FDA advisors endorse minimal residual disease (MRD) as accelerated approval endpoint for multiple myeloma

Quick background on multiple myeloma

• Myeloma is a blood cancer in which B cells, a type of immune cell, overproduce plasma cells. These plasma cells crowd out other types of blood cells in the bone marrow. Different subtypes of myeloma are labeled and staged according to the prevalence and histology of the plasma cells and the proteins they produce.
• When abnormal plasma cells build up in multiple locations in the bone marrow, the disease is called multiple myeloma (MM). This type of hematological malignancy affects fewer than 50,000 people in the U.S., is more common with increasing age and in males, African Americans, and those with a family history, among other risks.
• Treatment for MM may be lifelong, with patient and disease characteristics guiding the course of treatment. Patients frequently receive up to four therapies at once and go through multiple prolonged courses of treatment to keep recurrence and progression at bay. Commonly combined treatment options include antineoplastics (chemotherapies), radiation therapy, corticosteroids, immunomodulators, proteasome inhibitors, and monoclonal antibodies. Hematopoietic stem cell transplantation (HSCT) is another mainstay of MM treatment, though not all patients are eligible for this option.
• Over time, as more products have been approved for MM, the list of potential combinations has also grown. The specific combination that is recommended is informed by various patient-specific factors, such as whether the patient is newly diagnosed, is preparing to undergo autologous HSCT, or has already undergone one or more treatments without an adequate or durable treatment response.
• Despite the availability of treatments, there is still significant unmet need for patients with MM. Although therapeutics have contributed to improvements in survival, the five-year relative survival rate for patients with MM was just 57% between 2012 and 2018.

Regulatory context: Accelerated approval for MM treatments

• Quick background: FDA has authority to grant accelerated approval (AA) on the basis of surrogate or intermediate endpoints, for “drugs for serious conditions that filled an unmet medical need,” according to the FDA’s Accelerated Approval page. A surrogate or intermediate clinical endpoint stands in for a clinical outcome that improves how patients feel, survive, or function – the threshold for traditional approval. To receive approval, the surrogate or intermediate must be “considered reasonably likely to predict the clinical benefit of that drug.” Products granted accelerated approval are required to conduct confirmatory studies to verify their clinical benefit and convert the accelerated approval to a traditional approval. These postmarketing requirements (PMR) are usually agreed upon at the time of the marketing application review and approval, though execution of these requirements has been variable.
• Given that MM is a serious and life-threatening condition, many drugs have received accelerated approval for this indication. As of March 31, the FDA’s quarterly list included 12 products that had received accelerated approval for an MM-related indication. Of those 12 products, six have been converted to traditional approval, three have been withdrawn, and three are pending.
• The most recent accelerated approvals occurred in August 2023: Elrexfio (elranatamab-bcmm; Pfizer) and Talvey (talquetamab-tgvs; Janssen). These drugs were both approved for relapsed or refractory (RR) MM patients who had already received at least four prior lines of therapy. The approvals were granted on the basis of two intermediate endpoints – overall response rate (ORR) and duration of response (DOR), based on International Myeloma Working Group (IMWG) criteria.
• Due to the high rates of “complete response” seen with new treatment approaches, a 2016 update to the IMWG criteria recognized the need to identify “deeper” responses than those conventionally defined as complete response. To this end, the IMWG added new response categories of minimal residual disease (MRD) negativity to add an additional layer of depth to outcomes reporting both within and outside of clinical trials.

Regulatory context: Minimal residual disease (MRD) as an intermediate endpoint

• Minimal residual disease (MRD) is defined in the FDA’s 2020 guidance on the use of MRD for hematologic malignancies as: the detection of malignancies at low levels by measuring cell characteristics such as genetic mutations, cell surface markers or specific DNA gene rearrangements. As outlined by the FDA in the guidance, there are several technologies capable of measuring MRD, including multiparametric flow cytometry (MPFC), next-generation sequencing (NGS), quantitative reverse transcription polymerase chain reaction (RT-qPCR) of specific gene fusions, and allele specific oligonucleotide polymerase chain reaction (ASO-PCR). Potential challenges in the MRD approach include lower sensitivity than conventional morphologic methods and discrepancies due to laboratory variations.
• MRD’s acceptability as a diagnostic, prognostic, predictive, efficacy-response or monitoring biomarker is highly dependent upon its specific context of use. While the FDA considers MRD to be a “ general measure” of tumor burden, the agency’s opinion on MRD for regulatory use depends on the disease it is used to assess. For instance, when FDA issued the MRD guidance in 2020, it referred to MRD as “one of the most significant prognostic factors” in acute lymphoblastic leukemia, regardless of a host of disease and patient factors. In contrast, that guidance also stated that “the molecular heterogeneity of AML [acute myeloid leukemia] poses substantial challenges to the use of MRD as a biomarker.”
• MRD in multiple myeloma, specifically: The 2020 guidance states that “significant improvements in clinical outcomes of MM have spurred interest in the use of MRD as a potential surrogate endpoint to expedite drug development.” At that point, MRD had most often been used to evaluate the persistence of MM in newly diagnosed patients in the post-transplant setting. As a result, the guidance stated that “The relationship between MRD and clinical benefit and the test performance characteristics should be demonstrated in each disease setting (e.g., relapsed refractory (RR), newly diagnosed, nontransplant eligible, smoldering MM) to validate MRD as a surrogate endpoint in MM.”

Last week, FDA advisors weighed in on efforts to validate MRD across the spectrum of MM disease settings

• FDA’s Oncologic Drugs Advisory Committee (ODAC) met on April 12 to discuss “the use of minimal residual disease (MRD) as an endpoint in multiple myeloma clinical trials, including considerations regarding timing of assessment, patient populations, and trial design for future studies that intend to use MRD to support accelerated approval of a new product or a new indication.” ODAC was not tasked with discussing specific products, but rather with evaluating the adequacy of the data as a whole to support the use of the endpoint in this context.
• The committee heard results from the two independent research groups that submitted applications for the use of MRD as a surrogate endpoint in MM: The University of Miami Sylvester Comprehensive Cancer Center and the International Myeloma Foundation’s i2TEAMM. The University of Miami meta-analysis included phase 2 or 3 trials which enrolled patients with newly diagnosed MM (transplant eligible or transplant-ineligible) and performed validated MRD assays. The analysis ultimately included eight studies for a total of 4,907 subjects with individual patient-level data (IPD). The i2TEAMM meta-analysis had a broader scope, including previously untreated, newly diagnosed patients (transplant eligible or ineligible) as well as patients with RRMM. The analysis included 20 trials for 12,926 patients with IPD. Note: Both analyses focused only on studies of small molecules, antibodies and biologics; studies of CAR T therapies were not included.
• The meta-analysis protocols, which were pre-specified and cleared by the FDA, leveraged statistical methodologies that have previously been used to evaluate the correlation between other early endpoints and long-term clinical endpoints. Regression analyses were conducted to find the association between MRD negativity and progression-free survival (PFS) as a primary analysis and overall survival (OS) as a secondary analysis. The specific methods included correlation by Global Odds Ratio (OR), a bivariate copula model (R^2 Copula) and weighted least-squares regression (R^2 WLS).
• The FDA clarified that these meta-analyses are intended to “assess the strength of two associations: association at the individual level and association at the trial level.” An individual-level association assesses the relationship between those two endpoints within an individual patient. On the other hand, trial-level association assesses whether there is a correlation between the treatment effect on the earlier endpoint and the treatment effect on the long-term endpoint.
• At a high level, the groups came to concordant conclusions in their evaluation of individual- and trial-level associations of MRD with long-term endpoints. The groups found a similar strong association between MRD-negative complete response (CR) and PFS at the individual level in newly diagnosed populations. However, the association at the trial level was less robust, showing some association between MRD and PFS and no significant association with OS. Importantly, i2TEAMM found that MRD sensitivity level, or detection thresholds, affects this outcome, with increased association at the 10^-5 level compared to 10^-4.
• To validate the findings of these meta-analyses, the FDA conducted its own analysis of the data, removing duplicate trials and imputing missing data as non-responders. The FDA ultimately came to similar conclusions, but with some important differences: while MRD-negative CR at 9 or 12 months “ha[s] strong individual-level correlations with PFS and OS across multiple MM disease settings,” trial-level association with PFS was deemed to have a “weak-to-moderate association.” The agency also found no association at all between MRD negativity and OS at the trial level.

The committee sought to tease apart the clinical meaning and implications of the meta-analyses

• JORGE NIEVA, oncologist at the University of Southern California, asked the FDA to interpret the discordant individual-level versus trial-level associations. FDA’s JONATHAN VALLEJO, team leader in the Division of Biometrics IX (DBIX), described how the agency views the strength of evidence in endpoint development. “Probably the weakest rationale is just biological rationale. The next step up would be individual-level correlation. So, if you have biological rationale and individual patient-level association, that’s stronger. And the strongest criteria would be trial-level, so that’s the strongest measure point for surrogacy.” However, FDA clinical reviewer RACHEL ERSHLER clarified that because the goal is to validate MRD as an endpoint for accelerated approval, and not regular approval, strong trial-level association is not required.
• Digging into the data, the group asked for an explanation of the weaker association for certain subpopulations and contexts. In the RR populations, NEIL VASAN of Columbia University observed that “the R-squared values are lower for the transplant-eligible group compared to the ineligible group.” However, FDA reviewers cautioned against subdividing the data into distinct populations, since this reduces the power to detect associations.
• RAVI MADAN, oncologist at the National Cancer Institute and chair of the committee, expressed concern that the acceptance of MRD as an endpoint for accelerated approval could “change incentives” for MM therapeutic development. “All of a sudden, perhaps you could see a world where preclinical modeling is now more focused on the biologic and maybe less on the clinical, as would the phase 1 and 2 development.” He asked what this would mean for confirmatory trials, questioning “With PFS being negative, if MRD is positive, would that be sufficient to remove the accelerated approval?” NICOLE GORMLEY, associate director of Oncology Endpoint Development, explained that a decision to withdraw an accelerated approval relies on the totality of data with multiple types of analyses. In the proposed hypothetical scenario, she explained that there would also likely be OS data and other safety data to inform the decision.
• One participant in the Open Public Hearing (OPH) expressed concerns that seeped into the committee’s deliberations. VINAY PRASAD, hematologist-oncologist at San Francisco General Hospital offered an alternative perspective that, given the number of available treatments for MM, using MRD would result in premature drug approvals and potentially increase exposure to toxicity. “The biggest problem with MRD for approval is that you’re taking people who are doing pretty well – great overall survival for a decade or more – and you’re giving them drugs with very inadequate safety profiles,” he said.

Three discussion questions and a vote

DISCUSSION QUESTION: Discuss the adequacy of the available data to support the use of minimal residual disease (MRD) as an accelerated approval endpoint in multiple myeloma (MM).

• The committee members praised the work of the research groups and found the data compelling. “It’s rare to have this kind of discussion where it’s all academic and with a great amount of integrity of data,” said Stanford oncologist RANJANA ADVANI, continuing, “I think the patient level is very, very clear.” University of Colorado Cancer Center’s CHRIS LIEU said, “I think we’d all like to see the trial-level data show more correlation, but that’s not the bar that’s set for accelerated approval.” From the myeloma expert perspective, THOMAS MARTIN of the University of California-San Francisco emphasized the biological significance of MRD, saying “if you get that low with your MRD, you’re probably going to have a longer span.”
• Several committee members echoed Madan’s concerns about incentives and “gamesmanship” if industry’s focus is shifted to MRD at a single time point. Vasan agreed, but stated he is “assuaged” by recent legislative reforms under FDORA, which he thinks “gives the FDA a muscular policy to enforce accelerated approval with multiple safeguards in place to nudge companies to comply.” RICHARD PAZDUR, director of the FDA’s Oncology Center of Excellence (OCE), chimed in to say that “the safeguard really is the randomized trial that will happen subsequently.” He also emphasized the importance of “early randomized information on dosing” to prevent the removal of drugs from the market due to toxicity.

DISCUSSION QUESTION: Discuss whether the available data supports the use of MRD as an endpoint in the different MM disease settings: Newly diagnosed MM, Relapsed/Refractory MM

• The committee members agreed that MRD would be a useful endpoint in both frontline and RR settings. While the data in the RR setting was the weakest, the members acknowledged that the studies were smaller and not as highly powered. In addition, Martin pointed out that the studies in this population were skewed toward specific classes of therapeutics that do not necessarily represent all MM treatments.

DISCUSSION QUESTION: Discuss the acceptability of the timepoints for MRD assessment: 9-months, 12-months, MRD negative complete response at any time; Requirement for assessment of durability

• The committee members supported a flexible approach that would allow MRD assessment at 9 or 12 months based on the study design. A difference of three months can have a real relevance for some patients, as it would alter the number of cycles of induction, transplant and consolidation. On the other hand, newly diagnosed patients may be better suited for an earlier readout.
• While the durability of the effect could be useful to assess, members flagged concerns with the burden on patients. Nieva said, “While I think durability is a good thing to know, I have tremendous fear that this is going to be: Every myeloma protocol has a marrow biopsy every six weeks on the patients, forever.”

VOTING QUESTION: Does the evidence support the use of MRD as an accelerated approval endpoint in MM clinical trials?

• The committee voted unanimously in favor of MRD, with 12 yes votes and zero no votes. Lieu explained that the studies clearly fulfill the FDA’s stipulations regarding biological plausibility, prognostic impact, and clinical evidence (at the patient level). CHRIS HOURIGAN, hematologist-oncologist at Virginia Tech, pointed out that “There is harm to inaction,” saying, “We’re not currently curing people with multiple myeloma. I’m not willing to make patients wait for a theoretical perfect that may never come. Our responsibility is to accept ‘a little bit messy,’ and be agile enough to adapt and iterate on the evidence developed, and not create barriers to the work of discovering effective new therapies for patients.”

Analysis

• The committee’s endorsement of MRD as an accelerated approval endpoint will offer a new and potentially more commercially attractive option for MM drug developers. According to ClinicalTrials.gov, there are currently about 200 Phase 2 or 3 active clinical trials for MM treatments that include MRD as an outcome measure. This outcome may also have implications for manufacturers of MRD technology and assays; some remarks from FDA reviewers underscored the importance of consistent, highly sensitive MRD assays.
• As referenced during the committee meeting, the FDA received an expansion of its accelerated approval authority through the FY23 Omnibus legislation, also known as the Food and Drug Omnibus Reform Act (FDORA). First, the FDA received explicit authority to require, “as appropriate,” that a study or studies be underway prior to accelerated approval, or within a specified period. In fact, the agency recently flexed this ability in two March 2024 Complete Response Letters (CRLs). Second, the FDA received authority to use a new “ expedited procedure” which allows the FDA to withdraw the approval of a product that had been granted accelerated approval when the confirmatory study is not conducted or fails to verify benefit, or in other specific situations.
• Comments from FDA reviewers at the meeting highlighted the importance of these new legislative authorities as a method to safeguard against risks associated with the incomplete evidence inherent in accelerated approvals. These comments confirmed that applications that first receive approval based on MRD will be subject to strict scrutiny as confirmatory studies are planned and completed.
• This seems particularly important considering the poor correlation between OS and other accepted surrogate endpoints in MM. A study published in the British Journal of Hematology in 2022 evaluated the correlation between time-to-event surrogate endpoints in MM clinical trials conducted between 2005 and 2019. The analysis found that only 43% of the variance in OS was due to changes in PFS, and that the overall correlation between PFS and OS was weak (0.65). In RRMM, this association grew slightly stronger, with 58% of the variance in OS due to changes in PFS, representing a medium correlation (0.76). This is of particular note considering that the meta-analyses presented to the committee found the strongest correlation between MRD and PFS, as opposed to OS.
• The MM setting also offers clear examples of drug regimens that have ultimately yielded an OS detriment due to drug-related toxicities. Two studies evaluating the addition of pembrolizumab to an existing RRMM regimen were terminated early by the FDA due to an increased risk of death. Although these studies did not show a PFS benefit at the time of study termination, they also did not show a PFS detriment. OS outcomes, on the other hand, were strongly weighted in favor of the placebo arm, with OS hazard ratios of 1.61 in one study and 2.06 in the other. A separate study, the BELLINI trial, which evaluated a specific regimen in RRMM patients, demonstrated a PFS clearly in favor of the treatment arm (hazard ratio = 0.58). However, the median OS ultimately favored the placebo arm, with a hazard ratio of 1.19.
• Considering the extended lifespan that many MM patients are beginning to see as a result of newly available treatment options, treatment-related toxicities may be of particular concern. While the five-year relative survival rate for patients with MM was estimated to be 57% between 2012 and 2018, more recent research suggests that this number may be increasing rapidly. The PERSEUS trial, published in the New England Journal of Medicine in 2023, identified a 4-year survival rate of about 90%. And some estimates indicate that median patient survival may now be as high as 8-10 years. While this is an unquestionably positive development, it does introduce a new confounder. MRD as a surrogate endpoint would allow for shorter trials, which ultimately provide less safety data at the time of approval. This could result in exposing patients to drug-related toxicities that could turn out to either shorten survival time or introduce long-term sequelae.

Ongoing changes to advisory committee meeting structures were on display at this ODAC session

• As AgencyIQ has previously discussed, FDA leadership has been previewing changes to the Advisory Committee system for over a year. While many changes have been discussed at the agency level, Pazdur has articulated his own vision for the future of ODAC, which involves some separate reforms.
• Back to White Oak: The presenters and committee members participated in this meeting in-person from the FDA’s White Oak campus, facilitating some additional back-and-forth exchanges.
• A single briefing document: This meeting combined all relevant information into a single briefing document, using a “point-counterpoint” approach to organize the perspectives on various topics. Pazdur, who has recently advocated for this approach, explained at the meeting that he intends for this to be the “default position” for ODAC going forward, replacing the prior approach of providing two separate briefing documents which often include duplicative information. He emphasized that while the product is offered as a single “simplified” document, the drafting process is not collaborative and the inputs are separate. Acknowledging that this joint approach was particularly controversial in the context of a different advisory committee meeting, Pazdur encouraged members of the committee to reach out with any concerns regarding the adoption of this new approach.

Featuring previous research by Rachel Coe and Kari Oakes.

To contact the author of this item, please email Amanda Conti ( [email protected]).To contact the editor of this item, please email Chelsey McIntyre ( [email protected]).

Key Documents and Dates

• April 12, 2024 Meeting of the Oncologic Drugs Advisory Committee (all meeting materials and recording linked here)
• FDA Docket No.: FDA-2024-N-1179