PI3K/AKT/Mtor Pathway Alterations Promote Malignant Progression

Published OnlineFirst April 11, 2019; DOI: 10.1158/1078-0432.CCR-18-4144

Translational Cancer Mechanisms and Therapy Clinical Cancer Research PI3K/AKT/mTOR Pathway Alterations Promote Malignant Progression and Xenograft Formation in Oligodendroglial Tumors Kensuke Tateishi1,2,3,4, Taishi Nakamura1,4, Tareq A. Juratli3,5, Erik A. Williams3,6, Yuko Matsushita2, Shigeta Miyake1,4, Mayuko Nishi7, Julie J. Miller3,8, Shilpa S. Tummala3,5, Alexandria L. Fink3,5, Nina Lelic3,5, Mara V.A. Koerner3,5, Yohei Miyake1,4, Jo Sasame1,4, Kenji Fujimoto2, Takahiro Tanaka1, Ryogo Minamimoto9, Shigeo Matsunaga10, Shigeo Mukaihara11, Takashi Shuto1,10, Hiroki Taguchi12, Naoko Udaka13, Hidetoshi Murata1, Akihide Ryo7, Shoji Yamanaka13, William T. Curry5, Dora Dias-Santagata6, Tetsuya Yamamoto1, Koichi Ichimura2, Tracy T. Batchelor3,8, Andrew S. Chi14, A. John Iafrate3,6, Hiroaki Wakimoto3,5, and Daniel P. Cahill3,5

Abstract

Purpose: Oligodendroglioma has a relatively favorable retained the OD/AOD-defining genomic alterations prognosis, however, often undergoes malignant progres- (IDH1R132H and 1p/19q codeletion) and PIK3CAE542K, and sion. We hypothesized that preclinical models of oligo- displayed characteristic sensitivity to alkylating chemothera- dendroglioma could facilitate identification of therapeutic peutic agents. In contrast, a xenograft did not engraft from the targets in progressive oligodendroglioma. We established region that was clinically stable and had wild-type PIK3CA.In multiple oligodendroglioma xenografts to determine our panel of OD/AOD xenografts, the presence of activating if the PI3K/AKT/mTOR signaling pathway drives tumor mutations in the PI3K/AKT/mTOR pathway was consistently progression. associated with xenograft establishment (6/6, 100%). OD/ Experimental Design: Two anatomically distinct tumor AOD that failed to generate xenografts did not have activating samples from a patient who developed progressive anaplastic PI3K/AKT/mTOR alterations (0/9, P < 0.0001). Importantly, oligodendroglioma (AOD) were collected for orthotopic mutant PIK3CA oligodendroglioma xenografts were vulnera- transplantation in mice. We additionally implanted 13 tumors ble to PI3K/AKT/mTOR pathway inhibitors in vitro and to investigate the relationship between PI3K/AKT/mTOR path- in vivo—evidence that mutant PIK3CA is a tumorigenic driver way alterations and oligodendroglioma xenograft formation. in oligodendroglioma. Pharmacologic vulnerabilities were tested in newly developed Conclusions: Activation of the PI3K/AKT/mTOR pathway is AOD models in vitro and in vivo. an oncogenic driver and is associated with xenograft formation Results: A specimen from the tumor site that subsequently in oligodendrogliomas. These findings have implications for manifested rapid clinical progression contained a PIK3CA therapeutic targeting of PI3K/AKT/mTOR pathway activation mutation E542K, and yielded propagating xenografts that in progressive oligodendrogliomas.

1Department of Neurosurgery, Yokohama City University Graduate School of Note: Supplementary data for this article are available at Clinical Cancer Medicine, Yokohama, Japan. 2Division of Brain Tumor Translational Research, Research Online (http://clincancerres.aacrjournals.org/). National Cancer Center Institute, Tokyo, Japan. 3Translational Neuro-Oncology Laboratory, Massachusetts General Hospital, Harvard Medical School, Boston, Current address for A.S. Chi: Neon Therapeutics, Cambridge, Massachusetts. Massachusetts. 4Neurosurgical-Oncology Laboratory, Yokohama City Univer- sity, Yokohama, Japan. 5Department of Neurosurgery, Massachusetts General Corresponding Authors: Daniel P. Cahill, Massachusetts General Hospital/ Hospital, Harvard Medical School, Boston, Massachusetts. 6Department of Harvard Medical School, MGH Brain Tumor Center, 55 Fruit Street, Boston, MA Pathology, Massachusetts General Hospital, Boston, Massachusetts. 7Depart- 02114. Phone: 617-724-0884; Fax: 617-724-0887; E-mail: ment of Microbiology, Yokohama City University Hospital, Yokohama, Japan. [email protected]; Kensuke Tateishi, Graduate School of Medicine, Yoko- 8Stephen E. and Catherine Pappas Center for Neuro-Oncology, Department of hama City University, 3-9 Fukuura, Kanazawa, Yokohama, 2360004, Japan. Neurology, Massachusetts General Hospital Cancer Center, Harvard Medical Phone: 81-45-787-2663; Fax: 81-45-783-6121; E-mail: School, Boston, Massachusetts. 9Department of Radiology, Division of Nuclear [email protected]; and Hiroaki Wakimoto, Massachusetts General Medicine, National Center for Global Health and Medicine, Tokyo, Japan. Hospital/Harvard Medical School, MGH Brain Tumor Research Center, 55 Fruit 10Department of Neurosurgery, Yokohama Rosai Hospital, Yokohama, Japan. Street, Boston, MA 02114. E-mail: [email protected] 11Department of Neurosurgery, Fujisawa Municipal Hospital, Fujisawa, Japan. – 12Department of Neurosurgery, Taguchi Neurosurgery Clinic, Yokohama, Japan. Clin Cancer Res 2019;XX:XX XX 13 Department of Pathology, Yokohama City University Hospital, Yokohama, doi: 10.1158/1078-0432.CCR-18-4144 Japan. 14Laura and Isaac Perlmutter Cancer Center, NYU Langone Medical Center, New York University, New York, New York. 2019 American Association for Cancer Research.

www.aacrjournals.org OF1

Tateishi et al.

tumor progression within WHO2016-defined oligodendroglial Translational Relevance tumors (11, 18, 19). Oligodendroglial tumors comprise a relatively indolent sub- Recent studies have shown that, in OD/AOD, a subpopulation type of adult diffuse gliomas; however, the majority of oligo- of undifferentiated cells was enriched for proliferative potential dendrogliomas eventually develop an outgrowth of a subclone and expressed stem cell signature (20). This finding suggests that that has undergone malignant transformation. Modeling the only a subset of undifferentiated cells plays a pivotal role in the molecular mechanisms of oligodendroglioma tumor progres- maintenance of OD/AOD. Patient-derived glioma xenograft sion is crucial to identify therapeutic targets for malignant (PDX) models are derived from these undifferentiated cells, which disease. Here, we present novel patient-derived anaplastic manifest a cancer stem-like cell phenotype (21–23). Indeed, early- oligodendroglioma (AOD) orthotopic xenograft models. We passage orthotopic xenografts generated with glioma tumor- established 6 oligodendroglioma intracerebral xenograft mod- spheres, in most cases, recapitulate the phenotypic and genotypic els, all of which were found to harbor PI3K/AKT/mTOR path- characteristics of parent glioma, including critical driver gene way gene activating mutations, including PIK3CA at E542K and mutations (24). Our recent efforts generating IDH-mutant xeno- H1047L hotspot mutations. In contrast, OD/AOD tumors that grafts indicated that the establishment and propagation of these did not form xenograft did not have detectable activation of the xenografts was often dependent upon the presence of a "tertiary" PI3K/AKT/mTOR pathway. Importantly, we found progressive mutation (genomic alterations beyond disease-defining altera- tumor cells with mutant PIK3CA were vulnerable to alkylating tions in IDH, ATRX, TP53, and chromosomes 1p/19q), which was agents and PI3K/AKT/mTOR pathway inhibitors in vitro and a hallmark of clinically progressive behavior in the patient's in vivo. These findings suggest a critical role of PI3K/AKT/mTOR tumor (25). Given the indolent course of primary OD/AOD, it pathway activation in driving progression and xenograft for- is therefore not surprising that establishment of tumor xenograft mation in oligodendrogliomas and identify potential thera- models has been difficult, with only a few models reported (5, 26, peutic strategies for progressive tumors. 27). Taken together, our experience suggests that focusing efforts on clinically progressive OD/AOD may simultaneously achieve three important goals: (i) improving the chances of successful xenograft establishment; (ii) providing insights into the genomic evolutionary events that underlie malignant progression in oli- Introduction godendroglial tumors; and (iii) identifying novel therapeutic The World Health Organization (WHO) revised 2016 neuro- targets for progressive tumors. pathologic classification defines oligodendroglioma (grade 2) Herein, we describe a longitudinally monitored AOD and anaplastic oligodendroglioma (AOD, grade 3) as a specific patient,whomanifestedprogressionoftwoanatomicallysep- molecularly defined subtype of adult diffuse gliomas. Mutation of arate tumors that were resected for investigation. From a region the isocitrate dehydrogenase genes (IDH1/2) represents one of the of the tumor that subsequently rapidly progressed, we success- fundamental and earliest molecular events in the genesis of these fully established patient-derived AOD xenografts. This xeno- tumors (1, 2), and has therefore been investigated as a potential graft harbored a hotspot mutation in PIK3CA and activation of therapeutic target (3–5). OD/AOD are also characterized by the PI3K/AKT/mTOR pathway, findings that were also observed whole chromosome arm losses of 1p and 19q (6), and TERT within this region of the patient's tumor. In contrast, cells promoter mutations are almost universal (>95%), while CIC and harvested from a distinct tumor region, which was stable after FUBP1 are also frequently observed, irrespective of tumor surgery and had wild-type PIK3CA, did not induce xenograft grade (1, 6–8). These molecular markers enabled improved formation. Additional attempts at xenografting patient oligo- classification of gliomas for predicting patient outcome (9, 10). dendroglial tumors revealed a tight association between acti- Additional recurring mutations, in the genes PIK3CA, PIK3R1, and vating mutations in the PI3K/AKT/mTOR signaling pathway NOTCH1, have been identified in a subset of oligodendroglial and successful xenograft formation, supporting the pivotal role tumors (11–13). In contrast, TP53 and ATRX mutations, which of the PI3K/AKT/mTOR pathway in driving progression of are commonly seen in astrocytic gliomas with IDH mutation oligodendroglial tumors. (diffuse astrocytoma, anaplastic astrocytoma, and secondary glioblastoma), are exceedingly rare in OD/AOD. In aggregate, IDH-mutant gliomas have better prognosis than Materials and Methods IDH wild-type gliomas, across all WHO grades (12, 14). Within Creation of glioma tumorsphere lines, cell culture, and reagents IDH-mutant gliomas, the prognosis of oligodendroglial (1p/ This study was conducted in accordance with Declaration of 19q codeleted) tumors is more favorable than astrocytic (TP53 Helsinki. All tumor and blood samples were collected with patient and ATRX mutated) tumors. In addition, oligodendroglial consent under protocols approved at Yokohama City University tumors are highly responsive to radiation plus chemotherapy (YCU, Yokohoma, Japan) Hospital and Massachusetts General with procarbazine, lomustine (CCNU), and vincristine (known Hospital (MGH, Boston, MA) Institutional Review Boards. To as PCV), or temozolomide (15–17). No survival difference create glioma tumorsphere lines, fresh tumor specimens were could be distinguished between WHO grade 2 and grade 3 obtained from surgery and enzymatically dissociated with 0.1% oligodendroglial tumors in a large clinical cohort (12). While of Trypsin and DNase. Tumorsphere lines were cultured in serum- oligodendrogliomas can undergo malignant degeneration with free neural stem cell medium, as described previously (3, 25, 28). a fatal outcome (1, 15–17), strict diagnostic reclassification All tumorsphere lines were cryopreserved less than passage 3 to by WHO 2016 raises the possibility that the progression of use for in vitro experiments. Temozolomide (Sigma), Nimustine IDH-mutant, 1p/19q codeleted OD/AOD may be poorly char- (ACNU, Tokyo Chemical Industry), Lomustine (CCNU), Procar- acterized; indeed, few studies have identified predictors of bazine (Sigma), Vincristine (Sigma), FK866 (Sigma), LY294002

OF2 Clin Cancer Res; 2019 Clinical Cancer Research

PI3K Pathway Activation in Progression of Oligodendroglioma

(Sellek), GDC-0068 (Cayman), BYL719 (Cayman), Everolimus mentary Table S1. Methods for cell line fingerprinting, Sanger (Cayman), and AGI-5198 (Sigma) were used. sequencing, pyrosequencing, FISH, MLPA, multiplex PCR technology (SNaPshot), and microsatellite instability analysis are Orthotopic xenograft model described in Supplementary Methods. Cells (1–2 105) were orthotopically implanted into the right striatum of 7- to 9-week-old female SCID Beige mice (Charles Cell viability analysis River Laboratories) within 24 hours after dissociation of tumor To assess cell viability, tumorspheres were dissociated into samples. Mice were monitored daily and sacrificed when neuro- single cells and seeded into 96-well plates at 7,000 to 8,000 cells logic deficits or general conditions reached the criteria for eutha- per well. After 12 to 24 hours, chemical inhibitors were serially nasia. Brains were harvested for pathologic studies and for acutely diluted and added to wells. Cell viability was measured by dissociated tumor cells that were cryopreserved, cultured for CellTiter-Glo (Promega) assay at indicated time points. in vitro experiments, or repeatedly implanted into mice brains. All mouse experiments were approved by the Institutional Animal 11C-Methionine PET imaging Care and Use Committee at MGH and YCU. Methionine (MET)-PET CT was performed at National Center for Global Health and Medicine (Tokyo), as described previous- Histology and IHC ly (29). For semiquantitative interpretations, the SUVmax was Tumor tissue specimens were fixed in 10% neutral-buffered determined by a standard formula. The tumor/normal ratio formalin and embedded in paraffin. Hematoxylin and eosin (T/N ratio) of MET was calculated relative to the uptake in the staining was performed using the standard procedures. For IHC, contralateral frontal cortex. 5-mm thick sections were deparaffinized, treated with 0.5% H2O2 in methanol, rehydrated, and heated for 20 minutes for Statistical analysis antigen retrieval. After blocking with serum, tissue sections Statistical analysis was performed with JMP Pro12 software. For were incubated with primary antibody [1:100 for IDH1R132H parametric analysis, two-tailed t tests were used. Two-tailed Fisher (Dianova), Ki-67 (Dako), MSH6 (Proteintech), LKB1/STK11 exact test was performed for analysis of frequencies of nominal (Novus Biologicals), phospho-AKT (p-AKT, Ser473, Cell Signal- data. Survival analysis was performed using Kaplan–Meier meth- ing Technology), phospho-4EBP1 (p-4EBP1, Cell Signaling od, and log-rank test was used to compare survival differences. Technology), phospho-S6K (p-S6K, Cell Signaling Technology), Data were expressed as mean SEM. P < 0.05 was considered phospho-S6 (p-S6) ribosomal protein (Cell Signaling Technol- statistically significant. ogy), ATRX (Sigma), p53 (Leica), p16INK4a/CDKN2A (R&D Systems)] at 4C overnight. The next day, sections were washed with PBS, incubated with biotinylated secondary anti- Results body for 30 minutes at room temperature, and then incubated Case presentation with ABC solution (PK-6101, PK-6102; Vector Laboratories) for The clinical course and pathologic findings of a representative 30 minutes at room temperature. Finally, sections were incubated patient (YMG6) are shown (Fig. 1A; Supplementary Fig. S1A), and with DAB (Dako) and counterstained with hematoxylin. Only summarized as follows: A previously healthy 33-year-old man strongly stained cells were considered positive. Positive stained presented with headache and MRI of the brain revealed a non- cells were quantitatively analyzed and classified as strong (¼>20%), contrast enhancing cystic lesion located within the left frontal moderate (5%–20%), and weak (25%), respectively. lobe. This primary tumor (designated YMG6I) underwent radio- graphically complete resection and was diagnosed as AOD, with Western blotting 20% of Ki-67 positive cells. The patient then underwent 60 Gy of Cells were lysed in RIPA buffer (Sigma-Aldrich) with a Protease radiotherapy as an adjuvant treatment. Nine years later, MRI Inhibitor Cocktail Tablets (Roche). Fifty micrograms of protein revealed a small noncontrast enhancing lesion at the margin of was separated by 10% SDS-PAGE and transferred to polyvinyli- the original tumor site. 11C-methionine PET imaging, which has dene difluoride membranes (Millipore) by electroblotting. After utility for evaluating recurrent gliomas (30), revealed a relatively blocking with 1 or 5% nonfat dry milk in TBST [25 mmol/L Tris high uptake (Fig. 1A), consistent with tumor recurrence. The (pH, 7.4), 137 mmol/L NaCl, 0.5% Tween20], membranes were recurrent tumor was resected and the pathological diagnosis was incubated at 4C overnight with primary antibodies. After WHO grade 2 glioma (YMG6R1) consisting of reactive astrocytes washing and incubation with horseradish peroxidase–conju- admixed with atypical cells. Temozolomide was administered for gated secondary antibodies (Cell Signaling Technology), blots 24 cycles. Four years later, a contrast-enhanced recurrent lesion were washed, and signals were visualized with chemilumines- emerged, that was again resected (YMG6R2) and the diagnosis cent HRP substrate (Millipore). Primary antibodies used were was AOD (WHO grade 3). Temozolomide treatment resumed for p-AKT, p-mTOR (Cell Signaling Technology), p-4EBP1, p-S6K, additional 20 cycles. However, two distant contrast-enhancing Actin (Cell Signaling Technology), and Vinculin (Cell Signaling tumors with higher uptake of 11C-methionine developed: one at Technology). the left frontal base (YMG6R3F) and a second within the left medial temporal lobe (YMG6R3T), while the initial site of tumor Molecular analysis remained unchanged. Gross total resection was performed for Genomic and bisulfite-modified DNA was extracted using both distant tumors. Both tumors were diagnosed as WHO grade DNeasy Blood & Tissue and EpiTect Bisulfite Kits (Qiagen), per 3 glioma with Ki-67 index approximately 40%. Intriguingly, the manufacturer's protocol. PCR protocol and primer sequencing YMG6R3T (left temporal tumor) subsequently developed a rapid for IDH1 (exon 4), IDH2 (exon 4), TERT promoter, STK11 (exon local recurrence in 4 months, whereas the left frontal tumor 6), and PIK3CA (exons 8, 9, and 20) are described in Supple- (YMG6R3F) has remained stable for more than 24 months.

www.aacrjournals.org Clin Cancer Res; 2019 OF3

Tateishi et al.

Figure 1. Clinical, pathologic, and genomic characteristics of the patient YMG6. A, Clinical course of AOD YMG6 with brain MRI and 11C-methionine PET images. YMG6I 11 (treatment na€ve), YMG6R1 (oligodendroglioma, 1st relapse, C-methionine PET; SUVmax ¼ 2.6, T/N ratio ¼ 1.4), YMG6R2 (AOD, 2nd relapse, SUVmax ¼ 5.6, T/N ratio ¼ 4.8), YMG6R3F (3rd relapse at frontal lobe, SUVmax ¼ 5.3, T/N ratio ¼ 3.1), YMG6R3T (3rd relapse at temporal lobe, SUVmax ¼ 6.6, T/N ratio ¼ 4.3), YMG6R4T (4th relapse, SUVmax ¼ 6.9, T/N ratio ¼ 3.5). GTR, gross total resection; RT, radiotherapy; TMZ, temozolomide. B, Genetic analysis of YMG6R3T. Top, Sanger sequencing for IDH1 (arrow, c.395G>A) and TERT (arrow, c.-124C>T, C228T). Bottom: left, FISH for 1p36 (red) and 1q25 (green); right, FISH for 19q13 (red) and 19p13 (green). C, Multiplex PCR–based MGH SNaPshot assay and Sanger sequencing identifying mutations in blood (YMG6 BD), YMG6R2, YMG6R3F, YMG6R3T, and YMG6R4T DNA. D and E, Frequencies of G:C>A:T transitions (D)andC>T change at CpG or non-CpG sites (E) in YMG6R3F, YMG6R3T, and YMG6R4T. F, Sanger sequencing showing PIK3CA mutation (arrows, E542K) in YMG6R3T and YMG6R4T and not in YMG6R2 and YMG6R3F. G, IHC for p-AKT (Ser473, top), p-4EBP1 (middle), and p-S6K (bottom) in YMG6R2, YMG6R3F, YMG6R3T, and YMG6R4T. Bars, 50 mm.

Finally, the left temporal tumor was again resected (YMG6R4) evolution, we profiled each tumor using an NGS panel that inter- with the histopathology exhibiting similar findings (WHO rogates the exons of 94 cancer-related genes, direct DNA sequenc- grade 3). ing, and IHC analysis (Fig. 1C; Supplementary Fig. S1B; Supple- Genomic analysis identified mutations in IDH1 (R132H) and mentary Table S2). We found sequence variants in MSH6 and the TERT promoter (c.-124C>T, C228T) in all the tumors through- STK11 in normal blood DNA, not previously reported as patho- out the course including the initial tumor (YMG6I), consistent genic mutations, likely indicating germline polymorphisms (Sup- with these mutations being early events in gliomagenesis (Fig. 1B; plementary Table S2; Supplementary Fig. S3A). MSH6 and STK11 Supplementary Fig. S1B; refs. 1, 11, 12). FISH and Multiplex was positively expressed in YMG6I (Supplementary Fig. S3B and Ligation-dependent Probe Amplification (MLPA) assay demon- S3C). Germline mutations in the MSH6 and STK11 are associated strated whole chromosome arm losses of 1p and 19q (Fig. 1B; with Lynch syndrome and Peutz–Jeghers syndrome, respectively. Supplementary Fig. S1B and S1C). Altogether, the WHO 2016 However, upper gastrointestinal endoscopy, endoscopy colonos- integrated classification was AOD, IDH1 mutant and 1p/19q copy, and whole-body FDG-PET imaging did not find any findings codeleted. to suggest these diagnoses were present in this patient (Supple- mentary Fig. S3E–S3G). In addition, MSI was not found in all PIK3CA mutation activating PI3K/AKT/mTOR pathway tumor samples, including non-hypermutant YMG6R2 and hyper- in AOD mutant YMG6R3T (Supplementary Fig. S4; Supplementary Table We confirmed matched DNA identification in YMG6R3F and S3), further evidence against these germline mutations directly YMG6R3T cells (Supplementary Fig. S2). To evaluate genomic promoting AOD growth or treatment resistance in this case.

OF4 Clin Cancer Res; 2019 Clinical Cancer Research

PI3K Pathway Activation in Progression of Oligodendroglioma

Five mutations were detected in YMG6R2 (IDH1, TERT, MSH6, mide-associated DNA hypermutation (Fig. 1D; refs. 1, 31). STK11, and SMARCA4), while tumor specimens from subsequent We then focused on mutations other than the 5 detected in recurrences harbored a much greater number of mutations: the low-mutant load YMG6R2. Considering the strand on YMG6R3F (29 mutations), YMG6R3T (55), and YMG6R4T (27) which C>T change occurred, the majority of C>T changes were cells, respectively (Fig. 1C). Twenty-seven of 29 mutations observed at non-CpG sites, a signature of temozolomide- (93.1%) in YMG6R3F were shared by YMG6R3T, whereas 28 of associated mutation rather than spontaneous deamination, as 55 YMG6R3T mutations (50.9%) were private. YMG6R4T had all mutations at CpG sites were only 16.0%, 13.3%, and 19.0% in the 27 mutations that YMG6R3F and YMG6R3T shared but did YMG6R3F, YMG6R3T, and YMG6R4T, respectively (Fig. 1E). not have the 2 private mutations of YMG6R3F (Fig. 1C; Supple- Somatic mutation of MSH6Gly409Glu was identiﬁed in mentary Table S2). These ﬁndings imply that YMG6R3T and YMG6R4T in which MSH6 expression was negative (Supple- YMG6R4T arose from a subclone of YMG6R3F, and subsequently mentary Fig. S3B and S3D). Thus, the DNA hypermutation acquired additional mutations. phenotype of these later recurrences was consistent with temo- The elevated number of mutations in YMG6R3F, YMG6R3T, zolomide treatment–induced hypermutation, manifest in this and YMG6R4T were suggestive of the emergence of a DNA case after at least 24 cycles of temozolomide treatment. Recur- hypermutation phenotype. G:C>A:T transition mutations were rent glioma with temozolomide-associated hypermutation has overwhelmingly dominant, as observed in 28 of 29 (96.6%) in been shown to have increased MGMT methylation compared YMG6R3F, 51 of 55 (92.7%) in YMG6R3T, and 26 of 27 with initial untreated tumors (32). In this case, both nonhy- (96.3%) in YMG6R4T, a pattern consistent with temozolo- permutated and hypermutated tumors after temozolomide

Figure 2. PIK3CA-mutant patient-derived AOD orthotopic xenograft models recapitulating the phenotypic and genotypic characteristics of the patient. A, Hematoxylin and eosin (H&E) staining of YMG6R3F xenograft (YMG6R3Fsc1; left, overview with the tumor indicated by yellow dotted line; right, microscopic view). B, Overview: H&E (left) and IDH1R132H (right) staining in YMG6R3T xenograft (YMG6R3Tsc1). C, H&E (left) and Ki-67 (right) staining in YMG6R3Tsc1. D, Sanger sequencing for IDH1 (arrow, c.395 G>A, top) and TERT (arrow, c.-124 C>T, C228T, bottom) in YMG6R3Tsc1. E, FISH for 1p36 (red, top), 1q25 (green, top), 19q13 (red, bottom), and 19p13 (green, bottom). F, Overview of H&E-stained YMG6R4T orthotopic xenograft. The tumor is indicated by yellow dotted line. G, H&E (left) and Ki-67 (right) staining in YMG6R4T xenograft (YMG6R4Tsc1). H, Kaplan–Meier curves demonstrating survival differences between YMG6R3F (red line)- YMG6R3T (blue line)- and YMG6R4T (green line)-implanted mice. I, Multiplex PCR–based MGH SNaPshot assay identifying mutations in YMG6R3T xenograft (YMG6R3Tsc1) and YMG6R4T xenograft (YMG6R4Tsc1). Green bars, mutations not identiﬁed in YMG6R3T. Red bars, shared mutations with YMG6R3T. Blue bars, shared mutations with YMG6R4T. J, Sanger sequencing of PIK3CA (arrow, c.1624G>A, E542K) in YMG6R3Tsc1 and YMG6R4Tsc1. K, IHC for p-AKT (top), p-4EBP1 (middle), and p-S6K (bottom) in YMG6R3Fsc1, YMG6R3Tsc1, and YMG6R4Tsc1 xenograft tumors. Bars, 50 mm.

www.aacrjournals.org Clin Cancer Res; 2019 OF5

Tateishi et al.

treatment had high MGMT promoter methylation status (Sup- Histologically, glial tumor cells with atypical nuclei that re- plementary Table S4). sembling oligodendrocytes exhibited high proliferative activity Of note, among the 28 private mutations of YMG6R3T, only (Ki-67 index of 40%), consistent with the diagnosis of AOD PIK3CA (c.1624G>A, Glu542Lys) was retained in YMG6R4T (Fig. 2C). For genomic analysis, we first confirmed matched DNA (Fig. 1C and F; Supplementary Table S2), an observation suggest- identification between YMG6RT3 and the first-generation xeno- ing PIK3CAE542K, and not the other private mutations, may be a graft YMG6R3Tsc1 (Supplementary Fig. S2). YMG6R3Tsc1 har- driver genetic event of YMG6R3T and YMG6R4T. Interpreting the bored mutations in IDH1 (R132H) and TERT promoter functional consequences of individual mutations in cases with (c.-124C>T, C228T), and whole chromosome arm loss of 1p and hypermutation can be difficult, as the vast majority of alterations 19q (Fig. 2D and E; Supplementary Fig. S7A), fulfilling the WHO are nonselected passenger events (1). Nonetheless, PIK3CA muta- 2016 diagnostic criteria of oligodendroglioma. YMG6R4T cells tions at highly recurrent sites (E542K, E545K, and H1047R) have were also tumorigenic (YMG6R4Tsc; Fig. 2F and G), and recapit- been shown to activate oncogenic PIK3/AKT/mTOR pathways in ulated the genotype of AOD (Supplementary Fig. S7B). After serial cancer (33, 34). In contrast, mutations in ATRX and TP53 were transplantation in mouse brains, PDXs generated from YMG6R3T identified in YMG6R3T, not YMG6R4T, and IHC analysis dem- and YMG6R4T cells maintained the phenotypic and genotypic onstrated retained expression of ATRX and low expression of p53 characteristics of AOD (Supplementary Fig. S8A–S8C). YMG6R3T in YMG6R3T (Supplementary Fig. S5), indicating these may be and YMG6R4T xenografts were similarly and consistently lethal nonfunctional passenger mutations. Intriguingly, pyrosequen- within 130 days, whereas YMG6R3F xenografts did not cause cing identified PIK3CAE542K mutation in a small subset of tumor animal death (Fig. 2H). cells in YMG6I, which during subsequent treatment then fell below the detection limit in YMG6R1, YMG6R2, and YMG6R3F PI3K/AKT/mTOR pathway gene mutation is tightly associated (Supplementary Fig. S6A). No other PI3K/AKT/mTOR pathway with AOD orthotopic xenograft formation gene mutation was identified in YMG6R2 and YMG6R3F (Fig. 1C We identified 15 mutations in YMG6R3Tsc1 xenografts (Fig. 2I; and F; Supplementary Table S2). To assess the functional con- Supplementary Table S2). Among them, 8 mutations, including sequences of PIK3CAE542K mutation, we performed IHC for IDH1R132H, TERT promoter, PIK3CAE542K, were retained from the downstream effectors in the PI3K/AKT/mTOR pathway. Results parental primary tumor (YMG6R3T). Interestingly, we identified demonstrated strong expression of p-AKT as well as p-4EBP1 and only 5 mutations in YMG6R4Tsc1, derived from the tumor that p-S6K, in YMG6R3T and YMG6R4T, but not in YMG6R1, progressed after YMG6R3T, including IDH1R132H, TERT promot- YMG6R2, and YMG6R3F (Fig. 1G; Supplementary Fig. S6B). We er, and PIK3CAE542K (Fig. 2I; Supplementary Table S2), whereas did note scattered expression of p-4EBP1 and p-S6K in YMG6I, the other mutations (associated with hypermutation phenotype) consistent with PI3K/AKT/mTOR pathway activation in a subset that were lost were passengers. Importantly, we confirmed the of cells with PIK3CAE542K. maintenance of PIK3CAE542K in serially passaged xenografts by Sanger sequencing (Fig. 2J; Supplementary Fig. S8C). IHC analysis Orthotopic AOD xenograft models recapitulate patient revealed strong expression of p-AKT, p-4EBP1, and p-S6K in phenotypic and genotypic features YMG6R3T and YMG6R4T xenografts, as compared to nonlethal, slow-growing YMG6R3F xenografts (Fig. 2K; Supplementary Fig. Glioma sphere cell lines were established from YMG6R3F and E542K YMG6R3T and implanted into mice brains to assess the potential S8D). We therefore hypothesized that PIK3CA is the driver for engraftment. YMG6R3F implanted mice did not develop mutation that promotes oligodendroglioma progression in the neurologic symptoms during a 250-day follow-up period and patient and in orthotopic xenografts. autopsy revealed only a small tumor (YMG6R3F secondary cells, To test the hypothesis that genetic activation of the PI3K/AKT/ derived from first-generation xenograft, designated YMG6R3Fsc; mTOR pathway promotes xenograft formation in oligodendroglial tumors, we collected 12 additional oligodendroglioma speci- Fig. 2A). In contrast, YMG6R3T implanted mice rapidly developed R132H neurologic deterioration (124 days), and large nodular tumors mens that harbored mutations in IDH1 and the TERT expressing IDH1R132H were found (YMG6R3Tsc; Fig. 2B). promoter, and 1p/19q co-deletion, and performed orthotopic transplantation assays (Table 1). This cohort included 4 AODs with NGS-detected genetic alteration in the PI3K/AKT/mTOR Table 1. Characteristics of oligodendroglial tumors used in xenograft assays pathway (Supplementary Table S5): YMG23 (newly diagnosed Xenograft-induced Mutation in PI3K/Akt/mTOR AOD, PIK3CAHis1047Arg and CDKN2A-CDKN2B homozygous Cell line Histology lethality pathway deletion; Fig. 3A; Supplementary Fig. S9A and S9B); YMG5 YMG5 AOD Yes PIK3R1 I571NfsTer31 (recurrent AOD, PIK3R1Ile571AsnfsTer31; Fig. 3B; Supplementary Fig. YMG6R3T AOD Yes PIK3CA E542K S9C); MGG60 (newly diagnosed AOD, PIK3CAHis1047Leu; ref. 25); YMG6R4T AOD Yes PIK3CA E542K Ser2215Phe YMG23 AOD Yes PIK3CA H1047R MGG137 (newly diagnosed AOD, MTOR ; Fig. 3C). We MGG60 AOD Yes PIK3CA H1047L observed xenograft formation from YMG5, YMG23, MGG60, and MGG137 AOD Yes mTOR S2215F MGG137, which all harbored genetic alteration in the PI3K/AKT/ YMG6R3F AOD No None mTOR pathway (Fig. 3D–F). Mutations of IDH1, TERT, and YMG28 AOD No None PIK3CA as well as CDKN2A-CDKN2B homozygous deletion and YMG46 OD No PIK3CA E469G YMG53 AOD No None p16INK4a/CDKN2A inactivation were retained in YMG23sc1 – MGG78 OD No None xenografts (Supplementary Fig. S9D S9F). In contrast, we did MGG82 AOD No None not observe xenograft formation from YMG46, which harbored MGG109 AOD No None alteration of PIK3CAGlu469Gly, not reported in the COSMIC data- MGG126 AOD No None base (Fig. 3G and H; Supplementary Fig. S10A; Supplementary MGG130 AOD No None Table S5). Also, we did not detect PI3K/AKT/mTOR pathway

OF6 Clin Cancer Res; 2019 Clinical Cancer Research

PI3K Pathway Activation in Progression of Oligodendroglioma

www.aacrjournals.org Clin Cancer Res; 2019 OF7

Tateishi et al.

mutation in seven ODs/AODs: YMG28, YMG53, MGG78, proposal that either inactivated MSH6 or unmethylated MGMT MGG82, MGG109, MGG126, and MGG130 (Table 1; Fig. 3I and promoter can increase temozolomide resistance in glioma. J; Supplementary Fig. S10B and S10C; ref. 25). These ODs/AODs Intriguingly, in patient YMG6, a progressive contrast- without activating PI3K/AKT/mTOR pathway alteration did not enhancing lesion that emerged after removal of YMG6R4T form orthotopic xenografts (Table 1; Fig. 3I and J; Supplementary showed complete response to 4 cycles of procarbazine, nimus- Fig. S10D). Presence of PI3K/AKT/mTOR pathway gene activating tine (ACNU), and vincristine, the regimen called "PAV," which mutation was therefore tightly associated with xenograft lethality is analogous to PCV (Fig. 5B). Consistent with this clinical (6/6 PI3K/AKT/mTOR pathway mutant vs. 0/9 PI3K/AKT/mTOR observation, YMG6R4T tumorspheres also responded to pathway wild type; P < 0.0001; Table 1; Fig. 3K). There was no ACNU, CCNU, procarbazine, and vincristine (monotherapy) significant difference in Ki-67 index between xenograft generating in vitro (Fig. 5C). YMG23 cells also responded to ACNU, and nongenerating patient specimens (P ¼ 0.13; Supplementary CCNU, and procarbazine, but resisted vincristine. In contrast, Table S6). YMG36 (GBM, IDH-wild type) responded rather poorly to IHC of patient tumor specimens demonstrated detectable ACNU, CCNU, and vincristine (Fig. 5C). Clinically, PIK3/ expression of p-AKT, p-4EBP1, and p-S6K (or p-S6) in PI3K/ AKT/mTOR gene mutant tumors, YMG5 and MGG137, AKT/mTOR pathway–mutated YMG23, YMG5, and MGG137 showed durable response to chemotherapeutic agents (PAV (Fig. 4A–C; Supplementary Table S6), as opposed to weak expres- or temozolomide regimen; Supplementary Fig. S12C and sion in PI3K/AKT/mTOR pathway wild-type YMG28, YMG46, S12D). These observations indicate that alkylating agents can and YMG53 (Fig. 4D–F). The xenografts derived from AODs still retain effectiveness in PIK3/AKT/mTOR pathway mutant carrying activating alterations in the PI3K/AKT/mTOR pathway oligodendrogliomas, suggesting instead that alkylator sensi- similarly recapitulated increased expression of p-4EBP1 and tivity is dependent on MMR status or MGMT promoter sta- p-S6K or p-S6 (Fig. 4A–C; Supplementary Table S7). Phosphor- tus(37).Indeed,wedidnotfind significant differences ylation of AKT was also preserved in PIK3CA-mutant xenografts, between the overall survival of patients with PIK3CA/ YMG23sc1 and YMG5sc1. Serially passaged xenografts of YMG23 PIK3R1–mutant oligodendrogliomas as compared with retained mutations of IDH1, TERT, and PIK3CA as well as PI3K/ patients with PIK3CA/PIK3R1 wild-type oligodendrogliomas AKT/mTOR pathway activation found in the patient (Supplemen- (Supplementary Fig. S13; Supplementary Table S8). þ tary Fig. S11). Taken together, our results support a proposed role We recently reported that metabolic depletion of NAD using for activating PIK3/AKT/mTOR pathway alterations in promoting NAMPT inhibitors alone or in combination with temozolomide, þ tumor progression and xenograft formation in oligodendroglial which also lowers intracellular NAD levels, represents a potential tumors. therapeutic strategy for IDH1-mutant gliomas (3, 35, 38). As expected, OD/AOD tumorspheres were sensitive to NAMPT Therapeutic vulnerability of the PI3K pathway activated AOD inhibitor, and combination with temozolomide increased the The YMG6 patient experienced tumor progression on pro- NAMPT inhibitor effects (Fig. 5D; Supplementary Fig. S14A). On longed temozolomide therapy, suggestive of emergence of the other hand, OD/AOD tumorspheres were resistant to direct acquired resistance (Fig. 1A). Our PDX models offered an inhibition of the mutant IDH1 enzyme using AGI-5198 (Fig. 5E; opportunity to test therapeutic vulnerabilities in vitro and Supplementary Fig. S14B), consistent with previous observations in vivo.Wefirst used a cell viability assay to test temozolomide in gliomaspheres (3). response of tumor neurosphere cultures (tumorspheres) We next tested whether inhibition of PI3K/AKT/mTOR sig- derived from the patient tumor in vitro. Despite methylated naling could inhibit xenograft formation and induce selective MGMT promoter, MSH6 inactivated YMG6R4T cells were high- cytotoxicity in PI3K pathway altered AOD. To test whether ly resistant to temozolomide, in line with reported effects of inhibitors of the PI3K pathway could suppress the PI3K/ mismatch repair deficiency on temozolomide response AKT/mTOR pathway, YMG6R4T xenograft cells were treated (Fig. 5A; Supplementary Fig. S3D; Supplementary Table S2; with DMSO, 50 mmol/L of LY294002, or 5 mmol/L of GDC- refs. 35, 36). YMG23 (AOD, MGMT unmethylated) and 0068 for 12 hours. As expected, PI3K inhibitor LY294002 and YMG14 (GBM, IDH wild type, MGMT unmethylated) tumor- AKT inhibitor GDC-0068 potently blocked phosphorylation of spheres were also resistant to temozolomide, while YMG12 AKT, mTOR, 4EBP1, and S6K in YMG6R4T cells (Fig. 5F), (GBM, IDH wild type, MGMT methylated, MMR intact) and indicating on-target effects. LY294002 and GDC-0068 induced MGG152 (GBM, IDH1 mutant, MGMT methylated, MMR cytotoxic effects in PIK3CA-mutant YMG6R4T and YMG23 intact) tumorspheres were responsive. Indeed, YMG23 and cells, while they did not in PIK3CA wild-type YMG28 tumor- YMG14 patient tumors progressed on temozolomide (Supple- sphere cells (Fig. 5G). A similar profile was observed when mentary Fig. S12A and S12B). These results support the YMG6R4T and YMG23 (PIK3/AKT/mTOR pathway mutant;

Figure 3. PI3K/AKT/mTOR pathway gene alteration is associated with oligodendroglioma xenograft generation. A–C, Contrast-enhancing T1-weighted MR images, hematoxylin and eosin (H&E) staining, and IDH1R132H IHC of patient tumors. A, YMG23 carrying IDH1Arg132His, TERT (C228T), DDX3XArg503Gly, TP53Glu11Gln, and PIK3CAHis1047Arg. B, YMG5 carrying IDH1Arg132His, TERT (C228T), DDX3XArg351Trp, HNF1Ac.711T>C,andPIK3R1Ile57AsnfsTer31. C, MGG137 harboring IDH1Arg132His, TERT (C228T),andmTORSer2215Phe. D–F, H&E staining and IDH1R132H IHC of xenografts. D, YMG23 xenograft model (sc1). E, YMG5 xenograft model (sc1). F, MGG137 xenograft model (sc1). Left, H&E staining (overview). Right, H&E (top) and IDH1R132H (bottom). Tumor is indicated by yellow dotted line. G, YMG46 carrying IDH1Arg132His, TERT (C228T),andPIK3CAGlu469Gly. H, H&E-stained brain section derived from YMG46-implanted mouse. I, YMG28 harboring IDH1Arg132His, TERT (C228T),andCIC splice acceptor variant. J, H&E-stained brain section derived from YMG28-implanted mouse. Bars, 50 mm. K, Number of oligodendroglioma/ AOD cases stratiﬁed by PI3K/AKT/mTOR pathway gene alteration and xenograft lethality. A, B, and I, FISH for 1p36 (red, right, top), 1q25 (green, right, top), 19q13 (red, right, bottom), and 19p13 (green, right, bottom). G and I, T2-weighted MRI due to lack of enhancement.

OF8 Clin Cancer Res; 2019 Clinical Cancer Research

PI3K Pathway Activation in Progression of Oligodendroglioma

Figure 4. Activation of PI3K/Akt/mTOR signaling in oligodendroglial tumors and xenografts that harbor alteration in PI3K/Akt/mTOR pathway genes. IHC analysis demonstrating Ki-67, p-AKT (Ser473), p-4EBP1, and pS6K expression in xenograft generating oligodendroglial tumors and corresponding xenografts [YMG23 and YMG23sc1 (A), YMG5 and YMG5sc1 (B), and MGG137 and MGG137sc1 (C)], and oligodendroglial tumors that did not generate xenografts [YMG28 (D), YMG46 (E), and YMG53 (F)]. Bars, 50 mm.

sensitive) and YMG28 (resistant) were tested with another PI3K Discussion inhibitor (BYL719) and an mTOR inhibitor (everolimus; PI3K pathway genes are frequently mutated in oligodendroglial Fig. 5H). In contrast, YMG46 (nonactivating mutation of tumors, including PIK3CA (20%) and PIK3R1 (9%; ref. 13). These PIK3CA (E469K) was not sensitive to these compounds mutations are predominantly found in recurrent tumors when (Fig. 5I). Finally, we treated YMG6R4T cells with DMSO or paired primary and progressive tumors are compared (11), imply- LY294002 for 24 hours in vitro and implanted them orthoto- ing acquisition of PI3K pathway mutation is associated with pically (Fig. 5J). Together, these results suggest that activating progression, and therefore may be a driver for malignant trans- mutations in the PIK3/AKT/mTOR pathway may predict formation. However, beyond these temporal associations, no response to targeted inhibition of this signaling pathway. evidence exists supporting these mutations as promoters of tumor

www.aacrjournals.org Clin Cancer Res; 2019 OF9

Tateishi et al.

OF10 Clin Cancer Res; 2019 Clinical Cancer Research

PI3K Pathway Activation in Progression of Oligodendroglioma

progression in oligodendroglioma. We here provide functional whereas serially retained mutations, such as PIK3CAE542K,rep- evidence that PIK3/AKT/mTOR pathway activating mutation resent the functional drivers. promotes malignant progression using a series of patient- As few oligodendroglioma xenograft models have been estab- intracranial oligodendroglial xenografts and a patient whose lished to date, complete knowledge of the molecular drivers of two evolutionarily divergent and genetically distinct tumors oligodendroglioma progression remains unclear (5, 26, 27). A followed contrasting clinical and laboratory courses. In the previous study demonstrated that ectopic expression of patient case, the PIK3CAE542K mutant tumor rapidly recurred IDH1R132H in the subventricular zone of the mouse brain induced andformedlethalxenografts,whilethePIK3CA wild-type early IDH-mutant gliomagenesis, including 2-hydroxyglutarate tumor did not progress nor form progressive xenografts. production and a global DNA methylation phenotype, however, In our oligodendroglioma series, we found that hotspot muta- was unable to sustain tumor growth (40). We also reported that tion and activation of the PI3K/AKT/mTOR pathway was secondary mutations, including TP53 and ATRX in astrocytic tightly associated with progressive AOD xenografts, and found tumors or TERT promoter mutation and 1p/19q codeletion in that PI3K/AKT/mTOR pathway activated mutant tumor models oligodendroglioma, were not sufficient to form IDH1-mutant were selectively sensitive to PI3K/AKT/mTOR pathway xenografts in immunocompromised mice (25). Recently, Philip inhibitors. and colleagues demonstrated mutant IDH1 cooperated with PI3K/AKT/mTOR pathway is a canonical growth regulatory PDGFRA and loss of CDKN2A, ATRX, and PTEN to promote pathway in cancer. A pan-cancer proteo-genomic atlas demon- tumor development in vivo (41). These findings suggest that a strated that PI3K/AKT/mTOR pathway activation was correlated tertiary genomic event may be required for xenograft formation in with distribution of mutations in key PI3K/AKT/mTOR pathway oligodendrogliomas as well. genes (33). Highly recurrent mutations at E542, E545, and H1047 All of our progressive oligodendroglioma xenograft models in PIK3CA, nonsense/frameshift/indel mutations involving that successfully generated orthotopic xenograft carried function- PIK3R1, and hotspot mutation (e.g., S2215F) in MTOR all potent- al genetic alterations in PIK3CA, PIK3R1,orMTOR. In contrast, ly induced expression of p-AKT (Ser473). In contrast, tumors with none of PI3K pathway gene wild-type tumors or tumor with mutations not predicted to be functional demonstrated weak or PIK3CA nonfunctional mutation induced xenograft formation. no significant effect on p-AKT (Ser473; ref. 33). These prior Clinically, we observed rapid tumor progression after resection of reports, alongside our current work, collectively support the YMG6R3T and YMG6R4T (PIK3CA mutant), whereas YMG6R3F model that alterations in PI3K/AKT/mTOR pathway genes acti- (PIK3CA wild-type) remained stable postoperatively, mirroring vate the PI3K pathway and promote malignant progression in the inability to generate xenograft ex vivo. These findings are oligodendroglial tumors. consistent with previous studies that associated genomic altera- Previous reports indicate that recurrence of oligodendro- tions in the PI3K pathway, such as PIK3CA and PIK3R1 mutations, glioma is not accompanied by dramatic changes in the number with progression in IDH1-mutant gliomas (11, 25). A recent study of mutation and genome-wide methylation status (39). Inter- also demonstrated that hotspot mutations in PIK3CA potentiated estingly, our case study displayed DNA hypermutation after PI3K signaling and promoted tumorigenesis of immortalized tumor progression on temozolomide, and PIK3CA mutation normal human astrocytes in vivo (42). In support of these findings, emerged in one of the two recurrent lesions. Notably however, our study also demonstrated delayed xenograft growth by PI3K both PIK3CA-mutant (YMG6R3T) and PIK3CA wild-type pathway inhibition. These results support the crucial role of PI3K/ (YMG6R3F) tumors harbored a DNA hypermutation pheno- AKT/mTOR pathway gene alteration in the formation and pro- type, yet the former subsequently progressed and the latter did gression of oligodendroglioma xenografts. not, indicating the DNA hypermutation phenotype itself was Our oligodendroglioma xenograft models should facilitate not a direct promoter of malignant progression. In addition, research exploring therapeutic targets in progressive oligoden- the fact that the DNA hypermutation phenotype was lost in the droglial tumors. Using these models, we provide experimental subsequent progressive tumor (YMG6R4T) suggests the major- evidence that oligodendroglial tumors with PI3K/AKT/ ity of mutations found in YMG6R3T might be passengers, mTOR mutation are susceptible to selective inhibition of the

Figure 5. Alkylating chemotherapy and molecular targeted agents target PI3K pathway–mutant oligodendroglial tumors. A, Cell Titer-Glo cell viability assay after 6-day temozolomide (TMZ) treatment for glioma tumorsphere lines. , P < 0.05 for the difference between DMSO and TMZ at indicated concentration in MGG152 () and YMG12 (). B, Complete remission of recurrent AOD after 4 cycles (8 months) of a chemotherapeutic regimen consisting of procarbazine, nimustine (ACNU), and vincristine. Contrast-enhanced MRI showing rapid recurrence (yellow circles) 2 months after gross total resection (GTR) of YMG6R4T (left) and following chemotherapy (right). C, Cell Titer-Glo cell viability assay after 3-day treatment of AOD lines (YMG6R4T and YMG23, both PIK3CA mutant) and a glioblastoma line (YMG36) with ACNU, CCNU, procarbazine, or vincristine. , , P < 0.05 for the difference between DMSO and indicated chemotherapeutic agent in YMG6R4T (), YMG23 (), and YMG36 (). D, Relative cell viability of YMG6R4T (left) and YMG23 (right) cells after 3-day treatment with FK866 combined with DMSO control (blue bars) or TMZ (200 mmol/L, purple bars). , P < 0.05 for the difference between DMSO and FK866. , P < 0.05 for the difference between DMSO and FK866 plus TMZ. E, Cell viability of YMG6R3T, YMG6R4T, and YMG23 cells (all PIK3CA mutant) after 9-day exposure with AGI-5198 (IDH1R132H- speciﬁc inhibitor), relative to DMSO control. F, Western blot analysis of p-AKT, p-mTOR, p-4EBP1, and p-S6K expression in YMG6R4T cells after 12-hour treatment with DMSO control, LY294002 (PI3K inhibitor, 50 mmol/L, left), and GDC-0068 (AKT inhibitor, 5 mmol/L, right). G, Cell Titer-Glo cell viability assay after 3-day treatment of YMG6R4T, YMG23, and YMG28 (PI3K pathway gene wild type) AOD cells with LY294002 or GDC-0068. , P < 0.05 for the difference between DMSO and treatment at indicated concentration in YMG6R4T () and YMG23 (). H, Cell Titer-Glo cell viability assay after 3-day treatment of YMG6R4T, YMG23, and YMG28 (PI3K pathway gene wild type) AOD cells with BYL719 (PI3K inhibitor) or everolimus (mTOR inhibitor). , P < 0.05 for the difference between DMSO and treatment at indicated concentration in YMG6R4T () and YMG23 (). I, Cell Titer-Glo cell viability assay after 3-day treatment of YMG46 (PIK3CAGlu469Gly, not reported in COSMIC database) AOD cells with LY294002, GDC-0068, BYL719, or everolimus. P < 0.05 for the difference between DMSO and treatment at indicated concentration. J, Kaplan–Meier curves indicating survival difference between mice implanted with DMSO (24 hours, n ¼ 3) or LY294002 (50 mmol/L, 24 hours, n ¼ 3) pretreated YMG6R4Tsc2 cells (2 105 cells/mouse). Bars, SEM.

www.aacrjournals.org Clin Cancer Res; 2019 OF11

Tateishi et al.

PI3K/AKT/mTOR pathway. Interestingly, alkylating chemother- Authors' Contributions apeutics such as ACNU and procarbazine remained effective in the Conception and design: K. Tateishi, T. Nakamura, J.J. Miller, J. Sasame, patients described and our oligodendroglioma models after T.T. Batchelor, H. Wakimoto, D.P. Cahill developing resistance to temozolomide and developing temozo- Development of methodology: K. Tateishi, A.J. Iafrate, H. Wakimoto, D.P. Cahill lomide-induced hypermutation phenotype. Indeed, we did not fi – Acquisition of data (provided animals, acquired and managed patients, nd a survival difference of PIK3CA/PIK3R1 mutant oligoden- provided facilities, etc.): K. Tateishi, T. Nakamura, T.A. Juratli, drogliomas, when compared with wild-type oligodendrogliomas. E.A. Williams, S. Miyake, A.L. Fink, N. Lelic, M.V.A. Koerner, Y. Miyake, Also, we found in our index case, that although the primary T. Tanaka, R. Minamimoto, S. Matsunaga, S. Mukaihara, T. Shuto, tumor (YMG6I) harbored PIK3CAE542K, this subpopulation of H. Taguchi, N. Udaka, W.T. Curry, D. Dias-Santagata, T. Yamamoto, cells fell below the limit of detection during radiation and K. Ichimura, T.T. Batchelor, A.J. Iafrate, D.P. Cahill Analysis and interpretation of data (e.g., statistical analysis, biostatistics, temozolomide treatment. These findings imply that PI3K path- computational analysis): K. Tateishi, T.A. Juratli, Y. Matsushita, J.J. Miller, way gene alteration in oligodendrogliomas may promote K. Fujimoto, T. Tanaka, S. Yamanaka, D. Dias-Santagata, K. Ichimura, A.S. Chi, malignant progression, but not necessarily resistance to effective A.J. Iafrate, H. Wakimoto, D.P. Cahill cytotoxic chemotherapies. Early detection of subclonal PI3K path- Writing, review, and/or revision of the manuscript: K. Tateishi, J.J. Miller, way activating mutations could therefore potentially inform H. Murata, W.T. Curry, T. Yamamoto, K. Ichimura, T.T. Batchelor, A.S. Chi, the timing of initiating such treatments, although further trans- A.J. Iafrate, H. Wakimoto, D.P. Cahill Administrative, technical, or material support (i.e., reporting or organizing lational clinical studies will be needed to guide treatment deci- data, constructing databases): M. Nishi, S.S. Tummala, N. Lelic, A. Ryo, sions for patients with oligodendroglioma. S. Yamanaka In conclusion, we identified the critical role of PI3K/AKT/ Study supervision: K. Tateishi, H. Murata, H. Wakimoto, D.P. Cahill mTOR gene activating mutation in promoting malignant transformation and generating xenografts in oligodendroglial tumors. Acknowledgments These xenograft models provide insights into potential therapeu- The authors thank Mrs. Emi Hirata and Yasuko Tanaka for technical tic strategies, including PI3K/AKT/mTOR pathway inhibitors, in a assistance. fi subset of oligodendroglial tumors. This work was supported by Grant-Aid for Scienti c Research C (16K10765 to K. Tateishi), Princess Takamatsu Cancer Research Fund (to K. Tateishi), Takeda Science Foundation (to K. Tateishi), The Yasuda Medical Foundation Disclosure of Potential Conflicts of Interest (to K. Tateishi), Japanese Foundation for Multidisciplinary Treatment of Cancer (to K. Tateishi), Yokohama Foundation for Advancement of Medical Science T.T. Batchelor reports receiving commercial research grants from Pfizer and (to K. Tateishi), SGH Foundation for Cancer Research (to K. Tateishi), a Bristol- Oncoceutics, is a consultant/advisory board member for Genomicare, Cham- Myers Squibb research grant (to K. Tateishi), OligoNation (to D.P. Cahill), and pions Biotechnology, NXDC, Merck, Amgen, and Proximagen, and reports NIH R01CA227821 (to D.P. Cahill and H. Wakimoto). receiving other remuneration from CRICO, UpToDate, Oakstone, and Jiahui Health. A.S. Chi is an employee of Neon Therapeutics, and is a consultant/ advisory board member for Cota Healthcare. A.J. Iafrate reports receiving The costs of publication of this article were defrayed in part by the payment of commercial research grants from Sanofi, holds ownership interest (including page charges. This article must therefore be hereby marked advertisement in patents) in ArcherDx, and is a consultant/advisory board member for Chugai, accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Debiopharm, and Roche. D.P. Cahill is a consultant/advisory board member for Lilly and Merck. No potential conflicts of interest were disclosed by the other Received December 21, 2018; revised February 14, 2019; accepted April 8, authors. 2019; published first April 11, 2019.

References 1. Johnson BE, Mazor T, Hong C, Barnes M, Aihara K, McLean CY, et al. 9. Arita H, Yamasaki K, Matsushita Y, Nakamura T, Shimokawa A, Takami H, Mutational analysis reveals the origin and therapy-driven evolution of et al. A combination of TERT promoter mutation and MGMT methylation recurrent glioma. Science 2014;343:189–93. status predicts clinically relevant subgroups of newly diagnosed glioblas- 2. Yan H, Parsons DW, Jin G, McLendon R, Rasheed BA, Yuan W, et al. IDH1 tomas. Acta Neuropathol Commun 2016;4:79. and IDH2 mutations in gliomas. N Engl J Med 2009;360:765–73. 10. Eckel-Passow JE, Lachance DH, Molinaro AM, Walsh KM, Decker PA, 3. Tateishi K, Wakimoto H, Iafrate AJ, Tanaka S, Loebel F, Lelic N, et al. Sicotte H, et al. Glioma groups based on 1p/19q, IDH, and TERT promoter Extreme vulnerability of IDH1 mutant cancers to NADþ depletion. mutations in tumors. N Engl J Med 2015;372:2499–508. Cancer Cell 2015;28:773–84. 11. Bai H, Harmanci AS, Erson-Omay EZ, Li J, Coskun S, Simon M, et al. 4. Schumacher T, Bunse L, Pusch S, Sahm F, Wiestler B, Quandt J, et al. A Integrated genomic characterization of IDH1-mutant glioma malignant vaccine targeting mutant IDH1 induces antitumour immunity. Nature progression. Nat Genet 2016;48:59–66. 2014;512:324–7. 12. Suzuki H, Aoki K, Chiba K, Sato Y, Shiozawa Y, Shiraishi Y, et al. Mutational 5. Rohle D, Popovici-Muller J, Palaskas N, Turcan S, Grommes C, Campos C, landscape and clonal architecture in grade II and III gliomas. Nat Genet et al. An inhibitor of mutant IDH1 delays growth and promotes differen- 2015;47:458–68. tiation of glioma cells. Science 2013;340:626–30. 13. Cancer Genome Atlas Research N, Brat DJ, Verhaak RG, Aldape KD, Yung 6. Killela PJ, Reitman ZJ, Jiao Y, Bettegowda C, Agrawal N, Diaz LA Jr., et al. WK, Salama SR, et al. Comprehensive, integrative genomic analysis of TERT promoter mutations occur frequently in gliomas and a subset of diffuse lower-grade gliomas. N Engl J Med 2015;372:2481–98. tumors derived from cells with low rates of self-renewal. Proc Natl Acad Sci 14. van den Bent MJ, Smits M, Kros JM, Chang SM. Diffuse inﬁltrating U S A 2013;110:6021–6. oligodendroglioma and astrocytoma. J Clin Oncol 2017;35: 7. Arita H, Narita Y, Fukushima S, Tateishi K, Matsushita Y, Yoshida A, et al. 2394–401. Upregulating mutations in the TERT promoter commonly occur in adult 15. Buckner JC, Shaw EG, Pugh SL, Chakravarti A, Gilbert MR, Barger GR, et al. malignant gliomas and are strongly associated with total 1p19q loss. Radiation plus procarbazine, CCNU, and vincristine in low-grade glioma. Acta Neuropathol 2013;126:267–76. N Engl J Med 2016;374:1344–55. 8. Bettegowda C, Agrawal N, Jiao Y, Sausen M, Wood LD, Hruban RH, et al. 16. van den Bent MJ, Taphoorn MJ, Brandes AA, Menten J, Stupp R, Frenay M, Mutations in CIC and FUBP1 contribute to human oligodendroglioma. et al. Phase II study of ﬁrst-line chemotherapy with temozolomide in Science 2011;333:1453–5. recurrent oligodendroglial tumors: the European Organization for

OF12 Clin Cancer Res; 2019 Clinical Cancer Research

PI3K Pathway Activation in Progression of Oligodendroglioma

Research and Treatment of Cancer Brain Tumor Group Study 26971. J Clin L-methyl-[(11)C]methionine PET. AJNR Am J Neuroradiol 2014;35: Oncol 2003;21:2525–8. 278–84. 17. Cairncross JG, Ueki K, Zlatescu MC, Lisle DK, Finkelstein DM, Hammond 30. Tripathi M, Sharma R, Varshney R, Jaimini A, Jain J, Souza MM, et al. RR, et al. Specific genetic predictors of chemotherapeutic response and Comparison of F-18 FDG and C-11 methionine PET/CT for the evaluation survival in patients with anaplastic oligodendrogliomas. J Natl Cancer Inst of recurrent primary brain tumors. Clin Nucl Med 2012;37:158–63. 1998;90:1473–9. 31. Hunter C, Smith R, Cahill DP, Stephens P, Stevens C, Teague J, et al. A 18. Halani SH, Yousefi S, Vega JV, Rossi MR, Zhao Z, Amrollahi F, et al. Multi- hypermutation phenotype and somatic MSH6 mutations in recurrent faceted computational assessment of risk and progression in oligoden- human malignant gliomas after alkylator chemotherapy. Cancer Res droglioma implicates NOTCH and PI3K pathways. NPJ Precis Oncol 2018; 2006;66:3987–91. 2:24. 32. van Thuijl HF, Mazor T, Johnson BE, Fouse SD, Aihara K, Hong C, et al. 19. Aoki K, Nakamura H, Suzuki H, Matsuo K, Kataoka K, Shimamura T, et al. Evolution of DNA repair defects during malignant progression of low- Prognostic relevance of genetic alterations in diffuse lower-grade gliomas. grade gliomas after temozolomide treatment. Acta Neuropathol 2015;129: Neuro Oncol 2018;20:66–77. 597–607. 20. Tirosh I, Venteicher AS, Hebert C, Escalante LE, Patel AP, Yizhak K, et al. 33. Zhang Y, Kwok-Shing Ng P, Kucherlapati M, Chen F, Liu Y, Tsang YH, et al. Single-cell RNA-seq supports a developmental hierarchy in human oligo- A pan-cancer proteogenomic atlas of PI3K/AKT/mTOR pathway altera- dendroglioma. Nature 2016;539:309–13. tions. Cancer Cell 2017;31:820–32e3. 21. Luchman HA, Stechishin OD, Dang NH, Blough MD, Chesnelong C, Kelly 34. Kang S, Bader AG, Vogt PK. Phosphatidylinositol 3-kinase mutations JJ, et al. An in vivo patient-derived model of endogenous IDH1-mutant identified in human cancer are oncogenic. Proc Natl Acad Sci U S A glioma. Neuro Oncol 2012;14:184–91. 2005;102:802–7. 22. Wakimoto H, Kesari S, Farrell CJ, Curry WT Jr, Zaupa C, Aghi M, et al. 35. Tateishi K, Higuchi F, Miller JJ, Koerner MVA, Lelic N, Shankar GM, et al. Human glioblastoma-derived cancer stem cells: establishment of invasive The alkylating chemotherapeutic temozolomide induces metabolic stress glioma models and treatment with oncolytic herpes simplex virus vectors. in IDH1-mutant cancers and potentiates NADþ depletion-mediated cyto- Cancer Res 2009;69:3472–81. toxicity. Cancer Res 2017;77:4102–15. 23. Singh SK, Hawkins C, Clarke ID, Squire JA, Bayani J, Hide T, et al. 36. Yip S, Miao J, Cahill DP, Iafrate AJ, Aldape K, Nutt CL, et al. MSH6 Identification of human brain tumour initiating cells. Nature 2004;432: mutations arise in glioblastomas during temozolomide therapy and medi- 396–401. ate temozolomide resistance. Clin Cancer Res 2009;15:4622–9. 24. Wakimoto H, Mohapatra G, Kanai R, Curry WT Jr, Yip S, Nitta M, et al. 37. Choi S, Yu Y, Grimmer MR, Wahl M, Chang SM, Costello JF. Maintenance of primary tumor phenotype and genotype in glioblastoma Temozolomide-associated hypermutation in gliomas. Neuro Oncol stem cells. Neuro Oncol 2012;14:132–44. 2018;20:1300–9. 25. Wakimoto H, Tanaka S, Curry WT, Loebel F, Zhao D, Tateishi K, et al. 38. Shankar GM, Kirtane AR, Miller JJ, Mazdiyasni H, Rogner J, Tai T, et al. Targetable signaling pathway mutations are associated with malignant Genotype-targeted local therapy of glioma. Proc Natl Acad Sci U S A 2018; phenotype in IDH-mutant gliomas. Clin Cancer Res 2014;20:2898–909. 115:E8388–E94. 26. Navis AC, Niclou SP, Fack F, Stieber D, van Lith S, Verrijp K, et al. Increased 39. Aihara K, Mukasa A, Nagae G, Nomura M, Yamamoto S, Ueda H, et al. mitochondrial activity in a novel IDH1-R132H mutant human oligoden- Genetic and epigenetic stability of oligodendrogliomas at recurrence. droglioma xenograft model: in situ detection of 2-HG and alpha-KG. Acta Neuropathol Commun 2017;5:18. Acta Neuropathol Commun 2013;1:18. 40. Bardella C, Al-Dalahmah O, Krell D, Brazauskas P, Al-Qahtani K, Tomkova 27. Klink B, Miletic H, Stieber D, Huszthy PC, Campos Valenzuela JA, Balss J, M, et al. Expression of Idh1(R132H) in the murine subventricular zone et al. A novel, diffusely infiltrative xenograft model of human anaplastic stem cell niche recapitulates features of early gliomagenesis. Cancer Cell oligodendroglioma with mutations in FUBP1, CIC, and IDH1. PLoS One 2016;30:578–94. 2013;8:e59773. 41. Philip B, Yu DX, Silvis MR, Shin CH, Robinson JP, Robinson GL, et al. 28. Tateishi K, Iafrate AJ, Ho Q, Curry WT, Batchelor TT, Flaherty KT, et al. Myc- Mutant IDH1 promotes glioma formation in vivo. Cell Rep 2018;23: driven glycolysis is a therapeutic target in glioblastoma. Clin Cancer Res 1553–64. 2016;22:4452–65. 42. McNeill RS, Stroobant EE, Smithberger E, Canoutas DA, Butler MK, Shelton 29. Tateishi K, Tateishi U, Nakanowatari S, Ohtake M, Minamimoto R, Suenaga AK, et al. PIK3CA missense mutations promote glioblastoma pathogenesis, J, et al. (62)Cu-diacetyl-bis (N(4)-methylthiosemicarbazone) PET in but do not enhance targeted PI3K inhibition. PLoS One 2018;13: human gliomas: comparative study with [(18)F]fluorodeoxyglucose and e0200014.

www.aacrjournals.org Clin Cancer Res; 2019 OF13

PI3K/AKT/mTOR Pathway Alterations Promote Malignant Progression and Xenograft Formation in Oligodendroglial Tumors

Kensuke Tateishi, Taishi Nakamura, Tareq A. Juratli, et al.

Clin Cancer Res Published OnlineFirst April 11, 2019.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-18-4144

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2019/04/11/1078-0432.CCR-18-4144.DC1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/early/2019/05/21/1078-0432.CCR-18-4144. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.