Identification of two distinct regions of allelic imbalance on chromosome 18Q in metastatic prostate cancer.
ABSTRACT Like most cancers, prostate cancer (CaP) is believed to be the result of the accumulation of genetic alterations within cells. Previous studies have implicated numerous chromosomal regions with elevated rates of allelic imbalance (AI), using mostly primary CaPs with an unknown disease outcome. These regions of AI are proposed sites for tumor suppressor genes. One of the regions previously implicated as coding for at least one tumor suppressor gene is the long arm of chromosome 18 (18q). To confirm this observation, as well as to narrow the critical region for this putative tumor suppressor, we analyzed 32 metastatic CaP specimens for AI on chromosome 18q. Thirty-one of these 32 specimens (96.8%) exhibited AI at one or more loci on chromosome 18q. Our analysis using 17 polymorphic markers revealed statistically significant AI on chromosome 18q at 3 markers, D18S35, D18S64 and D18S461. Using these markers as a guide, we have been able to identify 2 distinct minimum regions of AI on 18q. The first region is between the genetic markers D18S1119 and D18S64. The second region lies more distal on the long arm of the chromosome and is between the genetic markers D18S848 and D18S58. To determine if 18q loss is a late event in the progression of CaP, we also examined prostatic intraepithelial neoplasia (PIN) and primary prostate tumors from 17 patients for AI with a subset of 18q markers. We found significantly higher AI in the metastatic samples. Our results are consistent with 18q losses occurring late in CaP progression.
- SourceAvailable from: fk.uwks.ac.id[Show abstract] [Hide abstract]
ABSTRACT: The process of cancer and prostate cancer metastasis is complex and requires fundamental changes to the behaviour of the parent cell. While the stage at which essential mutations for prostate cancer metastasis occur remains controversial, it is likely, based on current evidence, that an accumulation of genetic damage is required. However, the study of cancer metastasis is clearly dependent on the availability of suitable in vitro and in vivo models. Not every model represents the full in vivo situation in man, but a combination of these models is now becoming available in prostate cancer and should allow a more detailed assessment of the specific genes involved in metastasis and the preferential adhesion in bone. Identification of specific genes associated with particular pathology has also taken tremendous steps forward in the last few years. Differential expression analysis, of both the RNA and also protein levels are providing new targets for therapy, specifically directed against metastatic disease. However, for longer term prospects the ability to detect metastasis in a simple blood sample would offer the most hope of permanent treatment or indeed cure. Based on serum profiling, such methods should soon be available to the oncologist in the clinic. On-line catalogues of genes whose expression is perturbed in metastatic processes offer the first clues to the key events in this complex biological process. It is perhaps from these catalogues improved animal models and indeed the more global analysis of patient samples from bio-banks that the key events and a genetic basis will be identified.05/2007: pages 21-61;
- [Show abstract] [Hide abstract]
ABSTRACT: Cytogenetic, molecular cytogenetic, and molecular studies of prostate cancer have produced a large volume of data about chromosomal loci that are aberrant in prostate cancer. The cumulative data on prostate cancer reveal allelic losses on chromosome arms 2q, 3p, 5q, 6q, 7q, 8p, 9p, 10p, 10q, 11p, 11q, 12p, 13q, 16q, 17p, 17q, 18q, and 21q, but there is a great deal of variability between studies. In most cases, the frequency of allelic loss is higher in metastatic tissues or hormone-refractory tumors than in primary tumors. There also seem to be discrepancies in the genetic findings depending on methods employed. Molecular genetic studies, using polymerase chain reaction (PCR) analysis of microsatellite markers, demonstrated allelic loss at 7q31.1, whereas fluorescence in situ hybridization analysis showed a gain at the same region. Com-mon sites of allelic loss that are consistently observed by various methods seem to exist on chromosome arms 8p, 10q, 13q, and 16q. PTEN/MMAC1 has been identified on 10q23.3 and was found to be frequently mutated in advanced prostate cancer. Other regions are also considered to harbor genes associated with the development and progression of prostate cancer, and these could be included in the diagnostic methods for the substaging of prostate cancer.International Journal of Clinical Oncology 11/2000; 5(6):345-354. · 1.73 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Zusammenfassung Zahlreiche Erkenntnisse in der Prostatakarzinomforschung basieren auf Zellkulturergebnissen. Neuere genomische Untersuchungen zeigen jedoch, dass verwendete permanente Prostatakarzinomzelllinien sich deutlich von dem klinisch relevanten, primären Prostatakarzinom unterscheiden und damit die Übertragbarkeit dieser Zellkulturergebnisse auf die Klinik nur bedingt gegeben ist. Das Arbeiten mit Primärzellkulturen aus Gewebestückchen von Prostatektomiepräparaten bietet eine sehr gute Alternative, doch die Etablierung von Primärkulturen gestaltet sich schwierig und recht aufwändig. In dieser Arbeit wurde ein Primärzellkulturmodell mit einem Invasionssystem kombiniert. Hiermit gelang es nicht nur invasiv wachsende Zellpopulationen aus Primärkulturen zu selektionieren, sondern auch diese Zellen in einem 3D-Modell unter der Ausbildung von Sphäroiden weiter zu kultivieren. Zur Charakterisierung dieser Zellpopulation haben wir vergleichende genomische Hybridisierungen durchgeführt, die zahlreiche genetische Alterationen aufzeigen. Das hier dargestellte Modell ermöglicht es erstmalig, invasive Zellklone aus primärem Prostatakarzinomgewebe zu gewinnen und durch Kultivierung für weitere Untersuchungen zu verwenden.Der Urologe 09/2008; 47(9):1199-1204. · 0.46 Impact Factor
IDENTIFICATION OF TWO DISTINCT REGIONS OF ALLELIC IMBALANCE ON
CHROMOSOME 18Q IN METASTATIC PROSTATE CANCER
Susan S. PADALECKI1, Dean A. TROYER2, Marc F. HANSEN3, Tomo SARIC2, Barbara G. SCHNEIDER4, Peter O’CONNELL2and
Robin J. LEACH1*
1Department of Cellular and Structural Biology, University of Texas Health Science Center at San Antonio, San Antonio, TX, USA
2Department of Pathology, University of Texas Health Science Center at San Antonio, San Antonio, TX, USA
3Fels Institute for Cancer Research and Molecular Biology, Temple University School of Medicine, Philadelphia, PA, USA
4Department of Pathology, Louisiana State University Medical Center and Stanley Scott Cancer Center, New Orleans, LA, USA
Like most cancers, prostate cancer (CaP) is believed to be
the result of the accumulation of genetic alterations within
cells. Previous studies have implicated numerous chromo-
somal regions with elevated rates of allelic imbalance (AI),
using mostly primary CaPs with an unknown disease out-
come. These regions of AI are proposed sites for tumor
suppressor genes. One of the regions previously implicated
as coding for at least one tumor suppressor gene is the long
arm of chromosome 18 (18q). To confirm this observation,
as well as to narrow the critical region for this putative tumor
suppressor, we analyzed 32 metastatic CaP specimens for AI
on chromosome 18q. Thirty-one of these 32 specimens
(96.8%) exhibited AI at one or more loci on chromosome
18q. Our analysis using 17 polymorphic markers revealed
statistically significant AI on chromosome 18q at 3 markers,
D18S35, D18S64 and D18S461. Using these markers as a
guide, we have been able to identify 2 distinct minimum
regions of AI on 18q. The first region is between the genetic
markers D18S1119 and D18S64. The second region lies more
distal on the long arm of the chromosome and is between the
genetic markers D18S848 and D18S58. To determine if 18q
loss is a late event in the progression of CaP, we also exam-
ined prostatic intraepithelial neoplasia (PIN) and primary
prostate tumors from 17 patients for AI with a subset of 18q
markers. We found significantly higher AI in the metastatic
samples. Our results are consistent with 18q losses occurring
late in CaP progression. Int. J. Cancer 85:654–658, 2000.
© 2000 Wiley-Liss, Inc.
Prostate cancer (CaP) is the most common form of non-cutane-
ous malignancy in American men with an expected 179,300 new
cases in 1999 (http://www3.cancer.org/cancerinfo). It is estimated
that more than 37,200 American men will die as a result of CaP
this year (http: www3.cancer.org/cancerinfo). With the advent of
prostate-specific antigen (PSA) testing, more cancers are being
identified at earlier stages. However, there is currently no way to
distinguish life-threatening tumors needing definitive intervention
from those that would benefit from a “watch and wait” approach.
It is believed that like most cancers, the accumulation of genetic
lesions leads to tumor development and progression. One focus of
CaP research is to identify allelic losses as an indication of genetic
alterations in cancer-related genes such as oncogenes, tumor sup-
pressor, metastasis suppressor and DNA mismatch repair genes.
Molecular studies of CaP have identified multiple non-random
genetic alterations in tumor specimens. To date, the molecular
events leading to initiation and progression of this disease remain
unclear, in part because it is difficult to obtain appropriate samples
Karyotypic analyses of prostate tumors have been hindered by
the difficulty in obtaining good chromosome preparations from
tumor tissue. However, cytogenetic studies have reported losses of
the Y chromosome and deletions on chromosomes 6q, 7q, 8p, 9p,
10q, 13q, 16q, 17 and 18q, implicating these regions as sites of
putative tumor suppressor genes (Brown et al., 1994; Visakorpi et
al., 1995). In addition, a number of allelotyping analyses of human
CaPs have been performed in order to identify the probable loca-
tions of tumor suppressor genes. Although there has been some
variability in the number of markers used and the number of
tumors examined, the most frequent allelic losses in CaP involve
chromosome arms 8p, 10 (both p and q), 16q and 18q (Kunimi et
al., 1991; Gao et al., 1993; Brewster et al., 1994; Latil et al., 1994;
Crundwell et al., 1996; Cunningham et al., 1996; Ueda et al.,
1997; Jenkins et al., 1998; Saric et al., 1999).
A previous analysis of CaP specimens for loss of heterozygosity
(LOH) on chromosome 18 by Brewster et al. (1994) is interesting
because it revealed 2 non-overlapping regions of loss: one at
18q21.3 near the putative tumor suppressor gene Deleted in Colon
Cancer (DCC) and a second region at 18q22 near the genetic
marker D18S5. A PCR-based genome-wide study (Cunningham et
al., 1996) of primary CaP also implicated 2 regions of high allelic
imbalance (AI) on 18q at different markers than those reported by
Brewster et al. (1994). The first region, at marker D18S851, was
near the putative tumor suppressor gene Deleted in Pancreatic
Cancer 4 (DPC4) (Cunningham et al., 1996). The second more
distal region is flanked by markers D18S64 and D18S58. Neither
of these studies, however, has presented statistically significant
evidence for 2 loci on 18q. The results of this study implicate the
presence of at least 2 distinct putative tumor suppressor genes on
18q, both of which appear to map distal to the previously identified
tumor suppressor genes, DCC and DPC4.
Besides demonstrating that regions of chromosome 18q are lost
in CaP, some studies have also given an indication that loss of
chromosome 18q may be a late event in CaP progression (Brewster
et al., 1994; Crundwell et al., 1996; Ueda et al., 1997; Jenkins et
al., 1998; Saric et al., 1999). To further explore whether 18q loss
occurs late in the progression of CaP, we have examined 32
metastatic lesions taken from 26 patients upon autopsy (Saric et
al., 1999). Although earlier lesions from these patients were not
available, we have attempted to demonstrate the stage specificity
of this AI by comparing rates of AI at select markers on chromo-
some 18q in metastatic lesions with rates of AI seen in patients
with either prostatic intraepithelial neoplasia (PIN) or primary
CaP. Our findings support the hypothesis that loss of 18q loci is a
late event in CaP progression and that regions implicated here may
be involved in the process of metastasis.
MATERIAL AND METHODS
Subjects and tumor samples
Tissues used for these studies were formalin-fixed, paraffin-
embedded tissues processed by routine methods. Histological eval-
uation was performed using hematoxylin and eosin (H&E)-stained
sections. Primary CaPs with adjacent normal prostate tissue and
foci of high-grade PIN were obtained from 17 patients undergoing
radical prostatectomies at the Audie Murphy Veterans Adminis-
tration Hospital (San Antonio, Texas, USA) between 1989 and
1991 (Saric et al., 1999). Patients ranged in age from 54 to 72
*Correspondence to: Department of Cellular and Structural Biology, The
University of Texas Health Science Center at San Antonio, 7703 Floyd
Curl Drive, San Antonio, TX 78284-7703, USA. Fax: (210) 567-6781.
Received 4 August 1999; Revised 27 September 1999
Int. J. Cancer: 85, 654–658 (2000)
© 2000 Wiley-Liss, Inc.
Publication of the International Union Against Cancer
years (median 64 years). Histological diagnosis and tumor grade
were determined after surgery on H&E-stained 5-?m sections. All
primary CaPs were graded using Gleason’s grading system. Sev-
enteen primary prostate adenocarcinomas were of acinar type, 3
(18%) of them were well differentiated (Gleason’s scores 3–4), 13
(77%) were moderately differentiated (Gleason’s scores 5–7) and
1 (6%) was a poorly differentiated tumor (Gleason’s score 8).
Twenty-two foci of primary CaP (1–3 per case, median 1) and 49
foci of high-grade PIN (1–9 per case, median 3) have been ana-
lyzed in this study.
Metastatic tumors and matched normal tissue samples were
obtained post mortem from 26 autopsies performed at the Audie
Murphy and Kerrville Veterans Administration Hospitals (Kerr-
ville, Texas, USA) between 1979 and 1991 (age 55–89 years,
median 71 years). In 6 autopsies, multiple metastatic samples were
collected. Metastatic tissues were obtained from a variety of tis-
sues: lymph nodes (8; 25%), peritoneum (6; 18.75%), liver (6;
18.75%), bone (4; 12.5%), adrenal gland (4; 12.5%), lung (2;
6.25%) and periprostatic tissue (1; 3.13%). The site of one meta-
static lesion was unknown.
Three serial histological sections (5 ?M) of each block were
prepared using PCR precautions to avoid contamination of sam-
ples by changing water in the water bath and wiping the microtome
blade with bleach and alcohol between blocks. The first section
was stained with H&E and areas for microdissection were marked.
The H&E-stained slide was then overlaid with the unstained slides
to mark areas for microdissection. Areas selected for microdissec-
tion contained at least 75% of nuclei from tumor cells. A small
portion of razor blade was used for microdissection and precau-
tions were taken to prevent cross-contamination, including chang-
ing the blades between samples. PIN lesions and primary CaP
tumors were dissected as previously published (Saric et al., 1999).
In brief, we required that PIN lesions be separated from other
lesions by 1 cm, or that they be in separate blocks to avoid
sampling cancerous lobules or ducts as PIN.
Sections were deparaffinized using 100% N-Octane (Sigma, St.
Louis, MO), rinsed in 100% ethanol and followed by centrifuga-
tion. Cell pellets were allowed to air dry before resuspension in
50–100 ?l of cell lysis solution containing 1 mg/ml Proteinase K,
50 mM Tris-HCl buffer, pH 8.0, 1 mM EDTA and 0.45% Tween-
20. Cell lysis suspensions were incubated at 55°C overnight.
Proteinase K was inactivated by incubation at 95°C for 15 min.
DNA concentrations could not be determined because of the small
volume of DNA isolated. However, samples were diluted 1:15 in
sterile ddH2O for microsatellite marker analysis. Stock prepara-
tions and dilutions were stored at ?20°C and 4°C, respectively.
Microsatellite marker analysis
Metastatic samples were examined by PCR at 17 polymorphic
microsatellite markers specific for the long arm of chromosome 18.
These markers were spaced at approximately 5–10 cM intervals
along the long arm of chromosome 18. In each case, one primer of
each pair was end-labeled with32P-?-ATP using T4 polynucle-
otide kinase (Promega, Madison, WI). Five microliters of template
DNA were amplified in 20-?l PCR reactions. The mixture in-
cluded 100 nM of each primer, 200 mM of each dinucleotide
trisphosphate, 1 mM spermidine, MgCl2concentrations ranging
from 1.5 to 3.0 mM and 1 U Taq Gold (Perkin Elmer, Foster
Reactions were heated to 95°C for 16 min to denature the
template DNA and release the enzyme, then cycled at 95°C for 1
min, 52–60°C for 1 min and 72°C for 1 min for a total of 35
cycles, followed by final extension at 72°C for 7 min. DNA from
PIN lesions and primary tumors was amplified as previously
described (Saric et al., 1999). PCR products were mixed with a
denaturing dye and incubated at 95°C for 5 min. Five microliters
of each reaction were then separated by electrophoresis on 7%
acrylamide/7M Urea sequencing gels for 3–5 hr at 65 W. Dried
gels were exposed to Kodak X-OMAT AR film for 24 hr. The
relative intensity of each band was determined by quantitation of
dried gels with PhosphorImager using ImageQuant software (Mo-
lecular Dynamics, Sunnyvale, CA). An imbalance or AI ratio was
calculated as the ratio of allele intensities in the tumor relative to
the ratio of allele intensities in the accompanying normal tissue.
The cutoff value used for establishing AI was 1.5. This method for
determining AI has been validated previously by MacGrogan et al.
The AI at each locus was assessed for associations with AI at all
other loci. Two by two tables listing AI and no AI were produced
for each pair of loci. The expected number of AI events was then
calculated. A chi-square value was calculated and used if the
expected number of AI events was greater than or equal to 5 in all
4 cells (common criterion for the adequacy of the chi-square test).
If any expected value was less than 5, the exact significance level
was calculated using the Fisher’s exact test. This is done by firstly
sorting all possible outcomes of the sample space, defined as the
set of all possible number of AI events in the 2 groups. Next, the
points whose likelihood ratios for the null hypothesis of equal
proportions were greatest (i.e., those outcomes least supporting the
null hypothesis) were sorted so these outcomes were at the top.
Then, with the assumption that the probability of an AI event in
both groups was the same, the probability of the points in the
sorted sample could be summed from the top of the list to the data
point. This sum was the significance level of the observed outcome
assuming the common probability of AI events.
To identify loci on chromosome 18q that may be involved in the
progression of CaP, we examined the DNA of 32 formalin-fixed,
paraffin-embedded CaP metastases and paired normal tissues ob-
tained post mortem. These samples were analyzed for AI at 17
genetic markers which map to the long arm of chromosome 18.
These markers were chosen based on high heterozygosity of the
markers and small product size. We found that markers with
products larger than 200 bp did not work well with our metastatic
samples because many samples could not be amplified across this
large a region. Presumably, this is due to varying levels of DNA
degradation as a result of the initial processing of the samples at
autopsy. The predicted marker order with cytogenetic locations is
provided in Table I. The percent AI seen with each marker is
illustrated in Figure 1.
TABLE I – MICROSATTELITE MARKERS AND THEIR
CHROMOSOME 18Q LOSS IN METASTATIC CAP
Samples that were heterozygous for a given marker were con-
sidered informative at that locus. Thirty-one of 32 specimens
(96.8%) exhibited AI at one or more loci on the long arm of
chromosome 18. This high level of AI supports the existence of
one or more tumor suppressor genes on 18q and indicates that our
microdissection techniques were effective in the enrichment of
tumor cells. In some samples, AI appeared to span the entire long
arm of chromosome 18 whereas other tumors exhibited more
isolated loss. Representative examples of AI are shown in Figure 2.
Statistical analyses of these AI data indicated that there is
significant AI on 18q at polymorphic loci D18S35, D18S64 and
D18S461, compared with background levels seen with the markers
D18S19 and DCC. Two of these loci, D18S35 and D18S64,
mapped between DCC and D18S19. The third locus, D18S461,
mapped distal to D18S19. This implies that 2 distinct regions of AI
lie near these markers. These regions are separated by a region of
very low AI, indicating that they may be completely independent
of one another. Based on these findings, we further evaluated the
data to identify 2 minimal regions of AI on 18q. Because not every
marker is informative in every sample, the minimum region of AI
may not necessarily fall near the markers where the highest AI was
observed. In identifying these minimal regions of AI, metastatic
tumors from different sites within the same patient were not treated
independently but were counted as a single patient sample.
The proximal minimal region of AI lies between the polymor-
phic loci D18S1119 and D18S64. Twenty-two of 26 independent
tumors exhibited AI proximal to the genetic marker, D18S19. Of
these, 10 tumor samples appear to have undergone mitotic recom-
bination with breakpoints at or near these loci (Fig. 3a). In total, 18
of 22 tumors exhibited AI that included this minimal region. The
proximal border of this region is well defined with 4 tumors (9, 12,
22, 26) that retain heterozygosity at D18S1119 and exhibit AI at
FIGURE 1 – Percent AI at each microsatellite marker.
FIGURE 2 – AI at different microsatellite markers analyzing paired
normal (N) and metastatic (M) CaP samples. Examples from various
cases at microsatellites markers: a, D18S35; b, D18S862, c, D18S70;
FIGURE 3 – a) Patterns of AI in 10 metastatic CaP specimens defin-
ing the minimum region of AI between genetic markers, D18S1119
and D18S64. b) Patterns of AI in 11 metastatic CaP specimens
defining the minimum region of AI between genetic markers,
D18S848 and D18S58. White circles, informative, both alleles re-
tained; black circles, informative and detected AI; white boxes, unin-
formative; arrows indicate that our data indicates that AI in a particular
tumor could extend in that direction.
PADALECKI ET AL.
D18S64, and in some cases more distal markers. The distal border
for this region of AI is less well defined, only 2 samples (28, 31)
were heterozygous at D18S64, and exhibited AI at D18S1119.
However, there were 3 other cases (1, 13, 24) with no loss at
D18S862 and loss at D18S1119. Two of these cases (1, 24) were
uninformative at D18S64. Thus, the minimal region of AI appears
to be proximal to D18S64. The estimated size of this minimal
region of AI at 18q21 is approximately 7 cM according to mapping
data from the Whitehead Institute Center for Genome Research
The distal minimal region of AI in these samples is between the
genetic markers D18S848 and D18S58 (Fig. 3b). Figure 3b illus-
trates the genotypes of 11 metastatic tumors with apparent break-
points at or near these loci. Twenty-one of the 26 independent
metastatic tumors showed AI distal to D18S19. Eighteen of these
21 metastatic tumors exhibited AI between D18S848 and D18S58,
which is proximal to the statistically significant AI observed at
marker D18S461. This region is relatively well defined on both its
proximal and distal borders. On the proximal end, 4 independent
tumors (9, 10, 15, 17) retain both alleles at D18S848 but exhibit AI
at more distal markers on the chromosome. This would support the
idea that the minimal region of AI should be distal to D18S848. At
the distal border, D18S58, four tumors (13, 14, 25, 31) retained
heterozygosity at D18S58, yet showed AI at D18S848. According
to mapping data from the WICGR, the estimated size of the distal
minimal region of AI is 6 cM.
These 2 regions appear to be independent of one another as
evidenced by AI of only one or the other region with retention of
heterozygosity of the second region (data not shown). Samples 2,
12, 18, 22 and 23 exhibit AI in the more proximal region but
remain heterozygous at the more distal regions. In contrast, spec-
imens 4, 20, 21 and 32 retain heterozygosity in the more proximal
region while undergoing AI at the more distal region.
To investigate whether AI on 18q was a metastasis-associated
event, we also examined PIN and primary CaP tumors for AI using
markers D18S35, D18S64, D18S19 and D18S461. With the ex-
ception of D18S19, these are the markers exhibiting statistically
significant AI in the metastatic tumor samples. D18S19 maps
between the 2 regions of interest and is the marker with the lowest
level of AI seen in these samples. These results are summarized in
Figure 4. At genetic markers D18S35, D18S64 and D18S461,
there was a statistically significant increase in the rate of AI in
metastases compared with PIN and primary tumors. However, at
D18S19, no difference was detected between rates of AI in early
lesions vs. metastases. Both PIN and primary tumors exhibited low
levels of 18q loss.
There is currently no way to distinguish between CaP tumors
that are likely to metastasize and are a significant health risk and
those that warrant a more conservative treatment approach because
they are less likely to progress. Our understanding of CaP pro-
gression remains limited, in part, because appropriate tissue sam-
ples have been unavailable for analysis. This is in part due to the
way CaP is treated. Patients with more advanced disease are rarely
candidates for prostatectomy. Therefore, samples of these ad-
vanced tumors are seldom available for analysis. As a result, most
genetic studies of CaP have been limited to examining primary
tumors of unknown outcome that were localized to the prostate.
Studies of CaP progression have been limited by the availability of
later stage lesions, such as metastatic CaP tissues. Here, we report
on our results examining lesions on chromosome 18q in metastatic
CaP specimens, a rare resource. We examined 32 metastases from
26 patients for AI on the long arm of chromosome 18, using
polymorphic markers that were spaced approximately 5–10 cM
Initially, most studies of chromosome 18q in CaP were re-
stricted by the use of a few RFLP markers. Rates of LOH or AI on
18q in CaP in these studies ranged from 10% to 45% (Gao et al.,
1993; Brewster et al., 1994; Latil et al., 1994; Crundwell et al.,
1996; Cunningham et al., 1996; Ueda et al., 1997; MacGrogan et
al., 1997; Jenkins et al., 1998). In our study, AI at at least one
marker on 18q was found in 31 of 32 specimens (96.8%), a much
higher rate of allelic loss than seen in previous studies. The reasons
for this are 2-fold. Firstly, because work by others had indicated
that loss of 18q loci might be a late event in CaP progression, we
chose to focus on metastatic lesions. Most other studies have
looked primarily at CaPs that were not as advanced as those in our
subset or tumors from patients of unknown outcome. Secondly, we
examined this set of metastatic tumors at more 18q loci than most
of the other studies, which used a limited number of markers
concentrating on or near the DCC gene.
The focus of our work has been a detailed AI analysis of 18q in
metastatic tumors. None of the markers we tested exhibited AI in
100% of the specimens; however, 2 distinct minimal regions of AI
were identified. Overall, these data are consistent with previous
findings indicating the existence of 2 distinct regions of elevated
AI on 18q and provide strong evidence for this scenario in CaP
(Brewster et al., 1994; Cunningham et al., 1996).
There are 2 putative tumor suppressor genes that map to 18q,
DCC and DPC4. DCC was originally identified based on frequent
loss of one allele in colorectal cancer (Fearon et al., 1990). One
function of DCC has recently been elucidated in experiments
involving knock-out animal models (Kolodziej et al., 1996). It is
involved in the migration of developing axons (Kolodziej et al.,
1996). A second tumor suppressor gene, DPC4, is proximal to
DCC at 18q21.1. DPC4 is homozygously deleted in 30% of
pancreatic cancers and is frequently mutated in pancreatic tumors
(Hahn et al., 1996). Few mutations of DPC4 have been identified
in cancers that are not of gastrointestinal origin (Schutte et al.,
1996). Data from other investigators support the exclusion of
DPC4 as a candidate tumor suppressor gene in CaP. To date, no
mutations of the DPC4 gene have been detected in CaP specimens
of any stage (MacGrogan et al., 1997; Ueda et al., 1997). The DCC
gene has also been examined for mutations. To date, few, if any,
have been detected in CaP specimens (Gao et al., 1993).
In our own study, statistically significant AI was not seen with
a marker intragenic for DCC nor with markers that flank the DPC4
gene, indicating that they do not appear to play a role in CaP.
However, we cannot rule out that a tumor suppressor gene may
exist in this region that plays a role in CaP as it has been implicated
by another group (Ueda et al., 1997). Our data imply that a tumor
FIGURE 4 – Comparison of AI at 4 loci on chromosome 18q in PIN,
primary CaP and metastatic CaP. AI in metastatic lesions was signif-
icantly higher than in early lesions, PIN and primary CaP, respectively,
at D18S35 (p ? 0.0071), D18S64 (p ? 0.001; p ? 0.004), and
D18S461 (p ? 0.0001; p ? 0.007).
CHROMOSOME 18Q LOSS IN METASTATIC CAP
suppressor gene lies distal to DPC4 and DCC in the interval
between D18S1119 and D18S64, where 18 of 26 independent
tumors exhibited AI. To date, there are no putative tumor suppres-
sor genes mapped to this interval. Our region of loss is consistent
with data presented by others identifying a commonly deleted
region between the centromere to D18S19 (Latil et al., 1994).
The second minimal region of AI implicated here maps more
distal on 18q at 18q22-q23 between the genetic markers D18S848
and D18S58. At present, no previously isolated tumor suppressor
gene maps to this interval. Our data indicating AI in this region are
also consistent with the findings of others (Brewster et al., 1994;
Cunningham et al., 1996). Brewster et al. (1994) demonstrated that
the most commonly deleted region on 18q in CaP was at 18q22-
q23, distal to DCC. They reported one tumor with isolated loss at
18q23 and retention of heterozygosity at DCC and more proximal
loci, indicating that these loci may be independent of one another.
Other groups have presented data confirming loss of 18q loci in
this area in CaP specimens (Kunimi et al., 1991; Latil et al., 1994;
Cunningham et al., 1996), although no common regions of dele-
tion were identified. Furthermore, our data agree with studies by
others of 18q loss in other cancers. A tumor suppressor gene
mapped to this region has the potential to play a role in a number
of different tumor types. AI or LOH at markers in this region has
also been implicated in the progression of head and neck cancers
(Pearlstein et al., 1998), cohesive gastric cancer (Inoue et al.,
1998) and breast cancer (Huang et al., 1995).
To determine if 18q loss is a late event in CaP progression, we
examined 4 of these markers, D18S35, D18S64, D18S19 and
D18S461, in samples from PIN and primary prostatic tumors. Our
results support the previous reports that loss of 18q loci is a late
event (Brewster et al., 1994; Crundwell et al., 1996; MacGrogan et
al., 1997; Ueda et al., 1997; Jenkins et al., 1998; Saric et al.,
1999). This also appears to be the case for a number of other
cancers such as gastric cancer (Inoue et al., 1998), head and neck
cancer (Pearlstein et al., 1998), breast cancer (Huang et al., 1995)
and ovarian cancer (Chenevix-Trench et al., 1992). This could
prove to be extremely useful information as it may be utilized as
a biomarker for the progression of this deadly disease.
In summary, our findings refine the regions on the long arm of
chromosome 18 that may harbor tumor suppressor genes. They
support the notion that there may be 2 tumor suppressor genes on
the long arm of chromosome 18 that play a role in the progression
of CaP. However, these genes remain to be identified. Continued
precise mapping of the deletions of 18q in CaP is needed to narrow
the regions and to ultimately isolate these genes. Our results also
support the notion that the loss of these loci may result in an
increase in metastatic potential, because we have shown that there
are significant differences between the rates of AI on 18q in early
vs. late CaP specimens.
We thank Dr. C. Leach for editing this manuscript and Ms. B.
Reus for providing technical assistance. RJL and POC received
grants from the San Antonio Cancer Institute. SSP received a grant
from the National Institute of Aging (AG00165).
BREWSTER, S.F., BROWNE, S. and BROWN, K.D., Somatic allelic loss at the
DCC, APC, nm23-H1, and p53 tumor suppressor gene loci in human
prostatic carcinoma. J. Urol., 151, 1073–1077 (1994).
BROWN, J.A., ALCARAZ, A., TAKAHASHI, S., PERSONS, D.L., LIEBER, M.M.
and JENKINS, R.B., Chromosomal aneusomies detected by fluorescent in
situ hybridization analysis in clinically localized prostate carcinoma.
J. Urol., 152, 1157–1162 (1994).
CHENEVIX-TRENCH, G., LEARY, J., KERR, J., MICHEL, J., KEFFORD, R.,
HURST, T., PARSONS, P.G., FRIEDLANDER, M. and KHOO, S.K., Frequent loss
of heterozygosity on chromosome 18 in ovarian adenocarcinoma which
does not always include the DCC locus. Oncogene, 7, 1059–1065 (1992).
CRUNDWELL, M.C., CHUGHTAI, S., KNOWLES, M., TAKLE, L., LUSCOMBE,
M., NEOPTOLEMOS, J.P., MORTON, D.G. and PHILLIPS, S.M., Allelic loss on
chromosomes 8p, 22q and 18q (DCC) in human prostate cancer. Int. J.
Cancer, 69, 295–300 (1996).
CUNNINGHAM, J.M., SHAN, A., WICK, M.J., MCDONNELL, S.K., SCHAID,
D.J., TESTER, D.J., QIAN, J., TAKAHASHI, S.N., JENKINS, R.B, BOSTWICK,
D.G. and THIBODEAU, S.N., Allelic imbalance and microsatellite instability
in prostatic adenocarcinoma. Cancer Res., 56, 4475–4482 (1996).
FEARON, E.R., CHO, K.R., NIGRO, J.M., KERN, S.E., SIMONS, J.W., RUPPERT,
J.M., HAMILTON, S.R., PREISINGER, A.C., THOMAS, G., KINSLER, K.W. and
VOGELSTEIN, B., Identification of a chromosome 18q gene that is altered in
colorectal cancers. Science, 247, 49–56 (1990).
GAO, X., HONN, K.V., GRIGNON, D., SAKR, W. and CHEN, Y.Q., Frequent
loss of expression and loss of heterozygosity of the putative tumor sup-
pressor gene DCC in prostate carcinomas. Cancer Res., 53, 2723–2727
HAHN, S.A., SCHUTTE, M., HOGUE, A.T.M., MOSKALUK, C.A, DA COSTA,
L.T., ROZENBLUM, E., WEINSTEIN, C.L., FISCHER, A., YEO, C.J., HRUBEN,
R.H. and KERN, S.E., DPC4, a candidate tumor suppressor gene at human
chromosome 18q21.1. Science, 271, 350–353 (1996).
HUANG, T.H., YEH, P.L., MARTIN, M.B., STRAUB, R.E., GILLIAM, T.C.,
CALDWELL, C.W. and SKIBBA, J.L., Genetic alterations of microsatellites on
chromosome 18 in human breast carcinoma. Diagn. mol. Pathol., 4, 66–72
INOUE, T., UCHINO, S., SHIRAISHI, N., ADACHI, Y. and KITANO, S., Loss of
heterozygosity on chromosome 18q in cohesive-type gastric cancer is
associated with tumor progression and poor prognosis. Clin. Cancer Res.,
4, 973–977 (1998).
JENKINS, R. TAKAHASHI, S., DE LACEY, K., BERGSTRALH, E. and LEIBER, M.,
Prognostic significance of allelic imbalance of chromosome arms 7q, 8p,
16q and 18q in stage T3N0M0 prostate cancer. Genes Chromosomes
Cancer, 21, 131–143 (1998).
KOLODZIEJ, P.A., TIMPE, L.C., MITCHELL, K.J., FRIED, S.R., GOODMAN,
C.S., JAN, L.Y. and JAN, Y.N., frazzled encodes a Drosophila member of
the DCC immunoglobulin subfamily and is required for CNS and motor
axon guidance. Cell, 87, 197–204 (1996).
KUNIMI, K., BERGERHEIM, U.S.R., LARSSON, I..-L., EKMAN, P. and COLLINS,
V.P., Allelotyping of human prostatic adenocarcinoma. Genomics, 11,
LATIL, A., BARON, J.C., CUSSENOT, P., FOURNIER, G., SOUSSI, T., BOCCON-
GIBOD, L., LE DUC, A., ROU¨ ESS´ E, J. and LIDEREAU, R., Genetic alterations
in localized prostate cancer: identification of a common region of deletion
on chromosome arm 18q. Genes Chromosomes Cancer, 11, 119–125
MACGROGAN, D., LEVY, A., BOSTWICK, D., WAGNER, M., WELLS, D. and
BOOKSTEIN, R., Loss of chromosome arm 8p loci in prostate cancer:
mapping by quantitative allelic imbalance. Genes Chromosomes Cancer,
10, 151–159 (1994).
MACGROGAN, D., PEGRAM, M., SLAMON, D. and BOOKSTEIN, R., Compar-
ative mutational analysis of DPC4 (Smad 4) in prostatic and colorectal
carcinomas. Oncogene, 15, 1111–1114 (1997).
PEARLSTEIN, R.P., BENNINGER, M.S., CAREY, T.E., ZARBO, R.J., TORRES,
F.X., RYBICKI, B.A., and VAN DYKE, D.L., Loss of 18q predicts poor
survival of patients with squamous cell carcinoma of the head and neck.
Genes Chromosomes Cancer, 21, 333–339 (1998).
SARIC, T., BRKANAC, Z., TROYER, D.A., PADALECKI, S.S., SAROSDY, M.,
WILLIAMS, K., ABADESCO, L., LEACH, R.J. and O’CONNELL, P., Genetic
pattern of prostate cancer progression. Int. J. Cancer, 81, 219–224 (1999).
SCHUTTE, M. and 14 OTHERS, DPC4 in various tumor types. Cancer Res.,
56, 2527–2530 (1996).
UEDA, T., KOMIYA, A., EMI, M., SUZUKI, H., SHIRAISHI, T., YATANI, R.,
MASAI, M., YASUDA, K. and ITO, H., Allelic losses on 18q21 are associated
with progression and metastasis in human prostate cancer. Genes Chro-
mosomes Cancer, 20, 140–147 (1997).
VISAKORPI, T., KALLIONIEMI, A.H., SYV¨ ANEN, A.C., HYYTINEN, E.R.,
KARHU, R., TAMMELA, T., ISOLA, J.J. and KALLIONIEMI, O.-P., Genetic
changes in primary and recurrent prostate cancer by comparative genomic
hybridization. Cancer Res., 55, 342–347 (1995).
PADALECKI ET AL.