research preclinical and clinical estimates of the basal...

Pred

Preof aby S

Lian Z

Abst

Aposubseviabilprimacell sutrinsicmitoccellulsignaltion, tlished

AuthorOncoloSchool

Corresand ThoCancer80045.ucdenv

doi: 10

©2010

Clin C4478

D

Published OnlineFirst July 14, 2010; DOI: 10.1158/1078-0432.CCR-10-0859

Clinical
Canceresearch
ictive Biomarkers and Personalized Medicine

clinical and Clinical Estimates of the Basal Apoptotic RateCancer Predict the Amount of Apoptosis Induced

R

ubsequent Proapoptotic Stimuli

hang1, Brian D. Kavanagh2, Andrew M. Thorburn1, and D. Ross Camidge3

ractPur

to submeasunormaExp

linesstableand cfromRes

KDs inminalpredicdUTPcaspassensitM30 v(r = 0Con

he multipcancers

s' Affiliatigy, and 3Dof Medicine

ponding Auracic OncoCenter, MaiPhone: 720er.edu.

.1158/1078-

American A

ancer Res

ownloade

pose: We hypothesized that the basal apoptotic rate (BAR) of a cancer would predict sensitivitysequent proapoptotic stimuli. To explore this, preclinical and clinical BAR assays were developedring cumulative apoptotic events through ELISAs for soluble caspase-cleaved cytokeratin 18 (M30)lized to either cell number increase or total tumor volume, respectively.erimental Design: The BARs of A549, HCC44, and SW1573 non–small cell lung carcinoma cellwere measured following different pro/antiapoptotic manipulations. In isogenic wild-type andknockdown (KD) series, pretreatment BARs were correlated with response to proapoptotic stimuliompared with established apoptosis assays. Pretreatment and posttreatment serum was availablestereotactic body radiation therapy patients.ults: Caspase inhibition and p53 KDs reduced the BAR, whereas serum deprivation, XIAP, or Bcl2creased the BAR. The nontreated BAR rank ordering of the XIAP series recapitulated that with ter-deoxynucleotidyl transferase–mediated dUTP nick end labeling and caspase-3/7 activity assays, andted each line's sensitivity to TRAIL or irradiation. Terminal deoxynucleotidyl transferase–mediatednick end labeling, however, underestimated basal apoptosis during increased apoptotic stress, ande-3/7 activity detected minimal death in the media. P53 KDs with lower nontreated BARs were lessive to TRAIL and cisplatinum than wild-type. Stereotactic body radiation therapy increased serumalues, and the pretreatment clinical BAR strongly correlated with fold change in M30 on treatment.93).clusions: M30-based BAR assays reflect apoptosis accurately and are more amenable to clinicalation than existing apoptosis assays. The pretreatment BAR correlates with cell and/or tumor sensi-
applic
tivity to extrinsic and intrinsic apoptotic pathway stimulation. Prospective clinical exploration is warranted.Clin Cancer Res; 16(17); 4478–89. ©2010 AACR.

stressgenercers. Crecogwell ato canapoptmolecsignal

ptosis is characterized by caspase activation andquent cleavage of cellular proteins essential for cellity (1). Proapoptotic signaling activates caspasesrily through either the extrinsic pathway followingrface death receptor stimulation or through the in-pathway, involving cytochrome c release from thehondria following p53-dependent/independentar damage recognition (2). Although proapoptoticing can clearly arise through therapeutic interven-

le genetic aberrations present within estab-, coupled with the microenvironmental

of apmembablyapoptIt h

compa cellthe prold, binevitferentpheno

ons: Departments of 1Pharmacology, 2Radiationivision of Medical Oncology, University of Colorado,, Aurora, Colorado

thor: D. Ross Camidge, Developmental Therapeuticslogy Programs, University of Colorado Comprehensivel Stop F704, ACP 2256, 1665 Aurora Court, Aurora, CO-848-0449; Fax: 720-848-0360; E-mail: ross.camidge@

0432.CCR-10-0859

ssociation for Cancer Research.

; 16(17) September 1, 2010

Research. on February 1, 2clincancerres.aacrjournals.org d from

es of malignant growth, proliferation, and spreadate significant basal proapoptotic signaling in can-onsequently, resistance to apoptotic cell death is

nized as an important aspect of tumorigenesis, ass resistance to anticancer therapies (3, 4). In additioncers having the impetus to deselect functional pro-otic machinery directly, within the cell, a series ofular brakes on proapoptotic signaling (prosurvivaling) are also known to exist, including the inhibitorsoptosis proteins and the antiapoptotic Bcl2 familyers, which are upregulated in some cancers presum-as an additional/alternative means of evadingosis (5, 6).as been suggested that the overall balance betweeneting proapoptotic and antiapoptotic signaling withinat any particular time may be viewed as determiningoximity of a cell to a hypothetical apoptotic thresh-eyond which completed apoptotic cell death becomesable (7). Based on this model, in which multiple dif-
stimuli can be integrated together into a commontypic decision (“live or undergo apoptosis”), we
019. © 2010 American Association for Cancer

http://clincancerres.aacrjournals.org/

hypoton avdifficuformesize tantiapronmeand omity tapoptcer's oAlth

betwecancediothetreatmconsisheteroof apoin thesignifcells uto be cselvesWe

methotionscumuantibothe caintermlar epmunotissue

is botat combloodprecliapoptdrugzationits preWe

canceproblestablwith igests r(16).assaycell nBAR ovationBAR wpatienminedThe vtheseproaptactic

Mate

ReageRea

supplUnlesobtainchemobtaiwas pgenero

Cell lAll

at theRPMIvate, 2cose alines wrepeatpliedCultuand agMinnDallasSW15but ha

In vit

Translational Relevance

The proapoptotic and antiapoptotic signaling bal-ance determines the proximity of a cell to the apoptoticthreshold. Cancers closer to this threshold may have ahigher basal apoptotic rate (BAR) and a greater propen-sity to commit apoptosis in response to a proapoptoticstimulus. Using specific normalizations of the M30Apoptosense ELISA (which can assess cumulative celldeath in culture medium or serum) based on viable cellnumber or total tumor volume, we have estimated theBAR of cancer cell lines and clinical cancers, respective-ly. The M30 BAR correlates with apoptotic outcomesfollowing different proapoptotic stimuli, includingradiation treatment in patients. This assay may nowbe explored prospectively as a predictive marker of clin-ical radiation sensitivity and as a specific pharmaco-dynamic marker, looking for changes in the BAR topredict benefit from the direct apoptosis-manipulatingagents currently being explored within clinical trials.

Basal Apoptotic Rate Predicts Apoptotic Sensitivity

www.a

D


hesize that the distance a cell (or a cell population,erage) is from this threshold determines how easy orlt it will be to kill the cell following an additional uni-proapoptotic stimulus (Fig. 1A). Second, we hypoth-hat, given the natural variation in proapoptotic andoptotic stimuli occurring in the cellular microenvi-nt as a result of, for example, growth factor, nutrient,xygen level fluctuations, a cancer with a closer proxi-o the apoptotic threshold will have a higher basalotic rate (BAR), which provides a surrogate of the can-verall apoptotic threshold proximity (Fig. 1B).ough clinically some correlation has been reporteden the histologically derived apoptotic index ofrs and subsequent response to radiation/chemora-rapy in tumors accessible for pretreatment and post-ent biopsies/resections, findings have not beentent (8–11). Beyond the problems of intratumoralgeneity when biopsying, histopathologic estimatesptosis and in vitro methods based on detecting cellsprocess of apoptosis at any one time point may alsoicantly underestimate the true BAR. This is becausendergoing apoptosis are unlikely to remain availableounted but, rather, they will rapidly dismantle them-, undergo phagocytosis, and disappear from view.have developed a more representative and practicald for assessing the BAR using specific normaliza-of the M30 Apoptosense ELISA analysis to estimatelative cell death both in vitro and in vivo. M30 is andy directed against a neoepitope produced duringspase-mediated cleavage of cytokeratin 18, a type Iediate filament common in single layer and glandu-ithelial cells (12). In addition to being used for im-
histochemical detection of apoptotic cells in solids, the cleaved fragment detected by M30 antibody
Undone

acrjournals.org

Research. on February 1, 2clincancerres.aacrjournals.org ownloaded from

h soluble and stable, offering the potential to lookpleted apoptosis events in cell culture media or

. The M30 Apoptosense assay has been used to shownical and clinical pharmacodynamic increases inosis following both novel and traditional anticancertreatments (13–15). However, appropriate normali-of the M30 assay to explore a cancer's BAR and

dictive significance have not been described before.chose to predominantly explore non–small cell lungr (NSCLC) as an example of a major cancer healthem (16). The limited effectiveness of many of theished treatments in lung cancer patients, particularlyncreasing lines of therapy in advanced disease, sug-esistance to apoptosis is common in this populationHere, we describe how using the M30 Apoptosenseon unchanged culture media normalized to viableumber increase over time was used to assess thef NSCLC cells in vitro. We then extended our obser-s to a clinical setting in which a clinically derivedas assessed by normalizing M30 reactivity in cancerts' serum to the baseline total tumor burden deter-by volumetric computed tomography scanning.

alidation and potential predictive significance ofBAR assays in NSCLC cell lines undergoing differentoptotic stimuli and in patients undergoing stereo-body radiation therapy (SBRT) are reported.

rials and Methods

ntsgents were RPMI 1640 growth medium (HyClone)emented with 10% fetal bovine serum (HyClone).s otherwise noted, chemicals and reagents wereed from Sigma. z-VAD-fmk was from Alexis Bio-

icals. M30 Apoptosense ELISA kits (PEVIVA) werened from DiaPhama. Human recombinant TRAILurchased from R&D Systems. Lexatumumab wasusly given by Human Genome Sciences.

ine and cell culturesthe lung cancer cell lines were the gift of Barb HelfrichUniversity of Colorado, Aurora, CO, and cultured in1640 supplemented with 10% FCS, 1% sodium pyru-.0 mg/mL sodium bicarbonate, and 10 mmol/L glu-t 37°C with 5% CO2 unless specified otherwise. Cellere DNA profiled using polymorphic short tandemmarkers (Identifiler Plus PCR Amplification kit, Ap-Biosystems). A549 originally from the American Typere Collection was matched by profiling in May 2006ain in December 2009. HCC44 obtained from John D.a and Adi Gazdar (University of Texas Southwestern,, TX) was matched by profiling in December 2007.73 was obtained from H Broxterman (Amsterdam)s not yet been profiled at the University of Colorado.

ro BAR assay
less specified otherwise, the in vitro BAR assay wasas follows: 80,000 cells were seeded per plate and
Clin Cancer Res; 16(17) September 1, 2010 4479



grownmediaand thplatescount

M30 Aturer'sviable

Zhang et al.

Clin C4480

D


in the presence of 10% FCS without changinguntil cells reached confluence (all cells in contact,e entire surface of the plate was covered). Replicate
were sacrificed daily to collect the media and tothe cell number. The media were subjected to the
M30numb

ancer Res; 16(17) September 1, 2010


poptosense ELISA assay according to the manufac-instruction. Time points at approximately equivalentcell numbers before confluence were compared, with
values normalized to the corresponding exact celler at the time of harvesting.
Clinical Cancer Research



MTS cCel

96-weindicaend oMTS,optimwavelBenchLabor

ClonoFive

licatelexatuatumuwashiaddedfixeddiH2O20% ewere swas re

CaspaCel

densitlowedfor 4,perimto eacAfter 1tivitiecropla

In situdUTPEigh

triplicTRAILmeasution k

instruevery

SmallThe

viralOpencotraenvelposttrthroustoragwithwith(sc-29nologplasmto thecin (1stable

ImmuCel

ferrincondubodies(#287(sc-12

SBRTQu

fromBoardinclud10 Gyor livenancypatiencriteriratin

Fig. 1.of howproximidrive frthreshocompoA1 is noeither trestingpathwacells coThe dis“lift” (siz(the dostimuliat cum


www.a

D


ell viability assayls were plated at a density of 4,000 cells per well in all plate in triplicate wells. Cells were treated with theted concentrations of drugs for 24 to 48 hours. At thef the experiment, the cells were treated with 20 μL ofinner salt (Promega) and incubated at 37°C untilal color changes occur. The absorbance values atength of 490 nm were measured with a Bio-Radmark Plus Microplate Spectrophotometer (Bio-Radatories).

genic survival assayhundred cells were plated in six-well plates in trip-manner, followed by human recombinant TRAIL ormumab the next day. After treating with TRAIL or lex-mab for 18 to 24 hours, the drugs were removed byng cells with prewarmed PBS, and fresh media were. Following 7 to 10 days of incubation, colonies werewith a fixer (10% acetic acid, 10% methanol, 80%) and stained with crystal violet (0.4% crystal violet,thanol). After the images were scanned, the coloniesolubilized with 30% acetic acid, and the absorbancead at a wavelength of 540 nm.

se-3/7 activity assayls were plated in a 96-well black-walled plate at ay of 4,000 cells per well in a triplicate manner, fol-by the indicated concentrations of TRAIL treatment8, or 24 hours. One hour before the end of the ex-ent, Caspase-Glo 3/7 reagent (Promega) was addedh well according to the manufacturer's instructions.hour of incubation in the dark, the caspase-3/7 ac-

s of the samples were measured using a Veritas mi-te luminometer (Turner BioSystems).

terminal deoxynucleotidyltransferase-mediatedbiotin nick end-labeling assayty thousand cells were plated in 24-well plates inate wells. The next day, the cells were treated with(50 ng/mL) for 4, 8, or 24 hours. Apoptosis was
red by using the TiterTACS In situ apoptosis detec-it (R&D Systems) according to the manufacturer's
of growith

Proximity to the apoptotic threshold may influence propensity to apoptose in rescancer cells may avoid apoptosis (passage beyond the apoptotic threshold) in rties to the apoptotic threshold. In Cell 1, the prodeath signaling from insult A1 isom insult A is reduced (A2), for example through increased repair of drug-induceld, the cell remains alive. In Cell 3, the prodeath drive from insult A remains thesite of all its other prodeath and antideath signaling has shifted its resting positiow insufficient to reach the threshold. Other insults with stronger prodeath drive (B1he prodeath drive resulting from an insult (the downward arrows), the proximity oproximity is a composite of prodeath and antideath signaling from many different cys with the therapeutic stressors. B, schematic representation of how, given thentinually experience even under standard growth conditions, proximity to the aptance of the cell relative to the ground is represented as a balance between proae of the balloon). Contact with the ground represents commitment to cell deathtted balloons and the dotted fluctuation line) will reach the apoptotic threshold mmanifesting as a higher BAR compared with the other (solid) population of cells.ulative products of caspase-mediated cell death in the unchanged culture media

acrjournals.org


ctions except that the amount of reagents used instep was doubled.

hairpin RNA transfectionXIAP, BCL2 small hairpin RNA (shRNA) lenti-

constructs were obtained from Thermo ScientificBiosystems. The shRNA lentiviral constructs werensfected with packaging plasmid psPAX2 andope plasmid pMD2.G into 293T cells. Two daysansfection, the virus was harvested and filteredgh a 0.45-μm filter and aliquoted for infection ore at −80°C. A549 and SW1573 cells were infectedXIAP or BCL2 shRNA lentivirus and selectedpuromycin (2 μg/mL). The p53 shRNA constructs435-sh) were purchased from Santa Cruz Biotech-y. A549 Cells were transfected with p53 shRNAids using DharmaFect 1 (Dharmacon) accordingmanufacturer's instructions, followed by puromy-.5-3 μg/mL) selection for 2 to 3 weeks to establishtransfectants.

noblotting analysisl lysate preparation, electrophoresis, membrane trans-g, signal detection, and film developing werected as previously described (17). The primary anti-used were as follows: anti-XIAP (#2045) and anti-BCL22) were from Cell Signaling Technology. Anti-p536) antibody was from Santa Cruz Biotechnology.

-treated patient samplesantitative serum M30 assays were done on serumpatients treated on a prospective Institutional Review–approved protocol. Eligibility for this protocoled the planned delivery of SBRT (minimum dose,per fraction for three fractions) to one or more lungr primary or metastatic lesion(s) from a solid malig-. From the entire cohort enrolled on the protocol,ts eligible for the BAR analysis met the followinga: (a) epithelial tumors expected to contain cytoke-18 positive cells, (b) SBRT given to all known sites
ss disease, (c) no concurrent use of systemic agentsknown effects on cellular apoptosis (e.g., systemic
ponse to a proapoptotic stimulus. A, schematic representationesponse to a therapeutic insult. Black bars, i and ii, differingsufficient to induce completed apoptosis. In Cell 2, the prodeathd damage, and given the same resting proximity to the apoptoticsame as in Cell 1 (A1), but the cell does not die because then away from the apoptotic threshold (ii), and the drive from insult) may still kill Cell 3. NB-specific molecular factors could influencef the cell to the threshold (the black bars), or both, as theellular stressors that may or may not share death-related signalingnormal fluctuations in prodeath and prosurvival stimuli thatoptotic threshold may be estimated by measuring the BAR.poptotic “gravity” (size of the balloon's basket) and antiapoptotic. The population of cells closer to the apoptotic thresholdore often given the same fluctuation of pro/antiapoptoticC, schematic representation of the in vitro BAR assay lookingof growing cells.




steroidwere pof SBRsera ppairedclinicatotal greactidetermcoefficin all

Resu

The BusingunchaTo

plied

to dethat oto incof celCellsto calreachemateland thThe ediagraWh

undertectabof theresultalent

Fig. 2.inhibitoZ-VAD-were haproducadjusteto detecorrespserum

Zhang et al.

Clin C4482

D


anti-inflammatory agents). Serum collection timesretreatment and immediately after the final fractionT. Absolute M30 reactivity increases in the patients're- and post-SBRT were compared using two-tailedt testing. The correlation between the pre-SBRTlly determined BAR [baseline M30 divided by theross tumor volume (GTV)] and fold change in M30vity from pre-SBRT to immediately post-SBRT wasined by calculating the product-moment correlationient, r. Typically, the GTV was expanded 5 to 10 mmdirections to generate the planning target volume.

lts

AR of a cancer cell line can be measured in vitroa normalized M30 ELISA assay applied tonged culture media
assess the BAR of a cancer cell line in vitro, we ap- and d
onding cell number; bars, SD. P values were obtained using Student's t test of cconcentration. ***, P < 0.0001.



termine the cumulative caspase-related cell deathccurred during the time it took for the cell linerease its viable cell number from a fixed seedingls (n) to a second, larger, viable cell number (N).were grown to confluence, but media were takenculate the BAR at a time before confluence wasd. Growth times were adjusted to allow approxi-y the same number of viable cells to be presenten normalized to the exact number on harvesting.xperimental design is illustrated in the schematicm in Fig. 1C.en A549 NSCLC cells (adenocarcinoma) were grownoptimal growth conditions, M30 reactivity was de-le in the culture media. The addition of 25 μmol/Lpan-caspase inhibitor z-VAD-fmk to the media

ed in a faster growth rate of the cells, such that equiv-cell numbers were achieved ∼1 day earlier (Fig. 2A)
ecreased M30 reactivity normalized to cell number
the M30 ELISA assay to unchanged culture media before approaching confluence compared with nontreated

The BAR of a cell line changes in response to antisurvival and/or proapoptotic stimuli from the extracellular microenvironment. The pan-caspaser z-VAD-fmk increased the growth rate (A) and reduced the BAR (B) of A549 cells measured before confluence by the method depicted in Fig. 1C.fmk was added to the medium (final concentration, 25 μmol/L) the day after the cells were seeded and remained in the medium until the cellsrvested. The graphs represent typical experiments repeated at least once. Points, mean from three replicates; bars, SD. In contrast, serum starvationes the opposite results. C, growth curves of HCC44 cells in the presence of 10%, 2%, or 0.5% serum concentration. D, growth times wered to reach approximately similar numbers of viable cells before confluence, and M30 reactivity normalized to exact cell number on harvestingrmine the BAR of the cell lines under the different growth conditions. Columns, mean from three replicates after normalization to the

omparing the BAR in 10% versus the BAR in 2% or 0.5%




cells (as conincreaConsecalcultial phadjustlevels,

The Bantisuthe exSeru

apoptto thetal strewhethtotic/aBARsand S

the prcontraconce(Fig. 2with a

The Beitherby geaspecIf th

imitybalanthe cewe us(KD)shoulthe ce

Fig. 3.differenShRNAM30-baKD lines measured by the M30-based BAR assay. The M30 values presented are means ± SD from three replicates after normalization to the correspondingcell num in WT


www.a

D


Fig. 2B). There was a dramatic increase in the BARfluence was approached presumably due to thesed stress of cell crowding and nutrient depletion.quently, in all subsequent assays, the BAR wasated based on media taken during the early exponen-ase of growth (approximately 650-850,000 cells),ing the growth time as necessary to achieve thesenormalized to the exact cell number at harvesting.

AR of a cell line changes in response torvival and/or proapoptotic stimuli fromtracellular microenvironmentm deprivation has been shown to trigger the intrinsicotic pathway (18). Following the observation relatingpotential effects of cell density as a microenvironmen-ssor on the BAR of A549 cells (Fig. 2A and B)we testeder the BAR assay could reflect changes in the proapop-ntisurvival stimuli caused by serum deprivation. The

ber. P values were obtained using Student's t test of comparing the BAR

of two NCSLC cell lines, HCC44 (adenocarcinoma)W1573 (squamous carcinoma), were determined in

A549substa

acrjournals.org


esence of 10%, 2%, or 0.5% fetal bovine serum. Inst to the effects of z-VAD-fmk, reduction of serumntration resulted in slower growth rate in HCC44C) and SW1573 cells (data not shown), in conjunctionn increase in the BAR of both cell lines (Fig. 2D).

AR of a cancer cell line can be manipulatedtoward or away from the apoptotic threshold

netically changing molecules affecting differentts of the apoptotic balance within the celle BAR represents an overall index that reflects prox-to the apoptotic threshold, then manipulating thece of proapoptotic and antiapoptotic molecules inll should change the BAR. To test this hypothesis,ed shRNAs to produce cells with stable knockdownsof either XIAP or Bcl2, each of which in theoryd reduce antiapoptotic (prosurvival) signaling withinll (Fig. 3). Two different XIAP shRNAs (#2 and #4 in

versus the KDs. *, P < 0.01; **, P < 0.001; ***, P < 0.0001.

The BAR of a cancer cell line can be manipulated either toward or away from the apoptotic threshold by genetically changing molecules affectingt aspects of the apoptotic balance within the cell. A, BARs of the WT and the isogenic XIAP KDs measured by the M30-based BAR assay.XIAP KDs confirmed by Western analysis in A549 and SW1573 cells (inset). B, BARs of the WT and the isogenic BCL2 KDs measured by thesed BAR assay. C, ShRNA p53 KDs confirmed by Western analysis in A549 cells in response to doxorubicin. D, BARs of the WT and the isogenic p53

cells, and #4 and #5 in SW1573 cells) displayedntial KD in each cell line as determined by Western




blottidownthe isof Bclwith tTo

oppoexamptory mdoxorboththe A5(Fig. 3ing dothe BAline (F

CompBAR atranscaspaTo

pared

media3/7 acThe rameasucorrelBAR athe cethe apand Xreducthe apBAR ain allXIAP(Fig.compbasedthe toparedWe

for 8

Fig. 4.(50 ng/and ap8 h, folcell plucompacompa

Zhang et al.

Clin C4484

D


ng for the expressed protein (Fig. 3A). KnockingXIAP resulted in increased BARs compared withogenic wild-type (WT) lines (Fig. 3A). Stable KDs2 similarly increased the BAR of A549 comparedhe isogenic WT line (Fig. 3B).manipulate the pro/antiapoptotic balance in thesite direction, we then knocked down p53 as anle of a damage sensing (i.e., proapoptotic) stimula-olecule. Stabilization of p53 protein in response toubicin treatment in the p53 KDs (p53-1, p53-2) wasdelayed and reduced in magnitude compared with49 WT line, indicating efficient KD by the shRNAC). Consistent with our central hypothesis, knock-wn p53 (reducing proapoptotic signaling) reducedR of the cells compared with the isogenic A549 WTig. 3D).

arison of the M30-based cumulative cell deathssay with terminal deoxynucleotidylferase–mediated dUTP nick end labeling andse-3/7 activity-based apoptotic assays
further validate our M30-based BAR assay, we com-it to both terminal deoxynucleotidyl transferase–
tion iFig. 4

ring the TUNEL signal (A) or BAR (B) without versus with TRAIL treatment. The cred with the cell + media samples in D. *, P < 0.01; **, P < 0.001; ***, P < 0.0001



ted dUTP nick end labeling (TUNEL) and caspase-tivity assays in our A549 WT and XIAP KD cell lines.nk ordering of the apoptotic rates of the three linesred by in situ TUNEL staining of fixed cultured cellsated well with that determined by our M30-basedssay (compare Figs. 3A and 4A). However, whenlls were challenged with 50 ng/mL TRAIL for 8 hours,optotic rates measured by TUNEL decreased in WTIAP KDs compared with no treatment, with a greatertion in the XIAP KD lines (Fig. 4A). In contrast, whenoptotic rates were measured using the M30-basedssay, TRAIL treatment increased the apoptotic ratesthe cell lines with the greatest effects being in theKDs and in those with the highest nontreated BAR4B). These data suggest that the ability to measureleted apoptotic events as is provided by the M30-BAR assay provides a more accurate measure oftal amount of apoptosis that has taken place com-with the “snapshot” provided by TUNEL staining.next explored increasing concentrations of TRAILhours, followed by caspase-3/7 activity determina-
n the culture wells (media and cells combined;C). Unlike with TUNEL, caspase activity revealed
Comparison of the M30-based BAR assay with TUNEL and caspase-3/7 activity assays. A, A549 WT and XIAP KD cells were treated with TRAILmL) for 8 h, and apoptosis was then measured by TUNEL assay. B, A549 WT and XIAP KD cells were treated with TRAIL (50 ng/mL) for 8 h,optosis was then measured by the M30-based BAR assay. C, A549 WT and XIAP KD cells were treated with increasing concentrations of TRAIL forlowed by caspase-3/7 activity assessment in the cells and media. D, the caspase-3/7 activity was measured in the cells only, media only, ands media at the basal level (without any treatment). Columns, mean from three replicates; bars, SD. P values were obtained using Student's t test of

aspase-3/7 activity of the cell only or media only samples was.




identiKD liconcebe ducaspacell p3/7 athe mtranslbased

The Min resstimuTo

predicway sviabilof concells tsensitXIAPtheircorrellexatuRecepBAR ppabilicondument.isogenTRAILotherold, recellswith(Fig. 5

The Min resin vitHav

comeswe testimuwas ain thAlthoof cisptheir ntheirsurvivWe

manipof radWT, Xlinesinduc

we haBARspostexlowertionsTo

mentprimalast d(liverpatienscriptithreeto 7 dreactivSBRT.145 ±SBRTof poBAR [(all siBAR soptosof 0.9

Discu

Muto dewill irmitmin probe vithresha cell,old wproap(Fig. 1reticala cellELISAchangof casgrowtcaspaconflurectlyincreajust aviableMa

inhibithe Bhypotconcethresh


www.a

D


cal sensitivity rankings across the WT and XIAPnes to the M30-based rankings even at the highestntrations of TRAIL. To explore how much mighte to caspase activity in the media, we comparedse-3/7 activity in the media with activity in theellet. The results showed that almost all caspase-ctivity was restricted to the cell pellet and notedia (Fig. 4D), thus offering less potential forational application to body fluids than the M30-assay (see below).

30-based BAR predicts for apoptotic sensitivityponse to extrinsic apoptotic pathwaylation in vitrofurther address the question whether the BAR cant treatment outcome from extrinsic apoptotic path-timulation, MTS assays were done to measure cellity after 48 hours of TRAIL treatment across a rangecentrations. XIAP KD sensitized A549 and SW1573o TRAIL treatment (Fig. 5A). More importantly, theivity rank ordering by MTS assay in the WT andKDs in A549 and SW1573 cells correlated well withnontreated BAR ranking (Figs. 3A and 5A). Thisation also applied to the monoclonal antibodymumab, which recognizes and stimulates Deathtor 5 (data not shown). To determine whether theredicts for long-term survival and proliferation ca-ty after treatment, clonogenic survival assays werected with or without TRAIL/lexatumumab treat-Again, the cell lines with the highest BAR in theiric series correlated with the poorest survival rate onor lexatumumab treatment (Fig. 5B and C). On thehand, pushing cells away from the apoptotic thresh-ducing their BAR by knocking down p53, made themore resistant to the TRAIL treatment comparedthe WT parent line in clonogenic survival assaysD).

30-based BAR predicts for apoptotic sensitivityponse to intrinsic apoptotic pathway stimulationro and in patients undergoing SBRTing shown the role of the BAR in determining out-from extrinsic apoptotic pathway–directed stimuli,

sted its relevance to intrinsic apoptotic pathwaylation. The lower nontreated BAR in p53 KD linesssociated with less M30 induction following growthe presence of 1 μmol/L cisplatinum (Fig. 6A).ugh the rank ordering of the BARs in the presencelatinum in the p53 KD series did not recapitulateontreated BAR rank ordering exactly, it did parallel

phenotypic sensitivity to cisplatinum in clonogenical assays (Fig. 6B).also tested radiation exposure in vitro as a way ofulating the intrinsic apoptotic pathway. Two graysiation clearly increased the postexposure BARs acrossIAP (Fig. 6C), and p53 (data not shown) KD cell
compared with no treatment, indicating radiation-ed caspase activation in these cells. Similar to what
undermulat

acrjournals.org


d observed before, lines with higher nontreated(XIAP KDs) manifested higher elevations in theirposure BAR following radiation treatment, whereasnontreated BARs (p53 KDs) manifested lower eleva-in their postexposure BARs.test if these measurements apply during patient treat-, we analyzed serum samples from patients withry or metastatic cancer collected before and on theay of a course of SBRT to all known sites of diseaseand/or lung). The clinical characteristics of the SBRTts (n = 13) are shown in Table 1. The median pre-on dose to the planning target volume was 60 Gy infractions (range, 36-60), administered in a total of 3ays. Consistent with our in vitro work, absolute M30ity increased in the 13 patients' sera pre- and post-The baseline M30 levels (mean ± SEM, U/L) were14 and increased to 232 ± 34 immediately post-(P < 0.01 by paired t test). Plotting the M30 ratiost- to pre-SBRT against the clinically determinedpre-SBRT M30 divided by the total tumor volumetes) in mm3] showed that this clinically determinedtrongly correlated with the apparent amount of ap-is induced by the treatment (correlation coefficiency3; Fig. 6D).

ssion

ltiple proapoptotic and antiapoptotic signals interacttermine whether, at any particular moment, a cellrevocably commit to undergo apoptosis. The com-ent to die is likely to result from sufficient imbalanceapoptotic and antiapoptotic signaling, or what may

ewed as transition past a hypothetical apoptoticold. We hypothesized that the theoretical distanceor a cell population on average, is from this thresh-ill determine how easy it will be for a subsequentoptotic stimulus to push the cell past this thresholdA). Moreover, we have hypothesized that this theo-distance may be estimated by measuring the BAR ofpopulation (Fig. 1B). Using the Apoptosense M30to quantify cumulative apoptotic events in the un-ed culture media, it is clear that a significant amountpase-dependent apoptosis occurs during normalh in vitro (Fig. 2A and B). The effect of the pan-se inhibitor z-VAD-fmk on both the time to reachence and M30 reactivity suggests that the BAR di-influences the perceived “growth rate” of a cancer,ses in the BAR reducing the total viable cell number,s increases in the division rate would increase thecell number.

ny factors affect the BAR. In addition to caspasetors reducing the BAR, serum deprivation increasesAR (Fig. 2D). This is consistent with our primaryhesis. Specifically, cells grown in reduced serumntrations are likely to be closer to the apoptoticold and have a higher tendency to spontaneously
go apoptosis as shown by an increase in the cu-ive caspase-mediated cell death products detectable



Fig. 5.treatedassaysthe absusing SIn additin p53

Zhang et al.

Clin C4486

D


The M30-based BAR predicts for sensitivity to extrinsic apoptotic pathway stimulation. A, A549, SW1573, and their isogenic XIAP KD cells werewith the indicated concentrations of TRAIL for 48 h; the cell viability rates were determined by MTS assay. B, C, and D, Clonogenic survivaldone with 500 cells treated with TRAIL (25 ng/mL) or lexatumumab (50 ng/mL) for 48 h. The stained colonies were solubilized with 30% acetic acid;orbances were read at 500nm and normalized to the no treatment ones. Points, mean from three replicates; bars, SD. P values were obtainedtudent's t test of comparing the survival rates of WT versus KD in response to TRAIL or lexatumumab. #, P < 0.05; **, P < 0.001; ***, P < 0.0001.
ion, there was a significant difference between the KDs in each series (P < 0.01 in XIAP KDs in A549 and SW1573 background and P < 0.05KDs), suggesting that the BAR predicts for survival on TRAIL and lexatumumab treatment.
ancer Res; 16(17) September 1, 2010 Clinical Cancer Research

Research. on February 1, 2019. © 2010 American Association for Cancerclincancerres.aacrjournals.org ownloaded from


in theelevateffectof thecell li(Fig. 2tic anthe BAdirectas asOth

alreadfyingas apoassayiof thetial toTheselookswith tXIAP K

standdecrea(Fig. 4are lotion bM30-bsing Tpost-Tsuggeand, aing cecells wactivitdardfor traactivethan rDat

ptotic

Fig. 6.A, A549The mothe noB, the sline X-2delivereof the MM30 vaM30 rethe BAR m or r


www.a

D


media. Again, as a direct consequence of theired apoptotic rate, plus or minus any additionalof serum deprivation on the actual division ratecells, in the presence of serum deprivation, the

nes take longer to increase their viable cell numberC). Genetic manipulation of different proapopto-d antiapoptotic signaling molecules also movedR of lung cancer cell lines in vitro in the expected

ions. XIAP and Bcl2 KDs increased the BAR, where-p53 KDs reduced the BAR (Fig. 3).er in vitro methods for determining apoptosisy exist. Yet, many of these methods rely on quanti-apoptotic events at a particular time point. However,ptosis results in cell destruction, methods based onng apoptotic cells/bodies represent only a snapshottrue extent of apoptosis and therefore have the poten-significantly underestimate the true apoptotic rate.limitations are reduced with the M30-ELISA, whichat completed/cumulative apoptotic events. In line

(A and C) or survival rate (B) of WT versus KDs in response to cisplatinu

his, although TUNEL rank ordering of our differentDs correlated with the M30 BAR rank ordering under

BAR aapopt

acrjournals.org


ard growth conditions, TUNEL readings actuallysed in the presence of death receptor stimulationA). When apoptosis occurs in large amounts, cells

st for subsequent quantitation, thereby evading detec-y the TUNEL assay but still being detectable by theased BAR assay applied to the culture media. Asses-UNEL at 4 and 24 hours, in addition to 8 hours,RAIL exposure made no difference to the results,sting that apoptosis under these conditions was rapids expected based on recent single cell studies examin-ll death (19), occurred at different times for differentithin the population (data not shown). Caspase-3/7y tracked the M30 assay more robustly under stan-and prodeath conditions, but offers little potentialnslational extension (Fig. 4C and D) because thecaspase is found primarily in the cell remnants rathereleased into the media or body fluids.a shown in Figs. 4–6 support our hypothesis that apo-threshold proximity, estimated through M30-based

adiation. *, P < 0.01; ***, P < 0.0001; #, P < 0.05.

The M30-based BAR predicts for apoptosis induction following intrinsic apoptotic pathway stimulation in vitro and in patients undergoing SBRT.and the p53 KD cell lines p53-1 and p53-2 had their M30-based BAR determined following exposure to 1 μmol/L cisplatinum or no treatment (NT).rning after cells were seeded, a final concentration of 1 μmol/L freshly made cisplatinum in DMSO, or the same amount of DMSO only fortreatment control, were added to the cells. At comparable viable cell numbers, preconfluence cells and the media were collected for the M30 assay.ame cell lines and treatments as in A were explored for phenotypic correlates through clonogenic survival assays. C, A549 and the XIAP KD cell, SW1573, and the XIAP KD cell line X-4 had their M30-based BAR determined following 2 Gy of radiation or no treatment. The radiation wasd through a RS2000 irradiator; the morning after the cells were plated. The BAR values presented in the graphs are means ± SD from three replicates30 values after normalization to the corresponding cell number. D, a clinically determined BAR was calculated by normalizing the baseline

lue to the GTV (determined by volumetric computed tomography scanning used in the radiation planning). The correlation between the fold change inactivity on radiation treatment (post-SBRT/pre-SBRT) and the clinical BAR is shown. P values were obtained using Student's t test of comparing

ssays in vitro, correlates with propensity to undergoosis following an additional proapoptotic stimulus.




ChangincreahigherFigs. 4phenoXIAPnomabaseddetermrespocorrelation pM30-bderingfully midea ththe appotenthe cecell fatbalangratedSup

ing thpletewith ttypic oand irIn o

outcodate wing thward

to mosis. Ththe clopmeproximhas tvariabcorrelsubsetumoicallybloodvolumshowfollowing Nof thecell dtreatmestimaProspwith cand wBec

pase-ccontrobetwesystemcytokeof ourexpresploreda noncompexist.18 difbetweinM3cytokeenougpoptoclinicain addtributoptosindireOn

scribesure omolecbeingficatioeach c(24).cally dlook f

Table teristics of analyzpatie T

nMale/fMedia 63 (30Site tre

LiLu 8

Site ofLuO 8

HistoloSq 4

Med

*Ontong†Othcell

Zhang et al.

Clin C4488

D


es inM30 reactivitywithin the BAR assay on treatmentsed significantly more among those cell lines withpretreatment BARs and vice versa, and it is clear from–6 that this reflects muchmore than a simple additivemenon. For example, the pretreatment BARs in theKD series in adenocarcinoma and in squamous carci-NSCLC cell lines predict, not only changes in theM30-BAR on treatment, but also phenotypic cell deathined through both MTS and clonogenic assays in

nse to death receptor stimulation. Similar outcometions in the XIAP series were apparent for M30 induc-ostirradiation (Fig. 6C). From Fig. 3A, although theased BAR correlated well with outcomes, the rank or-of XIAPprotein reductionby the different KDs did notatch with their different BARs. These data support theat the absolute protein level of amolecule involved inoptotic pathway (which continues to be explored as atial predictive biomarker) may poorly correlate withlls sensitivity or insensitivity to apoptosis, as, instead,e decisionmaking is actually determinedby the overallce of multiple different signals interacting (as is inte-together in the M30-BAR assay readout).porting the idea that multiple different factors affect-e BAR all influence the propensity of cells to com-apoptosis, the BARs of the p53 KDs, comparedheir WT lines, also correlated with M30 and pheno-utcomes to death receptor stimulation, cisplatinum,radiation (Figs. 5 and 6).ur model, most biomarkers predictive of treatmentme in cancer patients that have been explored toill have primarily focused only on factors determin-

neuroendocrine tumor.

e initial proapoptotic drive (the size of the down-arrows in Fig. 1A) and not how far the cell has left

shorttheir o



ve toward the apoptotic threshold to induce apopto-us, the ability to translate our in vitro findings intoinic represents a novel direction in biomarker devel-nt. Specifically, we suggest that apoptotic thresholdity, assessed through a clinically determined BAR,

he potential to be explored as an independentle determining treatment outcome. Previous reportsating the apoptotic index in cancer biopsies withquent response to radiation/chemoradiotherapy inrs have been inconsistent (8–10). However, our clin-determined BAR assay (baseline M30 readout innormalized by total tumor volume determined byetric computed tomography scanning) seems tovery strong correlation with apoptosis inductioning SBRT across a range of different cancers includ-SCLC (Table 1; Fig. 6D). Although in vivo/clinical useM30 assay has been described before as a marker ofeath following standard or experimental anticancerents (15, 20, 21), its appropriate normalization tote a true clinical BAR represents a new development.ective validation of the clinically determined BARlinical end points of efficacy or toxicity from SBRTith other therapies is now required.ause the basis for the M30 assay is the detection of cas-leaved cytokeratin 18, in all of our in vitro work, welled for any differences in cytokeratin 18 expressionen cell lines by conducting our analyses in isogenics. Of note, although there were some differences inratin 18 expression between different cell lines, noneKDs or interventions directly affected cytokeratin 18sion (data not shown). In our clinical assay, we ex-fold change in M30 as our end point, both to createconfounded variable with the BAR estimate and toensate for any variation in cytokeratin 18 that mayClinically, we do not know the extent of cytokeratinferences between tumors. Yet the degree of correlationen the clinically determined BAR and the fold change0 following SBRT suggests either that within our series,ratin 18 variation is not significant, or not significanth to bias the result in the presence of a strong proa-tic stimulus such as SBRT. Possibly confounding thel observations following SBRT is the likelihood thatition to tumor cell apoptosis, there may also be a con-ion to the serum levels of M30 by endothelial cell ap-is, a phenomenon supported by direct preclinical andct clinical observations (22, 23).e additional potential application of the assays we de-here relates to their use as a pharmacodynamic mea-f the dependence of cancer on particular signalingules. Many direct manipulators of apoptosis are nowexplored within oncology clinical trials, and preidenti-n of patients likely to derive the most benefit fromlass of agent represents a major parallel research effortGiven our data on the feasibility of measuring a clini-etermined BAR, it may be possible to use our assays toor changes in the clinically determined BAR following

1. Clinical characnts treated with SBR

ed

13emale 8:5
n age, y (range) -82) ated ver 5 ng primary cancer ng 5 ther* gy uamous enocarcinoma Ad 5
Other† 4ian GTV, cc (range) 26 (4-40)

e case each: ovary, bile duct, penis, maxillary sinus,ue, esophagus, kidney, liver.er includes NSCLC, not otherwise specified, or large

exposure to such agents, allowing the patient to act aswn control. In theory, thosewhoweremost dependent




on XIAest chapoptidentidrugs

Disclo

No p

Grant

InteInvestig

Thepaymenadvertiindicate

Refe1. De

202. Fu

ant3. Ha

57–4. Jo

can5. Co

de6. Ric

be7. Lo

208. Pa

Apsur

9. RöKi-restion

10. RocelchePh

11. WhshiIBand

12. Leetioear

13. Cuofde


www.a

D


P or Bcl2, for example, might then manifest the great-ange in their BAR on exposure to an inhibitors ofosis proteinorBcl2 inhibitor, allowing the assay tohelpfy those who will derive the most benefit from suchalone, or in subsequent combinations, in the future.

sure of Potential Conflicts of InterestRece

mmings J, Hodgkinson C, Odedra R, et al. Preclinical evaluationM30 and M65 ELISAs as biomarkers of drug induced tumor cellath and antitumor activity. Mol Cancer Ther 2008;7:455–63.

14. Decha na p

15. Miofclin73

16. WetheOn

17. HotoxPL

18. VoCa

19. AlbSoap

20. Olobioto

21. UluofpaCa

22. Garad30

23. Yaimles

24. CaapEp

acrjournals.org


Support

rnational Association for the Study of Lung Cancer Youngator's Award (D.R. Camidge).costs of publication of this article were defrayed in part by thet of page charges. This article must therefore be hereby markedsement in accordance with 18 U.S.C. Section 1734 solely tothis fact.

ived 04/05/2010; revised 06/11/2010; accepted 06/18/2010;
otential conflicts of interest were disclosed. published OnlineFirst 07/14/2010.
rencesgterev A, Boyce M, Yuan J. A decade of caspases. Oncogene03;22:8543–67.lda S, Debatin KM. Extrinsic versus intrinsic apoptosis pathways inicancer chemotherapy. Oncogene 2006;25:4798–811.nahan D, Weinberg RA. The hallmarks of cancer. Cell 2000;100:70.hnstone RW, Ruefli AA, Lowe SW. Apoptosis: a link betweencer genetics and chemotherapy. Cell 2002;108:153–64.ry S, Adams JM. The Bcl2 family: regulators of the cellular life-or-ath switch. Nat Rev Cancer 2002;2:647–56.hter BW, Duckett CS. The IAP proteins: caspase inhibitors andyond. Sci STKE 2000;2000:E1.we SW, Cepero E, Evan G. Intrinsic tumor suppression. Nature04;432:307–15.xton JR, Bolger BS, Armour A, Symonds RP, Mao JH, Burnett RA.optosis in cervical squamous carcinoma: predictive value forvival following radiotherapy. J Clin Pathol 2000;53:197–200.del C, Grabenbauer GG, Rödel F, et al. Apoptosis, p53, bcl-2, and67 in invasive bladder carcinoma: possible predictors forponse to radiochemotherapy and successful bladder preserva-. Int J Radiat Oncol Biol Phys 2000;46:1213–21.del C, Grabenbauer GG, Papadopoulos T, et al. Apoptosis as alular predictor for histopathologic response to neoadjuvant radio-motherapy in patients with rectal cancer. Int J Radiat Oncol Biolys 2002;52:294–303.eeler JA, Stephens LC, Tornos C, et al. ASTRO Research Fellow-p: apoptosis as a predictor of tumor response to radiation in stagecervical carcinoma. American Society for Therapeutic RadiologyOncology. Int J Radiat Oncol Biol Phys 1995;32:1487–93.rs MP, Kolgen W, Bjorklund V, et al. Immunocytochemical detec-n and mapping of a cytokeratin 18 neo-epitope exposed duringly apoptosis. J Pathol 1999;187:567–72.

miray M, Ulukaya EE, Arslan M, et al. Response to neoadjuvantemotherapy in breast cancer could be predictable by measuringovel serum apoptosis product, caspase-cleaved cytokeratin 18:rospective pilot study. Cancer Invest 2006;24:669–76.cha D, Cummings J, Shoemaker A, et al. Circulating biomarkerscell death after treatment with the BH-3 mimetic ABT-737 in a pre-ical model of small-cell lung cancer. Clin Cancer Res 2008;14:04–10.iss GJ, Bunn PA, Jr., Camidge DR. From radiotherapy to targetedrapy: 20 years in the management of non-small-cell lung cancer.cology (Williston Park) 2006;20:1515–24.rita H, Frankel AE, Thorburn A. Acute myeloid leukemia-targetedin activates both apoptotic and necroptotic death mechanisms.oS One 2008;3:e3909.usden KH, Lu X. Live or let die: the cell's response to p53. Nat Revncer 2002;2:594–604.eck JG, Burke JM, Aldridge BB, Zhang M, Lauffenburger DA,rger PK. Quantitative analysis of pathways controlling extrinsicoptosis in single cells. Mol Cell 2008;30:11–25.fsson MH, Ueno T, Pan Y, et al. Cytokeratin-18 is a useful serummarker for early determination of response of breast carcinomaschemotherapy. Clin Cancer Res 2007;13:3198–206.kaya E, Yilmaztepe A, Akgoz S, Linder S, Karadag M. The levelscaspase-cleaved cytokeratin 18 are elevated in serum fromtients with lung cancer and helpful to predict the survival. Lungncer 2007;56:399–404.rcia-Barros M, Paris F, Cordon-Cardo C, et al. Tumor response toiotherapy regulated by endothelial cell apoptosis. Science 2003;0:1155–9.mada Y, Bilsky MH, Lovelock DM, et al. High-dose, single-fractionage-guided intensity-modulated radiotherapy for metastatic spinalions. Int J Radiat Oncol Biol Phys 2008;71:484–90.ll JA, Eckhardt SG, Camidge DR. Targeted manipulation of
optosis in cancer treatment. Lancet Oncol 2008;9:1002–11,ub 2008 Aug 27.



2010;16:4478-4489. Published OnlineFirst July 14, 2010.Clin Cancer Res Lian Zhang, Brian D. Kavanagh, Andrew M. Thorburn, et al. Subsequent Proapoptotic Stimuliof a Cancer Predict the Amount of Apoptosis Induced by Preclinical and Clinical Estimates of the Basal Apoptotic Rate

Updated version

10.1158/1078-0432.CCR-10-0859doi:

Access the most recent version of this article at:

Cited articles

http://clincancerres.aacrjournals.org/content/16/17/4478.full#ref-list-1

This article cites 24 articles, 5 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/16/17/4478.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] at

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

Permissions

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

.http://clincancerres.aacrjournals.org/content/16/17/4478To request permission to re-use all or part of this article, use this link

Research. on February 1, 2019. © 2010 American Association for Cancerclincancerres.aacrjournals.org Downloaded from


http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-10-0859

http://clincancerres.aacrjournals.org/content/16/17/4478.full#ref-list-1

http://clincancerres.aacrjournals.org/content/16/17/4478.full#related-urls

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

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/16/17/4478


research preclinical and clinical estimates of the basal...

Documents