Background

TNT previously reported that patients(pts) with advanced triple negative breast cancer(TNBC) and germline BRCA1/2(gBRCA+) mutations are more likely to respond to carboplatin(C) than to docetaxel(D), and have longer progression-free survival with C. Here, we report results from a pre-planned biological analysis to test whether BRCA silencing or methylation status confer sensitivity to platinum.

Methods

Pts eligible for TNT had ER-, PgR-, HER2- BC or were known gBRCA+(any ER/PgR/HER2). Pts were randomized to C(AUC 6 q3wk) or D(100mg/m2 q3wk) for 6-8 cycles or until disease progression if sooner, with crossover possible on progression. Blood and archival tissue samples were requested for BRCA1/2 genotyping, central analysis of TN and DNA repair deficiency biomarkers. Primary endpoint was intention-to-treat objective response rate(ORR) up to cycle 6. BRCA1 mRNA level was estimated from whole-genome total RNA sequencing(RNA-Seq). BRCA1 methylation analysis was quantified by bisulfite sequencing. A methylation score for the sample was computed as percentage of methylated reads relative to total number of reads that were either methylated or not methylated for CpG sites. A sample with a score >10% was considered methylated(BRCA1meth)? a BRCA1 mRNA level "silenced" subgroup(BRCA1 silencing) was defined as value of log2(mRNA)<8.4, based on bimodal distribution.

Results

212 and 191 pts' samples were assessed for BRCA1 methylation and mRNA levels. 33 pts (15.6%) were found to be BRCA1meth. Pts with BRCA1meth had a better response to D (42%, 95%CI: 20, 64) than C (21%, 95%CI: -0.1, 43)? absolute difference (C-D) -21% (95%CI: -52, 10)? p=0.28), but not statistically significant. There was no evidence of a difference when gBRCA+ tumours were excluded.

31 pts (16%) had BRCA1 silencing. Concordant with BRCA1meth, pts with BRCA1 silencing had a better response to D (65%, 95%CI: 42, 87) than C (29%, 95%CI: 4.9, 52)? absolute difference (C-D) -36% (95%CI: -69, -3.3)? p=0.07), with an odds ratio 0.22 (95%CI: 0.035, 1.2) in favor of D, interaction with treatment not significant (p=0.066).

19 pts with BRCA1meth and BRCA1 silencing had a better response to D (60%, 95%CI: 30, 90) than C (22%, 95%CI: -5.0, 49)? absolute difference (C-D) -38% (95%CI: -79, 2.9) in favor of D (p=0.17), and the interaction was not significant (p=0.15).

Further exploratory analyses examining any relationship between a response to C and new cutpoints for BRCA1meth or BRCA1 mRNA level did not suggest any evidence of a signal.

Conclusion

Our data did not support a priori hypotheses that BRCA1 methylation or silencing would be associated with a greater response to C than D, which is consistent with other reports in ovarian cancer. As BRCA1 methylation will not always lead to significant silencing or be the only cause of low BRCA1 gene expression, exploratory analysis is investigating the interaction of BRCA silenced group with D sensitivity. RNA-seq is ongoing and results from the final database will be presented.

Citation Format: Tutt A, Cheang MCU, Kilburn L, Tovey H, Gillett C, Pinder S, Lanchbury J, Abraham J, Barrett S, Barrett-Lee P, Chan S, Gazinska P, Grigoriadis A, Kernaghan S, Hoadley K, Gutin A, Harper-Wynne C, Hatton M, Owen J, Parker P, Roylance R, Shaw A, Smith I, Thompson R, Timms K, Wardley A, Wilson G, Harries M, Ellis P, Ashworth A, Perou C, Bliss J, Rahman N, Brown R, On Behalf of the TNT Trial Management Group and Investigators. BRCA1 methylation status, silencing and treatment effect in the TNT trial: A randomized phase III trial of carboplatin compared with docetaxel for patients with metastatic or recurrent locally advanced triple negative or BRCA1/2 breast cancer (CRUK/07/012) [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr S6-01.