Multiomics in primary and metastatic breast tumors from the AURORA US network finds microenvironment and epigenetic drivers of metastasis

The AURORA US Metastasis Project was established with the goal to identify molecular features associated with metastasis. We assayed 55 females with metastatic breast cancer (51 primary cancers and 102 metastases) by RNA sequencing, tumor/germline DNA exome and low-pass whole-genome sequencing and global DNA methylation microarrays. Expression subtype changes were observed in ~30% of samples and were coincident with DNA clonality shifts, especially involving HER2. Downregulation of estrogen receptor (ER)-mediated cell-cell adhesion genes through DNA methylation mechanisms was observed in metastases. Microenvironment differences varied according to tumor subtype; the ER+/luminal subtype had lower fibroblast and endothelial content, while triple-negative breast cancer/basal metastases showed a decrease in B and T cells. In 17% of metastases, DNA hypermethylation and/or focal deletions were identified near HLA-A and were associated with reduced expression and lower immune cell infiltrates, especially in brain and liver metastases. These findings could have implications for treating individuals with metastatic breast cancer with immune- and HER2-targeting therapies.

An ERBB1-3 Neutralizing Antibody Mixture With High Activity Against Drug-Resistant HER2+ Breast Cancers With ERBB Ligand Overexpression

Background

Plasticity of the ERBB receptor network has been suggested to cause acquired resistance to anti-human epidermal growth factor receptor 2 (HER2) therapies. Thus, we studied whether a novel approach using an ERBB1-3-neutralizing antibody mixture can block these compensatory mechanisms of resistance.

Methods

HER2+ cell lines and xenografts (n ≥ 6 mice per group) were treated with the ERBB1-3 antibody mixture Pan-HER, trastuzumab/lapatinib (TL), trastuzumab/pertuzumab (TP), or T-DM1. Downregulation of ERBB receptors was assessed by immunoblot analysis and immunohistochemistry. Paired pre- and post-T-DM1 tumor biopsies from patients (n = 11) with HER2-amplified breast cancer were evaluated for HER2 and P-HER3 expression by immunohistochemistry and/or fluorescence in situ hybridization. ERBB ligands were measured by quantitative reverse transcription polymerase chain reaction. Drug-resistant cells were generated by chronic treatment with T-DM1. All statistical tests were two-sided.

Results

Treatment with Pan-HER inhibited growth and promoted degradation of ERBB1-3 receptors in a panel of HER2+ breast cancer cells. Compared with TL, TP, and T-DM1, Pan-HER induced a similar antitumor effect against established BT474 and HCC1954 tumors, but was superior to TL against MDA-361 xenografts (TL mean = 2026 mm 3 , SD = 924 mm 3 , vs Pan-HER mean = 565 mm 3 , SD = 499 mm 3 , P = .04). Pan-HER-treated BT474 xenografts did not recur after treatment discontinuation, whereas tumors treated with TL, TP, and T-DM1 did. Post-TP and post-T-DM1 recurrent tumors expressed higher levels of neuregulin-1 (NRG1), HER3 and P-HER3 (all P < .05). Higher levels of P-HER3 protein and NRG1 mRNA were also observed in HER2+ breast cancers progressing after T-DM1 and trastuzumab (NRG1 transcript fold change ± SD; pretreatment = 2, SD = 1.9, vs post-treatment = 11.4, SD = 10.3, P = .04). The HER3-neutralizing antibody LJM716 resensitized the drug-resistant cells to T-DM1, suggesting a causal association between the NRG1-HER3 axis and drug resistance. Finally, Pan-HER treatment inhibited growth of HR6 trastuzumab- and T-DM1-resistant xenografts.

Conclusions

These data suggest that upregulation of a NRG1-HER3 axis can mediate escape from anti-HER2 therapies. Further, multitargeted antibody mixtures, such as Pan-HER, can simultaneously remove and/or block targeted ERBB receptor and ligands, thus representing an effective approach against drug-sensitive and -resistant HER2+ cancers.

Quantitative Mass Spectrometry Analysis of PD-L1 Protein Expression, N-glycosylation and Expression Stoichiometry with PD-1 and PD-L2 in Human Melanoma

Quantitative assessment of key proteins that control the tumor-immune interface is one of the most formidable analytical challenges in immunotherapeutics. We developed a targeted MS platform to quantify programmed cell death-1 (PD-1), programmed cell death 1 ligand 1 (PD-L1), and programmed cell death 1 ligand 2 (PD-L2) at fmol/microgram protein levels in formalin fixed, paraffin-embedded sections from 22 human melanomas. PD-L1 abundance ranged 50-fold, from ∼0.03 to 1.5 fmol/microgram protein and the parallel reaction monitoring (PRM) data were largely concordant with total PD-L1-positive cell content, as analyzed by immunohistochemistry (IHC) with the E1L3N antibody. PD-1 was measured at levels up to 20-fold lower than PD-L1, but the abundances were not significantly correlated (r² = 0.062, p = 0.264). PD-1 abundance was weakly correlated (r² = 0.3057, p = 0.009) with the fraction of lymphocytes and histiocytes in sections. PD-L2 was measured from 0.03 to 1.90 fmol/microgram protein and the ratio of PD-L2 to PD-L1 abundance ranged from 0.03 to 2.58. In 10 samples, PD-L2 was present at more than half the level of PD-L1, which suggests that PD-L2, a higher affinity PD-1 ligand, is sufficiently abundant to contribute to T-cell downregulation. We also identified five branched mannose and N-acetylglucosamine glycans at PD-L1 position N192 in all 22 samples. Extent of PD-L1 glycan modification varied by ∼10-fold and the melanoma with the highest PD-L1 protein abundance and most abundant glycan modification yielded a very low PD-L1 IHC estimate, thus suggesting that N-glycosylation may affect IHC measurement and PD-L1 function. Additional PRM analyses quantified immune checkpoint/co-regulator proteins LAG3, IDO1, TIM-3, VISTA, and CD40, which all displayed distinct expression independent of PD-1, PD-L1, and PD-L2. Targeted MS can provide a next-generation analysis platform to advance cancer immuno-therapeutic research and diagnostics.

Abstract LB-056: Quantitative mass spectrometry analysis of immune checkpoint protein expression and N-glycosylation in human melanoma

Quantitative assessment of key proteins that control the tumor-immune interface is one of the most formidable analytical challenges in immunotherapeutics. We developed an targeted mass spectrometry (MS) platform to quantify programmed cell death-1 (PD-1), programmed cell death 1 ligand 1 (PD-L1), and programmed cell death 1 ligand 2 (PD-L2) at fmol/microgram protein levels in formalin fixed, paraffin-embedded sections from 22 human melanomas. PD-L1 abundance ranged 50-fold, from approximately 0.03 to 1.5 fmol/microgram protein and the PRM data were largely concordant with total PD-L1-positive cell content, as analyzed by immunohistochemistry (IHC) with the E1L3N antibody. PD-1 was measured at levels up to 20-fold lower than PD-L1, but the abundances were not significantly correlated (r2 = 0.062, p = 0.264). PD-1 abundance was weakly correlated (r2 = 0.3057, p = 0.009) with the fraction of lymphocytes/histiocytes in sections. PD-L2 was measured from 0.03 to 1.90 fmol/microgram protein and the ratio of PD-L2 to PD-L1 abundance ranged from 0.03 to 2.58. In 8 samples, PD-L2 was present at more than half the level of PD-L1, which suggests that PD-L2, a higher affinity PD-1 ligand, is sufficiently abundant to contribute to T-cell downregulation. We also identified five branched mannose and N-acetylglucosamine glycans at PD-L1 position N192 in all 22 samples. Extent of PD-L1 glycan modification varied by approximately 10-fold and the melanoma with the highest PD-L1 protein abundance and most abundant glycan modification yielded a very low PD-L1 IHC estimate, thus suggesting that N-glycosylation may affect IHC measurement and PD-L1 function. Additional PRM analyses quantified immune checkpoint/co-regulator proteins LAG3, IDO1, TIM-3, VISTA and CD40, which all displayed distinct expression independent of PD-1, PD-L1 and PD-L2. Targeted MS can provide a next-generation analysis platform to advance cancer immuno-therapeutic research and diagnostics. (Supported by NIH grant U24CA159988)

Citation Format: Carlos Morales Betanzos, Hyoungjoo Lee, Paula I. Gonzalez Ericsson, Justin M. Balko, Douglas B. Johnson, Lisa J. Zimmerman, Daniel C. Liebler. Quantitative mass spectrometry analysis of immune checkpoint protein expression and N-glycosylation in human melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr LB-056. doi:10.1158/1538-7445.AM2017-LB-056