Text: Amino Observations
Hematological tumor drug development or welcome the new change.
On April 12, the FDA held its third Oncology Advisory Committee (ODAC) discussion of the year.
Instead of discussing clinical trials of specific drugs, the ODAC meeting focused on a topic that was a turning point in the clinical trial of a drug:
Does the available data support minimal residual disease (MRD) as an accelerated approval endpoint for multiple myeloma (MM) clinical trials?
After analyzing and discussing a series of data, the expert group voted 12:0 in favor of MRD as an alternative endpoint for the accelerated approval of new drugs for MM.
This not only means that after years of MRD certification, it is expected to significantly shorten the time to market for new MM drugs, but it may also have a profound impact on the development and validation of new anti-tumor trial endpoints.
While the results of the above vote do not yet represent the final decision of the FDA, given the overwhelming support of the panel and the FDA's active review prior to this meeting, it may be possible that MRD will soon be incorporated into FDA's regulatory practice.
Not only hematologic tumors, but also MRD monitoring has great potential in the field of solid tumors.
A study conducted in lung cancer demonstrated that more than 99% of MRD-negative patients did not have recurrence and that MRD predicted recurrence prior to routine imaging. This also means that MRD monitoring for lung cancer can play a very critical role in the prognosis management of patients.
Not only lung cancer, but also solid tumors such as colorectal cancer and breast cancer, MRD monitoring products are also developing rapidly. In the future, as research advances, the relationship between MRD and other tumor clinical endpoints may also be revealed.
Perhaps the exciting challenge in the coming years for the development of new drugs for MM or other oncology is how to better use MRD to guide clinical and treatment decisions.
This process is both promising and uncertain, but it will undoubtedly have a significant and meaningful impact on drugs that can bring hope for new treatments to patients earlier.
From ORR to MRD
In the field of MM, there are many good drugs, and a large number of recent clinical data show that the medium-term survival of both newly diagnosed patients and patients after relapse has been significantly extended compared with a few years ago.
This also puts forward new requirements for the clinical endpoints of new MM drugs.
In the past, marketing applications for new MM drugs were routinely approved through progression-free survival (PFS) and overall survival (OS), or accelerated approval through objective response rate (ORR). For example, in 2022, Johnson & Johnson/Legend Biotech's CAR-T therapy was approved for marketing, relying on an objective response rate of 98%.
But at the moment, both types of endpoints face their own problems.
First of all, the response rate of MM treatment is generally high, and the standard treatment for newly diagnosed MM has reached more than 90%, and it is difficult to accurately reflect the benefit of new treatment through ORR.
Previously, it has also been demonstrated that ORR data are not fully equivalent to overall survival benefit.
In the past, many oncology drugs were marketed with alternative endpoints such as PFS or ORR, but in the end, many drugs failed to equate these surrogate endpoints with the gold standard of OS. That's why the FDA is increasingly looking to pharmaceutical companies to provide reliable overall survival data to prove that new therapies can really help patients live longer.
For example, CAR-T therapies such as Abecma do have the possibility of OS damage in clinical practice. In a paper in a phase III clinical study in the New England Journal of Medicine, the researchers disclosed in an appendix that 30 percent of patients in the Abecma group died, compared to 26 percent in the standard treatment group.
To this end, the FDA has convened a special meeting to evaluate the risk-benefit of CAR-T therapy. Of course, in the end, the FDA experts voted 8-3 to determine that the long-term benefits outweigh the risks, i.e., Abecma's data support a positive risk-benefit profile when used in third- to fifth-line myeloma treatments.
On the other hand, with the emergence of various new therapies, the PFS of MM patients is significantly prolonged. Median PFS has been greater than 6 years and median OS has been greater than 10 years with standard therapy. As a result, new treatments require longer follow-up to show a statistically significant difference in PFS, plus enrollment and follow-up times, even beyond 10 years.
For pharmaceutical companies, it is difficult to accept that a clinical trial can take 8 or even 10 years, after all, the patent life of a new drug is limited, and it is difficult for patients to use new treatment drugs.
In this case, it is necessary to reform and develop new endpoints to accurately predict the long-term clinical outcomes of patients at an early stage, so that new effective therapies can be marketed in a timely manner, bringing benefits to patients with MM who are still unable to obtain a cure.
MRD stepped up.
MRD, or microresidual disease, refers to the amount of tumor cells that remain after cancer patients receive treatment, and the amount may be negligible, but it may still cause cancer recurrence.
Although MM is still incurable, treatment options are numerous. From chemotherapy and transplantation, to proteasome inhibitors, immunomodulators, and anti-angiogenic inhibitors, the prognosis of MM has been significantly improved, and now with the emergence of CD38 monoclonal antibodies, ADCs, and CAR-T therapy, MM tends to be a "chronic disease".
Therefore, monitoring of patient prognosis and early screening for recurrence are becoming increasingly important. Globally, it has long been a consensus that patients with hematologic malignancies need to be monitored for MRD after treatment.
Over the past few years, the medical community has also been investigating how high-quality MRD data and reporting can be used as a basis for improving the prognostic role of MRD and as a support for the evaluation of clinically meaningful endpoints.
Now, after oncology experts reviewed data supporting surrogate endpoints for drug approval for multiple myeloma, the 12 members of the FDA advisory committee voted unanimously in favor of using MRD as an alternative endpoint to accelerate approval of new treatments for multiple myeloma.
Some market participants said that given the overwhelming support of the panel and the FDA's active review before this meeting, MRD may soon be incorporated into the FDA's regulatory practice.
The time is ripe
With MRD, researchers use the most sensitive techniques, such as flow cytometry NGF or next-generation sequencing NGS, to measure disease burden.
NGF and NGS can increase the sensitivity of MRD detection to more than 10-5, respectively. MRD-negative means that less than 1 clonal plasma cell can be detected in 105 nucleated cells. i.e., MRD negative results, which represent deeper tumor clearance than ORR and CR.
It has been reported that a negative MRD is more accurate in indicating a patient's prognosis. Therefore, the International Myeloma Working Group IMWG has also added MRD efficacy evaluation in addition to the traditional efficacy evaluation. Some scholars have proposed that MRD negativity should become the new CR criterion.
In 2020, a large meta-analysis published in the journal Blood Advances involving 44 clinical studies of multiple myeloma and PFS data from nearly 8100 patients showed that achieving MRD negativity reduced the risk of disease progression or death by 67%. According to the paper, there were overall survival data for nearly 4,300 patients in 23 studies, and a negative MRD was associated with a 55% improvement in the risk of death.
The study caused a stir in the academic community, but the FDA was not convinced. On April 12, FDA experts convened an advisory committee meeting to review the new data, and after reviewing data from two independent meta-analyses conducted by EVIDENCE and i2TEAMM on past clinical trials, experts unanimously supported MRD as a surrogate endpoint.
Specifically, the EVIDENCE study selected 16 phase 2/3 randomized controlled studies in patients with newly diagnosed MM for meta-analysis, in which the MRD sensitivity of the selected studies reached more than 10-5, and the main purpose of the study was to evaluate whether MRD negativity can be used as a reasonable surrogate endpoint for long-term clinical benefit.
The results showed a strong correlation between individual patient-level MRD and long-term benefit, with a global OR (global odds ratio, the higher the global OR greater than 1, the stronger the correlation) with PFS in all newly diagnosed MM (NDMM) and in newly diagnosed patients who were eligible for transplantation (NDTE and NDTinE), respectively, showing a strong correlation between patient MRD and PFS. Similarly, the results also showed a strong correlation between patient MRD and OS outcomes.
In terms of the level of trials that reflect the correlation between MRD endpoints and the long-term endpoints of the study, the results showed that there was some degree of correlation between MRD and PFS/OS across all NDMMs. For example, in terms of OS, in all NDMMs, MRD was weakly to moderately correlated with OS endpoints due to post-progression crossover or treatment with other line counts, while in NDTinE, MRD was moderately to strongly associated with OS endpoints.
The i2TEAMM study included 20 multicenter randomized controlled trials (RCTs) that meta-analyzed data from more than 12,000 newly diagnosed and relapsed patients with refractory MM (NDTE, NDTinE, and RR) to assess whether the early MRD endpoint (9-month and 12-month MRD negative rates) could be used as an early endpoint predicting long-term clinical benefit.
The results showed a strong correlation between the 9-month MRD negative and PFS endpoints, with global OR values of 3.06, 9.80, and 8.24 in NDTE, NDTinE, and RR patients, respectively, and a strong correlation in OS endpoints.
At the trial level, the results showed that the 9-month MRD negativity rate showed a moderate correlation with PFS and OS endpoints.
Comprehensively, both EVIDENCE and i2TEAMM believe that MRD can be used as a reasonable endpoint to support MM accelerated approval.
A growing body of data analysis suggests that the time may be ripe for the use of MRD as a surrogate endpoint.
Overall, the FDA endorses the methods and results of the two independent meta-analyses described above. At the same time, based on all the data submitted by both applicants, the FDA conducted an additional meta-analysis to verify whether the use of all data would affect the results or conclusions.
The results showed that at the individual patient level, 9-month and 12-month MRD negativity showed a strong correlation with PFS and OS, respectively, at the trial level, MRD and PFS showed a weak to moderate correlation in most disease subgroups, a strong correlation in the NDTinE subgroup, and an overall weaker correlation between MRD and OS.
While improvement in MRD may not predict the risk of long-term clinical benefit, the FDA also notes that this is a risk of using any of the early endpoints.
A new turning point?
If the FDA finally agrees to use MRD as an alternative endpoint, it will mean a turning point in the development of new myeloma drugs, with the clinical time of pharmaceutical companies greatly shortened and drugs able to reach patients faster.
At the previous American Society of Hematology Annual Meeting, Peter Voorhees, MD, from Atrium Health, presented a hypothetical Phase 3 trial that would target Johnson & Johnson's Darzalex or Sanofi's Sarclisa in combination with Takeda's Velcade, Bristol-Myers Squibb's Revlimid, and the steroid dexamethasone in first-line patients eligible for stem cell transplantation.
In order to statistically show PFS benefit, the trial would need to enroll more than 1,700 patients and would take nearly 10 years to reach the final PFS analysis, according to Voorhees' calculations.
But if MRD is used as a surrogate endpoint, the situation is different. Rahul Banerjee, a physician-researcher who studies multiple myeloma at Fred Hatch Cancer Center, estimates that in some cases, the dissociation of MRD-based treatment effects may begin to show up as early as four months after treatment.
Dr. Landgren, who led the EVIDENCE study, believes that a meaningful, statistically significant MRD improvement can be read out three years after the trial began.
In the i2TEAMM study, with MRD as an early endpoint, results were available in 9 to 12 months, rather than more than 5 years.
So, how to use MRD to accelerate the development of new MM drugs?
According to the FDA's supplementary note, MRD can replace ORR in a single-arm study of the posterior line treatment of relapsed and refractory MM in the traditional "posterior single-arm-frontline randomized controlled" two-trial development approach, based on the response time of MRD and DOR to obtain accelerated approval, and continue to be further validated as a secondary endpoint in the randomized controlled study of front-line therapy.
The other approach is the FDA-recommended "single trial approach", which can be based on MRD assessment in a randomized controlled trial, which can be continued if the MRD is not positive, and then followed up on long-term clinical outcomes in the same trial, with routine approval based on long-term benefits such as PFS and OS.
Of course, there are still some limitations and uncertainties about this. For example, trials of CAR-T therapy were not included in the meta-analysis, possibly due to timing. In addition, the duration of evaluation and follow-up varied between trials.
In addition, the results may not reflect the relationship between MRD and PFS in other trials that did not meet the criteria, as trials needed to meet some pre-defined criteria for inclusion in the meta-analysis.
In an article published last year in Clinical Cancer Research, FDA officials, including oncology director Richard Pazdur, pointed to several problems with the use of drugs for hematologic malignancies using MRD in the past. For example, of the types of cancer evaluated, multiple myeloma "had the lowest acceptance rate of MRD data despite the highest number of submissions".
The FDA noted that the main problem stems from the test method used to detect MRD. In many cases, NGS-based assays fail to identify malignancies suitable for tracking. The FDA said the missing data made the overall results uninterpretable.
Another concern raised by the FDA panel was that opening up MRD approaches could have potentially unintended consequences for drug development. With too much focus on MRD, pharmaceutical companies may abandon effective drugs that do not show early MRD signals. That's why the FDA prefers a "single test" approach.
While there are still open questions about how best to use MRDs, the FDA also stated that the meta-analyses conducted represent a robust assessment of MRDs, support their prognostic value, provide information on the appropriate timing of MRD evaluations, and indicate that MRDs may be suitable as intermediate clinical endpoints to support accelerated approval.
However, MRDs, like other clinical endpoints that support accelerated approval, are all provided that long-term benefits can ultimately be demonstrated. With the change of regulation, MRD may also usher in a turning point of accelerated development.
Globally, it has long been a consensus that patients with hematologic tumors need to be monitored for MRD after treatment, and clinical research on MRD detection of solid tumors is also in full swing. Including lung cancer, colorectal cancer, breast cancer and other solid tumor fields, MRD monitoring products are also developing rapidly.
What is certain is that in the future, MRD testing will not be limited to the field of hematological tumors, but will also play a big role in the field of solid tumors. Of course, from the perspective of sensitivity, clonal heterogeneity, accessibility, and sample requirements, MRD in solid tumors has higher technical requirements.
At any time, MRD obtained by reliable detection technology is an important indicator of prognostic management, and on this basis, the relationship between MRD and tumor clinical endpoints and long-term patient benefit can be further studied and revealed.
This time, the unanimous support of FDA experts means that the MRD has finally been approved for many years, which may be the beginning of a change in the clinical endpoint of oncology drugs.
Perhaps the exciting challenge in the coming years for the development of new drugs for MM or other oncology is how to better use MRD to guide clinical and treatment decisions.