Regulatory Focus™ > News Articles > 2019 > 4 > Minimal Residual Disease as a Surrogate Endpoint for Product Development for Multiple Myeloma: Compa

Feature ArticlesFeature Articles
Posted 08 April 2019 | By Telly Chi, PharmD, MS, RACAditi Sinha, PhD  | ©

Minimal Residual Disease as a Surrogate Endpoint for Product Development for Multiple Myeloma: Comparing FDA and EMA Guidances

3809 This article discusses the need to update regulatory science to include the use of Minimal Residual Disease (MRD) as a specific measure of tumor burden for multiple myeloma at levels undetectable through conventional laboratory techniques. The authors suggest MRD can potentially be used as a clinical and regulatory endpoint to evaluate a drug’s effect both on patients’ risk of future relapse risk and subsequent treatment survival outcomes, thus serving as a surrogate endpoint for early drug approvals for patients with multiple myeloma. The authors review and compare FDA and EMA guidances and advocate for flexible regulatory approaches to data requirements for the use of MRD.
Although a rare disease, Multiple Myeloma (MM) is the second most common hematologic malignancy and is responsible for approximately 80,000 annual deaths worldwide.1 The estimated incidence of MM in 2012 worldwide was 114,000 patients, representing 0.8% of all cancers.2 The five-year prevalence of MM worldwide was estimated at 229,000 persons.3 Multiple myeloma is a disease afflicting older adults, with a median age at diagnosis of 69 years.4 Over the past 15 years, significant advances have been made in the treatment of MM with the regulatory approval of more effective novel agents with different Mechanisms of Action (MOA). These new treatments include alkylating agents, anthracyclines, corticosteroids, proteasome inhibitors, immunomodulators, monoclonal antibodies and histone deacetylase inhibitors. Several other novel agents with different mechanisms of action and clinical modalities are in development and will result in a crowded MM therapeutic landscape.
While available treatments typically induce remission, MM eventually relapses in most patients, causing a recurring pattern of remission and relapse.5,6 With each subsequent line of treatment, the patient’s prognosis worsens and relapse risk increases.7,8 Patients ultimately die due to MM-related complications, including monoclonal gammopathy, anemia, bone destruction, renal failure, hypercalcemia and increased susceptibility to infections. Despite recent advances, there remains a great unmet need to achieve long-term depth and duration of response in patients with MM.
FDA and EMA Endpoints Guidance Agreement
Currently, Progression Free Survival (PFS) is considered an appropriate primary endpoint for standard approval by both US Food and Drug Administration (FDA) and European Medicines Agency (EMA) for demonstrating clinically meaningful patient benefit in pivotal Phase 3 studies. However, with the aforementioned recent crowding of the treatment landscape in which new therapies demonstrate superior efficacy benefit over existing ones, the timeframe to attain clinically and statistically meaningful PFS benefit—when compared to approved therapies—has risen to well over five years. This beneficial outcome will likely be extended as newer and more innovative drugs are developed.9,10 However, this increased beneficial duration is a deterrent to innovation and can potentially prevent effective therapeutics from becoming available to patients, especially in the frontline MM setting where survival durations have been extended to over a decade. Delays in approving drugs that could improve patient outcomes are increasingly stifling innovation in new treatment options and, therefore, it is critical that the regulatory science evolve to accommodate the changing landscape of drug discovery in areas where traditional clinical trial approaches are no longer feasible.

One approach for preventing this is to use alternatives to currently used time-to-event variables so that novel investigational drugs can be evaluated earlier for efficacy. Minimal Residual Disease (MRD), used as a specific measure of tumor burden for MM at levels that are undetectable through conventional laboratory techniques, can potentially be used as a clinical and regulatory endpoint to evaluate a drug’s effect both on a patient’s future relapse risk and subsequent treatment survival outcomes. The International Myeloma Working Group (IMWG) has said that the value of MRD following treatment is the most relevant prognostic factor for relapse and long term survival.11 Further, recent data on patients with myeloma have supported the position that achieving MRD negativity after treatment is a strong predictor for both longer PFS and Overall Survival (OS) in patients newly diagnosed with MM, regardless of having received a transplant, their known myeloma risk profile or their disease stage at diagnosis.12 More compelling was the evidence that prolonged survival was observed in both high-risk and standard-risk myeloma patients who obtained MRD-negativity versus standard-risk MRD-positive patients.13

In July and October 2018 respectively, EMA and FDA issued draft guidances on the use of MRD as a clinical endpoint for drug development in MM. While FDA’s guidance covered the use of MRD in all hematological malignancies, the EMA guidance was more targeted toward MM. Both draft guidances discussed the use of MRD as a potential surrogate endpoint for accelerated (FDA) or conditional (EMA) approval for patients with MM.
FDA Draft Guidance Summary14
The two main takeaways from FDA’s draft guidance was the use of a meta-analysis for validation of MRD as a surrogate endpoint and MM-specific considerations for using MRD as an endpoint in clinical trials.

FDA elaborated on the evidence required to support the clinical validity of MRD as a surrogate endpoint for regulatory approval. While a meta-analysis strategy would be acceptable, FDA encouraged sponsors to discuss details of the meta-analysis plan with the agency. Key considerations for conducting a meta-analysis to validate MRD as a surrogate endpoint mentioned in the guidance included:
  • details of the trial designs (e.g., inclusion/exclusion criteria, disease setting, pool-ability of data)
  • inclusion of trials that include a patient population representative of the population in which the surrogate endpoint will ultimately be used
  • inclusion of an adequate number of randomized trials with sufficient follow-up time
  • inclusion of trials that demonstrated both positive and negative results
  • analysis based on individual patient-level data to allow an assessment of individual-level surrogacy
  • pre-specified criteria established based on trial-level and patient-level surrogacy measures
  • pre-specified timing and window of the MRD assessment
  • inclusion of long-term clinical outcome endpoints, such as Event-Free Survival/Progression-Free Survival (EFS/PFS) and OS that have been clearly and consistently defined across studies
  • discussion of missing MRD assessments and reasons for missing data
  • consideration of the statistical handling of unevaluable samples
  • potential confounding factors
  • sensitivity analyses to demonstrate the robustness of the surrogacy
  • discussion of different assay cutoffs (e.g., 10-4, 10-5). For assisting in the interpretation of the results, the sponsor can present analyses such as surrogate threshold effect.
Two important caveats were emphasized in FDA’s draft guidance. First, if a meta-analysis supports validation of MRD as a surrogate, applying these results to a new trial may not be applicable (e.g., applicability in subgroups, disease status, lines of treatment). Second, when a new molecule with a markedly novel MOA is under investigation the meta-analysis used to generate the initial MRD-PFS validation will be considered only supportive.
FDA acknowledged that while several trials have evaluated the relationship between MRD status and PFS/OS, additional MM-specific considerations should be incorporated into the design of future prospective MM trials. These include:
  • Relationship between MRD, clinical benefit and test performance characteristics must be demonstrated in each disease setting (e.g., Relapsed/Refractory (R/R), Newly Diagnosed Transplant-Ineligible and Transplant-Eligible settings (NDTI and NDTE), smoldering MM).
  • For MM, MRD should only be measured in patients with Complete Response (CR). MRD measurement in other response categories needs to be justified. However, in other disease settings outside of MM, MRD assessments should not be limited to MRD evaluable population; instead they should be based on intent to treat (ITT) population to support a regulatory submission.
  • Flow cytometry and Next Generation Sequencing (NGS) does not detect extramedullary (EM) disease; sponsor should discuss with FDA how EM disease will be assessed and if imaging should be incorporated.
  • MRD and clinical benefit in patients with different cytogenetics abnormalities and associated risks may have different outcomes independent of their MRD status.
  • FDA is neutral as to which technology platform is used in clinical trials assessing MRD; however, the sponsor should fully pre-specify the selected platform (in terms of assay procedure, reagents, and analysis) and analytically validate the platform for its context of use.
EMA Draft Guidance Summary15
Contrary to FDA guidance, the EMA’s guidance mentioned that early approval of drugs based on MRD as an intermediate endpoint may be considered on a “case-by-case basis.” However, confirmatory comprehensive data on PFS and OS from the same trial would need to be submitted. EMA also acknowledged that current available qualitative data are sufficiently convincing for MRD response rate to be used as an intermediate endpoint in randomized controlled trials so long as the benefit in terms of long-term efficacy outcome can eventually be confirmed. Key MM-specific considerations noted by EMA while conducting prospective MM trials using MRD as primary endpoint included:
  • A randomized, controlled trial is required, one that should be prospectively powered for PFS; all patients followed for OS.
  • MRD and PFS should be pre-planned and described in a Statistical Analysis Plan (SAP) and treatment effect will need to be estimated.
  • Difference in undetectable MRD response rate between study arms should be large enough to assume clinically meaningful PFS benefit on mature data.
  • Measurement of MRD should be conducted after each treatment stage: at the time of suspected response (Partial Response (PR), Very Good Partial Response (VGPR), Complete Response (CR) or stringent Complete Response (sCR)) following induction treatment and 100 days after ASCT in patients who receives transplantation. Patients with missing MRD assessment/ detectable MRD will be counted as MRD non-responders.
  • MRD will be considered undetectable if the proportion of malignant cells in the bone marrow is < 10-5.
  • MRD eradication needs to be confirmed in the extramedullary compartment.
  • Sustained undetectable MRD would be defined as undetectable MRD in patients in CR and with normal imaging that has lasted a minimum of one year.
Variation in FDA and EMA Draft Guidances
The most significant difference between the guidances provided by the two agencies is that the EMA will consider marketing applications using MRD as primary endpoint for early approval (i.e., conditional approval) on a case-by-case basis. By contrast, FDA’s draft guidance provides extensive guidance on the meta-analysis required to validate MRD as a surrogate endpoint for MM. However, FDA’s guidance did not address the considering approval based on significant improvement in MRD through its accelerated approval pathway.
While both agencies generally accepted that PFS data will be expected within the trials where MRD is used as a primary endpoint, only EMA specified that long term OS follow up would be necessary. Both FDA and EMA also highlighted that an analytically validated platform will be required, but generally they are neutral to which technology platform is used in assessing MRD. FDA and EMA agreed that tumor cell MRD should be confirmed in the extramedullary compartment and should be studied through imaging as a secondary endpoint. The agencies also differed on the advice regarding the timing of assessments. While FDA recommended MRD measurement only for patients who achieve CR, EMA guidance specified that MRD measurement should be conducted at the time of any suspected response, including PR, VGPR, CR or sCR. Contrary to FDA, EMA guidance provided clearer expectations regarding MRD threshold required to establish MRD negativity in clinical trials. EMA stated that MRD will be considered undetectable if the proportion of malignant cells in the bone marrow is < 10-5. FDA did not provide guidance around the expectation of durability or duration of response; EMA stated that sustained undetectable MRD as a secondary endpoint is defined as undetectable MRD in patients in CR and with normal imaging that has lasted a minimum of one year. Table 1 summarizes the comparison of FDA vs EMA draft guidances.
It is important to note that not all available clinical MM data will be compatible with the extensive criteria outlined for consideration in FDA guidelines. Some of the required data do not currently exist and will likely not exist within the time frame to support development of new agents in the frontline setting. Therefore, while FDA has provided clear guidance, and also encouraged sponsors to discuss details of the meta-analysis prior to submission, more specific guidance would be useful on the minimum data requirements to support use of MRD as a surrogate endpoint for regulatory approval. Similarly, while the EMA draft guidance did not discuss details of a meta-analysis submission approach and general requirements about study design, results, endpoints and methods while designing clinical trials for MM, it would have been useful for EMA’s draft guidance to have provided further details regarding under what circumstances the agency would consider applications on a case-by-case basis for early approval of medicinal product based on MRD as an intermediate endpoint.
Regarding the meta-analysis approach detailed in the FDA guidance (although not all the proposed FDA criteria may be fully me), there is currently substantial literature with MRD data supporting surrogacy for MM from various meta-analyses and demonstrating the correlation of MRD negativity with better survival outcomes.16-18 Questions remain over whether MRD can best be advanced as a regulatory surrogate endpoint without perfect data.
A potential answer lies in the ability to apply FDA’s accelerated approval pathway to therapies utilizing MRD as a potential surrogate endpoint. Surrogate endpoint, in the context of accelerated approval, is noted as a laboratory measurement, radiographic image, physical sign or other measure that is thought to “reasonably likely predict a real clinical benefit.”19 The current availability of literature supporting the idea that MRD is reasonably likely to predict clinical benefit is compelling from a clinical perspective. However, in the best interest of patients with MM, some additional considerations regarding gaps might need to be addressed by sponsors for MRD to be accepted as a surrogate endpoint.
Future MRD Considerations
Even with the preliminary draft guidance from both agencies on how MRD can be evaluated or utilized as a surrogate/intermediate clinical endpoint, there remain several challenges regarding how MRD can be utilized as a clinically meaningful endpoint in MM.
Technologies for detecting MRD are rapidly evolving. While Next Generation Sequencing (NGS) has been widely incorporated in recent clinical trials as an analytically validated platform to measure MRD, platforms capable of detecting MRD to a greater sensitivity than NGS may become available in the future. In addition, given the invasive nature of bone sampling for MRD measurement, the ultimate goal for measuring MRD would be to sample peripheral blood for disease detection. With changes in assay technologies, the assessment of MRD and subsequent regulatory approval might prove to be a “moving target.”
The optimal timing for MRD assessment is still unclear. Typically, bone marrow is sampled for MRD as suspected CR, at six months, 12 months and at sustained MRD. However, this frame is more applicable to patients with newly diagnosed MM. For relapsed and refractory MM, the MRD assessment timings remain unclear.
Predicting MRD negativity treatment effect in molecules with marked different mechanisms of action and different modalities can be a challenge, especially if a strong correlation between MRD negativity and PFS/OS has not been established in molecules with different mechanism of action. However, the above question can be addressed over time when MRD is considered an acceptable surrogate endpoint for early approval of drugs in MM.
There is a pressing need to update regulatory science to include the use of MRD as a surrogate endpoint for early drug approvals. While both the EMA and FDA have acknowledged this need and have actively engaged in several public workshops and symposiums on MRD, EMA may consider product approvals based on MRD as primary endpoint on a case-by-case basis. However, challenges remain prior to broad regulatory acceptability of marketing applications based on MRD as a surrogate or intermediate endpoint for approval in MM. The issuance of the draft guidances by FDA and EMA further highlights the importance of the potential use of MRD as a surrogate endpoint for early licensure/approval, especially in the frontline MM setting. The ultimate goal would be to have flexible regulatory approaches to data requirements for the use of MRD as a surrogate endpoint to support accelerated or conditional approval in products for newly diagnosed or relapsed and/or refractory MM. This would not only stimulate new drug development in this space, but would also bring potentially efficacious drugs quicker to patients with debilitating MM.
  1. Ferlay J, et al. “Cancer Incidence and Mortality Worldwide: Sources, Methods and Major Patterns in GLOBOCAN 2012.” International Journal of Cancer, 2015. 136(5): p. E359-E386.
  2. Ibid.
  3. Surveillance, Epidemiology and End Results Program Cancer Stat Facts: Myeloma. Bethesda, MD: National Cancer Institute; 2018. Accessed 5 April 2019.
  4. Ibid.
  5. Laubach J, et al. “Management of Relapsed Multiple Myeloma: Recommendations of the International Myeloma Working Group.” Leukemia, 2015. 30: p. 1005.
  6. Jakubowiak AJ, et al. “A Phase 1/2 Study of Carfilzomib in Combination With Lenalidomide and Low-Dose Dexamethasone as a Frontline Treatment for Multiple Myeloma.” Blood, 2012. 120(9): p. 1801-1809.
  7. Op cit 3.
  8. Yong K, et al. “Multiple Myeloma: Patient Outcomes in Real-World Practice.” British Journal of Haematology, 2016. 175(2): p. 252-264.
  9. Kumar SK, et al. “Continued Improvement in Survival in Multiple Myeloma: Changes in Early Mortality and Outcomes in Older Patients.” Leukemia, 2013. 28: p. 1122.
  10. Pozzi S, et al. “Survival of Multiple Myeloma Patients in the era of Novel Therapies Confirms the Improvement in Patients Younger Than 75 Years: a Population-Based Analysis.” British Journal of Haematology, 2013. 163(1): p. 40-46.
  11. Kumar S, et al. “International Myeloma Working Group Consensus Criteria for Response and Minimal Residual Disease Assessment in Multiple Myeloma.” The Lancet Oncology, 2016. 17(8): p. e328-e346.
  12. Perrot A, et al. “Minimal Residual Disease Negativity Using Deep Sequencing is a Major Prognostic Factor in Multiple Myeloma.” Blood, 2018. 132(23): p. 2456-2464.
  13. Ibid.
  14. FDA Guidance for Industry. Hematologic Malignancies: Regulatory Considerations for use of Minimal Residual Disease in Development of Drug and Biological Products for Treatment. FDA website. Accessed 5 April 2019.
  15. EMA Guideline on the use of Minimal Residual Disease as a Clinical Endpoint in Multiple Myeloma Studies. EMA website. Accessed 5 April 2019.
  16. van de Velde HJK, et al. “Complete Response Correlates With Long-Term Survival and Progression-Free Survival in High-Dose Therapy in Multiple Myeloma.” Haematologica, 2007. 92(10): p. 1399-1406.
  17. Munshi NC, et al. “Association of Minimal Residual Disease With Superior Survival Outcomes in Patients With Multiple Myeloma: A Meta-Analysis.” JAMA Oncology, 2017. 3(1): p. 28-35.
  18. Landgren O, et al. “Role of MRD Status in Relation to Clinical Outcomes in Newly Diagnosed Multiple Myeloma Patients: A Meta-Analysis.” Bone Marrow Transplantation, 2016. 51(12): p. 1565-1568.
  19. FDA Guidance for Industry Expedited Programs for Serious Conditions–Drugs and Biologics. 2014.
About the Authors
Telly Chi has more than 17 years of hands-on global regulatory experience. He has led teams from early to late development, through post launch and throughout lifecycle management. Chi earned an MS in regulatory science and a PharmD from the University of Southern California. He also completed a Post-Doctoral Pharmaceutical Industry Fellowship at Rutgers University. Chi holds the RAC (US) and can be contacted at
Aditi Sinha, PhD, is a manager at Amgen, Inc., with previous experience at Genentech, Inc. and Canary Medical, Inc. He has been an active leader in a wide range of responsibilities, including regulatory submissions, health authority interactions and regulatory strategy and compliance. He holds a PhD in bioengineering from Clemson University and can be contacted at
Cite as: Chi T and Sinha A. “Minimal Residual Disease as a Surrogate Endpoint for Product Development for Multiple Myeloma: Comparing FDA and EMA Guidances.” Regulatory  Focus. April 2019. Regulatory Affairs Professionals Society.


© 2022 Regulatory Affairs Professionals Society.

Discover more of what matters to you

No taxonomy
No data found