inhibition of pi3k/bmx cell survival pathway sensitizes to ...doxorubicin sensitivity and...

Small Molecule Therapeutics

Inhibition of PI3K/BMX Cell Survival PathwaySensitizes to BH3 Mimetics in SCLCDanielle S. Potter1, MelanieGalvin1, Stewart Brown1, Alice Lallo1,Cassandra L. Hodgkinson1,Fiona Blackhall2,3, Christopher J. Morrow1, and Caroline Dive1,4

Abstract

Most small cell lung cancer (SCLC) patients are initiallyresponsive to cytotoxic chemotherapy, but almost all undergofatal relapse with progressive disease, highlighting an urgentneed for improved therapies and better patient outcomes inthis disease. The proapoptotic BH3 mimetic ABT-737 thattargets BCL-2 family proteins demonstrated good single-agentefficacy in preclinical SCLC models. However, so far clinicaltrials of the BH3 mimetic Navitoclax have been disappointing.We previously demonstrated that inhibition of a PI3K/BMXcell survival signaling pathway sensitized colorectal cancercells to ABT-737. Here, we show that SCLC cell lines, whichexpress high levels of BMX, become sensitized to ABT-737upon inhibition of PI3K in vitro, and this is dependent on

inhibition of the PI3K-BMX-AKT/mTOR signaling pathway.Consistent with these cell line data, when combined withNavitoclax, PI3K inhibition suppressed tumor growth in bothan established SCLC xenograft model and in a newly estab-lished circulating tumor cell–derived explant (CDX) modelgenerated from a blood sample obtained at presentation from achemorefractory SCLC patient. These data show for the firsttime that a PI3K/BMX signaling pathway plays a role in SCLCcell survival and that a BH3 mimetic plus PI3K inhibitioncauses prolonged tumor regression in a chemorefractory SCLCpatient–derived model in vivo. These data add to a body ofevidence that this combination should move toward the clinic.Mol Cancer Ther; 15(6); 1248–60. �2016 AACR.

IntroductionSmall cell lung cancer (SCLC), an aggressive neuroendocrine

tumor with a 5-year survival rate of <5% (1), accounts forapproximately 15% of newly diagnosed lung cancer cases andapproximately 200,000deathsworldwide each year (2). Surgery isperformed on <5% cases, and standard of care (SOC) is platinum-based chemotherapy combined with etoposide � radiation (3).Without treatment,median overall survival (OS) of SCLCpatientsis 2 to 4 months (4). Treatment increases median OS to 8 to 20months (5). Although approximately 80% patients initiallyrespond to chemotherapy, virtually all relapse with progressivedisease within 3 to 18months (5–7). The molecular mechanismsunderlying this acquired drug resistance are poorly understood(8). Approximately 20%patients are classified as chemorefractorywith disease progression within 90 days of completing chemo-therapy (5, 7). Multiple targeted therapies have been evaluatedin SCLC, none improved patient outcomes (9), and SOC

chemotherapy remains three decades after its introduction. Abetter understanding of SCLC biology, improved preclinicalmodels, discovery of druggable targets, and biomarker-led clinicaltrials is warranted in this dismal disease.

Hallmark genetic aberrations in SCLC involve tumor suppres-sors TP53 and RB1 with bi-allelic losses in 100% and 93% of thecases, respectively (10), amplification ofMYC family genes (19%of patients; ref. 10) and of BCL-2 (�40% patients; refs. 11, 12).Commonly recurring, druggable oncogenic drivers are elusive,although mutually exclusive mutations are observed in geneswithin the PI3K pathway (36% of patients; ref. 13). Three land-mark studies comprehensively characterized the SCLC genomiclandscape, identifying alterations in genes encoding histonemodifying proteins and transcription factors, including SOX2and NOTCH family genes (10, 14, 15). These studies revealeda mutation rate of 5.5 to 8.6 coding mutations per Mb (10, 14,15). Only 22 of 8,000 protein coding mutations were observed in>1 SCLC tumor explaining the remarkable diversity (15). Con-sequently, identifying novel anticancer strategies for SCLC treat-ment remains challenging. Given the commonly observed upre-gulation of BCL-2 in SCLC (11, 12), one rational combinationtherapy under investigation is combining BCL-2 family apoptosisregulators with PI3K/mTOR survival signal inhibitors (14, 15).

The BCL-2 family subdivides into antiapoptotic proteins, proa-poptotic effectors, and proapoptotic BH3-only proteins. Proa-poptotic effectors, BAX and BAK, once activated form homo-oligomers creating pores that drive mitochondrial outer mem-brane permeabilization (MOMP) and release of cytochrome c(16). The antiapoptotic subfamily (BCL-2, BCL-xL, BCL-w, MCL-1, and A1) prevents MOMP via interactions with proapoptoticfamily members. BH3-only proteins further subdivide into acti-vators (BIM and BID) or sensitizers (BAD, BMF, HRK, NOXA, andPUMA). Activators directly activate BAX and BAK. Activators and

1Clinical and Experimental Pharmacology Group, Cancer Research UKManchester Institute, University of Manchester, Manchester, UnitedKingdom. 2Institute of Cancer Sciences, University of Manchester,Manchester, United Kingdom. 3Christie NHS Foundation Trust, Man-chester, United Kingdom. 4CRUK Lung Cancer Centre of Excellence,Manchester, United Kingdom.

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Corresponding Author: Caroline Dive, Cancer Research UK Manchester Insti-tute, Wilmslow Road, Withington, Manchester M20 4BX, United Kingdom.Phone: 0044-161-446-3036; Fax: 0044-161-446-3019; E-mail:[email protected]

doi: 10.1158/1535-7163.MCT-15-0885

�2016 American Association for Cancer Research.

MolecularCancerTherapeutics

Mol Cancer Ther; 15(6) June 20161248

on January 21, 2021. © 2016 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst March 29, 2016; DOI: 10.1158/1535-7163.MCT-15-0885

http://mct.aacrjournals.org/

sensitizers antagonize antiapoptotic family members releasingBAX and BAK for activation (17–20). The BH3 mimetic ABT-737(developed by AbbVie) binds BCL-2 and BCL-xL (with pooraffinity for MCL-1) to block BCL-2 family member interactions.In vitro, ABT-737 showed impressive single-agent efficacy acrossSCLC cell panels and in xenograft models (21–23), although highMCL-1 expression caused resistance (21, 22). Despite promisingpreclinical data (21, 23), clinical trials of Navitoclax (ABT-263,bioavailable ABT-737 analogue) as a single agent were disap-pointing (24), and rational combination strategies are nowconsidered critical for progress.

Phosphoinositide 3-kinases, a conserved family of lipidkinases, catalyze phosphorylation of D-3 hydroxyl groups oninositol rings of phosphatidlyinositol (Ptdlns) species. Activatedclass IA PI3K heterodimers (p110 catalytic and p85 regulatorysubunits) transduce signals regulating cell metabolism, prolifer-ation, polarity, and survival (25) and are implicated in tumori-genesis (26). We showed that BMX, a TEC family member ofnonreceptor tyrosine kinase and seldom studied PI3K down-stream effector, had an antiapoptotic function in colorectal cancercell lines where inhibition of PI3K/BMX signaling sensitized toABT-737 (27). BMX overexpression in SCLCH446 cells decreaseddoxorubicin sensitivity and upregulated BCL-2 and BCL-xL, andmoreover, BMX, BCL-2, and BCL-xL levels were correlated in 75%of SCLC patient samples (28). BMX expression was higher inSCLC chemoresistant H69AR compared with parental H69 cellswhere BMX knockdown re-sensitized to chemotherapeutics(including cisplatin and etoposide), potentially via downregula-tion of BCL-xL (29).

We sought validation evidence for combining BH3 mimeticswith PI3K inhibitors in SCLC and examined the role of BMXdownstreamof PI3Kusing establishedmodels and anovel patientderived Circulating tumor cell–Derived eXplant (CDX; ref. 30).

Materials and MethodsCell culture and drugs

COR-L103 (Anne White), NCI-H82, DMS79, NCI-H187,NCI-H146, NCI-H526, NCI-H1048, NCI-H524, and DMS114(American Type Culture Collection) cells were cultured in RPMImedia (Life Technologies) with 10% FBS (BioWest). All cell lineswere incubated in ahumidified atmosphere at 37�Cwith 5%CO2.Cell lines were authenticated using AmpFlSTR every 3 months(Applied Biosystems). ABT-737 (AbbVie), PI-103, KU-0063794(Merck), MK-2206, and Ibrutinib (Selleck) were dissolved inDMSO (10 mmol/L; Sigma) and stored at �20�C. Navitoclax(AbbVie) was formulated in 4�C in 10% ethanol, 30% poly-ethylene glycol 400 (Sigma), and 60% Phosal 50 PG (AmericanLecithinCompany).GDC-0941 (LancrixChemicals)was stored at�20�C in 10% DMSO, 5% Tween 20 (Sigma), and 85% sterilesaline. Once formulated, Navitoclax and GDC-0941 were storedat room temperature (RT) for up to 7 days.

Generation of CDX2 and CDX tumor dissociationCDX2 tumors (previously described; ref. 30) were disaggre-

gated (gentleMACS dissociator/human tumor kit; Miltenyi Bio-tec) immediately after sacrifice of the mouse host when tumorsreached 400 mm3. Red blood cells were lysed (RBC lysis buffer;G-Biosciences), remaining cellswerefiltered (through70mmpores),and cell viability was checked (trypan blue exclusion). CDX2 cellswere maintained in the culture conditions as described above.

Drug concentration responses and drug treatmentsH1048 and DMS114 cells were seeded at 7,300 cells/well,

H526 cells at 14,000 cells/well, H146 cells at 15,000 cells/well,and disaggregated CDX2 cells at 5,000 cells/well (96-well plates).Twenty-four hours after seeding, cells were treated and culturedfor 72 hours. H1048 and DMS114 adherent cells were stainedwith sulforhodamine B (SRB), whereas H146, H526, and CDX2cell suspensions were assessed by resazurin assay (previouslydescribed; ref. 31). To determine log GI50, log drug concentrationwas plotted against absorbance for nonlinear curve fit analysis(Prism, GraphPad Software). For display purposes, drug concen-tration was plotted versus normalized absorbance. Statisticalanalysis was carried out on three independent log GI50 readingsand transformed to [GI50].

To assess BCL-2 family expression, PI3K or apoptotic pharma-codynamic biomarkers after drug treatment, cells were plated at106 cells/well (6-well plate), and 24 hours later, cells were treatedwith [GI50] for 4 or 24 hours.

Measurement of mitochondrial cytochrome c releaseH526, H1048, and DMS114 cells (106) were pelleted and

washed in PBS after drug treatment(s). Cells were resuspendedand incubated inDTEBbuffer [135mmol/L trehalose, 10mmol/LHEPES, 20 mmol/L EDTA, 20 mmol/L EGTA, 5 mmol/L succinateacid, 0.1 % BSA, and 50 mmol/L KCl in sterile ddH2O at pH 7.5(KOH)] and 0.002% digitonin for 15 minutes at RT. Cells werefixed with formaldehyde (1% final concentration) at RT for 15minutes in the dark and then diluted with 2:1 with neutralizingbuffer (1.7 mol/L Tris and1.25 mol/L glycine at pH 9.1) at RT for15minutes in the dark. Fixed cells were stainedwith cytochrome cantibody (1:400, Alexa 488, Clone 6 h2.b4; BD Biosciences)diluted in staining buffer [1% saponin/10% bovine serum albu-mins (Sigma) in sterile ddH2O] overnight at 4�C and analyzednext day by flow cytometry (BD LSRFortessa analyzer) using Divasoftware (BD Biosciences).

In vivo efficacy of GDC-0941 and NavitoclaxIn vivo procedures were carried out in accordance with Home

Office Regulations (UK) and the UK Coordinating Committeeon Cancer Research guidelines using approved protocols (HomeOffice Project Licence no. 70/8252 reviewed by Cancer ResearchUK Manchester Institute Animal Welfare and Ethical ReviewBody). In vivo research was reported according to ARRIVE (Ani-mal Research Reporting of In Vivo Experiments) guidelines(2010). H1048 xenografts were grown by s.c. injection of 5 �106 cells in 0.2 mL of 1:1 RPMI:Matrigel (BD Biosciences) intothe mid-dorsal flank of 8-week-old female SCID-beige mice(C.B-17/IcrHsd-PrkdcscidLystbg-J; Harlan Laboratories). Tumorfragments of passage 4 CDX2 were implanted s.c into 8-week-old female SCID-beige mice. Six mice were housed together invented caging systems in a 12-hour light/12-hour dark environ-ment and maintained at uniform temperature and humidity.Mice were monitored biweekly for signs of tumor growth, andonce a palpable tumor formed, it was measured biweekly usingcallipers. Tumor volume was calculated as 0.5 � (longest mea-surement) � (shortest measurement)2. Seven days after implan-tation, mice bearing H1048 xenografts of 150 to 250 mm3 wererandomized into 4 groups of 10 mice. Eleven weeks afterimplantation of CDX2 fragments, mice bearing tumors of 150to 250 mm3 were randomized into 4 groups of 6 mice. Treat-ment groups for both H1048 xenograft and CDX2-bearing mice

PI3K/BMX Combination with BH3 Mimetic in SCLC

www.aacrjournals.org Mol Cancer Ther; 15(6) June 2016 1249




were: vehicle, 75 mg/kg GDC-0941, 100 mg/kg Navitoclax, 75mg/kg GDC-0941, and 1 hour later 100 mg/kg Navitoclax.Treatments were administered by oral gavage for 21 days. Tumormeasurements were continued 3x a week until tumor reachedfour times initial tumor volume (4xITV) or the mouse had beenon study for 6 months.

RNA interferencesiRNA SMARTpools or individual oligos (Supplementary Table

S1; Thermo Scientific) were transfected into H1048 cells usingLipofectamine RNAiMAX (Life Technologies) according to themanufacturer's instructions. Cells were transfected in 6-wellplates, reseeded into appropriate culture vessels 24 hours later,and drug treated after 24 hours.

Western blottingWestern blotting was performed as previously described (32).

Primary antibodies used were as follows: rabbit anti-AKT, rabbitanti-pS473 AKT, rabbit anti-pT308 AKT, rabbit anti-BAX, rabbitanti-BIM, rabbit anti–cleaved caspase 3, rabbit anti-PARP, rabbitanti-S6, rabbit anti-pS235/236 S6 (Cell Signaling Technology),mouse anti–BCL-2 (Dako), rabbit anti–BCL-xL, mouse anti-BMX,mouse anti–MCL-1 (Becton Dickenson), rabbit anti-BAD (R&DSystems), mouse anti-BAK, mouse anti–a-tubulin (Merck),mouse anti-Actin, and rabbit anti-PUMA (Sigma).

Phospho-kinase arrayBMX knockdown and inhibition of the PI3K pathway (4-hour

Ibrutinib treatment) were determined byWestern blot using PI3Kpathway pharmacodynamic biomarkers: rabbit anti-pS473AKTand rabbit anti-pS235/236S6. Lysates from nontargeting RNAi,BMX RNAi, or Ibrutinib-treated cells were then applied to aphospho-kinase array according to the manufacturer's instruc-tions (# ARY003B; R&D System).

Statistical analysisStatistical analysis of significant differences for in vitro studies

comparing treated and control groups was performed usingunpaired, two-tailed t tests. Mouse survival rates were assessedusing the Kaplan–Meier analysis; the log-rank test was used tocompare survival distributions. One-way ANOVA multiple com-parisons analysis was used to distinguishmultiple drug treatmentgroup effects on tumor-doubling time using GraphPad Prismsoftware. P < 0.05 was considered statistically significant.

ResultsBMX expression in SCLC

Inhibition of PI3K/BMXpathway signaling sensitizes colorectalcancer cells to ABT-737 (27). We reasoned that other cancer celltypes that express high levels of BMX may also employ thissurvival pathway, and combined PI3K inhibitor and BH3mimetics would prove beneficial. BMX mRNA expression data(�850 cancer cell lines) were obtained from the Broad Institute'sCancer Cell Line Encyclopedia. BMXmRNA was expressed at lowlevels in themajority of cancer cell lines, although for some cancertypes, specific cell lines had particularly high BMX expression.Most notable was SCLC with the top three BMX-expressing celllines (H211, H526, and CORL311; Supplementary Fig. S1A). Weexamined BMX protein expression across a panel of 9 SCLC celllines; five expressed BMX (H526, H1048, DMS114, H82, and

DMS79; Supplementary Fig. S1B). The combination of PI3K/BMXinhibitors with ABT-737 was assessed in H526, H1048, andDMS114 (BMX expressing) or H146 and H524 (BMX notexpressed). The H1048 SCLC cell line harbors an oncogenicactivating mutation in PIK3CA, the remainder are PIK3CAwild-type (33).

PI3K/BMX pathway inhibition sensitized SCLC cell lines toABT-737

PI-103, a dual PI3K/mTOR inhibitor, was used to inhibit PI3Ksignaling (34). The TEC kinase inhibitor Ibrutinib (PCI-32765;ref. 35) was used to inhibit BMX (similar potency against Bruton'sTyrosine Kinase, its primary target). PI-103 and Ibrutinib wereassessed in combination with ABT-737. All 4 SCLC cell lines wereresponsive to PI-103 in either the SRB assay (adherent cells),which measures cellular biomass, or the resazurin assay (suspen-sion cells), which measures metabolic activity. PI-103 [GI50] was175 nmol/L in H1048, 300 nmol/L in H526, 330 nmol/L inDMS114, 30 nmol/L in H146, and 449 nmol/L in H524 cells(Supplementary Fig. S2). Dephosphorylation of AKT and S6 wasobserved in response to PI-103 [GI50], confirming PI3K pathwayinhibition in H1048, H526, and DMS114 but not in H146 cellswhich expressed less phospho-AKT but more phospho-S6 (com-pared with untreated) indicating incomplete PI3K pathway inhi-bition (Fig. 1A).

SCLC cell lines were less sensitive to Ibrutinib; [GI50] was 4mmol/L in H1048, 15 mmol/L in H526, 12 mmol/L in DMS114, 8mmol/L in H146, and 12 mmol/L in H524 cells (SupplementaryFig. S2). Ibrutinib treatment inhibited phosphorylation of AKTand S6 in H1048 and H526, but had a similar effect to PI-103 inH146 cells (Fig. 1A). Ibrutinib had no effect on phospho-AKT andphospho-S6 levels compared with untreated DMS114 cells(Fig. 1A).

To determine whether PI3K/BMX pathway inhibition affectedsensitivity to ABT-737, H1048, H526, DMS114, H146, and H524cells were treated concomitantly with PI-103 or Ibrutinib and/orABT-737. All cell lines exhibited a concentration-dependentresponse to ABT-737 alone (Fig. 1B; Supplementary Fig. S3)consistent with previous studies (21, 22). ABT-737 [GI50] wasreduced significantly in a concentration-dependentmanner by PI-103 or Ibrutinib in H1048 and H526 but not in DMS114 andH146 cells (Fig. 1B; Supplementary Fig. S4), revealing that sen-sitization to ABT-737 to PI3K pathway inhibition was cell contextdependent. PI3K pathway inhibition sensitized to ABT-737 irre-spective of PIK3CA mutation status confirming findings in colo-rectal cancer cells (27).

PI3K pathway inhibition enhanced ABT-737–inducedapoptosis

We sought to confirm that increased sensitivity to ABT-737 inH1048 andH526 cells treated with PI-103 or Ibrutinib was due toincreased apoptosis. DMS114 cells (which showed no combina-tion response below 10 mmol/L) were included as a negativecontrol. The effect of PI-103 only, Ibrutinib only, ABT-737 only,PI-103 plus ABT-737, or Ibrutinib plus ABT-737 on cytochrome crelease was determined by flow cytometry (Fig. 2A). As singleagents, neither PI-103 or Ibrutinib caused significant mitochon-drial cytochrome c release in H1048 and H526 compared withuntreated cells; ABT-737 released cytochrome c in 20% of H1048(P ¼ 0.011) and 11% of H526 cells (P ¼ 0.0007). ABT-737–induced apoptosis increased significantly when combined with

Potter et al.

Mol Cancer Ther; 15(6) June 2016 Molecular Cancer Therapeutics1250




either PI-103 or Ibrutinib compared with ABT-737 alone inH1048 and H526 cells (Fig. 2A). Cytochrome c release was notdetected in DMS114 cells with any single agent or combination

(Fig. 2A). PI-103 and Ibrutinib treatment alone did not increasecellular levels of cleaved caspase 3 or cleaved PARP in H1048 andH526 cells. ABT-737 alone did increase these apoptosis

Figure 1.PI3K/BMX pathway inhibitionsensitized H1048 and H526 cells toABT-737. A, Western blot ofpharmacodynamic biomarkers ofPI3K/BMX pathway inhibition after4-hour treatment with pathwayinhibitors. H1048 cells (0.1 mmol/LPI-103 or 4 mmol/L Ibrutinib), H526,DMS114 (0.25 mmol/L PI-103 or8 mmol/L Ibrutinib), H146 cells(0.025 mmol/L PI-103 or 8 mmol/LIbrutinib), or DMSO control. Data arerepresentative of three independentexperiments. B, H1048, H526, DMS114,and H146 cells were treated with PI-103, Ibrutinib, or DMSO control andABT-737 for 3 days. H1048 andDMS114 cells were assessed by SRBassay, and H526 and H146 wereassessed by resazurin assay.Absorbance (SRB, 540 nm) orfluorescence (resazurin) relative tountreated (UnT) cells was determinedrelative to PI-103, Ibrutinib, or DMSOonly–treated cells as appropriate forindividual concentration responsecurves. Data represent mean,n ¼ 3 independent experimentsin triplicate � SEM.






biomarkers that were increased further when ABT-737 was com-binedwith either PI-103or Ibrutinib (Fig. 2B).No cleaved caspase3 or cleaved PARP was observed with any treatment of DMS114(Fig. 2B). These data suggest that the increased ABT-737 sensitivitycaused by PI-103 or Ibrutinib treatment in H1048 andH526 cellswas due to increased apoptosis.

PI-103 sensitizes to ABT-737 independent of MCL-1The PI3K pathway regulates multiple BCL-2 family members,

including BIM (36), BAD (37), and MCL-1 (38), and we hypoth-esized that changes in expression of these proteins could contrib-ute to ABT-737 sensitization. The effect of PI-103, Ibrutinib, MK-2206 (allosteric AKT inhibitor), and KU-0063794 (ATP compet-itive mTORC1/2 inhibitor) treatment on expression levels of 8BCL-2 family members, including BIM, BAD, and MCL-1, wasassessed after 24 hours of treatment in H1048, H526, andDMS114 cells. No change in expression of any of these BCL-2

familymembers after treatmentwithPI3Kpathway inhibitorswasseen (Supplementary Fig. S5A).

MCL-1 is an established resistance factor of ABT-737 andNavitoclax efficacy (22, 23) and is downregulated by mTORsignaling leading to Navitoclax sensitization (39). AlthoughPI3K pathway inhibitors did not downregulate MCL-1 inH1048 cells (Fig. 2C), the mTOCR1/2 inhibitor AZD8055 candownregulate H1048 cell MCL-1 (39). We investigated whetherPI-103 treatment could further sensitize to ABT-737 whenMCL-1 was reduced by RNAi cells as this would imply inhibi-tion of PI3K signaling can sensitize to ABT-737 independentlyof MCL-1. As expected, MCL-1 RNAi sensitized H1048 cells toABT-737. However, PI-103 caused a 4.1- and 9.1-fold sensiti-zation to ABT-737 in control cells and MCL-1 RNAi cells,respectively (Fig. 2D and E and Supplementary Fig. S6A),suggesting a MCL-1–independent component to this furtherincrease in sensitivity.

Figure 2.PI3K/BMX inhibition increasesABT-737–induced apoptosis insensitive SCLC cell lines. H1048 cells(0.1mmol/LPI-103, 4mmol/L Ibrutinib,2mmol/LABT-737), H526 andDMS114(0.25 mmol/L PI-103, 8 mmol/LIbrutinib, 4 mmol/L ABT-737) cellswere treated with single agents orABT-737 in combination withPI-103 or Ibrutinib for 4 hours. A,cytochrome c release as a measure ofapoptosis. Data are mean, n ¼ 3independent experiments � SEM.�� , P < 0.01; and �� , P < 0.001,according to two-tailed unpaired ttest. B, Western blot forlevels of cleaved caspase 3and cleaved PARP. Data arerepresentative of three independentexperiments. C, Western blot of MCL-1 and a-tubulin (loading control) forH1048 cells treated with 0.1 mmol/LPI-103, 4 mmol/L Ibrutinib, 0.5 mmol/LMK-2206, and 0.25 mmol/L KU-0063794 or DMSO equivalent for 24hours. D,Western blot of MCL-1 levelsin H1048 cells transfected withSMARTpool siRNA targeting MCL-1mRNA or nontargeting control 48hours after transfection. Data arerepresentative of three independentexperiments. E, effect at 3 days of asingle agent or combination onH1048cells treated 48 hours aftertransfection to knock down MCL-1.Cells were assessed as in Fig. 1B.Mean, n ¼ 3 independentexperiments in triplicate � SEM.

Potter et al.





PI-103 does not further sensitize BMX RNAi cells to ABT-737To determine whether BMX inhibition sensitized to ABT-737

downstream of PI3K, the effect of BMX RNAi combined with PI-103 treatment onABT-737 sensitivity was assessed inH1048 cells.Cells were transfected with SMARTpool siRNA targeting BMXmRNA or individual siRNA oligos of the SMARTpool. BMXRNAi cells were treated with ABT-737 (Fig. 3A), and BMX knock-down confirmed by Western blot (Fig. 3B). SMARTpool siRNAand oligos 1 and 3 caused a significant ABT-737 sensitization

(Supplementary Fig. S6B; P ¼ 0.033, 0.006, and 0.037, respec-tively) and gave the greatest degree of BMX knockdown (consis-tent with previous observations in colorectal cancer cells; ref. 27).If BMX acts downstream of PI3K, then BMX RNAi should notsignificantly further sensitize PI-103–treated cells to ABT-737.H1048 cells were transfected with either siRNA-targetingBMX or nontargeting control siRNA and treated concomitantlywith PI-103 and/or ABT-737 (Fig. 3C). As expected, BMX RNAiand PI-103 treatments alone significantly sensitized to ABT-737.

Figure 3.BMX RNAi does not further sensitizePI-103–treated H1048 cells to ABT-737. A–C, cells were transfected withSMARTpool siRNA or individual siRNAoligos targeting BMX mRNA ornontargeting control (NT). A, forty-eight hours after transfection, cellswere treated with ABT-737 for 3 days.Cells were processed as in Fig. 1B.Mean, n¼ 3 independent experimentsin triplicate � SEM. B, Western blot ofBMX levels 48 hours after transfectionin BMX and NT RNAi cells. Data arerepresentative of three independentexperiments. C and D, forty-eighthours after transfection, cells weretreated with 0.1 mmol/L PI-103 orDMSO equivalent and the indicatedconcentration of ABT-737 for 3 days.Cells were processed as in Fig. 1B. C,data are mean, n ¼ 3 independentexperiments in triplicate � SEM.D, ABT-737 GI50 values � SEM.� , P < 0.05 according to two-tailedunpaired t test; NS, not significant.E, phospho-antibody array showingalterations in signaling pathways inBMX RNAi (BMX), nontargeting RNAicontrol, and Ibrutinib-treated cellshighlighting seven keyphosphorylation sites. Eachantibody in the phospho-antibodyarray was carried out in duplicatespots. F, semiquantitative analysis ofphospho-antibody array data in E forthe seven indicated phosphorylationsites.






PI-103–treated cells were not significantly further sensitized toABT-737 by knockdown of BMX (Fig. 3D and Supplementary Fig.S6C). These data suggest that BMX acts downstream of PI3K in aprosurvival signaling pathway. A phospho-kinase array carriedout in BMX RNAi cells or Ibrutinib-treated cells confirmed thatBMX regulates AKT/mTOR pathway signaling inH1048 cells (Fig.3E and F). This array also suggested that BMX may regulateSTAT3 in these cells (Fig. 3E and F) consistent with previousresearch (40).

ABT-737 sensitization by PI3K pathway inhibition is AKT andmTOR dependent in SCLC

We next investigated AKT/mTOR pathway signaling to deter-minewhether inhibition sensitizes to ABT-737 in SCLC. Cell linesshowedmixed sensitivity to the allosteric AKT inhibitorMK-2206,but were all responsive to the ATP competitive mTORC1/2inhibitor KU-0063794 (Supplementary Fig. S2). The [GI50] ofMK-2206 was 500 nmol/L in H1048, 250 nmol/L in H526, 8.8mmol/L inDMS114, 12.8 mmol/L inH146, and 7 mmol/L inH524cells. The [GI50] of KU-0063794 was 200 nmol/L in H1048, 400nmol/L in H526, 400 nmol/L in DMS114, 350 nmol/L in H146,and 610 nmol/L in H524 cells. A decrease in AKT and S6phosphorylation confirmed inhibition of the PI3K pathway withMK-2206 and KU-0063794 treatment in H1048, H526, andDMS114 cells (Fig. 4A; Supplementary Fig. S2). KU-0063794treatment inhibited the PI3K pathway in H146 cells, but MK-2206 treatment decreased phospho-AKT and increased phospho-S6 (Fig. 4A) consistent with what was observed with PI-103 andIbrutinib inhibitors (Fig. 1A).

All cell lines were treated concomitantly with MK-2206 and/orABT-737, or with KU-0063794 and/or ABT-737. InH1048,H526,andH524but not inDMS114 andH146 cells, ABT-737 [GI50]wasreduced significantly in a concentration-dependent manner byMK-2206 or KU-0063794 (Fig. 4B; Supplementary Figs. S3 andS4). These data suggest that PI3K pathway inhibition can sensitizeH1048 and H526 cells to ABT-737 via both BMX (Figs. 1B, 2A–C,and 3A–C) and AKT/mTOR-dependent (Fig. 4B) mechanisms,and this is cell context dependent.

Effect of combined PI3K inhibition and BH3 mimetic in vivoPI-103 andABT-737 arenot orally bioavailable soGDC-0941, a

clinically relevant class I PI3K inhibitor (currently in clinical trials;clinicaltrials.gov), andNavitoclax (used for phase II trials in SCLC;ref. 24) were used for in vivo studies. The combination was welltolerated in SCID-beige mice with an average body weightdecrease of 5.3% after 21 days of treatment which rapidly recov-ered after treatment ceased (Supplementary Fig. S7). The combi-nation efficacy was assessed in mice bearing H1048 xenografts.Mice bearing H1048 xenograft reached 4xITV in 9 and 14 days forvehicle and single-agent GDC-0941 or Navitoclax, respectively.When GDC-0941 and Navitoclax were combined, there was asignificant increase in days taken for tumor growth to reach thespecified end point (32 days) compared with either drug alone orthe vehicle group (Vehicle, P¼0.0001; GDC-0941, P¼0.001;Navitoclax, P ¼ 0.001 according to log-rank test; Fig. 5A–C). Thecombination significantly increased tumor-doubling time com-pared with vehicle and either drug alone (Fig. 5D). Navitoclaxalone caused increased level of apoptosis compared with vehiclecontrol after 24 hours, which was further enhanced by combiningwith GDC-0941 (Fig. 5E and F). GDC-0941, alone or in combi-nation with Navitoclax, caused downregulation of phospho-S6

after 4 hours of treatment, consistent with its primary pharma-cology (Fig. 5E and G).

The clinical relevance of long-established SCLC cell lines isopen to question (41), and SCLC patient–derived xenografts(PDX) have been demonstrated to better reflect the underwhelm-ing clinical results seen with single-agent Navitoclax (42). Toexpand our results to amodel thatmore accurately effects patientstumors, we assessed efficacy of GDC-0941 and Navitoclax in aSCLC CDX2 derived from circulating tumor cells (CTC) enrichedprior to the donor patient's chemotherapy (30). This is a highhurdle for drug development as SCLC cells that derived themodelhad already invaded tissue, were disseminating, and were tumor-igenic in the mouse. CDX mouse models mirror the subsequentchemotherapy response of the donor patient (30). CDX2 wasderived from a chemorefractory patient whose disease progressedthroughout treatment. Treatment of CDX2 with cisplatin/etopo-side delayed tumor growth by only 19 days (30).

Initial ex vivo culture experiments suggested that disaggregatedCDX2 cells were sensitive to ABT-737 and that sensitivity wasenhanced by cotreatment with PI-103 (Fig. 6A). The efficacy ofGDC-0941andNavitoclax aloneand in combinationwas assessedin mice bearing CDX2 tumors. The combination caused thegreatest effect comparedwith either drug alone or vehicle (Vehicle,P ¼ 0.0002; GDC-0941, P ¼ 0.0003; Navitoclax, P ¼ 0.033according to log-rank test; Fig. 6B–D). In the combination group,no tumors reached 4xITV over the experimental time course, andmeanmaximum tumor regressionwas 94%.Navitoclax alone hada significant effect on tumor growth compared with vehicle andGDC-0941 (Vehicle, P¼ 0.0005; GDC-0941, P¼ 0.010 accordingto log-rank test; Fig. 6D), although only one tumor failed to reach4xITV over the experimental time course. Mean maximum tumorregression was 80%, but tumors began to regrow as soon as theNavitoclax treatment was ceased, whereas in the combinationgroup, tumors began to regrow only approximately 30 days aftertreatment cessation (Fig. 6B), and one mouse in the combinationgroup had no detectable tumor at end of the experiment (Fig. 6C).GDC-0941 caused a significant decrease in phospho-S6 at 4 hours,and the combination caused more apoptosis at 24 hours aftertreatment (Fig. 6E–G). These data suggest that combined GDC-0941 and Navitoclax is more efficacious in vivo than either drugalone and could prove a beneficial rational combination in SCLCpatients, including chemorefractory patients.

DiscussionImproved outcomes for SCLC patients remain elusive, despite

numerous clinical trials of targeted therapeutics. New approachesthat extend progression-free survival after initially successful che-motherapy and alternatives to chemotherapy in chemorefractorySCLC are urgently required. We demonstrate that inhibition ofPI3K pathway signaling can increase the sensitivity of SCLC cells(H1048 and H526) to a BCL-2/BCLxL targeting BH3 mimetic viaincreased apoptosis. Our data in SCLC cell lines in vitro wererecapitulated in vivo in theH1048 xenograft and a newly developedCDX from a chemorefractory patient (43). These data provideadditional weight to an increasing body of evidence to suggestSCLC clinical trials of combined PI3K/mTOR pathway inhibitorsand BH3 mimetics are warranted. There appear to be, however,multiple PI3K downstream effector signaling networks and mod-ulation of different BCL-2 family proteins that affect PI3K/mTORinhibitor/BH3 mimetic synergy that are cell context dependent.

Potter et al.





In our study, increased sensitivity to combined PI3K inhibitionand BH3 mimetic was dependent on BMX, AKT, and mTORdownstream of PI3K. Our data suggest for the first time that BMXcan regulate AKT/mTOR pathway activity in SCLC. Both inhibi-tion of BMX using Ibrutinib (Fig. 1B and Supplementary Fig. S4B)

and reduction of BMX level byRNAi (Fig. 3E and F SupplementaryFig. S4B) reduced activation of pathways downstream of BMX,including AKT and mTOR, consistent with on target effects. Themechanism by which BMX regulates AKT/mTOR signaling inSCLC is not yet clear, but precedents that link BMX to AKT/mTOR

Figure 4.AKT/mTOR pathway inhibitionsensitized H1048 and H526 cells toABT-737. A, Western blot forpharmacodynamic biomarkers of PI3Kpathway signaling in H1048 cells(treated with to 0.5 mmol/L MK-2206and 0.25 mmol/L KU-0063794), H526cells (treated with to 0.25 mmol/LMK-2206 and 0.4 mmol/L KU-0063794), DMS114 cells (treated with8 mmol/L MK-2206 and 0.4 mmol/L KU-0063794), and H146 cells (treated with8 mmol/L MK-2206 and 0.4 mmol/L KU-0063794) or DMSO control for 4 hours.Data are representative of threeindependent experiments. B, H1048,H526, DMS114, and H146 cells wereexposed to the indicated concentrationof either MK-2206 and KU-0063794 orDMSO equivalent and the indicatedconcentration of ABT-737 for 3 days.Cells were processed as in Fig. 1B. Dataare mean, n ¼ 3 independentexperiments in triplicate � SEM.






signaling exist in other cell types (43, 44). Addressing the mech-anistic role for BMX in regulating this signaling axis may aid indevelopment of predictive biomarkers as well as unveil a newtherapeutic opportunity for the treatment of SCLC.

We observed a significant increase in apoptosis 4 hours aftercombined drug treatment in vitro, suggesting that drug synergy

most likely involves posttranslational modification(s) due toPI3K inhibition. BMX was identified as the tyrosine kinase thatphosphorylates BAK, keeping it in the inactive conformation(45). BMX inactivation by PI3K/BMX inhibitors could thereforelead to BAK dephosphorylation in SCLC. In this theoreticalmodel, addition of ABT-737 by antagonism of BCL-2, BCL-xL,

Figure 5.Effect of GDC-0941 and Navitoclax incombination in vivo in the H1048xenograft model. A, mean relativetumor growth (10 animals pertreatment group) up to the day thefirst animal in that treatment groupreached the predefined 4 � initialtumor volume (ITV) endpoint. B,relative tumor volumes for individualanimals within each treatment group.C, Kaplan–Meier survival curve withendpoint of 4 � ITV. GDC-0941/Navitoclax combination wassignificant compared with all othertreatment groups, P < 0.0001calculated by log-rank test. D, tumor-doubling time for vehicle, GDC-0941,Navitoclax, and GDC-0941/Navitoclaxgroups. GDC-0941/Navitoclaxcombination tumor-doubling timewas significant compared withall other treatment groups;�� , P < 0.0001 according to one-wayANOVA multiple comparisons.E–G, pharmacodynamic biomarkers.Twenty SCID-beige mice wereimplantedwith 5� 106H1048 cells andrandomized into 8 groups of 5 micewhen tumors size reached between150 and 250 mm3. Group 1 and 5vehicle only, group 2 and 6 GDC-0941only, group 3 and 7 Navitoclax only,and group 4 and 8 GDC-0941, and 1hour later Navitoclax. Groups 1 to 4were culled 4 hours after dosing, andgroups 5 to 8 were culled 24 hoursafter dosing. E, tumors were stainedfor either cleaved caspase 3 (CC3, 24hours) or phosphoS235/236-S6 (pS6,4 hours). Scale bar, 50 mm. F and G,semiquantitative analysis of IHC forCC3 and pS6, respectively. � , P < 0.05;�� , P < 0.01; and �� , P < 0.001.

Potter et al.





or BCL-w would release BIM to activate available dephosphory-lated BAK, resulting in BAK homo-oligomerization, MOMP,cytochrome c release, and apoptosis, and this warrants furtherinvestigation.

Depending on cellular context, upregulation of PI3K signal-ing may allow cells to sustain proliferation without growthfactors and resist cell death, two hallmarks of cancer (46, 47).

Of the two cell lines which were sensitized to ABT-737 by PI3Kor BMX inhibition, only H1048 has a known activatingPI3K pathway mutation (although we cannot rule out thatH526 cells have activated PI3K signaling via a mechanism yetto be determined). Although CDX2 does not harbor knownPI3K pathway mutations, one copy of chromosome 10 hasbeen lost (30), which could decrease PTEN expression and

Figure 6.Effect of GDC-0941 and Navitoclaxin combination in CDX2 derivedfrom CTCs from a chemorefractorySCLC patient. A, CDX2 tumors weredisaggregated and 24 hours latertreated with indicatedconcentration of drug for 3 days.Cells were analyzed as in Fig. 1B.Data are mean, n ¼ 3 independentexperiments � SEM. B, CDX2 meanrelative tumor growth curves. Allgroups were treated every day for21 days (orange arrow indicates lastday of dosing), and the predefinedendpoint was 4� initial tumorvolume (4� ITV) or when themousehad been on study for 6 months. Noanimals reached 4� ITV in the GDC-0941/Navitoclax combinationgroup. C, relative tumor volumes forindividual animal within eachtreatment group. D, Kaplan–Meiersurvival curve for predefinedendpoint of 4x ITV. GDC-0941/Navitoclax combination wassignificantly comparedwith all othertreatment groups (Vehicle, P <0.0002; GDC-0941, P <0.0003;Navitoclax, P < 0.033 according tolog-rank test). E, pharmacodynamicbiomarkers. Animals were culled4 or 24 hours after administration ofa single dose of indicatedcompounds. Tumors were stainedfor either cleaved caspase 3 (CC3)or phosphoS235/236-S6 (pS6).Scale bar, 50 mm. F and G,semiquantitative analysis of IHC forCC3 and pS6, respectively. � , P <0.05; �� , P < 0.01; and ��, P < 0.001.






increase PI3K signaling (Supplementary Fig. S8 shows CDX2expression of BMX and PI3K/AKT pathway components). BMXis important in mutant PIK3CA-induced transformation viaphosphorylation (on Y705) and activation of STAT3 causinghomo- and hetero-dimerization to activate STAT3 transcrip-tional regulator function (40). Consistent with this function ofBMX, we observed decreased phospho-STAT3 (Y705 and S727)in BMX-depleted and Ibrutinib-treated H1048 cells (Fig. 3E andF). mTOR can phosphorylate and activate STAT3 (on S727;ref. 48). One hypothesis therefore is that PI3K/BMX/mTORinhibition sensitizes to ABT-737 via repression of STAT3. STAT3is a key regulator of BCL-xL expression, a key target of ABT-737(49). However, no treatment effects on BCL-xL expression werenoted by 24 hours (Supplementary Fig. S4A).

PI3K pathway inhibition-mediated sensitization to ABT-737has been shown in other cancer types, including our study incolorectal cancer cell lines, which showed this occurs via aPI3K/BMX pathway–dependent but AKT/mTOR-independentroute (27). PI3K/AKT/mTOR pathway–dependent sensitizationto ABT-737/Navitoclax has been seen in non–small cell lungcancer (50) and lymphoma (51) cell lines. Two recent studiesdemonstrate that PI3K/mTOR pathway inhibition sensitizes toBH3 mimetics in SCLC, consistent with our observations. Thereare both similarities and differences in the mechanism(s)underpinning the combination synergies observed between thetwo prior studies and the results we present here. Gardner andcolleagues showed that mTOCR1 inhibitor rapamycinincreased efficacy of ABT-737 in H146 cells and in several PDXmodels (52). In the Gardner and colleagues (15) study, rapa-mycin prevented ABT-737–mediated downregulation of BAXresulting in increased BAX-dependent apoptosis. BAX proteinlevels were not decreased in H1048, H526, or DMS114 cellsafter treatment with PI3K pathway inhibitors (PI-103, MK-2206, or KU-0063794) in our study (Supplementary Fig.S4A), consistent with observations of Faber and colleagues inH1048 cells and in a PDX model using the TORC1/2 inhibitorAZD8055 (39). Moreover, we saw no synergy in H146 cellstreated with ABT-737 plus any of the PI3K pathway inhibitorswe examined. In our hands, PI-103, Ibrutinib, and MK-2206resulted in mTOR pathway activation (increased pS6) in H146.

Faber and colleagues report a different mechanism to explainhow AZD8055 sensitizes to Navitoclax in SCLC cell lines, xeno-grafts, a PDX, and a genetically engineered model (14). Theydemonstrated in H1048 cells that AZD8055 sensitizes to Navi-toclax, consistent with our finding for all PI3K pathway inhibitorsin the same cell line (Figs. 1B and 4B). However, they showed thatsensitization was via MCL-1 downregulation, which was notobserved in H1048 cells or any cell line used in our study (Fig.2C and Supplementary Fig. S4A), or in the cell line panel and PDXstudied by Gardner and colleagues (15), Faber and colleaguesaddress this issue suggesting that as an mTORC1/2 ATP compet-itive inhibitor, AZD8055 is uniquely capable of downregulatingMCL-1, unlike rapamycin (14). This explanation does not recon-cile our data using KU-0063794 which is also anmTORC1/2 ATPcompetitive inhibitor, but did not decrease MCL-1 levels inH1048 cells (at the concentration and time-points studied). Weshow that enforced downregulation of MCL-1 with RNAi doesindeed sensitize to ABT-737 in H1048 cells, but that sensitivitywas significantly further increased byPI-103, suggesting aMCL-1–independent element for this effect (Fig. 2D and E and Supple-mentary Fig. S5A).

Approximately 20% SCLC patients are classified as chemor-efractory, where disease progression occurs within 3 months ofceasing chemotherapy (5, 6). We assessed combined GDC-0941 and Navitoclax in CDX2 derived from CTCs enriched atbaseline from a patient subsequently shown to be chemore-fractory to set a 'high hurdle' for this combination. We previ-ously showed that cisplatin (5 mg/kg day 1) plus etoposide (8mg/kg, days 1, 2, 3) administered to mice bearing CDX2 tumors(at 200–250 mm3) regressed to approximately 50%, but regrewwithin the subsequent 10 days at a rate not dissimilar tovehicle-treated controls, mirroring the response and rapidrelapse of the patient. Here, we show, dosing CDX2 bearingmice up to 21 days, that GDC-0941 alone slowed CDX2 growthand Navitoclax alone caused tumor regression during thedosing period, but with immediate regrowth after treatmentceased. However, the combination gave a durable regression>30 days after treatment. These data from a chemorefractorymodel of SCLC provide some optimism that the PI3K pathway/BH3 mimetic combination may have impact in the mostresistant cases of SCLC.

Our study with that of Gardner and colleagues (52) and Faberand colleagues (39) suggest that there are different mechanismsdownstream of PI3K signaling that suppress BH3 mimetic–induced apoptosis; we introduce a new component, BMX, thatacts downstream of PI3K and upstream of Bcl-2 family gate-keepers of apoptosis. Overall, despite cell line and pathwayinhibitor–dependent differential mechanisms, the efficacy ofGDC-0941 and Navitoclax in a CDX model derived from achemorefractory patient highlights the potential clinical utilityof this combination and strongly supports initiation of a clinicaltrial with a BH3 mimetic and a PI3K pathway inhibitor in SCLCpatients.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: D.S. Potter, C.J. Morrow, C. DiveDevelopment of methodology: D.S. Potter, C.L. HodgkinsonAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): D.S. Potter, M. Galvin, S. Brown, A. Lallo,C.L. Hodgkinson, F. BlackhallAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): D.S. Potter, A. Lallo, C.J. Morrow, C. DiveWriting, review, and/or revision of the manuscript: D.S. Potter, M. Galvin,F. Blackhall, C.J. Morrow, C. DiveAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): D.S. Potter, C.L. HodgkinsonStudy supervision: D.S. Potter, C.L. Hodgkinson, C.J. Morrow, C. Dive

AcknowledgmentsThe authors thank John Brognard, Kristopher Frese, and Stuart Williamson

for constructive advice.

Grant SupportC. Dive received CRUK core funding (C5759/A20971).The costs of publication of this article were defrayed in part by the

payment 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 30, 2015; revised February 24, 2016; accepted March 10,2016; published OnlineFirst March 29, 2016.


Potter et al.




References1. American Cancer Society. Lung cancer (small cell). 2016[cited 2016 Mar

15]. Available from: http://www.cancer.org/2. Arriola E, Canadas I, ArumiM, Rojo F, Rovira A, Albanell J. Genetic changes

in small cell lung carcinoma. Clin Transl Oncol 2008;10:189–97.3. Hansen HH, Dombernowsky P, Hirsch FR. Staging procedures and prog-

nostic features in small cell anaplastic bronchogenic carcinoma. SeminOncol 1978;5:280–7.

4. Kato Y, Ferguson TB, Bennett DE, Burford TH. Oat cell carcinoma of thelung. A review of 138 cases. Cancer 1969;23:517–24.

5. Lally BE, Urbanic JJ, Blackstock AW, Miller AA, Perry MC. Small cell lungcancer: Have we made any progress over the last 25 years? Oncologist2007;12:1096–104.

6. Huisman C, Postmus PE, Giaccone G, Smit EF. Second-line chemother-apy and its evaluation in small cell lung cancer. Cancer Treat Rev1999;25:199–206.

7. Fujii M, Hotta K, Takigawa N, Hisamoto A, Ichihara E, Tabata M, et al.Influence of the timing of tumor regression after the initiation of chemor-adiotherapy on prognosis in patients with limited-disease small-cell lungcancer achieving objective response. Lung Cancer 2012;78:107–11.

8. De Ruysscher D, Pijls-Johannesma M, Bentzen SM, Minken A, Wanders R,Lutgens L, et al. Time between thefirst day of chemotherapy and the last dayof chest radiation is the most important predictor of survival in limited-disease small-cell lung cancer. J Clin Oncol 2006;24:1057–63.

9. Arcaro A. Targeted therapies for small cell lung cancer: Where do we stand?Crit Rev Oncol Hematol 2015;95:154–64.

10. George J, Lim JS, Jang SJ, Cun Y, Ozretic L, Kong G, et al. Comprehensivegenomic profiles of small cell lung cancer. Nature 2015;524:47–53.

11. Ikegaki N, KatsumataM,Minna J, Tsujimoto Y. Expression of bcl-2 in smallcell lung carcinoma cells. Cancer Res 1994;54:6–8.

12. Jiang SX, Sato Y, Kuwao S, Kameya T. Expression of bcl-2 oncogene proteinis prevalent in small cell lung carcinomas. J Pathol 1995;177:135–8.

13. Umemura S, Mimaki S, Makinoshima H, Tada S, Ishii G, Ohmatsu H, et al.Therapeutic priority of the PI3K/AKT/mTOR pathway in small cell lungcancers as revealed by a comprehensive genomic analysis. J Thorac Oncol2014;9:1324–31.

14. Peifer M, Fernandez-Cuesta L, Sos ML, George J, Seidel D, Kasper LH, et al.Integrative genome analyses identify key somatic driver mutations ofsmall-cell lung cancer. Nat Genet 2012;44:1104–10.

15. Rudin CM, Durinck S, Stawiski EW, Poirier JT, Modrusan Z, Shames DS,et al. Comprehensive genomic analysis identifies SOX2 as a frequentlyamplified gene in small-cell lung cancer. Nat Genet 2012;44:1111–6.

16. Letai AG. Diagnosing and exploiting cancer's addiction to blocks inapoptosis. Nat Rev Cancer 2008;8:121–32.

17. Chen L,Willis SN,Wei A, Smith BJ, Fletcher JI, HindsMG, et al. Differentialtargeting of prosurvival Bcl-2 proteins by their BH3-only ligands allowscomplementary apoptotic function. Mol Cell 2005;17:393–403.

18. Chipuk JE, Moldoveanu T, Llambi F, Parsons MJ, Green DR. The BCL-2family reunion. Mol Cell 2010;37:299–310.

19. Chonghaile TN, Letai A. Mimicking the BH3 domain to kill cancer cells.Oncogene 2008;27Suppl 1:S149–57.

20. Kuwana T, Bouchier-Hayes L, Chipuk JE, BonzonC, Sullivan BA,GreenDR,et al. BH3 domains of BH3-only proteins differentially regulate Bax-mediated mitochondrial membrane permeabilization both directly andindirectly. Mol Cell 2005;17:525–35.

21. Oltersdorf T, Elmore SW, Shoemaker AR, Armstrong RC, Augeri DJ, BelliBA, et al. An inhibitor of Bcl-2 family proteins induces regression of solidtumours. Nature 2005;435:677–81.

22. Tahir SK, Yang X, AndersonMG,Morgan-Lappe SE, Sarthy AV, Chen J, et al.Influence of Bcl-2 family members on the cellular response of small-celllung cancer cell lines to ABT-737. Cancer Res 2007;67:1176–83.

23. Shoemaker AR, Mitten MJ, Adickes J, Ackler S, Refici M, Ferguson D, et al.Activity of the Bcl-2 family inhibitor ABT-263 in a panel of small cell lungcancer xenograft models. Clin Cancer Res 2008;14:3268–77.

24. Rudin CM, Hann CL, Garon EB, Ribeiro de Oliveira M, Bonomi PD,Camidge DR, et al. Phase II study of single-agent navitoclax (ABT-263)and biomarker correlates in patients with relapsed small cell lung cancer.Clin Cancer Res 2012;18:3163–9.

25. Engelman JA. TargetingPI3K signalling in cancer: opportunities, challengesand limitations. Nat Rev Cancer 2009;9:550–62.

26. Engelman JA, Luo J, Cantley LC. The evolution of phosphatidylinositol 3-kinases as regulators of growth and metabolism. Nat Rev Genet 2006;7:606–19.

27. PotterDS, Kelly P,DennenyO, Juvin V, Stephens LR,Dive C, et al. BMX actsdownstream of PI3K to promote colorectal cancer cell survival and path-way inhibition sensitizes to the BH3 mimetic ABT-737. Neoplasia2014;16:147–57.

28. Guo L, Chen P, Zhou Y, Sun Y. Non-receptor tyrosine kinase Etk is involvedin the apoptosis of small cell lung cancer cells. Exp Mol Pathol 2010;88:401–6.

29. Guo L, Zhou Y, Sun Y, Zhang F. Non-receptor tyrosine kinase Etkregulation of drug resistance in small-cell lung cancer. Eur J Cancer2010;46:636–41.

30. HodgkinsonCL,MorrowCJ, Li Y,Metcalf RL, Rothwell DG, Trapani F, et al.Tumorigenicity and genetic profiling of circulating tumor cells in small-celllung cancer. Nat Med 2014;20:897–903.

31. Harrison LR, Micha D, Brandenburg M, Simpson KL, Morrow CJ, DennenyO, et al. Hypoxic human cancer cells are sensitized to BH-3 mimetic-induced apoptosis via downregulation of the Bcl-2 protein Mcl-1. J ClinInvest 2011;121:1075–87.

32. Martin-Fernandez C, Bales J, Hodgkinson C, Welman A, WelhamMJ, DiveC, et al. Blocking phosphoinositide 3-kinase activity in colorectal cancercells reduces proliferation but does not increase apoptosis alone or incombination with cytotoxic drugs. Mol Cancer Res 2009;7:955–65.

33. Forbes SA, Beare D, Gunasekaran P, Leung K, Bindal N, Boutselakis H, et al.COSMIC: Exploring the world's knowledge of somatic mutations inhuman cancer. Nucleic Acids Res 2015;43:D805–11.

34. Hayakawa M, Kaizawa H, Kawaguchi K, Ishikawa N, Koizumi T, Ohishi T,et al. Synthesis and biological evaluation of imidazo[1,2-a]pyridine deri-vatives as novel PI3 kinase p110alpha inhibitors. Bioorg Med Chem2007;15:403–12.

35. Honigberg LA, Smith AM, Sirisawad M, Verner E, Loury D, Chang B, et al.The Bruton tyrosine kinase inhibitor PCI-32765 blocks B-cell activationand is efficacious inmodels of autoimmune disease and B-cellmalignancy.Proc Natl Acad Sci U S A 2010;107:13075–80.

36. Brunet A, Bonni A, ZigmondMJ, Lin MZ, Juo P, Hu LS, et al. Akt promotescell survival by phosphorylating and inhibiting a Forkhead transcriptionfactor. Cell 1999;96:857–68.

37. Datta SR, Dudek H, Tao X, Masters S, Fu H, Gotoh Y, et al. Akt phosphor-ylation of BAD couples survival signals to the cell-intrinsic death machin-ery. Cell 1997;91:231–41.

38. Letai A. Growth factor withdrawal and apoptosis: the middle game. MolCell 2006;21:728–30.

39. Faber AC, Farago AF, Costa C, Dastur A, Gomez-Caraballo M, Robbins R,et al. Assessment of ABT-263 activity across a cancer cell line collectionleads to a potent combination therapy for small-cell lung cancer. Proc NatlAcad Sci U S A 2015;112:E1288–96.

40. Hart JR, Liao L, Ueno L, Yates JR3rd, Vogt PK. Protein expression profiles ofC3H 10T1/2 murine fibroblasts and of isogenic cells transformed by theH1047R mutant of phosphoinositide 3-kinase (PI3K). Cell Cycle 2011;10:971–6.

41. Daniel VC,Marchionni L,Hierman JS, Rhodes JT,DevereuxWL, RudinCM,et al. A primary xenograft model of small-cell lung cancer reveals irrevers-ible changes in gene expression imposed by culture in vitro. Cancer Res2009;69:3364–73.

42. Hann CL,Daniel VC, Sugar EA, Dobromilskaya I,Murphy SC, Cope L, et al.Therapeutic efficacy of ABT-737, a selective inhibitor of BCL-2, in small celllung cancer. Cancer Res 2008;68:2321–8.

43. Zhang R, XuY, EkmanN,WuZ,Wu J, AlitaloK, et al. Etk/Bmx transactivatesvascular endothelial growth factor 2 and recruits phosphatidylinositol 3-kinase to mediate the tumor necrosis factor-induced angiogenic pathway.J Biol Chem 2003;278:51267–76.

44. Chau CH, Chen KY, Deng HT, Kim KJ, Hosoya K, Terasaki T, et al.Coordinating Etk/Bmx activation and VEGF upregulation to promote cellsurvival and proliferation. Oncogene 2002;21:8817–29.

45. Fox JL, Storey A. BMXnegatively regulates BAK function, thereby increasingapoptotic resistance to chemotherapeutic drugs. Cancer Res 2015;75:1345–55.

46. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000;100:57–70.






47. Hanahan D, Weinberg RA. Hallmarks of cancer: The next generation. Cell2011;144:646–74.

48. Yokogami K, Wakisaka S, Avruch J, Reeves SA. Serine phosphorylation andmaximal activation of STAT3 during CNTF signaling is mediated by therapamycin target mTOR. Curr Biol 2000;10:47–50.

49. LesinaM, KurkowskiMU, Ludes K, Rose-John S, TreiberM, Kloppel G, et al.Stat3/Socs3 activation by IL-6 transsignaling promotes progression ofpancreatic intraepithelial neoplasia and development of pancreatic cancer.Cancer Cell 2011;19:456–69.

50. Qian J, Zou Y, Rahman JS, Lu B, Massion PP. Synergy between phospha-tidylinositol 3-kinase/Akt pathway andBcl-xL in the control of apoptosis inadenocarcinoma cells of the lung. Mol Cancer Ther 2009;8:101–9.

51. Ackler S, Xiao Y, Mitten MJ, Foster K, Oleksijew A, Refici M, et al. ABT-263and rapamycin act cooperatively to kill lymphoma cells in vitro and in vivo.Mol Cancer Ther 2008;7:3265–74.

52. Gardner EE, ConnisN, Poirier JT, Cope L, Dobromilskaya I, Gallia GL, et al.Rapamycin rescues ABT-737 efficacy in small cell lung cancer. Cancer Res2014;74:2846–56.


Potter et al.




2016;15:1248-1260. Published OnlineFirst March 29, 2016.Mol Cancer Ther Danielle S. Potter, Melanie Galvin, Stewart Brown, et al. Mimetics in SCLCInhibition of PI3K/BMX Cell Survival Pathway Sensitizes to BH3

Updated version

10.1158/1535-7163.MCT-15-0885doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2016/03/26/1535-7163.MCT-15-0885.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/15/6/1248.full#ref-list-1

This article cites 51 articles, 16 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/15/6/1248.full#related-urls

This article has been cited by 5 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://mct.aacrjournals.org/content/15/6/1248To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-15-0885

http://mct.aacrjournals.org/content/suppl/2016/03/26/1535-7163.MCT-15-0885.DC1

http://mct.aacrjournals.org/content/15/6/1248.full#ref-list-1

http://mct.aacrjournals.org/content/15/6/1248.full#related-urls

http://mct.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mct.aacrjournals.org/content/15/6/1248


inhibition of pi3k/bmx cell survival pathway sensitizes to ...doxorubicin sensitivity and...

Documents