Download In Vitro Analysis of the Response of Multicellular

[CANCER RESEARCH 38,3595-3598.
0008-5472/78/0038-OOOOS02.00
November
1978]
In Vitro Analysis of the Response of Multicellular Tumor Spheroids
Exposed to Chemotherapeutic Agents in Vitro or in V/Vo1
John M. Yuhas,2 Anne E. Tarleton, and James G. Harman
Cancer Research and Treatment
Mexico 87131.
Center [J. M. Y., A. E. T., J. G. H.J and Department
ABSTRACT
Multicellular tumor spheroids (MTS) have been exposed
to Chemotherapeutic agents in vitro (nitrogen mustard) or
in vivo (cyclophosphamide) and analyzed in vitro in terms
of altered growth patterns. Whether the MTS were ex
posed in vitro or in vivo, the major effect of the drugs was
to induce a dose-dependent lag period before the normal
MTS growth rate resumed. Exposure of MTS in the perito
neal cavity to i.v. injection cyclophosphamide results in
patterns similar to the in vitro exposure system, except
that a host anti-MTS reaction was detected. In combina
tion, these two methods allow the study of the responses
of these complex tumor forms to chemotherapy.
INTRODUCTION
One of the more perplexing problems in clinical cancer
chemotherapy is the wide range of responses that a given
drug or drug combination produces in apparently identical
cases. Although both the patients and the tumors are
indistinguishable in terms of standard criteria, one tumor
will completely regress following treatment, whereas an
other will progress not only after treatment but also during
treatment. This individual variation in responsiveness can
be attributed, at least in part, to variation in drug metabo
lism (10), tumor-inhibiting (9) or tumor-promoting (5) fac
tors, or the responsiveness of the tumor to the drug (4).
Presently, it is difficult to evaluate the role of each of these
(and other) factors, but the responsiveness of the tumorper
se must play a major role. For evaluation of this role,
however, it will be necessary to study the tumors without
the confounding influences of the other factors, i.e., in
vitro. For several years monolayer cultures have been used
for this purpose, in spite of the fact that their only similarity
to in vivo tumors is that the cell type is the same. Growing,
3-dimensional organized masses of tumor cells, which
possess many of the characteristics of solid tumors in vivo,
have been available, but they were grown under conditions
(3, 7) that made Chemotherapeutic studies impractical, and
very few tumors were thus adapted. More recently, we (11,
12) and others (6) have observed that MTS3 can be grown
from a variety of solid tumors merely by placing the cells, in
liquid medium, into plates that had been base coated with
agar. These MTS lines show wide variations in growth rate
1 Research was supported by Contract NIH-N01-CB-74203 and Grant 1P30-CA-21074-01from the National Cancer Institute.
2 To whom requests for reprints should be addressed: at Cancer Research
and Treatment Center, University of New Mexico, Albuquerque, N. M. 87131.
3 The abbreviations used are: MTS, multicellular tumor spheroids; EBME,
Eagle's basal medium; HN2, nitrogen mustard; CYC, cyclophosphamide.
Received April 24,1978; accepted July 28,1978.
NOVEMBER
of Radiology
[J. M. Y.), University
of New Mexico,
Albuquerque,
New
and growth fraction (11), thereby offering a tool for the
study of tumor-drug interactions. We report the develop
ment of 2 assay methods whereby one can analyze the
response of MTS in vitro following exposure to Chemother
apeutic drugs in vivo or in vitro.
MATERIALS
AND METHODS
Cells. Two cell lines were used in the present studies: a
radiation-induced mammary carcinoma of the BALB/c
mouse, MCa-11 (12); and the Line 1 lung carcinoma from
the same mouse (12). The lines were maintained as monolayer cultures in EBME supplemented with 10% fetal calf
serum,4 50 units penicillin per ml, 50 /*g streptomycin per
ml (Grand Island Biological Co., Grand Island, N. Y.), and
10 /¿gsodium insulin per ml (Elanco Products Co., Indian
apolis, Ind.).
MTS. The methods for producing and studying the
growth of MTS are the same as those described elsewhere
(11, 12). Approximately 106 cells in 10 ml of EBME are
plated in 100-mm plastic Petri dishes that have previously
been base coated with 0.75% Noble agar (Difco Laborato
ries, Inc., Detroit, Mich.) in complete EBME. Within 9 and 6
days, respectively, MCa-11 and line 1 MTS measuring 250
to 350 /j.m develop and are readily isolated for study.
Following treatment the MTS are returned to individual
agar-based 16-mm wells along with 1 ml of complete EBME.
The MTS are sized at least 3 times weekly on a dissecting
microscope, and the medium is changed twice weekly.
The MTS "cure" experiments were performed by placing
MTS in standard 16-mm wells after treatment. At 2-day
intervals after treatment, the wells, each containing 1 MTS,
were examined with both dissecting and phase-contrast
microscopes. At each inspection each well was scored as
positive or negative for detectable outgrowth. The lack of
detectable outgrowth is equated with cure in this system.
Drugs. Two drugs were purchased commercially and
used in the present studies: HN2 (Mustargen; Merck Sharp
and Dohme, West Point, Pa.); and CYC (Cytoxan; Mead
Johnson & Co., Evansville, Ind.). Immediately before use
the drugs were diluted with complete EBME for the in vitro
exposures or with 0.9% NaCI solution for in vivo injections.
Two-month-old female BALB/c mice served as the hosts for
the in vivo assays.
For the in vitro exposures, 12 to 18 MTS were placed in
60-mm agar-based dishes along with 5 ml of complete
medium for a 24-hr pretreatment. This pretreatment was
used to avoid the variable and unknown effects of crowding
' Fetal calf serum was dialyzed and absorbed on dextran-coated charcoal
(1) to reduce endogenous insulin and steroid hormones to as low a level as
possible.
1978
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1978 American Association for Cancer Research.
3595
J. M. Yuhas et al.
these data is to calculate delay in time to reach a given size
versus drug concentration. Over the range of 400 to 700
¿¿m,
the shape and slope of the curve were independent of
the final size chosen for analysis and were linear, with a
slope of 14.5 ±1.36 (S.E.) days delayed/pig/ml. This inde
pendence of size chosen for analysis is a reflection of the
fact that the drug has induced a dose-dependent lag period
but has not altered the growth rate once growth is resumed
(Chart 1).
Similar studies involving a variety of drugs, MTS, and
exposure schedules have been performed in our laboratory
and will be reported elsewhere. In our experience the shape
of the dose-response curve and its slope are a complex
product of the interaction among the drug, MTS, and
exposure pattern especially for more complex drugs, which
operate at a specific phase of the cell cycle. The range of
the sensitivity estimates (slopes) has not exceeded 10% in
RESULTS
our experience thus far, for a given drug/MTS combination.
Chart 1 is a plot of the growth of 300-^m MCa-11 MTS as
The responses of MTS to chemotherapeutic agents can
a function of time after a 1-hr exposure to graded doses of also be quantitated in terms of cure. Cure in the present
HN2. In actual practice the MTS ranged from 250 to 325 context is defined as the inability to show growth, in
standard tissue culture plates, by some presettime. Chart 2
/urn. Since the diameter increase per day is a linear function
(12), we normalized all groups to an initial size of 300 ¿¿m.is a plot of the percentage of 300-/¿mMCa-11 MTS, which
The primary effect of exposure to HN2 is to produce a dose- showed viable cell outgrowth in standard tissue culture
dependent increase in the "lag" period before the normal
dishes at varying periods after treatment as a function of
growth rate resumes (Chart 1). As had been used in the past the concentration of HN2, which they were exposed to for 1
hr. As time after exposure increases, the HN2 concentration
for in vivo experiments (8), a convenient way to analyze
required for 50% cure increases to a maximum of 1.28 /u,g/
ml at Day 17 (Chart 2) and afterward (data not shown). This
is a reflection of the fact that the time required to produce
a detectable outgrowth is inversely related to the number of
700 viable cells in the transferred MTS, which is in turn inversely
related to the HN2 concentration. The plateauing of the
HN2 concentration required to inhibit outgrowth by Day 17
and afterward suggests that all cells that are capable of
outgrowth do so by this time. The time at which cure rates
plateau depends on the MTS line studied, but it appears
600
independent of the drug under consideration. As will be
discussed in a subsequent report, however, cures are
difficult to achieve with cycle-specific and phase-specific
drugs unless the exposure time is prolonged far beyond 1
hr.
in the plates used to mass produce the MTS. The groups of
MTS were then transferred to similar plates containing the
appropriate concentrations of the respective drugs and
returned to the incubator for 1 hr. Following treatment the
MTS were removed from the drugs, washed twice with
complete medium, and placed individually in 16-mm agarbased wells along with 1 ml of medium.
For in vivo exposure of MTS, approximately 30 MTS were
injected i.p. in 0.5 ml of EBME. Fifteen to 30 min later, the
drugs were injected i.V.; 4 or 24 hrs later, the mice were
killed and the MTS were harvested from the peritoneal
cavity by means of a sterile Pasteur pipet. Fifty % of the
MTS can usually be retrieved. Following washing as de
scribed previously, the MTS were transferred to 16-mm
agar-based wells with 1 ml of EBME.
500
100
400
300
025
5
10
15
20
DAYS POST-TREATMENT
25
30
Chart 1. Growth of MCa-11 MTS as a function of time after exposure in
vitro to graded concentrations of HN2 (N, 12/group); •¿
0 ^g/ml; O, 0.25
fig/ml; x , 0.5 /¿g/ml;A, 1.0 /¿g/ml.
3596
050
075
1OO
125
ISO
175
200
Chart 2. Percentage of SOO-^m MCa-11 MTS cured by a 1-hr in vitro
exposure to graded concentrations of HN2 (H/V,). Cure is defined as the lack
of detectable viable cell outgrowth when the MTS are placed in standard
tissue culture wells. Inspection for outgrowth was made at 2-day intervals,
starting on Day 1 after treatment, through Day 21. After Day 17 the data
remained constant. •¿,
assayed on Day 9; O, assayed on Day 13; and x,
assayed on Day 17.
CANCER RESEARCH VOL. 38
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1978 American Association for Cancer Research.
Chemotherapy
Similar studies (data not shown) in which MCa-11 MTS
were exposed to HN2 and placed in agar-based wells
yielded a similar estimate of the concentration
of HN2 (1-hr
exposure) required for a 50% cure; i.e., all MTS exposed to
I /¿g/ml regrew within 30 days, whereas none of the MTS
exposed to 2 /¿g/mldid so.
The strict in vitro system described previously (in vitro
exposure and in vitro analysis), although providing a high
level of control and precision, cannot simulate the rise,
plateau, and fall of drug concentrations
in vivo without
requiring esoteric apparatus or a knowledge of the propor
tionality constants relating the effects of exposure to a
constant concentration
versus exposure to a rising and
falling one (2). A second problem of the strict in vitro system
is that it requires the use of additional biological supple
ments (e.g., liver microsomes) to allow the study of drugs
such as CYC, which require metabolic activation (2). Lastly,
the in vitro system is not a perfect duplication
of in vivo
environments,
which includes such factors as hormones,
¡mmunologically reactive effectors, etc. Although it is our
hope to eventually simulate the in vivo conditions in vitro, it
will first be necessary to develop an understanding of the in
vivo patterns so that they can be duplicated in vitro.
Toward this end we have been studying the in vitro
growth of MTS, which were exposed to chemotherapeutic
agents in vivo. In brief, MCa-11 MTS are injected into the
peritoneal cavity of syngeneic BALB/c mice, and 15 to 30
min later, the drug is injected i.v. Four or 24 hr later, the
mice are killed and the MTS are harvested from the peritonal cavity, placed in agar-based wells along with 1 ml of
medium, and followed as described previously.
In the absence of drug injection, MCa-11 MTS, which had
been placed in the peritoneal cavity of syngeneic mice,
showed a lag period in growth of 3 to 7 days, after which
the normal growth rate resumed (Chart 3a). With another
syngeneic tumor, Line 1 (Chart 3b), no such lag effect of a
24-hr peritoneal exposure was observed. This sensitivity of
the MCa-11 MTS to peritoneal exposure is barely detecta
ble, if the peritoneal exposure is limited to 4 hr (data not
shown), and does not appear to be due to the trauma of the
procedure. Chart 4 is a plot of the delay in growth for MCaII MTS due to a 24-hr peritoneal exposure in syngeneic
BALB/c mice that had been given a 250 mg/kg injection of
CYC between 1 and 15 days earlier. Injection of CYC 1 to 2
days before placing the MCa-11 MTS in the peritoneal cavity
resulted in a reduction of this peritoneal exposure effect.
Between 5 and 9 days after CYC, the peritoneal exposure
effect was even more pronounced than it was in controls,
but by Day 15 after CYC, it had returned to normal.
Chart 5 is a plot of the growth (in agar-based wells) of
MCa-11 MTS as a function of time after they have been
exposed, in the peritoneal cavity of BALB/c, for 15 to 30
min prior to and for 4 hr after the i.v. injection of graded
doses of CYC. Controls received no drug injection but were
placed in the peritoneal cavity for 4 hr. A fraction of each of
the groups treated with CYC failed to grow during the 30day observation period. This fraction increased directly with
drug dose (Chart 5) and was not included in the regrowth
calculations. As was the case for the strict in vitro system
(Chart 1), increasing drug doses served to increase the lag
period before the normal growth rate resumed. The plot of
and MTS
400
350
-100
0
DAYS
Chart 3. Growth of MCa-11 and line 1 MTS after a 24-hr exposure in the
peritoneal cavity of syngeneic female BALB/c mice ( •¿
) compared to the
growth of similar MTS that remained in vitro (O). Data are plotted as growth
since the sizes of the 2 lines differed at the time of analysis, a, MCa-11, 275
to 325 ¿im;b, line 1, 400 to 450 ^m. Size was not a factor in this comparison
since subsequent studies showed a similar effect for MCa-11 MTS over the
size range of 200 to 550 urn
8
12
510
6
8
DAYS POST
Chart 4. Delay in growth of MCa-11 MTS, due to a 24-hr exposure in the
peritoneal cavity of BALB/c mice, as a function of the time since the mice
had received an i.v. injection of CYC (250 mg/kg).
, delay observed in
mice that had never received CYC.
delay in time to reach sizes of 400 or 500 /¿mversus injected
drug dose was linear, with a slope of 0.08 ±0.004 day/mg/
kg.
DISCUSSION
As a complex tissue composed of tumor cells in varying
physiological states, MTS do not provide the precise esti
mates of the kinetics of drug-induced
cell killing that one
NOVEMBER 1978
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1978 American Association for Cancer Research.
3597
J. M. Yuhas et al.
550-
4
6
8
10
12
14
POST-TREATMENT
(DAYS)
16
Chart 5. In vitro growth of 300-jim MCa-11 MTS, which had been placed
in the peritoneal cavity of BALB/c mice 15 to 20 min prior to and had been
allowed to remain for 4 hr after i.v. injections of O(O), 50 (x), 100(«),or 200
(A) mgCYC per kg.
obtains in monolayer experiments. In fact we have not
identified which subpopulations of cells in the MTS are
being injured or whether this injury is cell killing, mitotic
delay, both, or neither. Admittedly, the drug responses of
MTS must eventually be described in more precise terms,
but their major advantage is their similarity to in vivo tumors
(3, 7, 11,12). Whether this similarity will allow prediction of
in vivo tumor responses remains to be determined.
Each of the 2 methods for exposing MTS, described previ
ously, possesses certain advantages and disadvantages. The
strict in vitro system allows precise quantitation of drug dose,
but in doing so it fails to parallel the oscillating drug concen
trations that an in situ tumor would encounter in v/Vo.The in
vivo exposure system produces the rising and falling drug
kinetics, but the drug dose to the peritoneal contents may
or may not reflect or be proportional to the drug dose to an
in situ tumor. This is a particularly important consideration
in any comparison involving one type of MTS and a series
of drugs. The relationship between the amount of drug
delivered to an in situ tumor and that delivered to the
peritoneal contents may vary widely, among a series of
drugs, thereby leading to a false estimate of which drug
would be most effective for the tumor growing in situ if the
MTS responses were compared in terms of injected dose.
Determination of the relative amounts of drugs delivered to
in situ tumors and peritoneal contents are presently in
progress, and it should eventually prove possible to adjust
the injected dose for each drug so that the kinetics of drug
appearance and disappearance in the peritoneal cavity
mimic the kinetics at the in situ tumor site.
Although the advantages of using a common host are
obvious, this/n vivo exposure approach cannot be used for
tumors other than those of host origin unless it is known
that the pharmacokinetics of the drug are similar in the host
species and in the species of tumor origin. Likewise, it will
be necessary to determine whether xenogenic tumors suffer
3598
the effects of cross-species immunological reactions, and if
so, how these might be abrogated. We have, however,
initiated pilot studies with human breast cancer MTS, such
as those derived from the MDA-361 cell line (13), and have
been able to develop drug dose-response curves similar to
those presented previously, by using a 4-hr i.p. exposure in
both mice and rats.
In summary, the systems described previously provide a
simple means of studying the response of very complex
tumor cell populations to chemotherapeutic agents. Both
in vitro and in vivo exposure methods can contribute to our
understanding of these responses, with each providing a
different aspect of the process. However, before it will be
possible to compare the responses of MTS versus those of
in situ tumors within a species, the pharmacokinetics of
each drug must be understood in terms of tissue and site
differences. Further, use of this system to study MTS from
other than the host species will require a full knowledge of
the species differences in pharmacokinetics, as well as any
cross-species natural immunity.
REFERENCES
1. Armelin, H. A., Wishikawa, K., and Sato, G. H. Control of Mammalian
Cell Growth in Culture; The Action of Protein and Steroid Hormones as
Effector Substances. In: B. Clarkson and R. Baserga (eds.), Control of
Proliferation in Animal Cells, pp. 97-104. Cold Spring Harbor, N. Y.:
Long Island Biological Association, 1974.
2. Bagley, C. M., Jr., Bostick, F. W., and DeVita, V. T. Clinical Pharmacol
ogy of Cyclophosphamide. Cancer Res., 33: 226-233, 1973.
3. Folkman, J., Hachberg, M., and Knighton, D. Self Regulation of Growth
in Three Dimensions. In: B. Clarkson and R. Baserga (eds.), Control of
Proliferation in Animal Cells, pp. 833-842. Cold Spring Harbor, N. Y.:
Long Island Biological Association, 1974.
4. Jackson, R. C., Hart, L. I., and Harrap, K. R. Intrinsic Resistance to
Methotrexate of Cultured Mammalian Cells in Relation to the Inhibition
Kinetics of Their Dehydrofolate Reductases. Cancer Res., 36: 19911997, 1976.
5. Manni, A., Trujillo, J. E., and Pearson, O. H. Predominant Role of
Prolactin in Stimulating the Growth of 7,12-Dimethylbenz(a)anthraceneinduced Rat Mammary Tumors. Cancer Res., 37: 1216-1219,1977.
6. Steuer, A. F., and Ting, R. C. Formation of Larger Cell Aggregates by
Transformed Cells: An In Vitro Index of Cell Transformation. J. Nati.
Cancer Inst., 56: 1279-1280,1976.
7. Sutherland, R., and Durand, R. Radiation Response of Multiceli Sphe
roids—Anin Vitro Tumor Model. Intern. J. Radiation Biol.. 11: 87-139
1976.
8. Thomlinson, R. H. An Experimental Method for Comparing Treatments
in Intact Malignant Tumors in Animals and Its Application to the Use of
Oxygen in Radiotherapy. Brit. J. Cancer, 14: 555-563, 1963.
9. Van Wijk, R., Wicks, W. D., Severs, M. M., and Van Rijn, J. Rapid Arrest
of DNA Synthesis by W,O2'-Dibutyryl Cyclic Adenosine 3',5'-Monophosphate in Cultured Hepatoma Cells. Cancer Res.,33: 1331-1338, 1973.
10. Wan, S. H., Haffman, D. H., Azarnoff, D. L., Stephens, R., and Hoogstraten, B. Effect of Route of Administration and Effusions on Methotrexate
Pharmacokinetics. Cancer Res., 34: 3487-3491, 1974.
11. Yuhas, J. M., and Li, A. P. The Growth Fraction as the Major Determinant
of Multicellular Tumor Spheroid Growth Rates. Cancer Res., 38: 15281532, 1978.
12. Yuhas, J. M., Li, A. P., Martinez, A. O., and Ladman, A. J. A Simplified
Method for the Production and Growth of Multicellular Tumor Sphe
roids. Cancer Res., 37: 3639-3643, 1977.
13. Yuhas, J. M., Tarleton, A. E., and Molzen, K. B. Multicellular Tumor
Spheroid Formation by Breast Cancer Cells Isolated from Different Sites.
Cancer Res.. 38. 2486-2491, 1978.
CANCER
RESEARCH
VOL. 38
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1978 American Association for Cancer Research.
In Vitro Analysis of the Response of Multicellular Tumor
Spheroids Exposed to Chemotherapeutic Agents in Vitro or in
Vivo
John M. Yuhas, Anne E. Tarleton and James G. Harman
Cancer Res 1978;38:3595-3598.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/38/11_Part_1/3595
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1978 American Association for Cancer Research.