Meet the Expert: Prof. C. Ola Landgren discusses MRD as a key endpoint in haematological cancer trials

To better understand the use of minimal residual disease (MRD) as a surrogate marker in myeloma and other haematological cancers, Medicom interviewed Prof. C. Ola Landgren (University of Miami, FL, USA).



www.umiamihealth.org

What is the rationale for minimal residual disease (MRD) as an endpoint in clinical trials evaluating treatments for patients with multiple myeloma or other haematological cancers?

“The reason why MRD has become an important endpoint is because the historical endpoints have become obsolete. A long time ago, overall survival was the only endpoint. This was at a time when there were no effective therapies. So, if you could prove the patient lived longer, that was a success story, and you could use that to get drugs approved. But as drugs improved and could make a patient live longer and longer, it would take a very long time to prove superiority.

So, the FDA adopted progression-free survival many years ago. Progression-free survival means that patients who either die or progress, count as an event. If you could show that a new therapy had a longer progression-free survival, you could get it approved. Now, the problem is that current drugs cannot cure patients. There is no established cure. About 40% of patients don't live 5 years or longer. There is a huge unmet need. However, with this endpoint, if you were to design a study for newly diagnosed patients, you need to consider the statistical framework required to show superiority.

Firstly, you need to have a randomised study. Secondly, you need sufficient numbers to statistically test the hypothesis that the new therapy is superior to an old one. That would require large sample size, usually hundreds of patients. Identifying and enrolling all those patients would typically take at least 2 years. Then, to prove in the newly diagnosed setting that there is an improved progression-free survival, the dataset needs to mature, which takes about 10 years from randomisation.

Considering that there is still no cure for the disease, and about 40% of patients don't survive 5 years or longer, this creates a huge barrier for drug development. Patients need new drugs, but it takes 2 years to enrol patients and then over 10 years for the dataset to mature.

I saw this coming over 15 years ago when I worked at the National Cancer Institute (NCI) at the National Institutes of Health (NIH). I established contacts within the the NIH, National Cancer Institute, and the National Heart, Lung, and Blood Institute. We formed an interagency agreement with the FDA and worked together to form a task force within the federal government. After a few years, we had strong evidence that testing for MRD after about a year would be a very strong predictor of what would happen many years later.

The model was basically: if you enrol patients for 1 year and check after 1 year. If a patient has a deep response, you are very close to predicting what will happen 10 years later. My idea was if we could test all the patients 1 year out in both arms of the study, we would have a very strong prediction of the outcome. That is the background rationale.”

Would you say that blood cancers have the edge on most solid tumours, where MRD is still an emerging concept?

“Every disease has rules for determining a response. Over 20 years ago, we agreed that if you could reduce the disease by half using a biomarker in the blood, that would be called a partial response. If the bone marrow was also free from disease, it was a complete response. So, if you had at least a partial response, it was counted as an overall response.

The FDA decided that for relapsed refractory disease, you could use overall response as an early endpoint for accelerated approval. You could evaluate the overall response rate at a fixed time point in heavily pre-treated patients, and if over 30% of patients had a response, you could get accelerated approval. However, in newly diagnosed patients, most will have at least a 50% reduction of the disease, but that does not help the patient enough. We need a more stringent measure to eliminate the disease, making MRD a critical step forward.”

Can you briefly touch upon the key trials and pieces of evidence that brought this to be?

“I am the lead investigator for the evidence meta-analysis published in the journal Blood [1]. We reviewed the entire literature through all publicly available databases, looking for newly diagnosed and relapsed refractory multiple myeloma patients who had been treated in randomised clinical trials with MRD testing.

We recognised that over time, the stringency for MRD negativity has evolved. In 2016, we published the definition for MRD negativity [2]. Previously, every group had their way of defining it. Less sensitive tests might still call it MRD negative, but in the modern era, to be MRD negative means truly eliminating the disease. Therefore, we included only high-sensitivity data sets (1 cell in 100,000, or 10^-5) and excluded lower-quality data.

We worked with the FDA for a long time, developing a statistical analysis plan endorsed by the FDA. I filed an IND similar to how drug companies file for drug approval, based on FDA guidance.”

What implications does MRD's negative status have on patient management and treatment decisions? How do you balance the need for early intervention based on MRD status with the potential risks of overtreatment?

“Tests in clinical medicine can be used for many purposes: determining remission, picking up recurrence, monitoring, etc. MRD tests can similarly be used for various purposes. Everything we discussed is in the context of drug approval. We tested the hypothesis that MRD 1 year after randomisation predicts clinical benefit, and our hypothesis was correct.

Using MRD to manage patients is different from drug approval. There are ongoing trials to see if MRD status could change treatment or determine it in advance. For example, randomised trials are exploring different maintenance strategies based on MRD status. MRD will likely be a major marker for decision-making in the future, but it's important to understand that this is just another tool, similar to protein electrophoresis, free light chains, PET scans, and bone marrow results.

The current challenge is the lack of an established, validated blood-based MRD test. A highly accurate, widely available blood test would revolutionise the field.

That's my perspective.”

1. Landgren O, et al. Blood. 2024 May 20:blood.2024024371.
2. Kumar S, et al. Lancet Oncol. 2016 Aug;17(8):e328-e346.

Copyright ©2024 Medicom Medical Publishers



Posted on 10 July 202423 December 2024