NEW YORK – Researchers at the San Antonio Breast Cancer Symposium Thursday presented data demonstrating the ability of circulating tumor DNA testing to personalize treatment with CDK4/6 inhibitors and endocrine therapy in advanced breast cancer patients.
In the PADA-1 study, researchers led by François-Clément Bidard, head of the circulating tumor biomarkers laboratory at Institute Curie Hospitals, employed a novel first-line endocrine therapy "switch strategy" guided by "a molecular signal in the blood."
In that study, liquid biopsy monitoring for the presence of ESR1 ctDNA mutations in hormone receptor (HR)-positive, HER2-negative metastatic breast cancer patients allowed some to avoid resistance to aromatase inhibitor-based endocrine therapy while maintaining treatment with the CDK4/6 inhibitor palbociclib (Pfizer's Ibrance). The molecularly guided endocrine therapy switch strategy employed in PADA-1 may be a promising method for individualizing treatment in this setting, according to Bidard.
In the BioltaLEE trial, researchers led by Giampaolo Bianchini, a senior consultant at the San Raffaele Hospital and head of the Breast Cancer Group, evaluated ctDNA dynamics to try to identify best responders among HR-positive, HER2-negative advanced breast cancer patients on first-line treatment with the CDK4/6 inhibitor ribociclib (Novartis' Kisqali) and the aromatase inhibitor letrozole. This study showed that the presence of a detectable target mutation at baseline is a negative prognostic factor for these breast cancer patients. In patients with detectable mutations at baseline, ctDNA clearance upon ribociclib-letrozole treatment was associated with a lower risk of progression, while the continued presence of mutations on treatment was associated with a higher risk of progression.
In this analysis, researchers also established that testing patients' ctDNA at a specific time point in treatment — 15 days into the first cycle of ribociclib-letrozole — was the most meaningful when trying to predict which patients are likely to do best on treatment.
In this Phase III trial, researchers randomized HR-positive, HER2-negative metastatic breast cancer patients to either the combination of the CDK 4/6 inhibitor palbociclib plus the selective estrogen receptor degrader (SERD) fulvestrant or palbociclib with an aromatase inhibitor upon detection of circulating ESR1 mutations. Patients started off on PADA-1 receiving first-line treatment with an aromatase inhibitor and palbociclib, but as soon as ESR1 mutations were detected in blood ctDNA — but patients' tumors weren't progressing on imaging analysis — investigators switched out the aromatase inhibitor with the SERD while maintaining CDK4/6 inhibition.
Studies have shown that ESR1 mutations make estrogen receptor (ER)-positive metastatic breast cancer patients resistant to aromatase inhibitors, but breast tumors with these mutations continue to respond to SERDs. ESR1 mutations in blood are rare at diagnosis, showing up in less than 5 percent of metastatic breast cancer patients. However, these mutations occur in 30 percent to 40 percent of breast tumors that progress on a first-line aromatase inhibitor-based treatment.
Around 1,000 patients in PADA-1 started on an aromatase inhibitor-palbociclib regimen. Researchers used Bio-Rad Droplet Digital PCR-based cell-free DNA analysis to test patients for five ESR1 mutations upon inclusion in the trial, at one month, and then every two months. Based on this monitoring, Bidard described at the meeting how ultimately 172 patients with "rising" blood ESR1 mutations, but without disease progression on imaging, were randomized to either aromatase inhibitor-palbociclib or fulvestrant-palbociclib. Patients in the aromatase inhibitor arm were allowed to crossover to the fulvestrant arm upon progression.
In reviewing the data from PADA-1 at SABCS, Minetta Liu, a breast cancer specialist at the Mayo Clinic, noted that it's important to recognize that "rising" blood ESR1 mutations, as described by Bidard in his presentation, doesn't necessarily mean that mutations are quantitatively rising but just that ESR1 mutations were detected by ctDNA testing. "That's a distinction that needs to be made clear as we think about the data here," Liu said.
As of July 31, and a median follow-up of 34.5 months, investigator-assessed median progression-free survival was 5.7 months on the aromatase inhibitor-palbociclib regimen and 11.9 months on fulvestrant-palbociclib. "In the experimental arm, we observed more than a doubling of median progression-free survival with the early introduction of fulvestrant combined with maintenance of palbociclib," Bidard said. The absolute difference in median progression-free survival was 6.2 months, which he added was "very relevant from a clinical perspective."
There were 47 patients who crossed over from the aromatase inhibitor arm to the fulvestrant arm, and they had median progression-free survival of 3.5 months on second-line palbociclib-fulvestrant. This, according to Bidard, gives a sense of what the median progression-free survival could be if patients receive palbociclib-fulvestrant in the second-line setting after receiving an aromatase inhibitor-palbociclib regimen in the first line.
"PADA-1 is the first clinical trial of its kind, and may be not the last, to demonstrate the clinical utility of [blood ESR1 mutation] monitoring," Bidard concluded. "It allows for the personalization and optimization of the endocrine therapy partner for CDK4/6 inhibitors in first-line."
Bidard recognized that the observed clinical benefit in the study might be due to patients having a lower ESR1 mutation tumor cell burden when they started fulvestrant. Still, these results, he believes, justify incorporating this ctDNA-based treatment switching strategy in routine care.
Liu agreed that switching patients' endocrine treatment based on molecular indicators of progression, as opposed to overt progression on imaging, led to improved outcomes in patients. "These findings support serial ctDNA monitoring as a means to identify and react to treatment resistance as soon as possible," she said, but noted that it remains to be seen whether this strategy will translate to a meaningful survival benefit in patients.
Moreover, the progression-free survival benefit seen in the trial must be balanced by quality of life and resource considerations related to using oral aromatase inhibitors versus intramuscular injection of fulvestrant.
There is work ongoing, Bidard said, to refine understanding of blood ESR1 mutation kinetics, mutant allelic fraction, and different ESR1 mutation types using ctDNA analysis. "At a higher level we also believe that monitoring the rise of resistance-associated mutations could open up new opportunities for cancer therapy," such as with next-generation SERDs in metastatic breast cancer and in other clinical settings, beyond blood ESR1 mutations, Bidard said.
In discussing the rationale for BioltaLEE, Bianchini noted at the meeting that three Phase III MONALEESA trials have previously demonstrated that ribociclib significantly improves progression-free and overall survival in HR-positive HER2-negative advanced breast cancer patients regardless of the type of endocrine therapy given, the line of prior treatment, or patients' menopausal status. However, there has been limited biomarker research to try to better distinguish responders from non-responders to ribociclib and some of these efforts have yielded inconclusive results, he said.
In BioltaLEE, Bianchini and colleagues used ctDNA testing to try to characterize tumor biology, describe its evolution over time, and predict patients' prognosis and treatment response. "The primary objective of this study is to identify ctDNA alterations, characterize how they evolve during treatment, and evaluate their possible association with clinical outcome," Bianchini said.
Researchers enrolled 287 patients who underwent blood and tumor sample collection at baseline and then received liquid biopsy testing on day 15 of the first cycle of ribociclib-letrozole treatment, and again on the first day of the second cycle of treatment. Imaging was performed at around three months after starting treatment.
Bianchini's group conducted ctDNA analysis using a 533-amplicon custom Thermo Fisher Scientific AmpliSeq HD Panel that covered the coding exons of 39 breast cancer-associated genes. Target mutations detected at baseline comprised single-nucleotide variants or indels. When researchers identified multiple target mutations in a patient, they considered the mutation with the highest variant allele frequency, and they deemed a patient to have variant allele frequency clearance when there was a 100 percent decrease in presence of the target mutation.
Out of 287 enrolled patients, 263 provided baseline samples for liquid biopsy analysis. Researchers were also able to collect blood samples from around 91 percent of patients by day 15 of the first treatment cycle and from 92 percent of patients on the first day of the second treatment cycle; 76.4 percent of patients were imaged at around three months.
Bianchini reported that at baseline target mutations were detected in 113 patients, or 43 percent, while 150 patients, or 57 percent, had no detectable mutations. The median progression-free survival in the wild-type group was not reached at a median follow-up of 26.9 months, and 16.6 months in the group with target mutations detected at baseline. This, according to Bianchini, translates to a 59 percent decrease in the risk of progression for those with no detectable baseline mutations.
Once patients started on treatment, "a strong and robust ctDNA dynamic was observed during the first cycle," he said, noting that early variant allele frequency clearance was observed in 49 patients, or 47 percent, at day 15 of the first cycle of treatment. Variant allele frequency clearance was observed in 55 patients, or 52 percent, on the first day of the second cycle of treatment.
When researchers compared patients who had baseline target mutations but achieved variant allele frequency clearance at day 15 of the first treatment cycle with those who didn't achieve this clearance, they found a statistically significant improvement in median progression-free survival: 21.9 months versus 12.1 months, respectively. Bianchini noted that ribociclib-letrozole led to variant allele frequency clearance in around 50 percent of patients as early as two weeks and "this pharmacodynamic modulation of ctDNA was associated with a lower risk of progression, suggesting predictive value."
Among 150 patients who didn't have detectable mutations at baseline, 34 patients, or around 23 percent, had detectable mutations at later time points in the study. "This group, overall, had a lower risk of progression" compared to those with detectable baseline mutations, Bianchini said, but still fared worse than those who continued to be mutation free throughout the study. The median progression-free survival was not reached in those who didn't have post-baseline mutations versus 16 months for those with detectable mutations at any post-baseline time point.
He noted that both the lack of mutations at baseline and at day 15 of the first cycle provide independent predictive information as to which advanced breast cancer patients are likely to benefit from ribociclib-letrozole. Patients who didn't have mutations at either time point had the best outcomes in the study, while patients with detectable mutations at both time points had the worst outcomes on ribociclib-letrozole.
In reviewing the data from BioltaLEE, Liu from the Mayo Clinic observed that patients who didn't have a target mutation at day 15 "had an extremely favorable outcome," and that this appears to be the "most informative" single time point for ctDNA evaluation. The study results also raise the question, she continued, as to whether it's best to just look at day 15, or compare ctDNA results at this time point with results at baseline, since the presence or absence of baseline tumor mutations could allow for further stratification of patients in terms of predicting treatment outcomes.
Liu further pointed out that the study tracked the emergence or disappearance of a single alternation in ctDNA, but patients' outcomes may still be impacted by other acquired mutations that weren't tested for. "Were those ctDNA mutations themselves clinically significant?" she wondered. "The specifics may be critical to our understanding."
This study provides further validation of the prognostic power of ctDNA testing and demonstrates predictive potential, experts agreed. "In a broader perspective, in [the context of] other targeted therapies and immune checkpoint inhibitors, the early disappearance of ctDNA has been an early indicator of the activity of the drug," Bianchini said. "If we look at the findings [in this study] with that general perspective, it's quite expected that this information is able to describe an effect of the drug."
Further study is warranted, Bianchini said, of both pre-treatment and early treatment ctDNA dynamics as a prognostic and predictive biomarker in patients with HR-positive, HER2-negative advanced breast cancer treated with first-line ribociclib and letrozole.