According to the findings of the SAFIR02-BREAST study, presented at SABCS 2021, genomic profiling as a therapeutic decision tool for targeted therapy improves the outcome of patients with metastatic breast cancer if these patients carry alterations classified in the tiers I/II of the ESMO Scale for Clinical Actionability of molecular Targets (ESCAT). However, targeted therapies were found not to be effective when matched to alterations that did not rank as ESCAT I/II.
Genomic profiling is the profiling of the cancer’s genome in order to identify potential targetable alterations and treat the patient with matched targeted therapy. Thus far, it remained unclear whether profiling a high number of genes is effective in patients with metastatic breast cancer and how we can best interpret a genomic report. In order to address these questions, the SAFIR02-BREAST study was set up.
SAFIR02-BREAST study design
The SAFIR02-BREAST trial enrolled 1,462 patients with metastatic HER2-negative breast cancer. Patients who received more than one line of prior chemotherapy were not eligible to participate. When feasible, a biopsy was retrieved from their metastatic lesion. When fresh samples were not possible, a biopsy of less than 12 months old or a ctDNA sample was used in order to profile the cancer with next-generation sequencing (NGS) and a copy-number analysis. If patients achieved a complete response, partial response or stable disease after 6-8 cycles of chemotherapy, they entered the randomisation phase of the study.
The researchers performed a pooled analysis of this trial and the SAFIR-PI3K trial that compared a combination of the PI3Kα-specific inhibitor alpelisib and the oestrogen receptor antagonist fulvestrant with maintenance chemotherapy in patients with PIK3CA-mutated metastatic breast cancer. For patients who presented a targetable molecular alteration (N=238), randomisation was done between targeted therapy matched to genomics (N=157; olaparib, capivasertib, vistusertib, AZD8931, vandetanib, bicalutamide, AZD4547 or selumetinib in SAFIR02-BREAST or alpelisib/fulvestrant in SAFIR-PI3K) or maintenance chemotherapy (N=81). Genomic alterations in the patients’ tumours were classified using the ESCAT (ESMO Scale for Clinical Actionability of molecular Targets) scale, which ranks the likelihood of genomic alterations to serve as therapeutic targets, based on the strength of evidence from clinical studies. The investigators assessed the efficacy of the therapies in relation to their ESCAT ranking. A hierarchical testing was performed; first, the efficacy of targeted therapies matched to genomic alterations was tested in patients presenting an ESCAT I/II alteration. If a p-value of <0.1 was observed in the first step, analyses were then performed in the intent-to-treat population.
Among the 115 patients in whom an ESCAT I/II genomic alteration was found, the median PFS was 9.1 months in the matched targeted therapy arm and 2.8 months in the maintenance chemotherapy arm (HR[90%CI]: 0.41[0.27-0.61], p<0.001). In contrast, no significant difference in PFS was seen between the two arms in the overall population (HR[95%CI]: 0.77[0.56-1.06], p=0.109), and targeted therapies were not effective when matched to alterations that did not rank as ESCAT I/II. This showed that the ESCAT classification was highly predictive of the benefits of targeted therapies matched to genomic alterations (p=0.004).
Researchers also assessed which genetic copy number alterations were enriched in patients pre-treated with CDK4/6 inhibitors. Thirteen genes were found to be more frequently altered in patients pretreated with CDK4/6 inhibitors (ZMIZ1, FOXM1, AGR2, TACC1, CPNE3, ATG16L2, CDK4, LGR5, NFKBIA, CCL1, KCNG1, LIC00686 and NSL1). Interestingly, an amplification of CDK4 was observed in almost 10% of patients treated with CDK4/6 inhibitors. Finally, the researchers found that a loss of heterozygosity for BRCA or homologous recombination deficiency was associated with longer PFS in patients with a BRCA mutation treated with olaparib.
Genomic profiling has clinical utility in patients with metastatic breast cancer when driven by a validated framework of actionability (such as ESCAT). This implies that genomic reports must rank the genomic alterations according to a validated framework of actionability. In addition, oncologists should not administer a targeted therapy matched to a genomic alteration classified beyond ESCAT II, except in the context of exploratory therapeutic trials. Therefore, patients should be offered genomic testing to detect ESCAT I/II alterations.
André F, Goncalves A, Filleron T, et al. Clinical utility of molecular tumor profiling: Results from the randomized trial SAFIR02-BREAST. Presented at SABCS 2021; abstract GS1-10.