i frankrig. dokumentet er ikke fuldstændigt. dele af de manglende sider fra dokumentet er efterfølgende fremlagt af krka som bilag n. dokumentet består af løbende opdaterede versioner af dmf’en indleveret til de fran- ske myndigheder i henholdsvis marts 2010, september 2010 og januar 2011. beskrivel- sen af de nærmere procesbetingelser i krkas krystallationsproces er i disse versioner af dmf’en medtaget fuldstændigt og er således ikke anonymiseret. af den version af dmf’en, som krka har indleveret til de franske myndigheder i marts 2010, og som senest er opdateret den 23. december 2009, fremgår om krystallationspro- cessen bl.a.: ‛< process description: esomeprazole magnesium is dissolved in methanol at 25-30°c, optionally purified with carbon carcoal and finally filtered to get a clear solution. solution is concen- trated to assay 30-50 % of [esomeprazole]. concentrated solution is cooled to 20 ± 2 °c, acetone / water mixture (4:1) is gradually added to solution that temperature not exceeded 25°c. suspension stirred for about 12 hours at 20 °c, then about 3 hours at about 5°c, filtered and washed with acetone. wet product is dried in a vacuum or an air dryer with rotation first at temperature from 20 to 35°c for about 6 hours following gradually increase for 10°c up to 70°c if it is necessary, drying is controlled by kf and lod analyses. yield: not less than 70 %. <‛ det er uomtvistet, at denne beskrivelse af krkas krystallisationsproces i al væsentlig- hed svarer til procesbeskrivelsen i krkas tidligere dmf, som er fremlagt som bilag 14, og at beskrivelsen samtidig svarer til procesbeskrivelserne i de efterfølgende opdatere- - 21 - de dmf’er fra september 2010 (som senest er opdateret den 24. juni 2010) og fra januar 2011 (som senest er opdateret den 23. november 2010). det fremlagte uddrag af den version af dmf’en, som blev indleveret til de franske myndigheder i marts 2010, viser desuden analysedata fra en række batches af eso- meprazol ‛krka‛. det fremgår således, at seks ‛pilotbatches‛ (r40208, r40209, r40210, r40264, r40282 og r40283), som alle blev fremstillet af krka i løbet af 2008, havde et indhold af ‛impurity f‛, dvs. r-enantiomeren af omeprazol, på mindre end 0,05 %. det fremgår tilsvarende, at to 64 kg-batches, k44035 og k44048, som begge var produk- tionsbatches fremstillet i januar 2009, havde et meget begrænset indhold af r- enantiomeren af omeprazol (0,01 %). de to 100 kg-produktionsbatches k44323 og k44479, som blev produceret i henholdsvis maj og juli 2009, havde derimod et indhold af r-enantiomeren af omeprazol på 0,54 % og 0,38 %, jf. også de ovennævnte ‛certifi- cates of analysis‛. af dmf’en fremgår herom bl.a.: ‛< < the results of first pilot batches were presented and content of impurity f was be- low reporting limit. the requirement for enantiomeric purity was set to not more than 1.0 % of impurity f. after submission of registration documentation the tech- nological process was optimized with the aim to scale up the process. during optimi- zation of technological process in scale up, higher content (more than 0.2%) of impur- ity f was detected. with the aim to propose robust and optimal technology (with re- gards to quality, yield and current production equipment) the results for impurity f of production scale batch (100
10. kg)should be taken into account. during registration procedure the requirement for enantiomeric purity was additionally tightened to not more than 0.6 % of impurity f based on the batch analyses. since a reasonable range of expected analytical and manufacturing variability should be considered according to ich guideline q6a, additional reducing of the requirements is not acceptable. <‛  vidneerklæring af 16. februar 2011 fra krkas marko herga (bilag s). denne erklæring, som er udarbejdet til brug for en parallel engelsk sag mellem astra- zeneca og krka, indeholder en beskrivelse af den proces, som ifølge krka anvendes til fremstillingen af esomeprazol ‛krka‛. - 22 - af erklæringen fremgår følgende om krystallationsprocessen: ‛< step 3: crystallisation of esomeprazole magnesium salt 3.1 dissolution 28. the crude esomeprazole magnesium salt from step 2 is loaded into the reactor vessel with methanol (vmeoh [l] = 10 x mesome [kg]). active charcoal suspended in methanol (mac [kg] = 0.05 x mesome [kg], vmeoh [l] = 0.3 x mesome [kg]) is added to the suspension. 29. the suspension is heated to 25-30°c and stirred for 30 minutes at this tempera- ture. the stirring is then stopped and the temperature is maintained at 25-30°c for a further 30 minutes. 3.2 filtration 30. the solution is filtered through a‛cuno‛ or gaf filter and a safety (pall) fil- ter 0.2µm. 31. at the end of the filtration the reactor vessel and hose connections are washed with methanol (vmeoh [l] = 0.5 x mesome [kg]). 3.3 vacuum concentration 32. the filtrate from step 3.2 is evaporated under vacuum at maximum negative pressure and a maximum temperature of the heating medium of 35°c until the desired concentration of esomeprazole in solution is reached. the desired con- centration depends on the quantity of the r-enantiomer of omeprazole in the filtrate (which is measured by hplc), as follows: r-enantiomer ≥0.7% – concentrate to 30-45% r-enantiomer 0.5 to 0.7% – concentrate to 45-50% 3.4 crystallisation 33. the temperature of the concentrated solution from step 3.3 is adjusted to 20 ± 2°c. 34. a mixture of acetone and water (4/1) (vacetone [l] = 6.25 x mesome [kg], vwater [l] = 1.6 x mesome [kg]) is added gradually and uniformly over a period of 30 mi- nutes to 1 hour. the suspension is then stirred at a temperature of 20-25°c. 35. after 8 hours (approximately) of stirring, the suspension is sampled. the sam- ple is filtered and the cake made therefrom is tested using hplc to determine the optical purity (the process is not stopped when sampling is performed). if the optical purity of the sample is less than 99.6% e.e., step 3.5 is performed. - 23 - 36. if the optical purity of the sample is not less than 99.6% e.e., the process con- tinues running and the suspension is stirred for a further 4 hours (approx- imately). the suspension is then sampled and tested again. if the optical purity of the sample is less than 99.6% e.e., step 3.5 is performed. 37. if the optical purity of the sample is not less than 99.6% e.e., the process con- tinues running and the suspension is subjected to a further vacuum concentra- tion to evaporate 20% of the total mass. the evaporation is performed under vacuum with a maximum temperature of the heating medium of 35°c and the temperature of the reaction mixture at 15-20°c. the pressure is regulated de- pending on the temperature of the reaction mixture. 38. after the vacuum concentration, the suspension is sampled and tested again. if the optical purity of the sample is less than 99.6% e.e., step 3.5 is performed. 39. if the optical purity of the sample is not less than 99.6% e.e., the process con- tinues running and the suspension is cooled over a period of 3 hours to a tem- perature of 0-5°c. the suspension is sampled and tested again. if the optical purity of the sample is less than 99.6% e.e., step 3.5 is performed. 40. if the optical purity of the sample is not less than 99.6% e.e., the process con- tinues running and water (vwater [l] = 0.10-0.30 x mesome [kg]) is added to the suspension. the suspension is stirred at a temperature of 0-5°c for an addi- tional 3-6 hours. the suspension is sampled and tested again. if the optical purity of the sample is less than 99.6% e.e., step 3.5 is performed. 41. if the optical purity of the sample is not less than 99.6% e.e, the process contin- ues running and methanol (vmeoh [l] = 0.10-0.30 x mesome (kg]) is added to the suspension. the suspension is maintained at a temperature of 0-5°c and sam- pled at intervals until the optical purity of the sample is less than 99.6% e.e. when the optical purity of the sample is less than 99.6% e.e, step 3.5 is per- formed. 3.5 filtration / isolation 42. the following steps are not carried out unless the optical purity of the sample of the suspension is less than 99.6% e.e. 43. the suspension is transferred to a centrifuge and the product is spin-dried. the wet cake is washed on the centrifuge with cooled acetone (0-5°
11. c)(vacetone [l] = 1 x mesome [kg]) and spin-dried again. 44. the product is unloaded from the centrifuge and packed in a double pe bag. the product is kept protected from light and atmospheric influences at a tem- perature of up to 25°c for maximum 1 day. 3.6 drying and packaging 45. the product is dried in an air dryer or vacuum dryer until the water content is less than 5.7% and loss on drying is less than 1.5%. - 24 - 46. after drying, the product is cooled to 25 ± 5°c and granulated/sieved or milled before packaging.‛ det er ubestridt, at trin 37-41 i marko hergas erklæring ikke fremgår af krkas dmf (bi- lag 35b).  erklæring af 7. april 2015 fra rick longton (bilag 38). af erklæringen, der er udarbejdet efter et besøg på krkas produktionsfacilitet i slove- nien, fremgår bl.a.: ‛< 5) the final steps of krka’s manufacturing process include the crystallization and drying of esomeprazole magnesium. 6) attached hereto as exhibit a is the batch record (and certified english lan- guage translation) provided to me by krka which describes the crystallization of esomeprazole magnesium, lot k52016, which began on february 10, 2015. i observed many of the steps described in the batch record. 7) in the following paragraphs, i describe what i observed during the crystalliza- tion of esomeprazole magnesium at krka’s manufacturing plant in novo mes- to, slovenia on february 10-13, 2015. day 1—february 10, 2015—crystallization of esomeprazole magnesium 8) 7:35 am: i arrived at krka’s manufacturing plant in novo mesto. 9) 8:15 am: i observed the addition of methanol to reaction vessel r-09. prior to the addition of methanol to the vessel, i looked inside the vessel and it ap- peared to be clean and empty. this is described in step 3 on page 6 of the batch record, exhibit a. 10) 8:30 am: i observed the addition of crude esomeprazole magnesium (lots k51957 and k51958) to reaction vessel r-09. this is described in step 6 on page 6 of the batch record, exhibit a. 11) 9:00 am: i observed the addition of activated charcoal followed by methanol to reaction vessel r-09. this is described in steps 7-8 on page 6 of the batch record, exhibit a. 12) 10:05 am: i observed the filtration of the material in reaction vessel r-09 through a 5µm filter followed by a 2µm filter to remove the charcoal. the fil- - 25 - trate was charged into reaction vessel r-10. this is described in step 13 on page 7 of the batch record, exhibit a. 13) 10:30 am: i returned to a conference room provided by krka to wait for the fil- tration to be completed. i was informed that after the filtration step, the filtrate in reaction vessel r-10 would be concentrated by vacuum evaporation of the solvent over the course of several hours. this is described in step 15 on page 8 of the batch record, exhibit a. 14) 6:35 pm: i returned to reaction vessel r-10 to observe the final minutes of the vacuum evaporation of solvent. this is described in step 15 on page 8 of the batch record, exhibit a. 15) 7:20 pm: i observed a sample removed from reaction vessel r-10. this is de- scribed in step 16 on page 9 of the batch record, exhibit a. 16) 7:25 pm: i went to krka’s analytical laboratory to observe analysis of the sam- ple by laboratory technicians. 17) 8:15 pm: i returned to reaction vessel r-10 to observe the addition of acetone and water to the vessel. this is described in step 27 on page 11 of the batch record, exhibit a. 18) 9:00 pm: [crystallization begins] i looked into reaction vessel r-10 through a porthole and observed that the solution was slightly pink and appeared to have the consistency of a milkshake. i was informed by a krka employee that this was an indication that crystallization of esomeprazole magnesium had started and would proceed for the next several hours. 19) 10:10 pm: i observed the removal of sample ‛k52016 ipc-4 dod a 1 hr‛ from reaction vessel r-10. 20) 10:20 pm: i went to krka’s analytical laboratory where sample ‛k52016 ipc-4 dod a 1 hr‛ was processed. i left the plant and returned to my hotel. day 2—february 11, 2015—crystallization of esomeprazole magnesium 21) 4:45 am: i arrived back at the plant. 22) 5:00 am: [8 hour sample] i returned to reaction vessel r-10. i observed sample ‛k52016/ipc-4/doda/8h‛ (the 8 hour sample) removed from the reaction ves- sel. the sample was placed into 2 clear glass vials that had been labeled with the sample name. this is described in step 29 on page 12 of the batch record, exhibit a. 23) 5:10 am: i returned to the analytical lab where the 8 hour sample was given to a lab technician. i observed how the sample was analyzed as follows: a. 10 ml of the 8 hour sample was vacuum filtered through a büchner funnel with filter paper in a fume hood. the sample was washed with 5 ml of acetone. filtration and drying lasted about 5 minutes. - 26 - b. 5:20 am: the dried sample, which was a white powder, was taken to an analytical balance. 50.09 mg of the dry powder was added to a 25 ml volumetric flask. methanol (ultra gradient hplc grade jt baker) was added to the volumetric flask. one to two drops of acetic acid we- re added from a syringe into the volumetric flask. methanol was add- ed to the volumetric flask to 25 ml. the volumetric flask was inverted several times to mix. about 3 ml of the solution was then placed into a small hplc sample vial and taken to the hplc. c. 5:30 am: the hplc sample vial was placed into the sample tray of an agilent technologies hplc. using an auto injector, 20 µl of the sam- ple was injected into the hplc which was equipped with a diacel chiralpak ia column (5 micron particle size, 250 x 4.6 mm). the sol- vent system was hexane:ethanol(dehydrated):methanol (75:20:5). the run time was approximately 20 minutes. 24) attached as exhibit b is the analytical report i was given for the 8 hour sam- ple and a certified english translation of the report. this is described in step 29 of the batch record, exhibit a (the analytical report is designated order no. 61). 25) 6:00 am, 7:00 am, 8:00 am: [9, 10 and 11 hour samples] i observed three addi- tional samples taken from reaction vessel r-10 at approximately one hour in- tervals which i refer to as the 9, 10 and 11 hour samples. the samples were tested by the lab technician as described above in paragraph 23) a-c. attached as exhibits c, d and e are the analytical reports i was given for the 9, 10 and 11 hour samples, respectively. this is described in step 29 of the batch record, exhibit a (the analytical reports are designated order nos. 62, 63 and 64, re- spectively). 26) 8:40 am: i returned to reaction vessel r-10. 27) 9:00 am: i observed the beginning of the vacuum concentration which is de- scribed in step 30 on page 12 of the batch record, exhibit a. i then returned to the conference room. 28) 9:35 am: i returned to reaction vessel r-10 to observe the removal of sample ‛k52016 ipc-4/dodb/½ konc.‛ after removal of the sample, the vacuum con- centration continued. attached as exhibit f is the analytical report i was giv- en for this sample. this is described in step 32 on page 13 of the batch record, exhibit a (the analytical report is designated order no. 65). 29) 11:20 am: i returned to reaction vessel r-10 to observe the removal of sample ‛k52016 ipc-4/dodb/ 0h.‛ attached as exhibit g is the analytical report i was given for this sample. this is described in step 32 on page 13 of the batch record, exhibit a (the analytical report is designated order no. 66). 30) 12:15 pm: i left the plant and returned to my hotel. day 3—february 12, 2015—crystallization of esomeprazole magnesium - 27 - 31) 8:00 am: i arrived back at the plant and was informed that sampling and ana- lytical testing of the samples had [been] proceeding every hour since i had left the previous day and throughout the night. i was also informed that 30 kg of water was added to reaction vessel r-10 as described in step 33 of the batch record, exhibit a. 32) 8:25 am: i returned to reaction vessel r-10. when i looked inside the reaction vessel through a glass porthole, i observed heavy caking of esomeprazole magnesium crystals on the walls and the shaft of the stirring mechanism in the vessel. the crystals were above the level of the liquid in the reaction vessel. 33) 8:30 am-10:00 am: i visited the analytical laboratory and returned to the confe- rence room. i then returned to the hotel. 34) 3:20 pm: i arrived at the plant. 35) 3:30 pm: i returned to reaction vessel r-10 to observe the procedure used to harvest the esomeprazole magnesium crystals. i looked into the reaction vessel and observed heavy amounts of esomeprazole magnesium crystals caked on the walls of the reaction vessel and on the shaft of the stirring mechanism. krka employee marco herga confirmed that it was esomeprazole magnesium crystals caked on the walls of the vessel and the shaft of the stirring mechan- ism. the crystals were above the level of the liquid in the vessel. 36) 3:35 pm: i observed the contents of reaction vessel r-10 drain into a centrifuge which was located one floor below the reaction vessel. i was informed that the reaction vessel r-10 would be washed with acetone to remove crystals en- crusted on the walls of the reaction vessel once the liquid contents of the vessel had been drained into the centrifuge. 37) 3:40 pm: i arrived at centrifuge c-11 in room 119 which had already started spinning. a label indicated the sample in the centrifuge was lot k52016. 38) 4:00 pm: i returned to the conference room. while i was in the conference room, i examined the results of analytical reports (order nos. 67-93) for the testing that had taken place during the crystallization process. the table below summarizes the data i observed in the analytical reports: table 1: data from crystallization of krka esomeprazole magnesium lot k52016 hours of analytical approximate quantitated detected %s** calculated crystallization report sample time %r %r enantiomeric order no. omeprazole* omeprazole excess (e.e.) of esomeprazole magnesium*** addition of acetone (step 27 of batch record) 8 61 5:00 am ≤ 0.05 0.029 99.971 ≥ 99.9 11-feb-2015 9 62 6:00 am ≤ 0.05 0.029 99.971 ≥ 99.9 11-feb-2015 10 63 7:00 am ≤ 0.05 0.014 99.986 ≥ 99.9 11-feb-2015 - 28 - hours of analytical approximate quantitated detected %s** calculated crystallization report sample time %r %r enantiomeric order no. omeprazole* omeprazole excess (e.e.) of esomeprazole magnesium*** 11 64 8:00 am ≤ 0.05 0.014 99.986 ≥ 99.9 11-feb-2015 13 65 10:00 am ≤ 0.05 0.014 99.986 ≥ 99.9 11-feb-2015 15 66 noon ≤ 0.05 0.020 99.980 ≥ 99.9 11-feb-2015 16 67 1:00 pm ≤ 0.05 0.014 99.986 ≥ 99.9 11-feb-2015 17 68 2:00 pm ≤ 0.05 0.016 99.984 ≥ 99.9 11-feb-2015 18 69 3:pm ≤ 0.05 0.012 99.988 ≥ 99.9 11-feb-2015 addition of water (step 33 of batch record) 19 70 4:00 pm ≤ 0.05 0.010 99.990 ≥ 99.9 11-feb-2015 20 71 5:00 pm ≤ 0.05 0.011 99.989 ≥ 99.9 11-feb-2015 21 72 6:00 pm ≤ 0.05 0.013 99.987 ≥ 99.9 11-feb-2015 22 73 7:00 pm ≤ 0.05 0.010 99.990 ≥ 99.9 11-feb-2015 23 74 8:00 pm ≤ 0.05 0.021 99.979 ≥ 99.9 11-feb-2015 24 75 9:00 pm ≤ 0.05 0.012 99.988 ≥ 99.9 11-feb-2015 25 76 10:00 pm 0.09 0.09 99.91 99.8 11-feb-2015 26 77 11:00 pm ≤ 0.05 0.021 99.979 ≥ 99.9 11-feb-2015 27 78 midnight ≤ 0.05 0.041 99.959 ≥ 99.9 11-feb-2015 28 79 1:00 am ≤ 0.05 0.016 99.984 ≥ 99.9 12-feb-2015 29 80 2:00 am ≤ 0.05 0.033 99.967 ≥ 99.9 12-feb-2015 30 81 3:00 am ≤ 0.05 0.013 99.987 ≥ 99.9 12-feb-2015 31 82 4:00 am ≤ 0.05 0.015 99.985 ≥ 99.9 12-feb-2015 32 83 5:00 am ≤ 0.05 0.019 99.981 ≥ 99.9 12-feb-2015 33 84 6:00 am ≤ 0.05 0.020 99.980 ≥ 99.9 12-feb-2015 34 85 7:00 am ≤ 0.05 0.022 99.978 ≥ 99.9 12-feb-2015 35 86 8:00 am ≤ 0.05 0.04 99.96 ≥ 99.9 12-feb-2015 36 87 9:00 am ≤ 0.05 0.05 99.95 ≥ 99.9 12-feb-2015 37 88 10:00 am 0.09 0.09 99.91 99.8 12-feb-2015 38 89 11:00 am 0.1 0.1 99.9 99.8 12-feb-2015 39 90 noon 0.12 0.12 99.88 99.7 12-feb-2015 40 91 1:00 pm 0.16 0.16 99.84 99.6 12-feb-2015 41 92 2:00 pm 0.19 0.19 99.81 99.6 12-feb-2015 42 93 3:00 pm 0.21 0.21 99.79 99.5 12-feb-2015 * the level of quantitation in the hplc assay for %r is 0.05% ** %s = 100 - %r *** e.e. = {([s]-*r+)÷(*s++*r+)} x 100 < - 29 - 39) 4:30 pm: i left the krka plant and returned to my hotel. 40) 9:40 pm: i received a call from a krka employee informing me that the acetone wash of reaction vessel r-10 would begin at 10:30 pm. 41) 10:15 pm: i arrived at the plant. 42) 10:30 pm: i arrived at reaction vessel r-10. i looked inside the vessel and ob- served that all of the liquid had been drained to the centrifuge. i also observed that the walls and shaft of the stirring mechanism of the vessel were heavily caked with crystals of esomeprazole magnesium. acetone was added to the vessel and i could see the crystals of esomeprazole magnesium washing off the walls and down the drain leading to the centrifuge. 43) 10:40 pm: a second aliquot of acetone was added to reaction vessel r-10. i ob- served clear white streams of esomeprazole magnesium crystals washing off the walls and the shaft of the stirring mechanism. the crystals came off in clumps and formed many streams of particles going down the drain into the centrifuge. 44) 10:45 pm: the washing was completed and i proceeded to centrifuge c-37 which was labeled with lot k52016. i observed white solution entering the cen- trifuge through a pipe. i was informed that the pipe led back to reaction vessel r-10 where i had just been.‛  ‛batch record‛ fra fremstillingen af batch k52016, som bl.a. rick longton overværede i februar 2015 på krkas produktionsfacilitet i slovenien (bilag 39). krkas repræsentanter udleverede efter det oplyste både batch record samt de løbende prøveresultater til rick longton under besøget hos krka i februar 2015 (hvilket rick longton har vedlagt som ‛exhibit a‛ til sin rapport). rick longton fik dog ikke udle- veret bagsiden af side 15 af 28 i batch record; denne side er efterfølgende blevet frem- lagt af krka som bilag v. af batch record (bilag 39, inkl. bilag
12. v)fremgår bl.a. følgende om den gennemførte krystallisationsproces i februar 2015 (gengivelsen nedenfor indeholder alene udvalgte kolonner fra dokumentet): - 30 - ‛< no. step requirement actual time note/ipc regulation from to phase 2: crystallization of esomeprazole magnesium [salt] 1. starting working operations in 07:17 pas crystallization of esomeprazole magnesium [salt] 2. inertize the reactor r-09 with n2 07:21 07:35 < (put in a vacuum twice until reaching the depression of up to 200 mbar and introduce nitrogen until reaching atmospheric pres- sure) 3. batch into a dry and clean reactor mmethanol = mmethanol = 08:11 08:21 < r-09 1050±50 kg of methanol 1050±50 kg 1051 kg