CHICAGO – Tumor protein 53 (TP53) mutation, fibroblast growth factor receptor 1 (FGFR1) amplification, and tumor purity in plasma each predict early progression on palbociclib and/or fulvestrant in patients with advanced estrogen receptor–positive (ER+) breast cancer, according to genomic analyses of PALOMA-3 trial data.
Sharon Worcester/MDedge News
Dr. Ben O'Leary
Overall, the presence of one or more of these genomic changes identified 131 out of 310 patients from the phase 3 trial who had baseline samples available, Ben O’Leary, MBBS, said at the annual meeting of the American Society of Clinical Oncology.
“So, a significant minority of patients – 42.3% – potentially who fall into a more poor-prognosis group,” said Dr. O’Leary of the Institute of Cancer Research at the Royal Marsden Hospital in London.
The findings suggest that a “liquid biopsy” at the start of treatment could identify patients at risk for progression.
The PALOMA-3 trial randomized 521 patients with ER+, human epidermal growth factor receptor 2 (HER2)–negative advanced breast cancer who had previously progressed on endocrine therapy 2:1 to CDK4/CDK6 inhibition with palbociclib plus fulvestrant (P+F) or placebo plus fulvestrant (F), and it showed that adding palbociclib significantly improved progression-free survival (PFS) (N Engl J Med. Jul 16 2015;373:209-19).
For the current analysis, the investigators assessed circulating tumor DNA (ctDNA) in baseline plasma samples from 459 study participants in an effort to identify genomic biomarkers for progression, to examine the association between baseline tumor fraction and clinical outcome, and to explore differences in predictive markers by treatment arm. A custom amplicon-sequencing analysis was performed to look for mutations in 17 different relevant genes, and another was used to estimate tumor fraction by looking at about 800 common germline single-nucleotide polymorphisms and to assess copy-number gain in the amplification status in 11 different genes, Dr. O’Leary said.
Results for mutations and circulating nucleic acids were available in 203 and 107 patients from the P+F and F groups, respectively, and on multivariable analysis of all 310 patients (including palbociclib as a variable in the model and with ctDNA fraction as a continuous variable), higher baseline tumor purity in plasma was associated with highly significantly worse PFS (HR 1.2 per 10% increase in purity), and baseline TP53 mutation and FGFR1 amplification each were associated with significantly shorter PFS (HRs, 1.8 and 2.9, respectively).
“[It is] very important to note ... that we did look specifically for interaction between our genomic changes and treatment, and we didn’t find any evidence of a significant interaction, so these genomic markers [are] prognostic rather than predictive in terms of the two treatment arms of the trial,” he said.
A survival analysis showed a median PFS of 3.7 vs. 12.7 months in patients with vs. without TP53 mutation in the P+F arm, and 1.8 vs. 5.4 months, respectively, in the F arm, with similar HRs of 2.0 and 2.3 in the arms, respectively.
“Even in the [P+F] arm, you see almost half of the patients with a TP53 mutation ... have relapsed by 2 months, the earliest clinical assessment in the trial,” he noted.
For FGFR1, the PFS was 3.9 vs. 12 months with vs. without amplification in the P+F arms, and 1.8 vs. 5.8 months, respectively in th F arm, with HRs of 3.4 and 3.6, respectively.
These findings are notable because markers of early progression on endocrine therapy in combination with CDK4/6 inhibitors remain limited – despite the key role of these combinations in treating ER+ advanced breast cancer, Dr. O’Leary explained.
“Although many patients derive a great deal of benefit from these combinations, there are a subset of patients who will relapse relatively early, and ... we don’t have an established means of identifying those patients at the present,” he said. “From the technical perspective, liquid biopsies have emerged in recent years as a promising means of genotyping patients’ cancers from circulating tumor DNA, and in addition, the overall level of circulating tumor DNA – the fractional purity – has been associated with poor prognosis, specifically in the triple-negative breast cancer setting.”
The results, which require independent validation, could potentially inform future clinical trials of CDK4/6 inhibitor combinations in advanced ER+ breast cancer to identify a high-risk group of patients who require escalation of therapy, he concluded.
Dr. O’Leary reported receiving research funding from Pfizer to his institution.
SOURCE: O’Leary B et al. ASCO 2019, Abstract 1010.