Minimal Disease, Maximum Benefit | Inside Precision Medicine

Minimal residual disease (MRD) testing is not new. Pioneered by scattered research groups more than 30 years ago, it spread rapidly from the early 2000s to the mid-2010s, when it was adopted as standard practice for the clinical monitoring of virtually all childhood and many adult acute lymphoblastic leukemia (ALL) cases. Early MRD testing methods consisted of flow cytometry and PCR-based testing, and focused largely on hematological malignancies.

But as both flow cytometry technology and genomic sequencing have taken technological leaps forward, the sensitivity and specificity of testing has increased to include MRD for solid tumors like non-small cell lung cancer (NSCLC) and breast, colorectal, prostate, melanoma, bladder, and pancreatic cancers. Today, leveraging advances in next-generation sequencing (NGS), MRD test developers can look for thousands of disease markers and even detect one cancerous cell among a million healthy cells. This level of sensitivity provides evidence of disease recurrence months before traditional follow-up screening methods.

Informing precision treatments

The benefits of MRD are difficult to ignore. Detection of remaining disease post-surgical resection or known disease recurrence can help inform adjuvant therapy and even prompt treatment many months before a recurrence is detected through imaging. Measuring disease burden can inform dosing levels, helping to avoid patient toxicity and serious side effects that often occur when a maximum tolerable dose is administered. As the specificity of MRD testing improves, there are now opportunities to pause or even stop treatment for patients who appear free of disease.

“Minimum effective therapy is what physicians want, it’s what patients clearly want, it’s what payers, obviously, will want,” said Susan Bobulsky, chief commercial officer, MRD, at Adaptive Biotechnologies, which developed an NGS-based MRD test called clonoSEQ® for blood cancers.

This testing will challenge the existing treatment regimen for cancers such as multiple myeloma, for which the current paradigm is “forever maintenance,” whereby patients remain on therapy until they die. “But in practical terms, it is unusual for a patient to be able to stay on therapy their entire lives. It is simply too toxic, too expensive,” she added.

Richard Chen, MD, chief medical officer of Personalis, agrees that MRD testing will be increasingly used for de-escalating therapy to mitigate the adverse effects of cancer treatments. He also pointed out that the increasing sensitivity of MRD testing can improve patient outcomes by allowing for significantly earlier treatment of cancer recurrence.

The Personalis MRD test, called NeXT Personal®, is a tumor-informed whole-genome liquid biopsy that can detect as many as 1,800 molecular markers of cancer and has been shown to detect breast cancer recurrence up to 15 months earlier than current methods. Although this advance intuitively suggests an improvement in patient outcomes by enabling earlier treatment, more evidence is needed before the test is widely used.

“To further push that adoption, more clinical utility studies are warranted,” Chen said. “I think what’s clear is that our technology does work. It does detect the cancer and detects it earlier than imaging. It could be prognostic. But now, if you can detect it earlier, and you treat earlier, does that result in better outcomes for the patients?”

As MRD testing continues to improve in sensitivity and range of applicable malignancies, clinical adoption will likely require inclusion in guidelines from the National Comprehensive Cancer Network to change the standard of care. This will likely spawn a multitude of clinical studies.

“Achieving guidelines and commercial coverage requires a sizable, compelling collection of well-designed clinical studies that show a clear benefit of MRD over the standard of care,” said Dale Muzzey, PhD, chief scientific officer of Myriad Genetics. “Because cancer is so diverse from one tissue to another and because the experts and guideline authors often are highly specialized, I suspect that each intended use (e.g., neoadjuvant de-escalation, adjuvant monitoring and so on) will need its own set of studies for each cancer.”

Tumor-informed vs. tumor-naïve testing

As solid tumor MRD testing ramps up, diagnostics developers must decide which approach they want to take with their testing platform: tumor-informed or tumor-naïve.

The tumor-informed approach seeks to provide a true precision medicine approach—a solid tumor sample typically undergoes whole-genome sequencing to determine the unique molecular markers of an individual’s cancer. This allows for the design of a targeted testing panel containing only markers unique to that patient. By providing blood samples, the patient can undergo serial testing to check whether their cancer has recurred.

A tumor-naïve approach does not rely on prior tumor testing. Rather, it searches for known cancer mutations to determine the tumor burden and measures the prevalence of circulating tumor DNA (ctDNA) in the blood to determine the presence or absence of disease.

Most diagnostic developers do not see the two approaches as competing, but as complimentary. They may choose one or the other based on each cancer patient’s circumstances.

Initially, tumor-naïve approaches can save weeks of time, from conducting the original test to finding the answer. “The ‘informing’ process takes time: the tumor must first be profiled and then patient-specific enrichment molecules must be designed and synthesized,” Muzzey noted. “Collectively, this ‘informing’ time takes four to six weeks before the tumor-informed MRD is ready to test a plasma sample for the presence of ctDNA.”

Because of this initial lag time, a tumor-naïve assay can make more sense in certain situations. For instance, a patient with late-stage metastatic disease does not have the luxury of time to select the most appropriate treatment—physicians want to know the molecular makeup of their cancer quickly to start treatments as soon as possible.

“Treatment response assessment in patients with metastatic disease is an unmet need clinically right now,” said Jake Chabon, PhD, CEO of Foresight Diagnostics. “There are a lot of different immunotherapy options out there. Identifying whether a patient is responding to their immunotherapy quickly and, if not, moving them to an additional therapy faster is one of the benefits [of tumor-naïve tests].”

Another application of tumor-naïve testing could be in late-stage cancers, where collecting a sample may pose a significant risk to the patient, or when the tumor itself is hard to access, obviating the use of a tumor-informed test.

“People often like to pit these two against each other, but in my mind, there are different scenarios for these technologies. I don’t think they’re mutually exclusive,” Chen said. “I think we’ll be in a world where they coexist, and happily, and there’ll be uses for one or the other technology, depending on the specific situation.”

MRD in clinical trials

Oncology clinical trials seek two essential truths: is the drug being tested safe and is it effectively fighting the tumor, i.e., effectively reducing the presence of disease? MRD testing is uniquely suited to determining the efficacy of treatments, while also helping to identify and stratify patients for inclusion in the trial.

At Foresight Diagnostics, its PhasED-Seq MRD platform is currently being used to determine patient eligibility for Allogene’s ALPHA3 trial, which is testing a first line consolidation treatment for large B-cell lymphoma. Foresight’s partnership with Allogene in this trial highlights the company’s positioning of its MRD testing as a vital tool to build evidence for specific treatments.

“In the ALPHA3 trial in patients with large cell lymphoma, they’re receiving curative intent chemotherapy upfront, and then they have an MRD test at the end of therapy, and that’s being used to determine whether they should get allogeneic CAR T cell treatment,” said Chabon. “We’re going to use that same approach in solid tumors as well— it would be guiding adjuvant therapy following a surgical resection or radiation or other curative-intent treatment regimens in patients with solid tumor indications.”

Perhaps the most significant development affecting the use of MRD in clinical trials occurred this spring, when the Oncologic Drug Advisory Committee (ODAC) of the FDA voted 12-0 to approve MRD as an endpoint in multiple myeloma (MM). While the FDA would still need to formally approve the committee’s recommendation, it appears to be merely a formality given the unanimous ODAC vote.

The ODAC recommendation did not happen overnight. It was the culmination of 15 years of work in developing better MRD testing techniques and, more importantly, engaging stakeholders (including the FDA), and convincing pharmaceutical companies to share their data from registrational studies to build the case for MRD as an intermediate clinical endpoint.

“It was a pretty big ask,” admitted C. Ola Landgren, MD, PhD, director of the Myeloma Research Institute at the Sylvester Comprehensive Cancer Center, University of Miami Miller School of Medicine, who spearheaded the effort. Landgren noted that he was able to leverage his position as a chairman or a member of several independent clinical data monitoring committees to begin conversations with pharma stakeholders.

“I had relationships with most of the big companies through that role,” Landgren said. “So I reached out to many of the leaders and said: ‘Look, you will never be able to go to the FDA with your datasets and ask for MRD to be an endpoint, because the FDA will ask you to provide data from other drugs as well. None of the other companies will give the data to you and you also will not give your data to another company, therefore, it will never happen.’”

The solution, Landgren realized nearly 15 years ago while working as a senior investigator and chief of the Multiple Myeloma Section at the National Cancer Institute, was for him to act as an independent third party. In that role, he provided guarantees to the companies that none of their data would be shared with the other companies involved. This work, called the EVIDENCE meta-analysis study, eventually featured information from a dozen Phase II or III clinical trials, and showed a tight correlation between MRD and clinical outcomes in both newly diagnosed and relapsed MM patients.

Key researchers in the study were Maryalice Stetler-Stevenson, MD, PhD, a flow cytometry expert who worked with Landgren to improve flow cytometry MRD assays; Gerald Marti, MD, a hematologist with the National Heart, Lung, and Blood Institute, who also had an appointment with the FDA; and Sean Devlin, PhD, from the Memorial Sloan Kettering Cancer Center, who served as the lead data scientist.

Currently, 40% percent of MM patients die within five years of diagnosis, pointing to the vital need for new therapies to treat this cancer. A driving force for Landgren’s work was a realization that the accepted standards for measuring drug responses for MM, and the identification of drugs with responses in a very high percentage of patients, could discourage drug developers. One of the key endpoints at the time—overall response—only required half of the tumor to shrink.

“We were getting close to 99% of our patients having that response,” Landgren noted. “It was getting very hard to develop a drug to show superiority.”

But the greater sensitivity of MRD proved to be a game changer. With MRD as a new endpoint in MM clinical trials, the EVIDENCE study has paved the way for preliminary new drug approvals via an accelerated pathway. New MM drug candidates will still need to exhibit progression-free survival (PFS) to achieve full FDA approval, but according to Landgren, accelerated approval using MRD will shave years off the standard delays in getting new treatments to patients.

“Without MRD, if you keep [using] PFS, it could take about 10 years for the data to mature,” he said. “I would say MRD represents a five-to-ten-year saving [in] getting the drugs to patients. We don’t yet have a cure for this disease, so we need a much faster timeline for new therapies.”

Chris Anderson, a Maine native, has been a B2B editor for more than 25 years. He was the founding editor of Security Systems News and Drug Discovery News, and led the print launch and expanded coverage as editor in chief of Clinical OMICs, now named Inside Precision Medicine.