New data have come forward pointing to a link between achieving undetectable minimal residual disease (MRD) and improved treatment outcomes in multiple myeloma. The FDA appears receptive to the idea of using the measurement to support accelerated approvals of new drugs, even though the agency has a few questions.
In one meta-analysis conducted by a team led by Carl Ola Landgren, M.D., Ph.D., from the Sylvester Comprehensive Cancer Center, researchers showed that newly diagnosed multiple myeloma patients who achieved MRD negativity at 12 months were about 4.7 times more likely to be alive without disease progression than those who did not. Similarly, MRD-negative patients were about 4 times more likely to be alive, regardless of disease progression status.
In a similar meta-analysis performed by the i2TEAMM, a group brought together by the International Myeloma Foundation, the progression-free survival (PFS) odds ranged between 9 times to nearly 12 times in favor of MRD negativity after 12 months in newly diagnosed patients.
After performing two different statistical analyses, the i2TEAMM said that the data showed “very strong associations” between MRD negativity and the conventional PFS marker across various treatment line settings.
The meta-analyses are based on findings from prior multiple myeloma trials and are designed to examine whether MRD negativity can be linked to positive patient outcomes, regardless of the therapy being studied.
All of these data were published (PDF) by the FDA Wednesday and will be discussed this Friday at an advisory committee meeting. During the meeting, oncology experts will weigh in on using MRD as a surrogate endpoint in clinical trials to support accelerated approvals of multiple myeloma drugs.
The researchers apparently believe their data support the establishment of a new regulatory practice. Although data didn’t show MRD could fully replace PFS in phase 3 trials, they support that “treatment effects on MRD (tested at a sensitivity of 10-5 or better) are reasonably likely to predict treatment effects on PFS,” the i2TEAMM said in its conclusion.
“Based on the results observed in this meta-analysis of multiple large randomized studies, i2TEAMM believes there is sufficient evidence to support use of MRD as an endpoint for accelerated approval, with PFS maintained as a long-term endpoint for confirmation of clinical benefit,” the team added.
The FDA seems to agree.
“The available data appears to support the use of MRD as an intermediate clinical endpoint in MM clinical trials to support accelerated approval across different disease settings,” including newly diagnosed transplant-eligible and -ineligible, as well as relapsed or refractory patients, the FDA reviewers wrote in a document prepared for Friday’s meeting.
Should the FDA eventually adopt MRD as an early endpoint for accelerated approval, new drugs could be made available for myeloma patients much faster, Landgren said in an interview with Fierce Pharma ahead of the data release.
MRD measures disease burden by using next-generation sequencing or flow cytometry to search for cancerous clones in cells. The overall response rate trial endpoint, which measures tumor shrinkage, has been utilized to review myeloma drugs in late-line settings, and PFS is deemed appropriate for first-line approvals. MRD represents an even deeper level of tumor clearance than a complete response.
Why MRD-based reviews could benefit patients
Myeloma researchers have made significant advances over the years, with new therapies prolonging newly diagnosed patients’ median life expectancy from less than four years in the late 1990s to more than 10 years today, the FDA noted.
Nevertheless, even more efficacious therapies are needed because the disease is not curable. As it stands, the five-year relative survival rate is 60%, according to the FDA.
However, as existing standard-of-care drugs already help stave off cancer progression for a median of seven years, it takes newer therapies a long time to prove that they can add even more benefits based on the traditional PFS endpoint.
At the American Society of Hematology annual meeting in December, Peter Voorhees, M.D., from Atrium Health, proposed a hypothetical phase 3 trial that pits a bispecific antibody against an emerging standard combination. To be statistically powered to show a PFS benefit, the trial would take nearly 10 years to reach the final PFS analysis.
In contrast, an MRD readout could come out about three years into a trial in the first-line setting, Landgren estimated.
The Landgren-led team’s study, coded EVIDENCE, reviewed eight randomized, controlled clinical trials involving 5,130 patients with newly diagnosed multiple myeloma who were either eligible or ineligible for stem cell transplants. The studies met several criteria, including assay sensitivity to be able to detect MRD clones at 1 in 100,000 (10-5) cells or better.
The i2TEAMM analysis included 20 trials covering a total of nearly 13,000 patients. Again, this team also used several criteria to exclude studies that were non-randomized, had too small of a sample size, were published before 2006, or were excluded for other reasons. This analysis included studies with MRD sensitivity at just 10-4, but the data reported above only included trials with an MRD sensitivity at 10-5 or better.
One major difference between the two studies is that the i2TEAMM tossed out patients with missing MRD data, while the Sylvester one assigned these patients to be “MRD positive.” This could explain the difference in the two studies’ odds ratio numbers, as the Sylvester approach potentially counted some negative patients as positive.
For its part, the FDA conducted its own analysis based on data from both studies.
“In general, FDA agrees with the results presented by the applicants,” the agency said in the Wednesday document.
Still, the agency pointed out several limitations of the studies. For example, likely because of timing, trials of CAR-T therapies were not included in the meta-analyses. In addition, assessment times and follow-up durations varied across the trials.
Plus, because the trials need to meet some prespecified criteria to be included in the meta-analysis, the results may not reflect the relationship between MRD and PFS in other trials that didn’t meet the bar.
The FDA's own work found a “weak-to-moderate” association between 12-month MRD negativity and PFS—and “no” association between MRD and OS—in its trial-level analyses, despite “strong” patient individual-level correlations.
Nevertheless, the FDA, while acknowledging a risk that improvement in MRD may not predict long-term clinical benefit, noted that this is a risk with the use of any early endpoint. To mitigate the risk, early information on PFS and OS at the time of an accelerated approval would be key, the agency said.
Of course, the FDA would retain the authority to withdraw a product from the market if clinical benefit is not verified later, it added.
“While there are still outstanding questions on how to best use MRD, the meta-analyses conducted represent robust assessments of MRD that support its prognostic value, provide information regarding the appropriate timing of MRD assessment, and suggest that MRD may be appropriate to use as an intermediate clinical endpoint to support [accelerated approval],” the FDA wrote.