вход по аккаунту



код для вставкиСкачать
Human Tumor Xenografts in Nude Mice Are
Not Always of Human Origin
A Warning Signal
e read with interest the article by Pathak et al.1 and the accompanying editorial2 in Cancer. The article cautions the scientific
community working with human tumor xenografts in nude mice and
rats about the possibility that what is expected to be human material
can be something different. The authors report that after human
tumors are implanted into athymic rodents, transformed host cells
with a malignant phenotype may be harvested. The editorial reminds
us of the experience we had in the early 1980s, when few in vivo
human tumor lines were commercially available. At that time we
began to grow and establish human tumor lines in nude mice, starting from the surgical material. In general, we had few problems in
making human tumor specimens grow subcutaneously in the nude
mouse, at least for the tumor types we tested, i.e., colon carcinoma,
nonsmall and small cell lung carcinoma, and ovarian carcinoma.
What was a serious problem was maintaining in line the small cell
lung carcinoma xenografts, because in 4 –5 years we lost all (of more
than 15!) of the tumor lines we had “established.”3 Ten years ago, we
published a report of such a case, in which a murine fibrosarcoma
was harvested following subcutaneous injection into nude mice of a
human small cell lung carcinoma.4 Indeed, we are perfectly aware of
the risk that in the “human tumor/nude mouse” model, malignant
transformation of host cells by human tumor material may occur,
particularly for some tumor types (small cell lung carcinoma in our
experience and prostate carcinoma and osteosarcoma in the experience of Pathak et al.). Nevertheless, we believe that the danger of
confusing murine with human material is rather limited. In fact, in
our experience, every time a growing tumor “changed nature,” its
growth properties were markedly affected and the growth rate rapidly
increased, making it relatively easy to suspect the tumor that was “not
human.” Differences were rather impressive, and Figure 1 shows
representative growth curves in nude mice of a small cell lung carcinoma, PoFG, before and after its transformation. In our experience,
human tumor xenografts from surgery specimens grow slowly in
mice, with doubling times .10 days (in case of the human PoFG, it
was about 20 days). However, when the tumor was transformed,
growth became very rapid (the doubling time was about 4 days for the
murine PoFG). When transformation occurred, although the latency
time for the tumor to become visible in the mouse flank was sometimes very long (even longer than 3 months), in the next passage the
tumor appeared in a few days and grew rapidly. In many cases,
histologic observation of the harvested material did not reveal relevant changes in tumor tissue appearance.
In addition to careful observation of tumor growth properties,
serial control of the human origin of the material by analysis of its
© 1999 American Cancer Society
Pathak S, Nemeth MA, Multani AS. Human tumor xenografts in nude mice are not always of human origin: a
warning signal. Cancer 1998;83:1891–3.
Krishan A. When a human tumor xenograft is not a human
tumor. Cancer 1998;83:1889 –90.
Pratesi G, Tortoreto M, Corti C, Capranico G, Binaschi M,
Pilotti S, et al. Small cell lung cancer xenografts in drug
resistance studies. In: Fiebig HH, Berger DP, editors. Immunodeficient mice in oncology. Basel: Karger, 1992;42:
472– 4.
Soranzo C, Ingrosso A, Pratesi G, Lombardi L, Pilotti S,
Zunino F. Malignant transformation of host cells by a human small cell lung cancer xenografted into nude mice.
Anticancer Res 1989;9:361– 6.
Pesce AJ, Bubel HC, DiPersio L, Michael JG. Human lactic
dehydrogenase as a marker for human tumor cells grown in
athymic mice. Cancer Res 1977;37:1998 –2003.
Graziella Pratesi, Ph.D.
Monica Tortoreto, Bc.S.
Preclinical Chemotherapy and Pharmacology Unit
Istituto Nazionale Tumori
Milan, Italy
FIGURE 1. Growth curves are shown for a small cell lung carcinoma, PoFG,
xenografted into nude mice.
Author Reply
FIGURE 2. Patterns of LDH isoenzymes are shown. Lane 1: murine control;
lane 2: human control; lane 3: human tumor xenograft; lane 4: transformed
tumor xenograft.
lactic dehydrogenase isoenzyme pattern appears to
be a useful procedure. This is a very fast and inexpensive assay5 that we routinely perform for all human tumor lines. Because the material we analyze
grows in a murine host, a minor murine contamination is expected. However, the different patterns
of a human tumor and a transformed tumor are
clearly distinguishable (Fig. 2).
In conclusion, in addition to cytogenetic techniques that can be used for the identification of interspecies contamination, a careful observation of tumor
growth properties and a simple enzymatic assay may
help investigators to avoid wasting time, effort, and
money and to feel more confident about their experimental results.
ratesi and Tortoreto have made some interesting
suggestions, including the use of cytogenetic techniques, tumor growth properties, and enzymatic assay
for the identification of interspecies cell line contamination. They published an article in 1989 describing
their experience of murine host cell transformation
with human small cell lung carcinoma. We somehow
missed citing this work in our 1998 report in Cancer.
Their experience indicates that there are alterations in
growth properties, with doubling time drastically decreasing in the transformed host cell tumor. They also
suggest the use of a lactate dehydrogenase (LDH)
isoenzyme assay for distinguishing between the human and the transformed host cells.
In the many samples that we have analyzed since
early 1970, we have observed that this phenomenon is
limited to not only human cancer cells (breast, colon,
osteosarcoma, prostate, brain, melanoma, and renal
cell) and transformed human embryonic lung cells,
but also tumors of other mammalian species, including melanomas of swine origin, marsupial skin tumor
cells induced by ultraviolet light (our unpublished
data), murine leukemia cells, and transformed rat cells
injected into baby Syrian hamsters—all have the capability of host cell transformation.1–5 Furthermore, it
is not always the case that only 100% transformed host
cells are present. There could be a mixture of trans-
CANCER September 1, 1999 / Volume 86 / Number 5
formed host cells as well as the original injected human tumor cells in various proportions. It is, therefore, not easy to identify and differentiate the two
different cell populations based on growth properties
and isoenzyme patterns, especially when a situation
such as 95% human cells and 5% noncycling mouse
cells is present. We recently demonstrated that a human osteosarcoma xenograft in nude mice, when removed and cultured, contained all metaphases of human origin, but a total human DNA fluorescence in
situ hybridization (FISH) analysis showed that a small
percentage of noncycling murine interphase cells was
also present.6 In this study we clearly demonstrated
that conventional cytogenetics with banding was not
sufficient alone for identifying interspecies cell line
contamination. In this same tumor at a later passage,
we found both mouse and human metaphases. Only
at this stage would a visible change in growth properties be noticed or an isoenzyme assay be effective. We
observed a similar phenomenon in 1974, in distinguishing between rat and human contaminated tumor
cells that were received from Miami (our unpublished
data). These dynamic changes occurring in the cell
populations can be rapidly identified by conventional
as well as molecular cytogenetics. However, conventional cytogenetics alone may fail to identify the nondividing murine host cells due to the lack of such
metaphase spreads in the preparations.6
Our main intentions were to demonstrate that 1)
murine host cells can be transformed in some cases by
the tumor cells originally inoculated, and 2) this might
be happening in some human cancer patients as well.
It is, therefore, very important to analyze those metastatic human tumors that are either partially or completely nonregressing with conventional therapy
and/or radiation therapy. We therefore recommend
that human or nonmurine cancer cells that are to be
xenographed into nude mice or rats be studied not
only with conventional cytogenetics or isoenzyme
(LDH) assay, as suggested by Pratesi and Tortoreto,
but also by the molecular cytogenetic FISH technique
using species specific total DNA probes. In this way,
even a small percentage of nondividing cells of another species, if present, can be definitely identified.
Pathak S, Hsu TC, Trentin JJ, Butel JS. Nonrandom chromosome changes in Syrian hamster embryo cell clones transformed by various viruses. Sixteenth Annual Somatic Cell
Genetic Conference, Los Alamos, New Mexico, 1977.
Pathak S, Hsu TC, Trentin JJ, Butel JS, Panigrahy B. Nonrandom chromosome abnormalities in transformed Syrian
hamster embryo clones. In: Arrighi FE, Rao PN, Stubblefield
E, editors. Genes, chromosomes, and neoplasia. New York:
Raven Press, 1981:405–18.
Bowen JM, Cailleau R, Giovanella B, Pathak S, Siciliano MJ.
A retrovirus-producing transformed mouse cell line derived
from a human breast adenocarcinoma transplanted in a
nude mouse. In Vitro 1983;19:635– 41.
Ozen M, Multani AS, Kuniyasu H, Chung LWK, von Eschenbach AC, Pathak S. Specific histologic and cytologic evidence for in vivo malignant transformation of murine host
cells by three human prostate cancer cell lines. Oncol Res
1997;9:433– 8.
Pathak S, Nemeth MA, Multani AS, Thalmann GN, von Eschenbach AC, Chung LWK. Can cancer cells transform normal host cells into malignant cells? Br J Cancer 1997;76:
1134 – 8.
Multani AS, Pathak S. Conventional cytogenetics alone is
not sufficient for identifying interspecies cell line contamination. Anticancer Res. In press.
Sen Pathak, Ph.D.
Margit A. Nemeth, M.S.
Asha S. Multani, Ph.D.
Department of Cancer Biology
The University of Texas M. D. Anderson
Cancer Center
Houston, Texas
Front-Line Chemotherapy with
Cisplatin and Etoposide for Patients
with Brain Metastases from Breast
Carcinoma, Nonsmall Cell Lung
Carcinoma, or Malignant Melanoma
A Prospective Study
he article by Franciosi et al.,1 who investigated
front-line chemotherapy in 116 patients with brain
metastases (107 were included in their analysis)
treated at 9 different centers in Italy during a period of
6.5 years, merits comment. With appreciable effort the
Italian Oncology Group for Clinical Research performed what to their knowledge is one of the largest
trials in which chemotherapy was administered as
front-line treatment for brain metastases. Cisplatin
and etoposide were given for a maximum of six cycles
(median, three cycles) to patients not amenable to
surgery. Eligibility criteria included no prior treatment
for brain metastases, normal bone marrow and renal
function, and lack of severe cardiac disease. No restrictions were made with regard to established prognostic factors2,3 such as performance status, age, or
extracranial disease status. Follow-up included routine computed tomography (CT) scans of the brain
after the second course and at every other course
thereafter. Radiation therapy was allowed after the
end of protocol chemotherapy. However, the authors
did not provide any information regarding the number of patients receiving radiation therapy or any kind
of salvage treatment for progressive brain metastases.
As observed in previous clinical trials, the high response rate (complete plus partial remission [cr 1 pr]
in 55% of the patients) of their initial Phase II study
was not reproducible in a larger multiinstitutional setting (cr 1 pr in 38% of patients with brain metastases
from breast carcinoma and 30% of patients with brain
metastases from nonsmall cell lung carcinoma
[nsclc]). Approximately 50% of patients who showed a
cr or pr of brain metastases also responded at other
sites. Certainly, this study raises important questions
with regard to the future management of patients with
brain metastases, in particular those with breast and
lung carcinoma (including small cell carcinoma [sclc]
and nsclc), which were found to be the most common
primary tumors in large clinical series.2–5 The authors1
may be able to contribute further to the continuing
search for the best treatment strategies by providing
more detailed results of their study. Because their
chemotherapy regimen was found to be ineffective in
eight patients with brain metastases from malignant
melanoma (no objective response), the remainder of
my comments will focus on patients with breast and
lung primary tumors, including sclc. The latter was
not included in the study by Franciosi et al.,1 although
other groups reported comparable response rates for
sclc using, for example, teniposide (cr 1 pr in 33% of
patients).6 Therefore, it appears justified to include
sclc in the discussion, which hopefully will contribute
to the design of future studies comparing the effects of
chemotherapy and radiation therapy.
Clinical and Radiologic Response
Recently, radiologic response rates of 30 –35% have
been reported after chemotherapy for brain metastases from both sclc and nsclc.1,6 – 8 Summarizing our
own data, it could be speculated that slightly higher
radiologic response rates may be observed after radiation therapy (cr 1 pr in .60% of breast carcinoma
metastases and 40 –50% of squamous cell carcinoma
and adenocarcinoma metastases, respectively [the
latter included some nonlung primary tumors as
well]).4,9 This holds true when examining the results of
150 lesions, each of which was scored separately,9 as
well as results of 144 patients,4 which were evaluated
using the same criteria as in the study by Franciosi et
al.1 Certainly, this hypothesis can be verified only in a
prospective trial because various selection factors may
have influenced these differences. It may even be
more important to compare clinical response rates
and quality-adjusted survival, which are the most important criteria from the patients’ point of view. Unfortunately (and comparable to many radiation therapy studies), Franciosi et al.1 did not report any data
regarding palliation or quality of life. The same holds
true for location of failure (development of new lesions versus progression of treated lesions) and late
Survival and Time to Progression
The median survival was approximately 8 months for
both breast carcinoma and nsclc patients (1-year survival rates were approximately 33% and 25%, respectively).1 The median time to progression was approximately 4 months and the median duration of
response was approximately 8 months in both groups,
respectively. Three patients died from toxic events,
some of which may have been influenced by additional corticosteroid treatment.
Conversely, remission rates and time to progression appear at best to be equivalent to those reported
from radiation therapy trials. However, survival data
appear to be better, which possibly may be explained
by the activity of chemotherapy in extracerebral disease sites (a desirable effect because, e.g., patients
with uncontrolled extracranial disease treated with
stereotactic radiosurgery had a median survival of
only 4.4 months, despite a local control rate of 92% in
the brain),5 the administration of salvage therapy, or
just different distribution of prognostic factors (e.g.,
the chemotherapy-related inclusion criteria mentioned earlier usually are not considered in radiation
therapy trials). The latter explanation may be supported by the shorter survival times (e.g., 5 months7)
reported in other chemotherapy studies. Over the
course of .10 years, it increasingly has been discussed
that chemotherapy may be effective in the treatment
of brain metastases from breast and lung carcinoma.1,6 – 8,10,11 Active agents also include 5-fluorouracil, cyclophosphamide, and methotrexate, as well as
teniposide and carboplatin.6 – 8,10,11 Several recent trials reported chemotherapy-associated death rates of
3–10%.1,6,7 This is an important issue regarding the
palliative nature of the treatment. Franciosi et al.1
suggested that additional, especially randomized,
studies are warranted to compare the results of chemotherapy and radiation therapy. However, this suggestion has been made for several years without actually resulting in a properly designed trial. From my
point of view, the authors of the majority of chemotherapy trials missed the opportunity to provide detailed information regarding predictive factors for re-
CANCER September 1, 1999 / Volume 86 / Number 5
sponse, prognostic factors for survival, and the
influence of additional radiation therapy as well as
various salvage modalities on outcome. This information would be desirable for a rationally designed randomized study to decide for example whether to include patients without extracranial disease or those
pretreated with chemotherapy and to choose both a
suitable combination of drugs (or different combinations for different primary tumors) and a suitable radiation therapy arm for comparison. The number of
patients presented by Franciosi et al.1 as well as the
close CT follow-up in their study may provide a basis
for more detailed analyses of predictive and prognostic factors than presented in their article, possibly
helping us to move this issue further.
Franciosi V, Cocconi G, Michiara M, Di Costanzo F, Fosser
V, Tonato M, et al. Front-line chemotherapy with cisplatin
and etoposide for patients with brain metastases from
breast carcinoma, nonsmall cell lung carcinoma, or malignant melanoma: a prospective study. Cancer 1999;85:1599 –
2. Gaspar L, Scott C, Rotman M, Asbell S, Phillips T, Wasserman T, et al. Recursive partitioning analysis (RPA) of prognostic factors in three Radiation Therapy Oncology Group
(RTOG) brain metastases trials. Int J Radiat Oncol Biol Phys
3. Priestmann TJ, Dunn J, Brada M, Rampling R, Barker PG.
Final results of the Royal College of Radiologists’ trial comparing two different radiotherapy schedules in the treatment of cerebral metastases. Clin Oncol (R Coll Radiol)
1996;8:308 –15.
4. Nieder C, Nestle U, Walter K, Schnabel K. Dose/effect relationships for brain metastases. J Cancer Res Clin Oncol 1998;
124:346 –50.
5. Pirzkall A, Debus J, Lohr F, Fuss M, Rhein B, EngenhartCabillic R, et al. Radiosurgery alone or in combination with
whole-brain radiotherapy for brain metastases. J Clin Oncol
6. Postmus PE, Smit EF, Haaxma-Reiche H, Van Zandwijk N,
Ardizzoni A, Quoix E, et al. Teniposide for brain metastases
of small cell lung cancer: a phase II study. J Clin Oncol
1995;13:660 –5.
7. Minotti V, Crino L, Meacci ML, Corgna E, Darwish S, Palladino MA, et al. Chemotherapy with cisplatin and teniposide
for cerebral metastases in non-small cell lung cancer. Lung
Cancer 1998;20:93– 8.
8. Malacarne P, Santini A, Maestri A. Response of brain metastases from lung cancer to systemic chemotherapy with
carboplatin and etoposide. Oncology 1996; 53:210 –3.
9. Nieder C, Berberich W, Schnabel K. Tumor-related prognostic factors for remission of brain metastases after radiotherapy. Int J Radiat Oncol Biol Phys 1997;39:25–30.
10. Rosner D, Nemoto T, Warren WL. Chemotherapy-induced
regression of brain metastases in breast carcinoma. Cancer
11. Boogerd W, Dalesio O, Bais EM, Van der Sande JJ. Response
of brain metastases from breast cancer to systemic chemotherapy. Cancer 1992;69:972– 80.
Carsten Nieder, M.D.
Department of Experimental
Radiation Oncology
The University of Texas
M. D. Anderson Cancer Center
Houston, Texas
Author Reply
e appreciate the opportunity to reply to the comments of Dr. Nieder regarding our article.1 Dr.
Nieder pointed out that the median survival of approximately 8 months observed in our patients with
brain metastases (BM) from breast carcinoma (BC)
and nonsmall cell lung carcinoma (NSCLC) is better
than that reported in the majority of whole brain
radiotherapy (WBRT) trials (approximately 3– 6
months).2 This may be due to the activity of chemotherapy (CT) on extracerebral disease but also to the
administration of salvage treatment at the end of chemotherapy or to a different distribution of prognostic
factors in patients enrolled in trials of CT and of RT.
With regard to salvage treatment, we reviewed our
database concerning BC and NSCLC patients. Only a
small proportion received salvage brain RT (i.e., 21 of
99 patients [21%]); 71 patients (72%) (38 with BC and
33 with NSCLC) did not receive RT; we do not have
information regarding the remaining 7 patients. We
did not report analysis of additional CT at the end of
protocol CT, but to our knowledge this information
usually is not provided in studies enrolling patients
with metastatic disease, especially if they possibly
were pretreated with previous CT. Indeed, 50% of our
BC patients were pretreated at time of randomization.
However, it is reasonable to suppose that, in BC patients not previously treated, another salvage CT may
be administered on a personalized basis at the time of
progressive disease as a palliative measure. This cannot be refused even to patients pretreated with brain
RT. For patients with NSCLC, salvage CT did not exist,
at least at the date of our enrollment of the patients.
With regard to prognostic factors, we do not know
the predictive or prognostic factors requested by Dr.
Nieder in addition to those already reported in the
article. However, considering prognostic factors defined by Gaspar et al.,3 it is true that our series mostly
included patients age ,65 years with an Eastern Cooperative Oncology Group performance status of 0-2
and with a neurologic function status of 1-2. However,
we have to take into account that 45 BC patients (80%)
and 36 NSCLC patients (84%) had additional extracranial disease, that 36 BC patients (64%) and 24 NSCLC
patients (56%) had multiple BM, and that 27 NSCLC
patients (63%) had nonsquamous histology. All these
are defined as significant poor prognostic factors.
For these reasons we cannot exclude that salvage
treatment or a different distribution of prognostic factors could have influenced survival, but there is no
evidence against the hypothesis that front-line chemotherapy itself could have contributed to the longer
survival observed in our study compared with that
reported in the majority of RT studies. We have to take
into account that RT and CT are qualitatively different
modalities, having local and systemic activity, respectively. Therefore although it is possible, comparing
response data reported in different trials, to hypothesize that WBRT can induce more frequently objective
response in BM, compared with our CT, it even is
possible that the cytoreductive activity of the combination of cisplatin and etoposide even on extracerebral disease, may have contributed to a better survival.
We believe that a strong element in favor of our survival figures is the prospective centralized registration
of patients and the strict intention-to-treat analysis of
the results.
Unfortunately our database does not contain sufficient data concerning the palliation of symptoms
and we did not consider quality of life assessment at
time of study design, as happened for the majority
studies having approximately the same dates for patient enrollment. With regard to neurotoxicity, we did
not report analytic information that already was included in our previous trials utilizing the same CT,4,5
but we may say that, due to the fact that it is a cumulative and late toxicity, it was not a major problem for
our patients, many of whom received only a few cycles
and/or had a rather short survival. In general, we
agree that the toxicity of CT must be considered when
selecting treatment of patients with BM. However, in a
typical life-threatening situation such as that of the
appearance of BM in a patient with incurable disease,
informed patients could accept the risk of toxicity with
this CT combination against the risk of an early death.
We observed three toxic deaths (3%), two of which
possibly were related to the concomitant steroid treatment. The risk of a lethal consequence of this CT in a
critical clinical situation such as that of the appearance of BM should alert the physician to monitor
patients very carefully and possibly to exclude those in
whom a further increase in intracerebral pressure,
favored by acute hydration before cisplatin administration, should be considered to be contraindicated.
We do not agree with Dr. Nieder’s suggestion also
to include in the discussion the situation of BM from
small cell lung carcinoma (SCLC). This does not mean
that treatment with the combination of cisplation and
etoposide or with RT is not advisable in patients with
BM from SCLC. Rather, there are too many conceptual
and practical objections to discuss this situation together with those included in our article (i.e., BM from
BC and NSCLC). The former are related to a totally
different natural history, chemosensitivity, and possibly radiosensitivity and to the substantially different
number of active agents to be considered as previous
CT or possibly as late salvage CT. The latter mostly are
related to the different period in which BM occur in
the course of these different diseases. In SCLC the
appearance of late BM in patients still in complete
response in extracerebral disease is rather common as
a sign of primary treatment failure. These patients
may or may not have already had brain irradiation. In
this last case we believe that salvage RT to the brain is
the best front-line choice.
Franciosi V, Cocconi G, Michiara M, Di Costanzo F, Fosser
V, Tonato M, et al. Front-line chemotherapy with cisplatin
and etoposide for patients with brain metastases from
breast carcinoma, nonsmall cell lung carcinoma, or malignant melanoma: a prospective study. Cancer 1999;85:1599 –
Gelber R, Larson M, Borgelt B, Kramer S. Equivalence of
radiation schedules for the palliative treatment of brain
metastases in patients with favorable prognosis. Cancer
1981;48:1749 –53.
Gaspar L, Scott C, Rotman M, Asbell S, Phillips T, Wasserman T, et al. Recursive partitioning analysis (RPA) of prognostic factors in three Radiation Therapy Oncology Group
(RTOG) brain metastases trials. Int J Radiat Oncol Biol Phys
Cocconi G, Lottici R, Bisagni GC, Bacchi M, Tonato M,
Passalacqua R, et al. Combination therapy with platinum
and etoposide of brain metastases from breast carcinoma.
Cancer Invest 1990;8:327–34.
Crinò L, Clerici M, Figoli F, Carlini P, Ceci G, Cortesi E, et al.
Chemotherapy of advanced non-small-cell lung cancer: a
comparison of three active regimens. A randomized trial of
the Italian Oncology Group for Clinical Research (GOIRC).
Ann Oncol 1995;6:347–53.
Vittorio Franciosi, M.D.
Giorgio Cocconi, M.D.
Division of Medical Oncology
Azienda Ospedaliera
Parma, Italy
Без категории
Размер файла
88 Кб
Пожаловаться на содержимое документа