[CANÅ’R RESEARCH 56. 1493-1497. April I. I996|
Advances in Brief
Genetic Divergence in the Clonal Evolution of Breast Cancer1
Hiroaki Fujii, Carla Marsh, Paul Cairns, David Sidransky, and Edward Gabrielson2
Department of Pathology ¡H.F.. C. M., E. G./. Oncology Center [D. S.. E. G./, and Head and Neck Cancer Research Laboratory ¡P.C., D. S.¡.Johns Hopkins University School
of Medicine, Baltimore, Maryland 21224
Abstract
The progression of ductal carcinoma in situ (DCIS) to infiltrating and
metastatic cancer of the breast is thought to be a consequence of clonal
expansions of neoplastic cells with progressively more genetic alterations.
To study this progression, we first dissected multiple foci from each of 23
breast tumors with DCIS only and 20 cases with synchronous DCIS and
infiltrating cancer. We then tested microsatellite markers by PCR for
allelic losses in the individual foci for loci on chromosomes 6q, 9p. llq,
13q, 16q, I7i|. and 17p. The patterns of allelic losses identified in the in situ
cancers were generally conserved in the synchronous infiltrating tumors,
supporting the paradigm that the infiltrating tumors are clonally derived
from the in situ lesions. However, in 8 (40%) of the 20 cases with
synchronous in situ and invasive cancer, heterogeneous patterns of allelic
loss at one or more chromosomal loci were observed in adjacent DCIS
foci. Moreover, some of the allelic losses recognized in in situ portions of
the tumors were not conserved in the clonal progression to the synchro
nous invasive tumor. Such allelic loss heterogeneity was noted in only 1 of
the 20 infiltrating tumors and only 3 of the 23 cases of DCIS without
invasion that were studied in a similar manner. This heterogeneity indi
cates genetic divergence during the clonal evolution of breast cancer,
particularly at the time when in situ cancers progress to invasive cancers.
Introduction
Breast cancer is recognized by hisiological studies to progress
through a series of in situ stages prior to the development of infiltrat
ing and metastatic cancer. Although previous studies have identified
chromosomal loci with nonrandom allelic losses in cases of invasive
breast cancer [reviewed by Devilee et a/.(l)], the molecular changes
responsible for progression to advanced disease are not yet well
understood. Recently, tumors with synchronous /'/; situ and invasive
cancer were analyzed (2, 3), and the allelic losses observed in the HI
situ lesions were found to be conserved in the synchronous invasive
cancers. These data supported the paradigm of clonal derivation of the
invasive cancer from the in situ cancer.
Most of the previous genetic studies on HI situ breast cancer have
been performed on DNA from either single in situ lesions or aggre
gates of multiple HI situ lesions. We reasoned that this approach may
not be adequate to define the sequence of genetic events in the
progression of the individual tumor, since distinct foci at various
stages of progression may be present within one tumor mass. For this
study, multiple tumor foci from 20 cases with DCIS' and synchronous
invasive breast cancer and 23 cases with DCIS only were individually
microdissected from sections of archival breast cancer tissues. Critical
suppressor loci on chromosomes 6q, 9p. 1Iq, 13q, 16q, 17q, and 17p
Received 12/19/95: accepted 2/15/96.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked culvenisemeni in accordance with
18 U.S.C. Section 1734 solely to indicate this fact.
1This work was supported in part by Grant R21 CA/ES66204 from the National
Cancer Institute.
2 To whom requests for reprints should be addressed, at Department of Pathology.
Johns Hopkins Bayview Medical Center. 4940 Eastern Avenue, Baltimore, MD 21224.
' The abbreviations used are: DCIS. ductal carcinoma in situ: LOH, loss of
heterozygosity.
were then characterized for allelic loss using PCR-amplified micro-
satellite markers for each of the individual neoplastic foci.
Materials and Methods
Paraffin-embedded breast cancer tissues with synchronous DCIS and inva
sive ductal carcinoma, or DCIS only, were obtained from the surgical pathol
ogy files of the Johns Hopkins Hospital and the Johns Hopkins Bayview
Medical Center. All pathology was reviewed for consistency of diagnosis. For
each tissue block, multiple sections 12 /j.m in thickness were deparaffini/ed.
stained with H&E, visualized with an inverted microscope, and dissected using
a 26-gauge needle. Tissue was digested overnight in butter containing 0.5%
NP40 and 200 /xg/ml proteinase K at an approximate ratio of I /^l huftcr/20
cells. This lysate was used directly in PCR reactions.
PCR amplifications of microsutellite markers were performed using primers
for 6q21-25 (D6S292, D6S3II. D6S3IO. D6S473. and D6S255). 9p2l
(D9S17I), llql3-23 (Int2. DIÃOES29. D1ÃOES35. and D11S528). I3ql2-14
(DI3S260, DI3S263. and DI3SI55), 16ql2-24 (DÃOE6S54ÃOE, D16S4I5.
DJ6S265. and DI6S402). 17p (D17S513. CHRNBI. TP53. DI7S786, and
D17S122). and 17ql2-24 (THRAI. DI7S579, and D17S588) obtained from
Research Genetics (Huntsville. AL), and additional primers for microsalellite
markers on 9p21 (D9SÃOE748. D9SI749, D9SÃOE75I,and D9S1752) were prepared
as previously described (4). PCR reactions typically contained 1 /il lysate and
other reagents as previously described (5). PCR products were then separated
using denaturing gel electrophoresis, and allelic loss was determined by at least
a 75% reduction in the relative intensity of one alÃ-elein the tumor compared
to normal after autoradiography. Allelic loss was usually confirmed by obser
vation of LOH at multiple informative markers mapped to the same chromo
somal region. When only one informative marker was recognized for a
chromosomal region in a particular tissue sample. LOH was confirmed by
repeating the PCR amplification of that marker.
Results
We tested 7 chromosomal loci using 29 microsatellite markers in a
total of 224 individual tumor foci from the 43 cases. Each tumor focus
was tested for loss at a minimum of three chromosomal loci, and cases
with heterogeneity were tested at all seven loci. Allelic loss of at least
one of the chromosomal loci studied was recognized in each lesion,
reaffirming the ability to isolate neoplastic foci free of significant
germ line contamination. As expected, nonrandom allelic loss was
recognized in HI situ as well as invasive cancers at each of the
chromosomal loci examined (Table I ).
Unexpectedly, we found heterogeneity with respect to allelic loss
among different neoplastic foci of in situ cancer within 8 of the 20
tumors with synchronous HIsitu and invasive cancer. Two such cases
are illustrated in Fig. 1. For case B25, loss of the same alÃ-elesat
13ql 2-14 (e.g., marker D13S263), llq. 6q. and 17q was seen in all of
the in situ, invasive, and metastatic foci examined. In contrast, dif
ferent alÃ-elesof 9p2l («•.#.. marker D9SI749) were found to be lost in
distinct foci of in situ cancer of similar histológica! appearance. The
three foci of HI situ cancer that lost the alÃ-elecorresponding to the
lower marker of D9S1749 (IS4, IS5, IS6) also lost an alÃ-eleon
chromosome 17p. whereas the foci that lost the upper allelic marker
of D9SI749 retained 17p at the same loci examined. No heterogeneity
was observed among the four foci of infiltrating cancer and one focus
of metastatic cancer examined. These data are supportive of a model
1493
on June 2, 2015. © 1996 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from