[CANCER RESEARCH 56. 1W>0-1663, April I. 1996]

Measurement of Potential Doubling Time for Human Tumor Xenografts Using theCytokinesis-Block MethodLynn Hlatky,1 Mariusz Olesiak, and Philip Hahnfeldt

Joint Center for Rttdiution Therap\. Department of Radiation Oncology. Harvard Medical School, Boston. Massachusetts 02115

ABSTRACT

Estimates of the potential doubling time (/,„„!of seven human tumorxenografts were made using a cytokinesis-block method. This method is

currently being investigated as an alternative to flow cytometric assaysusing the administration of a thymidine analog in the measurement ofTpM. If perfected, the cytokinesis-block method of measuring Tfn, would

be advantageous as a predictive assay, in that no label is administered tothe tumors in situ. Xenografts were grown in nude mice, and followingtumor excision and disaggregation, tumor cells were cultured with thecytokinesis-blocking agent cytochalasin B. The flux of cells through mito

sis was marked by the accumulation of multinucleate cells. By countingthe total number of nuclei as a function of time, the effective populationgrowth was observed. Tpal values were obtained by fitting suitable exponential least squares curves to the data, with the doubling time indicatedby the fitted functions. For the seven tumors studied, a significant spreadin growth rates was observed. '/',„„values generated by this method ranged

from ~2 days for a rapidly growing squamous cell carcinoma of thepharynx, FaDu, to —¿�7.5days for the slower growing glioblastoma multi-forme U251-MG. These values are compared with standard 5-iodode-oxyuridine 7"polmeasures and volume doubling times obtained by Perez

et al. (Cancer Res., 55: 392-398, 1995) for the same tumor Xenografts.

Although individual T^, values varied between these methods, the ranking of the seven tumors in order of rp(ll times was the same regardless ofmethod. In addition to estimating '/',„„.for each of the tumors, the fraction

of clonogenically dead cells that was microscopically apparent, includingapoptotic cells and cells expressing micronuclei, was determined as afunction of time in culture. Tracking this in vitro cell loss rate providesinformation on the adjustment of these primary tumor cells to in vitroculture, a factor that needs to be addressed when determining how in vitromeasurements of 7"pl)1can be effectively related to in vivo measurements.

INTRODUCTION

Certain tumors exhibit rapid cellular proliferation, which may leadto reduced local control under standard therapeutic regimens. If rapidly proliferating tumors were identifiable prior to treatment, alternative treatment protocols, such as accelerated fractionation schemes,could be considered. Thus, much effort has been invested in characterizing the kinetic behavior of individual tumors in an attempt topredict their therapeutic responsiveness (1-10). The T^, (11) of atumor is a first-order estimate of the tumor growth rate and is con

sidered a better indicator of the tumor repopulation rate during radiotherapy than is the pretreatment volume doubling time. The 7polmeasure characterizes tumor growth in terms of cell cycle time andproliferation fraction. It ignores tumor cell loss, cell cycle diversity,and time-dependent changes in the growth fraction. For particular

histologies, e.g., tumors of the head and neck, individual tumors thatexhibit short Tp<nvalues are considered candidates for acceleratedfractionation (12). Presently, the predictive value of Tpotmeasures are

Received 6/27/95: accepted 2/1/96.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked aihenisemem in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1To whom requests for reprints should be addressed, at JCRT/SWRL. HarvardMedical School. 330 Brookline Ave.. Boston. MA 02215. Phone: (617) 432-2803; Fax:(617)432-3779.

2 The abbreviations used are: Tp,,,, potential doubling time; ATCC. American Type

Culture Collection; MGH, Massachusetts General Hospital; SCC, squamous cell carcinoma; IdUrd. 5-iododeoxyuridine; BrdUrd. bromodeoxyuridine.

being assessed in clinical trials using accelerated and conventionalradiotherapy schemes (4, 5. 13, 14).

Tpot is conventionally estimated by a single in situ labeling of thetumor with a thymidine analog such as BrdUrd or IdUrd, followedafter a lag time by a tumor biopsy and flow cytometric analysis(15-21). Because rpot seems to be a clinically useful parameter,

alternative noninvasive methods to measure TpM are being investigated (22, 23). Shibamoto and Streffer (23) demonstrated the potentialvalidity of a cytokinesis-block method for estimating rpol values of

human tumor xenografts. Here, we further investigate this method byexamining the kinetics of seven human tumor lines using the cytokinesis-block method, estimating rpm values, and further considering

the potential of this method for investigating the fine structure oftumor population kinetics. rpm values obtained by the cytokinesis-block method are compared with published IdUrd 7"potvalues for the

same tumor xenografts (24).

MATERIALS AND METHODS

Cell Culture. In vitro estimates of 7^,, were made for the following humantumor cells grown as xenografts: U87. a glioblastoma multiforme (ATCC);HGL9, a glioblastoma multiforme (MGH); U251-MG. a glioblastoma multi-

forme (Dr. Bigner, Duke University); FaDu a SCC of the pharynx (ATCC);HCT15, a colon adenocarcinoma (ATCC); STS26T. a schwannoma, soft tissuesarcoma (Dr. Little. Harvard University); and SCC21, a head-and-neck SCC

(MGH).Xenografts of each of these human tumor lines were grown in nude mice in

the laboratory of Dr. Herman Suit (MGH). The method of tumor transplantation has been described previously (25. 26). Tissue samples were excised anddisaggregated to single cells using a digestion mixture consisting of 0.1 N HC1with 0.4 mg/ml pepsin (Sigma Chemical Co.) for a treatment time of l h at37°C.Following this, 3.3 X IO5 cells/slide were plated on CellTak-coated

slides (Collaborative Biomedicai Products) in DMEM containing 20% FCS.0.2 mg/ml gentamycin sulfate, and 2.0 p.g/m\ cytochalasin B (previouslyaliquoted in DMSO at 1 mg/ml). Coating slides with CellTak ensures that thevast majority of the cell population adheres to the slides, thereby considerablyincreasing plating efficiencies. Cytochalasin B is a cytokinesis-blocking agent

widely used in the assaying of micronuclei following clastogenic damage(27-29) and more recently in the determination of Tpol values (23). At low

concentrations (<3.0 /ig/ml), as used in this study, it is considered minimallytoxic. Cytochalasin B blocks cell cytokinesis following mitosis but does notblock mitosis itself. The result is that cells that have traversed mitosis since theaddition of the drug are clearly detectable by the presence of two separate

nuclei within a single cytoplasm.At various times after plating, slides were fixed in 3:1 ethanol:acetic acid

and rinsed in 70% ethanol. Slides were stained using the Feulgen technique,staining only the DNA (27). At each time point, the fraction of multinucleatedcells in the population was counted. Between 500 and 3000 cells were countedper time point. The percentages of multinucleated cells plotted as a function ofculture duration provide the means of estimating the time-dependent fraction

of divided cells in the populations. In addition, for these tumors, the fractionof clonogenically dead cells that was microscopically apparent, includingapoptotic cells (scored by morphological criteria; see Ref. 30) and cellsexpressing micronuclei (by morphological criteria; see Ref. 27) was determined as a function of time in culture.

Theory. A determination of 7^,, from an in vitro assay must take intoaccount the special circumstances of in vitro growth if it is to preserve itsessential clinical meaning. In place of flow cytometry, growth in culture wasassessed with a minimum of perturbation to the cell population by using

1660

on June 3, 2020. © 1996 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

USING CYTOKINESIS BLOCK METHOD

cytochalasin B to block cytokinesis while permitting karyokinesis. The numberof divisions each cell would have undergone is thus readily ascertained byscoring nuclei per cell. A plating efficiency correction was also introduced,because this is a known feature bearing on ultimate culture growth. The"Appendix" details how these two parameters are finally introduced into a

slightly modified formalism for finding T^,.

RESULTS

Following disaggregation into single cells, the xenograft populations were plated on CellTak-coated slides and incubated in thepresence of cytochalasin B. At intervals of —¿�24h after plating, cells

were fixed, and the fraction of multinucleated cells was determined ateach time point. The rate of cell division for each xenograft populationwas visualized by charting increases in the number of nuclei (over afixed cell number, e.g., 1000) as a function of time in culture. Threereplicate experiments were scored for each tumor type. Reproduciblekinetic data for the xenograft populations studied was obtained by thismethod, as is shown in Fig. 1 for cells from the glioblastoma U87.Data for all the human tumor xenografts, with replicate experimentspooled, are shown in Fig. 2. These curves were fit with exponentialsto generate rpot values. A significant spread in growth rates wasobserved for the seven tumors studied.

For the purpose of rpot calculation, initial data points of the growthcurves were fitted by least squares to exponential functions (Fig. 2).

Table 1 Values ofTpt)t and divided fraction obtained with the cytokinesis-block

method, compiled for the seven xenografts and compared with published data of Perezet al. (24) on IdUrd Tp(Jtmeasures and volume-doubling times for the same xenografts

1100•¿�-

100040 60 80 100 120 140 160 180 200

HOURS

Fig. 1.Total number of nuclei normalized to 1000cells shown as a function of time inculture for three U87 replicate experiments. Between 500 and 3000 cells were counted pertime point.

0 20 40 60 80 100 120 140 160 180 200

HOURS

Fig. 2. Number of nuclei/1000 cells versus time for all cell lines. These are essentiallyin vitro "growth curves" for the seven human tumor xenografts. The curves reflect the

pooled counts from replicate experiments for each tumor line. These plots were fit withexponentials to generate 7"^, values.

TumorlineFaDu

HCT15STS26TU87SCC21HGL9U251-MGOriginSCCof pharynx

HumancolonadenocarcinomaSchwannoma(soft

tissuesarcoma)GlioblastomamultiformeSCCGlioblastomamultiforme

GlioblastomamultiformeCytokinesis-

block rpol(days)2.0

2.52.53.95.96.27.5±0.2±0.1±0.2±0.3±0.3±0.3±

0.5Divided

fraction79.39

73.0575.7846.4645.5526.3967.10IdUrd(days')"1.3±2.2±2.4

±2.5

±2.913.9

±10.5

±O.I

0.10.20.230.31.4*Volume-doubling(days)4.54.73.03.04.78.3

a L. Perez, unpublished data.This value has since been modified downward, to ~-4.2 (L. Perez, unpublished data).

Points after curve rollover were ignored, as described in "Appendix."

The time at which the fitted function indicated a doubling of thepopulation was taken to be the rpot for the associated cell line. Thismethod is similar to that of Shibamoto and Streffer (23), except thatthe fits here allow for a fixed, nonproliferating component (see "Appendix"). The form is number of nuclei = A + ß*exp(Q).Numerical

values of rpol determined by the cytokinesis-block method are compiled in Table 1. Tpal values ranged from ~2 days for a rapidlygrowing SCC of the pharynx, FaDu, to —¿�7.5days for the more slowlygrowing glioblastoma U251-MG. Also listed in Table 1 are thevolume-doubling times and rpot values determined in situ by Perez et

al. (24) for the same xenografts using standard IdUrd techniques. It isnoted that for this set of xenografts, the cytokinesis-block methodranked the 7potvalues in the same order as did the IdUrd method. TheTpo, values obtained here by the cytokinesis-block method are alllonger than those detected by Perez et al. (24) for these xenograftsusing the IdUrd method (taking into account the updated estimate byPerez et al. of a shorter 7polvalue for the U251-MG).

The divided-fraction graph for the U87 populations demonstrates a

bimodal kinetic behavior. This is indicative of subpopulations of cellswithin the tumor, which either have different cycling rates or resumecycling in distinct cohorts after being put into culture (e.g., as wouldbe expected in mixtures of oxygenated cycling and hypoxic quiescentsubpopulations). Such information on cycle heterogeneity is not available from standard rpot assays.

In scoring data for estimating rpot, cells in the population werecounted and classified as single nucleate, multinucleate, or lethallydamaged. Cells that were visibly lethally damaged (including cellswith visible apoptotic bodies, micronucleated cells, and cells withdetectable lethal chromosome aberrations, e.g., dicentric bridges ornuclear blebbing) were counted, and the fraction of lethally damagedcells was scored as a function of time in culture, Fig. 3. Thus, besidestracking cell division, one can also track a portion of the cell death inculture. Tracking this in vitro cell loss rate provides data on theadjustment of primary tumor cells to in vitro culture and providesinformation for judging how well in vitro clonogenic assays can bemade that reflect the total cell populations for these specific tumors.The colon adenocarcinoma xenograft populations HCT15 were observed to undergo a significant degree of apoptosis (as determined bythe presence of apoptotic bodies) on transition from the in vivo to thein vitro situation. This was not seen for the cells from the otherxenografts.

1661

on June 3, 2020. © 1996 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

USING CYTOKINF.SIS-Bl.OCK METHOD

50 T

40 60 80 100 120 140 160 180 200

HOURS

Fig. 3. Fraction of cells that showed visible lethal damage (c.,i,'..micronuclei. apoptoticbodies, and hlehhing). the lethal index is shown as a function of time for the tumor lines.

DISCUSSION

The cytokinesis-block method has been suggested as an alternativeto thymidine analogue labeling for the measurement of 7"^,,. In this

study, as in the studies of Shibamoto and Streffer (23), estimates ofTpj,, values were obtained for human tumor lines grown as xenograftsand evaluated in vitro after the addition of cytochalasin B. These Tpotmeasures were then compared with 7"p<,tvalues for the same xe

nografts determined in the standard way, via in vivo labeling with athymidine analogue. If perfected as a predictive assay for humantumor kinetics, the cytokinesis-block technique would offer the dis

tinct advantage that no label is administered to the tumor in vivo.alleviating the problems associated with thymidine analog injection.Only a microscope is required for analysis; therefore, the method isaccessible to all laboratories, and normal cells such as stroma ormacrophages infiltrating the tumor sample could be excluded byvisual inspection. In addition, information on individual tumor radi-

osensitivity, via micronuclei detection, could be obtained simultaneously when assessing cytokinesis-blocked cells (27-29, 31). If

useful information for estimation of Tp,,, can be obtained from theobservation of tumor cells in vitro, disadvantages of the method arethat it is labor intensive, and, unlike flow cytometric Tp,,, measurements, data are not rapidly available, because the cells must becultured on the order of 100 h after removal from the tumor. Even so,in vitro, culture periods of much longer duration are tolerated in thepredictive assay measurements of radiosensitivity via clonogenic assays such as SF2.

When possible, 7pl)1values were obtained directly from the dataplots by noting at what times the population size N = 2000 wasachieved. Otherwise, the times at which N = 2000 in curves fitted to

initial points were used. In these cases, the nuclei/cell versus timegraphs were fitted by least squares to exponential functions. Theindividual 7"pl„times generated via this method were both significantly different and longer than the 7"pottimes derived by IdUrd

labeling (24). Yet, when the seven human xenografts were ranked inorder of 7"pl)l.both methods yielded the same ranking. (Note that for

HCT15 and STS26T, there was essentially no difference in rank.) Inthe data of Shibamoto and Streffer (23) the Tp,,, values obtained foreight tumor xenografts (human and murine) by the cytochalasin Bmethod did not vary significantly (with the exception of one murinexenograft. HP melanoma in BALB/c mice) and were not systematically larger or smaller than 7"pl),values obtained using BrdUrd label

ing. For these same xenografts, excluding HP. the ranking based onTpol values between the two methods agreed.

By counting multinucleate verxux mononucleate cells in vitro, di

vided cell fractions are obtained for the populations. Information ondivided fraction versus time augments that gained from the growthcurve, i.e., total nuclei in the population verxux time. In this study,particular tumor lines, e.g., U87. demonstrate nonhomogeneous transitional behavior in their divided-fraction curves, which show inflection points separating concave-up and concave-down behaviors.

These results are indicative of subpopulations of cells within thexenografts. which either have different cycling rates or resume cyclingin distinct cohorts after transfer to in vitro culture. It is worth notingthat both of these conditions, different cycling rates and distinct cellcohorts (nonconstant growth fraction), are inconsistent with the assumptions of Tp,,,as standardly defined. By use of the divided-fraction

curves, diversity in the cell proliferation rate could be disentangledfrom the resumption of cycling of distinct cohorts. This cytochalasin-

block assay potentially has the ability to detect heterogeneous cellularbehavior and to distinguish between asynchronous and synchronouscycling patterns, uncovering diversity in cellular kinetics not revealedby flow cytometric studies. In theory, by taking sufficient data points,mixtures of oxygenated cycling cells and hypoxic quiescent subpopulations within tumors should be discernible due to the "resumptionkinetics" of the hypoxic tumor subpopulations when disaggregated

and exposed to oxygen in vitro (32). Although assays, such as thecytokinesis-block assay, that follow proliferation over time offer a

means of measuring certain aspects of intrutumor cycle diversity, theproblematic nature of tumor proliferation heterogeneity and the time-

dependent alterations in the proliferating fraction cannot be fullyaddressed in vitro.

There are not enough tumors examined in this study to see whethercharacteristic kinetic fingerprints exist for the different tumor histologies. However, this assay potentially provides detailed informationon the adjustment of tumor populations to culture and could therebyreveal characteristic kinetic structures of specific histologies («".#..the

existence of quiescent subpopulations). The glioblastomas. U87.HGL9, and U251-MG, seem relatively slow growing and exhibit

bimodal growth curves. Besides its value as an alternative method forestimating 7"pl>,values, fuller development and exploration of this

cytokinesis-block method should provide an enhanced understanding

of: (a) how individual primary tumor populations grow in culture: and(b) the extent to which proliferation heterogeneities in vitro indicateanalogous diversities in vivo. The documentation of intcrtumor cellgrowth differentials may be a useful augment to the concept of 7^,, inthe prediction of tumor growth and therapeutic response.

ACKNOWLEDGMENTS

We gratefully acknowledge the people at MGH who provided us with thexenografts. in particular. Dr. Alphonse Taghian. Dr. Ayman Allam. andMichael Duffy. We thank Dr. Luis Pere/ for sharing some unpublished data.We also thank Carlos Romero of the Joint Center for Radiation Therapy, whoassisted in the data collection.

APPENDIX

To determine 7"p(llvalues, the numbers of nuclei/1000 cytokinesis-blocked

cells were plotted as a function of time. Best exponential projections ofpopulation growth (nuclear proliferation) were then made on the basis of thesedata. The exponential tits include certain classic kinetic assumptions of tumorgrowth used in standard T",,,, derivations [e.g.. Steel (II)]. There is the

observation in vivi> that there is a fraction of the population that is non-

proliferating, but which nonetheless increases in size to maintain a constantproportion with the proliferating population (fixed-growth fraction). This non-

growing subpopulation increases in si/e by receiving an influx of cells that arenonproliferating progeny of dividing cells. The effect of the presence of sucha nonproliferating fraction is to extend the 7"^,, from T, to JJf. where 7", is the

1662

on June 3, 2020. © 1996 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

T.,, USING CYTOKINESIS BLOCK METHOD

cell cycle time for the proliferating cells, and/is ln(l + G/r)/ln(2). where GF

is the growth fraction. We do not expect equivalent behavior in vitro but notethat any such effect will be absorbed into the solution for a time coefficient,which appears in the in vitro kinetics statement to follow.

Beyond in vivo tumor cell kinetics, an additional consideration is evidencedby observations of the acclimation of tumor cells to culture conditions. It seemsthat a small subpopulation of the original plated population is incapable offurther division or is so delayed in the resumption of division that it may beconsidered dormant for the sake of describing the inilial kinetics. (Correctionfor this dormancy is analogous in principle to the plating efficiency adjustmentroutine in colony assay studies.) This subpopulation is of fixed size, distinctfrom the variable-size, nongrowing fraction seen in vivo.

An extension of standard tumor kinetic theory more suitable for capturingboth in vitro kinetic considerations just discussed is to now consider the rpolto be the time at which the population has "grown" to an average of 2000

nuclei/1000 cytokinesis-blocked cells, where N is the population size (average

nuclear count/1000 cells) at time t, A is the nongrowing portion of thepopulation, B is the coefficient of exponential growth, and C is a coefficientintroducing the proper time scaling and/or net growth retardation:

N = A + S*exp(O)

where, specifically

2000 = A + ß*exp(C7;„,)

For fitting the data. A, B. and C are solved for by using least squares methods.The fits are actually made to logl/V - A) = log(B) + Ct, inserting trial values

for A and in each instance solving for B and C by linear regression fits to theresulting linear equation in t. The values of A, B, and C that gave the leastsquares fit to this equation overall were ultimately determined. Portions of thecurves after "rollover" (negative curvature) are attributed to mitotic growth

inhibition in the population, a dependence not applicable in the usual meaningof Tp,,,. and, therefore, were not considered in the fits. The origins were alsoexcluded from the fits, except for the case of STS26T cells, for which theorigin was included to give the minimum three points necessary to accomplishthe fit.

References

1. Tsang, R. W.. Fyles. A. W.. Kirkbride. P.. Levin, W., Manchul. L. A.. Rawlings.G. A., Banerjee. D.. Pintilie. M.. and Wilson, G. D. Proliferation measurements with(low cytometry T^,, in cancer of the uterine cervix: Preliminary results. Int. J. Radiât.Oncol. Biol. Phys., 32: 1319-1329, 1995.

2. Begg. A. C. The clinical status of 7"^,, as a predictor? Or why no tempest in the 7"^,,!

Int. J. Radiât.Oncol. Biol. Phys.. 32: 1539-1541, 1995.

3. Speke, A. K.. and Hill, R. P. Repopulation kinetics during fractionated irradiation andthe relationship to the potential doubling time, rp,„.Int. J. Radiât.Oncol. Biol. Phys..31: 847-856. 1995.

4. Begg. A. C.. Holland. I.. Van Glabekke. M.. Bartelink, H., and Horiot. J. C. Predictivevalue of potential doubling lime for radiotherapy of head and neck tumor patients:results from the EORTC cooperative trial 22851. Semin. Radial. Oncol., 2: 22-25,1992.

5. Begg, A. C.. Holland. !.. Moonen. L.. Bartelink. H.. Schraub. S.. Bontemps, P.. LeFur. R.. Van Den Bogaert. W.. Caspers. R.. Van Glabbeke. M.. and Horiot. J. C. Thepredictive value of cell kinetic measurements in a European trial of acceleratedfractionation in advanced head and neck tumors: an interim report. Int. J. Radiât.Oncol. Biol. Phys.. 19: 1449-1453, 1990.

6. Withers, H. R., Taylor. J. M. G., and Maciejewski. B. The hazard of acceleratedtumour clonogen repopulation during radiotherapy. Acta Oncol.. 27: 131-146. 1988.

7. Fowler. J. F. Potential for increasing the differential response between tumors andnormal tissues: can proliferation rate be used? Int. J. Radial. Oncol. Biol. Phys., 12:641-645. 1986.

8. Maciejewski. B.. Preuss-Bayer. G.. and Trott, K. R. The influence of the number offractions and of overall treatment time on loca! control and late complication rate insquamous cell carcinoma of the larynx. Int. J. Radiât.Oncol. Biol. Phys., u: 321-328,1983.

9. Trott. K. R. Cell repopulation and overall treatment time. Int. J. Radiât.Oncol. Biol.Phys., 19: 1071, 1990.

10. Trott. K. R., and Kummermehr. J. What is known about tumor proliferation rates tochoose between accelerated fractionation or hyperfractionalion? Radiother. Oncol., 3:1-9. 1985.

11. Steel. G. G. Growth Kinetics of Tumours. Oxford, United Kingdom: Clarendon Press,1977.

12. Peters. L. J., Ang, K. K., and Thames, H. D. Accelerated fractionation in the radiationtreatment of head and neck cancer: a critical comparison of different strategies. ActaOncol., 27: 185-194. 1988.

13. Lochrin. C. A., Wilson. G. D., McNally, N. J., Dische, S.. and Saunders. M. I. Tumorcell kinetics, local tumor control, and accelerated radiotherapy: a preliminar)' report.Int. J. Radial. Oncol. Biol. Phys.. 24: 87-91. 1992.

14. Begg. A. C. Prediction of repopulation rates and radiosensitivity in human tumours.Int. J. Radial. Biol., 65: 103-108, 1994.

15. Begg, A. C.. McNally. N. J., Shrieve, D. C., and Karcher, H. A method to measureduration of DNA synthesis and the potenlial doubling lime from a single sample.Cytometry, 6: 620-626, 1985.

16. White, R. A., Terry, N. H. A., Meistrich, M. L.. and Calkins. D. P. Improved methodfor computing the potenlial doubling lime from flow cytometric data. Cytometry. II:314. 1990.

17. Ritter, M. A.. Fowler, J. F.. Kim, Y.. Lindstrom, M. J.. and Kinsella. T. J. Singlebiopsy, tumor kinetic analyses: a comparison of methods and an extension to shortersampling intervals. Int. J. Radiât.Oncol. Biol. Phys.. 23: 811-820, 1992.

18. Cooke, L. D., Cooke, T. G., Forster, G.. Jones. A. S.. and Stell, P. M. Prospectiveevaluation of cell kinetics in head and neck squamous carcinoma: the relationship totumour factors and survival. Br. J. Cancer, 69: 717-720. 1994.

19. Wilson, M. S.. West. C. M. L.. Wilson. G. D.. Roberts. S. A.. James. R. D., andSchofield. P. F. An assessment of the reliability and reproducibility of measurementof potential doubling times (7^,,) in human coloréela!cancers. Br. J. Cancer. 67:754-759, 1993.

20. Rew, D. A., Campbell. I. D.. Taylor, I., and Wilson, G. D. Proliferation indices ofinvasive breast carcinomas after in vim 5-bromo-2'-deoxyuridine labelling: a flow

cytometric study of 75 tumours. Br. J. Surg., 79: 335-339, 1992.

21. Bourhis, J., Wilson, G., Wibault, P., Bosq, J., Chavaudra, N., Janot, F.. Luboinski, B.,Eschwege, F.. and Malaise, E. P. In vivo measurement of the potential doubling timeby flow cytometry in oropharyngeal cancer treated by conventional radiotherapy. Int.J. Radiât.Oncol. Biol. Phys., 26: 793-799, 1993.

22. Shibuya, M., Ito, S.. Davis, R. L.. Wilson. C. B.. and Hoshino, T. A new method foranalyzing the cell kinetics of human brain tumors by double labeling with bromode-oxyuridine in silu and with ¡ododeoxyuridine in vilra. Cancer (Phila.). 71: 3109-3113. 1993.

23. Shibamoto. Y.. and Streffer, C. Estimation of the dividing fraction and potentialdoubling lime of tumors using cytochalasin B. Cancer Res., 51: 5134-5138. 1991.

24. Perez, L. A.. Dombkowski. D.. Efird, J.. Preffer, F.. and Suit, H. D. Cell proliferationkinetics in human tumor xenografts measured with iododeoxyuridine labeling andflow cytometry: a study of heterogeneity and a comparison between different methodsof calculation and other proliferation measurements. Cancer Res.. 5.5: 392-398. 1995.

25. Baumann. M.. DuBois. W.. Pu. A., Freeman, J.. and Suit, H. D. Response ofxenografts of human malignant gliomas and squamous cell carcinomas to fractionatedirradiation. Int. J. Radiât.Oncol. Biol. Phys., 23: 803-809, 1992.

26. Taghian. A., DuBois. W.. Budach. W.. Baumann, M.. Freeman. J., and Suit, H. D., Invivo radiation sensitivity of glioblastoma multiforme. Int. J. Radiât. Oncol. Biol.Phys.. 32: 99-104, 1995.

27. Hurwitz. S. J.. and Hlatky, L. Assessment of radiation response on a cell by cell basisusing in situ densitometric imaging of micronuclci. Radiât.Res.. /34: 112-116, 1993.

28. Shibamoto. Y., Streffer. C.. Fuhrmann. C., and Budach. V. Tumor radiosensitivityprediction by the cytokinesis-block micronucleus assay. Radial. Res., 128: 293-300,

1991.29. Wakata. A., and Sasaki. M. S. Measurement of micronuclei by cytokinesis-block

method in cultured Chinese hamster cells: comparison with types and rates ofchromosome aberrations. Mutât.Res., 190: 51-57, 1987.

30. Abend. M., Rhein. A.. Gilbertz, K. P.. Blakely. W. F, and Van Beuningen, D.Correlation of micronucleus and apoptosis assays with reproductive cell death. Int. J.Radial. Biol.. 67: 315-326, 1995.

31. Shibamoto. Y., Shibata, T., Miyatake. S.. Oda. Y.. Manahe. T., Ohshio. G.. Yagi. K.,Streffer. C., Takahashi. M.. and Abe. M. Assessment of the proliferative activity andradiosensitivity of human tumours using the cytokinesis-block micronucleus assay.Br. J. Cancer., 70: 67-71. 1994.

32. Hlatky, L.. Hahnfcldl, P.. and Ring. C. S.. Synchronized resumption of DNA synthesis in reoxygenated cells: spatial profiles in the sandwich tumor model (Abstract).Program of the 38th Annual Meeting of the Radiation Research Society, 1990.

1663

on June 3, 2020. © 1996 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1996;56:1660-1663. Cancer Res   Lynn Hlatky, Mariusz Olesiak and Philip Hahnfeldt  Xenografts Using the Cytokinesis-Block MethodMeasurement of Potential Doubling Time for Human Tumor

  Updated version

  http://cancerres.aacrjournals.org/content/56/7/1660

Access the most recent version of this article at:

   

   

   

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

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

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

  Permissions

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

.http://cancerres.aacrjournals.org/content/56/7/1660To request permission to re-use all or part of this article, use this link

on June 3, 2020. © 1996 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from