Key Considerations for the FDA’s Guidance on MRD End Point in Myeloma

Minimal residual disease (MRD) has been established not only as a tool for evaluating patients but as a goal in drug development. As therapies have become increasingly effective, both traditional long-term end points and surrogates such as overall response rate (ORR) have reached a ceiling, limiting their ability to distinguish meaningful benefit within feasible study designs. In response, the FDA recently issued draft guidance outlining how MRD and complete response (CR) may be used as end points to support accelerated approval in multiple myeloma, a milestone that reflects decades of clinical research and regulatory dialogue.¹

Few investigators have been as closely involved in shaping the scientific foundation for this shift as C. Ola Landgren, MD, PhD, of the University of Miami Sylvester Comprehensive Cancer Center. From early MRD studies conducted at the National Institutes of Health (NIH) to large, pooled analyses presented to the FDA’s Oncology Drug Advisory Committee (ODAC) meetings, Landgren has played a leading role in demonstrating the prognostic value of MRD negativity and its correlation with long-term outcomes such as progression-free survival (PFS). In April 2024, ODAC unanimously endorsed MRD as an acceptable end point for accelerated approval in multiple myeloma, paving the way for the draft guidance released in early 2026.

In this interview, Landgren discussed the significance of the FDA’s draft guidance, what it gets right, and where further clarification is needed. He reflected on the long path toward regulatory acceptance of MRD, the implications for clinical trial design and drug development, and ongoing debates around assessment timing, sustained MRD, and the inclusion of smoldering multiple myeloma.

Targeted Oncology: What is the significance of the FDA’s draft guidance for using MRD as an approval end point?

C. Ola Landgren, MD, PhD: We have worked on MRD as an early end point for accelerated approval in myeloma for a long time. The very first study started at the NIH in the intramural program back in 2009, and I was a lead investigator. Over time, a few years later, the [International Independent Team for Endpoint Approval of Myeloma Minimal Residual Disease (i2TEAMM) Group] study started, and eventually, those 2 studies were invited to ODAC and in 2024 in April, the 2 groups were invited to present. ODAC voted 12 to 0 in favor, and it took about 1.5 to 2 years for the FDA to come out with a guidance document.² I think this is how long it usually takes, at least historically, for the government to go through all the different steps in the process. I think that the document is well written overall. I think it's very reflective of all the data that we have generated for all these years.

We have worked on it for almost 20 years now. It was a long process. We had to gather all the datasets, build a solid statistical analysis plan, test our hypothesis, and deliver that. There are always more questions left, and every time you answer questions, there will be new questions. I think this guidance document summarizes the core of what we have been trying to do for a very long time. It helps to create that framework that will allow new drugs to be developed for patients with multiple myeloma.By now, over 200,000 [patients] are living with myeloma in America right now and worldwide, there are hundreds of thousands of [patients]. There is still no established cure for the disease.

In order for drugs to be developed with a curative intent, and also for newly diagnosed patients, where you have the best long-term outcome, you need a surrogate end point such as MRD for early accelerated approval, and this guidance document helps companies or anyone interested in developing clinical trials for regulatory purposes—academics also. It brings clarity, and it's based on data. It tells you how to design those studies, how you shall apply a reproducible, validated assay, and what type of time window you need to have. It gives you information on the sensitivity level, and you also need to show correlation to clinical outcome—in this case, PFS. If you have overall survival [OS], that's also good, but it's not necessary. Both PFS and OS will be important [confirmatory] end points.

The document also talks about the fact that you can do single-arm or randomized trials, [which] I think is great.¹ The FDA per standard has invited experts [or] anyone to provide feedback in about a 2-month window, and that's available online to submit those comments.

What remaining needs should be clarified?

I think the document needs to clearly also include smoldering myeloma. In November of 2025, the first drug for the indication of smoldering myeloma was approved by the FDA, and this document only talks about multiple myeloma.³ We can debate back and forth whether there really is a smoldering myeloma or it's just all myeloma, just earlier detection. But the way the textbook is currently written, it is labeled as a separate entity.

I also think that the window when the capture of MRD should be done follows the ODAC meeting. We were given the task of using 12 months plus/minus 3 months. The [i2TEAMM] group was given 9 months plus/minus 3 months, and looking at the more powerful immunotherapies, CAR T cells, and other combination therapies with bispecific or trispecifics in the frontline setting, you can see much quicker MRD negativity if you look at what's going on in our clinics.

Those studies are not yet fully compiled, and they're still not part of this analysis that was referring to, so the data [are] not there to give exact guidance. The way the guidance document is currently written. It says, “for example, 9 or 12 months from randomization or from treatment start.” But because it uses, “for example”, you could imply that if you have a shorter window [such as] 6 months, that would be good as well. I think the guidance document probably should also state that shorter or longer windows may be applicable in in given situations with more powerful or less powerful therapies, or every study will be interpreted individually…. But these are just small nuances. Overall, the document is very well written, and I applaud the FDA's work putting it together. I think it sets the framework for other cancers. Myeloma is now leading the field for early drug development.

Could you elaborate on the current data for deeper responses in smoldering myeloma?

The only trial we have currently is the AQUILA trial [NCT03301220]. That's a randomized study, but there are already ongoing studies with combination therapies. We have a study called the REVIVE study [NCT06100237], for example, that uses daratumumab [Darzalex] in combination with a bispecific antibody. If you look at the AQUILA trial, the complete response rate was about 8% for the treated patients and 0% for the control arm, and yet the PFS was in favor of the treatment.⁴ Even if you didn't have a CR, you had some degree of response…that was holding the disease back from progression.

But in multiple myeloma, both newly diagnosed and relapsed, it's evident that MRD negativity is a very strong predictor of long-term PFS. All my clinical experience tells me the same applies in all the settings, including smoldering myeloma. To disqualify smoldering myeloma from that would mean that there could not be any drugs approved there. That would not be fair to those patients, and it will completely stop the field for smoldering myeloma. So, [the guidance] has to include smoldering myeloma as well. There is no reason to believe that the relationship between no detectable disease and long PFS would be excluded in that setting. That would be not logical.

Why is the single-arm trial approach important for getting both accelerated and regular approval?

If you look at how the accelerated approval pathway has led the field to develop drugs in the relapsed space, historically, you did a single-arm study with an overall response [end point], and if you reached your goal, if you set it at 30% or 60%...then you got accelerated approval, and then you have to do a randomized trial and prove PFS.⁵ You could do that ORR for patients with 4 or more prior lines, but the PFS you could do for earlier lines. For example, look at how the registrational CAR T cell study strategies were done. First, they did single-arm studies showing high ORR in later lines, and then they moved to patients with 1 to 3 prior lines and showed better PFS compared to the control arm. You can see the same strategy with all the newer antibodies and other newer therapies. This’s how the field evolves.

What the guidance document says is that you can still use a single-arm study followed by a randomized trial along the same lines, or you could do that randomized trial right out of the gate and capture MRD at half a year or 1 year, or 9 months from randomization, and then you get your accelerated approval, and then the study just keeps on going. When your PFS data come back, then you get your full approval.

In the relapsed setting, it is not as easy, because what's the appropriate control therapy? The field is moving so much. Even in the newly diagnosed setting, you have these issues [about] what's the optimal control group. It's mostly true in the in the relapsed/refractory [setting], but the field is changing very fast. Companies need to have a lot of conversations with the agency when they do these things to make sure all is clear before starting the studies.

What is notable about the draft guidance on using CR as a study end point?

The problem with ORR is that these days...in newly diagnosed patients, probably 95% of patients have an ORR. If you have your experimental arm vs the control arm, if the control arm is 95%...you have only 5% to improve upon. The sample size has to be gigantic in order to have a sufficient number of patients going from 95% to 100%, so it becomes an impossible task to do a study in that space. If you go to relapsed/refractory, it's very similar. You could say, let's not do ORR, let’s say CR. You may have 80% of patients reaching a CR. Between 80% and 100%, you still have a very narrow window to improve upon. [If] you go from 80% to 90%, with that 10% improvement...again, the size of the study becomes so large, it's just not a feasible end point to use.

That is the reason why 20 years ago, we started working on MRD as an early end point for accelerated approval. We realized this is going to happen, and now it happened. At the time when we did it, we had let’s say 30% of our patients being MRD negative, and then it became 50%. It is logical to envision a scenario where MRD is also going to reach the same dilemma as ORR did years ago. When that happens, MRD will not be an accelerated approval end point. At that time, MRD will have to be the full approval end point, otherwise you're never going to be able to move the field forward.

I think the reason they put [CR] there is probably because they are thinking that that could work for the smoldering [myeloma] space. But I disagree with that, because if you use daratumumab, you have only 8% CR, but if you do a bispecific antibody, it is 95% or 99%, so you have to start over again.

In the smoldering space, CR a very short-lived end point. …I would be more pragmatic and pick MRD across the board.

Why did the FDA not support using sustained MRD negativity as an end point for approval?

I think that people are continuing to get confused—is MRD good enough? Is sustained [MRD] better? Obviously, if you're trying to develop a curative treatment, if you show continued, sustained MRD negativity, that will equally become the path toward a cure. Everyone wants that. But that doesn't change the fact that a single data point with MRD negativity is predictive of longer PFS. If you want to develop a drug, you [had to] wait for the clinical end point from randomization to 15 years out. You can shrink that to a year or less. That saves you many years, and patients can get access to the drugs much sooner. For drug development, if you have an end point that can predict [PFS], which MRD does, we have…already shown it. Sustained [MRD] is for developing cures. It's a way to tailor therapy. You can decrease or increase your treatment if you sustain or lose your MRD. It's very important, but for the development of drugs, you don't need that.

REFERENCES

1. Minimal residual disease and complete response in multiple myeloma: use as endpoints to support accelerated approval: guidance for industry. US FDA. January 20, 2026. Accessed February 10, 2026. https://tinyurl.com/4m5v6zwv

2. Oncologic Drugs Advisory Committee (ODAC) Meeting. US FDA. April 12, 2024. Accessed February 10, 2026. https://tinyurl.com/2rexatkp

3. FDA approves daratumumab and hyaluronidase-fihj for high-risk smoldering multiple myeloma. News release. US FDA. November 6, 2025. Accessed February 10, 2026. https://tinyurl.com/2vb674jn

4. Dimopoulos MA, Voorhees PM, Schjesvold F, et al. Phase 3 Randomized study of daratumumab monotherapy versus active monitoring in patients with high-risk smoldering multiple myeloma: primary results of the Aquila study. Blood. 2024;144(supplement 1):773. doi:10.1182/blood-2024-201057

5. Landgren O, Devlin SM. Minimal residual disease as an early endpoint for accelerated drug approval in myeloma: a roadmap. Blood Cancer Discov. 2025 Jan 8;6(1):13-22. doi: 10.1158/2643-3230.BCD-24-0292. PMID: 39630969