A Study of Effect of Energy Healing on in Vitro Tumor Cell Proliferation

8/3/2019 A Study of Effect of Energy Healing on in Vitro Tumor Cell Proliferation

1/8


2/8

THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINEVolume 5, Number 4,1999, pp. 359-365Mary Ann Liebert, Inc.

A Study of the Effect of Energy Healing onIn Vitro Tumor Cell ProliferationSONAL SHAH,lALFREDT. OGDEN,2CHRISTIAN M. PETTKER,2

ANTHONY RAFFO,Ph.D.,2SILVIUITESCU,MD., and MEHMETC. OZ, MD.2

ABSTRACTThis study examined the effect of energy healing on in vitro tumor cell growth using the cellculture model similar to that embraced by oncologists to assess the effect of chemotherapeu-tic agents. After selecting an energy healer based on his ability to influence this model, weassessed the effects of energy treatment compared to cells left at ambient temperature and toa control treatment consisting of a medical student mimicking the healer. A chi-square testcomparing a medical student's and the practitioner's ability to inhibit tumor cell growth by15%associates our practitioner with inhibition of tumor cell proliferation (p = 0.02).We alsofound that the magnitude of change was too close to the assay's intrinsic margin of error, thusmaking our quantitative data difficult to interpret. Although energy healing appears to in-fluence several indices of growth in in vitro tumor cell proliferation, these assays are limitedin their ability to define and prove the existence of this phenomenon. More sensitive bio-logical assays are needed for further study in this field.

INTRODUCTION

Allopathic practitioners of Western medi-cine attribute ailments to observable dys-functions of organs or to the body's regulatorymechanisms. Although the placebo effect sug-gests that the mind can cause myriad physicalchanges in the body, the mind-body relation-ship still remains a mystery. Medical scientistswould concede that a person's mental state caninfluence his or her physical state, but are moreskeptical of the idea that a person can mentallyinfluence aspects of another person's body.Western science does not currently have a par-adigm with which to understand therapies inwhich practitioners purport to deliver or chan-nel energy to a patient, or therapies in whichpractitioners, using physical or metaphysical

means, manipulate energy fields. For this rea-son, healthcare practitioners are generally moreopen to physically tangible forms of comple-mentary medicine such as massage therapy oryoga, but are suspect of healers who claim totreat patients through intense concentration ormeditation. This negative bias persists despitethe presence of over 150 studies (see Benor,1992 for a review) claiming the effect of inten-tional healing.

These studies often fail to convince physi-cians practicing at large academic centers. Oneproblem is inconsistent adherence to strict sci-entific method and analysis. For example, asham treatment, a criticalcontrol ofnonspecificor unintended interactions, was omitted fromone of the most compelling studies (Wirth,1992). Another study with impressive results

lWasrungton University School of Medicine, St. Louis, Missouri.2Columbia University, New York, New York.

359


3/8

360

failed to report thoroughly their experimentalprotocols (Chien et al., 1991).Secondly, there isdiminished credibility of studies in which in-vestigators have published without trying togenerate an adequate sample size (Rein, 1995)or have failed to report their findings in peer-reviewed journals (Chien et al., 1991; Rein,1995).Finally, we found that many allopathicphysicians are biased against studies that areperformed in small privately funded centersfor alternative medicine. Though the data andmethods of these studies may be sound, theirreplication at large allopathic academic centerscould boost their credibility in the eyes of tra-ditional allopathic physicians.

This study attempts to address some of thesemethodological problems by documenting theeffect of one form of intentional healing on asimple biological model. We chose an in vitrotumor cell proliferation assay because it is spe-cific and quantifiable, and because previousstudies suggest that energy healers may be ableto retard tumor cell growth (Rein, 1995;Snel etal., 1995;Chien et al., 1991).Furthermore, oneproposed mechanism of intentional healing isthat healers are emitting electromagnetic (orother) energy fields. Low-frequency electro-magnetic fields have been shown toaffect onco-genic mechanisms (Dees et al., 1996;Liburdy etal., 1993).

METHODSStudy overview

Phase I. Initially, healers were screened inor-der to find one that had the highest likelihoodof efficacywithin the constraints of our model.Prospective healers were given one of twoidentical plates of MCF-7 cells (see below) totreat using their modality of choice. Addition-ally, a student performed a sham treatment onan identical plate of cells by mimicking the ac-tions of the healer. This plate served as a neg-ative control. After 24hours of incubation, thecells were counted and compared. We consid-ered a healer effective if he or she could twoout of the three times demonstrate at least a15% decrease in cell number when comparedto the control.

SHAH ET AL.Phase II. The healer with the most impres-

sive results from the screening phase was usedin a more extensive experimental phase con-sisting of four cell lines and two more condi-tions: (1) a plate of cells which remained in theincubator at 37 C,ermed "untouched"; and (2)another left out in the laboratory at room tem-perature, termed "ambient." The four experi-mental conditions are as follows: (E): experi-mental (treatment); (C): control (shamtreatment); (U): untouched; and (A) ambient.

Treatment lasted 1 hour. At treatment time,the cells were removed from the incubator,brought to the experimental room, and placedon a desk with a chair in front of it. Our prac-titioner then held his hands over the cells,sometimes pointing at the wells, sometimesholding his palms over the cells. Once or twiceduring the treatment, he gently shook the cellswithout lifting them, using a slow back-and-forth movement. Also, he would often make awiping motion above cells that looked like hewas pulling something out of the air. For ap-proximately the last 20 minutes of treatment,the practitioner sat down in the chair and me-diated. During each treatment, a studentwatched the practitioner from the other side ofthe room and noted specificmovements. At notime did the practitioner remove the protectivecovering of the cell plates. After treatment, thestudent replaced the cell cultures in the incu-bator. Toprovide a negative control, either be-fore or after the experimental treatment (in arandomized fashion), a student performed asham treatment, in the same room and positionand for the same amount of time. Afterwards,the student replaced the cells in the incubator.Once all the cells were back in the incubator, ascientists not involved in the study labeled theplates in order to blind the investigators in allsubsequent steps. The plates and their corre-sponding treatments were matched only afterall the counts were read and analyzed.Pra ctitioner

Our practitioner, Frank Huo, has studiedYuanji medicine for 18 years and currentlypractices under the Yuanji ScienceWorldwideCorporation. Yuanji medicine is a traditionalChinese science dedicated to healing the dis-


4/8

IN VITRO TUMOR CELL PROLIFERATIONeased body and mind. The Yuanji practitionerfocuses on collecting universal energy andtransferring that energy from practitioner tosubject. Although bodily balance is the goal ofYuanji practices, our practitioner felt that hecould transfer energy to any subject.Addition-ally, he has received a B.S.in Material Scienceand an M.s. in Plastics and Processing Engi-neering (BeijingUniversity ofAeronautics andAstronautics).C ultu ring cellsFresh cells (pO)were frozeninmultiple aliquotsand stored at -80 C in fetal bovine serum(GibcoBRL,Gaithersburg, MD) supplementedwith 10%dimethyl sulfonate (DMSO)(Sigma,St. Louis, MO). Cells were cultured in RPMImedium supplemented with 10%fetal bovineserum and 100 U/mL penicillin and strepto-mycin (Gibco BRL).The cells were incubatedat 37 Cn 5%CO2 and were maintained for tenpassages before a new aliquot would bethawed from the parent batch.P ro to co l for p ractitioner scree ning (phase I)

Twenty-four hours before an experiment,MCF-7cellswere trypsinized (Trypsinin 0.05%EDTA,GibcoBRL)and passed on to two 6-welldishes at a concentration of 30,000 cells perwell. One dish would be used as the experi-mental and one would be the control. Twenty-four hours after treatment, cells weretrypsinized against, spun (1000rpm for 5 min-utes) and resuspended in 1 mL of phosphate-buffered saline (PBS).Cels were stained withtrypan blue (1:10trypan blue:cell suspension)and counted using a hemacytometer (Sigma).Assays were done in sets of six. The meancounts for the six wells were taken to be the to-tal cell count for that condition.P rotocol for experim ental trial (ph ase II)

Cells were trypsinized (except K562,whichare nonadherent), counted, and passed on to 6wells of four 96-well dishes at a concentrationof 3,000 cells per well. On the day of the ex-periment, one plate was left out in the experi-mental room for 5 hours, one plate was left inthe incubator, and two others were used as the

361

experimental and the control. Eighteen to 20hours before the time of assay 0.2mCi/well of3-methyl thymidine (Sigma)was added. At thispoint (24hours after treatment), cellswere har-vested (Wallac 1295-001 Cell Harvester,Gaithersburg, MD) and counted (Wallac 1205Betaplate Liquid Scintillation Counter). Rawnumbers from six independent wells pertain-ing to a particular celltype and treatment wereaveraged to obtain a single data point, themean cell count. In the caseof the positive con-trol, a cocktail of 4 micromolar cis-platinum(II)-diamine dichloride and 4 micromolar ta-moxifen (Sigma) was added to the cells priorto adding thymidine.C ell lin es

The following four human cell lines wereused: MCF-7,K562, MEL6,and LNCaP. MCF-7 cells have been reported to respond to weakelectromagnetic fields (Dees et al., 1993;Liburdy et al., 1996)and is an ER+, breast can-cer line. MEL6 is a melanoma line with a p53knockout. LNCaP is a prostate metastatic linethat was cultured from a patient's lymph node.K562 is a leukemia line that responds particu-larly well to natural killer cell activity. MCF-7,K562, and LNCaP are commercially available(ATCC). MEL6 cells were a kind gift of Dr.Tony Raffo, Department of Oncology, Colum-bia University.Statistica l m eth ods/d ata a nalysis

When averaging the counts for the sixwells,any number that was more than two standarddeviations away from the mean of the remain-ing five numbers was discarded. If more thanhalf of the wells were outliers, the entireexperiment was considered unreliable andwas discarded. Mean counts were applied tothe following equation (l-experimental/un-touched) x 100 to obtain a value for effect. Anegative number indicated growth relative tothe incubator; a positive number indicated in-hibition (or cell death) relative to the incuba-tion.

A two-tailed paired t-test compared controland experimental groups for significant dif-ferences. A one-way analysis of variance


5/8

362 SHAH ET AL.TABLE1. RESULTSFPRACTITIONERcREENING decrease in cell number (relative to the un-

Practitio ner typ e Tri al 1 Trial 2 Tria l 3 touched plate) the data was analyzed by con-verting our continuous data into binary dataTherapeutic touch 1 25a -10 0 and then using a chi-square test.Therapeutic touch 2 -17 -36 3Energy healer 15 -15 -6Yuanji healer 31 29 18 RESULTSaNumbers are expressed as percent inhibitions calcu-lated with the following equation: (1-treated/control)*100. Table 1demonstrates that the first three prac-

titioners did not meet our study requirements.(ANOVA) demonstrated whether there were TheYuanji healer, on the other hand appearedsignificant differences between the four to consistently inhibit proliferation. The firstgroups. To evaluate whether the experimental four columns of Table2 show the raw data fortreatment was significantly associated with a each experiment in Phase II. The two columns

TABLE2. RAw DATAANDCALCULATEDERCENTNHrBlTIONSOREACHTRIALRaw data Percent inh ib itio n

Untouched (U ) Ambient (A) Contr ol (C) Experimental (E ) c/Ua E/U EICMCF7 (n = 7) 1 513483b 370554 523792 367433 -2 28.4 29.92 222617 230787 240883 240555 -8.2 -8.1 .13 154585 157179 161581 145943 -4.5 5.6 9.74 617311 545501 570789 557955 7.5 9.6 2.25 113414 69093 103825 101915 8.5 10.1 1.86 554141 474517 589549 619468 -6.4 -11.8 -5.17 67768 54125 63535 67065 6.2 1 -5.6Mean 320474.1 271679.4 321993.4 300047.7 0.2 6.3 4.7SEM 87803.2 73848.5 87412.2 83653.4 2.7 5.0 2.1K562 (n = 7) 8 228691 241775 242937 229458 -6.2 -0.3 5.59 28739 19428 33200 25183 -15.5 12.4 24.110 409636 565457 584563 560361 -42.7 -36.8 4.111 156326 200674 203713 203247 -30.3 -30 .212 301664 148202 295489 248945 2 17.5 15.813 71509 43589 63104 67211 11.8 6 -6.514 89958 109854 112501 101380 -25.1 -12.7 9.9Mean 183789.0 189854.1 219358.1 205112.1 -15.1 -6.3 7.6SEM 51898.1 69430.3 70788.1 67413.3 7.2 7.9 3.8MEL 6 (n = 8) 15 151210 x 130329 103358 13.8 31.6 20.716 953 969 1147 1031 -20.4 -8.2 10.117 4034 3926 4742 3504 -17.6 13.1 26.118 21456 21379 21140 19338 1.5 9.9 8.519 18775 17227 18051 12193 3.9 35.1 32.520 15489 8983 8433 12203 45.6 21.2 -44.721 1292 974 1458 1565 -12.8 -21.1 -7.322 9915 8884 11164 13269 -12.6 -33.8 -18.9Mean 27890.5 8906.0 24558.0 20807.6 0.2 6.0 3.4

SEM 17834.2 2990.2 15327.0 12013.8 7.7 8.8 9.1LNCaP (n = 4) 23 165937 125727 165315 122680 0.4 26.1 25.824 92873 117845 115007 114945 -23.8 -23.8 .125 11797 6165 14262 8151 -20.9 30.9 42.826 14417 14904 17982 11287 -24.7 21.7 37.2Mean 71256.0 66160.3 78141.5 64265.8 -17.3 13.7 26.5SEM 36739.9 32205.3 37258.2 31538.6 5.9 12.6 9.5apercent inhibition for X/Y is the inhibition caused by caused by X relative to Y. Values are calculated with thefollowing formula: (1 - X /Y ) X 100.bValues are expressed as counts per minute.


6/8

IN VITRO TUMOR CELL PROLIFERA nONfollowing the raw numbers contain values forpercent inhibition calculated by comparingboth the control and the experimental to the un-touched cells. In the last column, percent inhi-bition was calculated by directly comparing theexperimental cells to the control cells.

An ANOVAdemonstrated no significant dif-ference between any of the four groups whenlooking at them either within a particular cellline, or when looking at the numbers as a com-plied set (data not shown). One can see in fig-ure 1 that the differences between the groupsare slight when compared to their standard er-ror. On the other hand, for each cell line, theexperimental counts were consistently lowerthan the control counts. Using a paired t-test(Table 3) to compare these groups, we foundthat our practitioner was not able tocause a sig-nificant change ( p < 0.05),although in the K562and the LNCaP cells,the difference approachedsignificance (p =0.08for both).

Q UntouchedlE I AmbientIi) Control Experimental

50 00 I i

400000

. . .

"5c~~!1e"8 20 00

30 00

10 00

oMCF7 K562 MEL6 LNCaP

Cell Line

FIG. 1. Mean ( SEM) counts per minute for each of thefour groups and each of the four cell lines.

363TABLE 3. CONTROL AND EXPERIMENTAL PERCENT

INHJBITION ( STANDARD DEVlATION) FOR EACH CELLLINE AND THElR P VALUES CALCULATED WlTH A

Two-TAILED P ARlED t-TESTCell line n control Experimental pMCF7 7 .2 7.7 5.0 9.2 0.34K562 7 -15.1 19.1 -6.3 21.0 0.08MEL6 8 .2 21.8 6 24.9 0.24LNCaP 4 -17.3 11.9 13.7 25.3 0.08

We (bymeans of a chi square test) then askedif inhibition was significantly associated with atreatment by our practitioner (Tables 4A and4B).The results for all cell lines were combinedand subsequently divided into two groups: in-hibition or no inhibition. Inhibition was de-fined using two standards, the experimental er-ror of the assay (5%) and the inhibitory effectof the chemotherapeutic agents used as a pos-itive control (15%).The results of analyses un-

TABLE 4. CHl SQUARE TEST ANALYZED IN VARIOUS WAYSa) Inhibition is defined as greater than 5% inhibition

p = 0.012Inhibition No Inhibition

ExperimentalControl

167

1019

b) Inhibition is defined as greater than 15% inhibitionp = 0.024

Inhibition No InhibitionExperimentalControl

81

1825

c) Growth is defined as greater than 5% growthp = 0.160

Growth No GrowthExperimentalControl

915

1711

d) Growth is defined as 15% growthp = 0.349

Growth No GrowthExperimentalControl 59 2116


7/8

364 SHAH ET AL.der both criteria are significant (p =0.012and both logically and statistically from makingp =0.024,respectively). conclusions about the magnitude of the

healer's effect.DISCUSSION

Figure 1demonstrates a trend associating in-hibition of tumor cell growth with Yuanjitreatment. The chi-square tests establish thenonrandomness of this relationship. The in-significant results of the paired t-tests may beattributable to the sample size particularly inthe LNCaP and K562cell lines where p = 0.08for both.

Figure 1also demonstrates a weakness in theexperimental model. We added the untouchedgroup and the ambient group to the study togive us a sense of the precision of the assay andso we could see the effect of moving the cellsout of the incubator. We expected the un-touched cells,which never left the incubator, tohave the highest rate of proliferation as theywere grown in optimal conditions. Similarly,we expected the ambient group, kept at roomtemperature for 5hours, tohave the lowest rateof proliferation. Accordingly, if the ambientand the untouched groups demonstrated sim-ilar levels of proliferation, it would be evidentthat these cell lines are not sensitive to move-ment and moderate changes in temperature. Infact, for three of the four cell lines, the ambi-ent group did show the least amount of prolif-eration.

This would suggest that the control (sham)group, which was exposed to ambient condi-tions for 1 hour, should have proliferated lessthan the untouched group. However, in two ofthe four cell lines, the control group showedhigher rates ofproliferation than the untouchedgroup and thus from our data, it is impossibletomake global statements about the effectof re-moving cells from the incubator and this ex-ample illustrates the inherent imprecision in themodel that may have contributed to our inabil-ity to quantify the energy healing effect.

This imprecision does not invalidate the re-sults of the chi-square test. It is highly im-probable that random error alone could gener-ate the association we observed between tumorcellinhibition and our healer's treatment. How-ever, the large random error does preclude us

CONCLUSIONOur study would have been more informa-tive if we could have asked our practitioner to

demonstrate both a positive and negative effecton the same cells. We intended to have a sec-ond experimental group in which our practi-tioner would attempt to potentiate tumor cellgrowth, but he was uncomfortable with thethought of using his abilities to cause "harm-ful change." In fact, all of the practitioners weinterviewed felt that they would only be capa-ble of effecting a therapeutic change. Mostpractitioners also expressed misgivings abouttheir abilities toaffectan isolated biological sys-tem like cultured tumor cell as they usuallywork towards restoring health at the organis-mal level. Even though simple models lendthemselves to rigorous scientific experimenta-tion, they might not be appropriate when eval-uating certain energy healing modalities.

This study demonstrates that further criticalscientific evaluation of this field is warrantedand necessary. In the future, investigatorscould adopt a bidirectional approach that ac-commodates both the investigator and thepractitioner. For example, one could measurethe effect of energy healing on T-cellprolifera-tion versus tumor cell proliferation. To assessenergy healers who are restricted to integratedbiological systems, whole animal models couldbe used.

ACKNOWLEDGMENTThe authors would like to thank Alan Wein-

berg, M.S., for his help with the statisticalanalysis of this data.

REFERENCESBenor D. Lessons from spiritual healing research andpractice. Subtle Energies 1992;3:73--88.Chien C, Tsuei J J , Lee S, Huang Y, Wei Y. Effect of emit-


8/8

IN VITRO TUMOR CELL PROLIFERA nONted bioenergy on biochemical functions of cells. Am JChin Med 1991;14:285-292.

Dees C, Garett S, Henley D, Travis C. Effects of 60-Hzfields, estradiol and xenoestrogens on human breastcancer cells. Radiat Res 1996;146:444-452.

Grad B. A telekinetic effect on plant growth. Int J Para-psychol 1964;6:473-498.

Liburdy RP, Sloma TR, Sokolic R, Yaswen P. ELF mag-netic fields, breast cancer, and melatonin: 60 Hz fieldsblock melatonin's oncostatic action on ER+ breast can-cer cell proliferation. J Pineal Res 1993;14:89-97.

Rein G. The scientific basis for healing with subtle ener-gies. In: Laskow L, ed. Healing with Love. San Fran-cisco: Harper, 1995.

Snel FW, Van Der SijdePc. The effectofparanormal heal-ing on tumor growth. Sci Exploration 1995;9:209-221.

Wirth D. The effect of non-contact therapeutic touch on

365the healing rate of full thickness dermal wounds. Sub-tle Energies 1992;1:1-20.

BIBLIOGRAPHYEisenberg DM, Kessler RC, Foster C, Norlock FE,Caleins

DR, Delbanco TL. Unconventional medicine in theUnited States. N Engl J Med 1993;324:246-252.

Address reprint requests to:S on al S ha h

6323 Rosebury, Apartment 1 WClayton, MO 63108

E-mail: [email protected]

Documents

A Study of Effect of Energy Healing on in Vitro Tumor Cell Proliferation