Surrogate Endpoints Acceptable in AML Trials, Says FDA

Pam Harrison

January 31, 2022

The US Food and Drug Administration (FDA) has been harshly critized for using surrogate endpoints in clinical trials in its approval of new drugs, and especially so in oncology, where critics have argued that the only truly meaningful endpoint is overall survival (OS).

But the FDA is now fighting back and is arguing that in certain cases, surrogate endpoints do translate to overall survival benefits. A case in point is in clinical trials of new treatments being investigated in patients newly diagnosed with acute myeloid leukemia (AML).

FDA investigators led by led by Kelly Norsworthy, MD, conducted an analysis of eight randomized clinical trials of intensive chemotherapy for the treatment of newly diagnosed AML and found that both complete remission (CR) and event-free survival (EFS) were strongly correlated with OS in all the trials studied.

The results show that these particular surrogate endpoints do have real value in predicting the efficacy of drugs for the treatment of newly diagnosed AML, they conclude.

"To our knowledge, our results represent the first direct examination of the relationship of response rate and EFS to OS using trial-level and patient-level data in patients with newly diagnosed AML treated with intensive induction chemotherapy," the FDA investigators comment.

"The results support the FDA's acceptance of EFS as a clinical benefit end point supportive of traditional approval for treatments with curative intent," they add.

The analysis was published online on December 10, 2021, in the Journal of Clinical Oncology.

"The central finding of the article, that both EFS and the CR rate are reliably associated with improved OS, is of immediate relevance and indicates that these parameters represent acceptable and appropriate surrogate end-points for clinical trials of AML," say Courtney DiNardo, MD, University of Texas MD Anderson Cancer Center, Houston, Texas, and Daniel Pollyea MD, University of Colorado, Aurora, Colorado, in an accompanying editorial.

"The establishment of CR and EFS as appropriate surrogate end points for patients with newly diagnosed AML receiving intensive chemotherapy will allow earlier evaluation of novel therapies and speed the delivery of safe and effective therapies to our patients," they add.

Analysis of Clinical Trials Submitted for Approval

The FDA investigators conducted an analysis of eight trials that had been submitted to the agency for licensing approval between 2007 and 2011. Together, the trials included a total of 4482 patients with newly diagnosed AML.

Five trials evaluated gemtuzumab ozogamicin (Mylotarg), dubbed GO, while two trials evaluated a daunorubicin and cytarabine liposome injection, also known as CPX-351, (Vyxeos), and one trial evaluated midostaurin (RYDAPT).

"All were approved in combination with, or for use as, intensive induction chemotherapy in patients with newly diagnosed AML," the team points out. Both trial-level and patient-level associations between responses, EFS, and OS were evaluated.

The association between the hazard ratio (HR) for OS and the odds ratio (OR) for CR at the trial level was "moderate." The association between the HR for OS and the OR for lesser response rates — namely, CR with incomplete hematologic recovery (CRi) and CR with incomplete platelet recovery (Crp) — was similarly moderate, as investigators note.

On the other hand, while the associations between the HR for OS and the HR for EFS were again moderate, "on the basis of the harmonized primary definition of EFS across trials, the association became stronger," the authors report. The harmonized definition of EFS across the trials included time from random assignment to treatment failure, relapse from CR, or death from any cause, whichever occurred earlier.

The FDA authors caution that a significant number of patients who relapsed did not die during the course of the clinical trial, resulting in a considerably longer OS compared with EFS in some patients. However, to further explore the relationship between response and survival, a patient-level analysis of response — namely, CR vs CRi or CRp vs no response — was performed."Patients who achieved a CR had a [27%] better OS compared with patients whose best response was CRi or CRp," the authors note.

Patients who achieved a CRi or a CRp as their best response still had a 54% better OS compared with patients who achieved no response, irrespective of the treatment received, they add.

Interestingly enough, a small number of patients who had no response to treatment also experienced prolonged survival, possibly because of successful second-line therapy, the authors speculate.

Effective Salvage Therapies

Commenting further in their editorial, DiNardo and Pollyea point out that, given that there are now multiple effective salvage therapies for the treatment of AML, an OS endpoint no longer solely reflects the effectiveness of an initial therapy, as survival will also be affected by subsequent lines of AML-directed treatment.

"Accordingly, OS should no longer be considered the sole determinant of the value of a new therapy," the editorialists emphasize.

Furthermore, as the treatment of AML is increasingly based on biologically defined and differentially targeted subsets, "the required sample sizes and timelines to run a proper randomized phase 3 study for an OS end point of a rare AML subset become logistically untenable," they write.

That said, the editorialists felt the fact that FDA employees performed this meta-analysis at all was "highly laudable." "It is, moreover, gratifying to know that the experiences of clinical trial participants can be maximized beyond the original contributions made to the studies in which they originally volunteered," the editorialists observe.

"In the United States, this example should inspire investigators and industry partners to prioritize similar analyses with their [own] data sets," they add.

Norsworthy, DiNardo, and Pollyea have disclosed no relevant financial relationships.

J Clin Oncol. Published online December 10, 2021. Abstract, Editorial

For more news, follow Medscape on Facebook, Twitter, Instagram, and YouTube.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.