Analysis of genomic and non-genomic signaling of estrogen receptor in PDX models of breast cancer treated with a combination of the PI3K inhibitor alpelisib (BYL719) and fulvestrant

BACKGROUND

Endocrine therapies targeting estrogen signaling have significantly improved breast cancer (BC) patient survival, although 40% of ERα-positive BCs do not respond to those therapies. Aside from genomic signaling, estrogen triggers non-genomic pathways by forming a complex containing methylERα/Src/PI3K, a hallmark of aggressiveness and resistance to tamoxifen. We aimed to confirm the prognostic value of this complex and investigated whether its targeting could improve tumor response in vivo.

METHODS

The interaction of ERα/Src and ERα/PI3K was studied by proximity ligation assay (PLA) in a cohort of 440 BC patients. We then treated patient-derived BC xenografts (PDXs) with fulvestrant or the PI3K inhibitor alpelisib (BYL719) alone or in combination. We analyzed their anti-proliferative effects on 6 ERα+ and 3 ERα- PDX models. Genomic and non-genomic estrogen signaling were assessed by measuring ERα/PI3K interaction by PLA and the expression of estrogen target genes by RT-QPCR, respectively.

RESULTS

We confirmed that ERα/Src and ERα/PI3K interactions were associated with a trend to poorer survival, the latter displaying the most significant effects. In ERα+ tumors, the combination of BYL719 and fulvestrant was more effective than fulvestrant alone in 3 models, irrespective of PI3K, PTEN status, or ERα/PI3K targeting. Remarkably, resistance to fulvestrant was associated with non-genomic ERα signaling, since genomic degradation of ERα was unaltered in these tumors, whereas the treatment did not diminish the level of ERα/PI3K interaction. Interestingly, in 2 ERα- models, fulvestrant alone impacted tumor growth, and this was associated with a decrease in ERα/PI3K interaction.

CONCLUSIONS

Our results demonstrate that ERα/PI3K may constitute a new prognostic marker, as well as a new target in BC. Indeed, resistance to fulvestrant in ERα+ tumors was associated with a lack of impairment of ERα/PI3K interaction in the cytoplasm. In addition, an efficient targeting of ERα/PI3K in ERα- tumors could constitute a promising therapeutic option.

Abstract 1718: Tumor immune gene profile before and after various targeted therapies in NSCLC PDXs

Introduction: The blockade of immune checkpoints with antibodies directed against CTLA-4, PD-1 and PD-L1 molecules, is now entering in the current management of non-small cell lung cancer (NSCLC) patients; those tumors are heterogeneous regarding driver oncogenes, which predict the efficacy of targeted therapies (TT). In this context, our study aimed to define the relationship between conventional TT used in NSCLC and the dynamic expression of immune genes.

Experimental procedure: Eight different NSCLC Patient-Derived Xenografts (PDXs) with defined mutations, were grafted in immunodeficient mice which were then treated by various TT (Table). About 3 tumors per group were collected at the end of the in vivo experiments. The expression of human and mouse genes, using species-specific primers was determined by qPCR and normalized to the expression of the human or mouse Total Binding Protein gene. Basal gene expression, predictive markers of response and resistance to tested therapies, as well as pharmacodynamics (PD) markers were analyzed.

Results: Basal immune gene expression analysis revealed various immune profiles in this panel of NSCLC PDXs, with clustering of both models and genes. Predictive and pharmacodynamics markers underlined the impact of human HLA class I and II gene expression in the response to erlotinib and everolimus (HLA-A, HLA-DRA, and HLA-DRB); of B7-H3 (CD276), PDL1 (CD274) and Tigit (PVRIG) in the response to erlotinib; and of TNFSF7 (CD70) and TNFSF9 in the response to BKM120. PD markers showed that PDL1, B7-H3, and Tigit expression was highly impacted by everolimus, erlotinib, and BKM120 treatments, respectively. Complete gene expression results can be presented at the meeting.

Conclusions: Using NSCLC PDXs, we have identified specific tumor immune profiles and a set of genes involved in the response to TT. Moreover, PD analysis of immune markers reveals potential combinations of TT with immune checkpoints blockade that may have important translational value.

NSCLC PDXsLCF4LCF9LCF12LCF15LCF25LCF29ML1ML5HistologyAdenocarcinomaXXXXXXXLarge cellsXMUTATIONSEGFRXXKRASXXPi3KCAXXXTREATMENTS (Targets)Cetuximab (1)EGFRXXXXErlotinib (2)EGFRXXXXAfatinib (3)EGFR/HER2XXXXBKM120 (4)Pi3KXXXXXXXXEverolimus (5)mTORC1XXXXSelumetinib (6)MEK1/2XXXXXXXX1 + 2/XXXX1 + 3/XXXX1 + 4/X2 + 4/X3 + 4/X5 + 6/XXXX

Citation Format: Elodie Montaudon, Sophie Vacher, Wilfrid Richer, Ahmed Dahmani, Caroline Guibaudet, Olivier Déas, Stefano Cairo, Ludmilla De Plater, Marine Dubois, Fariba Némati, Christine Sedlik, Jean-Gabriel Judde, Nicolas Girard, Ivan Bieche, Eliane Piaggio, Didier Decaudin. Tumor immune gene profile before and after various targeted therapies in NSCLC PDXs [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1718.

Abstract 1687: Vandetanib as a potential new treatment for ER negative breast cancers

Introduction: Recent studies have shown that the receptor tyrosine kinase RET is involved in the biology of ER positive breast cancers and in the response to endocrine treatment, but its role in ER negative tumors is unknown. Here we investigated the expression of RET in BC patients tumors and patient-derived xenografts (PDX) and evaluated the therapeutic potential of Vandetanib in ER negative BC PDX.

Methods: RET mRNA expression was analyzed in BC of 446 patients and 57 PDX by RT-PCR analysis. The activity of Vandetanib, a tyrosine kinase inhibitor targeting RET, EGFR and VEGFR2, was tested in three PDX of triple-negative breast cancer (TNBC) and one PDX of HER2+ BC with different levels of RET expression. Protein expression of P-RET, RET, EGFR, P-EGFR and c-KIT were determined by immunohistochemistry (IHC). Analyses of PI3K and MAPK pathways and angiogenesis were performed by IHC and RT-PCR in both untreated and Vandetanib-treated tumors.

Results: In both clinical samples and PDX, elevated levels of RET were found in ER+ and HER2+ tumors, and in a subgroup of TNBC tumors. In the HBCx5 (HER2+) and HBCx24 (TNBC) PDX, both with RET over-expression, treatment by Vandetanib resulted in tumor growth inhibition (TGI) of 90% and 98%, respectively. In both models, tumor regressions were observed in 50% of xenografts. The effect of Vandetanib was associated to a marked inhibition of RET phosphorylation. To determine whether the lack of RET over-expression was associate to Vandetanib resistance, we treated two additional TNBC PDX with low and no expression of RET: HBCx4B and HBCx14. In these models, treatment by Vandetanib still inhibited tumor growth with a TGI of 85%. Tumor regressions were registered in 42% of animals in the PDX model with low expression of RET (HBCx4B), while no tumor regression were observed in HBCx14. IHC analyses showed an over-expression of EGFR in the HBCx4B xenograft and inhibition of EGFR phosphorylation in treated tumors, suggesting that tumor response to Vandetanib could depend on EGFR inhibition in this tumor. Further analyses of treated tumors revealed a decreased expression of phospho-ERK in the 4 PDX models, indicating inhibition of MAPK pathway, while the phosphorylation status of the PI3K pathway markers S6 and 4EBP1 was unchanged. Finally, treatment by Vandetanib decreased expression of murine Vegf receptors and the endothelial marker Cd31 in the 4 PDX tested, indicating angiogenesis inhibition.

Conclusions: Treatment by Vandetanib resulted in strong tumor growth inhibition in ER negative PDX with over-expression of RET. This effect was associated to inhibition of RET phosphorylation and MAPK pathway and decreased tumor vascularization. The lack of RET over-expression did not predict Vandetanib resistance, and over-expression of EGFR was also associated to a marked tumor response. These preclinical results suggest that Vandetanib treatment could be useful for patients with ER negative breast cancers expressing Vandetanib's targets.

Citation Format: Elisabetta Marangoni, Rana Hatem, Dalila Labiod, Sophie Chateau-Joubert, Rania El Botty, Jean-Luc Servely, Ludmilla De Plater, Ivan Bièche. Vandetanib as a potential new treatment for ER negative breast cancers. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1687. doi:10.1158/1538-7445.AM2015-1687

Vasculature analysis of patient derived tumor xenografts using species-specific PCR assays: evidence of tumor endothelial cells and atypical VEGFA-VEGFR1/2 signalings

BACKGROUND

Tumor endothelial transdifferentiation and VEGFR1/2 expression by cancer cells have been reported in glioblastoma but remain poorly documented for many other cancer types.

METHODS

To characterize vasculature of patient-derived tumor xenografts (PDXs), largely used in preclinical anti-angiogenic assays, we designed here species-specific real-time quantitative RT-PCR assays. Human and mouse PECAM1/CD31, ENG/CD105, FLT1/VEGFR1, KDR/VEGFR2 and VEGFA transcripts were analyzed in a large series of 150 PDXs established from 8 different tumor types (53 colorectal, 14 ovarian, 39 breast and 15 renal cell cancers, 6 small cell and 5 non small cell lung carcinomas, 13 cutaneous melanomas and 5 glioblastomas) and in two bevacizumab-treated non small cell lung carcinomas xenografts.

RESULTS

As expected, mouse cell proportion in PDXs -evaluated by quantifying expression of the housekeeping gene TBP- correlated with all mouse endothelial markers and human VEGFA RNA levels. More interestingly, we observed human PECAM1/CD31 and ENG/CD105 expression in all tumor types, with higher rate in glioblastoma and renal cancer xenografts. Human VEGFR expression profile varied widely depending on tumor types with particularly high levels of human FLT1/VEGFR1 transcripts in colon cancers and non small cell lung carcinomas, and upper levels of human KDR/VEGFR2 transcripts in non small cell lung carcinomas. Bevacizumab treatment induced significant low expression of mouse Pecam1/Cd31, Eng/Cd105, Flt1/Vegfr1 and Kdr/Vefr2 while the human PECAM1/CD31 and VEGFA were upregulated.

CONCLUSIONS

Taken together, our results strongly suggest existence of human tumor endothelial cells in all tumor types tested and of both stromal and tumoral autocrine VEGFA-VEGFR1/2 signalings. These findings should be considered when evaluating molecular mechanisms of preclinical response and resistance to tumor anti-angiogenic strategies.

Abstract 1772: PDL192, a humanized anti-Tweak receptor monoclonal antibody, mediates antitumor effects in primary human breast carcinoma xenografts

Background: PDL192 is a humanized IgG1 monoclonal antibody that binds the human TWEAK receptor (TweakR). TweakR, a member of the TNFR (Tumor Necrosis Factor Receptor) superfamily, is overexpressed in several human cancers including breast cancer (BC). In BC, it may also play a role in the invasive and metastatic potential of the disease (Willis et al, Mol Cancer Res 2008). In TweakR-expressing cancer cell lines or mouse xenograft models, PDL192 has a potent antitumor effect (Culp et al., CCR 2010). All these data therefore suggest that anti-TweakR targeting could be a promising new therapeutic approach for human BC patients.

Material and methods: TweakR expression was assessed by IHC (immunohistochemistry) on 3 Tissue-Micro-Array (TMA) banks of BC samples (basal-like, ERBB2, and luminal A/B), and 25 primary human BC xenografts (HBCx). The cut-off of positivity was defined as at least 25% cells with membraneous or cytoplasmic staining or by a combined score of percentage of positive staining cells x intensity > 50. The in vivo antitumor effect of PDL192 was then assessed on 7 TweakR-positive models (10 mg/kg thrice a week for 3 weeks by intraperitoneal route) in which one in combination with chemotherapy as maintenance therapy, as previsouly reported (Marangoni et al., BJC 2009).

Results: TMA analyses showed that TweakR was expressed in 16/37 basal like BC (43%), 23/37 ERBB2-positive BC (62%), and 38/71 luminal BC (54%). A high TweakR expression was correlated with double estrogen receptor- and Her2-positive tumors. Moreover, 13/25 xenografts have been found to be TweakR-positive (52%). Nine human BC models have been treated with PDL192, with 4 models (44%) showing a tumor growth inhibition (TGI) ranging between 59% and 91%. No correlation has been observed between TweakR expression and in vivo TGI. Moreover, when PDL192 was administered in complete remission after chemotherapy (doxorubicin + cyclophosphamide), we observed a highly significant delay of relapse greater than 2 months.

Conclusions: TweakR is expressed in 77/145 human BC samples (53%). In in vivo experiments, PDL192 showed potent TGI in 4/9 models, and significantly delayed tumor relapses after chemotherapy-induced complete remission. All these data therefore support the use of anti-TWEAK receptor monoclonal antibodies in the treatment of TweakR-positive BC patients.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 1772. doi:10.1158/1538-7445.AM2011-1772

Abstract A27: Efficacy of estramustine + docetaxel in docetaxel-resistant human prostate cancer xenograft: a preclinical model of docetaxel resistance reversion

Introduction: Androgen-dependent prostate cancers (PC) are usually sensitive to docetaxel chemotherapy. Nevertheless, docetaxel-resistance frequently appears after several cycles of treatment, raising the treatment rescue of docetaxel-resistant PC patients. If the combination of estramustine to docetaxel prolongs the metastatic-free and overall survivals of patients with androgen-dependent PC, its use remains limited in patients because of unacceptable toxicity including venous thromboses. The aims of this study were (1) to evaluate the in vivo efficacy of estramustine combined to docetaxel since initial tumor growth and since appearance of docetaxel resistance in one androgen-dependent human PC xenograft, and (2) to evaluate the efficacy of estramustine in six human androgen-independent human PC variants.

Experimental procedures: The preclinical models included the androgen-dependent PAC120 tumor obtained from a Gleason-9 primary PC of a 51-years old patient (De Pinieux et al, Am J Pathol, 2001), and six hormone-independent variants derived from PAC120. Docetaxel was administrated at a dosage of 20 mg/kg every three weeks by intraperitoneal injection for at most six cycles, and estramustine was given intraperitoneally at a dosage of 12 mg/kg days 1 to 5 every 3 weeks until mice sacrifice. Tumor volume was measured twice a week and Relative Tumor Volume (RTV) from start of treatment were then calculated. In the PAC120 model, escape to docetaxel was defined at the beginning of each next cycle (n + 1) as a ratio of RTVn+1/RTVn ≥ 2. Resistant mice were then randomized into two groups, one receiving estramustine alone, and the other treated by a combination of docetaxel + estramustine, with an evaluation of the tumor growth delay for a 2-fold tumor size increase from randomisation (TGD2). In the 6 androgen-independent PC xenografts (HID), estramustine administered alone was compared to a control group.

Results: Estramustine alone did not induce significant tumor growth inhibition in both PAC120 and HID xenografts. In PAC120 model, tumor growth inhibition (TGI) after docetaxel alone and docetaxel + estramustine was 81% and 95%, with a mean response duration of 29 days [21–63] and 68 days [64–74], respectively (p < 0.05). In docetaxel-resistant tumor bearing mice, estramustine alone induced a TGD2 of 18 days, whereas estramustine + docetaxel induced a TGD2 of 49 days (p < 0.05). Metastatic lesions and venous thromboses into lung of tumor-bearing mice have also been evaluated by histopathological analyses and may be presented during the meeting.

Conclusions: Estramustine alone was not efficient in both human androgen-dependent and -independent PC xenografts. Inversely, the combination of estramustine + docetaxel in first treatment line and estramustine added to docetaxel in docetaxel-resistant xenografts was highly effective. These data therefore suggest that estramustine should be combined to docetaxel in PC patients, but its use could be delayed, particularly in elderly patients, to docetaxel refractory cases.

Citation Information: Mol Cancer Ther 2009;8(12 Suppl):A27.