Supported via arm's length funding from Gilead.

Highlights from

SABCS 2021

San Antonio Breast Cancer Symposium

San Antonio, Texas, 7 - 10 December 2021

Using genomic profiling to match treatment improves survival outcomes for certain patients with metastatic breast cancer

Take-home messages
  • Genomic profiling has clinical utility in patients with HER2-negative metastatic breast cancer (mBC)
  • A genomic landscape of mBC can be deduced using genomic profiling
  • New genomic markers of drug sensitivity can be identified using genomic profiling
"In conclusion, genomic profiling has clinical utility in patients with metastatic breast cancer only when the decision to administer targeted therapy is validated by a framework of actionability."

Professor Fabrice André, Research Director at Gustave Roussy Cancer Campus, Paris, France

A unique study aiming to determine if genomic profiling can be clinically utilised in patients with mBC was presented at this year's San Antonio Breast Cancer Symposium (SABCS). "I really cannot overstate how important this clinical trial is. This is the first trial that we have where we have used genomic alterations to actively change a patient's treatment plan and we have found an improvement in the patients' outcomes." These were the words SABCS co-director Dr Virginia Kaklamani used to describe the results of the SAFIR02-BREAST trial at a press conference the day before SABCS kicked off on 7 December.

The results of the phase II, randomised, open-label SAFIR02-BREAST trial were presented by Professor Fabrice André, Research Director at Gustave Roussy Cancer Campus, Paris, France. The trial included patients with HER2‑negative mBC (N=1,462). Patients were deemed eligible if they were resistant to endocrine therapy and hormone receptor (HR)-positive, or if they had previously been treated with no or one line of chemotherapy.

Tumour profiling of patients was carried out on biopsy samples using next generation sequencing and single nucleotide polymorphism arrays. Patients whose tumour profiling was successful received 6-8 cycles of chemotherapy. Only those who responded well to chemotherapy entered the randomisation phase of the study. Of these patients, 646 presented with a targetable molecular alteration. This group was randomised 2:1 to targeted therapies that matched genomic alterations or maintenance chemotherapy. The targeted therapies patients received were vistusertib, AZD4547, capivasertib, sapitinib, selumetinib, vandetanib, bicalutamide, olaparib or alpelisib.

The trial assessed treatment efficacy in relation to patients' ESCAT rankings and determined if targeted therapies matched to genomic alteration improved progression-free survival (PFS) in patients who were ranked ESCAT I or ESCAT II (n=115) and in the intention-to-treat (ITT) population (n=238), compared with maintenance chemotherapy.

ESCAT classification is a framework that ranks the likelihood of genomic alterations to serve as therapeutic targets based on the strength of evidence from clinical trials. Genomic alterations are classified as ESCAT I when their match with the therapy received is associated with improved outcomes in clinical trials. ESCAT II is the ranking when the alteration-drug match is associated with antitumour activity at an unknown magnitude of benefit.1

In the ESCAT I/II population, median PFS was 9.1 months in patients who received targeted therapy matched to genomic alteration versus 2.8 months in those who received maintenance chemotherapy (HR: 0.41, 95% CI: 0.27‑0.61, p<0.001).

In the ITT population, median PFS was 5.5 months in the study arm in which targeted therapy was matched to the genetic alteration, compared with 2.9 months in the maintenance chemotherapy arm (HR: 0.77, 95% CI: 0.56-1.06, p=0.109). Subgroup analysis in this population showed ESCAT I/II ranking provided predictive value for targeted therapies (HR: 0.48, 95% CI: 0.31-0.75).

The conclusion Professor André drew from these results was "genomics can improve outcomes in patients with metastatic breast cancer only in patients with ESCAT I/II genomic alterations."

Professor André and colleagues addressed two other questions in the SAFIR02-BREAST trial: "what is the genomic landscape of metastatic breast cancer?" and "can we identify new genomic markers for drug sensitivity?"

Analysis of the genomic landscape was specific to patients who had previously been treated with CDK4/6 inhibitors (n=102) and it identified 13 genes that were altered in patients who had previously received CDK4 inhibitors. "Interestingly, one of these gene alterations was amplification of CDK4 that was observed in around 10% of patients, [they were] pretreated and became resistant to CDK4 inhibitors" remarked Professor André.

As for the last question of identifying new genomic markers, the results showed that median PFS was longer in patients who presented loss of heterozygosity (LOH) for BRCA1/2 than those with no LOH for BRCA1/2 when they were treated with olaparib (11.7 months versus 5.5 months in patients with LOH versus no LOH, HR: 0.32, 95% CI: 0.14-0.73, p=0.005).

At the end of his presentation Professor André concluded "genomic profiling has clinical utility in patients with mBC only when the decision to administer targeted therapy is validated by a framework of actionability [such as ESCAT and OncoKB], meaning that genomic reports MUST rank the genomic alterations according to this framework of actionability."

The implications of this conclusion, according to Professor André, are that an oncologist should not administer a targeted therapy matched to a genomic alteration classified beyond ESCAT II, except in the context of exploratory therapeutic trials, and patients should be offered genetic testing to detect ESCAT I/II alterations.

The limitations of the trial were that it did not assess the technology used in genomic profiling and could not directly assess the impact of genomics in the overall population of patients because only patients presenting with a genetic alteration were randomised.

In response to a question in the Q&A session, Professor André disclosed that the ratio of ESCAT I patients to ESCAT II patients in the study was 1:1, which would not be expected to cause any bias in the results.

CDK: cyclin-dependent kinase; CI: confidence interval; ESCAT: ESMO Scale for Clinical Actionability of molecular Targets; ESMO: European Society for Medical Oncology; HER2: human epidermal growth factor receptor 2; HR: hazard ratio

Based on:
André F, Gonçalves A et al. Clinical utility of molecular tumor profiling: results from the randomized trial SAFIR02-BREAST (abstract GS1-10). Presented on Tuesday 7 December 2021

1. Mateo J, Chakravarty D et al. A framework to rank genomic alterations as targets for cancer precision medicine: the ESMO Scale for Clinical Actionability of molecular Targets (ESCAT). Ann Oncol 2018;29(9):1895-1902

Top image: rustycloud

Article image: filo

The content and interpretation of these conference highlights are the views and comments of the speakers/ authors.

Supported by