International Journal of Radiation Research, April 2016 Volume 14, No 2

Detection of breast cancer using non-invasive X-ray diffraction technique of hair: A preliminary study

INTRODUCTION

Earlydiagnosisofbreastcancer is importantfor making decision on proper treatmentplanning. Because of high and also increasingworldwide morbidity and mortality rates ofbreast cancer (1-3), there are many researcherswho interested in diagnosis and treatment ofthis type of neoplasm. Since the cure rate ofcancer is highly dependent on stage of thedisease, early detection of disorder is animportant subject to choose an effectivetreatment strategy. To obtain an accurate andalsopropertreatmentplanandstrategy,thereisstrongneedforreliableandalsoearlydiagnosismethods. The usual methods for detection ofbreast cancer after physical examination aremammography; which is gold standard,magnetic resonance imaging (MRI) and

ultrasound(3,4).Sincedetectionofdensebreastsisdif*icultinmammography(5),theradiographerand also the physician should be pro*icient (6).The sensitivity for mammography is about 70percent, and for de*inite diagnosis of cancerthere is a need of pathology and biopsycon*irmation. It seems that a simple, lessdangerousandsafemethodwithhighsensitivitycouldbeinitiatedandimplemented.

Therearemanymethodsforstudyinghumantissue as an indicator of breast cancer such asmolecular structure, protein crystallography orbiomarkers evaluation (7, 8). Diffraction effectproduces a characteristic scattering pattern ofthe tissue which has been irradiated. Thissignature is dependent upon the molecularcomposition of the target and hence could beusedtocharacterizethetissue(8-10).

James et al. (11) presented a method using

A. Maziar1, D. Shahbazi-Gahrouei1*, M.B. Tavakoli1, V. Changizi2, Z. Ghasemian1

1DepartmentofMedicalPhysics,SchoolofMedicine,IsfahanUniversityofMedicalSciences,Isfahan,Iran

2DepartmentofMedicalPhysics,TehranUniversityofMedicalSciences,Tehran,Iran

ABSTRACT

Background: An early diagnosis of breast cancer relates directly to an

accurate treatment plan and strategy. Early detec�on of breast cancer before

its development would be a significant reduc�on of morbidity and mortality

rates. The aim of this preliminary study is to inves�gate the sensi�vity of

Wide Angle X-ray diffrac�on (WAXRD) method on women hair samples of

healthy and breast cancer pa�ents in comparison with other modali�es such

as synchrotron based XRD beam and mammography. Materials and

Methods: Hair samples were taken from occipital region of skull from healthy

and breast cancer pa�ents (43 women) were analyzed using X-ray diffrac�on

and the results were analyzed and compared with mammography and

pathology reports. Results: The results of analyzed samples showed the

sensi�vity for purposed WAXRD method was 86% in comparison with

synchrotron based XRD beam (64%) and also with mammography (70%).

Conclusion: This non-invasive method is less harmful and is more sensi�ve

than the two other methods and help the physicians for choosing accurate

treatment plan.

Keywords: X-ray diffraction (XRD), breast cancer, hair, detection, non-invasive.

*Correspondingauthor:Dr.DaryoushShahbazi-Gahrouei,

Fax:+983137929032

E-mail:shahbazi@med.mui.ac.ir

Revised: Feb. 2015

Accepted: March 2015

Int. J. Radiat. Res., April 2016; 14(2): 153-158

► Technical note

DOI: 10.18869/acadpub.ijrr.14.2.153

synchrotron based X-Ray diffraction (XRD)whichproduceddiffractionpatternsofhairandshowed a good correlation between healthystate of hair samples and related pattern.Saengkaew et al. (12) presented an analysis ofhuman-hair microstructures by wide-angle X-raydiffractions(WAXRD)andsmall-angleX-rayscattering(SAXS).Inanotherstudy,Corinoetal.(13)madearesearchwithaspecialsampleholderand get the related results between molecularstructureof hair andpresenceofdisease.Theyreported that synchrotron-derived X-raydiffractionhasthepotentialabilitytoprovideanon-invasive method to show the presence ofbreastcancer(13).

Several studies by other researcherscon*irmedanassociationbetweentheXRDhairpatterns and the presence of breast cancer (14).Limitations including of the complexity ofinterpretation the data and lack ofsynchrotron based on X-ray diffractionapparatus in all medical experiments centers,caused evaluation of XRD patterns using afacilityother thansynchrotron.For thisreason,investigationofchangesindiffractionpatternofhair samples of healthy and breast cancerpatients using Wide Angle X-ray Diffraction(WAXRD)wasproposed.Attheend,advantagesand disadvantages of the purposed methodagainstprevioussynchrotronbasedmethodandmammographywasinvestigated.

MATERIALSANDMETHODS

In this study, hair samples of 43 women(including 14 healthy, 17 patient and 12suspicious)individualsinfourcancertreatmentcentersofTehranhospitalswereused.Healthyindividuals were women those physicalexaminations and mammography reported ashealthy, and cancer cases were those patientsthat had positive physical examination,mammographyandalsopathology reports.Thesuspicious individuals whom mammographyreports were reported normal but the clinicalevidencesandexaminationsweresuspicious.

All the sample donors were Iranian withArian origin and with the age between 33-75

yearsoldwithaverageof51.3±11.5years.Thehairsamplesobtainedfromwomenwhohadnocolored hair for 6 weeks or more and hairsamples of cancer cases were collected frompatients who had not delivered chemotherapyand medicine therapy. Hairs were cut fromoccipital region of skull and taking throughsample preparation method introduced by theInternationalAtomicEnergyAssociation (IAEA)(15). In order to cleaning of samples from anykind of external contamination, coded sampleswere taking washing method with ethanol anddistilledwaterforthreetimesandthendriedinoven about 30 minutes at 80 oC temperature.Then, the samples were cut with surgicalstainless steel scissors into *ine pieces about 2mm and using a mortar the samples becamepowderedanddryinginovenforabout10hoursatan80oCtemperature(16).

Wide Angle X-Ray Diffraction (WAXRD) wasusedasmeasurementmethod.Theprocedureofsample preparation was as follow. The amountof0.4gofeachpowderedsample*illedinspecialsample holders of XRDmeasuring system with10 × 10 mm2 exposure window and measuredwith X’ Pert Pro MPD (PANalytical Inc.,Netherlands)XRDsystemwithcopperanode in40 kV and 40 mA at 25 oC temperature. Thedetector was high resolution Germanium (Ge)semiconductor PIXcel (PANalytical Inc.,Netherlands) detector. Starting and endingpositionsofeachexposurewere2.02and89.97degrees,respectivelywith0.02ostepsizeand48secondsstepscantime.Theproportionalcurvesinintensity(counts)versus2θ(exposureangle)weredrawnandstudied.

The WAXRD results of coded hair sampleswere analyzed by XRD experts. Theseresultswerematchedwith donors’ health statedata. The results of healthy and cancer patientgroupswerecomparedwithmammographyandalsopathologyreportsandbasedontheresults,the sensitivity of the method versusmammographywerealsoevaluated.

XRD curve of healthy and also breast cancerpatientswereproceedandstatisticsparameterssuch asmean and standard deviation values ofpeak heights (counts) and peak positions (θ)were analyzed with IBM SPSS (version 22.0,

154

Maziar et al. / Detection of breast cancer using XRD

Int. J. Radiat. Res., Vol. 14 No. 2, April 2016

2013)software.To show normal distribution of data,

Shapiro-Wilktestwasused.Students’t-testandnon-parametric statistical tests were used tocompare the data. To check the assumption ofcorrelationdatawereusedPearsoncorrelationcoef*icientwasusedtoshowcorrelationofdata.P value <0.05 was considered as signi*icantlevel.

An example of X-ray diffraction pattern ofhealthywomanhairsandhairsfrompatientwithbreastcancerisshownin*igure1.Asthis*igureshows, twodistinguishablepeakswithdifferentheights (counts) arepresented.Also, diffractionpattern of normal human hair and hair from apatient with breast cancer using synchrotronbeamXRDmethodisshownin*igure2.

RESULTS

Results of WAXRD experiment of hairsamples belong to healthy and breast cancerpatientsareshownin*igure3.Asitseen,1stand2ndpeaksheightofbreastcancerpatientswhichare shown in counts (height) is more thancounts related to healthy samples. So, X-ray

diffraction patterns of hair from breast cancerpatientshavemoreintensitythanthatofnormal.

Figure 3 also demonstrates the comparisonbetween healthy and cancer patients. Inaddition, in each group, the results werecompared with mammography and alsopathologyreportofeachindividualandtheyareshownintable1.

Maziar et al. / Detection of breast cancer using XRD

Int. J. Radiat. Res., Vol. 14 No. 2, April 2016

Figure 2. Diffrac�on pa/ern of (A) normal human hair and (B) hair from a pa�ent with breast cancer using synchrotron beam

XRD. The diffuse ring that has shown in (B) indicates the presence of tumor.

Figure 1. Examples of X-ray diffrac�on pa/ern of healthy human hairs (a) and hairs from pa�ent with breast cancer (b) which

showing two dis�nguishable peaks but in different heights (counts).

155

156

Maziar et al. / Detection of breast cancer using XRD

breast cancer with changes in hair structureusingWAXRDmethodwereconsidered.

In this study, the average age of 43women was 51.3 ± 11.5, which is less thanscreening populations (16). Since inmammography there is radiation hazards topatients, and the sensitivity of

Table 1. Comparison of obtained results and their sensi�vi�es of different studied groups among three methods of WAXRD,

mammography and pathology.

# WAXRD Mammography Pathology Sensi vity

P* N** P N P N

Healthy 14 2 12 - 14 - 14 85.7

Suspicious 12 9 3 7 5 9 3 58.3

Cancer 17 17 - 17 - 17 - 100

Total 43

Note: *Pa�ent, **Normal

DISCUSSION

This study was performed to evaluate theef*iciency of WAXRD in detection of breastcancer as a safe and non-invasive method. Inaddition, comparing the results of pathology asgold standard and mammography in detecting

Figure 3. Comparison charts of normal (N) and breast cancer pa�ents (P) for 1st

peak (a) and 2nd

peak (b).

Int. J. Radiat. Res., Vol. 14 No. 2, April 2016

157

Maziar et al. / Detection of breast cancer using XRD

mammographyisdecreaseswithincreaseofage(16), the goal of this study was todetermine if differences between XRD curve ofhealthy women and breast cancerpatients can lead to early detection of breastcancerorpredictionofdiseasewithoutradiationhazardsornot.

In this research, thedifferencebetweenXRDcurve of healthy and cancer patients werestudied and compared. As *igure 1 showed,there are two separate and distinguishablepeaks in 9.8 AE and 4.4 AE which re*lectsd-spacing of hair structure. These data werecompatible with those obtained by previousstudy using synchrotron source which, ringsrelatedtod-spacingofhairstructurehavebeenreported (9-13). But, as shown in *igure 3, therewasanadditionalringre*lects4.7nmd-spacingincancerpatients (12, 14).The9.8AE re*lectiond-spacingreferstostructurepatternofthealpha-helix diameter and the 4.4 AE re*lection isreferring to the periodic distance of the 3.6-alpha-helixstructure(18).

Asitisshownin*igure3,theoverallheightof1st and also 2nd peak inpatient group ishigherthanthatofnormalwomenanddifferenceintheheight means of two group for 1st peak issigni*icant (P<0.05). Of the *irst 17 bars ofpatientrelatedbarsin*igure3-a,15ofthemhasgreater height in comparison with healthyindividuals. Hence, it can be assumed thatpresenceofdiseasecanmakedifferencesinhairstructure, especially themolecular structure ofalpha-helix (18). Formean of 2nd peak height inthetwoexaminedgroups,thedifferencewasnotsigni*icant(P<0.05).

The results in table1 indicated thatWAXRDcould provide a proper diagnostic test forpatient with breast cancer. In addition, in twogroups of suspicious and cancer, the resultswerecompletely thesameasresultsofWAXRDand pathology reports. In cancer group, theresults were the same among all three studiedmethods. In comparison between results ofWAXRD and pathology, the detection rate ofbreast cancer by WAXRD was 60.4%. Also,sensitivityofapproximately86%wasachieved.The results obtained here were more accuratethanthosereportedbyCorinoetal.(64%)(6).

When comparing the age of donors withresultsofthemethod,therewasnocorrelationbetween age and average height 1st peak ofWAXRD.Thisresultisagainsttheresultswhichreported by Corino et al. (17). It can be duo tolessnumberofsamplesthatweretestedinthiswork.

Comparing with mammography, the totalprocedure time, includes sample preparation;XRD examination time and the curveinterpretation, were taking about 48 hourswhich is longer than mammography. Fromexpense point of view, it was clear that theprice of XRD procedure is half of that formammography. As it is shown in table 1, thesensitivityofXRD(86%)washigher than thatofmammography(64%).

Therefore, identi*icationof breast cancer atan early stage, using XRD technique may beprovidedanalternativeearlylow-cost,withoutradiation hazards, non-invasive and moresensitive method. Ultimately, there are lookpromising in detecting breast cancer by XRDmethod.

Of course, further study should be done toexamine more hair samples of breast cancerpatients.Inaddition,theneedtoprovidemoreexperiments is of utmost importance. Inconclusion, the *indings of this work may beopenaneweratoinvestigatewhetherWAXRDorothermodalitiessuchasmammographyandalso synchrotron basedmethods are useful todiagnosis of early stages of breast cancerpatients.

ACKNOWLEDGMENT

This work is a part of PhD thesis which

-inancially supported by Isfahan University of

MedicalSciences(MUI)(GrantNo.391456).

The authors would like to acknowledge the

Radiation Oncology Department of Cancer

Institute, TehranUniversity ofMedical Sciences

forsupportingthisresearch.

Con�lictofInterest:Declarednone.

Int. J. Radiat. Res., Vol. 14 No. 2, April 2016

158

Applied Radiaon and Isotopes, 68(12): 2237–45.

10. Kidane G, Speller RD, Royle GJ, Hanby AM (1999) X-ray

sca/er signatures for normal and neoplas�c breast

�ssues. Phys Med Biol, 44: 1791–1802.

11. James V, Kearsley J, Irving T, Amemiya Y, Cookson D

(1999) Using hair to screen for breast cancer. Nature, 398

(6722): 33-4.

12. Saengkaew P, Ussawawongaraya W, Khaweerat S, Rugmai

S, Ouajai S, Luengviriya J, et al. (2011) A preliminary X-ray

study on human-hair microstructures for a health-state

indicator. World Academy of Science, Eng Technol,

59: 1945-9.

13. Corino GL and French PW (2008) Diagnosis of breast

cancer by X-ray diffrac�on of hair. Internaonal Journal of

Cancer, 122(4): 847-56.

14. Mistry DAH, Haklani J, French PW (2012) Iden�fica�on of

breast cancer associated lipid in scalp hair.

Breast Cancer: Basic and Clinical Research, 6: 113-123.

15. Interna�onal Atomic Energy Agency (IAEA) TECDOC-950

(1997) Sampling, storage and sample prepara�on

procedures for X-ray fluorescence analysis of

environmental materials.

16. Gholizadeh N, Kabiri Z, Kakuee O, Saleh-Kotahi M,

Changizi V, Fathollahi V, et al. (2013) Feasibility of breast

cancer screening by PIXE analysis of hair. Biological Trace

Element Research, 153 (1-3): 105-10.

17. Corino GL, French PW, Lee M, Ajaj MM, Haklani J, Mistry

DA, et al. (2009) Characteriza�on of a Test for Invasive

Breast Cancer Using X-ray Diffrac�on of Hair-Results of a

Clinical Trial. Breast Cancer: Basic and Clinical Research,

3: 83-90.

18. Pauling L, Corey RB, Branson HR (1951) The structure of

proteins; two hydrogen-bonded helical configura�ons of

the polypep�de chain. Proceedings of the Naonal

Academy of Sciences of the United States of America,

37(4): 205-11.

REFERENCES

1. WHO World cancer report 2012

2. Shahbazi-Gahrouei D (2003) Possible effect of background

radia�on on cancer incidence in Chaharmahal and

Bakh�ari province. Iran J Radiat Res, 1(3): 171-174.

3. Sasieni PD, Shelton J, Ormiston-Smith N, Thomson CS,

Silcocks PB (2011) What is the life�me risk of developing

cancer? The effect of adjus�ng for mul�ple primaries.

Br J Cancer, 105(3): 460-5.

4. Hagen AI, Kvistad KA, Maehle L, Holmen MM, Aase H, Styr

B, et al. (2007) Sensi�vity of MRI versus conven�onal

screening in the diagnosis of BRCA-associated breast

cancer in a na�onal prospec�ve series. The Breast, 16(4):

367-74.

5. Carney PA, MiglioreN DL, Yankaskas BC, Kerlikowske K,

Rosenberg R, Ru/er CM, et al. (2003) Individual and

combined effects of age, breast density, and hormone

replacement therapy use on the accuracy of screening

mammography. Annals of Internal Medicine,

138(3): 168-75.

6. Théberge I, Chang SL, Vandal N, Daigle JM, Guer�n MH,

Pelle�er E, Brisson J (2014) Radiologist interpre�ve volume

and breast cancer screening accuracy in a Canadian

organized screening program. J Natl Cancer Inst,

106(3): djt461.

7. Rawashdeh MA, Lee WB, Bourne RM, et al. (2013) Markers

of good performance in mammography depend on

number of annual readings. Radiology, 269(1): 61-7.

8. Maziar Asghar, Shahbazi-Gahrouei D, Tavakoli Mohammad

Bagher, Changizi Vahid (2015) Non invasive XRF analysis of

human hair for health state determina�on of breast �ssue.

Iran J Cancer Prev, 8(6): e3983.

9. Chaparian A, Oghabian MA, Changizi V, Farquharson MJ

(2010) the op�miza�on of an energy-dispersive X-ray

diffrac�on system for poten�al clinical applica�on.

Maziar et al. / Detection of breast cancer using XRD

Int. J. Radiat. Res., Vol. 14 No. 2, April 2016