intraoperative visualization of the tumor microenvironment ...mainly focused on tissue anatomy and...

SC I ENCE ADVANCES | R E S EARCH ART I C L E

CANCER

1Beckman Institute for Advanced Science and Technology, University of Illinois atUrbana-Champaign, Urbana, IL 61801, USA. 2Department of Electrical and Com-puter Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801,USA. 3Department of Bioengineering, University of Illinois at Urbana-Champaign,Urbana, IL 61801, USA. 4Carle Foundation Hospital, Urbana, IL 61801, USA.5Carle-Illinois College of Medicine, University of Illinois at Urbana-Champaign,Urbana, IL 61801, USA. 6Cancer Center at Illinois, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA.*Corresponding author. Email: [email protected]

Sun et al., Sci. Adv. 2018;4 : eaau5603 19 December 2018

Copyright © 2018

The Authors, some

rights reserved;

exclusive licensee

American Association

for the Advancement

of Science. No claim to

originalU.S. Government

Works. Distributed

under a Creative

Commons Attribution

NonCommercial

License 4.0 (CC BY-NC).

Intraoperative visualization of the tumormicroenvironment and quantification of extracellularvesicles by label-free nonlinear imaging
Yi Sun1,2, Sixian You1,3, Haohua Tu1, Darold R. Spillman Jr.1, Eric J. Chaney1, Marina Marjanovic1,3,Joanne Li1,3, Ronit Barkalifa1, Jianfeng Wang1, Anna M. Higham4, Natasha N. Luckey4,Kimberly A. Cradock4, Z. George Liu4, Stephen A. Boppart1,2,3,4,5,6*
Dow

nloaded fro

Characterization of the tumor microenvironment, including extracellular vesicles (EVs), is important forunderstanding cancer progression. EV studies have traditionally been performed on dissociated cells, lackingspatial information. Since the distribution of EVs in the tumor microenvironment is associated with cellularfunction, there is a strong need for visualizing EVs in freshly resected tissues. We intraoperatively imaged un-treated human breast tissues using a custom nonlinear imaging system. Label-free optical contrasts of thetissue, correlated with histological findings, enabled point-of-procedure characterization of the tumor micro-environment. EV densities from 29 patients with breast cancer were found to increase with higher histologicgrade and shorter tumor-to-margin distance and were significantly higher than those from 7 cancer-free pa-tients undergoing breast reduction surgery. Acquisition and interpretation of these intraoperative images notonly provide real-time visualization of the tumor microenvironment but also offer the potential to use EVs as alabel-free biomarker for cancer diagnosis and prognosis.

hm
on A
ugust 19, 2020ttp://advances.sciencem

ag.org/

INTRODUCTIONThe tumor microenvironment, host to cancer-associated events (1)such as angiogenesis (2, 3), production of cancer-associated fibro-blasts (CAFs) (4), and a reorganized extracellular matrix (ECM) (5),provides many potential biomarkers for cancer pathology. Tumor-associated extracellular vesicles (EVs), which play important roles inintercellular communication both inside and outside of the tumormicroenvironment (6), have been found to promote tumor progressionby directing cancer-associated events and changes (6–8), illustratingthe clinical significance of EVdetection. Various EVdetectionmethodsfor cancer diagnosis have been proposed and investigated, such as flowcytometry performed on circulating exosomes (9), immuno-based de-tection (10), and fluorescence label-based approaches for visualization(11). However, detection and imaging of EVs have not been performedin a spatially-resolved and label-free way to study the unperturbeddensity and distribution of EVs in ex vivo human tumor tissues. Novellabel-free multimodal multiphoton imaging technology has been im-plemented in a laboratory-based system to visualize the unperturbedEVs in an in vivo animal tumor model to evaluate its potential appli-cability to human breast cancer (12, 13). Few attempts, however, havebeen made to study human EVs at the point of procedure, such asduring interventional surgical procedures with a portable system de-signed for intraoperative label-free nonlinear optical imaging of freshuntreated or unstained human tumor tissues.

Intraoperative optical imaging and spectroscopy have previouslybeendemonstrated to address different diagnostic cancer needswith var-ious imaging modalities, such as tumor margin assessment (14–16),

tumor type differentiation (17, 18), and lesion/malignancy determi-nation (18, 19). Nonlinear optical imaging is particularly suitable forthe intraoperative visualization of human tissue specimens becauseof the molecular and structural specificity it can provide (17, 18).Previous intraoperative nonlinear optical imaging approaches haveused stimulated Raman scattering (18) and two-photon fluores-cence (2PF) with exogenous labeling agents (17), as well as second-harmonic generation (SHG) (17). However, these approaches weremainly focused on tissue anatomy and macroscopic tissue features,attempting to generate images that replicate hematoxylin and eosin(H&E)–stained histology. Few efforts have been devoted to imagingand characterizing the tumor microenvironment intraoperatively,which hosts many potential biomarkers for tumor-specific cells andfeatures, including EVs.

To visualize the unperturbed tumormicroenvironment in real time,we designed and built a custom portable multimodal system for label-free nonlinear optical imaging. On the basis of our methodology forperforming label-free imaging of EVs (12), the primary applicationof this intraoperative imaging systemwas to directly observe the spatialdistribution of EVs within the human tumor microenvironment. Fournonlinear optical imaging modalities were integrated into this systemand displayed in different colors in multimodal images: SHG was usedfor visualizing collagen fiber reorganization (20–22), displayed ingreen. 2PF was used for visualizing elastin fibers and flavin adeninedinucleotide (FAD)–containing cell cytoplasm (20, 23), displayed inyellow. Third-harmonic generation (THG) was used for highlightinginterface structures (20, 24) such as cell membranes, lipid boundaries,and EVs, displayed inmagenta, and 3PFwas used formappingNADH(reduced form of nicotinamide adenine dinucleotide) in the lipids(20, 25), displayed in cyan. EVs were visualized, characterized (fig. S1),and spatially coregistered with other visualized tissue features in thetumor microenvironment. The results of this study uniquely contrib-ute to a better understanding of the roles that EVs play in the tumormicroenvironment, their substantial clinical potential as a diagnosticand prognostic biomarker for cancer aggressiveness and progression,

1 of 10

http://advances.sciencemag.org/


and the potential for imaging and optically characterizing EVs in otherbiological samples such as liquid surgical waste, blood, and urine.

on http://advances.sciencem

ag.org/D

ownloaded from

RESULTSIntraoperative imaging of the unperturbedtumor microenvironmentThe unperturbed human breast tumor microenvironment in ex vivo–resected specimens was imaged intraoperatively and without the useof any exogenous stains or labels using the portable multimodal non-linear optical imaging system within 30 min after surgical excisionof the tissue. Abundant optical signals associated with tissue featureswere identified, including molecular signatures from autofluorescentNADH and FAD that are associated with metabolic activity withinindividual cells in the tumor microenvironment. These real-timeimage-based results reveal the substantial advantage of this label-freeintraoperative system over conventional H&E-stained or immuno-histochemically stained slides that often require several days to pre-pare. The multimodal images acquired from different sites (Fig. 1)demonstrate the capability of visualizing heterogeneous human breasttissue. The digital images of colocated H&E-stained histology slides(Fig. 1) were acquired for confirming the nonlinear optical signaturesof various tissue structures. In the multimodal nonlinear opticalimages, EVs were found near the breast cancer region (Fig. 2), suggest-ing their origin and their tumor-related functions (26).

The most common subtype of invasive carcinoma in breast tissue,invasive ductal carcinoma (IDC), is characterized by thick rows oflarge groups of tumor cells that construct a nest-like structure (Fig.1A), with an overall orientation marked by red arrows. These cells wereidentified by their yellow or magenta cytoplasm, which represents the2PF signal primarily generated fromFAD(27), and theTHGsignal gen-erated from membrane-bound structures such as EVs (12) and ribo-somes, respectively. In addition, the collagen fibers (green channel)were aligned in the same direction in response to the tumor cell in-filtration, as verified by the colocated histology images.


Mammary ducts and blood vessels in the tumor microenviron-ment are associated with DCIS and angiogenesis, respectively.These structures were visualized in our study and found to be re-siding with adipocytes inside the ECM (Fig. 1, B and C). Epithelialcells of mammary ducts and the interfaces of adipocytes are high-lighted in the THG channel. In addition, the strong 3PF signal wasmainly generated from NADH associated with lipids and localizedinside the adipocytes. The yellow-colored endothelial cells of a bloodvessel (Fig. 1B) were visualized by 2PF imaging. Last, the ECM vi-sualized by SHG supported the tissue structures discussed above.These images demonstrate the heterogeneity of the tumor micro-environment and the capability of our imaging system to faithfullyvisualize its major structural and molecular components. For com-parison, we also imaged healthy breast tissues (Fig. 1D) from cancer-free subjects undergoing breast reduction surgery. The healthy breasttissues can be readily distinguished from the cancerous breast tissuesby the prevalence of tissue areas with highly organized collagen fibersand a lack of cancer cells and angiogenesis.

In addition to macroscopic features (Fig. 1), micro- to nanoscalefeatures such as enrichment of EVs near a site of DCIS were also iden-tified (Fig. 2). Amultimodal label-free nonlinear optical image (Fig. 2A)includes multiple nonlinear optical contrasts, respectively, from SHG-visible ECM, 2PF/3PF-visible tumor cells within the region of DCIS,and THG-visible EVs. Most tissue features were confirmed by the co-located histology image (Fig. 2B), except for the EVs, which are lostduring histological preparation. In the nonlinear optical image, EVswere primarily visualized by THG imaging because of the strongphase-matching condition that themembrane structures provided, es-pecially because of their large surface-to-volume ratios (28). TheseEVs, appearing in the THG-contrast image (Fig. 2C) as diffraction-limited bright points, were segmented from the background (Fig. 2D)using a segmentation algorithm (Materials and Methods). A high den-sity of EVswere found in this specific region near the tumor (12), whichagreeswith previous observations that cancer cells are amajor producerof EVs in the tumor microenvironment (29). The quantification and

August 19, 2020

Fig. 1. Multimodal intraoperative label-free nonlinear optical images of human breast tissues and the corresponding histology. Multimodal label-free non-linear optical image (left) and the colocated histology (right) of (A) invasive ductal carcinoma (IDC) with an overall orientation of collagen alignment and tumor cellinfiltration (red dashed arrows), (B) adipocytes (red dashed arrows) and blood vessel (red solid arrows), (C) adipocytes (red dashed arrows) and mammary duct (red solidarrows), and (D) healthy breast tissue from breast reduction surgery. Scale bars, 100 mm.

2 of 10



on August 19, 2020


Dow

nloaded from

analysis of EVs reveal correlations betweenEVdensity and pathologicaldiagnoses, as discussed below.

Quantification of EVs and correlations withpathological diagnosisTissue specimens from 29 human breast cancer cases and 7 cancer-free breast reduction cases were imaged and analyzed for this study.An automated segmentation algorithm (Fig. 3A) was used to isolatethe EVs (Fig. 3B) and subsequently quantify EVdensity (Materials andMethods). The segmentation algorithm was applied on THG imagesof purified vesicles (Fig. 3B) to validate its accuracy (Materials andMethods). Using this algorithm to segment EVs from the intra-operative images of the breast tissues, we found a clear difference inEV density (Fig. 3C) between the cancer cases (average, 142 ± 56 nl−1)compared with the breast reduction cases (average, 23 ± 8 nl−1). Fur-thermore, among the breast cancer cases, the quantified EV densitieswere correlated with the pathological diagnoses, including the histo-logic grade of IDC, the nuclear grade of DCIS, and the tumor-to-margin distance. The grades of IDC andDCIS are used in breast cancerpathology to evaluate the apparent aggressiveness of the cancer cells(30). A higher grade typically implies a faster and more aggressivetumor growth, and it has been suggested that more aggressive tumorcells will produce a higher number of EVs distributed throughout thetumormicroenvironment (31). The quantified EVdensities are plottedagainst the tumor-to-margin distance (Fig. 3D). As expected, there is adecreasing trend of EV density, with increasing distance from the tu-mor to the closest surgical margin or surface of the excised tissuemassthat was imaged intraoperatively, indicating the diffusion-driven EV


distribution. Moreover, data points with different histologic gradesof IDC (1, 2, and 3) could be divided into three groups (shaded re-gions in Fig. 3D), where the curves of the region boundaries arerepresented by a diffusion model for EVs (Materials and Methods).By controlling the tumor-to-margin distance variation, we also per-formedmultiway analysis of variance (ANOVA) statistics (Materialsand Methods) to find that the quantified EV densities are highly re-lated to the histologic grade of IDC (Fig. 3E), but less related to thenuclear grade of DCIS.

Application of EV quantification and tumormicroenvironment visualization: Studyingcancer invasion near desmoplasiaDesmoplastic reaction (or desmoplasia), as a pathophysiologic eventoccurring in the stroma of breast tissue, is often used as a histo-pathological risk factor for cancer invasion (32–35). Multimodalnonlinear optical images were acquired at tissue sites that were iden-tified histologically, by a board-certified pathologist, as containingevidence of desmoplasia. From the nonlinear optical images, macro-scopic morphological features associated with cancer invasion wereidentified and validated by comparing with colocated histologicalimages. Our findings also show that the distribution of EV densities,which are highlighted in the THG images, matches the invasionphase, based on the macroscopic morphological features.

In the nonlinear optical and histologic images of an early phaseof desmoplasia (Fig. 4, A and B), the red dashed lines mark the in-terfaces between tumor and desmoplasia. The dense and thick col-lagen fibers below the interface were recognized as being associatedwith desmoplasia, while the regions with densely packed cells, lo-cated on the other side of the interface, were identified as the tumorregion. These dense collagen fibers of desmoplasia are tightly alignedto block tumor cell infiltration. That is because a desmoplastic re-action, at an earlier phase of carcinogenesis, is described as a sec-ondary reaction of the human body trying to “heal” the tumor byproducing dense fibrosis (35). However, it is still largely unknownwhether this reaction is initiated by the tumor or is a response reac-tion of the body (36). At this early phase, tumor cells have not startedto invade the dense ECM and, hence, are considered to be less ag-gressive. The EV density obtained from the THG-contrast imageof this imaging site was around 144 nl−1 (Fig. 4C) on both sides ofthe red dashed line.

A later phase of tumor invasion is initiated by the secretion ofECMdegrading enzymes such asmatrix metalloproteinases from tu-mor cells (37), which can break down and remodel the dense colla-gen fibers produced by the desmoplastic reaction. The tumor cellssubsequently use these fibers as a scaffold to facilitate their furthermigration (37). In this later phase (Fig. 4, D and E), small gaps wereobserved between the desmoplastic region located below the tumor-desmoplasia interface (red dashed lines), and a few groups of cells(white and yellow arrows) were identified within these gaps. In breastcancer histopathology, the presence of tumor cells and fibroblastsinside the newly formed ECM normally signifies a later phase of localtumor invasion (37, 38). These active tumor cells and CAFs rely onEVs to transfer intercellular information to complete targeted gene ex-pression (39) that facilitates further tumor progression. As expected,images reveal a greater number of EVs (575 nl−1) at this site (Fig. 4E),compared with the early phase (Fig. 4C), with most EVs found insidethe desmoplastic region, suggesting that more active intercellularcommunication is taking place.

Fig. 2. EV enrichment in the tumor microenvironment. (A) Multimodal label-free nonlinear optical image of a tissue site with ductal carcinoma in situ (DCIS)(boundarymarked by red dashed line). (B) ColocatedH&E histology. (C) THG-contrastimage visualizing the DCIS boundary and EVs. (D) Binary image of EVs segmentedfrom (C). Scale bars, 100 mm.

3 of 10



on August 19, 2020


Dow

nloaded from

There are several similarities and differences between the two phasesof cancer invasion near sites of desmoplasia. In both cases, the thickand straightened desmoplasia-associated collagen fibers can be easilydistinguished from the thin and wavy collagen fibers within the tumorregions, and there is a clear boundary separating the tumor region andthe desmoplasia-associated collagen region. In the later phase, tumorcells are observed to infiltrate the remodeled desmoplasia-associatedcollagen region. These similarities and differences help classify anddifferentiate desmoplasia from the tumor and enable the direct com-parison between cancer invasion phases. The concurrence of the mac-roscopic morphological features and the signature distributions of EVdensities directly reveal the relationship between the production anddistribution of EVs and macroscopic cancer invasion.


DISCUSSIONThe tissue features of the breast tumor microenvironment, includingcollagen fibers, cells, blood vessels,mammary ducts, and lipids, are typ-ically visualized by H&E-stained histology that requires labor- andtime-intensive tissue processing. Intraoperative multimodal nonlinearoptical imaging can visualize most of these features in ex vivo–resectedtissue specimens in real time without the aid of any labeling or tissuepreparation. The correlations with gold standard H&E-stained histo-logy were found to validate the nonlinear optical signatures of theseessential tissue features. The label-free multimodal nonlinear opticalimages can provide abundant details and molecular contrasts fromthe unperturbed breast tumormicroenvironment, especially the meta-bolic information associated with FAD and NADH that is rapidly lost

Fig. 3. Quantification and pathological correlations of EVs. (A) Flowchart of EV segmentation and quantification algorithm. (B) Representative THG-contrast imageacquired from the tumor microenvironment and processed binary image, highlighting the presence of EVs within the tumor microenvironment. (C) Comparison of EVdensity from breast cancer cases versus healthy breast reduction cases. The average EV density is 142 ± 55 nl−1 for the cancer cases, while it is only 23 ± 8 nl−1 for thehealthy breast reduction cases. ****P < 0.0001 (one-sided Student’s t test). (D) EV density data from each case are registered by the distance from tumor to closestsurgical margin and the cancer invasiveness grade. An overall decreasing trend of EV density is identified with increasing tumor-to-margin distance. Data points aredivided into three groups (shaded areas) representing different histologic grades of IDC. (E) Relationship between EV density and IDC histologic grade. To minimize theeffect of spatial heterogeneity, EV data were chosen from cases within a small range of margin distances (0 to 8 mm). Sample size of each IDC grade is indicated aboveeach bar. ***P < 0.001, **P < 0.01, *P < 0.1 (multiway ANOVA test, multiple comparison test).

4 of 10



on August 19, 2020


Dow

nloaded from

after tissue excision. Although the nonlinear optical signals from cellnuclei are not as distinguished as in stained histological sections, neg-ative contrast (dark regions) is still evident because of the intense signalfrom the cell cytoplasm (12, 20).

This multimodal nonlinear optical imaging method enables thedirect intraoperative observation of EVs in the unperturbed ex vivohuman breast tumor microenvironment. The enrichment of EVsidentified in the microenvironment surrounding the resected breasttumors from human subjects provided strong evidence for the cor-relation of observed EV distribution with macroscopic tumor tissueevents (12). It was shown that the trend of decreasing EV densitycorrelated well with increasing tumor-to-margin distance, and theEV density determined from tumor specimens of different histologicgrades of IDC can be separated by the spatial distribution curves ofthe EV diffusion model. Unexpectedly, EV density was more depen-dent on the histologic grade of IDC than on the nuclear grade ofDCIS.A possible reason is that the tumor cells inDCIS are not invasive, com-pared with those in IDC, and hence produce fewer EVs for tumor-stroma interactions (31). Therefore, the lower numberof EVs producedby DCIS offers a small contribution to the overall spatial distributionof EVs (31). A further study of EV densities and their distribution inother organ systems, and at sites of tumor metastasis, is expected toreveal more differences between IDC and DCIS in terms of EV pro-duction, kinetics, and distribution.

It is noted that in this feasibility study, the imaged tissue volumesfor each breast cancer case are relatively small considering the heter-


ogeneity of breast tissue specimens, the physical size of the resectedtissue/tumor mass, and the size of the human breast. Therefore, intra-operative delineation of the tumormargin is beyond the capabilities ofthis imaging system. Furthermore, a direct relationship between EVdensity and histologic grade has not been established with controlledtumor-to-margin distances due to the limited number of cases withsimilar tumor-to-margin distances and different cancer invasivenessgrades. The application of EV quantification for the assessment ofcancer invasion near sites of desmoplasia was also limited by the num-ber and size of images, and thus, wewere unable to providemeans andstandard deviations (SDs) for the quantified EV densities (Fig. 4). Thestatistical significance can possibly be improved further in the futureby implementing a fast-scanning microscope stage along with simul-taneous multichannel detection to increase the imaging volume andnumber of imaged sites within the limited acquisition time permittedin the operating room (~5 min) and by collecting EV density datafrom more human breast cancer cases of different histologic gradesand different tumor-to-margin distances.

The intraoperative imaging and visualization of the tumor micro-environment, and the quantification of EV density, demonstrate thestrengths of our portable nonlinear optical imaging system in cancerresearch and clinical applications.Without the need for tissue fixation,processing, sectioning, staining, and preparation, this imaging systemcan provide multimodal contrast in real time to identify many tissuefeatures that are diagnostically relevant in cancer, potentially reducingthe labor and time costs associated with current surgical breast cancer

Fig. 4. Determination of phase of tumor cell invasion around desmoplasia by EV distribution. (A) Multimodal label-free nonlinear image of desmoplasia at an earlyphase. There is an interface (red dashed line) between the tumor and the dense collagen fibers of desmoplasia, and the tumor cells are identified only in the tumor region(white arrows). (B) Colocated histology image of the early-phase desmoplasia. (C) Binary image of segmented EVs from the THG channel of (A). The average (AVG) EVdensity is quantified to be 144 nl−1, and there is no major difference (113 nl−1 versus 163 nl−1) between the EV counts from each side of the interface. (D) Multimodalnonlinear optical image of desmoplasia at a late phase. The interface between dense collagen fibers and the tumor is marked by a red dashed line, with infiltrating tumorcells being identified within the collagen region (white arrows). (E) Colocated histology image of this late-phase desmoplastic reaction. (F) Binary image of segmented EVsfrom the THG channel (A). The average EV density of the entire FOV is 575 nl−1, but the EV density within the dense collagen fibers (938 nl−1) is much higher than the EVdensity within the tumor (188 nl−1). Scale bars, 100 mm.

5 of 10



on August 19, 2020


Dow

nloaded from

diagnostics. Uniquely, the label-free in situ visualization of EVs in theex vivo human breast tumor microenvironment, validated by immu-nohistochemical labeling and colocated detection (fig. S1), enables theinvestigation and quantification of the spatially resolved properties ofEVs. As a result, the quantification of EV densities revealed relation-ships with pathological diagnoses, including tumor-to-margin dis-tance and cancer invasiveness. In addition, the EV density anddistribution near sites of desmoplasia were shown to be associatedwith macroscopicmechanisms and processes in carcinogenesis. Theseresults suggest the feasibility and future potential for implementing in-traoperative label-free nonlinear optical imaging to investigate the hu-man breast tumor microenvironment and the spatial EV distribution,both to improve our fundamental understanding of carcinogenesis andto potentially provide new biomarkers for tumor invasiveness.

MATERIALS AND METHODSStudy designThe primary objectives of this study were to intraoperatively visual-ize and characterize the human breast microenvironment ex vivousing label-free nonlinear optical imaging and to find the relation-ship between EV density and pathological diagnoses. Informed con-sents were obtained from all 29 patients with breast cancer and 7cancer-free healthy patients participating in this study. During thebreast cancer surgeries, the resected fresh human breast tissue speci-mens were imaged in the operating room with the portable system.Following guidance from the surgeon, the sites imaged on the surgi-cal margin were chosen to be closest to the tumor within the resected


specimen. Histological slides and pathological diagnoses were ob-tained postoperatively for correlation of image features and EV den-sities. Among the intraoperative images, only those collected fromcancer stromal tissue were included in EV density analysis, while theimages of pure adipose tissue were excluded. The inclusion criteria forthe subjects in this study included a biopsy-proven diagnosis of breastcancer (DCIS or IDC) in need of surgical treatment or elective breastreduction surgery in subjects with no history of cancer. No subjectswere excluded on the basis of age or race. The intraoperative imagingand EV density quantification were blindly implemented prior to ob-taining the pathological diagnoses of the cancer subjects. The patho-logical reports were assessed afterward to correlate with the collectedmultimodal images and the quantified EV densities.

Intraoperative real-time multimodal label-free nonlinearimaging systemA portable imaging system integrating four nonlinear optical imag-ing modalities (Fig. 5, A to C) was designed to collect label-free mul-timodal imaging data in the operating room during cancer surgeries.The laser source in this imaging system provided transform-limited55-fs laser pulses at a 70-MHz repetition rate and with a spectralrange of 1040 to 1100 nm, which can excite the four nonlinear opticalprocesses with high efficiency and avoid the potential laser damageand photo bleaching in the tissue specimens. With this excitationspectral range, the emitted nonlinear optical signals from the fourmodalities were separately detected in different spectral windows(Fig. 5D), achieved by four optical bandpass filters. The color codefor each imaging modality was chosen to represent the approximate

Fig. 5. Intraoperative label-free multimodal imaging system. (A) A photograph of the compact and portable intraoperative label-free multimodal nonlinear im-aging system. (B) Software interface of the imaging system. (C) System schematic. (D) Spectral range and display color of the four nonlinear optical imaging modalities.L, lens; GM, galvanometer-scanning mirror; DM, dichroic mirror; OBJ, objective; FW, filter wheel; PMT, photomultiplier tube. (Photo credit: Yi Sun, Biophotonics ImagingLaboratory, University of Illinois at Urbana-Champaign.)

6 of 10



on August 19, 2020


Dow

nloaded from

wavelengths of the emitted nonlinear optical signals from the specimens.Using a pair of galvanometer scanning mirrors, a nonlinear opticalimage (800 by 800 pixels, 500 by 500 nm2 per pixel) was acquiredwithin80 s for each imaging modality. By sequentially switching betweenfilters for each imaging modality, a complete set of images from asingle field of view (FOV) (400 × 400 mm2) was acquired in approx-imately 5 min.

To effectively observe EVs, the lateral resolution of THG imaging(~322 nm) was sufficiently high to visualize and resolve microvesi-cles (~500 to 1000 nm) and detect some exosomes (40 to 120 nm) (6).Under the spatial Nyquist sampling condition, the pixel size shouldbe set below half of the THG imaging resolution (~161 nm) to assuresufficient sampling. However, because of some residual image jittercaused by inevitable cart vibration from mechanical elements androom equipment, acquiring images with small pixel sizes would con-siderably reveal vibration-induced artifacts, yield low image qualityand fidelity, and increase image acquisition time. Therefore, the pixelsize was somewhat compromised and set to 500 nm by 500 nm, whichwas still sufficient to visualize EVs as diffraction-limited bright pointsin the acquired images due to the strong THG signal they emitted.Furthermore, to assure the depth-resolved sectioning of EVs, the ax-ial resolution of THG imaging in this imaging system was estimatedto be 1.0 mm, which was later used to calculate the imaging volumeand EV density.

The entire imaging system was housed in a compact and portablecart (90 cm by 90 cm by 120 cm, 90 kg) that was comparable in sizeand weight to other intraoperative equipment, such as intraoperativeultrasound, intraoperative optical coherence tomography, and anes-thesia carts used within the intraoperative working environment.The system was able to be readily moved throughout the hospitaland clinical environment and operated by a single person. The opticalcomponents were aligned in a robust way to withstand floor obstaclesand vibrations during transportation, eliminating the need for realign-ment before image acquisition. The system was designed so all opticalcomponents and electronics were contained within the cart enclosure,and imaging was performed in an inverted microscope configurationwhere the specimen was simply placed on a clear glass window inlaidin the top surface of the cart, covered with a light-tight box-shapedcover, and imaged using an objective located within the cart and belowthe glass window. Because of the strictly controlled lighting conditionsin the operating room, light concealment during imaging was con-sidered a priority when the imaging system was designed and built,so as to obviate the need for the surgeon and staff to change lightingconditions during surgery and delay the procedure. Specifically, switch-ing between modalities was automated using a motorized filter wheel,and focus adjustment was accomplished by a piezoelectric linear stageto eliminate the need to open the cart doors. As a result, noise frombackground light was minimized, and the laser beam was confinedwithin the imaging cart for laser safety. On the other hand, necessaryventilation was maintained to reduce the thermal noise of the PMT(H7421-40, Hamamatsu Photonics K.K.). With the minimized back-ground and thermal noise, the average signal-to-noise ratio (SNR)measured from the intraoperative THG images was approximately19 ± 4 dB, sufficient to visualize the tissue structures.

EV segmentationOn the basis of the characteristics of EVs in the THG-contrast images,an automated segmentation algorithm (Fig. 3A) was developed to ex-tract the EV signal from the background and subsequently quantify


EV density. The principles of this segmentation algorithm relied onthe spatial and nonlinear optical properties of the EVs: that they aresmall (40 to 2000 nm) (6), point-like, and generate exceptionally strongTHG signal due to the good phase-matching condition provided bytheir large surface-to-volume interface ratio (28). Therefore, an inten-sity threshold was used to segment the EVs from the background (Fig.3A). This threshold was automatically set to be the pixel intensity valueat a fixed percentage out of the intensity histogram generated fromeach THG image, and this specific percentage for all images was delib-erately determined to leave only the in-focus EVs while suppressingthe background noise. By applying this algorithm to THG-contrastimages of the tumor microenvironment (Fig. 2C), binary images(Fig. 2D) were generated to reveal the spatial distribution and densityof the EVs. These black points in the binary image were subsequentlyquantified to represent the density of EVs in each FOV.

To validate this imaging and segmentation method for EV detec-tion and quantification, we acquired THG-contrast images of EVspurified from human cancer cell lines with a known density of 3 ×1010ml−1, measured by a standardized technique (40) with a commer-cial instrument (NS3000, NanoSight Ltd.). A representative exampleof a THG-contrast image of purified EVs was processed to highlightthe EVs (Fig. 3B) using this segmentation algorithm. Considering theaxial resolution and imaging FOV, the three-dimensional imagingvolume of each image of purified EVs was approximately 100 mm by100 mm by 1 mm = 10−8 ml. The density of EVs was then calculated tobe 1.5 × 1010ml−1 using the EV counts (152 ± 10), quantified from fiveTHG-contrast images of purified EVs. To explain the density dis-crepancy, it is likely that some EVs underwent refractive indexmatch-ing due to the diffusion and permeation of glycerol through the EVmembrane and, thus, ceased to provide the phase-matching conditionnecessary for THG signal generation. Nevertheless, the EV densityquantified from the THG images using the segmentation algorithmwas of the same magnitude as the known density of EVs measuredby the commercial instrument. Furthermore, the isolated EVs weremixed with human cells in culture. The drastic increase of EV densityafter mixing was identified by the quantification algorithm based onthe THG images (fig. S2) and validated by the NanoSight measure-ment. Therefore, the EV densities quantified from the intraoperativeTHG-contrast images faithfully represented the distribution and den-sity of EVs in the tumor microenvironment.

EV diffusion modelTo help explain and demonstrate the relationship between EV densityand tumor-to-margin distance, the distribution of EVs in the tumormicroenvironment was treated as a process of diffusion for small par-ticles (41). Assuming that the EV distribution is spherically symmetricand emanating from the center of the tumor and that the tumor cellsare the only major source of EVs, we can derive the time-dependentthree-dimensional diffusion equation to describe EV density C(r, t)with the boundary conditions

D*1r2

∂∂r

r2∂∂r

Cðr; tÞ� �

¼ ∂Cðr; tÞ∂t

Cð0; tÞ ¼ A

Cð∞; tÞ ¼ 0

where D is the diffusion coefficient, r is the radial distance from thetumor boundary, t is time, and A is the initial number of EVs. Under

7 of 10



Table 1. Demographic data on human subjects and pathological diagnoses. A total of 29 breast cancer subjects and 7 healthy cancer-free subjectsundergoing breast reduction surgery were included. Healthy subjects are indicated by NML (normal) under the histologic grade column.

Sun

Histologic grade (IDC)

et al., Sci. Adv. 2018;4 : eaa

Distance to IDC (mm)

u5603 19 December 20

Nuclear grade (DCIS)

18

Distance to DCIS (mm)
Age Tumor tissue/healthy tissue size (cm)
1
6 — — 41 0.4 by 0.3 by 0.3
1
1 — — 69 1.2 by 1.0 by 0.6
1
1 I 1 75 0.8 by 0.5 by 0.4
1
3 I 3 65 0.8 by 0.6 by 0.5
1
1.6 II 5 46 3.7 by 2.0 by 2.0
1
6 II >10 70 1.6 by 1.1 by 0.8
2
5 — — 75 1.5 by 1.3 by 1.2
2
50 — — 71 2.1 by 1.5 by 0.9
D
ow 2 1.3 — 8 37 1.8 by 1.5 by 1.3
n
loa 2 10 — — 48 1.0 by 1.0 by 0.9
d
ed
2
8 I–II >10 70 2.6 by 4.6 by 0.5
f
rom 2 14 II 17 59 Not grossly identified http 2 3 II 1 75 1.5 by 1.0 by 0.6
:/
/ad 2 6 II 10 50 1.1 by 1.0 by 0.4
v
anc 2 10 II 10 67 5.2 by 3.3 by 2.4
e
s.s 2 12 II 14 54 1.5 by 1.2 by 1.1
c
ien 2 51 II 51 82 1.7 by 1.5 by 1.3
c
ema 2 11 II 14 57 1.5 by 1.4 by 1.0
g
.or 2 8 II 2 60 1.2 by 0.8 by 0.8
g
on/
2
6.2 II 6.3 68 1.0 by 1.0 by 0.8 Aug 2 25 II 25 48 1.5 by 1.2 by 0.7
u
st 2 10 III 10 51 1.1 by 0.9 by 0.8
1
9, 2 2 1 III no 67 2.0 by 2.0 by 1.8
0
20 3 10 III 1.2 64 0.8 by 0.6 by 0.5
3
12 III 12 76 1.9 by 1.8 by 1.7
3
10 III 10 62 2.0 by 1.3 by 1.0
3
5 III 0 (positive) 75 2.5 by 2.4 by 2.2
3
4 — — 52 2.1 by 1.2 by 1.1
3
1 — — 68 1.7 by 1.1 by 1.0
NML
— — — 41 15 by 14 by 5.5
NML
— — — 50 19 by 16 by 9.0
NML
— — — 45 6.5 by 4.1 by 1.1
NML
— — — 70 18 by 15 by 8.5
NML
— — — 19 9.0 by 4.3 by 3.2
NML
— — — 50 15 by 12 by 12
NML
— — — 40 14 by 13 by 4.5
8 of 10



on August 19, 2020


Dow

nloaded from

the same assumptions, the boundary conditions for this partial differ-ential equation were set to be a constant value A at the tumor bound-ary and zero at infinite distance. As for the initial condition of the EVdensity distribution, functions such as an exponential decay function,power function, and Gaussian function were compared with the fittedresults of EV density versus tumor-to-margin distance.

The exponential decay function was chosen as an initial conditionto generate the region boundary curves shown in Fig. 3D. Consid-ering the much longer time scale of tumor growth (42) comparedwith EV diffusion (41), it was also assumed that EV diffusion wasalways at an equilibrium state (changing extremely slowly over time)at the time the tissue specimenswere imaged. The boundary conditionvalue A was used to represent the cancer invasiveness based on theobservation that more aggressive cancer cells tend to produce moreEVs at the boundary (31). The two EV distribution curves calculatedfrom two different values ofA served as the boundary curves of the EVdensity data points from cases with different histologic grades of IDC(Fig. 3D).

Multiway ANOVAMultiway ANOVA (anovan; MATLAB) was performed to examinethe relationship between EV density and the corresponding gradeof IDC/DCIS. Noticing that the tumor-to-margin distance alsocontributes to the EV density, we only included in the multiwayANOVA the cases with a tumor-to-margin distance between 0 and8 mm to minimize the contribution by distance and maintain asufficient number of data points. Statistically significant correla-tion was found between the EV density and the histological gradeof IDC (P = 0.0002), but not with the nuclear grade of DCIS (P =0.1835). Furthermore, with statistical significance existing betweenthe EVdensity and the histologic grade of IDC, amultiple comparisontest (multcompare, critical value: “tukey-kramer,” MATLAB) wasused to analyze the differences of EV densities between different his-tologic grades of IDC (Fig. 3E).

Human tissuesHuman tissues were obtained under a protocol approved by the In-stitutional Review Boards at the University of Illinois at Urbana-Champaign and Carle Foundation Hospital, Urbana, Illinois. A totalof 29 breast cancer human subjects and 7 healthy (and no history ofcancer) human subjects undergoing breast reduction surgeries wereincluded in this study (Table 1). During the breast cancer surgeries,and immediately following resection, the fresh human breast tissuespecimens were directly passed to the research team member in theoperating room for ex vivo imaging. On the basis of guidance fromthe surgeon, the location on the surgical margin surface that wasdeemed to be closest to the tumor within the resected mass was iden-tified and selected as the site from which images were collected. Thedistance between the imaged surgical margin surface and the tumormass (tumor-to-margin distance) was later measured postoperativelyby the pathologists.

Themultimodal label-free images of the unperturbedhuman breastspecimens were acquired within a time window of less than 30 minbetween the time of surgical excision and tissue fixation for histo-pathological processing, without disrupting or delaying the surgicalprocedure. Following imaging, the imaged regions on the surgicalmargin surfaces of the breast tissue specimens were marked withsurgical ink for later registration with histological slides for imagefeature correlations. After intraoperative imaging, the intact speci-


mens were sent to the pathology laboratory for standard processingand diagnosis.

SUPPLEMENTARY MATERIALSSupplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/4/12/eaau5603/DC1Fig. S1. Validation of THG imaging of EVs by immunohistochemical-based detection.Fig. S2. Increase of EV density in cell culture by adding isolated EVs.

REFERENCES AND NOTES1. L. A. Liotta, E. C. Kohn, The microenvironment of the tumour–host interface. Nature 411,

375–379 (2001).2. P. Carmeliet, R. K. Jain, Angiogenesis in cancer and other diseases. Nature 407, 249–257

(2000).3. S. M. Weis, D. A. Cheresh, Tumor angiogenesis: Molecular pathways and therapeutic

targets. Nat. Med. 17, 1359–1370 (2011).4. A. Orimo, P. B. Gupta, D. C. Sgroi, F. Arenzana-Seisdedos, T. Delaunay, R. Naeem,

V. J. Carey, A. L. Richardson, R. A. Weinberg, Stromal fibroblasts present in invasive humanbreast carcinomas promote tumor growth and angiogenesis through elevated SDF-1/CXCL12 secretion. Cell 121, 335–348 (2005).

5. P. P. Provenzano, K. W. Eliceiri, J. M. Campbell, D. R. Inman, J. G. White, P. J. Keely, Collagenreorganization at the tumor-stromal interface facilitates local invasion. BMC Med. 4,38 (2006).

6. S. EL Andaloussi, I. Mäger, X. O. Breakefield, M. J. A. Wood, Extracellular vesicles:Biology and emerging therapeutic opportunities. Nat. Rev. Drug Discov. 12, 347–357(2013).

7. C. Ciardiello, L. Cavallini, C. Spinelli, J. Yang, M. Reis-Sobreiro, P. de Candia, V. R. Minciacchi,D. Di Vizio, Focus on extracellular vesicles: New frontiers of cell-to-cell communicationin cancer. Int. J. Mol. Sci. 17, 175 (2016).

8. G. Raposo, W. Stoorvogel, Extracellular vesicles: Exosomes, microvesicles, and friends.J. Cell Biol. 200, 373–383 (2013).

9. S. A. Melo, L. B. Luecke, C. Kahlert, A. F. Fernandez, S. T. Gammon, J. Kaye, V. S. LeBleu,E. A. Mittendorf, J. Weitz, N. Rahbari, C. Reissfelder, C. Pilarsky, M. F. Fraga,D. Piwnica-Worms, R. Kalluri, Glypican-1 identifies cancer exosomes and detects earlypancreatic cancer. Nature 523, 177–182 (2015).

10. D. Duijvesz, C. Y. L. Versluis, C. A. M. van der Fels, M. S. Vredenbregt-van den Berg, J. Leivo,M. T. Peltola, C. H. Bangma, K. S. I. Pettersson, G. Jenster, Immuno-based detection ofextracellular vesicles in urine as diagnostic marker for prostate cancer. Int. J. Cancer 137,2869–2878 (2015).

11. C. P. Lai, E. Y. Kim, C. E. Badr, R. Weissleder, T. R. Mempel, B. A. Tannous, X. O. Breakefield,Visualization and tracking of tumour extracellular vesicle delivery and RNA translationusing multiplexed reporters. Nat. Commun. 6, 7029 (2015).

12. H. Tu, Y. Liu, M. Marjanovic, E. J. Chaney, S. You, Y. Zhao, S. A. Boppart, Concurrenceof extracellular vesicle enrichment and metabolic switch visualized label-free in thetumor microenvironment. Sci. Adv. 3, e1600675 (2017).

13. S. You, H. Tu, E. J. Chaney, Y. Sun, Y. Zhao, A. J. Bower, Y.-Z. Liu, M. Marjanovic,S. Sinha, Y. Pu, S. A. Boppart, Intravital imaging by simultaneous label-freeautofluorescence-multiharmonic microscopy. Nat. Commun. 9, 2125 (2018).

14. F. T. Nguyen, A. M. Zysk, E. J. Chaney, J. G. Kotynek, U. J. Oliphant, F. J. Bellafiore,K. M. Rowland, P. A. Johnson, S. A. Boppart, Intraoperative evaluation of breast tumormargins with optical coherence tomography. Cancer Res. 69, 8790–8796 (2009).

15. S. J. Erickson-Bhatt, R. M. Nolan, N. D. Shemonski, S. G. Adie, J. Putney, D. Darga,D. T. McCormick, A. J. Cittadine, A. M. Zysk, M. Marjanovic, E. J. Chaney, G. L. Monroy,F. A. South, K. A. Cradock, Z. G. Liu, M. Sundaram, P. S. Ray, S. A. Boppart, Real-timeimaging of the resection bed using a handheld probe to reduce incidence of microscopicpositive margins in cancer surgery. Cancer Res. 75, 3706–3712 (2015).

16. A. S. Haka, Z. Volynskaya, J. A. Gardecki, J. Nazemi, J. Lyons, D. Hicks, M. Fitzmaurice,R. R. Dasari, J. P. Crowe, M. S. Feld, In vivo margin assessment during partial mastectomybreast surgery using Raman spectroscopy. Cancer Res. 66, 3317–3322 (2006).

17. Y. K. Tao, D. Shen, Y. Sheikine, O. O. Ahsen, H. H. Wang, D. B. Schmolze, N. B. Johnson,J. S. Brooker, A. E. Cable, J. L. Connolly, J. G. Fujimoto, Assessment of breast pathologiesusing nonlinear microscopy. Proc. Natl. Acad. Sci. U.S.A. 111, 15304–15309 (2014).

18. D. A. Orringer, B. Pandian, Y. S. Niknafs, T. C. Hollon, J. Boyle, S. Lewis, M. Garrard,S. L. Hervey-Jumper, H. J. L. Garton, C. O. Maher, J. A. Heth, O. Sagher, D. A. Wilkinson,M. Snuderl, S. Venneti, S. H. Ramkissoon, K. A. McFadden, A. Fisher-Hubbard,A. P. Lieberman, T. D. Johnson, X. S. Xie, J. K. Trautman, C. W. Freudiger, S. Camelo-Piragua,Rapid intraoperative histology of unprocessed surgical specimens via fibre-laser-basedstimulated Raman scattering microscopy. Nat. Biomed. Eng. 1, 0027 (2017).

9 of 10

http://advances.sciencemag.org/cgi/content/full/4/12/eaau5603/DC1

http://advances.sciencemag.org/cgi/content/full/4/12/eaau5603/DC1



on August 19, 2020


Dow

nloaded from

19. A. S. Haka, K. E. Shafer-Peltier, M. Fitzmaurice, J. Crowe, R. R. Dasari, M. S. Feld, Diagnosingbreast cancer by using Raman spectroscopy. Proc. Natl. Acad. Sci. U.S.A. 102,12371–12376 (2005).

20. H. Tu, Y. Liu, D. Turchinovich, M. Marjanovic, J. K. Lyngsø, J. Lægsgaard, E. J. Chaney,Y. Zhao, S. You, W. L. Wilson, B. Xu, M. Dantus, S. A. Boppart, Stain-free histopathology byprogrammable supercontinuum pulses. Nat. Photonics 10, 534–540 (2016).

21. E. Brown, T. McKee, E. diTomaso, A. Pluen, B. Seed, Y. Boucher, R. K. Jain, Dynamicimaging of collagen and its modulation in tumors in vivo using second-harmonicgeneration. Nat. Med. 9, 796–800 (2003).

22. J. Condeelis, J. E. Segall, Intravital imaging of cell movement in tumours. Nat. Rev. Cancer3, 921–930 (2003).

23. M. C. Skala, K. M. Riching, A. Gendron-Fitzpatrick, J. Eickhoff, K. W. Eliceiri, J. G. White,N. Ramanujam, In vivo multiphoton microscopy of NADH and FAD redox states,fluorescence lifetimes, and cellular morphology in precancerous epithelia. Proc. Natl.Acad. Sci. U.S.A. 104, 19494–19499 (2007).

24. D. Débarre, W. Supatto, A.-M. Pena, A. Fabre, T. Tordjmann, L. Combettes,M.-C. Schanne-Klein, E. Beaurepaire, Imaging lipid bodies in cells and tissues usingthird-harmonic generation microscopy. Nat. Methods 3, 47–53 (2006).

25. S. Maiti, J. B. Shear, R. M. Williams, W. R. Zipfel, W. W. Webb, Measuring serotonindistribution in live cells with three-photon excitation. Science 275, 530–532 (1997).

26. Q. Feng, C. Zhang, D. Lum, J. E. Druso, B. Blank, K. F. Wilson, A. Welm, M. A. Antonyak,R. A. Cerione, A class of extracellular vesicles from breast cancer cells activates VEGFreceptors and tumour angiogenesis. Nat. Commun. 8, 14450 (2017).

27. S. Huang, A. A. Heikal, W. W. Webb, Two-photon fluorescence spectroscopy andmicroscopy of NAD(P)H and flavoprotein. Biophys. J. 82, 2811–2825 (2002).

28. T. Y. F. Tsang, Optical third-harmonic generation at interfaces. Phys. Rev. A 52, 4116–4125(1995).

29. T. L. Whiteside, Tumor-derived exosomes and their role in cancer progression. Adv. Clin.Chem. 74, 103–141 (2016).

30. C. W. Elston, I. O. Ellis, Pathological prognostic factors in breast cancer. I. The value ofhistological grade in breast cancer: Experience from a large study with long-termfollow-up. Histopathology 19, 403–410 (1991).

31. A. Ginestra, M. D. La Placa, F. Saladino, D. Cassarà, H. Nagase, M. L. Vittorelli, The amountand proteolytic content of vesicles shed by human cancer cell lines correlates withtheir in vitro invasiveness. Anticancer Res. 18, 3433–3437 (1998).

32. L. Bochet, C. Lehuédé, S. Dauvillier, Y. Y. Wang, B. Dirat, V. Laurent, C. Dray, R. Guiet,I. Maridonneau-Parini, S. Le Gonidec, B. Couderc, G. Escourrou, P. Valet, C. Muller,Adipocyte-derived fibroblasts promote tumor progression and contribute to thedesmoplastic reaction in breast cancer. Cancer Res. 73, 5657–5668 (2013).

33. H. Kim, J. Watkinson, V. Varadan, D. Anastassiou, Multi-cancer computational analysisreveals invasion-associated variant of desmoplastic reaction involving INHBA, THBS2 andCOL11A1. BMC Med. Genomics 3, 51 (2010).

34. C. A. Iacobuzio-Donahue, P. Argani, P. M. Hempen, J. Jones, S. E. Kern, The desmoplasticresponse to infiltrating breast carcinoma: Gene expression at the site of primary invasion andimplications for comparisons between tumor types. Cancer Res. 62, 5351–5357 (2002).

35. M. A. Troester, M. H. Lee, M. Carter, C. Fan, D. W. Cowan, E. R. Perez, J. R. Pirone,C. M. Perou, D. J. Jerry, S. S. Schneider, Activation of host wound responses in breastcancer microenvironment. Clin. Cancer Res. 15, 7020–7028 (2009).

36. R. A. Walker, The complexities of breast cancer desmoplasia. Breast Cancer Res. 3,143–145 (2001).

37. H. Ohtani, Stromal reaction in cancer tissue: Pathophysiologic significance of theexpression of matrix-degrading enzymes in relation to matrix turnover and immune/inflammatory reactions. Pathol. Int. 48, 1–9 (1998).


38. M. W. Conklin, P. J. Keely, Why the stroma matters in breast cancer: Insights into breastcancer patient outcomes through the examination of stromal biomarkers. Cell Adhes.Migr. 6, 249–260 (2012).

39. Y. Lee, S. EL Andaloussi, M. J. A. Wood, Exosomes and microvesicles: Extracellular vesiclesfor genetic information transfer and gene therapy. Hum. Mol. Genet. 21, R125–R134(2012).

40. V. Filipe, A. Hawe, W. Jiskoot, Critical evaluation of Nanoparticle Tracking Analysis (NTA)by Nanosight for the measurement of nanoparticles and protein aggregates.Pharm. Res. 27, 796–810 (2010).

41. N. Kastelowitz, H. Yin, Exosomes and microvesicles: Identification and targeting byparticle size and lipid chemical probes. Chembiochem 15, 923–928 (2014).

42. H. M. Byrne, Dissecting cancer through mathematics: From the cell to the animal model.Nat. Rev. Cancer 10, 221–230 (2010).

Acknowledgments: We thank Carle Foundation Hospital and its physicians, surgeons,nursing, and research staff for their clinical collaboration and assistance with this translationalresearch study. Funding: Research reported in this publication was supported by theNational Institute for Biomedical Imaging and Bioengineering and the National CancerInstitute of the NIH under award numbers R01EB023232, R01CA166309, and R01CA213149.One hundred percent of the total project costs was financed with federal money, and0% of the total costs was financed by nongovernmental sources. The content is solely theresponsibility of the authors and does not necessarily represent the official views of theNIH. This work was also supported by an award from the Cancer Scholars for Translationaland Applied Research (CSTAR) program of Carle Foundation Hospital, the Universityof Illinois at Urbana-Champaign, and the Cancer Center at Illinois. Author contributions:H.T. and S.A.B. conceived the project of intraoperative nonlinear optical imaging. Y.S.,S.Y., and D.R.S.J. designed and built the portable imaging system. Y.S., S.Y., and J.W.conducted imaging in the operating room. E.J.C., M.M., J.L., and R.B. assisted with imagecollection and communicated with surgeons, nurses, and pathologists for specimenhandling and pathological reports. M.M. and S.A.B. wrote the protocol for intraoperativeimaging of breast cancer specimens. E.J.C. prepared the histological slides from imagingsites. Z.G.L. provided expert advice on pathological interpretation of the image data.A.M.H. and K.A.C. performed the breast cancer surgeries and provided surgical guidanceon tumor localization. N.N.L. performed the breast reduction surgeries. Y.S. obtainedhistological correlations for the intraoperative images and analyzed EV density data. Y.S.and S.A.B. wrote the manuscript. Competing interests: H.T., S.Y., and S.A.B. are inventorson patents filed by the University of Illinois at Urbana-Champaign (2018/0286044 A1,4 October 2018) related to the laser source technology and the imaging and quantificationof EVs. All other authors declare that they have no competing interests. Data andmaterials availability: All data needed to evaluate the conclusions in the paper are presentin the paper and/or the Supplementary Materials. Additional data available from authorsupon request.

Submitted 26 June 2018Accepted 19 November 2018Published 19 December 201810.1126/sciadv.aau5603

Citation: Y. Sun, S. You, H. Tu, D. R. Spillman Jr., E. J. Chaney, M. Marjanovic, J. Li, R. Barkalifa,J. Wang, A. M. Higham, N. N. Luckey, K. A. Cradock, Z. George Liu, S. A. Boppart,Intraoperative visualization of the tumor microenvironment and quantification ofextracellular vesicles by label-free nonlinear imaging. Sci. Adv. 4, eaau5603 (2018).

10 of 10


vesicles by label-free nonlinear imagingIntraoperative visualization of the tumor microenvironment and quantification of extracellular

Jianfeng Wang, Anna M. Higham, Natasha N. Luckey, Kimberly A. Cradock, Z. George Liu and Stephen A. BoppartYi Sun, Sixian You, Haohua Tu, Darold R. Spillman, Jr., Eric J. Chaney, Marina Marjanovic, Joanne Li, Ronit Barkalifa,

DOI: 10.1126/sciadv.aau5603 (12), eaau5603.4Sci Adv

ARTICLE TOOLS http://advances.sciencemag.org/content/4/12/eaau5603

MATERIALSSUPPLEMENTARY http://advances.sciencemag.org/content/suppl/2018/12/17/4.12.eaau5603.DC1

CONTENTRELATED http://stke.sciencemag.org/content/sigtrans/12/567/eaan8247.full

REFERENCES

http://advances.sciencemag.org/content/4/12/eaau5603#BIBLThis article cites 42 articles, 12 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.Science AdvancesYork Avenue NW, Washington, DC 20005. The title (ISSN 2375-2548) is published by the American Association for the Advancement of Science, 1200 NewScience Advances

License 4.0 (CC BY-NC).Science. No claim to original U.S. Government Works. Distributed under a Creative Commons Attribution NonCommercial Copyright © 2018 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of

on August 19, 2020


Dow

nloaded from

http://advances.sciencemag.org/content/4/12/eaau5603

http://advances.sciencemag.org/content/suppl/2018/12/17/4.12.eaau5603.DC1

http://stke.sciencemag.org/content/sigtrans/12/567/eaan8247.full

http://advances.sciencemag.org/content/4/12/eaau5603#BIBL

http://www.sciencemag.org/help/reprints-and-permissions

http://www.sciencemag.org/about/terms-service


intraoperative visualization of the tumor microenvironment ...mainly focused on tissue anatomy and...

Documents