By analyzing progression-free survival (PFS) as a primary endpoint in a pooled analysis of SAFIR02-BREAST and SAFIR-PI3K with hierarchical testing, starting with the population of patients with metastatic breast cancer and ESMO Scale for Clinical Actionability of molecular Targets (ESCAT) tier I/II alterations followed by the overall population, a group of French investigators led by Prof. Fabrice Andre of the Gustave Roussy in Villejuif and Université Paris Saclay in Kremlin Bicetre, France and colleagues reported on 7 September 2022 in the Nature that the use of genomics improves the outcome of patients who present with a match drug/alteration ESCAT I/II, but not for those presenting with ESCAT beyond these levels. The authors argue that reporting the results of genomics in the context of a framework of target actionability should therefore be considered as a standard of care. However, the results should be interpreted with caution, because a large part of the benefit observed with matched targeted therapies was derived from patients presenting with BRCA1/2 alteration and the small sample size does not allow exploration of the actionability of new genomic alterations.
The SAFIR02-BREAST is a prospective randomised study that compared targeted therapies matched to genomic alterations with the standard of care, maintenance chemotherapy in patients with metastatic breast cancer. Between April 2014 and October 2019, 1,462 patients with HER2-negative metastatic breast cancer signed informed consent to perform a biopsy and/or a genomic test in SAFIR02-BREAST and were included in the study. Among 1,462 screened patients, 646 presented with a targetable genomic alteration. Subsequently, 238 patients (16%) were randomised between maintenance chemotherapy (81 patients) and targeted therapy (157 patients).
From December 2017, a specific protocol (SAFIR-PI3K) was opened for patients with hormone receptor-positive, HER2-negative metastatic breast cancer presenting with a hotspot PIK3CA mutation detected in SAFIR02-BREAST. This protocol randomised alpelisib combined with fulvestrant versus chemotherapy.
The first step in the statistical plan was to analyse the efficacy of targeted therapies in patients presenting with an ESCAT I/II genomic alteration. Of the 115 patients with a genomic alteration classified as ESCAT I/II, 57 presented with BRCA1/2 alterations, 3 with a PALB2 alteration, 31 with a PIK3CA mutation, 16 with an AKT1 mutation, 5 with a PTEN mutation and/or deletion and 3 with HER2 mutation; 8 patients presented with a somatic-only BRCA1/2 genomic alteration.
Targeted therapies matched to genomics improved PFS when genomic alterations are classified as level I/II according to ESCAT (adjusted hazards ratio [HR] 0.41, 90% confidence interval [CI] 0.27–0.61, p < 0.001), but not when alterations are unselected using ESCAT (adjusted HR 0.77, 95% CI 0.56–1.06; p = 0.109). No improvement in PFS was observed in the targeted therapies arm (unadjusted HR 1.15, 95% CI 0.76–1.75) for patients presenting with ESCAT alteration beyond level I/II. A total, 49 patients with germline BRCA1/2 mutations derived high benefit from olaparib (germline BRCA1, HR 0.36, 90% CI 0.14–0.89; germline BRCA2, HR 0.37, 90% CI 0.17–0.78).
The authors commented that their study shows that the tools for clinical interpretation of sequencing are pivotal in achieving tangible clinical benefits. In the current study they tested ESCAT as a framework for actionability and found that DNA sequencing should lead to treatment administration only for patients presenting with genomic alterations ranked as level I/II.
The authors acknowledged several limitations of their study. It did not test a specific platform or technology of sequencing, but rather the utility of multiple genomic tests and the method used to interpret these. However, this is common in clinical studies in which, over the time required to complete enrolment, newer or constantly refined sequencing platforms are implemented.
Patients with germline BRCA mutations derived high benefit from matched targeted therapy, but subgroup analysis showed that targeted therapies matched to genomic alterations also reduce the risk of progression when excluding patients with germline BRCA1/2 mutations (HR 0.64, 90% CI 0.39, 1.06).
Design allowed the inclusion of only those patients who presented with sensitivity to chemotherapy. This could have biased the population toward a group of patients who do not present with genomic alterations involved in resistance. Furthermore, some drugs could be considered suboptimal. Design itself did not properly test precision medicine because the control arm included maintenance chemotherapy.
In a research highlight published on 5 October 2022 in the Nature Reviews Clinical Oncology, Diana Romero emphasized that after a median follow-up of at least 21.4 months, median PFS was significantly longer in patients receiving targeted therapy in the ESCAT I/II subgroup (9.1 months versus 2.8 months with chemotherapy), but not in the intention-to-treat population (5.5 months versus 2.9 months), or the ESCAT beyond level II (2.8 months versus 3.1 months). These results provide prospective evidence to support molecularly guided treatment decisions.
The study was funded by Fondation ARC, AstraZeneca, the IHU-B programme (PRISM) and the Breast Cancer Research Foundation. AstraZeneca also supported targeted therapy (capivasertib, AZD2014, AZD8931, AZD4547, olaparib, selumetinib, bicalutamide and vandetanib) supply and distribution to the study sites. Novartis supported the supply of targeted therapy (alpelisib).