Abstract 239P
Background
ER and PR loss might occur during metastatic progression of ER+/HER2- BC, but the underpinning molecular alterations remain elusive.
Methods
We accessed data from metastatic HER2- BC within the MSK-2018 dataset to compare outcome, tumor characteristics and genomic alterations of BC with loss of ER (ER+/- n= 66) to those maintaining ER positivity (ER+/+ n= 364) or ER negativity (ER-/- n= 50). We also compared metastatic ER+/HER2- BC with loss of PR (PR+/- n= 111) to those maintaining PR (PR+/+ n= 192).
Results
ER+/- represented 15% of ER+/HER2- BC and were PR- in >90% cases. Compared to ER+/+, ER+/- BC were significantly (p<0.05) associated with higher grade, lower ER levels, PR – (vs +) and HER2 0 (vs low) status of the primary BC, but not with the histologic subtype (ductal vs lobular vs other), site (bone vs other) or timing (before 1st-second vs >second-line) of metastatic biopsy. Outcome data and selected mutations (mut) according to ER classes are presented in the table. In line with previous data, median disease-free survival (DFS) and overall survival (OS) significantly differed among ER+/+, ER+/- and ER-/- groups. Interestingly, ESR1 mut and ER loss were mutually exclusive and ER+/- showed genomic context similar to ER-/-. PR +/- BC were 37% of ER+/PR+/HER2- primary BC and were significantly (p<0.05) associated with shorter DFS (51 vs 70 months; HR 1.4. 95% CI (1.1-1.7)), lower ESR1 mut (3% vs 24%) and, when selecting only biopsies obtained before first-line, with adjuvant aromatase inhibitors (51% vs 29% SERM) compared to PR+/+ BC. Of note, ER and PR loss were 20% and 43% of HR+/HER2- ESR1 wild-type BC, respectively. Table: 239P
| ER +/+ | ER +/- | ER -/- | p value +/+ vs +/- | q value +/+ vs +/- | p value all | q value all | |
| Outcome | |||||||
| DFS (months) | 60 | 42 HR 1.4. 95% CI (1.1- 1.8) | 28 HR 2.5, 95% CI (1.8-3.3) | p <0.0001 | |||
| OS (months) | 206 | 105 HR 2.6, 95% CI (1.7-4) | 60 HR 5.5, 95% CI (3.6-8.4) | p <0.0001 | |||
| GENE mut | |||||||
| TP53 | 97 (27%) | 43 (65%) | 44 (88%) | 1.9E-09 | 7.2E-07 | 5.5E-21 | 2.1E-18 |
| RB1 | 5 (1%) | 8 (12%) | 6(12%) | 1.7E-05 | 0.0031 | 1.8E-06 | 0.00033 |
| PIK3R1 | 3 (0.8%) | 4 (6%) | 6 (12%) | 0.01 | 0.34 | 5.8E-06 | 0.00072 |
| ESR1 | 59 (16%) | 0 | 0 | 0.0008 | 0.076 | 2.2E-05 | 0.0021 |
| CDKN2A/p14ARF | 0 | 3 (4.5 %) | 0 | 0.001 | 0.076 | 7.7E-05 | 0.0058 |
| PIK3CA | 144 (40%) | 16 (24%) | 7 (14%) | 0.025 | 0.62 | 0.0003 | 0.015 |
| GATA3 | 72 (20%) | 6 (9%) | 0 | 0.057 | 1 | 0.0004 | 0.020 |
| MAP3K13 | 2 (0.5%) | 4 (6%) | 0 | 0.003 | 0.18 | 0.0007 | 0.027 |
Conclusions
We show that ER+/- BC display a molecular profile similar to ER-/- and hypothesize that loss of ER and/or PR might be alternative to ESR1 mutations in the metastatic progression of ER+/HER2- BC. Further analyses on matched primary and metastatic samples are warranted.
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
L. Biganzoli: Financial Interests, Personal, Advisory Board: AstraZeneca, Daiichi Sankyo, Eisai, Pfizer, Gilead, Sanofi, Seattle Genetics, Exact Sciences, Amgen, Boehringer-Ingelheim, Pierre Fabre, Menarini; Financial Interests, Personal, Invited Speaker: Lilly, Novartis, Roche; Financial Interests, Institutional, Research Grant: Celgene, Genomic Health, Novartis; Non-Financial Interests, Member of Board of Directors: SIOG. M. Benelli: Financial Interests, Personal, Other, Personal fees for consultancy on bioinformatics analyses: Novartis. L. Malorni: Financial Interests, Personal, Invited Speaker: Pfizer, Lilly, Novartis, Menarini, Novartis; Financial Interests, Personal, Advisory Board: Lilly, Seagen, Roche, Menarini, Pfizer; Financial Interests, Institutional, Research Grant: Pfizer, Novartis; Non-Financial Interests, Institutional, Product Samples: Biovica International; Other, Principal investigator in one clinical trial, Co-Principal Investigator in one clinical trial: International Breast Cancer Study Group (IBCSG). All other authors have declared no conflicts of interest.