Oxygen-enhanced MRI Is Feasible, Repeatable, and Detects … · Precision Medicine and Imaging Oxygen-enhanced MRI Is Feasible, Repeatable, and Detects Radiotherapy-induced Change

Precision Medicine and Imaging

Oxygen-enhanced MRI Is Feasible, Repeatable,and Detects Radiotherapy-induced Change inHypoxia in Xenograft Models and in Patients withNon–small Cell Lung CancerAhmedSalem1,2,3, RossA. Little2, Ayse Latif4, AdamK. Featherstone2,MuhammadBabur4,Isabel Peset5, Susan Cheung2, Yvonne Watson2, Victoria Tessyman4, Hitesh Mistry1,4,Garry Ashton6, Caron Behan6, Julian C. Matthews7, Marie-Claude Asselin2,Robert G. Bristow1,3, Alan Jackson2, Geoff J.M. Parker2,8, Corinne Faivre-Finn1,3,Kaye J.Williams4, and James P.B. O'Connor1,9

Abstract

Purpose: Hypoxia is associated with poor prognosis and ispredictive of poor response to cancer treatments, includingradiotherapy. Developing noninvasive biomarkers that bothdetect hypoxia prior to treatment and track change in tumorhypoxia following treatment is required urgently.

Experimental Design: We evaluated the ability of oxygen-enhanced MRI (OE-MRI) to map and quantify therapy-induced changes in tumor hypoxia by measuring oxygen-refractory signals in perfused tissue (perfused Oxy-R). Clinicalfirst-in-human study in patients with non–small cell lungcancer (NSCLC) was performed alongside preclinical experi-ments in two xenograft tumors (Calu6NSCLCmodel andU87glioma model).

Results: MRI perfused Oxy-R tumor fraction measurementof hypoxia was validated with ex vivo tissue pathology in bothxenograft models. Calu6 and U87 experiments showed thatMRI perfused Oxy-R tumor volume was reduced relative to

control following single fraction 10-Gy radiation and frac-tionated chemoradiotherapy (P < 0.001) due to bothimproved perfusion and reduced oxygen consumption rate.Next, evaluation of 23 patients with NSCLC showed that OE-MRI was clinically feasible and that tumor perfused Oxy-Rvolume is repeatable [interclass correlation coefficient: 0.961(95% CI, 0.858–0.990); coefficient of variation: 25.880%].Group-wise perfused Oxy-R volume was reduced at 14 daysfollowing start of radiotherapy (P ¼ 0.015). OE-MRI detectedbetween-subject variation in hypoxia modification in bothxenograft and patient tumors.

Conclusions: These findings support applying OE-MRIbiomarkers to monitor hypoxia modification, to stratifypatients in clinical trials of hypoxia-modifying therapies, toidentify patients with hypoxic tumors that may fail treatmentwith immunotherapy, and to guide adaptive radiotherapy bymapping regional hypoxia.

IntroductionMost solid tumors contain subregions of hypoxia. The presence

and degree of hypoxia has long been recognized as an importantnegative prognostic factor in cancer (1, 2). Furthermore, hypoxiapredicts poor outcome following surgery (3), radiotherapy (4),and chemotherapy (5), and is associated with relapse and pro-gression (6). Therefore, patients with highly hypoxic tumors tendto have decreased overall cancer survival following conventionaltherapies (7). Recently, there is evidence that targeted therapiessuch as immunotherapy may be less effective in hypoxictumors (8, 9). For these reasons, there is an unmet need to targethypoxia to improve cancer outcomes (10).

Positron emission tomography (PET) methods have shownthat imaging can provide serial noninvasive in vivo measure-ment of tumor hypoxia, can track hypoxia modification, andhas potential to stratify patients and personalize their therapy(11–13). However, hypoxia PET approaches are limited currentlyto specialist centers, hindering widespread use in clinical trialsand adoption into clinical practice (14). Proton MRI is usedwidely in the clinic, making it an attractive alternative to PET for

1Division of Cancer Sciences, University of Manchester, Manchester,United Kingdom. 2Division of Informatics, Imaging & Data Sciences, Uni-versity of Manchester, Manchester, United Kingdom. 3Department ofClinical Oncology, The Christie Hospital NHS Trust, Manchester, UnitedKingdom. 4Division of Pharmacy, University of Manchester, Manchester,United Kingdom. 5Imaging and Flow Cytometry, Cancer Research UKManchester Institute, Manchester, United Kingdom. 6Histology, CancerResearch UK Manchester Institute, Manchester, United Kingdom. 7Divisionof Neuroscience and Experimental Psychology, University of Manchester,Manchester, United Kingdom. 8Bioxydyn Limited, Manchester, UnitedKingdom. 9Department of Radiology, The Christie Hospital NHS Trust,Manchester, United Kingdom.

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

K.J. Williams and J.P.B. O0Connor contributed equally to this article.

Corresponding Author: James P.B. O0Connor, University of Manchester,Oglesby Cancer Research Building, Wilmslow Road, Manchester M20 4GJ,United Kingdom. Phone: 4416-1446-3324; Fax: 4416-1275-5145; E-mail:James.O'[email protected]

Clin Cancer Res 2019;25:3818–29

doi: 10.1158/1078-0432.CCR-18-3932

�2019 American Association for Cancer Research.

ClinicalCancerResearch

Clin Cancer Res; 25(13) July 1, 20193818

on August 5, 2021. © 2019 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst May 3, 2019; DOI: 10.1158/1078-0432.CCR-18-3932

http://crossmark.crossref.org/dialog/?doi=10.1158/1078-0432.CCR-18-3932&domain=pdf&date_stamp=2019-6-11

http://clincancerres.aacrjournals.org/

measuring oxygen delivery and hypoxia in tumors. Oxygen-enhanced MRI (OE-MRI) measures the change in longitudinalrelaxation rate of tissue protons (R1; ref. 15). Preclinical studieshave shown that the OE-MRI biomarker "perfused Oxy-R" dis-tinguishes hypoxic tumor regions from well-oxygenated regions,where analysis is performed in perfused tumor subregions[identified by dynamic contrast-enhanced MRI (DCE-MRI);ref. 16].

We hypothesized that perfused Oxy-R could spatially mapand quantify changes in tumor hypoxia induced by radiother-apy and chemoradiotherapy. To test this, we evaluated changesin perfused Oxy-R induced following radiotherapy or chemor-adiotherapy in two xenograft models. We validated our find-ings with ex vivo IHC and oxygen consumption rate (OCR)assays. We then performed a first-in-human study in patientswith non–small cell lung cancer (NSCLC) to demonstratefeasibility and repeatability and to see whether equivalentfindings were observed. Finally, we investigated whether per-fused Oxy-R could detect intersubject heterogeneity in hypoxiamodification between patients to support a potential role forOE-MRI in personalized medicine.

Materials and MethodsPreclinical study design

The studies complied with UK guidelines on animal welfare incancer research (17) and received approval from the AnimalWelfare and Ethical Review Body. Preclinical experiments wereperformed in Calu-6 NSCLC xenografts and in U87 gliomaxenografts.

Tumors were propagated by injecting either 0.1 mL of Calu-6NSCLC cells (2 � 107 cells/mL) or 0.1 mL of U87 cells (5 �106 cells/mL) intradermally in the lower back of CD1 nude mice.When tumors reached >200 mm3 by caliper measurement, micewere entered into the study (assigned as day 0). They were imagedwhile anesthetized using 2% isoflurane carried initially inmedical

air (21% oxygen), before switching the 100% oxygen as part ofthe OE-MRI protocol. Core temperature was controlled at 36�C.

After initial imaging (day 0), mice were randomized to pathol-ogy validation only, control/sham, or given treatment withtumor-localized radiotherapy. This was administered using ametal-ceramic MXR-320/36 X-ray machine (320 kV, Comet AG)under ambient conditions to restrained, nonanesthetized miceheld in a lead-shielded support perpendicular to the source.Irradiation was delivered at a dose rate of 0.75 Gy/minute. Micewere turned around halfway through the procedure to ensureuniform tumor dose.

Preclinical MRI data acquisition and analysisFor Calu-6 xenograft MRI experiments (Supplementary

Fig. S1A), groups were:

(a) MRI pathology validation (N ¼ 7) with single MRI scan,followed by tumor harvest;

(b) Treatment effect: control group (N ¼ 9); treatment withsingle 10 Gy fraction of radiotherapy (N ¼ 9); treatmentwith fractionated 5 � daily 2 Gy radiotherapy with concur-rent cisplatin on day 0; chemoradiotherapy (N ¼ 6). InitialMRI was at day 0 after which therapy was administered.Subsequent MRI was performed at days 3, 6, and 10 in allgroups and then at day 14 (control), day 18 and 24 (radio-therapy), and day 18 (chemoradiotherapy).

For U87 xenograft MRI experiments (Supplementary Fig. S1B),groups were:

(a) MRI pathology validation (N ¼ 10) with single MRI scan,followed by tumor harvest;

(b) Treatment effect: control group (N ¼ 10); treatment withsingle 10 Gy fraction of radiotherapy (N ¼ 13). Initial MRIwas at day 0 after which therapy was administered. Subse-quent MRI was performed at day 3 only (control) or at days3, 6, 10, and 24 (radiotherapy).

Imaging was performed using a volume transceiver coil on a7 T Magnex instrument interfaced to a Bruker Avance III consoleand gradient system. After initial localizer and T2-weightedanatomy scans, mice underwent coronal imaging. OE-MRIconsisted of a variable flip angle (VFA) spoiled gradient echo(SPGR) acquisition to calculate native tissue R1 (1/T1; unit/s;TR ¼ 30 ms; TE ¼ 1.44 ms; a ¼ 5�/10�/20�). This was followedby 42 dynamic T1-weighted spoiled gradient recalled acquisi-tions in the steady state (SPGR; TR ¼ 30 ms; TE ¼ 1.44 ms; a ¼20�; temporal resolution: 28.8 seconds). After 18 acquisitions,gas delivered through a nose cone was switched to 100%oxygen. Details of preclinical DCE-MRI acquisition are describ-ed in Supplementary Methods.

Regions of interest (ROI) were defined by a research radiogra-pher (Y. Watson, 18 years' experience) using Jim 7 (XinapseSystems). For OE-MRI, voxel-wise DR1 were calculated for eachvoxel, where DR1¼ R1 while breathing oxygen (last 18 of 24 timepoints on 100% oxygen) minus R1 on breathing air (18 timepoints on medical air). Tumor average DR1 was calculated asthe median value. Voxels were classified as oxygen-enhancing(Oxy-E) if the DR1 was positive and significant (one-sided pairedsample t test between preoxygen and oxygen-breathing timepoints, P < 0.05). All other voxels were defined as refractory tooxygen challenge (Oxy-R).

Translational Relevance

Hypoxia is a negative prognostic indicator andpredicts pooroutcome for cancer treatments including radiotherapy, che-motherapy, and immunotherapy. Currently, no validated testsare routinely available to detect and track tumor hypoxia toguide clinical decision-making. Here, we report the first evi-dence to support using oxygen-enhanced MRI (OE-MRI) forthis role. Preclinical experiments in two xenograft modelsdemonstrated that the combined OE-MRI and dynamic con-trast–enhanced MRI (DCE-MRI) biomarker perfused Oxy-Rvolume can identify, map, and quantify (chemo)radiothera-py-induced tumor hypoxia change. We then translated thisimaging method into the clinic demonstrating feasibility andthat perfused Oxy-R volume is repeatable and can detectchemoradiotherapy-induced hypoxia changes in patients withnon–small cell lung cancer. Our data support applying OE-MRI to monitor hypoxia modification, to stratify patients inhypoxia-targeted drug trials, to identify patients with hypoxictumors that may fail anticancer treatment, such as immuno-therapy, and to guide adaptive radiotherapy dose painting bymapping regional hypoxia.

OE-MRI Detects Radiotherapy-induced Hypoxia Modification

www.aacrjournals.org Clin Cancer Res; 25(13) July 1, 2019 3819




DCE-MRI was then used to classify tumor voxels as nonper-fused [where initial area under the Gd contrast agent concentra-tion–time curve at 60 seconds (IAUC60; units mmol/L) was �0)]or perfused (where IAUC60 >0). Together, these coregisteredvoxel-wise OE-MRI and DCE-MRI data were combined to distin-guish perfusedOxy-E (normoxia), perfusedOxy-R (hypoxia), andnonperfused (necrosis) voxels, based on previously publishedmethods (ref. 18; Supplementary Fig. S2A).

Xenograft pathology analysisFor Calu-6 xenografts, three datasets were used. The first

pathology experiment compared the hypoxic fractions measuredon an IHC image from the center of each tumor with a single MRIslice acquired in the same plane from the corresponding tumorregion, to validate OE-MRI measurements of hypoxia (Supple-mentary Fig. S1A). The next two pathology experiments evaluatedchange in hypoxia following therapy. We evaluated tumor size–matched data obtained at tumor harvest from the mice thatunderwent the MRI study (details above; Supplementary Fig.S1A) and separate time-matched data acquired in nonimagedmice at day 10, guided by the MRI experiment. This additionalexperiment had control (N ¼ 9), and tumors treated with single10 Gy fraction of radiotherapy (N ¼ 6; Supplementary Fig. S1C).For U87 xenografts, size-matched data were obtained at tumorharvest from the mice that underwent MRI (details above; Sup-plementary Fig. S1B). Pimonidazole data acquisition and analysisdetails are described in Supplementary Methods.

Xenograft oxygen consumption rate analysisA fourth Calu-6 xenograft cohort was used separate to those

used for MRI and pathology analysis. Tissue biopsy preparationprotocols were adapted from a previously publishedmethod (19)and are described in Supplementary Methods. Separate mousecohorts were used: Calu-6 xenograft groups were time-matchedcontrol (N ¼ 7), size-matched control (N ¼ 5), and treated withsingle 10 Gy fraction of radiotherapy (N ¼ 6) (SupplementaryFig. S1D). Five consecutive basal OCR measurements were per-formed for all samples.

Because previouswork (19) has shown that percentage viabilityand necrotic fraction of tumors is directly proportional toobserved oxygen consumption, we used hematoxylin and eosin(H&E) staining to quantify necrotic fraction for each sample usinga scoringmethodbased on the level of nuclear staining. FollowingOCR measurements, the samples were imaged using OxfordOptronix GelCount (Oxford Optronix Ltd.). Necrotic areas werecalculated using ImageJ software. Whole-field H&E images, rep-resentative biopsy sections, and the scoring method devised toquantify necrosis are shown in Supplementary Fig. S3.

Clinical study design, patient population, and treatmentPatients with NSCLC eligible for curative-intent radiotherapy

alone or combined chemoradiotherapy (either concurrent orsequential) were recruited prospectively from The Christie NHSFoundation Trust following local research and development andinstitutional review board approval (reference: 15/NW/0264,CPMS ID: 18870). This study was conducted in accordance withthe Declaration of Helsinki.

All patients were �18 years old, had histologically or cytolog-ically confirmedNSCLC, and provided written informed consent.Eligible participants also had Eastern Cooperative OncologyGroup (ECOG) performance score of �2, serum creatinine

<120 mmol/L, or calculated creatinine clearance (Cock-croft�Gault) �30 mL/minute. Patients with distant metastasiswere included only if eligible for curative-intent therapy. Patientswith contraindications to MRI were excluded. Coexistent chronicobstructive pulmonary disease was allowed, but patients wererequired to have adequate lung function as part of standardradiotherapy workup (forced expiratory volume in 1 secondgreater than 1 liter, or >40% predicted value).

We examined safety (defined as no adverse events reported ordetected on clinical examination) and tolerability (defined ascompletion of the imaging protocol) in a development cohort(N ¼ 6 patients). On the basis of these data, the study recruitedan expansion cohort to evaluate MRI biomarker repeatabilityand sensitivity to treatment effect (N ¼ 17 patients). The studydesign is summarized in Supplementary Fig. S4.

Radiotherapy planning was performed using three-dimension-al or four-dimensional CT. Treatment was delivered with inten-sity-modulated radiotherapy on a linear accelerator (55 Gy in20 daily fractions or 60–66 Gy in 30–33 daily fractions). Patientsreceiving concurrent chemoradiotherapy had cisplatin and etopo-side administered. Patients receiving sequential chemoradio-therapy had cisplatin or carboplatin administered with eithergemcitabine (if squamous cell carcinoma) or pemetrexed (ifadenocarcinoma). In sequential chemoradiotherapy-treatedpatients, repeat imaging was performed after completion ofchemotherapy. Toxicity was assessed by a clinical oncologist(A. Salem; 8 years' experience) prior to and after each scan. Nopredefined standardized follow-up diagnostic scans were man-dated after completion of the research imaging protocol.

Clinical MRI data acquisition and analysisAll MRI data were acquired free breathing on a 1.5 T whole

body scanner (Philips Achieva, Philips Medical Systems) usingthe Q Body (OE-MRI) and Sense XL Torso (DCE-MRI) coils.After initial localizer and anatomic scans, patients underwentOE-MRI and DCE-MRI. Scan duration was approximately45 minutes (summarized in Supplementary Fig. S5A). Allsequences were colocalized and acquired in the coronal planewith FOV 450 mm � 450 mm � 205 mm; and 5-mm thickslices. Anatomic scans' in-plane resolution was 1.76 mm �1.76 mm, whereas OE-MRI and DCE-MRI images had in-planeresolution of 4.69 mm � 4.69 mm.

For OE-MRI, R1 was calculated using a series of coronal 3Dsingle-shot inversion recovery–prepared SPGR acquisitions(TR ¼ 2.1 ms; TE ¼ 0.50 ms; a ¼ 6�, number of excitations ¼5, TI: 10, 50, 300, 1,100, 2,000, 5,000ms). Dynamic images wereacquired at TI of 1,100 ms. For the first 6 patients in the protocoldevelopment cohort, 12 measurements were acquired whilebreathing medical air (21% oxygen) to determine native R1 priorto oxygen challenge. This was increased to 18 for all subsequentpatients following an interim study review. Measurements onair were followed, in all patients, by 48 measurements after gasswitch to 100% oxygen (flow rate: 15 L/minute) and a final30 following switch back to medical air (temporal resolution:10 seconds). Gases were delivered using a tight-sealed, non-rebreathing Intersurgical EcoLite Hudson facemask (IntersurgicalLtd) via a gas blender, allowing switching between gases. Atwo-step quality assurance procedure was applied to ensuresuccessful delivery of the gas challenge. First, 100% oxygenchallenge delivery was confirmed in all scans at the time of imageacquisition via a gas-sensing probe placed inside the facemask

Salem et al.

Clin Cancer Res; 25(13) July 1, 2019 Clinical Cancer Research3820




and connected to a gas analyzer (ADInstruments Pty Ltd; Sup-plementary Fig. S5B). Second, R1 wasmeasured in the descendingthoracic aorta, outlined by a clinical oncologist (A. Salem), toprovide a positive control by detecting indirect oxygen inputfunction to the tumor. Details of clinical DCE-MRI acquisitionare described in Supplementary Methods. Representative prepro-cessed OE-MRI and DCE-MRI acquisitions are shown in Supple-mentary Video S1A.

Tumor ROIs were defined by a clinical oncologist (A. Salem)and reviewed by a board-certified clinical radiologist (J.P.B.O0Connor; 14 years' experience), using Jim 7 on coronal pre-and post-Gd T1-weighted images, guided by diagnostic[18F]fluorodeoxyglucose (FDG) PET CT images.

For OE-MRI, whole tumor and voxel-wise DR1 were calculated,where DR1 ¼ R1 while breathing oxygen (mean of last 18 timepoints on 100% oxygen) minus R1 on breathing air (mean ofacquisitions onmedical air). Voxels were classified as Oxy-E if theDR1 was positive and significant (one-sided paired sample t testbetween preoxygen and oxygen-breathing time points, P < 0.05).All other voxels were defined as Oxy-R.

For DCE-MRI, tumor median IAUC60 was also calculated.Voxels were classified as perfused when the area under the Gdcontrast agent concentration curve was >0 (one-sided pairedsample t test, P < 0.005; ref. 20). Voxel-wise OE-MRI andDCE-MRI data were combined to distinguish perfused Oxy-E(normoxia), perfused Oxy-R (hypoxia), and nonperfusedvoxels (necrosis), based on previously published methods andwith an identical approach to that used in the preclinicalexperiments (ref. 18; Supplementary Fig. S2B), using datacorrected for breathing and bulk patient motion (details inSupplementary Methods; Supplementary Video S1B).

Statistical analysisAll statistical analyses were performed in IBM SPSS Statistics

version 22.0 (IBM Corp.). For all preclinical data, tumor growthrate comparisons were assessed using a log-rank test for time todouble tumor volume. Pearson correlation analysis was used toassess relationships between pretreatment MRI parameters andIHC. Two-sample unpaired t tests were performed to assessdifferences in mean subregion fractions and volumes (perfusedOxy-R, perfused Oxy-E, and nonperfused), IHC or OCR assaybetween control and treated groups at specific timepoints ofinterest. CorrespondingP values are reported andwere consideredstatistically significant when P < 0.05 (two-sided). No correctionsfor multiple comparisons were made.

For both preclinical and clinical MRI, changes in perfusedOxy-R volume of >50% were considered significant. Thisthreshold was based on similar studies of other functionalimaging biomarkers, where parameter changes of 40%–50%are considered significant (21). For clinical data, descriptivestatistics were presented using median and 95% confidenceintervals (CI) or SD.

For the clinical study, prestudy sample size calculations werenot performed as this was a first-in-human proof-of-conceptstudy. Single measures two-way mixed effects model absoluteagreement interclass correlation coefficient (ICC) and coeffi-cient of variation (CoV) were calculated for summary OE-MRIparameters and the hypoxia biomarker perfused Oxy-Rbetween repeat acquisitions (together with their respective95% CI). Recommendations (22) that interpret ICC valuesbetween 0.75 and 0.9 and greater than 0.9 as indicative of

good and excellent repeatability, respectively, were used. Inaddition, Bland–Altman analysis was performed to calculatebias and 95% limits of agreement (LoA ¼ 1.96 � SD).

ResultsPerfused Oxy-R identifies and quantifies hypoxic volumes inxenograft models

Calu-6 xenografts (23, 24) and U87 xenografts (25) werechosen for the study because these models contain moderate tohigh levels of hypoxia. We measured the MRI perfused Oxy-Rfraction (Supplementary Fig. S2A) on one central slice of thetumor to provide in vivo quantification of hypoxia. PerfusedOxy-R fraction correlated significantly with the hypoxic frac-tion measured by pimonidazole IHC, for both Calu-6 xeno-grafts (R2 ¼ 0.700; P ¼ 0.019; Fig. 1A) and U87 xenografts(R2 ¼ 0.447; P ¼ 0.035; Fig. 1B). No relationship was detectedbetween pimonidazole IHC measurement of hypoxia andIAUC60 (Supplementary Fig. S6A and S6B). These data providepathologic validation that the OE-MRI biomarker perfusedOxy-R fraction, but not DCE-MRI–derived IAUC60, quantifiestumor hypoxia in Calu-6 and U87 xenografts.

Data were obtained across a size range that was representa-tive of all subsequent experiments (tumor sizes: 191–974 mm3

measured by MRI volumetrics). No significant relationship wasseen between pimonidazole IHC measurement of hypoxia andtumor size (representative images in Fig. 1C and D; Supple-mentary Fig. S6C and S6D). This highlights that tumor size andhypoxia are independent of one another in both xenograftmodels used in this study.

Perfused Oxy-R detects radiotherapy-induced hypoxiamodification in xenograft models

Calu-6 xenografts treated with either radiotherapy (single frac-tion of 10 Gy) or chemoradiotherapy (5 � 2 Gy fractions pluscisplatin) exhibited significant growth delay relative to controlxenografts (P < 0.001), assessed by measuring time to double intumor volume.

To understand the dependence of three-dimensional growthinhibition on tumor hypoxia, we measured change in perfusedOxy-R volume within each tumor. Because growth inhibitionwas clearly apparent by day 10 (Fig. 2A), we assessed OE-MRI atthis time point. By day 10, perfused Oxy-R volume was reducedin xenografts treated with radiotherapy (P ¼ 0.029) or fraction-ated chemoradiotherapy (P ¼ 0.047), relative to control(Fig. 2B). Hypoxia modification persisted until tumor harvest(chemoradiotherapy group at day 18; radiotherapy group at day24; both P < 0.001). OE-MRI maps showed spatially coherentchanges in hypoxia (Fig. 2C). In distinction, median change inDR1, a commonly reported biomarker in OE-MRI (26), showedborderline significance only at day 6 and 10 (SupplementaryFig. S7).

We analyzed pimonidazole IHC performed at day 10 in aseparate cohort of Calu-6 xenografts. Lower hypoxic fraction wasseen in tumors treated with radiotherapy (P ¼ 0.026), relative totime-matched controls (Fig. 2D). Next, we performed pimonida-zole IHC analysis of the xenografts undergoing MRI. Lowerhypoxic fractions in radiotherapy (P ¼ 0.042) and chemora-diotherapy (P ¼ 0.041) were found in treated tumors, relativeto size-matched control (Fig. 2E and F; Supplementary Fig. S8).






Experiments were repeated in U87 tumors. Xenografts trea-ted with single fraction of 10 Gy radiotherapy exhibitedsignificant growth delay, relative to control xenografts (P <0.001; Supplementary Fig. S9A). Perfused Oxy-R volume wasdecreased in radiotherapy treated (P ¼ 0.017) xenografts,relative to control by day 3 and this reduction persisted untilday 10 (Supplementary Fig. S9B). Sample OE-MRI maps areshown (Supplementary Fig. S9C). Pimonidazole IHC con-firmed that lower hypoxic fractions were observed in radio-therapy-treated tumors (P ¼ 0.002), relative to size-matchedcontrol (Supplementary Fig. S9D). Collectively, these dataprovide pathologic validation that the OE-MRI biomarkerperfused Oxy-R volume detects hypoxia modification inducedby radiotherapy in two xenograft models.

Hypoxia modification detected by perfused Oxy-R is due toalterations in blood flow and oxygen consumption

To investigate the mechanistic basis for the reduction inhypoxia following treatment, we examined dynamic changein tumor vascular status measured on DCE-MRI from day 0 totumor harvest and also single measurement of OCR at day 10.DCE-MRI data showed that IAUC60 was increased significantlyat days 6–10 in Calu-6 xenografts in both the radiotherapy-and chemoradiotherapy-treated groups, relative to control(P < 0.001). These changes also persisted to day 24 in thegroup treated with single 10 Gy radiotherapy (Fig. 3A and B).Equivalent data were observed in U87 xenografts (Supplemen-tary Fig. S10).

Fresh biopsy samples were taken from Calu-6 xenografts inthree groups: control time-matched tumors (average volume637 � 52 mm3) at day 10 after sham treatment; control size-matched tumors (average volume 242 � 27 mm3); and radio-therapy-treated tumors at day 10 post single 10 Gy fraction

(average volume 271 � 15 mm3). The mean of the five consecu-tive basal OCR measurements performed over 30 minutes underambient conditions was significantly reduced by radiotherapycompared with time- (P ¼ 0.008) and size-matched (P < 0.001)controls (Fig. 3C). Necrotic scoring revealed no significant differ-ences between the irradiated and time- or size-matched controls(Fig. 3D), indicating equivalent levels of tissue viability betweenthe three groups.

Large intra- and intertumor heterogeneity of OCR measure-ments were observed within both time- and size-matched con-trol Calu-6 xenografts. Both types of heterogeneity were marked-ly reduced in tumors treated with radiotherapy (Fig. 3E). In all,31.3% of control biopsy samples at day 10 (time-matched) hadOCR of >25 pmol/minute per normalized unit and thisincreased to 59.4% in size-matched control biopsy samples. Indistinction, only 13.0% of radiotherapy-treated biopsy sampleshad residual OCR of >25 pmol/minute per normalized unit.These data provide evidence that the hypoxia modificationdetected by OE-MRI and pimonidazole IHC in Calu-6 is likelydue to reduced overall OCR, in particular removing thosesubregions with very high localized OCR.

Perfused Oxy-R is feasible, well-tolerated, and repeatable inpatients with NSCLC

Twenty-three patients with stage I–IV NSCLC were recruited(Supplementary Fig. S4; Supplementary Table S1). The pro-tocol was safe and well-tolerated (Supplementary Table S2).Significant changes in R1 were observed following oxygeninhalation in the aorta of all patients demonstrating tech-nique feasibility and providing quality control (Supplemen-tary Fig. S11).

Whole tumor median R1 increased with oxygen inhalation in11 of 15 patients (individual all P < 0.05 for 11 patients, oxygen

0

10

20

30

40

50

0 10 20 30 40 500

10

20

30

40

50

0 10 20 30 40 50

A Calu-6 U87B

% Hypoxia (Perfused Oxy-R fraction)

% Hypoxia (Perfused Oxy-R fraction)

% H

ypox

ia (P

imon

idaz

ole)

% H

ypox

ia (P

imon

idaz

ole)

C5 mm

IHC

imag

eM

RI m

ap

MR

I map

5 mm

IHC

imag

e

D

Perfused Oxy-E (normoxia) Perfused Oxy-R (hypoxia) Nonperfused

Figure 1.

The MRI perfused Oxy-R fractioncorrelated with pimonidazoleIHC-derived hypoxic fraction inboth untreated Calu-6 tumors (A)and (B) U87 tumors. Examplewhole-field pimonidazole IHC andMRI perfused Oxy-Rmaps areshown for the least hypoxic, middle,and most hypoxic Calu-6 tumors(C) and U87 tumors (D).

Salem et al.





inhalation versus air). However, all tumors demonstrated somedegree of intratumor spatial heterogeneity in oxygen-inducedDR1, with three patterns of tissue classification revealed. In 3 of15 tumors, the whole tumor median DR1 was significant and hadsimilar temporal evolution and magnitude as the DR1 seen in theaorta. These tumors had no measurable regional hypoxia

(absence of perfused Oxy-R; representative example in Fig. 4A).The remaining 12 of 15 tumors had spatially coherent regions ofhypoxia; this included 8 of 15 tumors with an overall significantDR1 that was partially attenuated compared with aorta DR1

(representative example in Fig. 4B), and 4 of 15 tumors wherethe proportion of hypoxic tissue was high enough to substantially

0

60

120

180

240

0 4 8 12 16 20 24

B

MR

I hyp

oxic

tum

or v

olum

e (m

m3 )

Days

Control Single 10 Gy5x 2 Gy + cisplatin

**

*

C Day 0 End

5x 2

Gy

+ C

ispl

atin

Con

trol

Sing

le

10 G

y

Perfused Oxy-E (normoxia)Perfused Oxy-R (hypoxia)Nonperfused

5x 2 Gy + cisplatin

A Control Single 10 Gy

Tum

or v

olum

e (m

m3 )

250

350

450

550

650

750

0 4 8 12 16 20 24Days

** **

% H

ypox

ia (P

imon

idaz

ole)

F5x

2 G

y +

5x 2

Gy +

10 G

y10

Gy

Cis

plat

inC

ontr

olSi

ngle

10

Gy

D

* **E

Time-matched day 10 Size-matched Size-matched

% H

ypox

ia (P

imon

idaz

ole)

0

10

20

30

40

0

10

20

30

40

Figure 2.

A, Radiotherapy induced significantgrowth delay in Calu-6 xenograftstreated with single fraction 10 Gy andfractionated 5� 2 Gy withconcurrent cisplatin, relative tocontrol. Tumor measurements wereby calipers, with day 0 being the firstday of MRI. B,Mean perfused Oxy-Rvolume was reduced in bothradiotherapy-treated groups at day10, relative to control tumors, and thisreduction persisted until day 18(chemoradiotherapy) or day 24(radiotherapy). C, Example maps ofperfused Oxy-E (normoxic tumor),perfused Oxy-R (hypoxic tumor), andnonperfused tumor for eachtreatment group.D, Time-matcheddata from pimonidazole adductformation IHC staining obtained atday 10 in an independent cohortshowed relatively lower hypoxicfraction in radiotherapy-treatedCalu-6 xenografts relative to control.E, Pimonidazole adduct formationIHC data acquired from thexenografts imaged with MRI-confirmed persistent reduction inhypoxic fraction at size-matchedtumor harvest, with correspondingsample images (magnification20�; F; �� , P < 0.001; � , P < 0.05).






attenuate the median whole tumor DR1 so that it was not signif-icant (representative example in Fig. 4C).

Ten patients underwent repeat MRI (within 7 � 5 days; Sup-plementary Fig. S4) before radiotherapy was administered tomeasure biomarker precision. The perfused Oxy-R volume, mea-suring tumor hypoxic volume, demonstrated excellent repeatabil-itywith ICCof 0.961 (95%CI, 0.858–0.990) andCoVof 25.880%(Fig. 4D). Additional repeatability results can be found in Sup-plementary Table S3.

Consistent classification of tumors as either entirely nor-moxic (n ¼ 3; perfused Oxy-R volume ¼ 0) or containing somehypoxia (n ¼ 7; perfused Oxy-R volume >0) was concordantbetween the two preradiotherapy scans. Finally, visual inspec-tion revealed that MRI hypoxia mapping was spatially repeat-able in tumor, nodal, and distant metastatic lesions across arange of tumor and hypoxic volumes (Fig. 4E). Collectively,these data demonstrate that OE-MRI can identify and maphypoxia in clinical NSCLC tumors, when subregional tissueclassification is performed using perfusion data.

Perfused Oxy-R detects therapy-induced changes in hypoxia inpatients with NSCLC

Twelve patients were imaged at day 14 � 4 of radiotherapy,in addition to pretreatment imaging (Supplementary Fig. S4).No significant change was detected in tumor volume at thistime point (P ¼ 0.159; Table 1), but we hypothesized thatradiation would induce measurable changes in hypoxia withinthis window, based on previous clinical PET studies in headand neck cancer and lung cancer (11–13).

The perfused Oxy-R volume, indicating hypoxic tumorvolume, decreased in the patient cohort from 4.16 cm3

(95% CI, 0–10.6 cm3) at baseline to 3.23 cm3 (95% CI, 0–9.41 cm3) at mid-treatment (P ¼ 0.015). In distinction, theincrease in median DR1 at day 14 was not significant (P ¼0.097). MRI parameter changes are summarized in Table 1.These data show that the MRI biomarker perfused Oxy-Rvolume detected reduction in tumor hypoxia in patients withNSCLC consistent with data in the two xenograft models ofcancer.

-80

0

80

160

240

0 4 8 12 16 20 24

BDays 0 3 6 10 14 18 24

s.mmol/L

20

0

5x 2

Gy

+ C

ispl

atin

Con

trol

Sing

le 1

0 G

y

Time-m

atched

(day 10

) contro

l

Size-m

atched

contro

l Single

10 G

y

Time-m

atched

(day 10

) contro

l

Size-m

atched

contro

l Single

10 G

y

Time-m

atched

(day 10

) contro

l

Size-m

atched

contro

l Single

10 G

y

A5x 2 Gy + cisplatin

Control Single 10 Gy%

Cha

nge

IAU

C60

Days

**

OC

R (p

mol

/min

/nor

m. u

nit)

EC*

D

Nec

rotic

sco

re

*

0

1

2

3

0

10

20

30

40 80

60

40

20

0

-20

**

**

OC

R (p

mol

/min

/nor

m. u

nit) **

Figure 3.

A, Oxygen delivery (indicated by the perfusion biomarker IAUC60) was increased significantly at days 6–10 in Calu-6 xenografts treated with eitherradiotherapy (RT) or chemoradiotherapy (CRT), relative to control. Increased perfusion persisted until day 24 in the RT treated group. B,Representative maps in one xenograft tumor from each cohort. C, Oxygen consumption rate (OCR) was significantly reduced in radiotherapy-treatedCalu-6 tumors relative to in time-matched and size-matched controls. Mean OCR values of each tumor (6 samples per tumor) over 5 real-timemeasurements are shown. D, There was no difference in biopsy sample necrosis in any of the three groups. E, Significant intratumor and intertumorheterogeneity in OCR was observed within time- and size-matched control Calu-6 xenografts, but this was markedly reduced in tumors treated withradiotherapy. Individual symbols representative different tumors. Error bars, SEM in all panels (�� , P < 0.001; � , P < 0.05).

Salem et al.





32% 68%

41% 27%

31%

100%

Aorta input function Whole tumor ΔR1 SubregionalΔR1 Tumor composition A

B

C

ΔR

1(s

-1)

Time point

ΔR

1(s

-1)

ΔR

1(s

-1)

Time point Time point

ΔR

1(s

-1)

Time point

ΔR

1(s

-1)

ΔR

1(s

-1)


ΔR

1(s

-1)

Time point

ΔR

1(s

-1)

ΔR

1(s

-1)


D E

0.1 1 10 100-20

-10

0

10

20

Average (cm3)

Scan

1 -

scam

2 (c

m3 )

Upper LoA 7.22 0

Lower LoA -1 0.5 00

Primary tumor

Distant metastasis

Nodal metastasis

Perfused Oxy-E (normoxia)

Perfused Oxy-R (hypoxia)

Nonperfused

Figure 4.

Three distinct patterns of tissue classification were seen on MRI. In each row, aortic input functions and whole tumor OE-MRI DR1 were derived from the samepatient. A, In 3 of 15 tumors, whole tumor median OE-MRI DR1 had similar temporal evolution andmagnitude as aortic DR1; these tumors had nomeasurableregional hypoxia (absence of perfused Oxy-R). Representative tumor displaying this distinct pattern of tissue classification is shown. B, In 8 of 15 tumors wholetumor median OE-MRI DR1, was partially attenuated by hypoxic regions, but was still significant. Representative tumor displaying this distinct pattern of tissueclassification is shown. C, In 4 of 15 tumors, the extent of hypoxia substantially attenuated the whole tumor DR1, which was not significantly different from zerochange. Representative tumor displaying this distinct pattern of tissue classification is shown. D, Bland–Altman plot for perfused Oxy-R volume with upper andlower limit of agreement (LoA). Tumor, nodal, and distant metastatic lesions are indicated. E,MRI mapping of tumor hypoxia was spatially repeatable to varyingextents in tumor, nodal, and distant metastatic lesions across a range of tumor and hypoxic volumes.






Perfused Oxy-R measurement of hypoxia has implications forpersonalized therapy

Previous PET studies have reported variable hypoxia modi-fication in patients with NSCLC and other cancer types(11, 27, 28). We examined the variation in hypoxiamodificationin Calu-6 xenografts at day 10. Those xenografts with >50%change in hypoxic volume were designated as exhibiting signif-icant hypoxia modification. While radiotherapy-treated tumorsshowed overall reduction in hypoxia at the cohort level, relativeto control, a >50% decrease in hypoxic volume was seen in only9 of 15 tumors (Fig. 5A). These "hypoxia-modified tumors" hadhigher perfusion and permeability (denoted by median DCE-MRI IAUC60) at baseline than tumors that did not demonstratehypoxia modification (P ¼ 0.035; Fig. 5B).

Similarly, variation was seen in response to (chemo)radiother-apy in the clinical tumors,with 8of 12 tumors having adecrease inhypoxic volume >50%. In distinction, two tumors had an increasein hypoxic volume >50% and two tumors showed no changeabove the 50% threshold (Fig. 5C). Tumors with significantreduction in hypoxia had higher median IAUC60 (P ¼ 0.011) atbaseline than other tumors (Fig. 5D). There was no difference inbaseline tumor size or hypoxic volume between the hypoxia"modified" and nonmodified tumors. These data show thatOE-MRI can distinguish tumors that demonstrate hypoxia mod-ification following radiotherapy from those that have persistenthypoxia.

DiscussionThere is an unmet need to develop noninvasive biomarkers of

tumor hypoxia to monitor response for anticancer treatments.This is particularly important for radiotherapy and chemora-diotherapy, as meta-analysis has identified the negative impactof hypoxia detected via imaging, on radiotherapy outcome (29).

Imaging enables repeated whole tumor sampling that over-comes the limitations of tissue-based hypoxia quantification(subsampling and single measurement) and biofluid assays(inability to distinguish heterogeneity between different tumorsin an individual). Therefore, translational imaging tests couldenable patient selection and stratification in trials of combinedradiotherapy and hypoxia-targeted therapies (10), to adapt radio-therapy dose intensification, and to monitor emergence of radio-resistant hypoxic cancer cells (14).

Here, we report the first-in-human evidence that perfused Oxy-R can identify, map, and quantify change in hypoxia induced by atherapeutic intervention. We chose to evaluate patients withNSCLC as a proof-of-principle study because lung cancer is theleading cause of cancer mortality worldwide (30); around 90% ofcases are of the NSCLC subtype (31, 32); and tumor hypoxia isassociated with poor survival in patients with NSCLC (33). Fur-thermore, radiotherapy plays an important role in the treatmentof all stages of NSCLC (34) and imaging biomarkers are attractive

in this setting due to limited access to tumor tissue material inradiotherapy-treated patients.

Our bench-to-bedside approach began by testing whetherOE-MRI biomarkers could detect radiotherapy-induced changesin tumor biology in two xenograft models of NSCLC (Calu-6)and high-grade glioma (U87). We showed that OE-MRI, com-bined with assessment of perfusion, identified and mappedregional differences in hypoxic and normoxic tumor in bothmodels, using IHC validation (18, 35, 36). Next we showed thatradiotherapy resulted in significant reduction in perfused Oxy-Rvolume in Calu-6 xenografts after 10 days, relative to controltumors, with both high-dose single fraction radiotherapyand a fractionated chemoradiotherapy regimen that moreclosely mimics clinical therapy. These differences persisted for18–24 days, indicating that while regrowth occurred eventuallyin the radiotherapy and chemoradiotherapy-treated tumors,they had less hypoxic tissue than untreated tumors of a similarsize. Our findings were validated by time- and size-matched IHCquantification of hypoxia and quantification of oxygen con-sumption. Confirmatory data were obtained in U87 tumors. Inall studies, an OE-MRI biomarker that was sensitive to spatialheterogeneity (37) between hypoxic and nonhypoxic tumorsubregions, perfused Oxy-R, was most closely related to hypoxiaon pathology validation and was most sensitive to therapeuticmodification of hypoxia.

These data are the first to show that perfused Oxy-R can trackgroup differences in hypoxia modification following therapyin vivo. Mechanistic insight was provided through analysis of theDCE-MRI data acquired during MRI treatment experiments inCalu-6 and U87 xenografts showed that increases in tumorperfusion (measured by IAUC60) were detected with radiation-based therapies, relative to control. Furthermore, ex vivo OCRanalysis in Calu-6 xenografts provided evidence that oxygenconsumption was reduced at day 10 following radiotherapy.These data suggest that both increased oxygen delivery andreduced oxygen consumption contribute to radiation-inducedhypoxia modification.

We then translated the OE-MRI technique to humans, dem-onstrating safety, tolerability, and measurement feasibility inpatients with NSCLC. Measurement precision was evaluated byassessing biomarker repeatability. Perfused Oxy-R (hypoxic) vol-ume had excellent test–retest precision with high ICC and com-pared favorably with previously published studies of MRI bio-marker precision in lung and other tumors (38–41). The intra-tumor spatial distribution of hypoxia was comparable betweenrepeat scans on visual inspection. Lack of absolute visual repeat-ability could be due tomethodologic factors (including imperfectimage registration or inconsistent target volume definition) orbiological factors (including cyclical hypoxia; ref. 14). Further-more, OE-MRI provided consistent classification of tumors ashypoxic or not; in this respect, it outperformed PET imaging withthe tracer [18F]FMISO in patients withNSCLC (42). These data arethe first to demonstrate repeatability of OE-MRI biomarkers thatquantify tumor hypoxia in patients with cancer.

Our clinical data provide the first evidence that OE-MRI cantrack tumor hypoxia modification induced by a therapy inpatients. This finding concurs with previous PET imaging stud-ies in patients with NSCLC and other cancers which haveshown that the volume of hypoxic tumor can increase, decrease,or remain unchanged within the initial few weeks followingchemoradiation (11).

Table 1. Baseline and mid-treatment imaging parameters in patients withNSCLC

Parameter (unit)Baseline(95% CI)

Mid-treatment(95% CI) P

Tumor volume (cm3) 36.7 (9.40–640) 31.7 (5.90–57.6) 0.159Perfused Oxy-R volume (cm3) 4.16 (0–10.60) 3.230 (0–9.410) 0.015DR1 (s

�1) 0.018 (0.013–0.023) 0.025 (0.016–0.033) 0.097

Salem et al.





Finally, because PET studies in patients with head and neckcancer suggest that the persistence of hypoxia during chemor-adiotherapy, rather than pretreatment levels of hypoxia, predictclinical outcome (12, 13), we investigated the potential forOE-MRI to have application in personalized medicine. Wedemonstrated that perfused Oxy-R distinguished those xeno-grafts and NSCLC patient tumors whose hypoxic tumor volumedecreases following radiotherapy-based treatments from those

that remained hypoxic (and in some cases worsened). Of note,hypoxia modification was observed in tumors with relativelyhigh pretreatment vascular perfusion and permeability, mea-sured by IAUC60. This suggests that a multimodal imaging,genomic, and tissue pathology biomarker panel, performedbefore and during treatment, may be required to fully under-stand, predict, and monitor hypoxia modification in the clin-ic (43). However, further prospective powered studies are

0

4

8

12

16

20

-100

-50

0

50

100

% C

hang

e O

xy-R

vol

ume

% C

hang

e O

xy-R

vol

ume

% C

hang

e O

xy-R

vol

ume

-100

-50

0

50

100

6/9 more hypoxic 1/6 more hypoxicA Single 10 Gy 5x 2 Gy +Cisplatin

5/9 less hypoxic 4/6 less hypoxic

Control

≥ ≥

B *

Hypoxia modified Hypoxia modified

Single 10 Gy 5x 2 Gy +Cisplatin

Hypoxia not modified Hypoxia not modified20

0s.mmol/L

Not modified

Modified

2/12 more hypoxic

-100

-50

0

50

100

8/12 less hypoxic

C

0

4

8

12

16

20

Not modified

Med

ian

IAU

C60

(s.m

mol

/L)

Med

ian

IAU

C60

(s.m

mol

/L)

*D

≥

Modified

Figure 5.

A,Waterfall plot data show that perfused Oxy-R volume quantified the variation in hypoxia modification in radiotherapy and fractionated chemoradiotherapy–treated Calu-6 xenografts at day 10 and in controls. B,We found that "modified tumors" had higher perfusion and permeability (denoted by median IAUC60) atbaseline than tumors that did not demonstrate hypoxia modification. Example IAUC60 maps from "modified" and nonmodified tumors shown. C,Waterfall plotdata show that perfused Oxy-R volume also quantified the variation in the 12 patients with NSCLC receiving (chemo)radiotherapy. Overall there was asignificant reduction in hypoxia in this cohort. D, Intertumor heterogeneity was also seen here with "modified tumors" having higher median IAUC60. Error bars,SEM (� , P < 0.05).






required to confirm any role for hypoxia imaging in stratifiedmedicine.

While this study focused on the role of perfused Oxy-R inmeasuring hypoxia changes induced by radiotherapy, these find-ings have broad implications. Perfused Oxy-R has potential tomonitor direct pharmacologic targeting of hypoxia (44) andto inform how resistance to immunotherapy relates to thehypoxic tumor stem cell niche (9). Furthermore, perfused Oxy-R may guide radiotherapy dose intensification (dose painting;refs. 45, 46) and has unique potential for clinical applicationssuch as real-time adaptive radiotherapyonMRLinac systems (47).For each of these applications, large studies are required to qualifythe prognostic value of perfused Oxy-R as a biomarker of hypoxiaand its ability to predict therapy response.

In summary, this study provides substantial new informationto advance clinical translation of the OE-MRI and DCE-MRIbiomarker perfused Oxy-R volume (48) and supports furthereffort to qualify this biomarker for use in clinical trials.

Disclosure of Potential Conflicts of InterestG.J.M. Parker is an employee of and holds ownership interest (including

patents) in Bioxydyn Limited. No potential conflicts of interest were disclosedby the other authors.

Authors' ContributionsConception and design: A. Salem, A. Jackson, G.J.M. Parker, C. Faivre-Finn,K.J. Williams, J.P.B. O0ConnorDevelopment of methodology: A. Salem, R.A. Little, A. Latif, Y. Watson,G.J.M. Parker, C. Faivre-Finn, J.P.B. O0ConnorAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): A. Salem, A. Latif, M. Babur, S. Cheung, Y. Watson,V. Tessyman, G. Ashton, K.J. Williams, J.P.B. O0Connor

Analysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): A. Salem, R.A. Little, A. Latif, A.K. Featherstone,I. Peset, Y. Watson, H. Mistry, J.C. Matthews, R.G. Bristow, C. Faivre-Finn,K.J. Williams, J.P.B. O0ConnorWriting, review, and/or revision of the manuscript: A. Salem, R.A. Little,A. Latif, M. Babur, S. Cheung, V. Tessyman, H. Mistry, J.C. Matthews,M.-C. Asselin, R.G. Bristow, A. Jackson, G.J.M. Parker, C. Faivre-Finn,K.J. Williams, J.P.B. O0ConnorAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): A. Salem, A.K. Featherstone, C. Behan,A. Jackson, J.P.B. O0ConnorStudy supervision: A. Salem, G.J.M. Parker, C. Faivre-Finn, K.J. Williams,J.P.B. O'ConnorOther (supervision of A.K. Featherstone): J.C. Matthews

AcknowledgmentsThis study was funded by the Engineering and Physical Sciences

Research Council and Cancer Research UK (CRUK; grant C8742/A18097; 2013–2018). This study was supported by researchers at theNIHR Manchester Biomedical Research Centre. A. Salem was supportedby an Early Careers Fellowship from Manchester Cancer Research Centreand The Christie Charity and by a University of Manchester PresidentialFellowship for Clinicians. J.P.B. O0Connor is supported by Cancer ResearchUK (C19221/A15267, C19221/A22746). Clinical OE-MRI was support-ed by Philips through a research agreement with The University ofManchester.

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

Received December 7, 2018; revised February 4, 2019; accepted March 14,2019; published first May 3, 2019.

References1. Brizel DM, Sibley GS, Prosnitz LR, Scher RL, Dewhirst MW. Tumor hypoxia

adversely affects the prognosis of carcinoma of the head and neck. Int JRadiat Oncol Biol Phys 1997;38:285–9.

2. Hockel M, Schlenger K, Aral B, Mitze M, Schaffer U, Vaupel P. Associationbetween tumor hypoxia and malignant progression in advanced cancer ofthe uterine cervix. Cancer Res 1996;56:4509–15.

3. Vergis R, Corbishley CM, Norman AR, Bartlett J, Jhavar S, Borre M, et al.Intrinsic markers of tumour hypoxia and angiogenesis in localisedprostate cancer and outcome of radical treatment: a retrospectiveanalysis of two randomised radiotherapy trials and one surgical cohortstudy. Lancet Oncol 2008;9:342–51.

4. Yang L, Taylor J, Eustace A, Irlam JJ, Denley H, Hoskin PJ, et al. A genesignature for selecting benefit from hypoxia modification of radiotherapyfor high-risk bladder cancer patients. Clin Cancer Res 2017;23:4761–8.

5. Holden SA, Teicher BA, Ara G, Herman TS, Coleman CN. Enhancement ofalkylating agent activity by SR-4233 in the FSaIIC murine fibrosarcoma.J Natl Cancer Inst 1992;84:187–93.

6. Overgaard J. Hypoxic radiosensitization: adored and ignored. J Clin Oncol2007;25:4066–74.

7. Vaupel P, Mayer A. Hypoxia in cancer: significance and impact on clinicaloutcome. Cancer Metastasis Rev 2007;26:225–39.

8. Lin SH, Koong AC. Breathing new life into hypoxia-targeted therapies fornon-small cell lung cancer. J Natl Cancer Inst 2018;110:1–2.

9. Taylor CT, Colgan SP. Regulation of immunity and inflammation byhypoxia in immunological niches. Nat Rev Immunol 2017;17:774–85.

10. Wilson WR, Hay MP. Targeting hypoxia in cancer therapy. Nat Rev Cancer2011;11:393–410.

11. Bollineni VR, KooleMJ, Pruim J, BrouwerCL,WiegmanEM,GroenHJ, et al.Dynamics of tumor hypoxia assessed by 18F-FAZA PET/CT in head andneck and lung cancer patients during chemoradiation: possible implica-

tions for radiotherapy treatment planning strategies. Radiother Oncol2014;113:198–203.

12. Lock S, Perrin R, Seidlitz A, Bandurska-Luque A, Zschaeck S, Zophel K, et al.Residual tumour hypoxia in head-and-neck cancer patients undergoingprimary radiochemotherapy, final results of a prospective trial on repeatFMISO-PET imaging. Radiother Oncol 2017;124:533–40.

13. Zips D, Zophel K, Abolmaali N, Perrin R, Abramyuk A, Haase R, et al.Exploratory prospective trial of hypoxia-specific PET imaging during radio-chemotherapy in patients with locally advanced head-and-neck cancer.Radiother Oncol 2012;105:21–8.

14. Salem A, Asselin MC, Reymen B, Jackson A, Lambin P, West CML, et al.Targeting hypoxia to improve non-small cell lung cancer outcome. J NatlCancer Inst 2018;110:14–30.

15. Young IR, Clarke GJ, Bailes DR, Pennock JM, Doyle FH, Bydder GM.Enhancement of relaxation rate with paramagnetic contrast agents in NMRimaging. J Comput Tomogr 1981;5:543–7.

16. DewhirstMW,Birer SR.Oxygen-enhancedMRI is amajor advance in tumorhypoxia imaging. Cancer Res 2016;76:769–72.

17. Workman P, Aboagye EO, Balkwill F, Balmain A, Bruder G, Chaplin DJ,et al. Guidelines for the welfare and use of animals in cancer research. Br JCancer 2010;102:1555–77.

18. O'Connor JP, Boult JK, Jamin Y, Babur M, Finegan KG, Williams KJ, et al.Oxygen-enhanced MRI accurately identifies, quantifies, and maps tumorhypoxia in preclinical cancer models. Cancer Res 2016;76:787–95.

19. Russell S, Wojtkowiak J, Neilson A, Gillies RJ. Metabolic profiling ofhealthy and cancerous tissues in 2D and 3D. Sci Rep 2017;7:15285.

20. Tofts PS, Brix G, Buckley DL, Evelhoch JL, Henderson E, Knopp MV, et al.Estimating kinetic parameters from dynamic contrast-enhanced T(1)-weighted MRI of a diffusable tracer: standardized quantities and symbols.J Magn Reson Imaging 1999;10:223–32.

Salem et al.





21. O'Connor JP, Jackson A, Parker GJ, Roberts C, Jayson GC. Dynamiccontrast-enhanced MRI in clinical trials of antivascular therapies.Nat Rev Clin Oncol 2012;9:167–77.

22. Portney LG, Watkins MP. Foundations of clinical research: applications topractice. Upper Saddle River, NJ: Prentice Hall Health; 2000.

23. Williams KJ, Telfer BA, Shannon AM, Babur M, Stratford IJ, Wedge SR.Combining radiotherapy with AZD2171, a potent inhibitor of vascularendothelial growth factor signaling: pathophysiologic effects and thera-peutic benefit. Mol Cancer Ther 2007;6:599–606.

24. Jiang Y, Verbiest T, Devery AM, Bokobza SM, Weber AM, Leszczynska KB,et al. Hypoxia potentiates the radiation-sensitizing effect of olaparib inhuman non-small cell lung cancer xenografts by contextual syntheticlethality. Int J Radiat Oncol Biol Phys 2016;95:772–81.

25. Joseph JV, Conroy S, Pavlov K, Sontakke P, Tomar T, Eggens-Meijer E, et al.Hypoxia enhancesmigration and invasion in glioblastoma by promoting amesenchymal shift mediated by the HIF1alpha-ZEB1 axis. Cancer Lett2015;359:107–16.

26. Matsumoto K, BernardoM, Subramanian S, Choyke P,Mitchell JB, KrishnaMC, et al. MR assessment of changes of tumor in response to hyperbaricoxygen treatment. Magn Reson Med 2006;56:240–6.

27. KohWJ, Bergman KS, Rasey JS, Peterson LM, Evans ML, GrahamMM, et al.Evaluation of oxygenation status during fractionated radiotherapy inhuman nonsmall cell lung cancers using [F-18]fluoromisonidazole pos-itron emission tomography. Int J Radiat Oncol Biol Phys 1995;33:391–8.

28. Vera P, Bohn P, Edet-Sanson A, Salles A, Hapdey S, Gardin I, et al.Simultaneous positron emission tomography (PET) assessment of metab-olism with (1)(8)F-fluoro-2-deoxy-d-glucose (FDG), proliferation with(1)(8)F-fluoro-thymidine (FLT), and hypoxia with (1)(8)fluoro-misoni-dazole (F-miso) before and during radiotherapy in patients with non-small-cell lung cancer (NSCLC): a pilot study. Radiother Oncol 2011;98:109–16.

29. Horsman MR, Mortensen LS, Petersen JB, Busk M, Overgaard J. Imaginghypoxia to improve radiotherapy outcome. Nat Rev Clin Oncol 2012;9:674–87.

30. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancerstatistics. CA Cancer J Clin 2011;61:69–90.

31. Govindan R, Page N, Morgensztern D, ReadW, Tierney R, Vlahiotis A, et al.Changing epidemiology of small-cell lung cancer in the United States overthe last 30 years: analysis of the surveillance, epidemiologic, and end resultsdatabase. J Clin Oncol 2006;24:4539–44.

32. Riaz SP, Luchtenborg M, Coupland VH, Spicer J, Peake MD, Moller H.Trends in incidence of small cell lung cancer and all lung cancer.Lung Cancer 2012;75:280–4.

33. WangQ,HuDF, Rui Y, Jiang AB, Liu ZL, Huang LN. Prognosis value ofHIF-1alpha expression in patients with non-small cell lung cancer. Gene 2014;541:69–74.

34. Johnson DH, Schiller JH, Bunn PA Jr. Recent clinical advances in lungcancer management. J Clin Oncol 2014;32:973–82.

35. Linnik IV, Scott ML, Holliday KF, Woodhouse N, Waterton JC, O'ConnorJP, et al. Noninvasive tumor hypoxia measurement using magnetic reso-

nance imaging in murine U87 glioma xenografts and in patients withglioblastoma. Magn Reson Med 2014;71:1854–62.

36. Little R, Jamin Y, Boult J, Naish J, Watson Y, Cheung S, et al. Combinedoxygen and gadolinium enhanced MR imaging of hypoxia in renal carci-noma: comparison with susceptibility MR imaging and pathology. Radi-ology 2018;288:739–47.

37. O'Connor JP, Rose CJ, Waterton JC, Carano RA, Parker GJ, Jackson A.Imaging intratumor heterogeneity: role in therapy response, resistance, andclinical outcome. Clin Cancer Res 2015;21:249–57.

38. Morgan B, Utting JF, Higginson A, Thomas AL, Steward WP, Horsfield MA.A simple, reproducible method for monitoring the treatment of tumoursusing dynamic contrast-enhanced MR imaging. Br J Cancer 2006;94:1420–7.

39. Ng CS, Raunig DL, Jackson EF, Ashton EA, Kelcz F, Kim KB, et al. Repro-ducibility of perfusion parameters in dynamic contrast-enhanced MRI oflung and liver tumors: effect on estimates of patient sample size in clinicaltrials and on individual patient responses. AJR Am J Roentgenol 2010;194:W134–40.

40. O'Connor JP, Carano RA, Clamp AR, Ross J, Ho CC, Jackson A, et al.Quantifying antivascular effects of monoclonal antibodies to vascularendothelial growth factor: insights from imaging. Clin Cancer Res 2009;15:6674–82.

41. Weller A, Papoutsaki MV, Waterton JC, Chiti A, Stroobants S, Kuijer J, et al.Diffusion-weighted (DW) MRI in lung cancers: ADC test-retest repeatabil-ity. Eur Radiol 2017;27:4552–62.

42. Grkovski M, Schwartz J, Rimner A, Schoder H, Carlin SD, Zanzonico PB,et al. Reproducibility of 18F-fluoromisonidazole intratumour distributionin non-small cell lung cancer. EJNMMI Res 2016;6:79.

43. Lalonde E, Ishkanian AS, Sykes J, Fraser M, Ross-Adams H, Erho N, et al.Tumour genomic and microenvironmental heterogeneity for integratedprediction of 5-year biochemical recurrence of prostate cancer: a retro-spective cohort study. Lancet Oncol 2014;15:1521–32.

44. Tap WD, Papai Z, Van Tine BA, Attia S, Ganjoo KN, Jones RL, et al.Doxorubicin plus evofosfamide versus doxorubicin alone in locallyadvanced, unresectable or metastatic soft-tissue sarcoma (TH CR-406/SARC021): an international, multicentre, open-label, randomised phase3 trial. Lancet Oncol 2017;18:1089–103.

45. Vera P, Thureau S, Chaumet-Riffaud P,Modzelewski R, Bohn P, VermandelM, et al. Phase II study of a radiotherapy total dose increase in hypoxiclesions identified by 18F-misonidazole PET/CT in patients with non-smallcell lung carcinoma (RTEP5 Study). J Nucl Med 2017;58:1045–53.

46. van ElmptW, De Ruysscher D, van der Salm A, Lakeman A, van der Stoep J,Emans D, et al. The PET-boost randomised phase II dose-escalation trial innon-small cell lung cancer. Radiother Oncol 2012;104:67–71.

47. Bainbridge H, Salem A, Tijssen RHN, Dubec M, Wetscherek A, Van Es C,et al. Magnetic resonance imaging in precision radiation therapy for lungcancer. Transl Lung Cancer Res 2017;6:689–707.

48. O'Connor JP, Aboagye EO, Adams JE, Aerts HJ, Barrington SF, Beer AJ, et al.Imaging biomarker roadmap for cancer studies. Nat Rev Clin Oncol 2017;14:169–86.






2019;25:3818-3829. Published OnlineFirst May 3, 2019.Clin Cancer Res Ahmed Salem, Ross A. Little, Ayse Latif, et al.

small Cell Lung Cancer−in Patients with NonRadiotherapy-induced Change in Hypoxia in Xenograft Models and

Oxygen-enhanced MRI Is Feasible, Repeatable, and Detects

Updated version

10.1158/1078-0432.CCR-18-3932doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2019/08/17/1078-0432.CCR-18-3932.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/25/13/3818.full#ref-list-1

This article cites 47 articles, 11 of which you can access for free at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://clincancerres.aacrjournals.org/content/25/13/3818To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-18-3932

http://clincancerres.aacrjournals.org/content/suppl/2019/08/17/1078-0432.CCR-18-3932.DC1

http://clincancerres.aacrjournals.org/content/25/13/3818.full#ref-list-1

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

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/25/13/3818


Documents

Oxygen-enhanced MRI Is Feasible, Repeatable, and Detects … · Precision Medicine and Imaging Oxygen-enhanced MRI Is Feasible, Repeatable, and Detects Radiotherapy-induced Change