Crit­i­cal Junc­tures in Gene Ther­a­py Clin­i­cal Ma­te­r­i­al Man­u­fac­tur­ing: Where Do I Start? And Will It Work?

Strate­gic ap­proach­es to suc­cess­ful ade­no-as­so­ci­at­ed virus (AAV) man­u­fac­tur­ing cam­paigns of­ten in­clude ro­bust and ear­ly an­a­lyt­i­cal char­ac­ter­i­za­tion, as well as con­sis­tent, da­ta-dri­ven re-eval­u­a­tion of man­u­fac­tur­ing plans through­out a pro­gram’s life­cy­cle. Holis­tic prod­uct de­vel­op­ment of ther­a­peu­tics uti­liz­ing AAV vec­tors re­quires the syn­er­gis­tic con­sid­er­a­tion of pre-clin­i­cal or clin­i­cal de­vel­op­ment plans over­layed with a thought­ful AAV man­u­fac­tur­ing strat­e­gy. Both prod­uct de­vel­op­ment and man­u­fac­tur­ing will re­quire an in­di­ca­tion-cen­tric and pa­tient-cen­tric ap­proach. How­ev­er, AAV man­u­fac­tur­ing pro­grams must al­so ac­com­mo­date ma­te­r­i­al needs, study time­lines, and prod­uct-spe­cif­ic re­quire­ments, of­ten dic­tat­ed by the route of ad­min­is­tra­tion, in­tend­ed dose, po­ten­cy, buffer com­pat­i­bil­i­ty, and tis­sue/or­gan sen­si­tiv­i­ty to the prod­uct. While ma­te­r­i­al need and avail­abil­i­ty may be quick­ly pre­dict­ed based on yield pro­jec­tions for the cho­sen serotype and study de­sign, po­si­tion­ing an AAV man­u­fac­tur­ing pro­gram for suc­cess al­so re­quires an un­der­stand­ing of the AAV prod­uct it­self and po­ten­tial process ad­just­ments that may en­able greater suc­cess. At Forge Bi­o­log­ics, we lever­age our be­spoke plat­form process cou­pled with Drug Mas­ter Files for the Forge AAV man­u­fac­tur­ing process and plas­mid de­signs and man­u­fac­tur­ing strate­gies, which pro­vide our clients an adapt­able start­ing point in de­vel­op­ing their man­u­fac­tur­ing process. Our ex­ten­sive ex­pe­ri­ence in man­u­fac­tur­ing nat­ur­al and nov­el AAV serotypes us­ing our plat­form tech­nolo­gies al­lows ac­cel­er­at­ed tran­si­tion from small-scale pi­lot pro­duc­tion through cGMP Clin­i­cal Tri­al Ma­te­r­i­al (CTM) pro­duc­tion. With nu­mer­ous de­vel­op­ment stud­ies, scales, and grades of AAV avail­able to build man­u­fac­tur­ing path­ways that will suc­cess­ful­ly reach pro­gram mile­stones, a com­mon two-fold ques­tion re­mains: ‘Where do I start, and will it work?’

While the start­ing point for most man­u­fac­tur­ing pro­grams will be a small-scale plat­form pi­lot run, the learn­ings from this pi­lot run may in­form a labyrinth of next steps. To es­tab­lish the vast scope of op­tion­al­i­ty avail­able when pur­su­ing prod­uct-spe­cif­ic process de­vel­op­ment, our tech­ni­cal teams have com­piled a list of more than fifty stud­ies that may be pur­sued to op­ti­mize drug sub­stance and drug prod­uct man­u­fac­tur­ing process­es. Op­ti­miz­ing yield—or in­creas­ing pu­ri­ty of the fi­nal prod­uct—is of­ten es­sen­tial to long-term pro­gram fea­si­bil­i­ty, but in­ter­est­ing­ly, two of the most im­pact­ful strate­gies to guide suc­cess are in­de­pen­dent of prod­uct, serotype, ther­a­peu­tic in­di­ca­tion, or in­tend­ed man­u­fac­tur­ing scale. Be­low, we in­ves­ti­gate the high-pow­ered strate­gies that pro­vide pro­grams a foun­da­tion for suc­cess by, 1) build­ing a pro­gram around ro­bust ear­ly an­a­lyt­ics, and 2) con­tin­u­ous re-eval­u­a­tion of your man­u­fac­tur­ing ap­proach as da­ta be­comes avail­able.

Build­ing a Pro­gram with Ro­bust Ear­ly An­a­lyt­ics

Where do I start?

To be­gin de­sign­ing a pro­gram that will gen­er­ate a ro­bust an­a­lyt­i­cal char­ac­ter­i­za­tion pack­age for reg­u­la­to­ry fil­ings, the qual­i­ty tar­get prod­uct pro­file (QTPP) is first draft­ed for the prod­uct. The QTPP, as de­fined by ICH Q8(R2), is ‘a prospec­tive sum­ma­ry of the qual­i­ty char­ac­ter­is­tics of a drug prod­uct that ide­al­ly will be achieved to en­sure the de­sired qual­i­ty, tak­ing in­to ac­count safe­ty and ef­fi­ca­cy of the drug prod­uct’¹. This doc­u­ment will out­line the tar­get char­ac­ter­is­tics and clin­i­cal re­quire­ments for the ther­a­peu­tic prod­uct, such as the route of ad­min­is­tra­tion, dose strength, and prod­uct sta­bil­i­ty, thus pro­vid­ing the foun­da­tion for iden­ti­fy­ing qual­i­ty at­trib­ut­es (QAs). QAs can then be as­sessed for their po­ten­tial im­pact to the prod­uct in or­der to de­ter­mine pre­lim­i­nary crit­i­cal qual­i­ty at­trib­ut­es (CQAs). CQAs are de­fined as phys­i­cal, chem­i­cal, bi­o­log­i­cal, or mi­cro­bi­o­log­i­cal prop­er­ties or char­ac­ter­is­tics that should be with­in an ap­pro­pri­ate lim­it, range or dis­tri­b­u­tion to en­sure the de­sired prod­uct qual­i­ty¹. CQAs are typ­i­cal­ly bro­ken in­to char­ac­ter­is­tics that re­late to the pu­ri­ty, safe­ty, quan­ti­ty, iden­ti­ty, and po­ten­cy of the drug prod­uct. For AAV, ex­am­ple CQAs may in­clude phys­i­cal titer, in­fec­tious titer, and en­do­tox­in lev­els, as an out-of-range re­sult of any sin­gle char­ac­ter­is­tic may neg­a­tive­ly im­pact prod­uct qual­i­ty or safe­ty. Af­ter a list of CQAs with pro­ject­ed tar­get ranges that com­ply with the guide­lines in the QTPP have been gen­er­at­ed, an as­say pan­el suit­able to eval­u­ate the process and prod­uct can be de­fined.

In prac­tice, align­ing on a QTPP and CQAs for a yet to be man­u­fac­tured AAV prod­uct is a dif­fi­cult task, and of­ten takes a retroac­tive route. An ear­ly draft may rep­re­sent a brief, ide­al­ized prod­uct pro­file, with an un­der­stand­ing that con­tin­u­ous re­vi­sions will be made through­out the pro­gram life­cy­cle. Defin­ing the ide­al cri­te­ria for the prod­uct is the start­ing point for en­sur­ing ear­ly as­say pan­els pro­vide suf­fi­cient in­sight to sup­port or jus­ti­fy re­vi­sion of your CQA tar­gets.

 Will it work?

With the QTPP and CQAs in place, ro­bust as­say pan­els—in­clu­sive of a phase-ap­pro­pri­ate po­ten­cy as­say² for prod­uct char­ac­ter­i­za­tion—will fur­ther help di­rect a path­way to cGMP man­u­fac­tur­ing readi­ness. From the first pro­duc­tion run on­ward, thought­ful as­say pan­el de­sign al­lows da­ta-dri­ven re­fine­ment of CQAs, with the goal of un­der­stand­ing the link be­tween po­ten­cy and process. Nar­row­ing CQA tar­get ranges and re­duc­ing the num­ber for CQAs is an ex­pec­ta­tion of pro­gram ma­tu­ri­ty and process con­trol as the pro­gram pro­gress­es to lat­er clin­i­cal and com­mer­cial stages. Char­ac­ter­iz­ing in-process and fi­nal sam­ples from ear­ly pro­duc­tion lots can pro­vide in­sight to guide de­vel­op­ment stud­ies that are nar­row in scope, rather than post-hoc analy­ses which can lead to time-con­sum­ing in­ves­ti­ga­tion ef­forts. With the goal of de­vel­op­ing a con­trolled AAV process and con­sis­tent prod­uct, ro­bust ear­ly an­a­lyt­ics pois­es a pro­gram for suc­cess by uti­liz­ing each pro­duc­tion batch to gain valu­able in­sight.

Con­tin­u­ous Re-Eval­u­a­tion of Man­u­fac­tur­ing Ap­proach

Where do I start?

Now that the AAV prod­uct de­vel­op­ment path­way and goals are clear­ly de­fined by the QTPP, CQAs, and an­a­lyt­i­cal plan, the pro­gram has an es­tab­lished base­line to mea­sure suc­cess. Con­tin­u­ous re-eval­u­a­tion of the man­u­fac­tur­ing path­way for an AAV prod­uct be­gins with build­ing a strong and adapt­able team. A well-round­ed team ca­pa­ble of or­ga­niz­ing da­ta, in­ter­pret­ing re­sults, crit­i­cal­ly think­ing, and redi­rect­ing when re­quired, may con­sist of pro­fes­sion­als span­ning pro­gram man­age­ment, tech­ni­cal pro­gram strat­e­gy, and reg­u­la­to­ry af­fairs, as well as man­u­fac­tur­ing sub­ject mat­ter ex­perts. A com­mon mis­con­cep­tion is that the man­u­fac­tur­ing path­way is in­flex­i­ble af­ter pro­gram ini­ti­a­tion. How­ev­er, pro­gram changes and re­align­ments should be viewed as con­tin­u­ous im­prove­ments, ap­plied af­ter learn­ing more about the AAV process and prod­uct. To this end, af­ter each pro­duc­tion run, da­ta should be com­piled and eval­u­at­ed metic­u­lous­ly against the QTPP, de­sired CQA tar­gets, and fu­ture ma­te­r­i­al needs. Uti­liz­ing a process in­for­ma­tion man­age­ment soft­ware (PIMS) or cre­at­ing an in-house so­lu­tion to da­ta or­ga­ni­za­tion is key for ef­fi­cient de­ci­sion-mak­ing and pro­gram eval­u­a­tion. Ad­di­tion­al­ly, in­cor­po­rat­ing at-scale man­u­fac­tur­ing runs will gen­er­ate the most rel­e­vant da­ta to bet­ter un­der­stand the in­her­ent vari­abil­i­ty of your prod­uct.

Will it work?

With a cross-func­tion­al team of ex­perts, ide­al­ly with rep­re­sen­ta­tion from the prod­uct spon­sor as well as the man­u­fac­tur­ing part­ner, con­sis­tent re-eval­u­a­tion of da­ta and pro­gram strat­e­gy will pro­vide ben­e­fit to any ther­a­peu­tic de­vel­op­ment pro­gram in the fol­low­ing ways:

• CQA tar­gets can be eval­u­at­ed for fea­si­bil­i­ty and proac­tive­ly ad­just­ed as need­ed for sub­se­quent man­u­fac­tur­ing runs, with the goal of achiev­able and ac­cu­rate tar­gets in place for a Tox­i­col­o­gy or cGMP run.
• CQA tar­get ranges can be nar­rowed and re­fined to re­flect process and prod­uct un­der­stand­ing, as well as pro­gram ma­tu­ri­ty—a re­quire­ment for lat­er stage clin­i­cal in­ves­ti­ga­tions and com­mer­cial­iza­tion.
• Ad­di­tion­al process de­vel­op­ment needs can be iden­ti­fied ear­ly in the man­u­fac­tur­ing scale-up path­way to pre­serve the ap­plic­a­bil­i­ty of sub­se­quent da­ta.
• Ear­li­er align­ment in man­u­fac­tur­ing scale ad­just­ments, or rec­om­mend­ed stud­ies to achieve ma­te­r­i­al needs can be proac­tive­ly im­ple­ment­ed.

 Case Study

To ex­plore the ben­e­fits of ro­bust prod­uct char­ac­ter­i­za­tion ear­ly in the man­u­fac­tur­ing path­way, the ex­am­ple be­low out­lines a case study fea­tur­ing two pro­grams forg­ing dif­fer­ent paths to ar­rive at cGMP Clin­i­cal Tri­al Ma­te­r­i­al. In this case, Pro­gram 1 uti­lized a very com­mon ap­proach of the min­i­mum vi­able path, which is suit­able for high yield­ing, read­i­ly man­u­fac­turable prod­ucts. How­ev­er, Pro­gram 1 en­tered a cGMP man­u­fac­tur­ing cam­paign with­out da­ta on sev­er­al crit­i­cal resid­ual as­says, risk­ing the over­all suc­cess of the cam­paign. On the oth­er hand, Pro­gram 2 in­vest­ed in ear­ly AAV process and prod­uct un­der­stand­ing at small scale, as well as a full-scale en­gi­neer­ing run. Pro­gram 2 em­ployed a more risk-ad­verse ap­proach, which may have been prompt­ed by ear­ly re­sults in con­flict with the goals out­lined in their QTPP. Both pro­grams uti­lized the same an­a­lyt­i­cal meth­ods for cGMP CTM re­lease; how­ev­er, Pro­gram 2’s ap­proach pro­vid­ed in­creased batch his­to­ry and repli­cate da­ta, which sub­se­quent­ly in­creased sta­tis­ti­cal pow­er to nar­row spec­i­fi­ca­tion ranges that would achieve the goals of their QTPP.

Case Study: Pro­gram 1

Case Study: Pro­gram 2

Defin­ing Pro­gram Suc­cess Cri­te­ria Prospec­tive­ly

At the out­set of a new AAV ther­a­peu­tics pro­gram, draft­ing suc­cess cri­te­ria, an an­a­lyt­ics plan, and as­sem­bling a ver­sa­tile team for con­sis­tent pro­gram eval­u­a­tion will es­tab­lish a strong foun­da­tion that may with­stand com­mon pres­sures en­coun­tered along the de­vel­op­ment path­way. With these strate­gies for suc­cess in place, events such as unan­tic­i­pat­ed re­sults or shift­ing dead­lines can be met with ra­tio­nal and da­ta-dri­ven pro­gram ad­just­ments. Prepar­ing pro­gram re­quire­ments be­fore man­u­fac­tur­ing da­ta is avail­able is a daunt­ing task, but the ben­e­fits of cre­at­ing a base­line against which to mea­sure suc­cess should not be over­looked. Uti­liz­ing re­sources such as aca­d­e­m­ic lit­er­a­ture, in­for­ma­tion on the in­di­ca­tion or pa­tient nat­ur­al his­to­ry, or knowl­edge from a plat­form man­u­fac­tur­ing process that has en­coun­tered many prod­uct types, may al­low a QTPP to be es­tab­lished which can be fur­ther honed through­out the prod­uct de­vel­op­ment jour­ney.

Time, risk, and cost are of­ten in op­po­si­tion when it comes to cre­at­ing an AAV man­u­fac­tur­ing path­way. On one ex­treme, the most risk ad­verse plan will be the least cost- and time-ef­fi­cient. On the oth­er ex­treme, the most cost- and time-ef­fi­cient plan of­ten car­ries most risk for the pro­gram spon­sor. To bal­ance this, an in­ter­me­di­ate ap­proach to gen­er­ate a suf­fi­cient amount of da­ta with­out sur­pass­ing ac­cept­able dead­lines or bud­get, is a com­mon way to bal­ance risk, suc­cess, and pa­tient safe­ty. Giv­en that most AAV gene ther­a­pies are cur­rent­ly en­deav­or­ing to cre­ate treat­ment op­tions for pa­tients with rare, and of­ten fa­tal dis­eases, the great­est risk may be not reach­ing the clin­ic at all. To be a good part­ner to pro­gram spon­sors and the pa­tient com­mu­ni­ties they serve, Forge pro­vides flex­i­bil­i­ty in man­u­fac­tur­ing so­lu­tions and client-fo­cused teams that can be tai­lored to the re­quire­ments of each in­di­vid­ual pro­gram. With a strong strat­e­gy in place from the out­set, when chal­lenges arise, ‘There is al­ways a so­lu­tion.’

Ad­di­tion­al con­tri­bu­tions from:

• An­gela Coy, Ph.D., Se­nior Man­ag­er, Reg­u­la­to­ry Af­fairs
• Ju­lianne Bartz, Man­ag­er, Reg­u­la­to­ry Af­fairs

Ref­er­ences

¹ ICH Q8(R2) https://data­base.ich.org/sites/de­fault/files/Q8_R2_Guide­line.pdf

² U.S. De­part­ment of Health and Hu­man Ser­vices. Food & Drug Ad­min­is­tra­tion. Cen­ter for Bi­o­log­ics Eval­u­a­tion and Re­search. De­cem­ber 2023. (Draft) Po­ten­cy As­sur­ance for Cel­lu­lar and Gene Ther­a­py Prod­ucts. https://www.fda.gov/me­dia/175132/down­load