NEW YORK – Researchers working with Roche-affiliate Foundation Medicine and its assay development partner Natera have collected new evidence supporting the potential of the company's FoundationOne Tracker Assays to predict prognosis in lung cancer patients receiving chemo-immunotherapy (chemoIO), a first step toward developing personalized care strategies based on patients' individual risk.
Published in Clinical Cancer Research earlier this month, the study, led by investigators at the Moffitt Cancer Center, retrospectively analyzed samples from more than 200 non-small cell lung cancer (NSCLC) patients from the phase III IMpower131 trial, demonstrating that ctDNA testing during induction chemoIO could distinguish patients with more favorable versus less favorable outcomes.
Bruna Pellini, the study's first author and a Moffitt thoracic oncologist, said that chemoIO has become a prevalent first-line treatment for advanced, driver-negative NSCLC, with various forms of maintenance therapy given after induction. But there is significant variability in the duration, dosing, and timing of this follow-on maintenance.
"Much like in the surgical space, there is a question in chemo-immunotherapy treatment regarding whether everyone should be treated the same," she said. "When the trials were designed, the duration of maintenance therapy was chosen in an arbitrary way to be two years. But why two years? Why not one year? Why not three years? Why not forever?"
The hope is that the field can find a way to know which patients are on track for a better or worse outcome, so that some intervention might be deployed.
"For higher-risk patients, we would hope to develop ways to change their therapy. On the other hand, if somebody has a lower risk of disease progression, there is the potential to study de-escalation of therapy. Maybe these patients don't really need two years of therapy. Maybe one year is enough," Pellini said.
"We have a wealth of regimens that we can use, but we don't know the duration that is optimal for each patient and which treatment strategy is the best in this maintenance period."
Prior studies have shown that ctDNA monitoring is "an excellent prognostic biomarker," and this new work has reiterated that for chemo-immunotherapy.
In their new publication, Pellini and her colleagues utilized what Foundation Medicine calls FoundationOne Tracker, a tumor-informed, patient personalized assay strategy where upfront tumor tissue sequencing is used to select up to 16 mutation targets that are then combined into a panel for testing circulating cell-free DNA obtained from blood samples. They tested samples from 221 patients, with some having material from multiple time points.
A primary goal was to establish the ideal sampling time point to support the design of clinical trials comparing new strategies to the current standard of care. "What we demonstrated is that even if you only have one time point, which we chose as right at the beginning of the fourth cycle [of chemoIO], … it is enough to risk-stratify these patients," said Pellini.
Although induction regimens vary up to six cycles, this would mean that patients could be successfully tested even if they only undergo four, with enough lead time to have the results when next steps are being set in stone. It is also beneficial from a cost effectiveness standpoint, she added.
Overall, the study investigators found that a ctDNA decrease from baseline to day one of cycle four was associated with improved outcomes across multiple cutoffs for patients treated with chemoIO. Even including patients with missing plasma or those who were ctDNA-negative at baseline, a negative test at that fourth cycle time point was associated with better progression-free survival, with a median PFS of 8.8 months compared to 3.5 months for ctDNA-positive patients.
"Now the next step … is to design clinical trials where we are comparing the standard of care where everybody gets the same treatment to a personalized approach based on this pipeline," Pellini said.
She stressed the importance of recognizing that ctDNA monitoring assays are very reliable in terms of specificity, meaning if you get a positive result, you can feel confident that it is real. But the possibility of false negatives is real. "I think most of us feel comfortable with escalation because study after study, for decades now almost, have shown [that specificity]," she said. De-escalation, however, is harder "because you may have a negative test but it can be a false negative because the patient is not shedding DNA from their cancer cells at the moment," Pellini added.
"The way I explain this when I'm giving biopsies is that I think there are actually three risk groups," she said. "There is the low risk, which is somebody that is negative from the get-go, continues to be negative, and has amazing outcomes. Then you have the intermediate where they are positive and then treatment clears that ctDNA; and then you have the high risk who is positive and continues to be positive or is negative but becomes positive during treatment."
Pellini said that to make strides for that third group, it will be necessary to develop novel agents or combinations because "giving more of the same" has not shown to have a meaningful impact.
For example, in the CCR study, investigators did an exploratory analysis comparing patients that got four cycles of chemotherapy with immunotherapy versus patients that got six cycles. "We saw that with progression-free survival, it didn't change for the patients that got more chemo-immunotherapy," Pellini said, though she cautioned that that data was exploratory, featuring only 33 patients from the larger cohort.
According to Pellini, the Moffitt team and others are currently designing clinical trials that will use ctDNA assays to randomize patients to a specific escalated treatment — either a novel agent, or a new therapy combined with immunotherapy — but she said specific details are confidential.
"For us to really implement this in clinical practice we have to prove that changing what we're doing actually improves outcomes, or with de-escalating, that it does not affect their outcomes negatively, so if you asked me if I use this in clinical practice right now, I don't, because [we don't yet know] what to do with the results," Pellini said.
Foundation Medicine said in an email that it has been amassing additional data on the use of FoundationOne Tracker for late-stage cancer monitoring, as well as in residual disease detection after treatment of early-stage cancers.
Research published in Molecular Oncology recently featured results from the OMICO-MoST study, which showed that a decline in ctDNA as early as four weeks after treatment was predictive of improved overall survival in multiple tumor types treated with immunotherapy. Additionally, in two patients with complete response, ctDNA clearance, or complete removal of ctDNA, preceded any response by scan with a median lead time of 11.5 months.
Another study of patients in the IMvigor010 trial comparing immunotherapy treatment to observation after bladder cancer surgery showed that ctDNA-positive patients had improved disease-free survival and overall survival with atezolizumab versus observation.
FoundationOne Tracker is currently available through an early-access program to what the company called a limited number of clinical customers, and the firm has plans to expand to broad availability in the US later in 2023.
A burgeoning number of similar technologies has emerged in recent years, with tumor-informed products currently marketed in the US by Natera, Invitae, NeoGenomics, Myriad Genetics, and Personalis, among others, and tumor-naive, universal assays also gaining adoption from companies like Guardant Health.
Pellini said that it's not clear which ctDNA residual disease or monitoring technology is better than others, but she said her team has had better luck with tumor-informed versus agnostic technologies. The late-stage cancer monitoring space offers a slightly easier target for assays, she added, while minimal residual disease detection in early-stage tumors may be a setting where the field begins to see more differentiation between assays with smaller or larger target panels.
"In our situation, we chose this platform because it's in the guidelines that everyone with stage IV lung cancer should get comprehensive molecular testing, period," meaning that a technology that uses that standard of care as its upfront tumor-information source has already fought half the battle, Pellini said.
"I think because we're already planning to have that test done it's so easy to just do everything in one company if you're sending out."