Luminal androgen receptor role and pathological complete response rate to neoadjuvant chemotherapy in triple negative breast cancer

Background

Triple negative breast cancer (TNBC) is a heterogeneous disease with distinct molecular subtypes. A luminal androgen receptor subgroup dependent on AR expression has been recently defined in a gene-expression study by Lehmann et al. We aimed to explore the clinical relevance of this AR dependent subtype in TNBC determining differences in response to neoadjuvant chemotherapy.

Methods

In a population of 116 patients (39 [34%] from GEICAM/2006-03 trial) treated with neoadjuvant anthracyclines and taxanes + /-carboplatin, tumor DNA was obtained from FFPE pre-treatment tumor biopsies. Lehmann subtypes were determined by gene expression profiling with HTA2.0 arrays (Illumina) and the classification tool TNBCtype. Breast cancer intrinsic subtypes according to PAM50 test were also determined with an nCounter Analysis System (Nanostring Technologies). The association of the different subtypes with pathologic complete response (pCR) was explored using Fisher's exact test and logistic regression.

Results

The global rate pCR of TNBC patients was 38.8%, and it was unevenly distributed within Lehmann's subtypes. Basal-like subtypes had the highest rates (BL1 = 53%, BL2 = 46%) and the luminal-androgen receptor (LAR) the lowest (14%). As it has been previously described that MSL is enriched in normal tissue, we performed this analysis with and without MSL subgroup, obtaining a significant association between LAR subtypes and pCR when MSL was excluded (p = 0.35 and p = 0.045, respectively). Most patients were classified as basal-like according to PAM50 except those included in LAR subtype that were also HER2-enriched (33.3%) and Luminal A (11.1%).

Conclusions

Our results suggest that there is a high genetic diversity within TNBC, mainly due to a luminal androgen receptor subtype that includes an elevated percentage of non-basal-like tumors. The LAR subtype is associated with a lower rate of pCR probably because almost half of them don't have basal-like characteristics (HER2 without anti-HER2 therapy and Luminal). Taking into account this specific subtype it could be necessary to use a new TNBC classification.

Clinical trial identification

Legal entity responsible for the study

FIMABIS-GEICAM

Funding

FIMABIS

Disclosure

All authors have declared no conflicts of interest.