We need faster end­points for tar­get­ed can­cer drugs – here’s one we can use

I spent more than a decade re­view­ing mar­ket­ing sub­mis­sions at the EMA, and I think we need faster end­points to mea­sure how tar­get­ed can­cer drugs can ben­e­fit pa­tients. We can cre­ate one by com­bin­ing two com­mon­ly used mea­sures.

Ob­jec­tive re­sponse rate (ORR) is the pro­por­tion of pa­tients whose tu­mors shrink by a pre­de­fined amount for a min­i­mum pe­ri­od. The du­ra­tion of re­sponse (DOR) is the time from treat­ment re­sponse to pro­gres­sion or death. The com­bi­na­tion of these two end­points—some­times called “durable ORR”—can es­tab­lish the du­ra­tion of pa­tients’ re­sponse to an ex­per­i­men­tal drug.

In the right cir­cum­stances, durable ORR could sup­port stan­dard ap­proval just as well as pro­gres­sion-free sur­vival (PFS), which the FDA and EMA cur­rent­ly re­ly on. Durable ORR can demon­strate an agent’s clin­i­cal ac­tiv­i­ty even where there is not enough PFS da­ta, giv­ing on­col­o­gists ad­di­tion­al treat­ment choic­es and im­prov­ing pa­tients’ lives.

What’s wrong with cur­rent­ly used end­points?

The ide­al out­come for can­cer pa­tients is to live longer with a rea­son­able qual­i­ty of life (QoL). That’s why the gold stan­dard clin­i­cal end­point for can­cer tri­als is over­all sur­vival (OS), the time from the start of treat­ment to death. But prov­ing a clear OS ben­e­fit re­quires:

• En­rolling a large num­ber of pa­tients (dif­fi­cult for or­phan and rare can­cers)
• Fol­low­ing pa­tients for years (in the case of some slow-grow­ing tu­mors)
• Re­strict­ing pa­tients to a sin­gle treat­ment dur­ing fol­low-up (which is un­eth­i­cal), and
• Con­duct­ing a ran­dom­ized, con­trolled tri­al in which some pa­tients would get a place­bo or a ther­a­py with lit­tle im­pact on OS

So since the 1970s, reg­u­la­tors have ac­cept­ed PFS and ORR as prox­ies for OS.

For stan­dard ap­provals, reg­u­la­tors use PFS be­cause it al­lows small­er, short­er stud­ies with less fol­low-up. But it is flawed. For ex­am­ple, it does not cor­re­late well with OS, es­pe­cial­ly for tar­get­ed can­cer drugs, and it is not mean­ing­ful in non-ran­dom­ized tri­als. PFS can­not ac­count for the fact that a slow in­crease in tu­mor size and growth does not ag­gra­vate symp­toms in some tu­mor types. And for first-line treat­ment of some can­cers, PFS re­quires an ex­tend­ed fol­low-up.

For ac­cel­er­at­ed ap­provals (AAs), reg­u­la­tors use ORR on the ra­tio­nale that it’s an in­ter­me­di­ate end­point that could rea­son­ably pre­dict clin­i­cal ben­e­fit.

Why durable ORR is a use­ful end­point

1. It’s clin­i­cal­ly rel­e­vant for pa­tients. A high re­sponse rate that does not last long is mean­ing­less to pa­tients and may be off­set by tox­i­c­i­ties. But if a pa­tient ex­pe­ri­ences a com­plete or par­tial tu­mor re­sponse that lasts, it’s com­mon sense that their QoL will be bet­ter for the same rea­sons that sup­port PFS use. The pa­tient is alive, with­out pro­gres­sion of the dis­ease, and prob­a­bly with a re­duc­tion in symp­toms.
2. It could work in sup­port of reg­u­lar ap­proval. Durable ORR could be used to es­tab­lish ef­fi­ca­cy un­der the fol­low­ing con­di­tions: 1) There is no ex­ist­ing drug that sig­nif­i­cant­ly im­proves PFS/OS; 2) The ex­per­i­men­tal agent pro­duces a dra­mat­ic and durable ORR com­pared with es­tab­lished treat­ments, and; 3) There is a high un­met med­ical need.
3. There is reg­u­la­to­ry prece­dent for us­ing it. The FDA and EMA have grant­ed full ap­proval to sev­er­al drugs that used ORR as the pri­ma­ry end­point with DOR as a key sec­ondary end­point to show ef­fi­ca­cy, in­clud­ing two pre­ci­sion med­i­cines for ROS1-pos­i­tive metasta­t­ic non-small cell lung can­cer; Xalko­ri (crizo­tinib) in 2016, and Ro­z­lytrek (en­trec­tinib) in 2019. The FDA and EMA al­so ap­proved two chimeric anti­gen re­cep­tor T-cell (CAR T-cell) ther­a­pies in 2017 and 2018 based on ORR and re­mis­sion rates, an end­point that close­ly aligns to durable ORR. Kym­ri­ah (ti­s­agen­le­cleu­cel) was ap­proved for acute lym­phoblas­tic leukemia and Yescar­ta (axi­cab­ta­gene ciloleu­cel) for dif­fuse large B-cell lym­phoma.

Durable ORR could en­able physi­cians to bring bet­ter care to more can­cer pa­tients more quick­ly in the right cir­cum­stances.

For more on­col­o­gy in­sights, down­load our eBook “The new im­per­a­tives: adapt­ing on­col­o­gy drug de­vel­op­ment to a post-COVID world.”