Supplementary Table 1. Clinical characteristics of the 42 USC patients whose tissue specimens were utilized for HER2 immunohistochemical and FISH analyses.
Supplementary Figure 3. Mouse weight changes over the course of each four-arm treatment study. In all experiments, lapatinib and trastuzumab treatments did not significantly affect mouse weights. Mice were weighed weekly, and the percent changes in weight compared to baseline weights (100%) are shown. Error bars represent the standard error of the mean.
Supplementary Table S2: p-values of RT-PCR analysis of RAD51 gene family in CD133+ cells and CD133- CD117- cells FACS purified (Figure 6A and 6B).
Supplementary Table S1: p-values of RT-PCR analysis of PROM1 (CD133), KIT (CD117), SOX2 and POU5F1 (OCT4) genes (Supplementary Figure SC).
Supplementary Table 1. Clinical characteristics of the 42 USC patients whose tissue specimens were utilized for HER2 immunohistochemical and FISH analyses.
Supplementary Figure Legends. The supplementary figure legends are in this document as requested.
Supplementary Figure 2. Anti-tumor activity of single agent lapatinib treatment was observed in HER2 amplified ARK2 xenografts (p = 0.02), while no response was seen in non-HER2 amplified SPEC2 xenografts. Mice harboring tumors derived from ARK2 or SPEC2 cell lines were treated with either vehicle or lapatinib. Tumors were measured twice weekly, and the percent change in tumor volume compared to baseline volume (100%) is shown. Error bars represent the standard error of the mean.
Supplementary Figure 5. HER2 inhibition in mice harboring non-HER2 amplified SPEC2 (A) or EnCa2 (B) xenografts did not affect tumor cell apoptosis, as shown by TUNEL analysis. Representative images of TUNEL (green) and DAPI (blue) stained sections of the differently treated xenografts are shown. Scale bars: 50 µm.
Supplementary Figure 1. HER2 expression levels in USC models. HER2 protein expression (A) and gene amplification (B) status of the USC cell line derived and patient derived xenografts that were utilized in this study, as assessed by IHC and FISH. HER2 protein over-expression as well as HER2 gene amplification were observed in ARK1, ARK2 and EnCa1 xenografts, while low HER2 protein expression and normal HER2 gene status were found in SPEC2 and EnCa2 xenografts.
Supplementary Figure 4. Ki-67 staining of non-HER2 gene amplified USC xenografts. Unlike the HER2 amplified models, no changes in proliferation were seen upon treatment with lapatinib and/or trastuzumab in SPEC2 (A) and EnCa2 (B) xenografts. Tumors harvested at the completion of each in vivo treatment study were subjected to Ki-67 immunohistochemical analysis. Scale bars: 100 µm.
AbstractPARP inhibitors (PARPi) are FDA-approved monotherapy agents for the treatment of recurrent ovarian cancer in patients with and without a BRCA mutation. Despite promising response rates, not all patients derive benefit, and the majority develop resistance. PARPi treatment in vitro and in vivo induced an enrichment of CD133+ and CD117+ ovarian cancer stem cells (CSC). This effect was not affected by BRCA mutation status. In the CSC fractions, PARPi induced cell-cycle arrest in G2–M with a consequent accumulation of γH2AX, RAD51, and uniquely DMC1 foci. DNA damage and repair monitoring assays demonstrated that CSCs display more efficient DNA repair due, in part, to activation of embryonic repair mechanisms which involved the RAD51 homologue, DMC1 recombinase. Preserved and induced homologous repair (HR) could be a mechanism of an inherent resistance of CSCs to the synthetic lethality of PARPi that likely promotes disease recurrence.Implications:Treatment with PARPi fails to significantly affect ovarian cancer CSC populations, likely contributing to recurrent disease. Ovarian cancer CSCs stabilize genomic integrity after PARPi treatment, due to a more efficient inherent DNA repair capacity. PARPi-induced DMC1 recombinase and HR proficiency provide CSCs the opportunity to repair DNA damage more efficiently.Visual Overview: http://mcr.aacrjournals.org/content/molcanres/17/2/431/F1.large.jpg.
One of the most significant therapeutic challenges in the treatment of ovarian cancer is the development of recurrent platinum-resistant disease. Cancer stem cells (CSCs) are postulated to contribute to recurrent and platinum-resistant ovarian cancer (OvCa). Drugs that selectively target CSCs may augment the standard of care cytotoxics and have the potential to prevent and/or delay recurrence. Increased reliance on metabolic pathway modulation in CSCs relative to non-CSCs offers a possible therapeutic opportunity. We demonstrate that treatment with the metabolic inhibitor CPI-613 (devimistat, an inhibitor of tricarboxylic acid (TCA) cycle) in vitro decreases CD133+ and CD117+ cell frequency relative to untreated OvCa cells, with negligible impact on non-CSC cell viability. Additionally, sphere-forming capacity and tumorigenicity in vivo are reduced in the CPI-613 treated cells. Collectively, these results suggest that treatment with CPI-613 negatively impacts the ovarian CSC population. Furthermore, CPI-613 impeded the unintended enrichment of CSC following olaparib or carboplatin/paclitaxel treatment. Collectively, our results suggest that CPI-613 preferentially targets ovarian CSCs and could be a candidate to augment current treatment strategies to extend either progression-free or overall survival of OvCa.
Ovarian cancer (OvCa) is the most lethal gynecological malignancy in the United States primarily due to lack of a reliable early diagnostic, high incidence of chemo-resistant recurrent disease as well as profuse tumor heterogeneity. Cancer stem cells (CSCs) continue to gain attention, as they are known to resist chemotherapy, self-renew and re-populate the bulk tumor with undifferentiated and differentiated cells. Moreover, CSCs appear to readily adapt to environmental, immunologic and pharmacologic cues. The plasticity and ability to inactivate or activate signaling pathways promoting their longevity has been, and continues to be, the challenge faced in developing successful CSC targeted therapies. Identifying and understanding unique ovarian CSC markers and the pathways they utilize could reveal new therapeutic opportunities that may offer alternative adjuvant treatment options. Herein, we will discuss the current state of ovarian CSC characterization, their contribution to disease resistance, recurrence and shed light on clinical trials that may target the CSC population.
Abstract Inhibition of the phosphatidylinositol 3-kinase (PI3K) pathway has been shown to induce defective homologous repair in breast carcinoma leading to heightened sensitivity to poly-ADP ribose polymerase (PARP) inhibitors. Our objective was to assess the effects on DNA repair pathways following single agent and combined PI3K and PARP inhibition utilizing endometrial cancer (EnCa) patient derived xenografts (PDX) with and without identified PIK3CA mutations. Under an institutional protocol, tumors from consenting patients with endometrioid carcinoma were collected at the time of primary surgery and genotyped for common oncogenic mutations utilizing the SNaPshot® platform. NOD/SCID mice bearing xenografts derived from two primary human endometrioid endometrial tumors with and without PIK3CA gene mutations (ENCA1 & ENCA2 respectively) were divided into four arm cohorts with equivalent tumor volumes. Vehicle or NVP BKM-120 (30 mg/kg) was administered alone and in combination with veliparib (25 mg/kg) and endometrial xenograft tumor volumes were assessed. Immunoblotting was used to assess downstream PI3K pathway and DNA repair pathway alterations in post treatment samples at specific time points. While NVP BKM120 as a single agent failed to induce significant anti-tumor activity compared to vehicle in either model, decreased pAKT, RAD51 and increased γ-H2AX levels were observed in response to NVP BKM120 alone. PIK3CA mutation status did not affect these molecular alterations. Dual NVP BKM-120 and veliparib therapy significantly impeded tumor growth (p < 0.01) in the ENCA1 model, while in ENCA2, significant tumor regression was observed (p < 0.001). No changes in mouse weight or premature death were observed in the various treatment arms. Combination therapy was associated with reconstitution of baseline RAD51 and increased γ-H2AX levels. In conclusion, these results suggest that PI3K inhibition induces defects in homologous repair mediated by RAD51 and increases reliance on PARP mediated DNA repair mechanisms. Our in vivo data suggest that PI3K inhibition sensitizes endometrial tumors with and without PIK3CA mutations to a PARP inhibitor. Citation Format: Bo Rueda, Celeste Digloria, Tracilyn Hall, Rosemary Foster, Darrell Borger, Whitfield Growdon. PI3K inhibition induces homologous repair defects in endometrioid endometrial cancer patient derived xenografts leading to synergistic anti-tumor activity with PARP inhibitors. [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 2666. doi:10.1158/1538-7445.AM2015-2666
Objectives: Inhibition of the phosphatidylinositol 3-kinase (PI3K) pathway has been shown to induce defective homologous repair in breast carcinoma, leading to heightened sensitivity of poly-ADP ribose polymerase (PARP) inhibitors. Our objective was to assess the in vivo antitumor effects of combined PI3K and PARP inhibition using patient-derived xenografts (PDX) with and without identified PIK3CA mutations.
Abstract Uterine serous carcinoma (USC) is an aggressive subtype of endometrial cancer that commonly harbors HER2 gene amplification. Clinical trial has demonstrated that USC is impervious to trastuzumab therapy, though the mechanism is poorly understood. Since HER3 mediated signaling has been implicated in trastuzumab resistance in breast cancer, we sought to understand the relevance of HER3 activation in USC xenografts derived from the HER2 gene amplified (HER2:Chr 17 > 15) non-immortalized USC cell line ARK2. Cohorts of mice harboring xenografts derived from ARK2 were treated with either vehicle, trastuzumab (10 mg/kg IP BIW), lapatinib (150 mg/kg QD oral gavage) or the combination of trastuzumab and lapatinib for 21 days. Acute and chronic post treatment tumor samples were assessed for downstream signaling alterations. Single agent trastuzumab had no impact on xenograft growth compared to vehicle. Lapatinib alone resulted in significant tumorstatic effects (p < 0.01) and dual therapy with trastuzumab and lapatinib induced synergistic activity that significantly decreased tumor volume compared to all other arms (p < 0.01). Single agent trastuzumab was associated with rapid elevation in pHER3 levels with unchanged pAKT and pERK expression compared to vehicle. These elevated pHER3 levels were similarly elevated after the 21 day treatment. Anti-tumor activity observed in the lapatinib and dual lapatinib/trastuzumab arm was not associated with any alterations in pHER3 levels following 21 day treatment, though acute elevations in pHER3 were noted in the dual HER2 blockade arm at 24 hours after treatment. In conclusion, trastuzumab alone failed to impact USC xenograft growth and this innate resistance was associated with an elevation in pHER3 levels that was still evident in xenografts at the end of the treatment period. Unlike single agent trastuzumab, treatment arms that utilized lapatinib demonstrated significant anti-tumor activity with no increase in pHER3 above vehicle over the course of treatment. These data highlight HER3 activation as a possible trastuzumab resistance mechanism in USC that can be abrogated through the addition of lapatinib. Citation Format: Silvia F. Hernandez, Celeste DiGloria, Jolijn Groeneweg, Darrell Borger, Rosemary Foster, Bo Rueda, Whitfield Growdon. Phosphorylated HER3 levels associated with trastuzumab resistance in HER2 gene amplified uterine serous carcinoma xenograft tumors. [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 658. doi:10.1158/1538-7445.AM2015-658
BACKGROUND:Subsets of high grade endometrial cancer (EnCa) over-express HER2 (ERBB2), yet clinical trials have failed to demonstrate any anti-tumor activity utilizing trastuzumab, an approved platform for HER2 positive breast cancer (BrCa). A truncated p95HER2 variant lacking the trastuzumab binding site may confer resistance. The objective of this investigation was to characterize the expression of the p95HER2 truncated variant in EnCa.MATERIALS AND METHODS:With institutional approval, 86 high grade EnCa tumors were identified with tumor specimens from surgeries performed between 2000 and 2011. Clinical data were collected and all specimens underwent tumor genotyping, HER2 immunohistochemistry (IHC, HercepTest®), HER2 fluorescent in situ hybridization (FISH), along with total HER2 (H2T) and p95HER2 assessment with VeraTag® testing. Regression models were used to compare a cohort of 86 breast tumors selected for equivalent HER2 protein expression.RESULTS:We identified 44 high grade endometrioid and 42 uterine serous carcinomas (USC). IHC identified high HER2 expression (2+ or 3+) in 59% of the tumors. HER2 gene amplification was observed in 16 tumors (12 USC, 4 endometrioid). Both HER2 gene amplification and protein expression correlated with H2T values. High p95HER2 expression above 2.8RF/mm2 was observed in 53% (n=54) with significant correlation with H2T levels. When matched to a cohort of 107 breast tumors based on HercepTest HER2 expression, high grade EnCa presented with higher p95 levels (p<0.001).CONCLUSIONS:These data demonstrate that compared to BrCa, high grade EnCa expresses higher levels of p95HER2 possibly providing rationale for the trastuzumab resistance observed in EnCa.
Abstract Purpose: Uterine serous carcinoma (USC) is an aggressive subtype of endometrial cancer that commonly harbors HER2 gene amplification. We investigated the effectiveness of HER2 inhibition using lapatinib and trastuzumab in vitro and in xenografts derived from USC cell lines and USC patient-derived xenografts. Experimental Design: Immunohistochemistry and FISH were performed to assess HER2 expression in 42 primary USC specimens. ARK1, ARK2, and SPEC2 cell lines were treated with trastuzumab or lapatinib. Cohorts of mice harboring xenografts derived from ARK2 and SPEC2 cell lines and EnCa1 and EnCa2 primary human USC samples were treated with either vehicle, trastuzumab, lapatinib, or the combination of trastuzumab and lapatinib. Acute and chronic posttreatment tumor samples were assessed for downstream signaling alterations and examined for apoptosis and proliferation. Results:HER2 gene amplification (24%) correlated significantly with HER2 protein overexpression (55%). All models were impervious to single-agent trastuzumab treatment. Lapatinib decreased in vitro proliferation of all cell lines and in vivo growth of HER2-amplified xenografts (ARK2, EnCa1). In addition, dual therapy with trastuzumab and lapatinib resulted in significant antitumor activity only in ARK2 and EnCa1 tumors. Dual HER2 therapy induced on target alteration of downstream MAPK and PI3K pathway mediators only in HER2-amplified models, and was associated with increased apoptosis and decreased proliferation. Conclusions: Although trastuzumab alone did not impact USC growth, dual anti-HER2 therapy with lapatinib led to improved inhibition of tumor growth in HER2-amplified USC and may be a promising avenue for future investigation. Clin Cancer Res; 20(24); 6517–28. ©2014 AACR.