Support us

Human breast cancer development and progression

My group studies human breast cancer development and progression, as well as therapy response and resistance, in genetically engineered mouse models (GEMMs) and patient-derived tumor xenograft (PDX) models. We have developed mouse models for BRCA1/2-associated breast cancer and invasive lobular carcinoma (ILC), which are used to (1) study tumor cell-intrinsic and -extrinsic mechanisms of breast cancer development and progression; (2) develop novel therapeutic strategies for prevention and treatment of breast tumors; (3) study mechanisms of acquired resistance to targeted therapeutics.


Driver genes in BRCA1-associated breast cancer
We have developed several GEMMs for somatic modeling of BRCA1-associated breast cancer using intraductal injection of lentiviral vectors for stable overexpression of exogenous genes and CAS9-mediated disruption or APOBEC-CAS9n-UGI (BE3)-mediated base editing of endogenous genes. We have used these GEMMs to validate RB, PTEN, PIK3CA, MYC and MCL1 as bona fide driver genes in BRCA1-associated breast cancer. Moreover, MCL1 inhibition potentiated the in vivo efficacy of the PARP inhibitor (PARPi) olaparib, underscoring the therapeutic potential of this combination for treatment of BRCA1-associated cancer patients with poor response to PARPi monotherapy.

Key publications:
1. Annunziato et al., Nat Commun 2019
2. Annunziato et al., EMBO J 2020 

Therapy resistance in BRCA1/2 deficient breast cancer
BRCA1/2 deficient cancers are defective in homologous recombination repair and therefore hypersensitive to DNA-damaging agents, including platinum drugs and PARP inhibitors (PARPi). However, these treatments do not result in tumor eradication and eventually resistance develops. To study mechanisms of PARPi resistance, we combined functional genetic screens in BRCA1/2 deficient cells with multi-omics analysis of PARPi-resistant tumors from our GEMMs of BRCA1/2 deficient breast cancer. These studies have shown that PARPi resistance can be induced by (1) overexpression of drug efflux transporters, (2) re-activation of BRCA1 via (epi)genetic mechanisms, (3) hypomorphic actvity of BRCA1, or (4) loss of components of the 53BP1-RIF1-SHLD or CST complexes that govern protection of DNA double-strand breaks. PARPi resistance in BRCA2-deficient tumors can be induced by loss of the poly(ADP-ribose) glycohydrolase PARG. Recent work in our group shows that BRCA1-deficient tumors that have acquired PARPi resistance via 53BP1 loss are hypersensitive to ligase 3 inhibition, offering opportunities to combat this form of PARPi resistance in the clinic.

Key publications:
1. Rottenberg et al., PNAS 2008 
2. Bouwman et al., Nat Struct Mol Biol 2010 
3. Drost et al., Cancer Cell 2011 
4. Jaspers et al., Cancer Discov 2013 
5. Xu et al., Nature 2015 
6. Henneman et al., PNAS 2015 
7. ter Brugge et al., J Natl Cancer Inst 2016 
8. Drost et al., J Clin Invest 2016 
9. Barazas et al., Cell Rep 2018 
10. Gogola et al., Cancer Cell 2018 
11. Noordermeer et al., Nature 2018 

Driver genes in ILC
ILC accounts for 10-15% of all breast cancers and shows frequent inactivation of E-cadherin. Using in vivo transposon-based insertional mutagenesis in mammary-specific E-cadherin knockout mice, we have found that overexpression of hyperactive truncated forms of MYPT1/2 and ASPP2 reduces actomyosin contractility and thereby promotes malignant transformation of E-cadherin deficient mammary epithelial cells, resulting in ILC formation. Importantly, PPP1R12B, TP53BP2 and MYH9 are also frequently mutated in human breast tumors, suggesting that this novel oncogenic pathway may also be relevant for human cancer development. This work highlights actomyosin hypercontractility induced by E-cadherin loss as a critical barrier to ILC development.

Key publications:
1. Kas et al., Nat Genet 2017 
2. Schipper et al., Nat Commun 2019 
3. Schipper et al., Cancer Res 2020 

In vivo models of DCIS
Since the advent of breast screening, Ductal Carcinoma In Situ (DCIS) accounts for 25% of all breast neoplasms detected. This increased detection rate has resulted in overtreatment since most DCIS lesions will not progress into invasive breast cancer. Better insight into the biology of DCIS is required to distinguish indolent lesions from potentially hazardous ones. To this end, we are using intraductal injections of lentiviruses encoding (combinations of) DCIS driver genes to generate genetically engineered rat models of DCIS. We are also using intraductal injections of patient-derived DCIS cells to generate mouse PDX models of DCIS. These approaches will enable the identification of DCIS driver genes and yield models to study disease progression and response to targeted therapeutics.

This site uses cookies

This website uses cookies to ensure you get the best experience on our website.