PDK1 Mediates NOTCH1-Mutated Head and Neck Squamous … · Translational Cancer Mechanisms and Therapy PDK1 Mediates NOTCH1-Mutated Head and Neck Squamous Carcinoma Vulnerability

Translational Cancer Mechanisms and Therapy

PDK1 Mediates NOTCH1-Mutated Head and NeckSquamous Carcinoma Vulnerability toTherapeutic PI3K/mTOR InhibitionVaishnavi Sambandam1, Mitchell J. Frederick2, Li Shen3, Pan Tong3, Xiayu Rao3,Shaohua Peng1, Ratnakar Singh1,4, Tuhina Mazumdar1, Chenfei Huang2, Qiuli Li5,Curtis R. Pickering6,7, Jeffery N. Myers6,7, Jing Wang3,7, and Faye M. Johnson1,7

Abstract

Purpose: Head and neck squamous cell carcinoma(HNSCC) is driven largely by the loss of tumor suppressorgenes, including NOTCH1, but lacks a biomarker-driventargeted therapy. Although the PI3K/mTOR pathway isfrequently altered in HNSCC, the disease has modest clin-ical response rates to PI3K/mTOR inhibitors and lacksvalidated biomarkers of response. We tested the hypothesisthat an unbiased pharmacogenomics approach to PI3K/mTOR pathway inhibitors would identify novel, clinicallyrelevant molecular vulnerabilities in HNSCC with loss oftumor suppressor function.

Experimental Design: We assessed the degree to whichresponses to PI3K/mTOR inhibitors are associated with genemutations in 59HNSCC cell lines. Apoptosis in drug-sensitivecell lines was confirmed in vitro and in vivo. NOTCH1 pathwaycomponents and PDK1 were manipulated with drugs, geneediting, knockdown, and overexpression.

Results: PI3K/mTOR inhibition caused apoptosis anddecreased colony numbers in HNSCC cell lines harboringNOTCH1 loss-of-function mutations (NOTCH1MUT) andreduced tumor size in subcutaneous and orthotopic xenograftmodels. In all cell lines, NOTCH1MUT was strongly associatedwith sensitivity to six PI3K/mTOR inhibitors. NOTCH1 inhi-bition or knockout increasedNOTCH1WTHNSCC sensitivity toPI3K/mTOR inhibition. PDK1 levels dropped following PI3K/mTORinhibition inNOTCH1MUTbutnotNOTCH1WTHNSCC,and PDK1 overexpression rescued apoptosis in NOTCH1MUT

cells. PDK1 and AKT inhibitors together caused apoptosis inNOTCH1WT HNSCC but had little effect as single agents.

Conclusions: Our findings suggest that NOTCH1MUT pre-dicts response to PI3K/mTOR inhibitors, which may lead tothe first biomarker-driven targeted therapy for HNSCC, andthat targeting PDK1 sensitizes NOTCH1WT HNSCC to PI3K/mTOR pathway inhibitors.

IntroductionAlthough head and neck squamous cell carcinoma (HNSCC)

has a high mutation rate, genomic sequencing has not detecteddruggable driver mutations in the disease. Rather, HNSCC isdominated bymutations in nontargetable tumor suppressor genessuch as TP53 (72%),NOTCH1 (18%), KMT2D (16%), andAJUBA(6%; refs. 1, 2). The most frequently altered mitogenic signaling

pathway in HNSCC is the PI3K/mTOR pathway, with 54% ofpatients having mutations or copy number alterations in PIK3CA(35%), PTEN (6%),RICTOR (7%),AKT1 (3%), PIK3R1 (2%), andMTOR (3%; ref. 3). PIK3CA, the third most frequently mutatedgene in HNSCC (18%), has frequent hotspot mutations in thehelical (E542K or E545K) and kinase (H1047R) domains (3, 4).

PI3K signaling and its effectors protein kinase B (AKT), mTOR,and phosphoinositide-dependent kinase 1 (PDK1), which playcritical roles in cell proliferation and survival, are validated ther-apeutic targets in other cancers (5). With the exception of PI3Kinhibitors that are approved for the treatment of hematologicmalignancies and target of rapamycin complex 1 (TORC1) inhi-bitors that are approved for renal cell carcinoma, PI3K/AKT/mTORinhibitors have elicited only modest response rates in solidtumors (5). Strategies guiding the selection of patients for theseagents also remain elusive.Whether PIK3CAmutant (PIK3CAMUT)tumors have increased sensitivity to PI3K/AKT/mTOR inhibitors isunclear (4, 6, 7). PI3K/AKT/mTOR inhibitors have a low clinicalresponse rate in PIK3CAMUT tumors (�30%; ref. 6) comparedwithother targeted therapies such as the epidermal growth factorreceptor inhibitor osimertinib, which has a response rate of about80% in EGFRMUT lung cancers (8). Consistent with these clinicalfindings, PIK3CAMUT HNSCC cell lines and patient-derived xeno-grafts (PDX) aremore sensitive to PI3K/mTORpathway inhibitorsthanPIK3CAwild-type (PIK3CAWT)HNSCCcells are; however, thedrugs cause only cell-cycle arrest with no apoptosis in the mutantcell lines (4, 9–12).

1Department of Thoracic/Head & Neck Medical Oncology, The University ofTexas MD Anderson Cancer Center, Houston, Texas. 2Department of Otolaryn-gology, Baylor College of Medicine, Houston, Texas. 3Department of Bioinfor-matics andComputational Biology, TheUniversity of TexasMDAndersonCancerCenter, Houston, Texas. 4Department of Head and Neck Surgery, The Universityof TexasMDAnderson Cancer Center, Houston, Texas. 5Department of Head andNeck Surgery, Sun Yat-sen University Cancer Center, Guangzhou, Guangdong,China. 6Department of Head and Neck Surgery, The University of Texas MDAnderson Cancer Center, Houston, Texas. 7The University of Texas GraduateSchool of Biomedical Sciences, Houston, Texas.

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Current address for R. Singh: University of Illinois, Urbana-Champaign, Illinois.

Corresponding Author: Faye M. Johnson, The University of Texas MD AndersonCancer Center, 1515 Holcombe Boulevard, Unit 432, Houston, TX 77030.Phone: 713-792-6363; Fax: 713-792-1220; E-mail: [email protected]

Clin Cancer Res 2019;25:3329–40

doi: 10.1158/1078-0432.CCR-18-3276

�2019 American Association for Cancer Research.

ClinicalCancerResearch

www.aacrjournals.org 3329

on June 17, 2021. © 2019 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst February 15, 2019; DOI: 10.1158/1078-0432.CCR-18-3276

http://crossmark.crossref.org/dialog/?doi=10.1158/1078-0432.CCR-18-3276&domain=pdf&date_stamp=2019-5-13http://clincancerres.aacrjournals.org/

Molecular therapies targeting activated oncogenes are com-mon, but only one such therapy has been approved for cancersdriven by the loss of tumor suppressor function: PARP inhibitorsfor BRCA1/BRCA2MUT breast cancer (13). Perhaps because offeedback pathways, targeting pathways activated downstream oftumor suppressors has not been effective for cancer therapy. Todate, the new genomic information available for this disease hasnot been translated into clinical care largely because the genomiclandscape is dominated by tumor suppressors. In this study, wetested the hypothesis that an unbiased pharmacogenomicsapproach would identify novel, translationally applicable molec-ular vulnerabilities to HNSCC with the loss of tumor suppressorfunction. To address the lack of effective molecularly targetedtherapies for HNSCC, we used an unbiased pharmacogenomicsapproach to integrate drug sensitivity data for seven diverse PI3K/mTOR pathway inhibitors in a panel of 59 molecularly charac-terized HNSCC cell lines (14, 15). We identified a striking cor-relation betweenNOTCH1 loss-of-function (LOF)mutations andsensitivity to PI3K/mTOR pathway inhibitors in HNSCC that weconfirmed with both in vitro and in vivo studies.

To the best of our knowledge, this is the first study to establish atherapeutic vulnerability of NOTCH1MUT HNSCC to any class ofdrugs. Thus, our findings have the potential to advance theapproval of a biomarker-driven targeted therapy for HNSCC.Further, based on the mechanism of sensitivity of NOTCH1MUT

HNSCC, we have identified a combination therapy that may beeffective against NOTCH1WT HNSCC.

Materials and MethodsCells and reagents

We obtained and subjected 50 human papilloma virus(HPV)-negative and 9 HPV-positive HNSCC cell lines towhole-exome sequencing, reverse-phase protein array analysis,and gene expression profiling as described previously (14, 16,17). All the cell lines were genotyped by short tandem repeatanalysis, and all cell lines were mycoplasma-free at the time oftesting with a Mycoplasma Detection Kit (Lonza). UMSCC49parental and NOTCH1 knockout (KO) cells were purchased

from Dr. Chad Brenner at the University of Michigan. Anerythromycin ribosomal methylase (ERM) plasmid expressingPDK1-GFP was a gift from Dr. Gordon Mills. Cells weretransfected with the PDK1-expressing plasmid with Lipofecta-mine 3000 (Life Technologies) for 6 hours and selected with 1to 2 mg/mL G418 (Sigma). To create the PJ34, FaDU, andMDA686LN NOTCH1 KO lines, we transfected the parental celllines with a NOTCH1 CRISPR/Cas9 KO plasmid (sc-421930;Santa Cruz) using GenJet DNA transfection reagent (Signagen).The transfected cells were sorted based on green fluorescentprotein expression, and individual clones were obtained.PQR309 was provided by PIQUR Therapeutics AG. All otherdrugs were purchased from Selleck Chemicals and prepared as10 mmol/L stock solutions in DMSO.

Cell viability assaysHNSCC cell lines were treated with DMSO (vehicle) or PI3K/

mTOR pathway inhibitors at seven different concentrations(0.018–9.613 mmol/L) for 72 hours. A CellTiter-Glo luminescentcell viability assay (Promega) was performed as described previ-ously (30, 31). Inhibitory concentration (IC) and AUC valueswere calculated using the drexplorer R package with a best-fitdose–response model (32). The combination indices were calcu-lated using the Chou–Talalaymethod (33) in CalcuSyn (Biosoft).We tested the reproducibility and robustness of the data generatedusing three quality control parameters as described previous-ly (28). These parameters were the concordance correlation coef-ficient, location shift, and maximum SD between two biologicalreplicates for three technical and two biological replicates. Basedon heuristics from our previous screening studies in lung can-cer (28), the cut-offs for reproducibility were a concordancecorrelation coefficient greater than 0.8, a location shift less than0.9, and a SD less than 0.23 based on the normal mixture fitmodel. Experiments not satisfying these criteria were repeated.The replicate with the smallest experimental variation measuredby the median of SD was chosen as a representative of thereplicates, and its IC values served as the final values for subse-quent analysis.

Western blot analysisWestern blot analysis was performed as described previous-

ly (27). In brief, cells were lysed with ice-cold lysis buffer, and thelysates were centrifuged at 20,000� g for 10 minutes at 4�C. Cellsamples containing equal amounts of protein were resolved usingSDS-PAGE, transferred to nitrocellulose membranes, and immu-noblotted with different primary antibodies. Protein expressionwas detected using a horseradish peroxidase–conjugated second-ary antibody (Bio-Rad) and electrochemiluminescence reagent(AmershamBiosciences). The antibodies are listed in Supplemen-tary Table S1.

Apoptosis and cell cycle assaysTo measure apoptosis, we performed TUNEL staining with an

APO-BRDU Kit (BD Biosciences) and Annexin V/propidiumiodide staining with an FITC Annexin V Apoptosis Detection Kit(BD Pharmingen) as described previously (35). For the cell-cycleanalysis, cells were harvested, fixed, incorporated with bromo-deoxyuridine (BrdUrd), and stained with 7-aminoactinomycin Dusing aBrdUFlowKit (BDBiosciences). Datawere acquiredwith athree-laser, 10-color Gallios flow cytometer (Beckman Coulter)and analyzed using Kaluza software (Beckman Coulter). All

Translational Relevance

The genomic landscape of head and neck squamous cellcarcinoma (HNSCC) is dominated by nontargetable tumorsuppressor genes such as NOTCH1, which is mutated in 18%of HNSCC. Molecularly targeted therapies directed towardsactivated oncogenes are common for cancers generally but rarefor cancers driven by the loss of tumor suppressor function.Our in vitro and in vivo studies revealed that PI3K inhibitioncauses apoptosis in only HNSCC with NOTCH1 loss-of-function mutations, independent of PIK3CAmutation status,via differential regulation of PDK1. This study, the first toestablish a therapeutic vulnerability of HNSCCwithNOTCH1mutations to any class of drugs, may inform the developmentof the first biomarker-driven targeted therapy for HNSCC.Because NOTCH1 loss-of-function mutations are common inother squamous cell carcinomas, including lung (8%) andesophageal (21%) carcinomas, our findings likely have impli-cations beyond HNSCC.

Sambandam et al.

Clin Cancer Res; 25(11) June 1, 2019 Clinical Cancer Research3330



http://clincancerres.aacrjournals.org/

apoptosis assays were performed in triplicate, and each test wascompleted twice on different days.

Colony formation assaysHNSSC cells were seeded in 60-mm plates. One day later, the

cells were treated with DMSO or the indicated drugs for 48 hours.The medium was changed, and the cells were incubated in drug-free medium for 14 to 21 days. The cell colonies were thenwashed, fixed in 10% formaldehyde, and stained with crystalviolet (0.5%, w/v). Colony images were taken with a GelCountTumour Colony Counter (Oxford Optronix Ltd.). The total col-onynumber andareawere counted andanalyzedusing the ImageJsoftware program (NIH, Bethesda, MD). Assays were performedin triplicate, and each test was completed twice on different days.

Mouse modelsThis study was performed in accordance with the Guide for the

Care and Use of Laboratory Animals of the National Institutes ofHealth and approved by MD Anderson's Institutional AnimalCare and Use Committee. Subcutaneous xenograft and orthoto-pic nude mouse tongue models were created as described previ-ously (18). GSK2126458 was administered by oral gavage 5 daysper week at 1 mg/kg (orthotopic model) or 3 mg/kg (subcuta-neous xenograft model). PQR309 was administered by oralgavage 5 days per week at 30 mg/kg (orthotopic model).

TUNEL tissue stainingHarvested tumor tissueswerefixed in10%formalin, embedded

in paraffin, cut into 5-mm sections, and stored until further use.Deparaffinized and rehydrated tissue sections were processed forDNA labeling by terminal deoxynucleotidyl transferase, and thenstreptavidin-conjugated horseradish peroxidase/diaminobenzi-dine was detected using a TUNEL Apoptosis Detection Kit (Trevi-gen). Images were taken at 40� magnification, and Image Jsoftware was used to calculate the percentage of TUNEL-positivecells.

Statistical analysisWeused the b-uniformmixturemodel to control the FDR (16).

To identify differentially expressed features between groups, weapplied modified two-sample t-tests using the Limma package inR. We used the Fisher exact test and Wilcoxon rank-sum test toevaluate associations between molecular characteristics and drugsensitivities. In vitro experiment results were compared using atwo-sample t test and two-way analysis of variance corrected formultiple comparisons with the Tukeymethod inGraphPad Prism7. In vivo experiment results were analyzed using a linear mixedmodel in R.

ResultsHNSCC cell lines with LOFNOTCH1mutations are sensitive toPI3K/mTOR pathway inhibitors in vitro

To identify novel predictive therapeutic vulnerabilities linkedto PI3K/mTOR pathway inhibition, we treated 59 HNSCC celllines with seven different PI3K/mTOR pathway inhibitors thatwere in clinical development at the start of this study (19–26). Ofthe drug–cell line combinations, 95%met our predefined qualitycontrol cutoff. The HNSCC cell lines exhibited diverse sensitivityto PI3K/mTOR pathway inhibitors (Supplementary Fig. S1A andS1B; Supplementary Table S2). Because the dose–response curves

for PI3K/mTOR pathway inhibitors often plateau near the IC50values (9), we used the more robust IC70 and AUC values asparameters for drug potency as described previously (27). Celllines were classified as sensitive based on IC70 values less than thepeak plasma concentration for each drug. To determine if cross-comparison of the PI3K/mTOR inhibitors as a class was feasible,we compared their drug sensitivity patterns with those of threecell-cycle kinase inhibitors we tested previously (17). The PI3K/mTOR pathway inhibitors clustered separately from the cell-cycleinhibitors (Supplementary Fig. S1C).

We then compared drug sensitivity to common driver genesidentified in HNSCC tumors from The Cancer Genome Atlas(TCGA; ref. 17). Our cell lines did not have any mutations in11 of the top 50 mutated genes, and six genes were mutated inonly one cell line, precluding comparison. Of the 515 sequencedTCGA HNSCC tumors, 92% had a mutation in at least one of theremaining 33 genes, demonstrating that our cell lines were geno-mically representative of patients with HNSCC. Of the 33 genes,only mutated NOTCH1 and KRTAP5-5 were significantly corre-latedwith sensitivity to six of the seven drugs (Fig. 1A).We did notstudyKRTAP5-5 further because it ismutated in only two cell linesand 6% of patients with HNSCC (1, 2).

Because PIK3CAmutations predict pathway inhibitor responsein HNSCC (9–11), we specifically examined their relationshipwith drug sensitivity.We found no consistent correlation betweenPIK3CA mutations and drug sensitivity when we included all 12PIK3CAMUT cell lines (Fig. 1A). Compared with PIK3CAWT celllines, PIK3CAMUT HNSCC cell lines (excluding three cell lineswithuncharacterizedmutations)were significantlymore sensitiveto four drugs and substantially more sensitive to two drugs(Fig. 1B). Themedian IC70 values of six drugs for theNOTCH1

MUT

lines were significantly lower than those of the drugs for theNOTCH1WT lines; the seventh drug, BKM120, was almost uni-versally effective in HNSCC cell lines, likely owing to its multipleoff-target effects (28). Two cell lines harbored bothNOTCH1 andPIK3CA mutations and were sensitive to all the inhibitors tested.Therewas nomutual exclusivity of these twomutations in the 515sequenced TCGA HNSCC patient samples, but there was a ten-dency towards co-occurrence (www.cbioportal.org: accessed onSeptember 15, 2018).

HNSCC patient samples have frequent truncating NOTCH1mutations (2).We found the distribution of thesemutations to bevery different from that of the activatingmutations found in T-cellacute lymphoblastic leukemia (T-ALL); mutations in HNSCC celllines were similar to the mutation pattern in HNSCC patientsamples (Fig. 1C andD; ref. 29).Of the 17HNSCC lines, we foundto harbor NOTCH1 mutations, six had truncating mutations(nonsense mutations and deletions) and 11 had missense muta-tions.Of those 11, threewere excluded fromanalysis because theirNOTCH1 alterations occurred in the nuclear regulatory domainand more likely represented SNPs. Of the 14 HNSCC lines withNOTCH1 LOF mutations (Fig. 1E), two had both bona fideNOTCH1 and PIK3CA mutations. Both NOTCH1 and cleavedNOTCH1 [NOTCH intracellular domain (NICD)] protein levelswere lower in NOTCH1MUT lines than in NOTCH1WT lines,although the differencewas not significant, likely owing to diverseprotein expression levels inNOTCH1WT lines.However, we founda significant positive correlation between NICD and totalNOTCH1 protein levels in the cell lines (Fig. 1F). TCGA HNSCCpatient samples also showed diverse NOTCH1protein expressionlevels regardless of NOTCH1 mutation status (Supplementary

NOTCH1-Mutant Squamous Cancer Sensitivity to PI3K Inhibition

www.aacrjournals.org Clin Cancer Res; 25(11) June 1, 2019 3331



www.cbioportal.orghttp://clincancerres.aacrjournals.org/

Sambandam et al.





Fig. S2A). NOTCH1 and NICD levels did not consistently corre-late with drug sensitivity in NOTCH1MUT or NOTCH1WT lines(Supplementary Fig. S2B–S2E).

Basal activation of the PI3K/mTOR pathway and HPV status donot predict sensitivity to PI3K/mTOR pathway inhibitors

The diverse drug sensitivity suggested that subsets of cell lineshave inherent molecular therapeutic vulnerabilities. Althoughthere was a trend for PIK3CAMUT cell lines to be more sensitiveto PI3K/mTOR pathway inhibitors, none of the 28 mutations inthe PI3K/mTOR pathway described previously (4) consistentlycorrelated with drug sensitivity (Supplementary Fig. S3A). Like-wise, we found no significant correlation between the basalactivation of the PI3K/mTOR pathway and drug sensitivity(Supplementary Fig. S3B and S3C; Supplementary Table S3). Ofnote, drug sensitivity did not correlate withRICTOR amplificationor PDK1 protein or gene expression (data not shown). AlthoughHPV-positive HNSCC is molecularly distinct from HPV-negativeHNSCC, drug sensitivity did not correlate with HPV status (Sup-plementary Fig. S4A and S4B).

We also compared basal gene and protein expression levelsbetween resistant and sensitive cell lines (Supplementary TablesS4 and S5) and two additional proteomic scores (27). There wasno robust correlation between DNA repair and epithelial tomesenchymal transition scores and sensitivity to drugs (r coeffi-cient < 0.4; Supplementary Fig. S4C and S4D). We did not studythese differentially expressed genes or proteins or these proteomicscores further owing to both inconsistency of the correlations anda lack of prior validation of the scores in HNSCC.

PI3K/mTORpathway inhibition induces apoptosis and reducesclonogenic growth in NOTCH1MUT cell lines

We then validated the susceptibility ofNOTCH1MUTHNSCC toPI3K/mTOR inhibitionwithGSK2126458.GSK2126458 inducedsignificant apoptosis in allNOTCH1MUT lines tested; no apoptosiswas detected in the two NOTCH1WT/PIK3CAWT lines or threeNOTCH1WT/PIK3CAMUT lines (Fig. 2A and B). BAY806946 andPQR309 elicited similar results (Supplementary Fig. S5A andS5B). Consistent with these data, GSK2126458, even at concen-trations well below the peak plasma concentration value, signif-icantly reduced the number of colonies formed and the totalcell number in NOTCH1MUT lines compared with NOTCH1WT

lines (Fig. 2C). GSK2126458 induced G1 arrest in all cell linesbut increased the sub-G0 fraction in only NOTCH1

MUT lines

(Supplementary Fig. S5C). We confirmed that GSK2126458 haddose-dependent target inhibition in representative NOTCH1WT

and NOTCH1MUT lines (Fig. 2D).

PI3K/mTOR pathway inhibition reduces tumor growth in bothorthotopic tongue and subcutaneous xenograft models ofNOTCH1MUT HNSCC

To assess the antitumor effect of PI3K/mTOR inhibition inNOTCH1MUT HNSCC, we first used two NOTCH1MUT lines tocreate subcutaneous xenograft models. The GSK2126458-treatedtumors regressed,whereas the vehicle-treated control tumors grewsignificantly. TUNEL staining showed that, consistent with thein vitro data, tumors treated with GSK2126458 had significantapoptosis (Fig. 3A and B).

We also used an orthotopic xenograft model of HNSCC, whoselocal growth patterns and histology are similar to those of humanHNSCC (18). In both NOTCH1MUT models, GSK2126458 andPQR309 significantly reduced tumor size compared with vehiclealone. TUNEL staining showed that the NOTCH1MUT tumorsunderwent apoptosis at the end of treatment (Fig. 3C–F).

CRISPR-Cas9 NOTCH1 KO sensitizes NOTCH1WT HNSCC toPI3K/mTOR pathway inhibitor�mediated apoptosis

We used four NOTCH1WT lines with NOTCH1 KO to test thehypothesis that abrogating NOTCH1 signaling rendersNOTCH1WT cells more sensitive to PI3K/mTOR pathway inhibi-tion. Compared with their parental cell lines, the PJ34 andUMSCC49 lines with NOTCH1 KO showed decreased cell viabil-ity after GSK2126458 treatment, with IC70 values that droppedfrom 0.1 to 0.024 mmol/L and from 1.48 to 0.36 mmol/L, respec-tively (Fig. 4A); had a significantly higher rate of apoptosis whentreated with PI3K/mTOR inhibitors (Fig. 4B and C); and formedsignificantly fewer colonies after PI3K/mTOR inhibition(Fig. 4D). However, NOTCH1 KO did not significantly increasethe sensitivity of the NOTCH1WT lines FaDU and MDA686LN toPI3K/mTOR pathway inhibition (Supplementary Fig. S6).

We also used the g-secretase inhibitors N-[N-(3,5-difluoro-phenacetyl-L-alanyl)]-S-phenylglycine t-butyl ester and diben-zazepine to inhibit the intra-membrane proteolytic cleavage/activation of NOTCH1. The combination of PI3K/mTOR andNOTCH1 signaling inhibitors decreased cell viability in fourNOTCH1WT lines (Supplementary Fig. S7A and S7B), with mostcombination indices of less than 1 for a range of fractionsaffected (Supplementary Table S6). Consistent with these

Figure 1.NOTCH1 LOFmutations and NOTCH1 signaling in HNSCC are correlated with sensitivity to PI3K/mTOR pathway inhibitors. A, HNSCC cell lines (n¼ 59) wereincubated with seven different PI3K/mTOR pathway inhibitors. The drugs' IC70 and AUC values in wild-type cell lines were compared with those in cell lines withmutations in 33 common driver genes using a modified two-sample t test at an FDR of 0.05. Positive and negative correlations between the presence of amutation and drug sensitivity are indicated by blue and red data points, respectively, with larger points representing significant differences. B, The IC70 values incell lines with wild-type or mutant NOTCH1 and PIK3CAwere compared using a Kruskal–Wallis test corrected for multiple comparisons by the Dunn test; P valuesare indicated in the figure. The dashed black lines indicate each drug's peak plasma concentration. N.S., not significant. C, The frequency and location ofmutations inNOTCH1 in HNSCC cell lines (upper gene map) are compared with those of T-ALL (lower map, below the gene) and HNSCC patients (upper map,above the gene) based on data from the Catalogue of Somatic Mutations in TCGA. Oncogenic T-ALL NOTCH1mutations occur in a hotspot of mostly missensemutations within the negative regulatory heterodimerization (NRD) domain or in a second hotspot of mostly truncating mutations near the C-terminus, deletingthe proline-glutamic acid-serine-threonine (PEST) domain and increasing the stabilization of activated NOTCH1 in the nucleus. In contrast, the truncatingmutations in HNSCC are scattered throughout the gene but not in the PEST domain, and the missense mutations cluster in the extracellular EGF ligand-bindingdomains.D,Most mutations in the NOTCH1 gene in TCGA HNSCC patient samples are missense mutations outside the C-terminal NRD and PEST domains. E,From our panel of 59 HNSCC cell lines, we identified 24 harboring NOTCH1 and/or PIK3CAmutations. Orange text corresponds to nonsense mutations outsidethe PEST domain; red text, frameshift mutations outside the PEST domain; blue text, missense mutations outside the NRD and PEST domains; and gray text, in-frame deletions. ¥, noncanonical mutations; †, suspected SNPs excluded from NOTCH1 LOFmutants. F, The basal expression levels of NOTCH1 and NOTCHintracellular domain (NICD) proteins in NOTCH1WT and NOTCH1MUT cell lines were compared using a two-sample t test. The correlation between the expressionlevels of NICD and NOTCH1 was assessed by Spearman correlation.






findings, the combination also increased apoptosis (Supple-mentary Fig. S7C–S7E). As with the NOTCH1-KO lines, com-bined g-secretase and PI3K/mTOR inhibition was not syner-gistic in two NOTCH1WT lines, likely because of the diversemolecular mechanisms of resistance.

PDK1mediates resistance toPI3K/mTORpathway inhibition inNOTCH1WT HNSCC

To identify the mechanism of sensitivity in NOTCH1MUT

HNSCC, we inhibited components of the PI3K/mTOR pathwayin bothNOTCH1MUT andNOTCH1WT cell lines. All cell lines wereresistant to the mTOR inhibitors (rapamycin and ridoforolimus)and the AKT inhibitor (MK-2206), but both the dual PI3K/mTORinhibitor (GSK2126458) and pan-PI3K inhibitor (BAY806946)had differential effects on NOTCH1MUT and NOTCH1WT cells

(Supplementary Fig. S8A). These results suggest that an importantsignaling node that is downstream of PI3K but independent ofAKT and mTOR is responsible for the differential sensitivity.

We then investigated PDK1's role in the differential sensitivity,because PDK1 is activated by PI3K, mediates resistance to PI3Kinhibition in breast cancer (30), and can regulate p70S6K, whichwas inhibited more robustly inNOTCH1MUT than inNOTCH1WT

HNSCC (Supplementary Fig. S8B). PI3K inhibition using threedifferent drugs reduced PDK1 expression in only NOTCH1MUT

lines (Fig. 5A and B; Supplementary Fig. S9A). To assess PDK1'srole in PI3K/mTOR inhibition–induced apoptosis in NOTCH1-MUT HNSCC, we induced PDK1's overexpression inNOTCH1MUT

HNSCC cells. Compared with parental lines, NOTCH1MUT cellsoverexpressing PDK1 had higher expression levels of p-AKT, p-S6,and p-4EBP1 (Fig. 5C), and they had decreased apoptosis after

Figure 2.

PI3K/mTOR signaling inhibitioninduces cell death in NOTCH1MUT

but not NOTCH1WT HNSCC cell linesin vitro. A, Annexin V/propidiumiodide or BrDU-TUNEL assays wereused to measure apoptosis inHNSCC cells with the indicatedNOTCH1 and PIK3CAmutations,which were treated with 50 nmol/LGSK2126458 (GSK212) for 24 or 48hours, respectively. Values are themean� SDs of three independentexperiments. � , P < 0.05, unpairedtwo-tailed Student t test. B,Western blotting of cleaved PARP(Cl-PARP) and cleaved caspase 3(Cl-Caspase 3) levels in mutant andwild-type cells after GSK212treatment for 48 hours. C,NOTCH1MUT and NOTCH1WT HNSCCcells were plated sparsely andtreated with GSK212 at theindicated concentrations for 48hours. At 14 to 21 days aftertreatment, the cells were fixed andstained, images were capturedusing the GelCount Tumour ColonyCounter, and colony numbers andthe colorimetric intensity of thecolonies were determined bymeasuring the optical density (OD)at 570 nm. Values are the mean�SDs of three independentexperiments. � , P < 0.001, unpairedtwo-tailed Student t test.D, ThreeNOTCH1WT lines (black text andbars) and three NOTCH1MUT lines(red text and bars) were treatedwith increasing concentrations ofGSK212 for 4 hours, and levels ofPI3K/mTOR pathway proteins weremeasured byWestern blot analysis.

Sambandam et al.





GSK2126458 treatment (Fig. 5D and E). As expected, PDK1overexpression did not inhibit the effect of GSK2126458 onphosphatidylinositol (3,4,5)-trisphosphate (PIP3)-dependentmolecules such as p-AKT (T308; Fig. 5E). In contrast, PDK1overexpression did inhibit the effect of GSK2126458 on PIP3-independent molecules such as c-Myc and p-TSC2 (Supplemen-tary Fig. S9B).

PDK1 inhibition with GSK2334470 alone had little effect oncell viability or apoptosis regardless of NOTCH1 status (Fig. 6).Therefore, we hypothesized that both AKT and PDK1 inhibitionare necessary for apoptosis following PI3K inhibition. In all celllines tested, AKT inhibition plus PDK1 inhibition decreased cellviability with combination indices consistent with a greater-than-additive effect (Fig. 6A; Supplementary Fig. S9C and S9D; Sup-plementary Table S7). The combination also led to apoptosis inall cell lines tested (Fig. 6B andC). InNOTCH1MUT lines, the effectof the combinationwas similar to that achievedwith PI3K/mTORinhibition alone (Fig. 6B).

DiscussionWith this study, we are the first to establish NOTCH1 LOF

mutations in HNSCC as a significant therapeutic vulnerabilityto PI3K/mTOR pathway inhibition. We found that, unlike

PIK3CAMUT HNSCC cell lines, which underwent only cell-cyclearrest following PI3K/mTOR inhibition, NOTCH1MUT HNSCCcell lines also underwent apoptosis. Likewise, PI3K/mTOR inhi-bitors decreasedNOTCH1MUTHNSCC tumor size in vivo. SelectivemTOR or AKT inhibitors did not have differential effects onNOTCH1WT or NOTCH1MUT HNSCC. This led us to examinePDK1, whose expression decreased following PI3K/mTOR inhi-bition in only NOTCH1MUT HNSCC, supporting the hypothesisthat persistent PDK1 activation leads to PI3K/mTOR inhibitionresistance in NOTCH1WT HNSCC despite robust AKT inhibition.Consistent with this hypothesis, a combination of drugs thatinhibit both PDK1 and AKT led to apoptosis.

Two other lines of evidence support our findings. First, thePI3K inhibitor PX-886 significantly reduced tumor growth intwo NOTCH1MUT HNSCC PDX models (10). Second, NOTCHpathway activation confers resistance to PI3K/mTOR inhibitorsvia c-Myc activation in breast cancer (31). Although we foundthat basal c-Myc gene or protein expression did not predictresponse, and c-Myc expression did not differ accordingto NOTCH1 mutation status (data not shown), c-Myc maybe differentially regulated downstream of PDK1 in a PIP3-independent manner.

Although activatingmutations in PIK3CA also occur inHNSCCand other solid tumors, PI3K/mTOR inhibitors have had limited

Figure 3.

PI3K/mTOR pathway inhibitors cause apoptosis and reduce tumor size inNOTCH1MUT HNSCC in vivo.A and B, The NOTCH1MUT cell lines HN31 and UMSCC22A(UM22A) were injected subcutaneously into athymic nude mice. After tumors reached 150mm3, the mice were randomized to receive either vehicle control(2-hydroxypropyl-b-cyclodextrin) or 1 mg/kg GSK2126458 (GSK212) by oral gavage daily for 28 days. Tumor sizes were measured and tumor volumes werecalculated at the indicated times, and the percent change in tumor volume at the end of therapy was calculated using the formula (final tumor volume� initialtumor volume/initial tumor volume)� 100%. The percentages of TUNEL-positive cells were calculated from image quantification at 40�magnification. C–F,HN31 and UM22A were implanted orthotopically (OT) into the anterior tongues of athymic nude mice. After tumors reached 1.5 to 2.0 mm3, the mice wererandomized to receive either vehicle control (16% 2-hydroxypropyl-b-cyclodextrin for GSK212 or 2-hydroxypropyl-b-cyclodextrin for PQR309) or GSK212(3 mg/kg) or PQR309 (30mg/kg) by oral gavage daily for 28 days. Tumor sizes were measured, and tumor volumes and percent changes in tumor volumes werecalculated as above. For the tumor growth curve analysis, we applied a linear model using generalized least squares with two variables (treatment and day) andan autocorrelation structure of order 1 in residuals. The percentages of TUNEL-positive cells were calculated as described above. Data are mean� SDs from fivemice per group; P values were calculated using an unpaired t test with theWelch correction.






clinical success. Compared with PIK3CAWT HNSCC, PIK3CAMUT

HNSCC cell lines and PDX models are more sensitive to PI3K/mTOR pathway inhibitors (4, 9–12, 32). However, PI3K/mTORinhibition does not cause significant apoptosis in PIK3CAMUT

HNSCC cell lines (9). These published findings are consistentwith those of this study and with clinical findings demonstratingthat PIK3CAMUT tumors tend to bemore sensitive than PIK3CAWT

tumors to PI3K/mTOR pathway inhibitors but still have modestclinical responses (4, 6, 7, 33). Because HNSCC tumors withNOTCH1 mutations undergo cell death (rather than simplegrowth arrest) in response to PI3K/mTOR inhibitors, we hypoth-esize that drugs targeting the PI3K/mTOR pathway have greaterclinical efficacy in this genomic subtype that in PIK3CAMUT orNOTCH1WT HNSCC.

Resistance toPI3K/mTORpathway inhibitorsmanifests as a lackof sustained pathway inhibition despite the use of potent agents.For example, mTORC1 inhibition blocks S6 kinase�dependentinhibition of insulin receptor substrate 1, leading to insulin-like

growth factor 1 receptor and then PI3K/AKT activation (34). Asecond feedback loop occurs because AKT activation inhibitsthe nuclear localization of forkhead box class O, a transcrip-tional driver of receptor tyrosine kinase (RTK) expression. AKTinhibition relieves this feedback suppression, leading to RTKexpression and subsequent ERK activation (35). Similarly, inHNSCC, the addition of MEK/ERK or RTK pathway inhibitionenhances antitumor effects of PI3K/mTOR inhibition (9, 36–38). However, we did not find differential effects on p-ERK afterPI3K/mTOR inhibition based on NOTCH1 status (data notshown). A third mechanism is the sustained PDK1 signalingthat mediates residual mTORC1 activity despite potent PI3Kainhibition in PI3Ka inhibitor�resistant PIK3CAMUT breastcancers (30). Our findings suggest that sustained PDK1 activityis a mechanism of resistance to PI3K/mTOR inhibition inNOTCH1WT HNSCC. Additionally, other factors may influenceresponse of HNSCC to PI3K/mTOR pathway inhibitors such asTP53 and TGFb (39).

Figure 4.

NOTCH1 KO sensitizes twoNOTCH1WT HNSCC cell lines to PI3K/mTOR pathway inhibition in vitro. TwoNOTCH1WT HNSCC cell lines [PJ34 and UMSCC49(UM49)], parental cell lines, and NOTCH1 KO cell lines were treated with PI3K/mTOR inhibitors. A, The CellTiter-Glo assay was used to assess the viability of cellstreated for 72 hours with the indicated concentrations of GSK2126458 (GSK212). B,Western blot analysis of apoptosis markers, NOTCH1, and NOTCH intracellulardomain (NICD) was performed after treatment with the indicated concentrations of GSK212 for 48 hours. C, Apoptosis in cells treated with 50 nmol/L GSK212 or200 nmol/L BEZ235 for 48 hours was measured using an Annexin V/propidium iodide assay. Values are the mean� SDs of three independent experiments. �, P <0.05, unpaired two-tailed Student t test. D, Cells were treated with 50 nmol/L GSK212 for 48 hours and then incubated in drug-free medium for 14 to 21 days.Colony areas as percentages of the control were measured using ImageJ. Data are presented as the mean� SDs of three independent experiments. � , P < 0.05,unpaired two-tailed Student t test.

Sambandam et al.





The mechanism underlying this differential effect on PDK1 inHNSCC is unknown. Although crosstalk between the PI3K/mTORand NOTCH1 pathways has been extensively studied in T-ALL, inwhichNOTCH1 acts as an oncogene (29), the way in which thesepathways interact in solid tumors is unknown. In T-ALL, theactivation of hes family bHLH transcription factor 1 (HES1), adownstream target of NOTCH1, suppresses PTEN expression,leading to PI3K activation and resistance to NOTCH1 inhibi-tion (29). In this study, NOTCH1 inhibition did not affect AKTactivation in HNSCC. We speculate that in NOTCH1WT HNSCC,PDK1 is activated independently of PI3K. In T-ALL (40) andthymocytes (41), PDK1 activation downstream of NOTCH1signaling mediates survival and proliferation.

One candidate through which PDK1 becomes activated iscollagen 9 type alpha 1 (COL11A1). In ovarian cancer cells,

COL11A1 binds to PDK1 and attenuates its ubiquitination anddegradation, leading to chemoresistance (42); and in mouseembryonic fibroblasts, COL11A1 expression is downregulatedafter NOTCH1 inhibition (43). A second candidate is peroxisomeproliferator-activated receptors, which are transcription factorsinduced byNOTCH1 inhibition in fatty liver (44) and can activatePDK1, leading to increased keratinocyte survival (45). In addi-tion, the NOTCH1 pathway can confer drug resistance through avariety of other mechanisms, including epithelial-to-mesenchy-mal transition (46), and by promoting the cancer stem cellphenotype (47), which leads to tumor proliferation despite PI3Kpathway inhibition.

In our study, NOTCH1 inhibition did not sensitize someNOTCH1WT HNSCC cell lines to PI3K/mTOR inhibitors. Wespeculate that other diverse mechanisms mediate resistance to

Figure 5.

PDK1 mediates resistance to PI3K/mTOR pathway inhibition inNOTCH1WT HNSCC lines. A, TwoNOTCH1WT lines and fourNOTCH1MUT lines were treated with50 nmol/L GSK2126458 (GSK212)for 24 hours and then subjected toWestern blot analysis with theindicated antibodies to assess theexpression of 3-phosphoinositidedependent kinase 1 (PDK1) andphosphorylated PDK1. B, Proteinexpression fromWestern blottingwas quantified using ImageJ andnormalized using b-actin as acontrol, and the relative fold controlfrom untreated was calculated. Barsindicate mean� SEMs from twoNOTCH1WT cell lines and fourNOTCH1MUT cell lines. � , P < 0.05,one-tailed Student t test. C,Weestablished NOTCH1MUT lines (HN31,PCI15B, and UMSCC22A) with PDK1overexpression, which wasconfirmed byWestern blot analysis.D and E, Parental and PDK1-overexpressing NOTCH1MUT cellswere treated with 50 nmol/LGSK212 for 24 or 48 hours, andapoptosis was assessed bymeasuring levels of cleaved PARP(Cl-PARP) and cleaved caspase 3(Cl-Caspase 3) by Annexin V/propidium iodide assay (D) andWestern blot analysis (E). Thepercentages of apoptotic cells areexpressed as the mean� SDs ofthree independent experiments.� , P < 0.001, two-way ANOVAcorrected for multiple comparisonsby the Tukey test.






PI3K/mTOR pathway inhibitors, including PTEN loss (48) andAXL activation (12). However, the consistent sensitivity ofNOTCH1MUTHNSCC tomultiple PI3K/mTORpathway inhibitorssupports the vulnerability of this molecular subtype despite thevaried effects of NOTCH1manipulations inNOTCH1WT HNSCC.

This is the first study to establish a therapeutic vulnerability ofNOTCH1MUT HNSCC to any class of drugs. Because NOTCH1LOF mutations are common in other squamous cell carcinomas,including lung (8%) and esophageal (21%) carcinomas (49), ourfindings have wide-ranging implications. Our finding thatNOTCH1 LOFmutations predict response to PI3K inhibitors maylead to the first biomarker-driven targeted therapy for HNSCC. Inaddition, targeting PDK1 may sensitize NOTCH1WT HNSCC toPI3K/mTOR pathway inhibitors.

Disclosure of Potential Conflicts of InterestF.M. Johnson reports receiving other commercial research support from

PIQUR and Trovagene. No potential conflicts of interest were disclosed by theother authors.

AcknowledgmentsWe thank Doriano Fabbro, PhD, of PIQUR Therapeutics AG for guidance

regarding PQR309 use in experiments and Joe Munch in MD Anderson'sDepartment of Scientific Publications for editing the manuscript. This workwas supported by philanthropic contributions to The University of Texas MDAnderson Cancer Center's Oropharynx Discovery Program (to J.N. Myers,F.M. Johnson); by grants from the National Institutes of Health/NationalInstitute of Dental and Craniofacial Research (U01DE025181, to J.N. Myersand M.J. Frederick; R01 DE024179-01A1, to M.J. Frederick); and by PIQURTherapeutics AG. This work used the services of MD Anderson's FlowCytometry and Cellular Imaging Core and Bioinformatics Shared Resource,which are supported by the NIH through MD Anderson's Cancer CenterSupport Grant (P30CA016672).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received October 16, 2018; revised December 20, 2018; accepted February11, 2019; published first February 15, 2019.

Figure 6.

AKT inhibition and PDK1 inhibition synergistically induce apoptosis in NOTCH1WT HNSCC lines. A, NOTCH1WT HNSCC cell lines were treated with increasingconcentrations of MK2206, GSK2334470 (GSK233), or the MK2206-GSK233 combination at a fixed 1:1 ratio for 72 hours, and cell viability was measured with theCellTiter-Glo assay. The combination index (CI) values were calculated with CalcuSyn, and the fraction affected (Fa) and CI are plotted as heatmaps for all fourcell lines. B,NOTCH1WT (black text) and NOTCH1MUT (red text) HNSCC cells were treated with 50 nmol/L GSK2126458 (GSK212), 10 mmol/L GSK233, 10 mmol/LMK2206, or the GSK233-MK2206 combination for 48 hours. Western blot analysis showed that the combination increased cleaved PARP (Cl-PARP) and cleavedcaspase 3 (Cl-Casp3) in the NOTCH1WT lines. C, Four NOTCH1WT HNSCC cell lines were treated with 50 nmol/L GSK212, 10 mmol/L GSK233, 10 mmol/L MK2206, orthe GSK233-MK2206 combination for 48 hours. Apoptosis was measured by Annexin V/propidium iodide assay, and the percentages of apoptotic cells areexpressed as the mean� SDs of three independent experiments. � , P < 0.001; �� , P < 0.05, two-way analysis of variance corrected for multiple comparisons withthe Tukey test.

Sambandam et al.





References1. Cancer Genome Atlas Network. Comprehensive genomic characterization

of head and neck squamous cell carcinomas. Nature 2015;517:576–82.2. Agrawal N, Frederick MJ, Pickering CR, Bettegowda C, Chang K, Li RJ, et al.

Exome sequencing of head and neck squamous cell carcinoma revealsinactivating mutations in NOTCH1. Science 2011;333:1154–7.

3. Iglesias-Bartolome R, Martin D, Gutkind JS. Exploiting the head and neckcancer oncogenome: widespread PI3K-mTOR pathway alterations andnovel molecular targets. Cancer Discov 2013;3:722–5.

4. Lui VW,HedbergML, LiH, Vangara BS, Pendleton K, Zeng Y, et al. Frequentmutation of the PI3K pathway in head and neck cancer defines predictivebiomarkers. Cancer Discov 2013;3:761–9.

5. Rodon J, Tabernero J. Improving the armamentarium of PI3K inhibitorswith isoform-selective agents: a new light in the darkness. Cancer Discov2017;7:666–9.

6. Janku F,Wheler JJ,Westin SN,Moulder SL, Naing A, Tsimberidou AM, et al.PI3K/AKT/mTOR inhibitors in patients with breast and gynecologic malig-nancies harboring PIK3CA mutations. J Clin Oncol 2012;30:777–82.

7. Jimeno A, Shirai K, Choi M, Laskin J, Kochenderfer M, Spira A, et al. Arandomized, phase II trial of cetuximab with or without PX-866, anirreversible oral phosphatidylinositol 3-kinase inhibitor, in patients withrelapsed or metastatic head and neck squamous cell cancer. Ann Oncol2015;26:556–61.

8. Soria JC, Ohe Y, Vansteenkiste J, Reungwetwattana T, Chewaskulyong B,Lee KH, et al. Osimertinib in untreated EGFR-mutated advanced non-small-cell lung cancer. N Engl J Med 2018;378:113–25.

9. Mazumdar T, Byers LA, Ng PK, Mills GB, Peng S, Diao L, et al. A compre-hensive evaluation of biomarkers predictive of response to PI3K inhibitorsand of resistance mechanisms in head and neck squamous cell carcinoma.Mol Cancer Ther 2014;13:2738–50.

10. Keysar SB, Astling DP, Anderson RT, Vogler BW, Bowles DW, Morton JJ,et al. A patient tumor transplant model of squamous cell cancer identifiesPI3K inhibitors as candidate therapeutics in defined molecular bins.Mol Oncol 2013;7:776–90.

11. Li H, Wawrose JS, Gooding WE, Garraway LA, Lui VW, Peyser ND, et al.Genomic analysis of head and neck squamous cell carcinoma cell lines andhuman tumors: a rational approach to preclinical model selection.Mol Cancer Res 2014;12:571–82.

12. ElkabetsM, Pazarentzos E, JuricD, ShengQ, Pelossof RA, Brook S, et al. AXLmediates resistance to PI3Kalpha inhibition by activating the EGFR/PKC/mTOR axis in head and neck and esophageal squamous cell carcinomas.Cancer Cell 2015;27:533–46.

13. Tutt A, Robson M, Garber JE, Domchek SM, Audeh MW, Weitzel JN, et al.Oral poly(ADP-ribose) polymerase inhibitor olaparib in patients withBRCA1 or BRCA2 mutations and advanced breast cancer: a proof-of-concept trial. Lancet 2010;376:235–44.

14. Kalu NN, Mazumdar T, Peng S, Shen L, Sambandam V, Rao X, et al.Genomic characterizationof humanpapillomavirus-positive and -negativehuman squamous cell cancer cell lines. Oncotarget 2017;8:86369–83.

15. ZhaoM, SanoD, Pickering CR, Jasser SA,Henderson YC, ClaymanGL, et al.Assembly and initial characterization of a panel of 85 genomically vali-dated cell lines from diverse head and neck tumor sites. Clin Cancer Res2011;17:7248–64.

16. Kalu NN, Mazumdar T, Peng S, Tong P, Shen L, Wang J, et al. Compre-hensive pharmacogenomic profiling of human papillomavirus-positiveand -negative squamous cell carcinoma identifies sensitivity to aurorakinase inhibition in KMT2D mutants. Cancer Lett 2018;431:64–72.

17. Zhang M, Singh R, Peng S, Mazumdar T, Sambandam V, Shen L, et al.Mutations of the LIM protein AJUBA mediate sensitivity of head and necksquamous cell carcinoma to treatment with cell-cycle inhibitors.Cancer Lett 2017;392:71–82.

18. Myers JN, Holsinger FC, Jasser SA, Bekele BN, Fidler IJ. An orthotopic nudemouse model of oral tongue squamous cell carcinoma. Clin Cancer Res2002;8:293–8.

19. Munster P, Aggarwal R,HongD, Schellens JH, van derNoll R, Specht J, et al.First-in-human phase I study of GSK2126458, an oral pan-class I phos-phatidylinositol-3-kinase inhibitor, in patients with advanced solid tumormalignancies. Clin Cancer Res 2016;22:1932–9.

20. Bendell JC, Kurkjian C, Infante JR, Bauer TM, Burris HA 3rd, Greco FA, et al.A phase 1 study of the sachet formulation of the oral dual PI3K/mTOR

inhibitor BEZ235 given twice daily (BID) in patients with advanced solidtumors. Invest New Drugs 2015;33:463–71.

21. Liu N, Rowley BR, Bull CO, Schneider C, Haegebarth A, Schatz CA, et al.BAY 80-6946 is a highly selective intravenous PI3K inhibitor with potentp110alpha and p110delta activities in tumor cell lines and xenograftmodels. Mol Cancer Ther 2013;12:2319–30.

22. Bendell JC, Rodon J, Burris HA, de JongeM, Verweij J, Birle D, et al. Phase I,dose-escalation study of BKM120, an oral pan-Class I PI3K inhibitor, inpatients with advanced solid tumors. J Clin Oncol 2012;30:282–90.

23. De Buck SS, Jakab A, Boehm M, Bootle D, Juric D, Quadt C, et al.Population pharmacokinetics and pharmacodynamics of BYL719, a phos-phoinositide 3-kinase antagonist, in adult patients with advanced solidmalignancies. Br J Clin Pharmacol 2014;78:543–55.

24. Salphati L, Pang J, Plise EG, Lee LB, Olivero AG, Prior WW, et al. Preclinicalassessment of the absorption and disposition of the phosphatidylinositol3-kinase/mammalian target of rapamycin inhibitor GDC-0980 and pre-diction of its pharmacokinetics and efficacy in human. DrugMetab Dispos2012;40:1785–96.

25. Wicki A, Brown N, Xyrafas A, Bize V, Hawle H, Berardi S, et al. First-inhuman, phase 1, dose-escalation pharmacokinetic and pharmacodynamicstudy of the oral dual PI3K and mTORC1/2 inhibitor PQR309 in patientswith advanced solid tumors (SAKK 67/13). Eur J Cancer 2018;96:6–16.

26. Beaufils F, Cmiljanovic N, Cmiljanovic V, Bohnacker T, Melone A, MaroneR, et al. 5-(4,6-Dimorpholino-1,3,5-triazin-2-yl)-4-(trifluoromethyl)pyridin-2-amine (PQR309), a potent, brain-penetrant, orally bioavailable,pan-class I PI3K/mTOR inhibitor as clinical candidate in oncology. J MedChem 2017;60:7524–38.

27. Ferrarotto R,Goonatilake R, Yoo SY, Tong P,Giri U, Peng S, et al. Epithelial-mesenchymal transition predicts polo-like kinase 1 inhibitor-mediatedapoptosis in non-small cell lung cancer. ClinCancer Res 2016;22:1674–86.

28. Bohnacker T, Prota AE, Beaufils F, Burke JE, Melone A, Inglis AJ, et al.Deconvolution of Buparlisib's mechanism of action defines specific PI3Kand tubulin inhibitors for therapeutic intervention. Nat Commun 2017;8:14683.

29. Hales EC, Taub JW, Matherly LH. New insights into Notch1 regulation ofthe PI3K-AKT-mTOR1 signaling axis: targeted therapy of gamma-secretaseinhibitor resistant T-cell acute lymphoblastic leukemia. Cell Signal 2014;26:149–61.

30. Castel P, Ellis H, Bago R, Toska E, Razavi P, Carmona FJ, et al. PDK1-SGK1signaling sustains AKT-independent mTORC1 activation and confers resis-tance to PI3Kalpha inhibition. Cancer Cell 2016;30:229–42.

31. MuellnerMK,Uras IZ,GappBV, KerzendorferC, SmidaM, LechtermannH,et al. A chemical-genetic screen reveals a mechanism of resistance to PI3Kinhibitors in cancer. Nat Chem Biol 2011;7:787–93.

32. Zumsteg ZS, Morse N, Krigsfeld G, Gupta G, Higginson DS, Lee NY, et al.Taselisib (GDC-0032), a potent beta-sparing small molecule inhibitor ofPI3K, radiosensitizes head and neck squamous carcinomas containingactivating PIK3CA alterations. Clin Cancer Res 2016;22:2009–19.

33. Cai Y, Dodhia S, Su GH. Dysregulations in the PI3K pathway and targetedtherapies for head and neck squamous cell carcinoma. Oncotarget 2017;8:22203–17.

34. Wang Z, Valera JC, Zhao X, Chen Q, Silvio Gutkind J. mTOR co-targetingstrategies for head andneck cancer therapy. CancerMetastasis Rev 2017;36:491–502.

35. Chandarlapaty S, Sawai A, Scaltriti M, Rodrik-Outmezguine V, Grbovic-Huezo O, Serra V, et al. AKT inhibition relieves feedback suppression ofreceptor tyrosine kinase expression and activity. Cancer Cell 2011;19:58–71.

36. Wang Z, Martin D, Molinolo AA, Patel V, Iglesias-Bartolome R, Degese MS,et al. mTOR co-targeting in cetuximab resistance in head and neck cancersharboring PIK3CA and RAS mutations. J Natl Cancer Inst 2014;106:1–11.

37. Yamaguchi K, Iglesias-Bartolom�e R, Wang Z, Callejas-Valera JL, Amorn-phimolthamP,Molinolo AA, et al. A synthetic-lethality RNAi screen revealsan ERK-mTOR co-targeting pro-apoptotic switch in PIK3CA(þ) oral can-cers. Oncotarget 2016;7:10696–709.

38. Mohan S, Vander Broek R, Shah S, Eytan DF, Pierce ML, Carlson SG, et al.MEK inhibitor PD-0325901 overcomes resistance to PI3K/mTOR inhibitorPF-5212384 and potentiates antitumor effects in human head and necksquamous cell carcinoma. Clin Cancer Res 2015;21:3946–56.






39. Herzog A, Bian Y, Vander Broek R, Hall B, Coupar J, Cheng H, et al.PI3K/mTOR inhibitor PF-04691502 antitumor activity is enhancedwith induction of wild-type TP53 in human xenograft and murineknockout models of head and neck cancer. Clin Cancer Res 2013;19:3808–19.

40. Lee K, Nam KT, Cho SH, Gudapati P, Hwang Y, Park DS, et al. Vital roles ofmTOR complex 2 in Notch-driven thymocyte differentiation and leuke-mia. J Exp Med 2012;209:713–28.

41. Kelly AP, FinlayDK,HintonHJ, Clarke RG, Fiorini E, Radtke F, et al. Notch-induced T cell development requires phosphoinositide-dependent kinase1. EMBO J 2007;26:3441–50.

42. Wu YH, Chang TH, Huang YF, Chen CC, Chou CY. COL11A1 conferschemoresistance on ovarian cancer cells through the activation of Akt/c/EBPbeta pathway and PDK1 stabilization. Oncotarget 2015;6:23748–63.

43. Abad M, Hashimoto H, Zhou H, Morales MG, Chen B, Bassel-DubyR, et al. Notch inhibition enhances cardiac reprogramming byincreasing MEF2C transcriptional activity. Stem Cell Reports 2017;8:548–60.

44. Song NJ, Yun UJ, Yang S, Wu C, Seo CR, Gwon AR, et al. Notch1 deficiencydecreases hepatic lipid accumulation by induction of fatty acid oxidation.Sci Rep 2016;6:19377.

45. Di-Poi N, Tan NS, Michalik L, Wahli W, Desvergne B. Antiapoptotic role ofPPARbeta in keratinocytes via transcriptional control of the Akt1 signalingpathway. Mol Cell 2002;10:721–33.

46. Liu W, Zhang J, Gan X, Shen F, Yang X, Du N, et al. LGR5 promotesepithelial ovarian cancer proliferation, metastasis, and epithelial-mesen-chymal transition through the Notch1 signaling pathway. Cancer Med2018 May 18 [Epub ahead of print].

47. Venkatesh V, Nataraj R, Thangaraj GS, Karthikeyan M, Gnanasekaran A,Kaginelli SB, et al. Targeting Notch signalling pathway of cancer stem cells.Stem Cell Investig 2018;5:5.

48. Juric D, Castel P, GriffithM,GriffithOL,WonHH, Ellis H, et al. Convergentloss of PTEN leads to clinical resistance to a PI(3)Kalpha inhibitor. Nature2015;518:240–4.

49. Zhang M, Biswas S, Qin X, Gong W, Deng W, Yu H. Does Notch play atumor suppressor role across diverse squamous cell carcinomas? CancerMed 2016;5:2048–60.


Sambandam et al.




2019;25:3329-3340. Published OnlineFirst February 15, 2019.Clin Cancer Res Vaishnavi Sambandam, Mitchell J. Frederick, Li Shen, et al. Carcinoma Vulnerability to Therapeutic PI3K/mTOR Inhibition

-Mutated Head and Neck SquamousNOTCH1PDK1 Mediates

Updated version

10.1158/1078-0432.CCR-18-3276doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2019/02/15/1078-0432.CCR-18-3276.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/25/11/3329.full#ref-list-1

This article cites 48 articles, 19 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/25/11/3329.full#related-urls

This article has been cited by 1 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/25/11/3329To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-18-3276http://clincancerres.aacrjournals.org/content/suppl/2019/02/15/1078-0432.CCR-18-3276.DC1http://clincancerres.aacrjournals.org/content/25/11/3329.full#ref-list-1http://clincancerres.aacrjournals.org/content/25/11/3329.full#related-urlshttp://clincancerres.aacrjournals.org/cgi/alertsmailto:[email protected]://clincancerres.aacrjournals.org/content/25/11/3329http://clincancerres.aacrjournals.org/

/ColorImageDict > /JPEG2000ColorACSImageDict > /JPEG2000ColorImageDict > /AntiAliasGrayImages false /CropGrayImages false /GrayImageMinResolution 200 /GrayImageMinResolutionPolicy /Warning /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 300 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict > /GrayImageDict > /JPEG2000GrayACSImageDict > /JPEG2000GrayImageDict > /AntiAliasMonoImages false /CropMonoImages false /MonoImageMinResolution 600 /MonoImageMinResolutionPolicy /Warning /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 900 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict > /AllowPSXObjects false /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (None) /PDFXOutputConditionIdentifier () /PDFXOutputCondition () /PDFXRegistryName () /PDFXTrapped /False

/CreateJDFFile false /Description > /Namespace [ (Adobe) (Common) (1.0) ] /OtherNamespaces [ > /FormElements false /GenerateStructure false /IncludeBookmarks false /IncludeHyperlinks false /IncludeInteractive false /IncludeLayers false /IncludeProfiles false /MarksOffset 18 /MarksWeight 0.250000 /MultimediaHandling /UseObjectSettings /Namespace [ (Adobe) (CreativeSuite) (2.0) ] /PDFXOutputIntentProfileSelector /NA /PageMarksFile /RomanDefault /PreserveEditing true /UntaggedCMYKHandling /LeaveUntagged /UntaggedRGBHandling /LeaveUntagged /UseDocumentBleed false >> > ]>> setdistillerparams> setpagedevice

Documents

PDK1 Mediates NOTCH1-Mutated Head and Neck Squamous … · Translational Cancer Mechanisms and Therapy PDK1 Mediates NOTCH1-Mutated Head and Neck Squamous Carcinoma Vulnerability