| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Immunology |
Departments of Pathology [G. R., E. M., P. S. M., G. Z., A. P., A. B., D. L., A. S.], Surgery [M. F., P. P.] and Medicine [G. T.], Università di Verona, I-37134 Verona, Italy, and Department of Pathology, Università di Brescia, Brescia I-25100, Italy [G. R., P. G.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Little is known about the molecular anomalies occurring in PETs. Mutations of the MEN-1 gene at chromosome 11q13 have been found in roughly 30% of sporadic PETs (7 , 8) . The four genes frequently altered in common ductal adenocarcinoma, i.e., K-ras, p53, p16, and DPC4 (9) , also have been examined in PETs. The K-ras and p53 genes have no significant role in the pathogenesis of PETs (10 , 11) , whereas alteration of the p16 and DPC4 genes have been reported to be frequent events (12 , 13) despite the fact that allelic losses on chromosomes 9p and 18q are infrequent in these tumors (11 , 14) . However, at least in the case of DPC4, these data are controversial inasmuch as a recent study found no mutations of this gene in either sporadic or MEN1-associated PETs (15) . Although allelic losses at chromosomes 3p and 17p are found in a proportion of sporadic PETs, neither the von Hippel-Lindau nor the p53 genes are the mutational targets of these deletions (11 , 16) .
The identification of commonly deleted chromosomal regions by LOH analysis may aid in the localization of tumor suppressor genes. When this analysis is extended to multiple chromosomal arms, a distinct allelotype is generated. The only allelotype study available on PETs suggested that these tumors are characterized by a low frequency of chromosomal allelic deletions, and that putative tumor suppressor loci implicated in their pathogenesis are located on chromosomes 3q, 11p, 16p, and 22q (14) . This allelotype was based on the analysis of two markers per chromosomal arm in 28 cases and only included 7 NF-PETs (14) .
Here we report a high resolution allelotype for NF-PETs obtained by the analysis of 16 cases, each probed with 394 microsatellite markers, on DNA obtained from frozen tissue. We also performed LOH analysis on chromosomes 1, 3p, 6q, 9p, 11, 17p, and 18q for an additional 16 cases for which only paraffin-embedded material was available. This selection was based on the observations that: (a) chromosomes 6q and 11 showed the highest frequency of allelic loss in our allelotype study; (b) losses on chromosomes 1, 3p, and 17p have been suggested to be of some prognostic value (11 , 16 , 17) ; and (c) chromosomal arms 9p and 18q harbor the p16 and DPC4 genes, which have been implicated in NF-PET tumorigenesis (12 , 13) . The resulting allelotype for NF-PETs showed differences with respect to that of F-PETs and exocrine tumors of the pancreas (14 , 18, 19, 20) and allowed for the identification of two molecular subgroups of tumors that may be distinguished on the basis of the ploidy status. The latter was revealed as an independent prognostic factor for NF-PET outcome.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Personal and family histories were obtained by direct interview of
patients, two of which (NF15 and NF20) had a family history suggestive
of MEN1 syndrome. Diagnosis of PET was established by histopathological
criteria and cell marker analysis and classified according to WHO
criteria (21)
. All tumors were considered nonfunctional
because none gave rise to any clinical symptom related to hormonal
hypersecretion regardless of the presence of immunostaining for any
hormone. All tumors were characterized using a panel of monoclonal
antibodies recognizing panendocrine markers (chromogranin A,
synaptophysin, and nonspecific enolase), gastrointestinal hormones
(insulin, glucagon, somatostatin, pancreatic polypeptide, gastrin,
serotonin, and vasoactive intestinal peptide), and Ki-67 antigen
(22)
. Twelve tumors were considered benign and 20 were
considered malignant in accordance with the respective absence or
presence of invasion of the neighboring organs and/or distant
metastases as evaluated by imaging techniques, surgical, and
pathological examinations. A neoplastic cellularity of at least 90%
was obtained in all cases by either cryostat enrichment or
microdissection (23)
. DNA was prepared as described
(23)
.
|
Analysis of Selected Chromosomal Regions in an Additional 16
NF-PETs.
Sixteen NF-PETs for which only formalin-fixed, paraffin-embedded
material was available (cases NF17–NF32), were analyzed for allelic
loss on chromosomal arms 1p and 1q, 3p, 6q, 9p, 11p and 11q, 17p, and
18q using two microsatellite markers per arm as follows:
D1S2667 (1p36.23–36.13) and D1S2841 (1p33–32.2); D1S2878 (1q22–23.1)
and D1S413 (1q32.1); D3S2338 (3p24.3) and D3S1263 (3p25.1); D6S287
(6q22.1) and D6S441 (6q25.1); D9S171 and D9S161; D11S904 (11p14.3) and
D11S902 (11p15.3); D11S898 (11q22.1) and D11S908 (11q22.3–23.1);
D17S799 (17p12) and D17S1857 (17p12); and D18S474 (18q12.3–21.1) and
D18S1102 (18q12.3).
Ploidy Status Analysis.
The ploidy status was assessed by flow cytometry on cell suspensions
from formalin-fixed, paraffin-embedded sections according to Hedley
et al. (24)
.
Statistical Analysis.
Univariate analysis was performed using the 
2
test or Fisher’s exact test to evaluate categorical variables, whereas
the Mann-Whitney nonparametric U test was used for
continuous variables. The Spearman correlation test was used to
evaluate the association between variables. In particular, we tested
the possible associations between LOH on single chromosomes (1p, 1q,
3p, 6q, 9p, 11p, 11q, 17p, and 18q) and the clinicopathological
features including sex, age, tumor size, local invasion, metastasis,
and the presence of hormone immunoreactivity.
For the survival analysis, the primary statistical outcome in this study was overall survival measured from the date of surgery. Death from cancer was the end point. The only drop-out case was lost at follow-up immediately after hospital discharge. No other case was lost at follow-up, which was updated at August 2000. Overall survival distribution was calculated by the product-limit method and analyzed using the Mantel-Cox test. Multivariate survival analysis was performed using the Cox proportional-hazard model (25) . To select the more parsimonious model, we used a backward elimination procedure including all conventional factors (age, sex, size, local invasion, and metastases), proliferation activity measured as Ki-67 index, and ploidy status. FAL status was not included in the analysis because this would have restricted the analysis only to the subgroup of 16 patients for which this data were available. The final model only included the factors consistently retaining significant P (<0.05). The SPSS release 9.0 (SPSS, Inc., Chicago, IL) statistical program was used.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, where it is evident that the complete loss of one allele is indicative
of the success of our enrichment for neoplastic cellularity. The
allelic status of these 16 NF-PETs at each chromosomal arm is shown in
Fig. 2A
. All cases were informative for at least two markers on
each chromosomal arm. The comprehensive allelotype is shown in Fig. 2B
and the detailed results are shown in Fig. 3
. Frequent LOH was found on chromosome arms 6q and 11q in >60% of
cases, whereas about 50% of cases had losses on 11p, 20q, and 21.
Chromosomes 2p and 10p showed loss in 44%, whereas the remaining
chromosome arms had allelic losses in <40% of cases.
|
|
|
). Of
210 chromosomal arms which showed allelic losses, 106 were complete
losses for an entire chromosomal arm or chromosome. In particular, the
group with a FAL of 0.55 showed a total of 154 chromosomal arms
affected by allelic losses, 100 of which involved the entire arm. In
contrast, a total of 56 chromosomal arms were affected by allelic
losses in the group of nine cases showing a mean FAL of 0.15; only 6 of
these losses involved the entire chromosomal arm. Our high-resolution allelotyping suggests that NF-PETs may be divided into two groups: one showing a high degree of large chromosomal allelic deletions; and the second with a low number of scattered losses with no apparent specific localization.
Analysis of Selected Chromosomal Regions in an Additional 16
NF-PETs.
DNA from 16 paraffin-embedded NF-PETs was analyzed for allelic loss on
chromosomal arms 1p, 1q, 3p, 6q, 9p, 11p, 11q, 17p, and 18q using two
microsatellite markers per arm, selected among those yielding PCR
products <250 bp and therefore adequate to amplify the partially
degraded DNA from paraffin-embedded material. Chromosomes 1p, 1q, 3p,
and 17p were chosen because their allelic status has been reported to
be of prognostic value (11
, 14
, 17)
, whereas chromosomes
6q, 11p, and 11q were chosen because they were the most frequently
affected by losses in our large-scale allelotype. Chromosomal arms 9p
and 18q were included in the study because they harbor the tumor
suppressor genes p16 and DPC4, respectively,
which are reported to be frequently altered in NF-PETs (12
, 13)
. The results are reported in Table 2
. In these additional cases, LOH on chromosomes 6q and 11q was found in
66% and 69% of cases, respectively, confirming the high frequency of
LOH observed in the large-scale allelotype study. Allelic loss on
chromosomes 9p and 18q was relatively infrequent, found in 
30% and
22% of cases, respectively.
|
. Among the 16 cases for which FAL
status was available, 5 of the 7 high-FAL cases (71%) were aneuploid
(NF1, NF3, NF9, NF10, and NF11) and 2 cases were multiploid (NF9 and
NF11), whereas all 9 low-FAL cases were diploid (data not shown). The
ploidy status was significantly associated with FAL status (Fisher’s
exact test; P < 0.005), with a high value of
correlation (Spearman correlation test, 0.76; P < 0.001).
Univariate Analysis.
Twenty-two patients (69%) were living after a median follow-up time of
6.3 years, whereas 10 patients died of disease after a median time of
3.7 years. As expected, the three known prognostic parameters,
i.e., local invasion (P < 0.001),
metastasis (P < 0.001), and Ki-67 index
(P < 0.001) were significantly associated
with outcome (Refs. 22
, 26
; Fig. 4
). Ploidy status was also of prognostic value. In fact, 70% (7 of 10)
of patients who died of their disease had aneuploid tumors, whereas
only 24% (5 of 21) of surviving patients had aneuploid tumors
(P < 0.02). At five years, 50% of patients
with aneuploid tumors were alive compared with 85% of those with
diploid tumors (log-rank test, P < 0.01).
|
Multivariate Survival Analysis.
Multivariate analysis was carried out including all conventional
factors (age, sex, tumor size, presence of hormone immunoreactivity,
local invasion, and metastasis), Ki-67 index, and ploidy status. Only
Ki-67 index and ploidy status emerged as independent prognostic factors
(Table 3)
, and both showed a proportional risk in the product-limit survival
analysis (Fig. 4)
. As shown in the last graph of Fig. 4
, the survival
analysis considering the combination of these two variables showed that
none of the patients with diploid tumors and a Ki-67 index <2% died,
whereas patients with aneuploid tumors and a Ki-67 index >2% had a
poor prognosis. Intermediate results were obtained in the other two
groups of patients, which showed a nonsignificantly different
trend in survival.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
50% of cases. However, even
metastatic diseases may have a surprisingly prolonged clinical course.
A first step toward unraveling the clinical enigma that PETs represent
is a greater understanding of the molecular events underlying their
pathogenesis. The present study allowed for the following conclusions:
(a) the allelotype of a NF-PET is characterized by frequent
allelic losses of 6q and 11q; (b) the NF-PET allelotype
differs from those of either F-PET or pancreatic exocrine tumors,
suggesting a different array of genes involved in its genesis and
progression of malignancy with respect to the latter tumor types;
(c) two subgroups of NF-PETs were identified on the basis of
the degree of chromosomal derangement as measured by the FAL status,
which is statistically correlated with ploidy status; and
(d) ploidy was shown to be an independent prognostic factor
at multivariate survival analysis. The allelotype of NF-PET is characterized by highly frequent allelic losses involving chromosomal arms 6q and 11q; chromosomes 20p and 21 also had relatively frequent allelic loss, whereas the remaining autosomes were affected in <40% of cases. The analysis of selected loci on an additional 16 NF-PETs confirmed the high frequency of LOH on chromosomal arms 6q and 11q. This finding is also supported by a recent study of 44 PETs using comparative genomic hybridization, in which the highest frequency of losses was found on 6q and 11q (39% and 36% of cases, respectively) (27) .
The allelotype of NF-PET differs from that of F-PET, which showed the highest frequency of LOH on chromosomal arms 3q, 11p, 16p, and 22q in a previous allelotype study (14) . In that study, LOH on chromosome 6q was found in only one of 21 F-PETs and in 4 of 7 NF-PETs (14) . Thus, whereas allelic deletions involving chromosome 11 are characteristic of both F- and NF-PETs, the participation of chromosome 6q is a distinguishing feature of NF-PETs.
Deletions involving the long arm of chromosome 6 are frequent chromosomal aberrations in non-Hodgkin’s lymphomas, acute lymphoblastic leukemias, and gastric carcinomas and, interestingly, have been reported to be a frequent event in pancreatic intraductal papillary-mucinous neoplasms (28, 29, 30) . This is therefore indicative of the existence of tumor suppressor loci on 6q that contribute to the pathogenesis of NF-PETs and other malignancies, although a target gene in this chromosomal region has yet to be identified. Genetic losses involving chromosome 11 most often encompassed the entire chromosome, even for those cases showing a low FAL, e.g.,. case 16. Whereas the MEN1 gene is located on this chromosome and is obviously involved in the familial syndrome, a proportion of sporadic NF-PETs also have somatic mutations of this gene (7 , 8) . As the losses observed often involve the entire chromosome, this would suggest that there are additional tumor suppressor genes residing on chromosome 11. One such candidate is the ATM gene located at 11q22–q23 (31) .
As pancreatic acinar carcinomas may express neuroendocrine markers, and mixed exocrine and endocrine pancreatic cancers exist (21) , one interesting question is whether PET shares molecular events with acinar or ductal carcinoma. Ductal carcinomas have a consistent, well-characterized allelotype consisting of highly frequent losses at chromosomes 1p, 9p, 17p, and 18q (18) . Acinar cell carcinomas have a markedly different allelotype, which is characterized by highly frequent LOH on chromosomes 4q and 16q (20) . Thus PETs are separate entities with molecular mechanisms distinct from pancreatic exocrine cancers. Moreover, chromosomal losses at 9p and 18q were not frequent in NF-PETs, confirming earlier reports (14 , 15) . Given the recent reports of high frequencies of alterations in NF-PETs of the p16 and DPC4 genes (12 , 13) , located on these chromosomes, a higher frequency of LOH would be expected if they are indeed involved in PET pathogenesis, as is found in pancreatic ductal cancers (18 , 23) . For DPC4, another report found no mutations in either sporadic or MEN1-associated PETs (15) . Furthermore, we found no alterations in either p16 or DPC4 in 41 PETs, including 30 nonfunctional and 11 functional tumors (32) . Taken together, the data suggest that these two genes are not likely to play a significant role in NF-PET pathogenesis.
LOH on chromosomes 1, 3p, and 17p was analyzed, as these losses have
been reported to be of possible prognostic significance (11
, 16
, 17)
. Whereas the moderate frequencies of allelic loss (
40%)
on these chromosomes in our series of 32 NF-PETs confirm previous
studies (11
, 16
, 17)
, we found no correlation between
these losses and any clinicopathological parameter, including outcome.
The frequency of allelic losses on these chromosomes was similar to
that found on most other chromosomes and may merely reflect a
background of genetic instability.
Our allelotype of 16 primary tumors concluded that there are two
subgroups of NF-PETs. The first shows a large number of allelic
deletions usually involving entire chromosomes, and the second shows a
minute number of indiscriminate losses. This distinction is reflected
in dramatic differences in the FAL, which interestingly also correlated
with the ploidy status: high-FAL tumors were aneuploid or multiploid,
whereas low-FAL tumors were diploid. Although not suggested by previous
authors, the existence of two molecular PET subtypes can also be
inferred from published data. In fact, the majority of functional PETs
can be classified as belonging to the low-FAL subtype inasmuch as only
20% of cases show a high FAL (14)
. In our series of 16
NF-PETs, about half of cases were of either subtype. The existence of
two molecular phenotypes among PETs can furthermore be deduced by
reports on the ploidy pattern and a comparative genomic hybridization
study of these tumors (27
, 33
, 34)
.
The apparent division of NF-PETs into two subgroups on the basis of the dramatic differences in the frequency and type of allelic losses is of particular interest. In the high-FAL subgroup, the large number of losses of entire chromosomes may be indicative of defects involving mitotic segregation (35) . The losses in the low-FAL subgroup were not localized to any particular region and were more or less evenly distributed throughout all 22 autosomes. At present, it remains unclear what precise mechanisms may underlie these two molecular phenotypes. Interestingly, whereas the high FAL/aneuploid phenotype may be analogous to that seen in colorectal cancers showing chromosomal instability, no PETs have been found to show microsatellite instability (Refs. 11 , 14 , and the present report) of the type attributable to the alteration of mismatch repair genes, which cause only subtle sequence instabilities (for review see Ref. 35 ).
Our study also has clinical implications. In fact, evaluation of the malignant potential of PETs is difficult by histological criteria alone, and patients with metastatic diseases may have prolonged survival. We found a strong correlation between the degree of chromosomal derangement, as assessed by the FAL status and the ploidy pattern in 16 frozen tumors (P < 0.005). The ploidy status, assessed in 31 tumors, was selected by multivariate survival analysis as a prognostic indicator with a content of information possibly higher than that furnished by classic clinicopathological parameters associated with tumor aggressiveness and adding valuable clinical information to that given by the well-recognized Ki-67 proliferation index.
|
FOOTNOTES |
|---|
1 This study was supported by grants from the
Associazione Italiana Ricerca Cancro to A. S., Milan, Italy; Consorzio
Studi Universitari di Verona, Italy; cofinanced grant from Verona and
Brescia University and Ministero Università e Ricerca Scientifica
e Tecnologica (Cofin 9806151968 and 9906218982), Rome, Italy;
Fondazione Cassa di Risparmio di Verona, Verona, Italy; Ministero
Sanita’ (Ricerca finalizzata d.lgs.229/99), Rome, Italy; and European
Community Grant BIOMED 2 CE-Contract BMH4-CT98-3805. 
2 To whom requests for reprints should be
addressed, at Dipartimento di Patologia-Sezione Anatomia Patologica,
Università di Verona, Strada Le Grazie, I-37134 Verona, Italy.
Phone: (39) (045) 8074-822; Fax: (39) (045) 8027-136; E-mail
a.scarpa@univr.it.
3 The abbreviations used are: PET, pancreatic
endocrine tumor; F-PET, functional pancreatic endocrine tumor; NF-PET,
nonfunctional pancreatic endocrine tumor; MEN1, multiple endocrine
neoplasia type 1; FAL, fractional allelic loss; LOH, loss of
heterozygosity.
Received 1/11/00. Accepted 11/ 1/00.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
m G., Westin G., Skogseid B. Genetic alterations on 3p, 11q13, and 18q in nonfamilial and MEN 1-associated pancreatic endocrine tumors. Genes Chromosomes Cancer, 26: 258-264, 1999.[Medline]
ppel G., Scarpa A. Allelotype of pancreatic acinar cell carcinoma. Int. J. Cancer, 88: 772-777, 2000.[Medline]
ppel, G., and Sobin, L. Histological typing of endocrine tumors. In: WHO Classification of Endocrine Tumors. Berlin, Heidelberg, New York: Springer-Verlag, 2000.
chain. Virchows Arch. A Pathol. Anat. Histopathol., 419: 463-468, 1991.[Medline]
This article has been cited by other articles:
|
|
 |
|
  R. Bettini, L. Boninsegna, W. Mantovani, P. Capelli, C. Bassi, P. Pederzoli, G. F. Delle Fave, F. Panzuto, A. Scarpa, and M. Falconi Prognostic factors at diagnosis and value of WHO classification in a mono-institutional series of 180 non-functioning pancreatic endocrine tumours Ann. Onc., May 1, 2008; 19(5): 903 - 908. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Piani, G. M Franchi, C. Cappelletti, M. Scavini, L. Albarello, A. Zerbi, P. Giorgio Arcidiacono, E. Bosi, and M. F Manzoni Cytological Ki-67 in pancreatic endocrine tumours: an opportunity for pre-operative grading Endocr. Relat. Cancer, March 1, 2008; 15(1): 175 - 181. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y M H Jonkers, S M H Claessen, A Perren, A M Schmitt, L J Hofland, W de Herder, R R de Krijger, A A J Verhofstad, A R Hermus, J A Kummer, et al. DNA copy number status is a powerful predictor of poor survival in endocrine pancreatic tumor patients Endocr. Relat. Cancer, September 1, 2007; 14(3): 769 - 779. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Vilar, R. Salazar, J. Perez-Garcia, J. Cortes, K. Oberg, and J. Tabernero Chemotherapy and role of the proliferation marker Ki-67 in digestive neuroendocrine tumors Endocr. Relat. Cancer, June 1, 2007; 14(2): 221 - 232. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Nagano, D. H. Kim, L. Zhang, J. A White, J. C Yao, S. R Hamilton, and A. Rashid Allelic alterations in pancreatic endocrine tumors identified by genome-wide single nucleotide polymorphism analysis Endocr. Relat. Cancer, June 1, 2007; 14(2): 483 - 492. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Panzuto, S. Nasoni, M. Falconi, V. D. Corleto, G. Capurso, S. Cassetta, M. Di Fonzo, V. Tornatore, M. Milione, S. Angeletti, et al. Prognostic factors and survival in endocrine tumor patients: comparison between gastrointestinal and pancreatic localization Endocr. Relat. Cancer, December 1, 2005; 12(4): 1083 - 1092. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Tomassetti, D. Campana, L. Piscitelli, R. Casadei, D. Santini, F. Nori, A. M. Morselli-Labate, R. Pezzilli, and R. Corinaldesi Endocrine pancreatic tumors: factors correlated with survival Ann. Onc., November 1, 2005; 16(11): 1806 - 1810. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. A. Kaltsas, G. M. Besser, and A. B. Grossman The Diagnosis and Medical Management of Advanced Neuroendocrine Tumors Endocr. Rev., June 1, 2004; 25(3): 458 - 511. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Iacobuzio-Donahue, M. S. van der Heijden, M. R. Baumgartner, W. J. Troup, J. M. Romm, K. Doheny, E. Pugh, C. J. Yeo, M. G. Goggins, R. H. Hruban, et al. Large-Scale Allelotype of Pancreaticobiliary Carcinoma Provides Quantitative Estimates of Genome-Wide Allelic Loss Cancer Res., February 1, 2004; 64(3): 871 - 875. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Furlan, R. Cerutti, S. Uccella, S. La Rosa, E. Rigoli, A. Genasetti, and C. Capella Different Molecular Profiles Characterize Well-Differentiated Endocrine Tumors and Poorly Differentiated Endocrine Carcinomas of the Gastroenteropancreatic Tract Clin. Cancer Res., February 1, 2004; 10(3): 947 - 957. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Maitra, D. E. Hansel, P. Argani, R. Ashfaq, A. Rahman, A. Naji, S. Deng, J. Geradts, L. Hawthorne, M. G. House, et al. Global Expression Analysis of Well-Differentiated Pancreatic Endocrine Neoplasms Using Oligonucleotide Microarrays Clin. Cancer Res., December 1, 2003; 9(16): 5988 - 5995. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Wild, A. Ramaswamy, P. Langer, I. Celik, V. Fendrich, B. Chaloupka, B. Simon, and D. K. Bartsch Frequent Methylation-Associated Silencing of the Tissue Inhibitor of Metalloproteinase-3 Gene in Pancreatic Endocrine Tumors J. Clin. Endocrinol. Metab., March 1, 2003; 88(3): 1367 - 1373. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-J. Chen, A. Vortmeyer, Z. Zhuang, S. Huang, and R. T. Jensen Loss of Heterozygosity of Chromosome 1q in Gastrinomas: Occurrence and Prognostic Significance Cancer Res., February 15, 2003; 63(4): 817 - 823. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. N. Hochwald, S. Zee, K. C. Conlon, R. Colleoni, O. Louie, M. F. Brennan, and D. S. Klimstra Prognostic Factors in Pancreatic Endocrine Neoplasms: An Analysis of 136 Cases With a Proposal for Low-Grade and Intermediate-Grade Groups J. Clin. Oncol., June 1, 2002; 20(11): 2633 - 2642. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
