| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Oncology, Markers, Clinical Correlates |
Departments of Radiology [Y. H., K. H., S. S., Y. T.] and Mathematics [S. A.], Kansai Medical University, Moriguchi City, Osaka 570-8506, and Department of Biology, Nara Medical University, Nara 634 [T. O], Japan
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The involvement of the HPV3 in the development of cervical cancer has been firmly established. Because HPV infection does not always lead to cervical cancer, other genetic alterations must also play a role in tumor development. A LOH, which points to a role for TSGs, oncogene amplification, and point mutations, are all thought to be involved, but there is as yet no complete picture of the relative roles for each of these genetic changes in patients with cervical carcinomas. To play a role in tumorigenesis, both copies of a TSG must be inactivated. The loss of one allele in a chromosome region may point to the presence of a TSG in that region.
Several studies have shown that LOH at specific chromosomal sites is frequently associated with the recurrence of various cancers, e.g., 13q14.3 in oral carcinoma (3) , 10q in human lung cancer (4) , and 11p15 in breast cancer (5) . Although cytogenetic studies of cervical cancer are relatively few, they have revealed frequent, nonrandom chromosomal changes (6) . Studies of LOH in patients with cervical carcinoma have also reported a high frequency of allelic deletions affecting 3p (7, 8, 9) , 5p (10) , 17p (11, 12, 13, 14) , and 18q (15) . A LOH on chromosome 6p has also been reported in patients with cervical carcinoma (12 , 13 , 16 , 17) . However, the importance of LOH on chromosome 6p in the recurrence of cervical cancer after radiotherapy remains unknown.
In this study, we evaluated the incidence of LOH on 6p21.2 in the DNA of patients with cervical cancer and assessed the impact on the recurrence after radiotherapy.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The follow-up for the surviving
patients ranged from 1 to 55.8 months, with a mean of 24.6 months. The
mean patient age was 66 years (range, 27–90 years). The primary tumors
ranged in diameter from 1 to 11.9 cm (mean, 5.7 cm), as measured with
MRI. The typical examination included a medical history, a physical
examination, routine blood counts, a blood chemistry profile, a chest
radiograph, an i.v. urogram, colonoscopy, and MRI. Each patient’s
tumor was staged jointly by the radiation oncology and gynecological
oncology staff according to the classification of the International
Federation of Gynecology and Obstetrics (18)
, with
modifications. The tumors consisted of 55 squamous cell carcinomas and
7 adenocarcinomas. No patient received chemotherapy prior to the
radiotherapy. Each of the patients signed a consent form approved by
the Kansai Medical University Review Board describing the experimental
nature of the treatment.
|
Clinical Response.
The response of the tumor to the treatment was defined as follows: CR
when no tumor was detected by physical examination or MRI and
cytological or biopsy studies were negative for malignant cells for at
least 1 month after treatment; PR when the tumor mass was reduced by
50%; and NC when the reduction in the tumor mass was <50%.
Detection and Typing of HPV.
Tumor samples were obtained from the 62 patients by punch biopsy
prior to radiotherapy. Samples were taken from two to four different
parts of each tumor and frozen immediately at -80°C. Genomic DNA was
extracted from each tumor according to standard protocols
(20)
.
Tumor DNA was amplified by PCR with primers specific for HPV types 16, 18, 33, and 58 E6, as described previously (21) . PCR was carried out for 40 cycles at 95°C for 1.5 min, 48°C for 1.5 min, and 70°C for 2 min, using a BioGene PHC-1 system (Techne, Cambridge, United Kingdom).
Investigation of p53 Status.
Mutations of the p53 gene were identified by a SSCP analysis
and DNA sequencing of tumor samples (22)
with primers
flanking the evolutionarily conserved regions of the gene from exon 5
to exon 8. Each exon was amplified separately using sense and antisense
oligonucleotide primers flanking the exon as shown: exon 5, sense
5'-TTCCTCTTCCTGCAGTACTC-3' and antisense
5'-GCCCC-AGCTGCTCACCATCG-3'; exon 6, sense
5'-CACTGA-TTGCTCTTAGGTCTG-3' and antisense
5'-AGTTGCAAA-CCAGACCTCAG-3'; exon 7, sense
5'-CCAAGGCGCACT-bpGGCCTC-3' and antisense 5'-GCGGCAAGCAGAGGCTGG-3';
and exon 8, sense 5'-CCTATCCTGAGTAGTGGTAA-TC-3' and antisense
5'-GTCCTGCTTGCTTACCTCGC-3'. All mutations were confirmed by sequencing
or an SSCP analysis of a second independent PCR reaction. The products
of this reaction were subcloned into a TA vector (TA Cloning kit;
Invitrogen). A mixture of at least 62 subclones was used as the
template for DNA sequencing using a T7 sequencing kit (Boehringer
Mannheim), and the results were visualized by exposure to Kodak XAR
film with an intensifying screen at -80°C.
Investigation of LOH on 6p21.2.
Regions of the genomic DNA extracted from the blood and tumor of each
patient were amplified. PCR was used to detect the presence of LOH on
chromosome 6p with the three microsatellite markers D6S276,
D6S1624, and D6S1583 (6p21.2). The PCR of each region
was cycled 35 times at 94°C for 1 min, 55°C for 1 min, and 72°C
for 1 min using each fluorescent set as described previously
(23)
. The PCR products were loaded on a 6% polyacrylamide
gel and were analyzed with automatic sequencing (ALFred; Pharmacia,
Uppsala, Sweden). Allelic losses were scored as decreases in the
intensity of one allele relative to the other, as determined from a
comparison of tumor and normal DNA. The shift was indicated by either
an addition or a deletion of one or more repeat units, resulting in the
generation of novel microsatellite alleles. The analysis in patients
was repeated at least twice, and the results were highly reproducible.
Statistical Methods.
Survival was measured as the time (days) from the start of
radiotherapy. The relationship between the presence or absence of
genetic alterations on chromosome 6p21.2 and the radioresponse was
analyzed with Fisher’s exact test. The tumor size data were analyzed
with a Wilcoxon rank sum test. Actuarial survival was estimated by the
Kaplan-Meier (24)
method, and differences in survival were
analyzed with the log-rank test (25)
. The statistical
analyses were performed with Stata 4.0 Software (Stata Statistical
Software, Release 4.0; Stata Corp., College Station, TX).
P < 0.05 was considered significant.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Relationships among Stage, Tumor Size, Clinical Response, and
Outcome.
There was a correlation between the stage of disease and the patients’
response. Twelve of 13 patients with stages I and II showed CR. In
contrast, 21 of the 49 patients with advanced stage cancer
(i.e., stage III or IV) showed PR or NC (P =
0.02, Fisher’s exact test). The tumor sizes were 3.9 ± 1.8 cm in
the stage I-II patients and 6.1 ± 2.2 cm in stage III-IV
patients. Statistical analysis using the Wilcoxon rank sum test
revealed significant differences in tumor sizes between stage I-II
versus stage III-IV (P = 0.002). The tumor
size increased with the stage of disease. The tumor sizes of patients
with CR was 4.8 ± 1.9 cm; it increased to 7.2 ± 2.2 cm in
patients with PR or NC. The difference in size was significant between
the CR and PR/NC groups (P = 0.0002, Wilcoxon rank sum
test). The tumor size of the patients without recurrence was 4.9 ± 2 cm; it increased to 6.7 ± 2.3 cm in patients with
recurrence. The difference in size was significant between the presence
and absence of recurrence (P = 0.003, Wilcoxon rank sum
test; Table 2
). The mean tumor diameter
of all patients was 5.7 ± 2.3 cm. We divided the patients into
two groups, those with tumors <5.7 cm in diameter (n =
35) and those with tumors 5.7 cm in diameter (n =
27). There was a significant difference in overall survival
(P = 0.0004, log-rank test) and disease-free survival
(P = 0.01, log-rank test) between these groups; the
former group survived significantly longer than the latter group.
|
). There was also no
significant difference in overall survival (log-rank test) and
disease-free survival (log-rank test) between HPV-positive and
HPV-negative patients.
Relationships between p53 Status and Stage, Tumor
Size, Clinical Response, and Outcome.
p53 mutations were detected in seven patients (11.3%) with
SSCP analysis, in five patients with stage III, and two with IV. We
observed p53 mutations in several codons: in codon 245,
changing GGC to AGC and causing the substitution of serine for glycine;
in codon 146, TGG to TAG, resulting in the substitution of a stop codon
for tryptophane; in codon 189, GCC to GTC, causing the substitution of
valine for alanine; in codon 286, GAA to CAA, resulting in the
substitution of glutamine for glutamic acid; in codon 285, GAG to AAG,
resulting in the substitution of lysine for glutamic acid; in codon
213, CGA to TGA, resulting in the substitution of a stop codon for
arginine; in codon 146, TGG to TAG, resulting in the substitution of a
stop codon for tryptophane; and in codon 193, CAT to CGT, resulting in
the substitution of histidine for arginine.
The tumor size was 7.7 ± 3.2 cm in the mutant p53
tumors versus 5.4 ± 2.1 cm in the wild-type
p53 tumors; there was a correlation between tumor size and
p53 status (P = 0.049, Wilcoxon rank sum
test). Of the patients with mutant p53, four had CR and
three had NC. Of the patients with wild-type p53, 36 had CR,
10 had PR, and 9 had NC. Thus, there was no significant relationship
between any group and the p53 status (Fisher’s exact test).
Of the patients with mutant p53, three had recurrence and
four had no recurrence. Of the patients with wild-type p53,
24 had recurrence and 31 had no recurrence. Thus, there was no
correlation between recurrence and the patient’s p53 status
(Fisher’s exact test; Table 2
). There was also no significant
difference in overall survival (log-rank test) and disease-free
survival (log-rank test) between patients with wild-type p53
and those with mutant p53.
Relationships between LOH on Chromosome 6p21.2 and Stage, Tumor
Size, Clinical Response, and Outcome.
Chromosome arm 6p21.2 was involved in LOH in 46.8% (29 of 62) of the
informative carcinomas. The tumor size was 5.9 ± 2.1 cm in
patients with LOH versus 5.5 ± 2.5 cm in the 33
patients without LOH; there was no correlation between tumor size and
LOH status (Wilcoxon rank sum test). LOH on 6p21.2 was seen in 1
patient with stage I disease, 7 stage II, 14 stage III, and 7 stage IV.
There was no correlation between the patient’s LOH status and the
tumor stage (stage I-II versus stage III-IV; Fisher’s exact
test).
Of the patients with LOH, 15 had CR, 6 had PR, and 8 had NC. Of the
patients without LOH, 25 had CR, 4 had PR, and 4 had NC. The difference
in LOH status was significant between the CR and PR/NC groups
(P = 0.04; Fisher’s exact test). Of the patients with
LOH, 18 had recurrence and 11 had no recurrence. Of the patients
without LOH, 9 had recurrence and 24 had no recurrence. The difference
was significant between the presence/absence of recurrence and the LOH
status (P = 0.006, Fisher’s exact test; Table 2
). In
addition, there was a significant difference in overall survival
(P = 0.02, log-rank test; Fig. 1A
) and disease-free survival
between patients with LOH and those without (P = 0.002,
log-rank test; Fig. 1B
). The latter group survived
significantly longer than the former.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In recent studies of cytogenetics and allelotypes, many observations of allelic losses at specific chromosomal loci in a variety of human cancers have implicated the presence of putative TSGs on some chromosomes. In particular, deletions and rearrangements of the short arm of chromosome 6 (6p) are known to be frequent in cancers of the colon (28) , lung (29) , ovary (30) , and kidney (31) . Published allelotype analyses of cervical cancer identify LOH on chromosome 6p in 28% (12) , 34% (16) , 43% (13) , and 47% (15) of specimens. However, the importance of LOH on chromosome 6p in the recurrence of cervical cancer after radiotherapy remains unknown. In this study, we investigated the correlation between recurrence after radiotherapy and the rate of LOH on chromosome 6p21.2 using three microsatellite markers (D6S276, D6S1624, and D6S1583) in both tumor tissue and blood of 62 patients with cervical carcinoma.
In this study, LOH on 6p21.2 occurred in 46.8% (29 of 62) of the patients. Patients with LOH had significantly poorer responses to radiotherapy (P = 0.04) and shorter disease-free survival (P = 0.002) compared with those without LOH. LOH on 6p21.2, where WAF1 is situated, has been described in many cancers (28, 29, 30, 31) , suggesting that WAF1 may be inactivated by a two-hit process in the corresponding tumors.
The WAF1 gene is a cyclin-dependent kinase inhibitor that acts primarily as a negative regulator of cell proliferation at the G1 cell cycle checkpoint, i.e., to allow the replicating cell time to repair damaged DNA (32 , 33) . The inability of the cell to properly observe this interval increases the risk of incorporating DNA damage and could lead to the accumulation of gene mutations with resultant genomic instability and altered regulation of cellular proliferation (34 , 35) . Alterations in the expression or function of the WAF1 gene can be expected to lead to aberrant control of cell proliferation and may predispose certain individuals to cancer. Tumor-associated mutation of the coding region of the WAF1 gene is rare, although it has been reported in a small number of patients with Burkitt’s lymphoma (36) , prostate adenocarcinoma (37) , and breast cancer (38) . We also detected the somatic mutation of the WAF1 gene in 9.1% (2 of 22) patients with cervical cancer.4 In addition, Shiohara et al. (39) have reported that the absence of WAF1 alterations in a large series of 14 human malignancies, including cervical cancers, suggested that WAF1 mutations may not play an important role in either the onset or the progression of these malignancies. Therefore, we hypothesized that other genes in this region might also be involved.
The smallest region of deletion is between 6p21 and 6p23. Kersemaekers et al. (17) observed a high percentage (54%) of LOH on chromosome 6p with primers at 6p22–23, which was also in the smallest region of overlap between D6S105 and tumor necrosis factor. The HLA region is in this smallest area of deletion. HLA molecules are required for the immunological response to HPV infection, and tumors may thus evade the immune defense by a loss of HLA expression. The loss of HLA expression is observed in many tumor types, including cervical cancer (40) . A genetic defect may explain this loss of expression, in view of the frequent LOH found in this region. Enlund et al. (41) reported that locus D6S276 was associated with the HLA region. Although we have used the same microsatellite marker, we hypothesized that the HLA gene might be partially associated with the recurrence of advanced cervical cancers. Further mapping with more markers is needed to elucidate this issue.
The importance of the p53 gene in clinical oncology has been
reviewed (42)
. A poor response of human malignancies, such
as breast cancer (43)
and colon cancer (44)
,
to different therapies is often associated with mutations of the
p53 gene. In contrast to many other tumors in humans,
p53 mutations are only rarely detected in cervical cancer
(42)
. In our study, 7 of the 62 patients (11.3%) were
found to have mutations in the p53 gene, as evaluated by an
SSCP analysis of genomic DNA. The seven tumors with mutant
p53 had G:C
A:T mutations. These data confirm the report
of a high frequency of G:CA:T mutations in patients with cervical
carcinoma (45)
.
Most cervical carcinomas have been shown to contain HPV DNA sequences, including the high-risk HPV 16 and HPV 18 types (46) . The binding of high-risk HPV type E6 viral protein to the p53 protein has been shown to result in a rapid ubiquitin-dependent proteolytic degradation of p53 (47) . Therefore, the presence of high-risk forms of HPV is believed to result in a loss of the p53-mediated control of cell growth (48) . In general, the presence of high-risk HPV in tumor cells is thought to be associated with a poor response of cervical cancer to treatment. It has been shown, however, that HPV-positive cancer cells may preserve the p53 protein in its functional form (48) . This observation may partially explain the discrepancies in the results of studies examining the correlation between the presence of various types of HPV DNA in tumors and the treatment outcome in patients with cervical cancer (49 , 50) . In the case of cervical carcinoma, there is an obvious need to study the effects of HPV infection on intrinsic tumor cell radiosensitivity (51) . However, the literature detailing the differences in prognosis between patients with HPV-positive and HPV-negative tumors is incomplete (52 , 53) . In our study, HPV-positive tumors were found in 41.9% of the patients (26 of 62). This low rate is consistent with the report by Matsumoto et al. (54) , who showed that HPV was detected in 49% of cervical cancers obtained from Japanese women as compared with 80% in Western countries (50) . The expression of the E6 or E7 gene of high-risk HPVs (types 16, 18, and others) seems to be an essential but not a sufficient factor for the malignant conversion of the cervical epithelium in Japanese patients (55) . No correlation was found between HPV infection and treatment outcome in our study.
In this study, tumor size was found to be the important determinant of the response to radiotherapy and a poor overall survival (P = 0.0004, log-rank test) and disease-free survival (P = 0.01) after radiotherapy, which confirms data from previous reports (2) . The difference in size was significant for the presence or absence of recurrence (P = 0.003). However, LOH on 6p21.2 was found to be the most significant determinant of the relapse-free survival time (P = 0.002). In addition, we found no correlation between LOH status and the size or stage of tumor. Thus, the results of this study suggest that LOH on 6p21.2 is an important predictor of recurrence after radiation treatment in patients with cervical carcinomas independent of tumor size. The higher rate of LOH on 6p21.2 in tumors may explain, at least in part, the poor prognosis of patients with cervical carcinoma after radiotherapy.
More recently, investigators have been looking for a way to improve local control by combining radiation therapy with chemotherapy (56 , 57) . The Gynecological Oncology Group (56) has explored the role of radiation therapy and concurrent chemotherapy with hydroxyurea, cisplatin, and 5- fluorouracil. In their study, the rate of local recurrences was significantly lower with cisplatin-based regimen than with hydroxyurea regimen, whereas the rate of distant recurrences was only reduced slightly. These results suggested that the principal effect of cisplatin is radiosensitization. In our series, the patients with local recurrence might be influenced by chemotherapy.
In this study, we evaluated the recurrence of cervical cancer, including both local recurrence and distant metastases after radiotherapy. In the future, we hope to investigate the correlation between LOH status and treatment failure according to local recurrence and distant metastasis.
The analysis of key gene-related LOH on 6p21.2 in patients with cervical carcinoma might help clarify the mechanisms of recurrence after radiotherapy in some individuals.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 This study was supported by a Grant-in-Aid for
Scientific Research from the Ministry of Education, Science, Sports and
Culture, Japan. 
2 To whom requests for reprints should be
addressed, at Department of Radiology, Kansai Medical University, 10-15
Fumizono-cho, Moriguchi City, Osaka 570-8507, Japan. Phone:
81-6-6992-1001, extension 3345; Fax: 81-6-6993–3865; E-mail: harima{at}takii.kmu.ac.jp
3 The abbreviations used are: HPV, human
papillomavirus; LOH, loss of heterozygosity; TSG, tumor suppressor
gene; MRI, magnetic resonance imaging; CR, complete response; PR,
partial response; NC, no change; SSCP, single-strand conformation
polymorphism.
Received 10/ 6/99; revised 12/13/99; accepted 12/16/99.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
nen S., Syrjnen K. Implications of the p53 tumor-suppressor gene in clinical oncology. J. Clin. Oncol., 13: 1009-1022, 1995.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  A. Chatterjee, H. A. Pulido, S. Koul, N. Beleño, A. Perilla, H. Posso, M. Manusukhani, and V. V. V. S. Murty Mapping the Sites of Putative Tumor Suppressor Genes at 6p25 and 6p21.3 in Cervical Carcinoma: Occurrence of Allelic Deletions in Precancerous Lesions Cancer Res., March 1, 2001; 61(5): 2119 - 2123. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
