continued malignant cell proliferation in head and neck...

Vol. 2, 1453-1460, Septemnber 1996 Clinical Cancer Research 1453

Continued Malignant Cell Proliferation in Head and Neck Tumors

during Cytotoxic Therapy’

Harvey D. Preisler,2 Valery M. Kotelnikov,

Suzanne LaFollette, Samuel Taylor IV,

Suneel Mundle, Nancy Wood, John S. Coon IV,

James Hutchinson, William Panje,

David D. Caldarelli, and Katherine Griem

Rush Cancer Institute [H. D. P., V. M. K., S. L., S. T., S. M.] and

Departments of Pathology [N. W., J. S. C.], Otolaryngology [J. S. C.,

J. H., W. P.. D. D. C.], and Radiation Oncology [K. G.], Rush-

Presbyterian-St. Luke’s Medical Center, Chicago, Illinois 60612-3833

ABSTRACT

The effect of cytotoxic therapy on the proliferation of

squamous cell carcinoma of the head and neck in vivo in

patients was evaluated before and 15-35 days after the start

of therapy. To accomplish this, iododeoxyuridine was ad-

ministered at t = 0, and bromodeoxyuridine was adminis-

tered 15-35 days later during treatment with a tumor biopsy

obtained for study immediately after each pyrimidine infu-

sion. Monoclonal antibodies specific for the halogenated

pyrimidines were used to identify cells that were in the

S-phase at the time of the infusions. Eleven patients were

studied prior to treatment. Of those, the intratreatment

biopsy of eight patients contained tumor tissue. In the other

three patients, tumor tissue was not present in the second

biopsy. Continued precursor incorporation into DNA-syn-

thesizing cells during treatment was detected in six of eight

tumor specimens. In two tumor specimens, an increase inthe percentage of S-phase cells was noted, in two specimens

tumor cells synthesizing DNA were not detected, and in four

specimens the percentage of S-phase tumor cells was lower

than that in the pretherapy specimen. Patients in whom

there were no S-phase cells detected during treatment or in

whom no tumor was detected in the second biopsy had afavorable treatment outcome in comparison to those pa-

tients in whom continued tumor proliferation during treat-

ment was detected. The number of cells in S-phase prior to

the initiation of treatment was not predictive of whether or

not proliferation would continue during cytotoxic therapy.

Evidence for reentry of kinetically quiescent cells into the

cycle during treatment was noted. Additionally, cytotoxic

therapy altered the proliferation pattern of normal-appear-

ing mucosa as well. The results of this study demonstrate

that tumor cell proliferation does continue in some squa-

mous cell carcinoma of the head and neck during intensive

cytotoxic therapy.

INTRODUCTION

Malignant cell repopulation during cytotoxic therapy is

believed to be a major impediment to the successful treatment of

patients with rapidly proliferating malignant disease ( 1-5). This

belief is based on a large body of indirect evidence, especially

in the area of SCCHN ( 1 , 6), which suggests that malignant cells

continue to proliferate during radiotherapy, thereby offsetting, at

least in part, the reduction in cell numbers produced by treat-

ment. There is also indirect evidence that in some eases tumor

proliferation rates increase during cytotoxic treatment, further

compromising treatment outcome (7). Based primarily on the

studies of experimental tumors (8-I 1) and on scant direct

human data, new radiation therapy schedules have nevertheless

been devised to overcome repopulation resistance ( I 2). Addi-

tionally, indirect evidence has recently been presented that re-

population or regrowth resistance is also a significant impedi-

ment to successful treatment outcome in patients receiving

chemotherapy (13).

Attempts to demonstrate the phenomenon of continued

proliferation in patients during treatment using in vitro labeling

with tritiated thymidine have been hampered by methodological

problems. For example, in two studies investigators removed

tumor tissue during treatment and labeled tumor fragments in

vitro with 13H]thymidine to assess the effects of treatment on the

proliferative rate (14, 15). Under these conditions, only a thin

rim on the periphery of the tumor fragment incorporated the

DNA label, thus limiting the information which can be obtained

regarding the proliferative behavior of the tumor. Additionally,

another group of investigators labeled tumor cell suspensions

obtained by fine-needle aspiration, also relying upon in vitro

uptake of [3Hlthymidine (16). To study accelerated (or continu-

ing) proliferation of tumor cells in patients, one must have the

ability to measure proliferation parameters in vivo both before

and during (or after) therapy. By administering IdUrd3 to pa-

tients before treatment and BrdUrd to patients after the treat-

ment, in combination with immunohistochemical analyses of

tumor biopsies obtained after each halogenated pyrimidine in-

fusion, these measurements can be made ( I 7).

In this article, we describe studies of the effects of cyto-

toxic therapy on SCCHN in vivo in patients. These studies have

demonstrated that: (a) proliferation does continue unabated in

some tumors during intensive cytotoxic therapy: (b) high pre-

Received 10/24/95; revised 6/5/96; accepted 6/1 1/96.

I Supported by funds from the Rush Cancer Institute.

2 To whom requests for reprints should be addressed, at Rush Cancer

Institute, 1725 West Harrison Street, Suite 809, Chicago, IL 60612-

3833. Phone: (312) 563-2190; Fax: (312) 455-9635.

3 The abbreviations used are: IdUrd, iododeoxyuridine; BrdUrd, bro-

modeoxyuridine: 5-FU, 5-fluorouracil; CR, complete remission: PD,

persistent disease; GMA, glycol methacrylate; ISEL, in situ end label-

ing; SCC, squamous cell carcinoma; SCCHN, squamous cell carcinoma

of the head and neck.

Research. on August 25, 2019. © 1996 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

1454 Continued Tumor Cell Proliferation during Therapy

Table 1 Patient and tumor characteristics

Patient Age (yr) Sex Tumor site Differentiation

New or

recurrent TNM classification

1 36 F’ Tonsil Well Recurrent 4, 0, 0

2 58 M Floor of mouth Moderate New 4, 2, 0

3 66 M Maxillary sinus Moderate New 4, 0, 0

4 64 M Larynx Moderate New 4, 0, 0

S 53 M Tongue Moderate New 3, 0, 26 68 M Parotid Poor New 4, 0, 07 56 F Larynx Well Recurrent 4, 0, 0

8 50 M Soft palate Moderate New 4, 0, 0

9 78 F Tongue Moderate New 1, 1, 0

10 41 M Tonsil Moderate New I, 3, 0

11 66 M Tonsil Moderate New 2, 0, 0

a F, female; M, male.

therapy proliferative rates are not necessarily associated with

continued proliferation during treatment; (c) some malignant

cells which apparently leave the cell cycle shortly after the

initiation of treatment resume proliferation 3-4 weeks later; and

(d) the effects of cytotoxic therapy on the proliferative behavior

of normal oral mucosa adjacent to a tumor can be markedly

different from the effects on the tumor.

MATERIALS AND METHODS

Patient Characteristics and in Vivo Studies. Eleven

squamous cell carcinomas of the head and neck were studied at

t = 0 and 15-36 days later. The characteristics of the patients

and their tumors are provided in Table 1. The patients received

an infusion of IdUrd (100 mg/m2 over 1 h). At the end of the

infusion, a biopsy of the tumor was obtained with care being

taken to avoid necrotic areas. The biopsy specimen was imme-

diately fixed as described below (t = 0 biopsy). The patients

were then treated as described in Table 2. Six patients (patients

4, 5, 7, 8, 9, and 10) were treated with concurrent chemotherapy/

radiotherapy in a week-on/week-off schedule. Chemotherapy

consisted of: cisplatin, 60 mg/m2 i.v. bolus on day 1 and 5-FU,

800 mg/m2 continuous infusion on days 1-5 of weeks 1 , 3, 5, 7,

9, 1 1, and 13. Radiation was given at 2 Gy/day on days 1-5 of

weeks I , 3, 5, 7, 9, 1 1 , and 13. Each patient completed a

13-week course with a total irradiation dose of 70 Gy.

Patients 2 and 3 received alternating chemotherapy/radia-

tion (18). Chemotherapy consisted of: cisplatin, 20 mg/rn2 iv.

bolus on days 1-4, and 5-Hi, 200 mg/m2 iv. bolus on days 1-5

of weeks 1, 4, 7, and 10. Radiation was given at 2 Gy/day on

days 1-5 of weeks 2, 3, 5, 6, and at 1.5 Gy twice a day on days

1-S of weeks 8 and 9. Total treatment duration was 10 weeks,

with a total radiation dose of 70 Gy. Patient 1 received one cycle

of chemotherapy only (laxol, 200 mg/m2 iv. infusion on day 1,

and cisplatin, 75 mg/m2 iv. bolus on day 2). Patient 1 1 was

treated with radiotherapy only (as above), and patient 6 received

neutron beam therapy. Cytotoxic therapy began on days 3-5

after the first biopsy.

According to theoretical calculations, tumor cell repopula-

tion may begin 3-6 weeks after the beginning of cytotoxic

therapy (1). Based on these data, we chose to obtain the second

biopsy (intratreatment biopsy) 15-36 days after the beginning of

the cytotoxic therapy. Before the second biopsy, the patient

received iv. infusion of BrdUrd (100 mg/rn2 over 1 h). With

respect to the second biopsy, care was taken to biopsy nonne-

erotic tissue at the site of the original tumor biopsy. In eight

patients, tumor tissue was present in the second biopsy. Tumor

tissue was not present in the second biopsy of the other three

patients.

The results of treatment were assessed 6 weeks after the

completion of the entire treatment course. At this time, the

patients underwent endoscopic examination and computed to-

mographic scan. Patients without signs of tumor were catego-

rized as being in CR. PD was characterized by gross tumor seen

endoscopically or on computed tomographic scan 6 weeks after

therapy.

Percentage of S-Phase Measurements. Detailed de-

scriptions of the methods used in these studies can be found in

previous publications (17, 19). In brief, the biopsy specimens

were fixed and embedded in GMA or paraffin. Two-jim-thin

sections were prepared from the plastic-embedded sections, and

7-jim sections were prepared from the tissues embedded in

paraffin. The sections were rehydrated and treated with 3%

H2O2, Pronase, and 4 N HC1. For specimens obtained after the

infusion of IdUrd, monoclonal antibody 3D9 (a gift from

Dr. G. L. Mayers, Roswell Park Memorial Institute, Buffalo,

NY) was applied (1 :200) for 30 mm followed by biotinylated

antimouse IgG (1 :200) for 30 mm, and then the ABC reagent is

added (Vectastain Elite ABC kit). If the patient was infused with

BrdUrd, tumor sections were stained according to the procedure

described above using the rat anti-BrdUrd antibody (Sera Labs,

Crawley Down) diluted 1 : 100 followed by staining with a

Vectastain kit as described above. At the final step, all speci-

mens were stained with 0.025% 3,3’-diaminobenzidine tetrahy-

drochlonde in 0.003% H2O2 and counterstained with Harris’

hematoxylin solution. Sections treated without primary antibody

were used as controls. In repeated staining sessions, it has been

confirmed that the anti-BrdUrd monoclonal antibody is specific

for BrdUrd and does not cross-react with IdUrd (19). In contrast,

monoclonal antibody 3D9 recognizes both IdUrd and BrdUrd.

In a previous study, 19 patients were infused with both IdUrd

and BrdUrd, with a 30-mm interval between infusions. Sections

from these specimens were stained with either 3D9 or rat

anti-BrdUrd monoclonal antibodies. The resulting labeling in-

dex values were very close to each other, indicating that both



Clinical Cancer Research 1455

Table 2 Response to therapy and follow-up

Treatment

Time of the second biopsy

Response 4 weeks

after completion

of treatment Follow-upPatient Chemotherapy Radiotherapy

1 Cisplatin/Taxol No t = 15; after 1 cycle of CT” PD Persistent disease, progression, died at

13 months

2 CisplatinlS-FU Yes t = 25: after 2 cycles of CT and 2

cycles of RIPD Progression, died at 4 months

3 CisplatinlS-FU Yes t = 22; after 1 cycle of CT and 2

cycles of RIPD Relapsed 4 months and died at 7

months4 CisplatinlS-FU Yes t = 36; after 3 cycles of concurrent

CT/RI

CR Relapsed 10 months, alive 18+ months,

with PD

S CisplatinlS-FU Yes t = 3 1 ; after 3 cycles of concurrentCT/RI

PD Died 4 months after study

6 No Neutron beamtherapy

t 17; during the RI PD Relapsed and died S months after study

7 CisplatinlS-FU Yes t = 22; after 2 cycles of concurrentCT/RI

CR Died 19 months from treatment-relatedcomplication (hemorrhage)

8 Cisplatin/5-FU Yes t = 21; after 2 cycles of concurrent

CT/RICR 9+ months, CR

9 CisplatinlS-FU Yes t = 31; after 3 cycles of concurrent

CT/RI

CR Died 14 month later from unrelated

disease

10 CisplatinlS-FU Yes t 20; after 2 cycles of concurrent

CT/RICR 27+ months, CR

1 1 No Yes t = 22; during the 3d cycle of RI CR 10+ months, CR

a CT, chemotherapy; RI, radiation therapy.

halogenated pyrimidines are incorporated into DNA in the same

manner (19).

The percentage of S-phase cells was determined in four to

six randomly selected fields of the section. From 2000 to 5000

cells were counted in each section. In the second biopsies, where

BrdUrd labeled cells were limited to some areas of the speci-

men, the percentage of S-phase values were calculated in these

areas only. Special care was taken to distinguish between tumor

tissue and dysplastic normal or interstitial tissue. Iwo experi-

enced histopathologists read the slides independently, and con-

sensus was reached on every specimen.

Some tumor cells in the second biopsy were the progeny of

cells that were in the S-phase at the time of the IdUrd infusion

just before the first biopsy. To be able to visualize these cells,

we applied the double-labeling technique. GMA sections were

first stained with a rat anti-BrdUrd antibody (Sera Labs), fol-

lowed by the ABC reagent (Vectastain Elite ABC kit) and DBA.

At this step, all DNA sites labeled with BrdUrd were blocked by

anti-BrdUrd antibody. At the second step, 3D9 antibody (1:50 in

PBS containing 0.5% Iween 20) was applied. This antibody

bound to IdUrd only. This was followed by the sequential

treatment of the sections with rabbit anti-mouse immunoglobu-

un and alkaline phosphatase-antialkaline phosphatase-conju-

gated mouse monoclonal antibody (DAKO Corp). The sections

were then treated with fast blue substrate for final staining. The

above procedure results in IdUrd-containing cells having blue

nuclei, BrdUrd containing nuclei staining brown, and nuclei

containing both BrdUrd and IdUrd staining both brown and

blue. Cells which had not incorporated IdUrd or BrdUrd were

unstained. The double-labeling technique can be performed on

GMA-embedded tissue only. Since only four of eight plastic-

embedded second biopsies contained tumor tissue, these double-

label IdUrd/BrdUrd studies were available in only one half of

the specimens. In the other four tumors, single labeling to detect

BrdUrd was performed in paraffin-embedded specimens.

Simultaneous Detection of S-Phase Cells and Apoptosis.

ISEL was carried out on GMA-embedded tumor (2 mm) see-

tions as described in our previous report (20). Briefly, following

pretreatment with sodium chloride-sodium citrate solution at

80#{176}Cand with 0.5% pepsin (Sigma, St. Louis, MO) in HC1 (pH

2) at 37#{176}C,the sections were incubated with a mixture of dAIP,

dCIP, and dGIP (0.01 M; Promega, Madison, WI), biotinylated

dUIP (0.001 M; Sigma), and Escherichia coli DNA Poll (20

units/ml; Promega) at 18#{176}Cfor 2 h. Incorporation of biotiny-

lated dIJIP was finally visualized using avidin-biotin-peroxi-

dase conjugate (Vectastain Elite ABC kit; Vector Laboratories,

Burlingame, CA) and diaminobenzidine. Thus, cells labeled

positively for ISEL showed brown staining in their nuclei under

light microscopy. Treatment of the tumor sections with mono-

clonal anti-IdUrdlBrdUrd antibody 3D9 to label cells in the

S-phase was then carried out. Avidin-biotin-peroxidase conju-

gate (ABC kit) and diaminobenzidine were used to visualize

ISEL labeling, and alkaline phosphatase-anti alkaline phospha-

tase (DAKO, Carpinteria, CA) and nitroblue (Genius I kit;

Promega) were used to visualize the S-phase cells. Thus, cells

undergoing apoptosis (ISEL positive) exhibit brown staining in

their nuclei, whereas the nuclei of the S-phase cells appear to be

dark purple.

RESULTS

Effects of Treatment on Proliferative Rate. Eleven pa-

tients with advanced 5CC of the head and neck were studied

before the initiation of treatment (t = 0 study) and on days

15-36 of treatment (intratreatment study). Of the latter, only

eight had tumor tissue in the second biopsy. Six of these patients



, #7

,,- �..

. �5!.

__J”� p.-. -

- � .

. :�‘ � � �

�---�. �

. --, ,. .� I.- .- � . , ‘�( .

V.�--,- �... �,‘. t�/ � _ � �.

#{128}�, .,i .� J

�: � � �,

‘, � .� ::‘::_� ,� :;�fr’ � _I� � ,� #{149}� � �r2

� � � ,,� � I�,. � a. #{231}- � � ,

� - - � � # - Si .� � .

# .� .� -� --�

) � -,

; .

.�J.S1�vrFig. I S-phase cells in biopsies obtained before and after treatment (paraffin, A and B; 3D9 (anti-IdUrd)/immunoperoxidase, C and D; anti-BrdUrd/

immunoperoxidase staining, counterstained with hematoxylin). Patient 3 had received an infusion of ldUrd before the first biopsy. Note numerous

(33%) IdUrd-labeled cells in the first biopsy (A). The second biopsy. preceded by BrdUrd infusion (B), was obtained 22 days later, after two cyclesof radiation (20 Gy) and chemotherapy (cisplatin + 5-FU). The tumor tissue contains a large number of S-phase BrdUrd-labeled cells (24%). Patient

4 received three cycles of radiation (30 Gy) and chemotherapy (cisplatin + 5-FU). Before the start of treatment, the tumor contained 19% S-phase

cells labeled with IdUrd (C). Thirty-six days later most of the tissue at the primary tumor site was fibrotic. Small islands of tumor cells (D, arrow)

survived with many cells synthesizing DNA, as demonstrated by in i’is’o BrdUrd labeling. A-D, X200.


.4_’ -.-�- . � ‘i__ -, I

� � �

.:�. .‘.. .

. - I, �..� . , L:;iI;. � .

., -

%x� �

C � � ,�I, � � � � 1;

- �t’._ -�-‘ t

4�T :�‘�

had newly diagnosed disease, and two had recurrent disease.

Table 2 provides details regarding the treatment that each pa-

tient received, the relationship of the second proliferation study

to the treatment being administered, and the response of the

tumor to treatment. The intratreatment biopsy was obtained

either in the last 1 or 2 days of cytotoxic therapy or within a few

days of the completion of a course of therapy.

As can be seen in Fig. 1 , A and C, prior to treatment the

S-phase cells were distributed throughout the biopsy specimen.

Treatment significantly disturbed the morphology of the tumor

cells. In every treated tumor, the number of tumor cells was

decreased. Many of the remaining cells were large with irregular

shapes containing giant, often multilobulated nuclei. In seven of

the eight tumors studied, the S-phase cells in the intratreatment

biopsy were distributed throughout the tumor tissue (Fig. lB). In

one intratreatment specimen (Fig. lD, patient 4), the proliferat-

ing tumor cells appeared to be organized into islands surrounded

by fibrotic tissue. In every intratreatment biopsy, even in those

that did not contain viable tumor tissue, there were many Br-

dUrd-labeled lymphocytes.

. - - - � �

Bc’.’ ‘�-�“ ‘-�: � ,,, � .-

� ,,� � . : �‘�: � �., .5, �,.‘ p :i�-.�. � � � . .� . ,dpi. b�,,,,

� .‘:�: �

V.’ � r,”.”�.. . .1.-.-� � ...,.�#{149} ... ._

.�.. � .-�.� a.

t.� .--, ... * -..,.� . -.

�. � � � � _; , . �‘4i #{149}‘ #{149}_,. . -....� ..‘ � _

. . �l �. . �1

�1 .1?-.’..,.,,- � � - .

,. , ‘�-#{149}:� �‘e.w1 ..

Table 3 provides data regarding the effects of cytotoxic

therapy on tumor proliferation. In four specimens, the percent-

age of S-phase cells fell by 27, 37, 68, and 71% of the initial

value (patients 3, 4, 5, and 6, respectively). In two specimens

(patients 7 and 8), it fell to 0, and in two specimens, the

intratreatment percentage of S-phase cells was greater than the

pretherapy value (increases of 17% in patient 1 and 19% in

patient 2). It is important to note that, as illustrated in Table 3,

a high pretherapy percentage of S-phase cells was not an mdi-cation that a tumor was likely to continue to proliferate during

cytotoxic therapy. In fact, a high pretherapy percentage of

S-phase cells could be associated with a greater or lesser per-

centage of S-phase cells relative to the pretreatment value.

Five of six patients whose tumors contained proliferating

cells in the second biopsy have died within 13 months after

being studied, and one has relapsed. In contrast, two of three

patients whose intratreatment biopsy did not contain tumor

tissue are alive in complete remission (10+ and 27+ months,

respectively), whereas the third patient died 14 months later

from an unrelated cause. Similarly, one of two patients whose



Tumor

proliferation

(LI.” #{182}/)

A

41

25.2

23.7

12

8.5

6.3

0

I 34.5

2 20.6

3 32.8

4 19.1

5 31.2

6 22

7 32.2

8 36.4

9 26.4

10 14.1

11 38

na.

na.

0.9

2.2

I 0.1

na.

10.9

5.9

25.8

I 3.8

B

BrdUdr

Patient only

ldUdr + IdUdr only IdUdr only

BrdUdr (fragmented) (nonfragmented)

0.8

0

0.5

I)


Table 3 Effects of treatment on percentage of S-phase cells

Proliferation of normal mucosa ( LI. C�

Basal layer Suprabasal layers

t=0 t=x” t=0 t=x t=0 t=x

Patient_(IdUdr)(BrdUdr) (IdUdr) (BrdUdr) (IdUdr) (BrdUdr)

na. 8.7 na. 13.8

7.3 32.5 8.2 14.8

na. na.

28.2 na.

11.1 24.5

28 10.7

8.1 16.2

0 na. na. na. na.

nt. na. na. na. na.

nt. 0.3 na. I 8.4 na.

nt. 0.5 na. 39.8 na.

(, LI, labeling index; na., not available: nt.. no tumor in the second

biopsy.

I, � = x, intratreatment biopsy performed 15-36 days after the start

of cytotoxic therapy.

Table 4 Labeling index values for tumor cells denionstrating

different IdUdr/BrdUdr labeling patterns” in the intratreatment biopsy

specimen

2 5.9 19.1 14.1

3 12.8 11.2 18.4

5 4.4 4.6 8.5

7 0 0 18

“ Cells labeled with BrdUdr were in the S-phase during the second

(intratreatment) biopsy. The parent cells of those labeled with both

IdUdr and BrdUdr were in the S-phase at t 0. Cell nuclei demon-

strating fragmented IdUdr label are the progeny of cells labeled with

IdUdr at t = 0, which divided subsequent to labeling. Cells with

nonfragmented ldUdr label were in the S-phase at t = 0 and probably

failed to divide immediately thereafter.

tumor in the intratreatment biopsy did not contain proliferating

cells is alive and in CR, whereas the second patient died tumor

free from an intercurrent cause (Table 2).

Evidence That the Same Cells in the Tumor Proliferate

before and after Several Weeks of Treatment (Double La-

beling of the Second Biopsies). In four of the second biopsies

(four others were unavailable for this study for reasons de-

seribed in “Materials and Methods”). double labeling demon-

strated that many tumor cells contained IdUrd despite the IS-

36-day hiatus between administration of the IdUrd and the

second biopsy. In some of these cells (Table 4, patients 2 and 5),

the intensity of staining for IdUrd was comparable to the inten-

sity of staining of labeled cells immediately after the IdUrd

infusion at t = 0 (Fig. 2A. In previous studies, we have found

that the pyrimidine label becomes fragmented in cells that

proliferate after being labeled with halogenated pyrimidines

(19). Hence, the persistence of undiluted IdUrd label in the

nuclei of tumor cells approximately 1 nionth after IdUrd admin-

istration strongly suggests that the IdUrd-containing cells in

these tumors had undergone few. if any, divisions between the

first and second biopsies.

Fig. 2 Intratreatment biopsy of patient 5 (GMA. anti-BrdUrdJ3D9 dou-ble labeling. A. cells labeled with BrdUrd only (hrow,i) and with IdUrd

only (blue). B. cells labeled with both IdUrd and BrdUrd. These cells (or

their parental cells) were synthesizing DNA before treatment when

IdUrd was infused. The persistence of IdUrd labeling 3 1 days after the

IdUrd infusion indicates that these cells underwent few or no cell

divisions during this time interval. The labeling of some of these cells

with BrdUrd indicates that these cells resumed proliferation at the time

infusion of the second label. A.C, Normarski; X l(XX).

Immediately after the intratreatment administration of Br-

dUrd, some tumor cells contained both IdUrd and BrdUrd (Fig.

2B). These double-labeled cells were in the S-phase during the

pretreatment IdUrd infusion and were proliferating during the

BrdUrd infusion as well, despite cytotoxic therapy. These dou-

ble-laheled cells represented 4.6-19.1% of the tumor cell pop-

ulation (Table 4). In some of these cells. residual IdUrd label

appeared to he undiluted and nonfragmented. The most likely

explanation for this labeling pattern is that these cells stopped

proliferating almost immediately after incorporating IdUrd and

then, at approximately the time of the second biopsies, resumed

proliferation.

In all tumors studied using the double-labeling methodol-

ogy (Table 4), 8.5-18.4% of tumor cells contained fragmented

IdUrd label. These cells had incorporated IdUrd during the first

infusion before the start of the treatment and apparently had

undergone one or more divisions since the IdUrd infusion. In

one intratreatment biopsy (patient 7), after the administration of



ldUdR -‘-#{149}-“-�

Day 0 biopsy

:,

B

.�‘j

.- -

. I�%

©

BrdUdr Q�o�O�_�i

2nd biopsy la

6�’ ‘:� .�..

w #{231}-. a

‘

. .m#, �.

C �

,‘, “I,

0

BrdUrd, all labeled cells contained fragmented IdUrd label only,

and no tumor cells were labeled with BrdUrd (Fig. 3, A and B).

As described below, although the cells of this tumor did not

incorporate any BrdUrd at the time of the second labeling,


A . - s.d.. %,�

� 4? -

�

.,�p .. ._s�

-- �

.1

. - ..

‘� �I’ : -‘

Fig. 3 Intratreatment biopsy of patient 7. A field of dying tumor cells,as indicated by the large swollen nuclei (A, GMA). The fragmentation ofthe idUrd label in some of these nuclei indicates that these cells weresynthesizing DNA during the IdUrd infusion (B, GMA, anti-BrdUrd/3D9 double labeling). C, ISEL-positive nuclei identify cells undergoing

apoptosis. Note large labeled swollen nucleus and several normal-sized

nuclei also undergoing apoptosis (GMA). A, hematoxylin; X400; B and

C, Normarski; xl000.

>� .--.

0. �‘-#{149}#{149}#{149}-�#{149}--#{149}CD0)

.CI- #{149}�.�-.-

‘4

3 3 lblc4

Fig. 4 Schematic representation of the origin of cells with different

labeling patterns in the intratreatment biopsy. Ia, cell with nonfrag-

mented IdUrd label was in the S-phase at t = 0 and probably failed to

divide thereafter; ib, cell nucleus demonstrating fragmented IdUrd label

is the progeny of a cell labeled with IdUrd at t = 0; ic, with more cell

divisions IdUrd will not be detectable at days 15-35 in the progeny of

these cells. The progeny may or may not be labeled with BrdUrd; 2,

doomed swollen cell which was labeled with IdUrd at t = 0; 3, cells

labeled with both IdUrd and BrdUrd. Their parent cells were in the

S-phase at t = 0; 4, cells labeled with BrdUrd were in the S-phase during

the second (intratreatment) biopsy. Past proliferation history is not

discernible.

adjacent normal mucosal cells incorporated BrdUrd. This result

demonstrates that the pyrimidine was available and that the

failure of tumor cells to incorporate BrdUrd is an indication that

the tumor cells were not engaged in DNA synthesis.

The different types of labeled cells in biopsies taken during

cytotoxic therapy are schematically represented in Fig. 4.

Apoptosis during Cytotoxic Therapy. Simultaneous

cell cycle and apoptosis assessments were performed on the intra-

treatment biopsies of seven tumors. In none of the tumors were any

S-phase cells undergoing apoptosis (ISEL positive), demonstrating

that proliferating tumor cells were viable and not in the process of

dying. In contrast, ISEL-positive cells were common in the fields of

dying cells (Fig. 3C). In two specimens, a few double-labeled cells

were detected (S-phase cells undergoing apoptosis) and were lo-

cated in normal epithelium or in interstitial tissue.Changes in the Percentage of S-Phase Cells in Normal-

appearing Mucosa during Treatment. Normal mucosa in the

second biopsy always contained the S-phase (BrdUrd-labeled)

cells. Pretherapy and intratherapy normal-appearing mucosa was

present in four of the eight t = 0 and intratreatment biopsies. The

percentage of the S-phase cells in the suprabasal region of normal

mucosa was greater than that of the basal layer at t = 0 (Table 3).

In contrast, in three of four specimens the percentage of S-phase

cells in the basal layer of the intratreatment specimen was signifi-

cantly greater than that of the t = 0 specimen (Fig. 5). In this small

number of specimens, there appeared to be no consistent effect of

cytotoxic therapy on the percentage of S-phase cells in the supra-

basal layers of normal mucosa.

Double labeling of the second biopsy in three patients

(patients 2, 5, and 7) shows that 25-36% of the labeled normal



A

f_ -

:-� #{149}G’ � F

. ,

:

. .‘.----,


B

S:

Fig. 5 Noninvolved normal-appearing epithelium adjacent to the tu-

mor tissue in patient 2. A, pretreatment biopsy of normal mucosa labeledwith IdUrd. Note that only suprabasal layers contain S-phase cells(GMA, 3D9). B, intratreatment biopsy obtained 25 days later after two

cycles of radiation (20 Gy) and chemotherapy (cisplatin + 5-FU). MostS-phase BrdUrd-labeled cells are now located in the basal layer (GMA,anti-BrdUrd). A and B, hematoxylin; X400.

mucosal cells contained both IdUrd and BrdUrd labels. This

observation suggests that, as for tumor cells, normal mucosal

cells which were in the S-phase before the initiation of treatment

may either continue to divide during cytotoxic therapy or leave

the S-phase to resume proliferation 3-4 weeks after the start of

treatment.

Table 3 shows that there is no apparent relationship between

the effects of treatment on the percentage of S-phase cells in the

tumor and in adjacent mucosa. The independence of the S-phase

behavior of the tumor and the adjacent normal mucosa is illustrated

by the extreme example of the intratreatment specimen 7, in which

the tumor cells failed to become labeled with BrdUrd whereas the

normal tissue contained a high percentage of S-phase cells.

DISCUSSION

It is a central tenet of radiobiology that the continued

proliferation of tumor cells during radiotherapy reduces the

effectiveness of treatment. This phenomenon has been named

repopulation resistance (3). The basis for the belief that repopu-

lation during treatment compromises treatment outcome comes

from indirect evidence: that the prolongation of a course of

radiotherapy or interruptions in a course of radiotherapy reduce

the likelihood of local disease control in SCCHN ( 1 , 2). Addi-

tionally, clinical observations suggest that the rate of malignant

cell proliferation may increase after 3-6 weeks of radiotherapy,

further compromising treatment outcome ( I).

The studies described here are the first to demonstrate that

the tumors of some patients do indeed continue to proliferate

during cytotoxic therapy. Although the number of patients stud-

ied to date is small, it is sufficient to establish the reality of

repopulation during treatment. Additionally, the studies de-

scribed here also demonstrate that some kinetically quiescent

tumor cells reenter the cell cycle 2-6 weeks after the start of

treatment, providing at least one explanation for the apparent

acceleration of tumor proliferation postulated by others (I).

Moreover, in ISEL/BrdUrd double-labeled specimens, we

did not see any tumor S-phase cells undergoing apoptosis. In

other words, proliferating cells in our study did not belong to the

population of dying cells. The absence of ISEL positivity among

the S-phase cells in the SCCHN intratreatment specimens dem-

onstrates that these cells are viable and are likely to be respon-

sible for treatment failure in the future.

Of potential clinical importance is the observation that a

high pretherapy proliferative rate is not necessarily associated

with continued rapid proliferation during treatment (patients 5,

6, 7, and 8, Table 3). This observation suggests that some

patients assigned to accelerated radiotherapy regimens on the

basis of rapid pretherapy tumor proliferative rates may not

require this type of therapy, a therapy associated with increased

toxicity. The data in Tables 2 and 3 suggest that patients in

whom the second tumor biopsy fails to demonstrate proliferat-

ing cells and patients in whom the second biopsy fails to reveal

malignant tissue have a better prognosis than patients in whom

the second biopsy demonstrates proliferating tumor cells.

The radiation tolerance of normal oral mucosa (which de-

pends on cell proliferation rates among other factors) is an impor-

tant aspect of cytotoxic therapy of SCCHN (21). The studies

described here demonstrate that the rate of proliferation of the basal

layers of normal mucosa, as predicted from animal studies in the

literature (4), increases in response to cytotoxic therapy. In a larger

series of normal mucosa samples, we have found that even though

the basal layer of mucosa in untreated patients contained only

1.6 ± 0.2% S-phase cells, after cytotoxic therapy the value was

significantly higher (19.0 ± 3.5%, P < 0.0001). In the suprabasal

layers of treated patients, the frequency of S-phase cells (15.3 ±

2.6%) was significantly lower than that of untreated samples

(31 .6 ± 3.1%, P = 0.0002). The increase in the percentage of

S-phase cells in the basal cell compartment after cytotoxic treat-

ment is part of the proliferative response of stem cells to the

damage inflicted by irradiation and chemotherapeutic agents. The

discordance of the effects of cytotoxic therapy on the proliferative

characteristics of the tumor and adjacent nontumorous mucosa in

some patients demonstrates that treatment strategies designed to

maximize the toxic effects of therapy on the tumor while minimiz-

ing effects on the adjacent normal mucosa will have to be based on

knowledge of the response to treatment of these tissues in individ-

ual patients.

Our studies provide evidence that SCC tumor of the head and




neck can continue to proliferate both during the administration of

cytotoxic therapy and between courses of therapy. Additionally, we

have presented strong evidence that SCC cells at times become

kinetically quiescent and are capable of resuming proliferation at a

later time. The latter data are virtually identical to those described

in the past for acute leukemia cells which were shown to become

kinetically quiescent in vivo in patients and to resume proliferation

in response to cytotoxic therapy (22). These observations, taken

together with similar observations in acute myelogenous leukemia

(23) and with the adverse prognosis associated with rapid pre-

therapy proliferative rates in both SCC and acute myeloid leukemia

and in many other malignancies (24-29), strongly suggest that

repopulation or regrowth resistance is a common phenomenon in

malignant disease.

Finally, the development of strategies for reducing the

regrowth of malignant cells, both during and between courses of

cytotoxic therapy, would increase the therapeutic efficacy of

currently available cytotoxic regimens (30). This approach

would compliment the more common approach of dealing with

resistant disease by increasing the intensity of cytotoxic therapy.

We have already demonstrated that biological agents can slow

the proliferation of malignant cells in vivo in patients (17).

REFERENCES

I. Withers, H. R., Taylor, J. M. G., and Maciejewski, B. The hazard of

accelerated tumor clonogen repopulation during radiotherapy. Acta On-col.,27: 131-146, 1988.

2. Irott, K., and Kummermelur, J. What is known about tumor prolif-eration rates to choose between accelerated fractionation or hyperfrac-tionation? Radiother. Oncol., 3: 1-9, 1985.

3. Tubiana, M. Repopulation in human tumors. A biological back-

ground for fractionation in radiotherapy. Acta Oncol., 27: 83-88, 1988.

4. Kummermehr, J., D#{246}rr,W., and Trott, K. R. Kinetics of accelerated

repopulation in normal and malignant squamous epithelia during frac-

tionated radiotherapy. Radiation science of molecules, mice and men.Br. J. Radiol. Suppl., 24: 193-199, 1992.

5. Bentzen, S. M., and Thames, H. D. Clinical evidence for tumor

clonogen regeneration: interpretations of the data. Radiother. Oncol.,

22: 161-166, 1991.

6. Peters, L., Ang, K., and Thames, H. Accelerated fractionation in theradiation treatment of head and neck cancer. A critical comparison of

different strategies. Acta Oncol., 27, 185-194, 1988.

7. Bourhis, J., Wilson, G., Wibault, P., Janot, F., Bosq, J., Armand,

J. P., Luboinski, B., Malaise, E. P., and Eschwege, F. Rapid tumor cell

proliferation after induction chemotherapy in oropharyngeal cancer.Laryngoscope, 104: 468-472, 1994.

8. Hermens A. F., and Barendsen, G. W. Changes of cell proliferation

characteristics in a rat rabdomyosarcoma before and after X-irradiation.Eur. J. Cancer, 5: 173-189, 1969.

9. Ramsay, J., Suit, H. D., Preffer, F. I., and Sedlacek, R. Changes inbromodeoxyuridine labeling index during radiation treatment of anexperimental tumor. Radiat. Res., 116: 453-461, 1988.

10. Rockwell, S., and Kaliman, R. F. Cyclic radiation-induced varia-tions in cellular sensitivity in a mouse mammary tumor. Radiat. Res.,

57: 132-147, 1974.

I 1. Braunschweiger, P. G., Schenken. L. L., and Schiffer, L. M. The

cytokinetic basis for the design of efficacious radiotherapy protocols.

Int. J. Radiat. Oncol. Biol. Phys., 5: 37-47, 1979.

12. Ang, K. K., and Peters, L. J. Altered fractionation in radiation

oncology. Princ. Pract. Oncol. 8: 1-15, 1994.

13. Preisler, H., and Gopal, V. Regrowth resistance in leukemia and

lymphoma: the need for a new system to classify treatment failure and

for new approaches to treatment. Leuk. Res, 18: 149-160, 1994.

14. Courdi, A., lubiana, M., Chavaudra, N., and Malaise, E. P.

Changes in labeling indices of human tumors after irradiation. Int. J.

Radiat. Oncol. Biol. Phys., 6: 1639-1644, 1980.

15. Silvestrini, R., Molinari, R., Costa, A., Volterrani, F., and Gardani,

G. Short-term variation in labeling index as a predictor of radiotherapy

response in human oral cavity carcinoma. Int. J. Radiat. Oneol. Biol.

Phys., 10: 965-970, 1984.

16. Elequin, F. I., Muggia, F. M., Ghossein, N. A., Ager, P. J., andKrisnaswamy, V. Correlation between in vitro labeling indices (LIs) and

tumor regression following radiotherapy. Int. J. Radiat. Oncol. BioI.

Phys.,4: 207-213, 1978.

17. Preisler, H. D., Raza, A., and Larson, R. Alteration of the prolifer-

ative rate of acute myelogenous leukemia cells in tivo in patients.

Blood, 80: 2600-2603, 1992.

18. Taylor, S. G., Murthy, A. K., Vannetzel, J-M., Cohn, P., Dray, M.,

Caldarelli, D. D., Shott, S., Vokes, E., Showel, J. L., Hutchinson, J. C.,

Witt, I. R., Griem, K. L., Hartsell, W. F., Kies, M. S., Mittal, B.,

Rebischung, J-L., Coupez, D. J., Desphieux, J-L., Bobin, S., and LePa-

jolec, C. Randomized comparison of neoadjuvant cisplatin and fluorou-

raeil infusion followed by radiation versus concomitant treatment in

advanced head and neck cancer. J. Clin. Oncol., 12: 383-395, 1994.

19. Kotelnikov, V. M., Coon, J., Haleem, A., Taylor, S., Hutchinson, J.,

Panje, W., Caldarelli, D., and Preisler, H. D. Cell kinetics of head and

neck cancers. Clin. Cancer Res., 1: 527-537, 1995.

20. Raza, A., Mundle, S., Ifticar, A., Gregory, S., Marcus, B., Khan, Z.,Alvi, S., Shetty, V., Dameron, S., Wright, V., Adler, S., Loew, J., Shott,S., Ali, N., and Preisler, H. D. Simultaneous assessment of cell kineticsand programmed cell death in bone marrow biopsies of myelodysplas-

tics reveals extensive apoptosis as the probable basis for ineffective

hematopoiesis. Am. J. Hematol., 48: 143-154, 1995.

21. Fowler, J. F. Apparent rates of proliferation of acutely responding

normal tissues during radiotherapy of head and neck cancer. Int. J.

Radiat. Oncol. Biol. Phys., 21: 1451-1456, 1991.

22. Gabutti, V., Pileri, A., larocco, R. P., Gavosto, F., and Cooper,

E. H. Proliferative potential of out-of-cycle leukemie cells. Nature

(Lond.), 224: 375-376, 1969.

23. Preisler, H. D, Raza, A., Larson, R., Godberg, J., licot, G., Carey,M., and Kukla, C. Some reasons for the lack of progress in the treatment

of acute myelogenous leukemia: a review of three consecutive trials of

the treatment of poor prognosis patients. Leuk. Res., 15: 773-780, 1991.

24. Raza, A., Yousuf, N., Bokhari, S. A. J., Abbas, A., Lampkin, B..

Pancoast, J., Bismayer, J., Siegrist, C., Browman, 0., Bennett, J.,

Goldberg, J., Grunwald, H., Larson, R., Tricot, G., Vogler, R., Gartside,

P., and Preisler, H. D. Cell cycle characteristics: alterable determinant ofremission duration in a study of 179 standard risk newly diagnosed

patients with acute myeloid leukemia. Int. J. Oncol., 2: 301-307, 1993.

25. Begg, A., Hofland, I., Van Glabekke, M., Bartelink, H., and Horiot,

0. Predictive value of potential doubling time for radiotherapy of head

and neck patients: results from the EORTC cooperative trial 2285 1.

Semin. Radiat. Oncol., 2: 22-25, 1992.

26. Corvo, R., Giaretti, W., Sanguineti, G., Geido, E., Orecchia, R.,

Guenzi, M., Margarino, G., Bacigalupo, A., Garaventa, G., Barbieri, M.,

and Vitale, V. In s’ivo cell kinetics in head and neck squamous cell

carcinomas predicts local control and helps guide radiotherapy regimen.

J. Clin Oncol., 13: 1843-1850, 1995.

27. Meyer, J. C., and Lee, J. Y. Relationship of S-phase fraction of

breast carcinoma in relapse to duration of remission, estrogen receptor

count, therapeutic responsiveness, and duration of survival. Cancer Res.,

40: 1890-1896, 1980.

28. Christensson, B., Tribueant, B., Linden, I. L., Ullman, B., and

Biberfeld, P. Cell proliferation and DNA content in non-Hodgkin’slymphoma. Cancer (Phila.), 58: 1295-1304, 1986.

29. Preisler, H. D., Raza, A., Bonomi, P., Taylor, S., LaFollette, S.,

Leslie, W., and Lincoln, S. Regrowth resistance is a likely significant

contribution to treatment failure in drug sensitive neoplastic diseases.

Cancer Invest., in press, 1996.

30. Preisler, H. D. Resistance to cytotoxic therapy: A speculative over-

view. Ann. Oncol., 6: 651-657, 1995.



1996;2:1453-1460. Clin Cancer Res H D Preisler, V M Kotelnikov, S LaFollette, et al. during cytotoxic therapy.Continued malignant cell proliferation in head and neck tumors

Updated version

http://clincancerres.aacrjournals.org/content/2/9/1453

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

[email protected] 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://clincancerres.aacrjournals.org/content/2/9/1453To request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



continued malignant cell proliferation in head and neck...

Documents