Ad­dress­ing the ‘Ca­pac­i­ty Crunch’ with a Scal­able Plat­form Process Ap­proach

The field of gene ther­a­py has been dili­gent­ly mov­ing for­ward over the past sev­er­al decades to bring po­ten­tial­ly life-sav­ing treat­ments to pa­tients with ge­net­ic dis­eases. In ad­di­tion to two ap­proved ade­no-as­so­ci­at­ed vi­ral (AAV) gene ther­a­pies, there are more than 250 AAV gene ther­a­pies in var­i­ous clin­i­cal tri­al stages.1 AAV vec­tors re­main the most fre­quent­ly used vec­tor for de­liv­er­ing ther­a­peu­tic trans­genes to tar­get tis­sues due to their demon­strat­ed and last­ing clin­i­cal ef­fi­ca­cy and ex­ten­sive safe­ty track record. As AAV ther­a­pies ad­vance through clin­i­cal tri­als and in­to com­mer­cial­iza­tion, many biotech com­pa­nies are turn­ing to con­tract de­vel­op­ment and man­u­fac­tur­ing or­ga­ni­za­tions (CD­MOs) to pre­pare their pro­grams for late-stage clin­i­cal and com­mer­cial scale man­u­fac­tur­ing. Giv­en the scope and scale of the man­u­fac­tur­ing needs that will ac­com­pa­ny reg­u­la­to­ry ap­provals for these as­sets, CD­MOs con­tin­ue to ex­pand their ca­pac­i­ty to meet the needs of in­creas­ing preva­lent pa­tient pop­u­la­tions. How­ev­er, de­spite rapid growth, pro­ject­ed gene ther­a­py man­u­fac­tur­ing de­mands still out­pace the col­lec­tive ca­pac­i­ty of the CD­MO in­dus­try.


Glob­al Mar­ket De­mand. Da­ta from an in­de­pen­dent third-par­ty re­port shows the per­sis­tent un­met gene ther­a­py man­u­fac­tur­ing de­mand, de­spite an in­crease in CD­MO ca­pac­i­ty over the next sev­er­al years.

The Bat­tle Be­tween Time and Cost with a Cus­tom vs. a Plat­form Process

En­cour­age­ment from reg­u­la­to­ry agen­cies has prompt­ed more CD­MOs to adopt a plat­form process ap­proach. As knowl­edge and ex­pe­ri­ence from the past 40 years is ap­plied, man­u­fac­tur­ing strate­gies can be ac­cel­er­at­ed to in­crease ac­cess to life-sav­ing gene ther­a­pies. How­ev­er, terms such as ‘plug and play,’ or ‘turnkey,’ are over-sim­pli­fied ex­pla­na­tions of this ap­proach. In­stead, a plat­form ap­proach with the flex­i­bil­i­ty to adapt to in­di­vid­ual client needs can pro­vide a head start on de­vel­op­ment that saves both time and mon­ey. On av­er­age, ini­ti­at­ing de­vel­op­ment us­ing a pre-ex­ist­ing man­u­fac­tur­ing plat­form can re­sult in a process reach­ing cGMP readi­ness ap­prox­i­mate­ly 30% to 60% faster than a process with dif­fer­ent lev­els of cus­tomiza­tion. An ad­di­tion­al ad­van­tage of a plat­form process is that the up­front in­vest­ment of time and ef­fort to char­ac­ter­ize and con­tin­u­ous­ly un­der­stand the process is the re­spon­si­bil­i­ty of the CD­MO part­ner. This al­lows a client the flex­i­bil­i­ty to mod­i­fy dif­fer­ent vari­ables, which may be rec­om­mend­ed on a prod­uct-spe­cif­ic ba­sis, and en­ables be­spoke out­puts. In this way, a plat­form process pro­vides the best of both worlds: an ap­proach that al­lows a quick path from the bench­top to the clin­ic but can ad­dress each prod­uct’s unique con­sid­er­a­tions.

When is a Plat­form Process Rec­om­mend­ed?

  1. When a plat­form process can be uti­lized to jump-start man­u­fac­tur­ing with a known, eas­i­ly pro­duced prod­uct (such as a sin­gle-strand­ed, eas­i­ly pack­aged AAV9), com­pared to a one-size-fits-all ap­proach for po­ten­tial­ly dif­fi­cult prod­ucts (such as a self-com­ple­men­tary, over­stuffed AAV2)
  2. When a pro­gram re­quires small dos­es to achieve ther­a­peu­tic out­comes
  3. When a pro­gram is tar­get­ing a dis­ease with low in­ci­dence in the pop­u­la­tion

An ex­am­ple of a pro­gram with a low vec­tor genome need is an AAV gene ther­a­py for treat­ing re­tini­tis pig­men­tosa, an in­her­it­ed reti­nal dis­ease, which may re­quire as lit­tle as 1e11 vg/eye. Yields from a small batch size (~50L work­ing vol­ume) us­ing a plat­form process would pro­vide enough ma­te­r­i­al to treat hun­dreds of pa­tients with a low­er pro­duc­ing AAV2 vec­tor, or thou­sands of pa­tients with a high­er pro­duc­ing AAV8 vec­tor. In this in­stance, the im­me­di­ate time sav­ings and speed of ther­a­peu­tic op­tions for pa­tients out­weigh fu­ture cost ef­fi­cien­cies.

A dif­fer­ent and more chal­leng­ing ex­am­ple is a pro­gram re­quir­ing a large vec­tor quan­ti­ty, such as one tar­get­ing a pa­tient pop­u­la­tion with mus­cu­lar dis­ease. This may re­quire an up­front in­vest­ment of time to in­crease trans­fec­tion ef­fi­cien­cy and pro­duc­tiv­i­ty to max­i­mize yield per batch. Per­form­ing this cus­tom op­ti­miza­tion may in­crease the fea­si­bil­i­ty of fu­ture prod­uct com­mer­cial­iza­tion by re­duc­ing batch sizes or pro­duc­tion fre­quen­cy. For pro­grams with high vec­tor needs, this long-term vi­sion and in­vest­ment in­to process de­vel­op­ment should be seen as an as­set be­ing primed to be­come a suc­cess­ful com­mer­cial prod­uct. In this case, af­ter up­stream op­ti­miza­tions are per­formed, the pro­gram could still ben­e­fit from a plat­form down­stream process. For ex­am­ple, each plat­form process unit step uti­lized for a prod­uct rather than a cus­tom-de­vel­oped step (such as ly­sis op­er­a­tion, clar­i­fi­ca­tion pro­ce­dure, or emp­ty/full sep­a­ra­tion tech­nique) may save an av­er­age of 2-2.5 months.

How a Plat­form Process Ac­cel­er­ates Scale-up

Mov­ing a pro­gram through re­search pro­duc­tion, in­to clin­i­cal scale, and even­tu­al­ly com­mer­cial pro­duc­tion re­quires a man­u­fac­tur­er to en­able plat­form process syn­er­gies among process de­vel­op­ment, an­a­lyt­i­cal de­vel­op­ment, and scale. Forge Bi­o­log­ics has demon­strat­ed scal­a­bil­i­ty from 1L to 1000L with mul­ti­ple AAV serotypes through mul­ti­ple it­er­a­tions of reg­u­la­to­ry in­ter­ac­tion, fa­cil­i­ty de­vel­op­ments, PD and AD eval­u­a­tions and demon­strat­ed cGMP pro­duc­tions. To date, Forge’s pro­pri­etary HEK 293 Ig­ni­tion Cell LineTM has scaled mul­ti­ple nat­ur­al and nov­el AAV serotypes suc­cess­ful­ly.

Two as­sets that re­duce prod­uct-to-prod­uct vari­abil­i­ty in the per­for­mance of the plat­form process are Forge’s Ig­ni­tion Cell LineTM and pEM­BRTM Ade­n­ovirus Helper Plas­mid. Re­duc­ing the vari­abil­i­ty of ma­te­ri­als in­put in­to the sys­tem al­lows a smooth and pre­dictable tran­si­tion from 1L→5L→50L→500L→1000L in sin­gle-use biore­ac­tors.

  • The Ig­ni­tion Cell LineTM is a clon­al­ly de­rived HEK 293 sus­pen­sion cell line that was screened for high AAV pro­duc­tiv­i­ty and pu­ri­ty across nu­mer­ous serotypes and se­lect­ed for pro­duc­ing a high per­cent­age of full AAV cap­sids.
  • pEM­BRTM is a pro­pri­etary Ad Helper plas­mid with re­duced ade­n­ovi­ral el­e­ments that has demon­strat­ed equiv­a­lent AAV pro­duc­tiv­i­ty, and an in­creased safe­ty pro­file com­pared to com­mer­cial­ly avail­able Ad Helpers.

The Sci­ence of Scal­ing

Bi­o­log­ic man­u­fac­tur­ing process­es, specif­i­cal­ly AAV, are di­vid­ed in­to up­stream and down­stream events, each with di­verse needs and goals. Up­stream man­u­fac­tur­ing fo­cus­es on scal­ing biore­ac­tor pa­ra­me­ters to al­low cell growth and ex­pan­sion, re­sult­ing in ef­fi­cient and re­li­able AAV pro­duc­tion, usu­al­ly from a sus­pen­sion cul­ture. In con­trast, down­stream man­u­fac­tur­ing fo­cus­es on pu­rifi­ca­tion and re­cov­ery of a high­ly con­cen­trat­ed, in­fec­tious AAV.

Biore­ac­tor pa­ra­me­ters such as pow­er in­put/vol­ume ra­tio (P / V), pH set­point, tem­per­a­ture set­point, and DO set­point are held con­stant dur­ing up­stream AAV man­u­fac­tur­ing. Sim­i­lar­ly, back­ground air sparge, back­ground air over­lay, O2, and CO2 sparge rates are scaled pro­por­tion­al­ly to the work­ing vol­ume with­in the biore­ac­tor. Biore­ac­tor in­puts such as plas­mid mo­lar ra­tio, vi­able cell den­si­ty at time of trans­fec­tion, trans­fec­tion reagent to DNA ra­tio, DNA to vi­able cell ra­tio, and trans­fec­tion cock­tail work­ing vol­ume can all be pre-de­ter­mined for a plat­form process. These in­puts al­so scale pro­por­tion­al­ly with the work­ing vol­ume. In process­es up to 500L, the nec­es­sary con­sid­er­a­tions for scal­ing biore­ac­tor pa­ra­me­ters are be­ing in­creas­ing­ly demon­strat­ed and un­der­stood through­out the in­dus­try. In process­es from 1000L to 5000L, on­ly a few groups, or in the case of 5000L vol­ume, no groups to our knowl­edge, have yet at­tempt­ed gene ther­a­py man­u­fac­tur­ing at such am­bi­tious scales. At the 1000L+ scale, flu­id dy­nam­ics and the co­or­di­na­tion of mov­ing, mix­ing, and stor­ing large vol­umes of liq­uids be­come para­mount.

Dur­ing down­stream AAV pu­rifi­ca­tion, care­ful as­sess­ment of yield re­cov­ery, pu­ri­ty and prod­uct qual­i­ty are crit­i­cal, as larg­er vol­umes of bio­mass need to be processed. If up­stream pa­ra­me­ters have been held con­sis­tent in a man­ner that pro­duces equiv­a­lent bio­mass per vol­ume, then scal­ing the clar­i­fi­ca­tion and affin­i­ty cap­ture steps can be done by in­creas­ing fil­ter sur­face area or resin vol­ume to match the load­ing ra­tios (L / m2 or vg/mL) pre-de­ter­mined by the plat­form process stud­ies. How­ev­er, as bio­mass in­creas­es, fil­ter size and/or col­umn size will need to in­crease to pre­vent fil­ter or col­umn foul­ing. A plat­form ap­proach to affin­i­ty col­umn chro­matog­ra­phy does not nec­es­sar­i­ly im­pact scal­ing. Still, it af­fects the trans­la­tion of the plat­form to mul­ti­ple AAV serotypes, by us­ing uni­ver­sal buffer ma­tri­ces com­pat­i­ble with var­i­ous col­umn types. Emp­ty/full cap­sid sep­a­ra­tion by CsCl den­si­ty gra­di­ent ul­tra­cen­trifu­ga­tion is an ef­fi­cient step for process­es up to 500L, re­quir­ing a sin­gle day for pro­cess­ing. How­ev­er, for pro­cess­ing 1000L+, there is val­ue, or ne­ces­si­ty, in re­duc­ing the vol­ume of ma­te­r­i­al af­ter affin­i­ty elu­tion to fit in­to 1-2 days of pro­cess­ing via an ad­di­tion­al TFF con­cen­tra­tion step. The fi­nal con­cen­tra­tion UF/DF step is again scaled lin­ear­ly by ad­just­ing the fil­ter sur­face area to the vol­ume of ma­te­r­i­al.


AAV Plat­form Process Scale Up. AAV man­u­fac­tur­ing pro­duc­tions were per­formed in a 1L shake flask, or biore­ac­tors with 5L to 500L work­ing vol­ume. Vi­able cell den­si­ty (A.) and vi­a­bil­i­ty (B.) were mea­sured on days one through four post-trans­fec­tion with the Vi-Cell BLU cell counter. At day four post-trans­fec­tion, cul­tures were har­vest­ed and pu­ri­fied through Forge’s plat­form down­stream process and ddPCR titers of AAV in the clar­i­fied lysate ma­te­r­i­al were ob­tained (C.). Ex­trap­o­lat­ed val­ues rep­re­sent­ing clar­i­fied lysate titers for 1000L and 5000L pro­duc­tion vol­umes (D.) were gen­er­at­ed based on 1L to 500L vol­ume AAV pro­duc­tion trends.

Reach­ing the End Goal

Most gene ther­a­py com­pa­nies aim to pro­vide pa­tients with safe and ef­fec­tive treat­ment op­tions as fast and cost-ef­fec­tive­ly as pos­si­ble. Un­for­tu­nate­ly, in the fast-paced biotech­nol­o­gy in­dus­try, the mis­sion can of­ten be­come buried in the day-to-day ex­cite­ment of hit­ting a mile­stone or re­ceiv­ing a promis­ing proof-of-con­cept re­sult. While each step of the pre-clin­i­cal path­way is es­sen­tial, a re­minder of the end goal is al­ways war­rant­ed. The ac­cel­er­at­ed time­line that ac­com­pa­nies uti­liza­tion of an ex­ist­ing plat­form man­u­fac­tur­ing ap­proach, and an ex­pe­ri­enced CD­MO, can be the dif­fer­ence that leads to time­ly treat­ment for many more pa­tients with a long-term plan for man­u­fac­tur­ing suc­cess.


Ref­er­ences:
1Mendell JR, Al-Zaidy SA, Rodi­no-Kla­pac LR, et al. Cur­rent clin­i­cal ap­pli­ca­tions of in vi­vo gene ther­a­py with AAVs. Mol. Ther. 2021; 29: 464–488.


Con­tribut­ing Au­thors:
Danielle Sex­ton, Sci­en­tist II, Process De­vel­op­ment, Forge Bi­o­log­ics

Author

Brianna Barrett

Ph.D., Associate Director, Technical Sales, Forge Biologics