Abstract PO5-22-04: Genomic analysis of local recurrences following risk adapted breast radiotherapy in the IMPORT trials defines 'true recurrences' and 'new primaries'

Abstract

Abstract Background: Most ipsilateral breast tumor recurrences (IBTR) following breast conserving surgery occur near the tumor bed (Veronisi et al, Ann Oncol 2001) and are considered to represent ‘true recurrence’ (derived from residual malignant cells of the index cancer). A smaller number occur elsewhere in the breast and are thought to be ‘new primaries’ (independently occurring cancer). The IMPORT LOW (Coles et al, The Lancet 2017) and IMPORT HIGH (Coles et al, The Lancet 2023) trials investigated adaptation of radiotherapy dose-volume to this spatially varying risk, testing partial breast irradiation and simultaneous integrated boost in patients with tumors at low and high risk of IBTR respectively. We analysed genomic relationships between index cancers and recurrences in these trials, to ascertain the frequency of true recurrences and unrelated new primaries, and their spatial distribution. Methodology: FFPE blocks were obtained from 137 patients who developed subsequent ipsi- or contralateral breast cancer: 66 from IMPORT HIGH and 71 from IMPORT LOW (4 with bilateral second cancers). DNA extracted from index and subsequent cancers underwent shallow whole genome sequencing (sWGS) using the Illumina NovaSeq 6000 system. Copy number profiles were derived from sWGS data using the R package QDNAseq(Scheinin et al, Genome Res 2014); those that failed QC criteria were removed. Relatedness of tumor pairs was determined using the R package Breakclone (Lips et al, Nat Genet 2022). This approach uses individual copy number aberration breakpoint position rather than events at chromosomal arm level. The p value cutoff for relatedness was 0.01; values between 0.01 and 0.05 were considered ambiguous. Researchers were blinded to clinical information prior to results. Results: In 80/137 tumor pairs both copy number profiles met QC criteria and clonal relationship could be ascertained. In total 26/80 were considered related, 20/80 ambiguous and 34/40 non-related (Table 1). 16/26 ipsilateral subsequent cancers in IMPORT HIGH were clonally related to the primary tumor and could be considered true recurrences. In 5/26 it was impossible to accurately call clonality as pairs shared only common copy number changes such as 1q or 8q gain and 16q loss. 5/26 were considered new primaries. In the lower risk IMPORT LOW cohort, subsequent cancers appear to have similar chances of being related or independent. As expected, most contralateral tumors in IMPORT LOW showed no evidence of a clonal relationship to the index tumors. In both trials an ambiguous subgroup of pairs shared some common copy number changes suggesting similar phenotype.5/14 contralateral tumors in IMPORT HIGH, in which some women were also at high risk of systemic relapse, shared very similar copy number profiles to the index tumor and may be metastases rather than new primaries. Discussion: A considerable number of tumors classed as IBTR appear genomically unrelated to the index cancer particularly in low risk cancer; this biology is relevant to clinical management both initially and at ‘recurrence’. The finding that 5/14 contralateral cancers in the high risk group appear to be clonally related suggests that some may be metastases. Results are being confirmed by targeted sequencing and will be correlated with recorded spatial and clinicopathological data and patient outcome. Future work will analyse spatial relationships more precisely using deformable image registration. *From the 77 patients with baseline circulating tumor DNA samples, 2 patients were excluded because their samples failed during assay quality control. Table. PIK3CA and ESR1 alterations detected in circulating tumor DNA at baseline and post abemaciclib treatment Citation Format: Sara Lightowlers, Maria Roman-Escorza, Elena Provenzano, Judith Bliss, Jason Carroll, H Y Charlie Chan, Clare Griffin, Joanne Haviland, Monica Jefford, Anna Kirby, Navita Somaiah, Mark Sydenham, Jenny Titley, John Yarnold, Charlotte Coles, Elinor Sawyer. Genomic analysis of local recurrences following risk adapted breast radiotherapy in the IMPORT trials defines 'true recurrences' and 'new primaries' [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO5-22-04.