Landgren on MRD as an End Point for Multiple Myeloma Trials

C. Ola Landgren, MD, PhD

On April 12, 2024, the FDA’s Oncologic Drug Advisory Committee (ODAC) voted unanimously that minimal residual disease (MRD) could serve as an end point in clinical trials in multiple myeloma for accelerated approval, setting a precedent for the field to potentially speed up drug development and offer treatments to patients sooner.

C. Ola Landgren, MD, PhD, director of the Sylvester Myeloma Institute, Sylvester Comprehensive Cancer Center at the University of Miami, presented data from the EVIDENCE meta-analysis that supported the use of MRD as a surrogate for overall survival (OS) and progression-free survival (PFS). Landgren started the MRD initiative in myeloma in 2009 as an interagency collaboration with the FDA, National Cancer Institute (NCI), and National Heart, Lung, and Blood Institute.

MRD and PFS and MRD and OS were strongly associated at an individual level, according to a Copula global odds ratio with a 95% confidence interval. Results from the meta-analysis supported the consideration of MRD as an early clinical end point reasonably likely to predict clinical benefit in multiple myeloma that could be used to support accelerated approval.

In an interview with Targeted Oncology^TM, Landgren talked more about the use of MRD as a clinical end point in myeloma, what implications the ODAC vote holds, and the future of research in this space.

Targeted Oncology: Could you provide some background on your work in this area and what you presented at the ODAC meeting?

Landgren: This project has been going on for a long time. I was presenting the EVIDENCE meta-analysis, [which] was started in 2009. It started at the National Cancer Institute, and I was the lead investigator. We worked for all these years to develop the statistical analysis plan. I think it is important that people understand this is not something that was cooked up last week or so. We worked for many years to develop the analysis plan to get access to all the data sets from all the companies around the world.

I also launched in 2014 [the Miami Myeloma MRD Meeting] where all the key leaders in the field interested in MRD, the patient organizations, the International Myeloma Foundation, Multiple Myeloma Research Foundation, and the HealthTree Foundation for Multiple Myeloma, as well as the FDA, have participated. The FDA has been there every year. On May 9, we will have the 11th annual meeting, where I hope that the FDA will tell us that they now have implemented the ODAC decision.

How much sooner can MRD be evaluated compared with the years-long wait times we see with PFS and OS?

The current drug approval pathway for multiple myeloma includes the regular approval and the accelerated approval pathway in the relapsed/refractory space. For multiple myeloma, overall response rate is the end point that has been used for the accelerated approval pathway. As we showed at the ODAC meeting, the number of patients that can achieve an overall response in a newly diagnosed setting is close to 98%. That is also true in the relapsed/refractory setting that the rates are high. They are over 90% In the modern era. So that makes it hard to design studies, because you have to beat that 90-plus percent. We have only 10% or less to improve upon.

First, you need to do a very large study to have sufficient numbers to have sufficient statistical power. Then, you will also have to have sufficient follow-up time to be able to benchmark toward the standard of care in a control arm if you do a randomized study. If you do a newly diagnosed study, enrolling all those patients that you need for statistical power will take anywhere from 2 to up to 4 or 5 years. Then you have to wait for the data to mature to be the benchmark for the control arm. We have done extensive modeling, and we have showed in our models that that could take 8 to 10 or more years [for] the totality of identifying, enrolling patients, and then you start the clock for the data to mature, for the last patient [that] needs to reach that point. That is somewhere between 10 and 15 years.

Now, if you look at MRD as an end point for accelerated approval, we worked closely with the FDA to develop a statistical analysis plan that has been approved by the FDA as part of our [investigational new drug (IND)] application. It is part of the IND statistical analysis plan. The FDA guidance we received was to check for MRD 1 year after randomization. What that means is in the setting of what I [mentioned], if it takes 2 years to enroll all the patients and the last patient has been enrolled, that's 2 years later, the first patient is day 1, but the last patient is 2 years later. Then you add 1 more year to check for MRD. That means that the first patient was enrolled 2 years ago, and 1 year later, you did the MRD, so that patient will have 3 years of follow-up. [For] the last patient that was enrolled, [we] have to still wait for 1 year. So, it is 3 years out [that] you will have the data in this setting, so you shrink 10 to 15 years down to 3 years.

Let's say the enrollment takes 4 or 5 years, then you could add 1 more year to capture MRD. The reason I [mention] these details is because you speed up the approval process for accelerated approval using MRD as an early end point, but what you don't do [is] lower the bar for safety because the patient would have to first be enrolled, which takes quite a time, several years. You will have a lot of patients that have been monitored and that then go in with the data for both MRD [and] progression-free and overall survival to the FDA as a totality, and they will review everything, then the data will be reviewed.

If MRD now becomes approved by the FDA—this was an ODAC vote—that means that a company in that setting could be granted accelerated approval, but they still have to capture progression-free survival with a longer window, and they have, for that reason, to statistically power the study to be able to answer that question. [One] cannot only enroll 100 patients, 50 on each arm, and capture MRD and hope for the best. That is not going to work. You need to have appropriate power to capture PFS.

In the relapsed setting, instead of [it taking] 10 to 15 years, I think we probably are somewhere between 7 to 10 years these days for that same exercise, and you can probably shrink that down to, say, 3 years. We are talking for both [the] newly diagnosed and relapsed [settings], thinking 10 to 15 years down to 3 to 5 years. Basically, you triple the speed of drug development, and you give patients access to drugs much faster, which is what we need.

Do you see the potential for similar decisions regarding guidelines for accelerated approval to transfer to other cancer types?

The brief answer is yes. The longer answer is the FDA has taken a lot of action to help drug approval for patients, to give patient access to drugs in the United States. The accelerated approval process was initially introduced in 1992, and it was codified into in by Congress in 2012. The accelerated approval process has been around for a long time, and it was developed by the FDA to help speed up the approval process.

If you work for overall survival, if drugs can last for a longer period of time, or you can go to 1 drug to another drug to another drug, it is going to be hard to prove survival benefits because every additional line of therapy could give patients additional survival. But it slows down the approval of new drugs. Then they switched to progression-free survival, but then the same problem happened there for the same reason. Right now, the drugs are sort of too good to allow new drugs to be approved, but they are not good enough to offer a cure to patients.

I think the unmet need is huge in myeloma. In myeloma, about 40% of patients do not live beyond 5 years. There is no cure; there's a huge unmet need, and there is no path forward to develop new drugs for myeloma. I saw this back in 2009, when I worked at the NIH as a federal government employee in the intramural program, and I started this program 15 years ago. I worked on this evidence meta-analysis for 15 years, and I collaborated with the FDA, the NCI and the National Heart, Lung, and Blood Institute. I held the IND, and we invited a lot of other groups. Eventually, there was another group that started working on it, also the I2TEAMM, and they started, I think, in 2014 or 2015. The 2 teams have worked then in parallel, and we could together go to the FDA and show our results.

Could [MRD as an end point] go to other diseases? The same situation has happened in many other diseases or is happening. The drugs are blocking development for the same reason. I think you can make the same case in leukemia, you can make the same case in lymphoma, you can make the same case in many other solid tumors, but the FDA will not grant approval for this just because we went to the ODAC. You have to do all the work. I spent 15 years.