On­col­o­gy drug de­vel­op­ment can of­ten feel more like a wait­ing game than a race to­wards break­through in­no­va­tion. Imag­ing can take months to de­fin­i­tive­ly re­veal whether a tu­mor is re­spond­ing to treat­ment, while end­points such as pro­gres­sion-free and over­all sur­vival can take years to ma­ture.

In­creas­ing­ly, drug de­vel­op­ers are look­ing to cir­cu­lat­ing tu­mor DNA (ctD­NA) as a bio­mark­er for ear­li­er in­di­ca­tion of ther­a­py re­sponse and ef­fi­ca­cy. Since ctD­NA has a half-life of ap­prox­i­mate­ly one hour, it can pro­vide drug de­vel­op­ers with a re­al-time, quan­ti­ta­tive mea­sure of dis­ease bur­den.¹ ctD­NA clear­ance, de­fined by the shift from de­tectable to un­de­tectable lev­els of ctD­NA, can of­fer an ear­ly read on whether a ther­a­py is work­ing.

ctD­NA re­duc­tion and clear­ance has been as­so­ci­at­ed with im­proved sur­vival across 100+ pub­li­ca­tions span­ning tu­mor types and treat­ment modal­i­ties.²

Fig­ure 1. Se­r­i­al ctD­NA test­ing over a pa­tien­t's treat­ment jour­ney, where ctD­NA clear­ance pro­vides an ear­ly in­di­ca­tor of treat­ment re­sponse.

In the IN­SPIRE prospec­tive Phase II study of pem­brolizum­ab across ad­vanced sol­id tu­mors, changes in ctD­NA af­ter two treat­ment cy­cles of the im­mune check­point in­hibitor (ICI) ther­a­py were as­so­ci­at­ed with over­all sur­vival (OS).³

Fig­ure 2. IN­SPIRE study: ctD­NA dy­nam­ics pre­dict­ed ben­e­fit from ICI af­ter just two cy­cles of treat­ment in a pan-can­cer study ³

A re­cent re­al-world study in metasta­t­ic breast can­cer found that ear­ly on-treat­ment ctD­NA dy­nam­ics, as­sessed with­in the first six weeks of ther­a­py, cor­re­lat­ed with time to next treat­ment (TTNT) across sub­types and reg­i­mens, with ctD­NA clear­ance yield­ing the longest in­ter­vals.⁴

As the body of ev­i­dence has grown rapid­ly, drug de­vel­op­ers are in­creas­ing­ly in­cor­po­rat­ing ctD­NA mon­i­tor­ing in­to ac­tive pro­grams. In re­frac­to­ry melanoma, Di­akonos On­col­o­gy is prospec­tive­ly track­ing ctD­NA dy­nam­ics in a Phase 1/2 tri­al eval­u­at­ing a Fast Track-des­ig­nat­ed in­ves­ti­ga­tion­al im­munother­a­py, where con­ven­tion­al imag­ing has well-doc­u­ment­ed chal­lenges in as­sess­ing re­sponse.⁵ An­oth­er ex­am­ple is Ex­elix­is, which has in­cor­po­rat­ed Sig­nat­era in­to STEL­LAR-316, its Phase 3 tri­al of zan­za­l­in­tinib for pa­tients with re­sect­ed stage II/III col­orec­tal can­cer us­ing lon­gi­tu­di­nal ctD­NA clear­ance mon­i­tor­ing as a key sec­ondary end­point.⁶

In the cu­ra­tive-in­tent set­ting, where time to re­lapse and fol­low-up pe­ri­ods can stretch for years, the po­ten­tial is even greater. As the field evolves, mea­sur­ing the clear­ance of mol­e­c­u­lar resid­ual dis­ease (MRD) via ctD­NA is mov­ing from an ex­plorato­ry sig­nal to­ward a reg­u­la­to­ry-grade end­point backed by ac­cu­mu­lat­ing ev­i­dence across sol­id tu­mors and hema­to­log­ic ma­lig­nan­cies.

Mul­ti­ple myelo­ma set an im­por­tant prece­dent in 2024, when the FDA's On­co­log­ic Drugs Ad­vi­so­ry Com­mit­tee vot­ed unan­i­mous­ly to sup­port the use of min­i­mal resid­ual dis­ease as an ac­cel­er­at­ed end­point.⁷ Fol­low­ing draft guid­ance is­sued in ear­ly 2026, iber­do­mide (Zen­bexus) be­came the first drug ap­proved us­ing an MRD-neg­a­tive com­plete re­sponse as its pri­ma­ry ef­fi­ca­cy end­point, mark­ing a ma­jor reg­u­la­to­ry mile­stone.^8,9

Could large B-cell lym­phoma be next? In Al­lo­gene Ther­a­peu­tics’ piv­otal Phase 2 AL­PHA3 tri­al eval­u­at­ing an in­ves­ti­ga­tion­al al­lo­gene­ic CAR T prod­uct as a 1L con­sol­i­da­tion treat­ment for LB­CL pa­tients, MRD clear­ance da­ta at Day 45 was com­pelling enough to sup­port both RMAT and Fast Track des­ig­na­tions for cemacab­ta­gene ansegedleu­cel (ce­ma-cel) from the FDA.^10,11

Fig­ure 3. AL­PHA3 in­ter­im fu­til­i­ty analy­sis: MRD clear­ance rate of 58.3% by Day 45. MRD analy­sis per­formed us­ing Nat­er­a's Clar­i­ty™ uti­liz­ing Phased Vari­ant Tech­nol­o­gy.¹⁰

As that ev­i­dence builds and clear­ance takes on greater weight in de­vel­op­ment de­ci­sions, it rais­es the ques­tion of: what does "un­de­tectable" ac­tu­al­ly rep­re­sent?

At very low tu­mor bur­den, an as­say can miss mol­e­c­u­lar-lev­el dis­ease that re­mains present — when ctD­NA vol­umes fall be­low the as­say's an­a­lyt­i­cal lim­it of de­tec­tion. This rais­es the ques­tion of what a ctD­NA-neg­a­tive re­sult ac­tu­al­ly re­flects: true mol­e­c­u­lar dis­ease clear­ance, or a sig­nal that has fall­en be­low the as­say's lim­it of de­tec­tion (LOD).

If clear­ance is go­ing to in­form de­vel­op­ment de­ci­sions or even­tu­al­ly serve as a sur­ro­gate end­point, de­vel­op­ers need con­fi­dence that a neg­a­tive re­sult rep­re­sents clear­ance of dis­ease, rather than the lim­its of the test used to mea­sure it.

Cer­tain set­tings place greater de­mands on that con­fi­dence than oth­ers. In ear­ly-stage dis­ease, af­ter de­fin­i­tive treat­ment, and in low-shed­ding tu­mor types such as breast and re­nal cell car­ci­no­ma (RCC), ctD­NA con­cen­tra­tions can be par­tic­u­lar­ly low and the sen­si­tiv­i­ty re­quire­ments of the as­say need to be ad­e­quate for the use case.^12,13

Back­ground se­quenc­ing er­ror in con­ven­tion­al as­says cre­ates noise that may be dif­fi­cult to dis­tin­guish from true low-lev­el ctD­NA sig­nal, and that noise sets the floor be­low where most as­says can re­li­ably de­tect.

Phased Vari­ant Tech­nol­o­gy ad­dress­es this noise is­sue by re­quir­ing two mu­ta­tions to co-oc­cur on the same DNA mol­e­cule be­fore a vari­ant is called, a con­fig­u­ra­tion that se­quenc­ing er­ror es­sen­tial­ly can­not pro­duce.¹⁵ That drops the er­ror rate from ap­prox­i­mate­ly one in 10,000 for sin­gle nu­cleotide vari­ants to one in 100 mil­lion for phased vari­ants, as demon­strat­ed in a Na­ture Biotech­nol­o­gy study en­abling de­tec­tion at con­cen­tra­tions that fall well be­low the sen­si­tiv­i­ty thresh­old of con­ven­tion­al as­says and turn­ing an un­cer­tain re­sult in­to a con­fi­dent one.^14,15

Fig­ure 4. Phased Vari­ant Tech­nol­o­gy pro­vides con­fi­dence across tracked vari­ants, en­abling mar­ket-lead­ing de­tec­tion.^14,15,16

Mon­i­tor­ing ctD­NA dy­nam­ics can of­fer drug de­vel­op­ers valu­able in­sight in­to whether a ther­a­py is work­ing, with­out wait­ing years for sur­vival end­points to ma­ture.

In ear­ly phase I/II de­vel­op­ment, ctD­NA re­sponse can sup­port ear­li­er dose se­lec­tion and go-for­ward de­ci­sions. In the cu­ra­tive in­tent set­ting, ground­work is be­ing laid to­ward MRD clear­ance as an end­point across mul­ti­ple tu­mor types and set­tings, dri­ven by part­ner­ships be­tween aca­d­e­m­ic, phar­ma, and di­ag­nos­tics.

Gen­er­at­ing ro­bust clin­i­cal ev­i­dence can ac­cel­er­ate the path to reg­u­la­to­ry ap­proval, bring­ing promis­ing treat­ments to reach the pa­tients who need them soon­er.

Learn more about how ctD­NA re­sponse mon­i­tor­ing can ben­e­fit de­vel­op­ment

Ref­er­ences: