Oncogenic PIK3CA Mutations Research Articles

Abstract B37: Understanding DCIS initiation and progression using PDX and genetically engineered mouse and rat models

Abstract Ductal carcinoma in situ (DCIS) is a starting lesion in the milk duct of the breast, which accounts for 25% of all “breast cancers” detected. DCIS is usually treated by surgery combined with radiotherapy, which can have a large impact on the life of patients. However, there is little to no evidence that treatment of low- and intermediate-grade DCIS reduces mortality, while women diagnosed with DCIS do perceive their risk of dying to be the same as patients with invasive disease. To reduce the negative perception and the overtreatment of DCIS, but assure proper treatment for high-risk DCIS, it is crucial to understand the biology underlying DCIS. To model DCIS and the progression to invasive ductal carcinoma (IDC), we took two approaches. 1) We established patient-derived xenograft (PDX) models by intraductally injecting fresh patient DCIS material into the mammary ducts of female immunocompromised mice. To date, 100 human patient DCIS samples were engrafted and incubated in vivo for a period of 12 months. The first 25 of these PDX models had a 72% take rate and show both indolent and invasive tumor growth. Histology and molecular subtyping by PAM50 classification are well preserved in the outgrown PDX lesions compared to the original human counterpart, ensuring that our models are a good representation of the patient disease. On top of this, we profiled copy numbers for both PDX and original patient lesions. Finally, using whole mount imaging, we were able to create 3D sections of the injected mammary glands extracted from our PDX models, showing two distinct growth patterns correlated with invasion. 2) We generated genetically engineered mouse and rat models by intraductal injection of lentiviral vectors (over)expressing known breast cancer driver genes and mutations, or combinations thereof, which are suspected to play a role in DCIS initiation and/or its progression. We were able to successfully and selectively transduce the mammary epithelial cell layer in both mice and rats and are therefore able to somatically engineer the mammary gland in vivo. DCIS formation presented in mice with overexpression of Ccnd1 and Myc only. However, the in situ stage could be observed in rat models with overexpression of the latter genes, as well as oncogenic point mutations in Pik3ca and p53 and all combinations of the tested genetic perturbations. These models will eventually allow us to study the biologic processes underlying DCIS progression. This knowledge could then be used to predict DCIS evolution and distinguish patients with high-risk DCIS from those with low-risk DCIS, which will aid in better care. Citation Format: Stefan Hutten, Catrin Lutz, Colinda Scheele, Madelon Badoux, Timo Eijkman, Stefano Annunziatio, Jelle Wesseling, Jos Jonkers. Understanding DCIS initiation and progression using PDX and genetically engineered mouse and rat models [abstract]. In: Proceedings of the AACR Special Conference on the Evolving Landscape of Cancer Modeling; 2020 Mar 2-5; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2020;80(11 Suppl):Abstract nr B37.

Molecular subtypes in canine hemangiosarcoma reveal similarities with human angiosarcoma.

Angiosarcoma (AS) is a rare neoplasm with limited treatment options and a poor survival rate. Development of effective therapies is hindered by the rarity of this disease. Dogs spontaneously develop hemangiosarcoma (HSA), a common, histologically similar neoplasm. Metastatic disease occurs rapidly and despite chemotherapy, most dogs die several months after diagnosis. These features suggest that HSA might provide a tractable model to test experimental therapies in clinical trials. We previously reported whole exome sequencing of 20 HSA cases. Here we report development of a NGS targeted resequencing panel to detect driver mutations in HSA and other canine tumors. We validated the panel by resequencing the original 20 cases and sequenced 30 additional cases. Overall, we identified potential driver mutations in over 90% of the cases, including well-documented (in human cancers) oncogenic mutations in PIK3CA (46%), PTEN (6%), PLCG1(4%), and TP53 (66%), as well as previously undetected recurrent activating mutations in NRAS (24%). The driver role of these mutations is further demonstrated by augmented downstream signaling crucial to tumor growth. The recurrent, mutually exclusive mutation patterns suggest distinct molecular subtypes of HSA. Driver mutations in some subtypes closely resemble those seen in some AS cases, including NRAS, PLCG1, PIK3CA and TP53. Furthermore, activation of the MAPK and PI3K pathways appear to be key oncogenic mechanisms in both species. Together, these observations suggest that dogs with spontaneous HSA could serve as a useful model for testing the efficacy of targeted therapies, some of which could potentially be of therapeutic value in AS.

Abstract P1-03-11: In vivo rapid discovery of tumor suppressors in breast cancer

Abstract In breast cancer, genome sequencing of patient have identified hundreds of genetic mutations associated with cancer, but which of these mutations drive tumor formation and which are simply bystanders is largely unknown. This is especially true for mutations which individually occur in a small percentage of patients, but when considered together these less common mutations make a significant proportion of the mutational landscape of breast cancer. To functionally test these genes, we performed a CRISPR screen in vivo to identify which of these genes function as tumor suppressors in a mouse model of breast cancer. To do this, we created a library of sgRNAs targeting 215 putative tumor suppressor genes that are found to be mutated in human patients but had not been previously validated as tumor suppressors. We delivered this library via lentivirus to the adult mammary glands of tumor susceptible mice so each lentiviral infected cell has a knockout of one of the genes in the library as well as an oncogenic PIK3CA mutation. Importantly, each infected cell is surrounded by unmodified normal cells, mimicking the initiation of human breast tumors. As expected, mice injected with the putative tumor suppressor library formed tumor significantly faster and reached endpoint sooner than control library injected mice. Tumors from tumor suppressor library injected mice were deep sequenced which revealed known tumor suppressors such as APC and NF1, as well as an enrichment in multiple epigenetic regulators previously unidentified as tumor suppressors in breast cancer. Further validation using independent individual sgRNAs targeting each of the 4 most commonly targeted epigenetic regulators confirmed their role as tumor suppressors. To test the role of these genes in a human context, we used CRISPR to knockout these epigenetic regulators in normal human mammary epithelial MCF10A cells carrying the same PIK3CA mutation. MCF10A lines with epigenetic regulator knockout formed abnormal and protrusive spheres when seeded in matrigel and were capable of forming tumors when engrafted into immunocompromised mice unlike the parent MCF10A line, indicative of transformation into cancerous cells. To identify the mechanism of action of these epigenetic regulators as tumor suppressors, RNA-seq and ATAC-seq were performed on tumors with knockout of epigenetic genes. When compared to control tumors without epigenetic knockout, pathway analysis revealed a role for these epigenetic regulators in EMT, differentiation and cell metabolism. Analysis of ATAC-seq data is currently underway to identify enhancers associated with these transcriptional changes. Additionally, primary cell lines were isolated from knockout tumors and used to perform a drug screen to find drug sensitivities specific to these regulators. This screen identified several compound classes, including CBP/p300 inhibitors which are more effective at killing epigenetic knockout tumor cell lines than control tumor cell lines or normal mouse mammary epithelial (NMUMG) cells. Overall, these data are the first to identify several epigenetic regulators as tumor suppressors in breast cancer and to test them in a parallel and syngeneic background to establish their mechanism and drug susceptibility. By studying these genes, we seek to not only find treatments that target tumors in which these genes are mutated but also to understand the greater role of epigenetic regulators in tumor suppression. Citation Format: Ellen R Langille, Sampath Loganathan, Daniel Schramek. In vivo rapid discovery of tumor suppressors in breast cancer [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P1-03-11.

Precision medicine for ovarian clear cell carcinoma based on gene alterations.

Ovarian clear cell carcinoma (OCCC) is a histological subtype of epithelial ovarian carcinoma prevalent in Asians. No clear therapeutic selection based on molecular profile has been implemented for this disease. Oncogenic PIK3CA mutation, which activates the PIK3CA/AKT/mTOR signaling pathway, is a promising druggable alteration in OCCC. Recent studies by our group and others have identified the ARID1A mutation as another alteration linked to therapeutic selection based on synthetic lethality: deleterious ARID1A mutations, resulting in ARID1A deficiency, make OCCC cells sensitive to drugs targeting poly (ADP-ribose) polymerase and EZH2, as well as to glutathione inhibitors. In addition, we recently obtained evidence that ARID1A-deficient OCCC could benefit from gemcitabine treatment. Precision medicine based on gene alteration profiling might improve the prognosis of OCCC patients.

High Frequency of PIK3CA Mutations in Low-Grade Serous Ovarian Carcinomas of Japanese Patients.

The frequency of KRAS/BRAF mutations associated with low-grade serous ovarian carcinoma (LGSC)/serous borderline tumors (SBTs) in Japan is unknown. We aimed to identify genetic variations in KRAS, BRAF, PIK3CA, and ERBB2 in LGSC/SBT/serous cystadenomas (SCAs) in a Japanese population. We performed a mutation analysis (by Sanger sequencing) of 33 cases of LGSC/SBT/SCA and 4 cases of LGSC with synchronous SBTs using microdissected paraffin-embedded sections. Immunohistochemistry of p53 and ARID1A was also performed. The frequency of oncogenic mutations in PIK3CA was 60.0% (6/10) in LGSCs, 63.6% (7/11) in SBTs, and 8.3% (1/12) in SCAs. All cases harbored wild-type KRAS. The frequency of BRAF mutations was 20.0% (2/10) in LGSCs, whereas all SBTs and SCAs harbored the wild-type allele. The frequency of ERBB2 mutations was 30.0% (3/10) in LGSCs, 0.0% (0/11) in SBTs, and 16.7% (2/12) in SCAs. ARID1A staining was positive in all cases. p53 staining was positive in 0% (0/10) LGSCs, 9.1% (1/11) SBTs, and 0.0% (0/12) SCAs. One LGSC case had two PIK3CA mutations (G1633A and G3149A) in both LGSC and SBT lesions, but a BRAF mutation was detected only in an LGSC lesion. These results suggest that, compared with the values in Western populations (16–54%), the KRAS mutation frequency in LGSCs/SBTs is lower and that of PIK3CA mutations in LGSCs/SBTs is much higher in Japanese populations. Therefore, the main carcinogenesis signaling pathways may be different between Japanese and Western LGSCs. Molecular therapies targeting the PIK3CA/AKT pathway may be effective in LGSCs in Japan.

High MET Overexpression Does Not Predict the presence of MET exon 14 Splice Mutations in NSCLC: Results From the IFCT PREDICT.amm study.

MET proto-oncogene (MET) exon 14 splice site (METex14) mutations were recently described in NSCLC and has been reported to correlate with efficacy of MET tyrosine kinase inhibitors. High diversity of these alterations makes them hard to detect by DNA sequencing in clinical practice. Because METex14 mutations induce increased stabilization of the MET receptor, it is anticipated that these mutations are associated with MET overexpression. We aim to determine whether NSCLC with high MET overexpression could define a subset of patients with a high rate of METex14 mutations. From The French Cooperative Thoracic Intergroup PREDICT.amm cohort of 843 consecutive patients with a treatment-naive advanced NSCLC who were eligible for a first-line therapy, 108 NSCLC samples with high MET overexpression defined by an immunochemistry score 3+ were tested for METex14 mutations using fragment length analysis combined with optimized targeted next-generation sequencing. MET copy number analysis was also derived from the sequencing data. METex14 mutations were detected in two patients (2.2%) who also displayed a TP53 mutation and a PIK3CA mutation, respectively. An MET gene copy number increase was observed in seven additional patients (7.7%). Next-generation sequencing analysis revealed inactivating mutations in TP53 (52.7%) and PTEN (1.1%), and oncogenic mutations in KRAS (28.6%), EGFR (7.7%), PIK3CA (4.4%), BRAF (4.4%), NRAS (2.2%), GNAS (1.1%), and IDH1 (1.1%). The rate of METex14 mutations in NSCLC with high MET overexpression was similar to that found in unselected NSCLC. Moreover, we observed a high frequency of driver alterations in other oncogenes. Consequently these findings do not support the use of MET immunohistochemistry as a surrogate marker for METex14 mutations.

Perspective: Potential Impact and Therapeutic Implications of Oncogenic PI3K Activation on Chromosomal Instability.

Genetic activation of the class I PI3K pathway is very common in cancer. This mostly results from oncogenic mutations in PIK3CA, the gene encoding the ubiquitously expressed PI3Kα catalytic subunit, or from inactivation of the PTEN tumour suppressor, a lipid phosphatase that opposes class I PI3K signalling. The clinical impact of PI3K inhibitors in solid tumours, aimed at dampening cancer-cell-intrinsic PI3K activity, has thus far been limited. Challenges include poor drug tolerance, incomplete pathway inhibition and pre-existing or inhibitor-induced resistance. The principle of pharmacologically targeting cancer-cell-intrinsic PI3K activity also assumes that all cancer-promoting effects of PI3K activation are reversible, which might not be the case. Emerging evidence suggests that genetic PI3K pathway activation can induce and/or allow cells to tolerate chromosomal instability, which—even if occurring in a low fraction of the cell population—might help to facilitate and/or drive tumour evolution. While it is clear that such genomic events cannot be reverted pharmacologically, a role for PI3K in the regulation of chromosomal instability could be exploited by using PI3K pathway inhibitors to prevent those genomic events from happening and/or reduce the pace at which they are occurring, thereby dampening cancer development or progression. Such an impact might be most effective in tumours with clonal PI3K activation and achievable at lower drug doses than the maximum-tolerated doses of PI3K inhibitors currently used in the clinic.

Abstract LB-294: Targeting glutamine addiction of PIK3CA mutant colorectal cancers: From preclinical models to clinical trials

Abstract Glutamine addiction is a major metabolic reprogramming event that occurs in cancer cells. Many tumors exhibit oncogene-dependent addiction to glutamine. As a metabolite, glutamine is first converted to glutamate by glutaminase (GLS) and then to α-ketoglutarate (α-KG) by either a transaminase or a glutamate dehydrogenase. PIK3CA, which encodes the p110α catalytic subunit of phosphatidylinositol 3-kinase α, is frequently mutated in human cancers including 20 to 30% of colorectal cancer (CRC). We found that oncogenic PIK3CA mutations render CRCs addicted to glutamine through an upregulation of glutamate pyruvate transaminase 2 (GPT2), thereby facilitating conversion of glutamate to α-KG. Here, we report that CB-839, a potent GLS inhibitor developed by Calithera Biosciences, inhibits xenograft tumor growth of PIK3CA mutant, but not WT, colorectal cancers. Remarkably, combination of CB-839 with 5-fluorouracil (5-FU) induces regression of xenograft tumors in four different PIK3CA mutant CRC models. These exciting preclinical data prompted us to perform a Phase I and Phase II clinical trial to assess combination of CB-839 with capecitabine, an oral prodrug of 5-FU. For the Phase I trial, patients with advanced solid tumors and disease progression on standard treatment, or for whom capecitabine is an acceptable treatment option, were enrolled. A standard 3+3 design was used to assess escalating doses of CB-839 (400-800 mg) and capecitabine (500-1000 mg/m2) in four dose levels. CB-839 was given orally twice daily continuously and capecitabine was given orally twice daily on days 14/21 of a 21 day treatment cycle. Dose-limiting toxicity period was 21 days. Restaging CT scans were obtained every 9 weeks with response assessment per RECIST. Adverse events were assessed per CTCAEv4. Sixteen patients were enrolled with 7 patients being treated at the final dose level (CB-839 800 mg and Capecitabine 1000 mg/m2). No dose-limiting toxicity were observed. Ten patients achieved stable disease as their best response with 9 being fluoropyrimidine resistant. Among the 16 patients, 12 were CRC patients who were 5-FU resistant. Consistent with our preclinical data, the median progress-free survival (PFS) of seven PIK3CA mutant CRC patients was 25 weeks, whereas the median PFS of five PIK3CA WT CRC patients was 16 weeks. Notably, the PFS of three PIK3CA mutant CRC patients was over six months (8, 9 and 13 months respectively). A Phase II trial of the drug combination in metastatic PIK3CA mutant CRC patients is ongoing. In summary, our preclinical and clinical studies suggest that targeting glutamine metabolism may be a promising approach to treat CRCs with PIK3CA mutations. Citation Format: Yiqing Zhao, Jennifer Eads, David Bajor, zhenghe wang. Targeting glutamine addiction of PIK3CA mutant colorectal cancers: From preclinical models to clinical trials [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr LB-294.

Abstract 3917: Double PIK3CA mutations in cis enhance oncogene activation and sensitivity to PI3K alpha inhibitors in breast cancer

Abstract Activating mutations in PIK3CA, the gene coding for the catalytic subunit (p110α) of phosphoinositide 3-kinase (PI3K), are the most frequent oncogenic alterations in all cancers, including estrogen receptor-positive (ER+) breast cancer. There are many distinct oncogenic PIK3CA mutations including major hotspot mutations (e.g. E542K, E545K, H1047R) which are predictive of response to PI3Kα inhibitors. While responses to PI3Kα inhibitors can be variable, some patients with single hotspot mutant tumors derive a deep and prolonged clinical benefit. In a comprehensive analysis of the genomic data available in breast cancer, we observed double PIK3CA mutations and investigated their potential biological relevance and potential correlation with sensitivity to PI3Kα inhibitors. We detected double PIK3CA mutations in 10-15% of all PIK3CA-mutant cancers including breast cancer. Double PIK3CA mutations in breast cancer are defined by a pattern of co-expression of a major hotspot mutation associated with a recurrent second-site PIK3CA mutation (‘minor mutation’). We have found that these double PIK3CA mutations are compound, that is in cis on the same allele, using single molecule real time sequencing (SMRT-seq) of fresh breast tumor samples. Double compound PIK3CA mutations result in increased PI3K activity and downstream signaling compared to single hotspot mutants in nontransformed cells and in ER+ breast cancer cells. Double compound mutations also increase cell proliferation and xenograft growth compared to single hotspot mutants. Biochemical experiments using recombinant PI3K compound mutant protein complexes reveal a combination mechanism of action, with simultaneous PI3K destabilization and increased membrane binding compared to single hotspot mutants, leading to increased kinase activity. Importantly, these compound mutations predict for increased sensitivity to PI3Kα inhibitors compared to single hotspot mutants in both preclinical models and also in patients with breast cancer. Together, our data support that double compound PIK3CA mutations enhance the oncogenicity of single hotspot PIK3CA mutations and this increased dependency results in increased sensitivity to PI3Kα inhibitors compared to single hotspot mutations. We propose that double compound activating mutations are an additional mechanism of oncogenic transformation in addition to other known mechanisms such as gene amplification, point mutation, or gene translocation. We speculate that double compound mutant PIK3CA may function as a clinical biomarker of increased sensitivity to PI3K-directed targeted therapies and may improve the therapeutic window of PI3K inhibitors in ER+ breast cancer and other PIK3CA-mutant tumor histologies. Citation Format: Neil Vasan, Pedram Razavi, Jared Johnson, Hong Shao, Hardik Shah, Alesia Antoine, Erik Ladewig, Alexander Gorelick, Eneda Toska, Guotai Xu, Abiha Kazmi, Matthew T. Chang, Barry S. Taylor, Maura N. Dickler, Komal Jhaveri, Raul Rabadan, Ed Reznik, Melissa L. Smith, Robert Sebra, Lewis C. Cantley, Maurizio Scaltriti, Jose Baselga. Double PIK3CA mutations in cis enhance oncogene activation and sensitivity to PI3K alpha inhibitors in breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 3917.

Establishment and genomic characterization of gingivobuccal carcinoma cell lines with smokeless tobacco associated genetic alterations and oncogenic PIK3CA mutation

Smokeless tobacco associated Gingivobuccal squamous cell carcinoma (GB-SCC) is a major public health problem but available oral cancer cell lines are mostly from smoking associated tongue SCC raising the need for pertinent GB-SCC cell line models. As part of the International Cancer Genome Consortium (ICGC) Project, 4 novel cell lines, namely, Indian Tata Memorial Centre Oral Cancer (ITOC) −01 to −04 were established and characterized with conventional methods, karyotyping, ultrastructure, in vivo tumourigenicity, Whole exome sequencing (WES) and RNA sequencing. These hyperploid cell lines form xenografts in mice and show metabolically active and necrotic areas on fluorodeoxyglucose-positron emission tomography (FDG-PET) imaging. WES of ITOC cell lines recapitulate the genomic tumor profile of ICGC GB-SCC database. We further identified smokeless tobacco associated genetic alterations (PCLO, FAT3 and SYNE2) and oncogenic PIK3CA mutation in GB-SCC cell lines. Transcriptome profiling identified deregulation of pathways commonly altered in cancer and down-regulation of arachidonic acid metabolism pathway, implying its possible role in GB-SCC. Clinical application of high throughput sequencing data depends on relevant cell line models to validate potential targets. Extensively characterized, these oral SCC cell lines are particularly suited for mechanistic studies and pre-clinical drug development for smokeless tobacco associated oral cancer.

Histomolecular characterization of intracranial meningiomas developed in patients exposed to high-dose cyproterone acetate: an antiandrogen treatment.

BackgroundMeningiomas are the most common primary intracranial tumors in adults. The relationship between meningiomas and exogenous sex hormones such as cyproterone acetate (CPA) is well documented, yet the underlying mechanisms remain unknown. Defining the histomolecular status of meningiomas developed on CPA would help us to better understand the oncogenesis of these tumors.MethodsWe identified 30 patients operated for a meningioma after long-term high-dose CPA therapy and with a history of CPA discontinuation before establishing the indication for surgical intervention. We used array-comparative genomic hybridization (to characterize copy number changes in those 30 meningiomas and subsequently performed next-generation sequencing with the National Institute of Cancer (INCa) solid tumor panel, which is a targeted panel of clinically actionable genes. We also examined grade, type, and clinical features.ResultsWe identified AKT1 mutations or PIK3CA mutations in 33.3% of CPA meningiomas. AKT1 and PIK3CA mutations were mutually exclusive. Enrichment in oncogenic PIK3CA mutations in the CPA cohort was detected. CPA meningiomas showed chromosomal stability and were located mainly in the skull base. Ninety percent of CPA meningiomas were low-grade meningiomas and 63.4% were meningotheliomas. Half of our CPA cohort had microcystic components.ConclusionOur study shows that low-grade meningothelial meningiomas of the skull base are predominant in CPA meningiomas. We identified PIK3CA/AKT1 pathway as a hypothetical actor in onco-pharmacological interaction between meningiomas and CPA. This signaling pathway could be an interesting target for precision medicine trials in meningioma patients who have been subjected to CPA. Our results could invite the scientific community to review the current classification of meningiomas and to evolve toward more specific histomolecular classification.

PIK3CA mutations in vascular malformations.

Recently, it has been discovered that a subset of vascular malformations, of the lymphatic and venous type, are caused by oncogenic mutations in the PIK3CA gene. Now, efforts have been focused in the understanding of the molecular and cellular consequences of these mutations and the opportunities for novel-targeted therapies for these diseases. Here, we review the latest findings in the biology of oncogenic PIK3CA mutations in the pathogenesis of vascular malformations. We focus on the recent development of in-vitro and in-vivo tools for the study of PIK3CA-mutant vascular malformations with special interest in preclinical models for drug testing. Also, we review the latest advances in phosphoinositide 3-kinase (PI3K) inhibitors in the clinic and their repurposing for the treatment of lymphatic malformations and venous malformations. Oncogenic mutations on PIK3CA causing lymphatic malformations and venous malformations are also frequently found in epithelial cancer. Thus, fundamental research done in the cancer field during the past decades might be applied to the understanding of lymphatic malformations and venous malformations. Likewise, repurposing PI3K pathway inhibitors that are currently in cancer clinical trials can be used as a novel strategy for the treatment of these diseases. Here, we also open a debate for the consideration of lymphatic malformations and venous malformations as developmental tumours.

Oncogenic PIK3CA mutations increase dependency on the mRNA cap methyltransferase, RNMT, in breast cancer cells.

Basic mechanisms in gene expression are currently being investigated as targets in cancer therapeutics. One such fundamental process is the addition of the cap to pre-mRNA, which recruits mediators of mRNA processing and translation initiation. Maturation of the cap involves mRNA cap guanosine N-7 methylation, catalysed by RNMT (RNA guanine-7 methyltransferase). In a panel of breast cancer cell lines, we investigated whether all are equivalently dependent on RNMT for proliferation. When cellular RNMT activity was experimentally reduced by 50%, the proliferation rate of non-transformed mammary epithelial cells was unchanged, whereas a subset of breast cancer cell lines exhibited reduced proliferation and increased apoptosis. Most of the cell lines which exhibited enhanced dependency on RNMT harboured oncogenic mutations in PIK3CA, which encodes the p110α subunit of PI3Kα. Conversely, all cell lines insensitive to RNMT depletion expressed wild-type PIK3CA. Expression of oncogenic PIK3CA mutants, which increase PI3K p110α activity, was sufficient to increase dependency on RNMT. Conversely, inhibition of PI3Kα reversed dependency on RNMT, suggesting that PI3Kα signalling is required. Collectively, these findings provide evidence to support RNMT as a therapeutic target in breast cancer and suggest that therapies targeting RNMT would be most valuable in a PIK3CA mutant background.

Abstract ES6-2: PI3K in breast cancer, glutamine changes and methionine addiction

Abstract Mutations in the PI 3-kinase and AKT signaling pathway are highly prevalent in breast cancer. In luminal, estrogen receptor-positive breast cancer patients, oncogenic mutations in PIK3CA, the gene that encodes the catalytic subunit of class 1a PI3K occur with a frequency of approximately 40%. By contrast, inactivating mutations and loss of heterozygosity in the tumor suppressors PTEN and INPP4B that normally terminate PI3K signaling, are found more frequently in basal-like, triple negative breast tumors. Regardless, in all cases the net result is hyperactivation of PI3K/AKT signaling leading to induction of phenotypes associated with malignancy. In spite of the frequency of genetic lesions in the PI3K pathway and the development of potent and specific small molecule inhibitors targeting the PI3K pathway, results from clinical trials with these inhibitors have been disappointing most often due to dose-limiting on-target toxicities. Consequently, numerous combination trials with PI3K inhibitors and other targeted therapies are underway. Separate pre-clinical and clinical efforts have also been initiated targeting PI3K oncogene addition with therapies that impact cellular states and programs, including metabolic pathways and enzymes. The concept that metabolic reprogramming is a feature of all cancer cells, and not merely a byproduct of accelerated growth, goes back to the days of Otto Warburg who in the 1920s proposed that increased levels of aerobic glycolysis followed by lactate fermentation occurs in tumor cells undergoing rapid growth. This so-called 'Warburg effect' spurred the recent development of numerous drugs that target glycolysis in human tumors, many of which are in clinical trials. Equally promising are strategies that target non-essential amino acid metabolism. The concept of metabolic starvation, whereby one or more metabolite is limiting or removed has gained traction in the field, with the finding that certain non-essential amino acids, particularly glutamine, can be synthesized by normal cells, whereas tumor cells rely on an extracellular supply to the extent that glutamine deprivation sensitizes tumor cells to death. Glutamine is the most widely studied nutrient, other than glucose, in the field of tumor metabolism. The presentation will focus on the intersection of PI3K, AKT and glutamine metabolism in breast cancer, as well as how these pathways modulate methionine dependency in breast cancers that display PI3K pathway hyperactivation. A greater understanding of the mechanisms by which glutamine and methionine sustain cancer cells may lead to the development or repurposing of other successful targeted metabolic therapies and use of the pathway enzymes and metabolites as biomarkers. Citation Format: Toker A. PI3K in breast cancer, glutamine changes and methionine addiction [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr ES6-2.

Abstract P4-01-11: Genomic alterations of cell-free DNA in early breast cancer patients with recurrence

Abstract Cell-free DNA (cfDNA), as a non-invasive strategy, provides substantial benefit to overcome tumor heterogeneity. Surveillance of recurrence after standard treatment in early breast cancer (BC) using cfDNA, enables to detect minimal residual disease (MRD), also to identify genomic alterations driving recurrences. We aimed to assess the role of cfDNA in detecting MRD by investigating genomic alterations of 1)primary, recurred tumor and 2)cfDNA at time of recurrence using deep targeted sequencing. Fifty-four early BC patients were enrolled prospectively between 2014 and 2017 at time of recurrence. Median disease free interval was 28.5 months (rage 6.2-49.8). 62.7% (32/51) were hormone receptor (HR) positive (28 HRpos/HER2neg, 4 HRpos/HER2pos), 11.8% (6/51) were HRneg/HER2pos and 25.5% (13/51) were triple negative BCs. 59.3% (32/54) patients developed loco-regional recurrence (15 local recurrence only, 13 regional only, 4 with both) and distant metastasis was observed among 40.7% (22/54) patients. Cell-free DNA was extracted from 5cc blood at time of recurrence. Deep targeted sequencing was performed using customized NGS panel –encompassing 426 cancer-related target coding region, 242 fusion and amplification-related region- of cfDNA and FFPE(formalin fixed paraffin embedded) tumor samples archived from surgical resection or biopsy. Deep targeted sequencing data was successfully performed in 72.1% (31/43) plasma samples and sequencing yield was significantly lower when stored for more than 2yrs (46.2% vs 83.3%). Mutations of cfDNA and tumor (primary, recurred) were analyzed. Mean sequencing depth of cfDNA and FFPE were x425.7 and x777.6 respectively. Median number of pathogenic mutations found in primary tumor, cfDNA and recurred tumor were 27(range 12-99), 25(range 8-85) and 9(range 0-23). Among mutations found in primary tumor, 27.4% were shared mutations (range 8.1%-72.7%) with recurred tumor and 26.1% were shared mutations (range 4.7%-69.2%) observed in cfDNA sample. Among mutations found in recurred tumor, 40.9% were observed in cfDNA (range 17.7-87.5%). In primary tumor, median number of mutations with allelic fraction (MAF)>10% were 12 (range 4-21) and at least one mutation was found in cfDNA at time of recurrence. Among mutations with MAF>10%, 59.4% and 69.1% were found in cfDNA and recurred tumor. Known oncogenic mutations of PIK3CA, TP53, GATA3, AKT1, ESR1, RELN, ERBB2, ERBB3, BRCA1 mutation were found. PIK3CA gene (p.H1047R) was found in two cases both in primary tumor and cfDNA at recurrence (MAF 11.4% vs 5.3% and 12.3% vs 15.4%) suggesting de novo driver mutation. One patient developed regional recurrence during adjuvant aromatase inhibitor with ESR1 V392I mutation in both cfDNA and recurred tumor (MAF 48.1 and 54.5%), while another patient's recurred tumor during aromatase inhibitor harbored ESR1 D538G mutation exclusively in recurred tumor with MAF <1%. Both patients had no ESR1 hotpot mutation in primary tumor. Our data showed sequencing yield of 83.3% in plasma samples within 2yr. Pathogenic mutations in primary tumor, especially when MAF>10%, half of them was observed in cfDNA at time of recurrence. ESR1 mutation should be included in cfDNA surveillance for patients undergoing endocrine therapy even absent in primary tumor. Citation Format: Kim J, Jo WK, Kim KY, Kim BJ, Lee SB, Lee HJ, Yu JH, Kim HJ, Chung IY, Ko BS, Kim S-B, Jung KH, Ahn JH, Chang S, Lee JW, Son BH, Ahn SH. Genomic alterations of cell-free DNA in early breast cancer patients with recurrence [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P4-01-11.

PI3K oncogenic mutations mediate resistance to afatinib in HER2/neu overexpressing gynecological cancers

ObjectiveAberrant expression of HER2/neu and PIK3CA gene products secondary to amplification/mutations are common in high-grade-serous-endometrial (USC) and ovarian-cancers (HGSOC). Because scant information is currently available in the literature on the potential negative effect of PIK3CA mutations on the activity of afatinib, in this study we evaluate for the first time the role of oncogenic PIK3CA mutations as a potential mechanism of resistance to afatinib in HGSOC and USC overexpressing HER2/neu. MethodsWe used six whole-exome-sequenced primary HGSOC/USC cell-lines and three xenografts overexpressing HER2/neu and harboring mutated or wild-type PIK3CA/PIK3R1 genes to evaluate the role of PI3K-mutations as potential mechanism of resistance to afatinib, an FDA-approved pan-c-erb-inhibitor in clinical trials in USC. Primary-USC harboring wild-type-PIK3CA gene was transfected with plasmids encoding oncogenic PIK3CA-mutations (H1047R/E545K). The effect of afatinib on HER2/PI3K/AKT/mTOR pathway was evaluated by immunoblotting. ResultsWe found PI3K wild-type cell-lines to be significantly more sensitive (lower IC50) than PI3K-mutated cell-lines p = 0.004). In vivo, xenografts of primary cell-line USC-ARK2, transfected with the PIK3CA-H1047R or E545K hotspot-mutations, exhibited significantly more rapid tumor growth when treated with afatinib, compared to mice harboring ARK2-tumors transfected with wild-type-PIK3CA (p = 0.041 and 0.001, respectively). By western-blot, afatinib effectively reduced total and phospho-HER2 proteins in all cell-lines. However, H1047R/E545K-PIK3CA-transfected-ARK2-cells demonstrated a greater compensatory increase in phosphorylated-AKT proteins after afatinib exposure when compared to controls ARK2. ConclusionsOncogenic PI3K mutations may represent a major mechanism of resistance to afatinib. Combinations of c-erb with PIK3CA, AKT or mTOR inhibitors may be necessary to more efficiently block the PIK3CA/AKT/mTOR pathway.

New Mechanisms of mTOR Pathway Activation in KIT-mutant Malignant GISTs

A great majority of gastrointestinal stromal tumors (GISTs) are primarily driven by gain-of-function KIT receptor tyrosine kinase mutations that subsequently lead to activation of phosphatidiylinositol 3-kinase (PI3K)/mammalian target of rapamycin (mTOR) pathway, a downstream effector of KIT signaling. KIT tyrosine kinase inhibitor, imatinib mesylate, has been successfully used for the treatment of primary, advanced, and disseminated GISTs. Recently, activation of mTOR pathway independent of KIT signaling was demonstrated in imatinib mesylate naïve malignant GISTs and treatment-resistant metastatic tumors. This activation was attributed to oncogenic mutations in PIK3CA encoding PI3K 110α subunit, or to the inactivation of PTEN tumor suppressor, a potent mTOR negative regulator. In this study, mTOR pathway genes were evaluated in 14 imatinib mesylate naïve, KIT-mutant, malignant small intestinal GISTs using next-generation sequencing. Mutations were detected in 3 (21%) of 14 analyzed tumors: (1) c.3200A>T substitution in PIK3CB encoding PI3K 110β subunit, (2) c.1040A>G substitution in tuberous sclerosis complex (TSC2) encoding tuberin, mTOR down-regulator (3) c.6625C>G substitution in mTOR. At the protein level, these changes were predicted to cause, respectively, PIK3CB p.D1067V, TSC2 p.K347R, and mTOR p.L2209V mutations. Previously reported "in vitro" experiments with mouse 3T3 fibroblasts demonstrated oncogenic potential of PIK3CB p.D1067V and mTOR p.L2209V mutants; whereas, PolyPhen-2 software analysis predicted TSC2 p.K347R mutation to likely have a damaging impact on tuberin function. The results of this and previous studies indicate diversity of genetic changes leading to activation of PI3K-AKT-TSC-mTOR pathway in malignant GISTs. Extensive genotyping of the genes involved in mTOR pathway demonstrates common alterations that need to be considered in targeted treatment.

E-cadherin loss induces targetable autocrine activation of growth factor signalling in lobular breast cancer

Despite the fact that loss of E-cadherin is causal to the development and progression of invasive lobular carcinoma (ILC), options to treat this major breast cancer subtype are limited if tumours develop resistance to anti-oestrogen treatment regimens. This study aimed to identify clinically targetable pathways that are aberrantly active downstream of E-cadherin loss in ILC. Using a combination of reverse-phase protein array (RPPA) analyses, mRNA sequencing, conditioned medium growth assays and CRISPR/Cas9-based knock-out experiments, we demonstrate that E-cadherin loss causes increased responsiveness to autocrine growth factor receptor (GFR)-dependent activation of phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K)/Akt signalling. Autocrine activation of GFR signalling and its downstream PI3K/Akt hub was independent of oncogenic mutations in PIK3CA, AKT1 or PTEN. Analyses of human ILC samples confirmed growth factor production and pathway activity. Pharmacological inhibition of Akt using AZD5363 or MK2206 resulted in robust inhibition of cell growth and survival of ILC cells, and impeded tumour growth in a mouse ILC model. Because E-cadherin loss evokes hypersensitisation of PI3K/Akt activation independent of oncogenic mutations in this pathway, we propose clinical intervention of PI3K/Akt in ILC based on functional E-cadherin inactivation, irrespective of activating pathway mutations.

The combination of everolimus and terameprocol exerts synergistic antiproliferative effects in endometrial cancer: molecular role of insulin-like growth factor binding protein 2.

Oncogenic PIK3CA mutations are common in endometrial cancers, and the PI3K/AKT/mTOR pathway is targetable by drugs. We sought to investigate whether the combination of an mTOR inhibitor, everolimus (RAD001), and an AKT inhibitor, terameprocol (M4N), exerts better antiproliferative effects in endometrial cancer. The molecular mechanisms underlying their pharmacological action were also examined. The combination of RAD001 and M4N exerted in vitro synergistic effects on cell viability, apoptosis, and expression of IGFBP2 in endometrial cancer cells. Mechanistically, the Sp1 site on the IGFBP2 promoter was required for RAD001- and M4N-induced downregulation. IGFBP2 protein expression was higher in endometrial cancer than in the normal endometrium (P < 0.001). Furthermore, elevated IGFBP2 histoscores were significantly associated with a lower overall survival (P = 0.021). In conclusion, our in vitro results demonstrate that RAD001 and M4N exert synergistic antiproliferative effects against endometrial cancer cells, which appeared to be mediated by the inhibition of IGFBP2, a key anti-apoptotic regulator. Further clinical studies of this drug combination in patients with endometrial cancer may be warranted, especially in the presence of PIK3CA and IGFBP2 aberrations. KEY MESSAGES: RAD001 and M4N synergistically suppress endometrial cancer growth. IGFBP2 is overexpressed in endometrial cancer. The combination of RAD001 and M4N significantly reduces IGFBP2 overexpression. Sp1 binding site on the IGFBP2 promoter is required for RAD001- and M4N-induced downregulation. High IGFBP2 histoscore in endometrial cancer portends a poor prognosis.

Abstract 5349: Research in correlation of ADAR and AIMP2 DX2 using pan-cancer analysis

Abstract AIMP2 is an interacting multifunctional protein complex of aminoacyl transferRNA synthase (ARS), a protein synthetase that binds to tRNA by activating amino acids. AIMP2 plays an important role as a tumor-suppressor gene, forming a complex with p53 and exon 2 deficient AIMP2 generated by alternative splicing mechanism (AIMP2-DX2) promotes carcinogenesis and loses its important role as a signaling modulator. On the other hand, the ADAR gene is a dsRNA-specific binding protein that triggers adenosine-to-inosine editing according to the hydrolytic adenosine deaminase activity and play an important role maintaining normal splicing activity. We studied the association between ADAR activity and AIMP2-DX2 generation. We used International Cancer Genome Consortium (ICGC) genomic data of cancer patients for this analysis: COAD-US (n=44), LUAD-US (n=42), and LUSC-US(n=48). Tophat and Cufflinks were used to map the raw RNA sequencing data of three carcinomas to a human reference and analyze for alternative transcription and splicing. The clustering algorithms-based Database for Annotation, Visualization and Integrated Discovery (DAVID) was used to find enrichment in pathway with lists of genes from each group. The difference in the genetic variation patterns between the two groups were determined using "deconstructSigs" package in R which best interprets the sample's mutational profile. In whole ICGC RNA sequencing data, the amount of AIMP2 DX2 negatively correlated with ADAR expression. Subgroup analysis was performed for lung cancer and colon cancer samples based on AIMP2 and AIMP2-DX2 expression levels. Inflection point of AIMP2-DX2 expression curve was used as a cut-off value, and this value divided samples into two groups (7 samples with high AIMP2-DX2 and 121 samples with low AIMP2-DX2). We could observe obvious contribution of downregulated ADAR in high AIMP2-DX2 samples. When we examined mutations enriched in each group and the mutational pattern differences, mutations of tumor-suppressor genes including TP53 (71.4%) and APC (57.1%) were enriched in group of high AIMP2-DX2. The oncogenic mutations of KRAS (42.9%) and PIK3CA (28.6%) were also enriched in AIMP2-DX2 high patient group. The mutations enriched in group of highly expressed AIMP2-DX2 were further analyzed by DAVID functional annotation analysis, and MAPK signaling pathway and BASE excision repair signaling was discovered. Furthermore, the difference in the genetic variation patterns between the two groups was inspected based on COSMIC database. Mutational signature 9 related to hyper-mutation in immunoglobulin was expressed higher in AIMP2-DX2 elevated samples. Signatures 17 and 18 were also expressed with DX2 high samples. We compared the mutational enrichment and mutational pattern differences between the two groups according to the expression of the ADAR gene and verified the effect of the relationship between ADAR gene and AIMP2 DX2 on cancer development. Citation Format: Seulki Song, Daeyoon Kim, Yoosik Kim, Sung-Soo Yoon, Sunghoon Kim, Youngil Koh. Research in correlation of ADAR and AIMP2 DX2 using pan-cancer analysis [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 5349.