Citations

Abstract PARP inhibitor (PARPi) resistance in high-grade serous ovarian carcinoma (HGSOC) can be acquired as a result of restored homologous recombination (HR) due to secondary or reversion mutations in HR genes, such as BRCA1, BRCA2, and RAD51C, or due to loss of BRCA1 promoter methylation (meBRCA1). We have demonstrated that homozygous meBRCA1 can be lost or reverted to heterozygous methylation following treatment with platinum-based chemotherapy, resulting in HR-competent PARPi-resistant tumors. RAD51C promoter methylation (meRAD51C) is detected in approximately 2% of HGSOC cases and, as with meBRCA1, is associated with gene silencing and HR deficiency. We are exploring PARPi response in meRAD51C preclinical models to determine clinical relevance. Two patient-derived xenograft (PDX) models of HGSOC with RAD51C gene silencing caused by meRAD51C have distinct meRAD51C profiles (measured by methylation-specific high-resolution melt analysis and targeted bisulfite next-generation sequencing), and different responses to PARPi treatment pressure. PDX PH039 loses methylation and regains RAD51C expression after only 2 cycles of PARPi retreatment (niraparib), resulting in PARPi-refractory tumors by cycle 3-4. Illumina EPIC methylation array analysis of PH039 revealed increasing global methylation losses following each round of PARPi treatment. Lack of meRAD51C stability and rapid development of PARPi resistance in PH039 may be due to the high degree of meRAD51C heterogeneity within the tumor favoring selection of pre-existing HR-competent clones under PARPi pressure. In contrast, PDX 183 has a relatively homogeneous and stable meRAD51C profile. We have multiple examples of using unique PDX models to demonstrate various important features of PARPi resistance, including loss of promoter methylation for BRCA1 vs. RAD51C. Thus, meRAD51C confers response to PARPi in HGSOC, but PARPi treatment pressure can cause loss of methylation and drug resistance in some tumors. The contrasting PARPi responses of these PDX provide a platform for the study of meRAD51C stability in vivo and may present therapeutic opportunities to improve meRAD51C durability and PARPi responses in patients. Citation Format: Kasenija Nesic, Rachel M Hurley, Cordelia McGehee, Olga Kondrashova, Maria I. Harrell, Giada V. Zapparoli, Ashan Musafer, Ming E. Wong, John Weroha, Xiaonan Hou, Hu Li, Vivian Negron, Kevin Peterson, Paula Schneider, Elizabeth M. Swisher, Melissa Southey, Alexander Dobrovic, Matthew Wakefield, Scott H. Kaufmann, Clare L. Scott. DNA repair gene promoter methylation patterns adapt and influence PARP inhibitor response [abstract]. In: Proceedings of the AACR Special Conference on Advances in Ovarian Cancer Research; 2019 Sep 13-16, 2019; Atlanta, GA. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(13_Suppl):Abstract nr IA02.

Abstract About 50% of high-grade serous ovarian cancers (HGSOC) have defects in homologous recombination (HR) DNA repair. Intriguingly, HR-defective tumors are highly susceptible to poly(ADP-ribose) polymerase (PARP) inhibitors and platinum-based chemotherapy. Mutations in BRCA1/2 or epigenetic silencing of BRCA1 via promoter hypermethylation contributes to ~30% of the HR defects, but other genetic alterations that cause HR deficiency in HGSOC remain unclear. To identify novel genes that may regulate HR, we studied Zinc Finger CCCH-Type Containing 18 (ZC3H18), encoded by ZC3H18. Notably, ZC3H18 lies within 16q24.2, a region with frequent copy number loss in HGSOC. Here we demonstrate that BRCA1 is transcriptionally activated by a previously unknown function of ZC3H18. We show that ZC3H18 interacts with an E2F site in the BRCA1 promoter where it facilitates recruitment of E2F4 to an adjacent E2F site to promote BRCA1 transcription. Consistent with the role of ZC3H18 in promoting BRCA1 expression, ZC3H18 depletion induces BRCA1 promoter methylation, reduces BRCA1 expression, disrupts HR, and sensitizes ovarian cancer cells to PARP inhibitors and platinum agents. Moreover, in patient-derived xenografts and primary HGSOC tumors, ZC3H18 mRNA levels were positively correlated with BRCA1 mRNA levels, further supporting the role of ZC3H18 in regulating BRCA1. Together, these findings suggest that loss and/or reduced ZC3H18 expression may help identify HGSOC patients most likely to benefit from PARP inhibitor and platinum-based therapies. Summary of the Unpublished Data/Significance/Innovation: These studies are innovative on several fronts. First, although BRCA1 methylation is a frequent event that mediates tumorigenesis and therapeutic responses in HGSOC, how BRCA1 methylation is regulated remains unknown. These studies will be the first to provide a mechanistic basis for how BRCA1 is methylated in HGSOC. Second, our studies are the first to show a role for ZC3H18 in any cancer. Third, we show that ZC3H18 regulates HR and that its depletion sensitizes ovarian cancer cells to PARP inhibitors and platinum therapy. Thus, these studies will provide important basis for developing ZC3H18 loss/decreased abundance as a novel biomarker for ovarian cancers that may respond to PARP inhibitors and platinum agents. Citation Format: Arun Kanakkanthara, Catherine J. Huntoon, Xiaonan Hou, Minzhi Zhang, Ethan P Heinzen, Ann L. Oberg, John S. Weroha, Scott H. Kaufmann, Larry M. Karnitz. Loss of ZC3H18 disrupts homologous recombination repair and sensitizes ovarian cancer cells to PARP inhibitors and DNA cross-linking agents [abstract]. In: Proceedings of the AACR Special Conference on Advances in Ovarian Cancer Research; 2019 Sep 13-16, 2019; Atlanta, GA. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(13_Suppl):Abstract nr A33.

Abstract Purpose: Gene expression–based molecular subtypes of high-grade serous tubo-ovarian cancer (HGSOC), demonstrated across multiple studies, may provide improved stratification for molecularly targeted trials. However, evaluation of clinical utility has been hindered by nonstandardized methods, which are not applicable in a clinical setting. We sought to generate a clinical grade minimal gene set assay for classification of individual tumor specimens into HGSOC subtypes and confirm previously reported subtype-associated features. Experimental Design: Adopting two independent approaches, we derived and internally validated algorithms for subtype prediction using published gene expression data from 1,650 tumors. We applied resulting models to NanoString data on 3,829 HGSOCs from the Ovarian Tumor Tissue Analysis consortium. We further developed, confirmed, and validated a reduced, minimal gene set predictor, with methods suitable for a single-patient setting. Results: Gene expression data were used to derive the predictor of high-grade serous ovarian carcinoma molecular subtype (PrOTYPE) assay. We established a de facto standard as a consensus of two parallel approaches. PrOTYPE subtypes are significantly associated with age, stage, residual disease, tumor-infiltrating lymphocytes, and outcome. The locked-down clinical grade PrOTYPE test includes a model with 55 genes that predicted gene expression subtype with >95% accuracy that was maintained in all analytic and biological validations. Conclusions: We validated the PrOTYPE assay following the Institute of Medicine guidelines for the development of omics-based tests. This fully defined and locked-down clinical grade assay will enable trial design with molecular subtype stratification and allow for objective assessment of the predictive value of HGSOC molecular subtypes in precision medicine applications. See related commentary by McMullen et al., p. 5271

Abstract Background: We have previously described a resistance to the normal process of apoptosis in neutrophils of patients with myelofibrosis with myeloid metaplasia (MMM) (Blood2003;102:11). Most recently, an activating mutation of JAK2 (V617F) has been described in approximately half of the patients with MMM as well as in variable proportion of patients with other myeloproliferative disorders (MPD). In the current study, we investigated the correlation between JAK2 V617F mutation status and neutrophil apoptosis in MMM. Methods: Neutrophils were isolated by density centrifugation from patients with MMM, other MPDs, and normal controls and assessed for apoptosis at baseline and after 24 hours in culture (IMDM with 20% sterilized fetal calf serum to simulate spontaneous apoptosis). Apoptosis was quantified using three-color flow cytometry using CD45 (to confirm leukocyte presence), annexin V (AN) (marker of apoptosis; detects aberrant externalization of phosphatidylserine during apoptosis), and propidium iodide (PI) (marker of dead cells). Mutation analysis for JAK2 V617F was performed in DNA derived from the isolated neutrophils using genomic DNA amplified by PCR, or extracted from cytogenetic pellets in archived specimens. Apoptotic rates after 24 hours in culture were correlated between patients and controls for both JAK2 mutation status and clinical parameters. Immunoblotting was performed on a subset of patients for correlation of JAK2 mutation status and downstream phosphorylation of the JAK2 target, STAT3, which transcriptionally activates several antiapoptotic genes. Results: Spontaneous neutrophil apoptosis was significantly decreased in MMM patients (n=50; median % apoptotic cells at 41%) compared to both healthy volunteers (n=9; 66%) and patients with other MPD (n=11; 53%) (p=0.002). Resistance to apoptosis in MMM correlated with both anemia (p=0.01) and the presence of the JAK2 V617F mutation (p=0.01). Furthermore, the specific abnormality was more pronounced in patients with homozygous JAK2 V617F; median % apoptotic cells of 47% for patients with wild-type allele (n=22) vs. 39% for heterozygotes (n=23) vs. 22% for homozygotes (n=5; p=0.008). The JAK2 mutation status did not appear dependent on other peripheral blood or clinical features. Neutrophils from 14 MMM patients were assessed simultaneously for both JAK2 mutation and STAT3 phosphorylation status by immunoblotting. Strong expression of phosphorylation of STAT3 was seen in all 3 homozygotes and 4 of 5 heterozygotes, but only 1 of 6 with wild-type allele (p=0.026). Conclusions: Impaired neutrophil apoptosis in patients with MMM correlates with the functional presence of JAK2 V617F in an allele-dose dependent manner and STAT3 phosphorylation. The current observation supports a pathogenetic role for the specific mutation in sustaining clonal myeloproliferation.

Abstract T is a Farnesyltransferase inhibitor with clinical activity in elderly adults with newly diagnosed AML who are not candidates for TCC. For adults ≥ age 65 with poor-risk AML, T 600 mg BID for 21 out of every 28–63 days induces complete remissions (CR) in 15%, median duration 7.3 mos, median survival 18 mos. In vitro, T+E exhibits synergistic anti-AML effects. In an attempt to increase CR rates in elderly AML pts, we are conducting a Phase I trial of oral T+E, with escalating doses of both drugs and 14 vs. 21 days of T every 28–63 days (median time to beginning cycle 2 is Day 31, range 29–43). To date, 60 adults ≥ age 70 (median 77, range 71–91) have been treated at 10 dose levels of T (300, 400 or 600 mg BID, 14 or 21 days) + E (100, 150 or 200 mg daily days 1–3 and 8–10) and have received 145 cycles (median 3, range 1–7). A total of 35 (58%) have had secondary AML and 31 (52%) have had adverse cytogenetics. Marrow blast percent was median 55% (range 20–96%), with 75% >30,000 blasts. Nonhematologic toxicities include oropharyngeal mucositis (15% ≥ grade 2), CNS (overall 30%, with 80% of those ≤ grade 2), renal (6%), hyperbilirubinemia (6%), rash (20%) and fatigue (20%). Toxicities occurred across all T+E dose levels tested to date. Documented infections occurred in 16 (27%) pts during cycle 1 and 22/145 (14%) cycles. Twenty six pts (43%) required hospitalization during cycle 1 for median 7 days (range 2–28). Overall hospitalization rate for all cycles was 28% (41/145). Ten pts (17%) died on study: 6 during cycle 1 (4 infection, 1 emphysema, 1 cerebrovascular) and 4 during cycles 2–4 (3 acute myocardial infarctions, 1 ventricular arrhythmia). For 56 evaluable pts, 12 (21%) have achieved CR and 11(20%) have achieved partial remission/hematologic improvement (PR/HI). All pts achieving CR have done so within 2 cycles (9/12 after cycle 1). Eight of 30 pts (27%) receiving T ≥ 400 mg BID have achieved CR. Median age of CR pts is 79 (range 71–86), 8 (67%) have secondary AML, and 5 (42%) have complex cytogenetics. CR durations to date range from 2+ to 13.5+ mos, with median 7+ months. Three of 12 have relapsed at 3, 10, and 11 mos, 2 have been retreated with T+E, and both have achieved 2nd CRs. In sum, the oral regimen of T+E is tolerable and feasible on an outpatient basis for elderly adults with newly diagnosed AML who are not candidates for TCC, with the suggestion of improvement of CR rates over T alone.

Abstract Background: Myelofibrosis with myeloid metaplasia (MMM) currently has limited therapeutic options. BAY 43-9006 (sorafenib; Bayer Pharmaceuticals) is a oral multikinase inhibitor with both inhibitory properties against the Raf/ MEK survival pathway and VEGF that are a potential therapeutic targets for the myeloproliferation and apoptotic resistance seen in this disorder. We evaluated the activity of this agent through inhibition of aberrant in vitro myeloid colony growth, as we have previously successfully utilized in this disorder (Leukemia2003;17(5):849). Methods: Peripheral blood mononuclear cells (PBMC) were isolated and plated on the Methocult™ (StemCell Technologies) colony assay media. Dilutions of BAY 43-9006 (0–10 μM) (Phase I trial sustained concentrations of 10 μM (JCO2005;23:965) was added for either a 24 hour incubation, or for 14 days. Myeloid colonies were then quantified according to standards established by the manufacturer. Subsequent assays were performed combining BAY 43-9006 with the pro-apoptotic agents arsenic trioxide (ATO) or adaphostin (ADA) (both at 1.25μM). Finally, immunoblotting was performed on protein extracted from treated primary MMM cells to quantify the downregulation of Mcl-1 (shown to be crucial for the pro-apoptotic effects of BAY 43-9006; Yu et. al. Oncogene 2005). Results: PBMC were isolated from 12 MMM patients and 10 controls (normal n=5, other myeloproliferative (MPD) n=5). Twenty-four hour exposure of cells to the agent was insufficient to suppress aberrant colony growth. Continuous exposure (for 14 days) of the agent led to a median inhibitory concentration-50% (IC50) of 12 μM for MMM (compared to 7.1μM for normals, and 5μM for MPD controls). Although no specificity could be deduced from these latter results, 4 MMM patients had an IC50 of <10μM (range 1.25 to 8.6μM). In contrast, the other 8 patients were quite resistant IC50 of > 10μM (higher than clinically achievable ranges). Subsequently we evaluated whether there might be benefit to use of BAY43-9006 in combination with a pro-apoptotic agent (i.e. ATO or ADA). No significant activity was seen combining ADA with BAY43-9006. When BAY43-9006 was combined with ATO, the median IC 50 for the BAY compound decreased to 2μM (range 1.25 to 16), however there was not a significant increase in colony inhibition over ATO alone. In order to evaluate whether the lack of activity of BAY43-9006 observed in colony formation assays was secondary to a lack of downregulation of the established target Mcl-1, we immunoblotted for Mcl-1 (after in vitro incubation) in a subset of MMM patients. Western blots demonstrated successful downregulation of Mcl-1 by Bay 43-9006 in MMM samples, even in samples with clear in vitro resistance to the drug (i.e. IC50 > 10 μM). Conclusions: The multi-kinase inhibitor BAY43-9006 appears to have minimal activity against primary cells isolated from MMM patients. Although, suppression of aberrant colony growth was noted in a subset of patients the sensitivities would not support specificity when compared to controls. Although BAY 43-9006 was observed to down regulate the anti-apoptotic target Mcl-1, in patient samples, no detectable inhibition of colony formation was observed in 8/12 MMM samples. These preliminary results do not appear to support a clinical trial of BAY 43-9006 in MMM, however combination of this multi-kinase inhibitor with other pro-apoptotic agents should be evaluated.

Abstract Purpose: Mantle cell lymphoma (MCL) is characterized by a t(11;14) resulting in overexpression of cyclin D1, a member of the phosphatidylinosital 3 kinase (PI3K) pathway. This study tested whether CCI-779, which inhibits the PI3K pathway at the level of the mammalian target of rapamycin (mTOR) could produce tumor responses in patients (pts) with MCL. Patients and Methods: Eligible pts had biopsy-proven, cyclin D1 positive MCL and had relapsed or were refractory to therapy. Pts received CCI-779 250 mg IV every week as a single agent. Pts were re-staged after 1 cycle (4 doses) and every 3 cycles thereafter. Pts with a tumor response after 6 cycles were eligible to continue drug for a total of 12 cycles or 2 cycles after complete remission (CR) and then were observed. Results: Thirty-five pts were enrolled and evaluable for toxicity; 1 patient had MCL by histology but was cyclin D1 negative and ineligible for efficacy evaluation. The median age was 70 years (range, 38–89), 91% were stage 4, and 69% had ≥ 2 extranodal sites. Pts had received a median of 3 prior therapies (range, 1–11) and 54% were refractory to their last treatment. The overall response rate was 38% (13/34) with 1 CR (3%) and 12 PRs (35%), surpassing the pre-defined criteria for a promising agent. Responses tended to occur rapidly with median time to response of 1 month (range, 1–8). To date, 26 patients have progressed, with a median time-to-progression of 6.8 months (95% CI: 3.8 – 9.7). Median duration of response for the 13 responders was 5.7 months (95% CI: 5.2 – 13.2). Overall, 32 out of 35 patients who received treatment had grade 3 or 4 toxicity. The most common toxicities were hematologic with grade 3 (n=24) or grade 4 (n=4). Thrombocytopenia was the most frequent grade 3/4 toxicity (n=25) and the largest cause of dose-reductions, although counts typically recovered within one week. Only 4 patients could tolerate sustained 250 mg per week throughout their treatment (including one who went on to alternate treatment after 1 cycle) and the median dose/month was 175 mg. Conclusions: Single-agent CCI-779 has substantial anti-tumor activity in relapsed MCL. This study demonstrates that agents, which selectively target cellular pathways dysregulated in MCL cells can produce therapeutic benefit. The high response rate warrants further studies of this agent in MCL, but the high incidence of hematologic toxicity suggests that a lower dose should be explored. CCI-779 at 25mg is currently being evaluated in MCL through an NCCTG trial

e18506 Background: Sapanisertib (SAP) is an oral small molecule dual mTOR complex 1 and 2 (TORC1/2) inhibitor that showed activity in lymphoid neoplasms. It blocks a feedback loop involving mTORC2 activation when TORC1 is selectively inhibited (Yun S et al, Blood 2016). SAP induces apoptosis of malignant lymphoid cells in vitro and in xenografts, with greatest effects in ALL cell lines. We aim to assess potential efficacy of SAP when used as a single agent in relapsed/refractory ALL. Methods: Adults (age ≥18) with ECOG performance status 0-2 and normal organ function (creatinine/total bilirubin <1.5 upper limit of normal) with relapsed and/or refractory ALL (B- or T-cell) were enrolled. Newly diagnosed patients (pts) unfit for aggressive treatments and BCR-ABL+ ALL were allowed if there was no suitable standard treatment. SAP was given at 3 mg orally daily for 21 days every 28 days. Dose escalation was allowed after 2 cycles. Hydroxyurea and corticosteroids were not allowed after starting SAP. The primary end point was complete remission (CR) or CR with incomplete count recovery (CRi). This is a one stage Simon design study with interim analysis after 11 pts completed treatment. Enrollment was continued until interim analysis was done. Results: A total of 16 pts were enrolled on the study with a median age of 46 years old, 56% being males. 1/16 (6%) was treatment naïve. Median number of prior therapies is 3 (range, 1-9). The mean duration on SAP was 14.7 days (range 1-22). Three pts were able to start cycle 2. Safety was evaluable in 14 pts. All 14 pts had at least one G3 AE, while 10 pts had at least one G4 AE. G3/4 related non-hematological adverse events (AE) included 3 mucositis, 1 hepatic failure, 1 sepsis, 3 hyperglycemia, 1 anorexia, 1 seizure, 1 confusion, and 1 pneumonitis. One pt had G5 AE with malignancy. G3/4 hematological AEs included 2 with lymphopenia, 1 thrombocytopenia, 3 with leukopenia, and 1 with neutropenia. No pt achieved a CR/CRi. Best response was stable disease in 2 (12.5%) pts. Five pts had progressive disease while 9 were not evaluable because of early removal. Conclusions: SAP exhibited the expected safety profile. Although several pts developed cytopenias on drug, there was no evidence of clinically meaningful single-agent activity against ALL. Analysis of paired bone marrow aspirates from this trial is being performed to assess whether mTOR was substantially inhibited at therapeutically achievable drug concentrations. Clinical trial information: NCT02484430.

5546 Background: CHFR, an E3 ubiquitin ligase that regulates Aurora A and Pololike Kinase 1, plays a critical role in the cellular response to mitotic stress. In particular, cells containing CHFR arrest in G2 when microtubule dynamics are disrupted. Conversely, low CHFR expression is associated with hypersensitivity to mitotic spindle poisons in breast cancer cell lines. Despite the extensive use of taxanes in front line treatment of epithelial ovarian cancer (EOC), little is known about CHFR expression in this disease. We examined CHFR expression and its relationship with clinical characteristics and outcome in women with EOC. Methods: Pretreatment EOC samples from women enrolled in the Mayo Clinic Biospecimen Resource for Ovarian Cancer Research who underwent initial surgical debulking between 1999-2009 were stained in triplicate with anti-CHFR antibodies. Samples were scored in a blinded fashion for CHFR expression. Associations between CHFR staining and histology, grade, and stage were assessed using chi-squared tests; associations between CHFR and overall survival (OS) or time to disease recurrence (TTR) were assessed using Cox proportional hazards models. Results: Samples from 354 women were stained; median age at diagnosis was 60.5 (range 21-93). At a median follow-up of 67 months, 221 women (62%) had died. Across all patients, moderate/strong CHFR staining was associated with poor OS (HR = 1.39, 95% CI: 1.07-1.81, p = 0.015). Moreover, moderate/strong CHFR expression was strongly associated with serous histology (p < 0.0001), high grade (p < 0.0001) and advanced stage (p = 0.0002). To assess whether the association between CHFR expression and OS was related to taxane sensitivity, a subgroup analysis was performed in the subset of patients (N=131) with high-grade serous EOC who received initial platinum/taxane chemotherapy. Moderate/strong CHFR expression failed to correlate with OS (HR=0.91, 95% CI: 0.57-1.45, p = 0.68) or TTR from the start of chemotherapy (HR = 0.85, 95% CI: 0.55-1.31, p = 0.55) in this subset. Conclusions: CHFR expression is associated with poor OS and adverse clinical characteristics in patients with EOC but does not appear to be related to taxane sensitivity in high-grade serous tumors.

5073 Background: Recent molecular exploration of high-grade epithelial ovarian cancer (OC) has revealed potential targets for novel therapy based on altered DNA repair function, deregulated pathways and recurrent amplifications (Cancer Genome Atlas Research Network. 2011. Nature 474). Improved pre-clinical models allowing analysis of specific molecular subsets of ovarian cancer are urgently required to test novel treatment strategies. Methods: We have generated a novel xenograft model of human high-grade serous OC (HG-SOC). Histologic, functional and molecular analysis of the novel xenograft cohort (at baseline and following xenotransplantation) allows stratification of individual HG-SOC for testing with appropriate targeted therapy. We perform functional analysis of in vitro Homologous Recombination (HR) DNA repair and drug response capabilities on fresh human HG-SOC immediately following surgical resection. Molecular classification (similar to Tothill [Clin Canc Res. 2008;14]); analysis of NHEJ pathway (Proc Natl Acad Sci. 2011;108) and other DNA repair genes (Proc Natl Acad Sci USA 2011;108) is performed. In vivo drug response is studied in murine xenografts. Results: Sixteen chemotherapy-naive potentially HG-SOC samples and associated clinical data have been collected. Functional evidence of DNA repair (HR) capability and response to DNA damaging agents will be presented, including IHC for markers of DNA damage (gH2AX), DNA repair (RAD51AP1) and apoptosis (capsase 3 cleavage). Molecular classification, DNA repair gene and DNA repair pathway analyses are underway. Twelve HG-SOC have been transplanted and 6 of the first 8 have successfully xenografted, with serial transplantation and phenotyping of xenograft derivatives underway. In vivo drug response will be presented. Conclusions: This xenograft model will enable us to address hypotheses generated by recent molecular analyses of human HG-SOC (Cancer Genome Atlas Research Network. 2011. Nature 474; Clin Canc Res. 2008;14). Clinical, functional and molecular annotation will allow pre-clinical drug testing based on the plausible hypothesis approach.