|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2001 American Society for Clinical Oncology Human Papillomavirus and Prognosis of Invasive Cervical Cancer: A Population-Based StudyFrom the Programs in Biostatistics and Epidemiology, Division of Public Health Sciences, and Program in Cancer Biology, Division of Human Biology, Fred Hutchinson Cancer Research Center; and Departments of Biostatistics and Epidemiology, School of Public Health and Community Medicine, and Departments of Microbiology and Pathology, School of Medicine, University of Washington, Seattle, Washington. Address reprint requests to Stephen M. Schwartz, PhD, Fred Hutchinson Cancer Research Center, 1100 Fairview Ave N (MP-381), PO Box 19024, Seattle, WA 98109-1024.
PURPOSE: To determine the association between human papillomavirus (HPV) type and prognosis of patients with invasive cervical carcinoma. PATIENTS AND METHODS: Patients diagnosed with International Federation of Gynecology and Obstetrics (FIGO) stage IB to IV cervical cancer between 1986 and 1997 while residents of three Washington State counties were included (n = 399). HPV typing was performed on paraffin-embedded tumor tissue using polymerase chain reaction methods. Patients were observed for a median of 50.8 months. Total mortality (TM) and cervical cancer–specific mortality (CCSM) were determined. Hazards ratios (HR) adjusted for age, stage, and histologic type were estimated using multivariable models. RESULTS: Eighty-six patients had HPV 18–related tumors and 210 patients had HPV 16–related tumors. Cumulative TM among patients with HPV 18–related tumors and among patients with HPV 16–related tumors were 33.7% and 27.6%, respectively; cumulative CCSM in these two groups were 26.7% and 18.1%, respectively. Compared with patients with HPV 16–related cancers, patients with HPV 18–related cancers were at increased risk for TM (HRTM, 2.2; 95% confidence interval [CI], 1.3 to 3.6) and CCSM (HRCCSM, 2.5; 95% CI, 1.4 to 4.4). The HPV18 associations were strongest for patients with FIGO stage IB or IIA disease (HRTM, 3.1; 95% CI, 2.3 to 4.2; and HRCCSM, 5.8; 95% CI, 3.9 to 8.7), whereas no associations were observed among patients with FIGO stage IIB to IV disease. Virtually identical associations were found in the subset of patients with squamous cell carcinoma (n = 219). CONCLUSION: HPV 18–related cervical carcinomas, particularly those diagnosed at an early stage, are associated with a poor prognosis. Elucidating the mechanism or mechanisms underlying this association could lead to new treatment approaches for patients with invasive cervical carcinoma.
APPROXIMATELY 12,800 women were diagnosed with invasive cervical cancer in the United States during 2000, and more than 4,500 died from this disease.1 Stage-specific survival of cervical cancer patients has not improved since the 1960s. As a result, the 5-year survival rate for the average cervical cancer patient in the United States is only 67%.2 The average cervical cancer patient who dies loses approximately 25 years of life, more years than are lost in all other cancers occurring in women, with the exception of Hodgkin’s disease.2 Identification of molecular tumor markers that are predictive of prognosis may lead to improved treatment of cancer patients. Infection with oncogenic types of human papillomavirus (HPV) is the initiating event of all cervical carcinoma,3 and HPV DNA can be detected in nearly all invasive lesions.4 Of the oncogenic HPV types, HPV 16 and HPV 18 are found with the highest frequency in invasive cervical carcinomas in most populations.5 Accumulating research suggests that HPV 18 infection is associated with a more aggressive form of cervical neoplasia than is HPV 16 infection,6-8 raising the possibility that the prognosis of invasive cervical cancer could be related to the HPV type found in the tumor. Although some investigators have reported that HPV 18–related tumors carry a poorer prognosis,9-17 other researchers have not noted such a difference.18-23 One difficulty in interpreting these studies is that many did not use survival analysis techniques, or used such methods but did not account for known cervical cancer prognostic factors or included only small numbers of patients (either overall or with HPV 18–related tumors). In addition, none of these studies was population-based. We report on the prognosis of cervical cancer in relation to HPV type in a large, population-based series of patients from western Washington State.
Study Population This investigation was conducted as part of an ongoing population-based epidemiologic study of HPV and invasive cervical cancer, for which the methods have been previously described.24 Briefly, patients in whom invasive cervical carcinoma was diagnosed between January 1, 1986, and June 30, 1998, and who resided in King, Pierce, or Snohomish County, Washington, were eligible for inclusion in the epidemiologic study. These patients were identified from the files of the Cancer Surveillance System (CSS) of western Washington, a population-based cancer registry that is part of the Surveillance, Epidemiology, and End-Results program of the National Cancer Institute.25 As part of the epidemiologic study, we attempted to obtain archived tumor tissue blocks from biopsy or surgery, to determine the presence and type of HPV DNA. In the present investigation, eligible cervical cancer patients were a subset of those eligible for the epidemiologic study. This subset of women had had International Federation of Gynecologic Oncology (FIGO) stage IB to IV disease diagnosed between January 1, 1986, and December 31, 1997. Of the 776 patients meeting these eligibility criteria, 577 (74.3%) were recruited into the epidemiologic study or died before recruitment; the remaining women were not recruited into the original epidemiologic case-control study, at the request of their physicians or because of refusal to participate. We obtained paraffin-embedded tumor specimens from diagnosing laboratories for 531 (92.0%) of the 577 patients, and we completed HPV detection assays of specimens from 425 patients (80.0%). The tumor samples from 26 patients yielded DNA inadequate for viral typing (see the next section). Therefore, analyzable HPV data existed for 399 patients. The CSS files provided data on patient characteristics (age and race), tumor characteristics (histologic type, FIGO stage, tumor differentiation, nodal status, and tumor size), the first course of cancer-directed therapy (surgery [including specific procedures], radiotherapy, and chemotherapy), and vital status (alive or dead; if dead, the underlying cause of death, coded according to the International Classification of Diseases, 9th Revision.26) The CSS determines the vital status of all patients through annual inquiries of follow-up physicians and computerized linkages to Washington State computerized death certificate and motor vehicle administration records. As of December 31, 1997, the vital status was known for 88.9% of the 776 eligible women.
Specimen Preparation and HPV Nucleic Acid Detection For 293 recent patients eligible for our study (247 of whom had HPV results available), one of us (P.L.P.) made and reviewed slides (hematoxylin and eosin–stained specimens) from the retrieved tissue blocks to provide a standardized classification of histologic type against which we could judge the histology coding in the CSS database. The standardized review confirmed the CSS classification of squamous cell carcinoma, pure adenocarcinoma, and adenosquamous carcinoma in 99.5%, 95.6%, and 89.5% of these cases, respectively. Because we were unable to conduct the standardized pathology review for all of the cases with HPV DNA data, and because the sample of cases for which the standardized review was completed indicated that the CSS classification is highly accurate, we relied on the CSS classification for our statistical analyses.
Statistical Analysis Our analysis focused on the relationship between HPV type (18 v 16) and prognosis. In most of these analyses, we grouped patients whose tumors contained both HPV 18 and HPV 16 DNA with patients having only HPV 18 DNA, but we conducted subanalyses in which patients with both HPV types were excluded. Because some studies have suggested that the inability to detect HPV DNA (any type) might be associated with a poor prognosis, we also conducted analyses in which patients with HPV DNA–negative tumors were compared with patients with HPV DNA–positive tumors, as well as with patients with HPV 16–related tumors. To examine unadjusted associations between characteristics (HPV type, detection or nondetection of HPV DNA) and prognosis, we computed and plotted Kaplan-Meier estimates of overall and cervical cancer–specific survival. Unadjusted differences between groups were assessed statistically by computing the log-rank test and corresponding P values. To estimate the association between HPV type and total mortality (TM) and cervical cancer–specific mortality (CCSM) while controlling for other characteristics, we used proportional hazards models to calculate adjusted hazards ratio (HR) estimates and associated 95% confidence intervals (CIs) from the standard errors of the model coefficients and the normal approximation. The covariates included in these models were FIGO stage, age at diagnosis, and histologic type. For histologic type, we grouped the small number of cases of adenosquamous carcinoma (n = 17) with pure adenocarcinoma cases (n = 87), and the small number of cases of other or unspecified epithelial carcinomas (n = 20) with squamous cell carcinoma cases (n = 275) to minimize the number of terms in the model, particularly for subgroup analyses. In a limited number of models in which we explored the impact of aggregating histologic type in this way, we did not find any differences in point estimates compared with the more detailed coding, but in general, the CIs were narrower when the grouped histologic-type data were used. We examined the influence of individual observations on HRs by computing delta beta statistics. To determine whether the proportional hazards assumption was reasonable, in models comparing HPV 18- and HPV 16–related cancers we tested for a nonzero slope of scaled Schoenfeld residuals.33 These analyses did not provide statistical evidence of departures from the proportional hazards assumption or of unusually influential observations. We computed HRs separately for patients with early-stage disease (FIGO stage IB or IIA) and for patients with more advanced disease (FIGO stage IIB and higher), because it might be expected that any association between HPV DNA and prognosis would be more easily identified in patients with the more favorable prognosis.12 We computed the stage-specific HR estimates within the context of a single model that included covariates plus indicator terms for stage, HPV type, and the product of these factors. Statistical differences in HRs between patients with early-stage disease and patients with late-stage disease were calculated using likelihood ratio tests. In addition, with regard to patients with early-stage disease, we performed subanalyses limited to women who were treated with radical hysterectomy and pelvic lymphadenectomy, because earlier studies of HPV type and prognosis focused on this group of women.14 Subanalyses were also conducted in which we computed HRs separately for patients with squamous cell carcinomas and patients with carcinomas containing glandular elements (pure adenocarcinomas and adenosquamous carcinomas). Data on tumor differentiation, nodal status, and tumor size were missing from the CSS records for approximately 21.8%, 50.1%, and 55.3% of patients with HPV data, respectively. Therefore, with regard to all women with HPV data, we were unable to analyze the impact of adjusting for these characteristics when examining relationships between HPV type and prognosis. Instead, for tumor differentiation we examined potential confounding in Cox regression models restricted to the subset of patients with both HPV and tumor differentiation data. For tumor size and nodal status, we did not have sufficient numbers of women with data on these characteristics and on HPV for similar subanalyses. Because confounding could occur only to the extent that these known prognostic factors are related to HPV type, we examined the association between tumor size, nodal status, and HPV type (18 v 16) to assess whether relationships between HPV type and prognosis were likely to be independent of these characteristics.
The median length of follow-up was 50.8 months (interquartile range, 16.6 to 86.5 months; maximum, 144 months) for patients with HPV DNA data available and 41.3 months (interquartile range, 14.1 to 78.0 months; maximum, 144 months) for patients without HPV DNA data available. Among patients with HPV DNA results available, there were 128 deaths, of which 90 were due to cervical cancer; the corresponding numbers for patients without HPV DNA results available were 123 and 76. The most common noncervical cancer causes of death in the cohort were other malignancies (n = 38), coronary heart disease and related circulatory conditions (n = 10), and nonmalignant respiratory system conditions (n = 8); this distribution was similar for patients with and patients without HPV DNA results available. Additional characteristics of the cervical cancer patients with and those without HPV DNA results are listed in Table 1. The distributions of age, race, histologic type, and FIGO stage were similar in the two groups, as were the associations between prognosis and either stage or histology. There was no difference in age-, stage-, or histology-adjusted mortality between patients with and patients without HPV data available, either for all causes (HRTM, 1.1; 95% CI, 0.4 to 1.4) or for deaths due to cervical cancer (HRCCSM, 1.2; 95% CI, 0.9 to 1.7).
We detected HPV DNA in the tumors of 85.7% of the patients in the study ( Table 2). The mean interval between diagnosis and testing was similar for HPV-positive (66 months) and HPV-negative (61 months) tumors. HPV 16 DNA, either alone or in combination with other HPV types, was detected in 68.2% of HPV-positive tumors and was more common in squamous cell carcinomas than adenocarcinomas. The great majority of the tumors in which HPV DNA was detected (282 of 342 [82.5%]) contained only one HPV type (HPV 16, followed by HPV 18, in the majority of these cases). The most common multiple infection involved both HPV 16 and HPV 18.
Kaplan-Meier survival curves for TM and CCSM in relation to HPV type are shown in Fig 1. HPV 18–related tumors were associated with nonsignificantly poorer overall survival (Fig 1A: log-rank P = .191) and cervical cancer–specific survival (Fig 1B: log-rank P = .090). The poorer overall survival and cervical cancer–specific survival among patients with HPV 18–related tumors were statistically significant for patients with early-stage disease (Fig 1C: log-rank P = .025; Fig 1D: log-rank P = .001), whereas there was little difference in the unadjusted survival by HPV type for patients with more advanced disease (Fig 1E: log-rank P = .796; Fig 1F: log-rank P = .541). Patients with HPV 18–related tumors had a similar FIGO stage distribution (IB, II, III, IV), as did patients with HPV 16–related tumors (  23 = 2.44, P = .486).
In proportional hazards regression, adjusting for age, stage, and histologic type, the risk of dying was increased two-fold for patients whose tumors contained HPV 18 DNA compared with patients with tumors containing HPV 16 DNA ( Table 3). The results were similar when we restricted the analysis to women with squamous cell carcinomas. There were only 12 deaths overall (five cervical cancer deaths) among the patients with either pure adenocarcinoma or adenosquamous carcinoma, but within this group there was a suggestion of an increased mortality associated with HPV 18 DNA (HRTM, 3.6; 95% CI, 0.8 to 16.5). Among women with FIGO stage IB or IIA disease, there was a three-fold association between HPV 18 detection and death, whereas among patients with later-stage disease there was no association (Table 3). The difference between the HRTM for the patients with early-stage disease and that for patients with late-stage disease was statistically significant (P = .037). The stage-specific results were essentially unchanged when we restricted the analyses to squamous cell carcinomas (FIGO stage IB or IIA: HRTM, 3.5; 95% CI, 1.3 to 9.1; FIGO stage IIB to IV: HRTM, 1.4; 95% CI, 0.6 to 2.9). Among the patients with early-stage disease who were coded by the CSS as having undergone radical hysterectomy and pelvic lymphadenectomy (n = 144), the HRTM was 5.4 (95% CI, 1.8 to 17.0) and the HRCCSM was 10.5 (95% CI, 2.7 to 40.2).
All of the adjusted associations were similar or stronger when we analyzed death due to cervical cancer as the outcome. The difference in the HRCCSM for HPV 18 DNA by stage was statistically significant (P = .002). Further, HPV 18 DNA was not associated with death due to causes other than cervical cancer (HR, 1.1; 95% CI, 0.3 to 3.5). When we excluded patients whose tumors contained both HPV 16 and HPV 18 DNA, the associations were similar or somewhat stronger than those for all patients with tumors containing HPV 18 DNA. For example, for all stages combined, the HRTM was 2.5 (95% CI, 1.5 to 4.3) and the HRCCSM was 3.1 (95% CI, 1.7 to 5.6). We repeated all analyses after restricting the sample to those women for whom data on tumor grade were available (n = 231) and including adjustment for this characteristic. The results were essentially unchanged (eg, HRTM, 2.0; 95% CI, 1.1 to 3.5). Tumor size did not differ between patients with HPV 18–related carcinomas (mean [± SD], 4.0 ± 2.2 cm; n = 44) and those with HPV 16–related carcinomas (mean, 3.7 ± 2.1 cm; n = 91) (P = .43); we obtained similar results when we conducted these comparisons separately for early- and late-stage tumors. Data on lymph node involvement were available for 61 patients with HPV 18–related tumors and 131 patients with HPV 16–related tumors. The proportion of patients with lymph node involvement did not differ by HPV type (24.5% for HPV 18–related tumors v 20.6% for HPV 16–related tumors, P = .657) There was no association between prognosis and the ability to detect HPV DNA (HRTM, 0.7; 95% CI, 0.5 to 1.1; HRCCSM, 0.7; 95% CI, 0.4 to 1.2). In a model in which HPV 18–related, HPV 16–related, and HPV-negative tumors were jointly analyzed (with HPV 16–related tumors as the reference group), the association between HPV-negative tumors and overall survival (HRTM, 1.5; 95% CI, 0.9 to 2.5) was between that for HPV 16– and HPV 18–related tumors (HRTM, 2.0; 95% CI, 1.2 to 3.2).
Among those cervical cancer patients whose tumors contained HPV DNA, detection of HPV 18 DNA was associated with a two-fold increase in the risk of death. Notably, this association was limited entirely to patients with early-stage disease; in this group, HPV 18 DNA was associated with an approximately three-fold increase in the risk of death. Further, the associations were specific to deaths in which cervical cancer was the underlying cause. These associations seemed to be independent of established cervical carcinoma prognostic factors, such as age, FIGO stage, histologic type, tumor differentiation, and tumor size. Although our study had several strengths compared with most previous studies, it was limited in several potentially important respects. First, we were able to study only slightly more than half of the eligible patients in the population. The patients we were unable to include did not seem to differ systematically from those we did include with respect to prognosis; distributions of age, race, stage, and histologic type; or associations of stage and histologic type with prognosis. Although these results do not ensure that the poorer prognosis we observed associated with HPV 18 DNA is unaffected by selection bias, they do suggest that the possibility of substantial error is low. Second, whereas optimal sample preparation and rigorous testing with a variety of primers should identify HPV DNA in more than 99% of cervical carcinomas,4,5 we were able to detect HPV DNA in only 85% of the tumors we studied. It is highly likely that the majority of the underascertainment of HPV DNA was due to our reliance on paraffin-embedded specimens, from which genetic material can be sufficiently degraded to impair amplification of viral DNA sequences. Although older tissue specimens were more likely to yield DNA that could not be tested for HPV (as assessed by the ability to amplify beta-globin DNA sequences), we did not find evidence that HPV-negative tumors and HPV-positive tumors differed in regards to the interval between diagnosis and HPV testing in our lab. This suggests little if any bias in our results from this source. Our comparisons of HPV 18–related and HPV 16–related cervical carcinomas nonetheless could have been affected by underascertainment of HPV DNA if these types had had different associations with prognosis among the tumors in which HPV DNA could not be detected. Results of our joint analysis of HPV 16–related, HPV 18–related, and HPV DNA–negative tumors, however, suggest that the HPV DNA–negative tumors likely included HPV 16–related and HPV 18–related tumors with outcomes similar to outcomes for those tumors we were able to type successfully. Finally, we were able to determine only whether a patient had died from cervical carcinoma based on coding of electronic death certificate data. Information on underlying cause of death is subject to error but should not have affected our results, unless the error was differential according to HPV DNA status of the tumor, which seems unlikely. HPV 18 DNA is found particularly often in cervical adenocarcinoma, but we did not find an association between this histologic type and poorer prognosis. Although institution-based studies involving relatively few patients have provided conflicting evidence concerning the prognosis of women with adenocarcinoma versus that of women with squamous cell carcinoma,34 population-based data do not indicate any difference in survival, except possibly among patients with late-stage disease.35 Divergent impressions about the association between histologic type and cervical cancer prognosis therefore might arise if patient series in institution-based studies are more heavily weighted toward patients with late-stage disease than toward the general population. Because our study included relatively few patients with adenocarcinomas, we were unable to explore the relationship between HPV 18 DNA and prognosis according to FIGO stage of adenocarcinoma to determine whether HPV type accounts for some or all of the reported poorer prognosis associated with adenocarcinomas among patients with late-stage disease.35 Combination chemotherapy regimens in conjunction with surgery and/or radiotherapy dramatically improve survival of cervical cancer patients, regardless of stage at diagnosis,36-39 but the issue of whether the efficacy of this regimen is influenced by HPV type has received limited attention. In a small study involving 33 patients with "bulky" early-stage cervical carcinoma who were receiving combination chemotherapy, both tumor response and survival were significantly lower among patients with HPV 18–related tumors than among patients with HPV 16–related tumors.40 However, the authors did not account for known prognostic factors in their analysis. Because only 11% of our study patients with early-stage disease received chemotherapy as part of the first course of treatment, we were unable to examine whether HPV type influenced the outcome of such patients.
Early studies examining the relationship between HPV type and cervical cancer prognosis9-11,18-20,41 include some that found a poorer outcome among patients with HPV 18–related tumors.9-11 These studies involved relatively small numbers of patients, did not use failure-time methods, and/or did not account for potential confounding by known prognostic factors. Further, a recent study’s finding of no difference in prognosis according to HPV type was based on unadjusted survival estimates.23 In contrast, results from several relatively recent studies that used multivariate failure-time methods, but involved small numbers of patients ( Whether the poorer prognosis associated with HPV 18–related cervical carcinomas can be capitalized on to improve the outcome of cervical cancer patients is likely to depend, at least in part, on identification of the mechanism or mechanisms by which this association occurs. In vitro studies have found several features of HPV 18 infection that seem to differ from HPV 16 infection, such as higher frequencies of integration into the host genome,42-44 enhanced E7 phosphorylation,45 enhanced transformation,7,8 and lower rates of apoptosis.46 Further, some reports have suggested that HPV 18 infection is associated with more rapid development of cervical neoplasia and with invasive carcinomas exhibiting poor prognostic features.47-49 Taken together, these findings support the existence of molecular mechanisms that would lead HPV 18–related tumors to have, on average, a more aggressive phenotype than that of HPV 16–related tumors. It might be expected that the aforementioned phenomena might manifest in tumors of a more advanced stage at diagnosis and thus that HPV type would not be associated with prognosis independent of FIGO stage. However, in our study as well as those of others,13-16 patients with HPV 18–related tumors were not more likely to receive a diagnosis at a later stage than patients with HPV 16–related tumors, and consequently the studies cited found little evidence that stage mediates the relation between HPV 18 DNA and outcome. One possible explanation for this apparent paradox is that HPV 18 is indeed related to disease stage but understaging due to insufficient diagnostic work-up (eg, failure to examine enough lymph nodes) has resulted in misclassification that obscures the association. Alternatively, among patients with early-stage disease, HPV 18 could be related to more finely measured aspects of disease spread that influence prognosis, such as lymphatic space invasion. Although we did not obtain data on this characteristic, findings from a previous study suggest that controlling for lymphatic space invasion does not influence the relation between HPV 18 and prognosis.14 HPV 18 might also influence prognosis through disease spread, but identification of this process would be possible only through molecular analysis of lymph nodes or margins. Finally, HPV 18–related tumors may express novel factors that enhance tumor growth or inhibit host response to the tumor, resulting in a growth advantage (possibly detectable only at the molecular level at the time of diagnosis) compared with HPV 16–related tumors. Additional research is needed to determine whether there are functional aspects of the viral genome, or cellular factors that act downstream of viral oncoproteins, that are characteristic of HPV 18–related tumors and that mediate the development of clinically aggressive cervical carcinoma. For example, sequence variation in the noncoding region of HPV 16 has been associated with more rapidly progressing preinvasive cervical lesions.50,51 Therefore, one possibility is that poorer survival among patients with HPV 18–related cervical cancers reflects a higher prevalence of tumors with specific HPV 18 noncoding region variants that produce aggressive behavior. If mechanisms are identified that explain the association between HPV 18 and poorer prognosis, it may be possible to develop novel therapies to counteract the development of aggressive disease among patients with HPV 18–related tumors. These yet unidentified mechanisms may be operating, but at lower levels, in cervical carcinomas containing HPV 16 and other oncogenic HPV types. If so, potential novel therapies could reduce mortality among all patients with invasive cervical cancer.
Supported by Program Project grant no. P01 CA 42792, contract N01-CN-05230, from the National Cancer Institute and by additional funds from the Fred Hutchinson Cancer Research Center, Seattle, WA.
1. Greenlee RT, Murray T, Bolden S, et al: Cancer statistics, 2000. CA Cancer J Clin 50: 7-33, 2000[Abstract] 2. Kosary CL, Ries LAG, Miller BA, et al: SEER Cancer Statistics Review, 1973-1992: Tables and Graphs. Bethesda, MD, National Cancer Institute, NIH publication 96-2789, 1995 3. Schiffman MH, Brinton LA, Devesa SS, et al: Cervical cancer, in Schottenfeld D, Fraumeni JF Jr (eds): Cancer Epidemiology and Prevention ( ed 2 ). New York, NY, Oxford University Press, 1996, pp 1090-1116 4. Walboomers JM, Jacobs MV, Manos MM, et al: Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol 189: 12-19, 1999[Medline]
5.
Bosch FX, Manos MM, Munoz N, et al: Prevalence of human papillomavirus in cervical cancer: A worldwide perspective—International biological study on cervical cancer (IBSCC) Study Group. J Natl Cancer Inst 87: 796-802, 1995 6. Barnes W, Delgado G, Kurman RJ, et al: Possible prognostic significance of human papillomavirus type in cervical cancer. Gynecol Oncol 29: 267-273, 1988[Medline] 7. Barnes W, Woodworth G, Waggoner S, et al: Rapid dysplastic transformation of human genital cells by human papillomavirus type 18. Gynecol Oncol 38: 343-346, 1990[Medline] 8. Villa LL, Schlegel R: Differences in transformation activity between HPV-18 and HPV-16 map to the viral LCR-E6-E7 region. Virology 181: 374-377, 1991[Medline] 9. Walker J, Bloss JD, Liao SY, et al: Human papillomavirus genotype as a prognostic indicator in carcinoma of the uterine cervix. Obstet Gynecol 74: 781-785, 1989[Medline] 10. Burnett AF, Barnes WA, Johnson JC, et al: Prognostic significance of polymerase chain reaction detected human papillomavirus of tumors and lymph nodes in surgically treated stage IB cervical cancer. Gynecol Oncol 47: 343-347, 1992[Medline] 11. Zhang J, Rose BR, Thompson CH, et al: Associations between oncogenic human papillomaviruses and local invasive patterns in cervical cancer. Gynecol Oncol 57: 170-177, 1995[Medline] 12. Rose BR, Thompson CH, Simpson JM, et al: Human papillomavirus deoxyribonucleic acid as a prognostic indicator in early-stage cervical cancer: A possible role for type 18. Am J Obstet Gynecol 173: 1461-1468, 1995[Medline] 13. Hagmar B, Christensen JJP, Johansson B, et al: Implications of human papillomavirus type for survival in cervical squamous cell carcinoma. Int J Gynaecol Cancer 5: 341-345, 1995[Medline]
14.
Burger RA, Monk BJ, Kurosaki T, et al: Human papillomavirus type 18: Association with poor prognosis in early stage cervical cancer. J Natl Cancer Inst 88: 1361-1368, 1996 15. Nakagawa S, Yoshikawa H, Onda T, et al: Type of human papillomavirus is related to clinical features of cervical carcinoma. Cancer 78: 1935-1941, 1996[Medline] 16. Lombard I, Vincent-Salomon A, Validire P, et al: Human papillomavirus genotype as a major determinant of the course of cervical cancer. J Clin Oncol 16: 2613-2619, 1998[Abstract] 17. Lai HC, Sun CA, Yu MH, et al: Favorable clinical outcome of cervical cancers infected with human papilloma virus type 58 and related types. Int J Cancer 84: 553-557, 1999[Medline] 18. King LA, Tase T, Twiggs LB, et al: Prognostic significance of the presence of human papillomavirus DNA in patients with invasive carcinoma of the cervix. Cancer 63: 897-900, 1989[Medline] 19. Leminen A, Paavonen J, Vesterinen E, et al: Human papillomavirus types 16 and 18 in adenocarcinoma of the uterine cervix. Am J Clin Pathol 95: 647-652, 1991[Medline] 20. Ip EW, Collins RJ, Srivastava G, et al: Human papillomavirus and its prognostic significance in invasive carcinoma of the cervix in young patients. Int J Gynecol Pathol 11: 258-265, 1992[Medline]
21.
Viladiu P, Bosch FX, Castellsague X, et al: Human papillomavirus DNA and antibodies to human papillomaviruses 16 E2, L2, and E7 peptides as predictors of survival in patients with squamous cell cervical cancer. J Clin Oncol 15: 610-619, 1997 22. Kristensen GB, Karlsen F, Jenkins A, et al: Human papilloma virus has no prognostic significance in cervical carcinoma. Eur J Cancer 32A: 1349-1353, 1996 23. van Muyden RC, ter Harmsel BW, Smedts FM, et al: Detection and typing of human papillomavirus in cervical carcinomas in Russian women: A prognostic study. Cancer 85: 2011-2016, 1999[Medline] 24. Daling JR, Madeleine MM, McKnight B, et al: The relationship of human papillomavirus-related cervical tumors to cigarette smoking, oral contraceptive use, and prior herpes simplex virus type 2 infection. Cancer Epidemiol Biomarkers Prev 5: 541-548, 1996[Abstract]
25.
Hankey BF, Ries LA, Edwards BK: The surveillance, epidemiology, and end results program: A national resource. Cancer Epidemiol Biomarkers Prev 8: 1117-1121, 1999 26. International Classification of Diseases, 9th Revision, Clinical Modification: ICD-9-CM (ed 4). Washington, DC, U.S. Department of Health and Human Services, Public Health Service, Health Care Financing Administration, DHHS publication no. (PHS) 91-1260, 1991 27. Wright DK, Manos MM: Sample preparation from paraffin-embedded tissues, in Innis MA, Gelfand DH, Sninsky JJ, et al (eds): PCR Protocols: A Guide to Methods and Applications. San Diego, CA, Academic Press, 1990, pp 153-158 28. Manos MM, Ting Y, Wright DK, et al: Use of polymerase chain reaction amplification for the detection of genital human papillomaviruses. Cancer Cells 7: 209-214, 1989 29. Jenison SA, Yu X-P, Valentine JM, et al: Evidence of prevalent genital-type human papillomavirus infections in adults and children. J Infect Dis 162: 60-69, 1990[Medline] 30. Anderson S, Shera K, Ihle J, et al: Telomerase activation in cervical cancer. Am J Pathol 151: 25-31, 1997[Abstract]
31.
Bauer HM, Ting Y, Greer CE, et al: Genital human papillomavirus infection in female university students as determined by a PCR-based method. JAMA 265: 472-477, 1991 32. Lungu O, Wright TC Jr, Silverstein S: Typing of human papillomaviruses by polymerase chain reaction amplification with L1 consensus primers and RFLP analysis. Mol Cell Probes 6: 145-152, 1992[Medline]
33.
Grambsch PM, Therneau TM: Proportional hazards tests and diagnostics based on weighted residuals. Biometrika 81: 515-526, 1994 34. Kilgore LC, Helm W: Adenocarcinoma of the uterine cervix. Clin Obstet Gynecol 33: 863-871, 1990[Medline] 35. Smith HO, Tiffany MF, Qualls CR, et al: The rising incidence of adenocarcinoma relative to squamous cell carcinoma of the uterine cervix in the United States: A 24-year population-based study. Gynecol Oncol 78: 97-105, 2000[Medline]
36.
Keys HM, Bundy BN, Stehman FB, et al: Cisplatin, radiation, and adjuvant hysterectomy compared with radiation and adjuvant hysterectomy for bulky stage IB cervical carcinoma. N Engl J Med 340: 1154-1161, 1999
37.
Morris M, Eifel PJ, Lu J, et al: Pelvic radiation with concurrent chemotherapy compared with pelvic and para-aortic radiation for high-risk cervical cancer. N Engl J Med 340: 1137-1143, 1999
38.
Rose PG, Bundy BN, Watkins EB, et al: Concurrent cisplatin-based radiotherapy and chemotherapy for locally advanced cervical cancer. N Engl J Med 340: 1144-1153, 1999
39.
Whitney CW, Sause W, Bundy BN, et al: Randomized comparison of fluorouracil plus cisplatin versus hydroxyurea as an adjunct to radiation therapy in stage IIB-IVA carcinoma of the cervix with negative para-aortic lymph nodes: A Gynecologic Oncology Group and Southwest Oncology Group study. J Clin Oncol 17: 1339-1348, 1999 40. Tseng CJ, Tseng LH, Lai CH, et al: Identification of human papillomavirus types 16 and 18 deoxyribonucleic acid sequences in bulky cervical cancer after chemotherapy. Am J Obstet Gynecol 176: 865-869, 1997[Medline] 41. van Bommel PF, van den Brule AJ, Helmerhorst TJ, et al: HPV DNA presence and HPV genotypes as prognostic factors in low-stage squamous cell cervical cancer. Gynecol Oncol 48: 333-337, 1993[Medline]
42.
Cullen AP, Reid R, Campion M, et al: Analysis of the physical state of different human papillomavirus DNAs in intraepithelial and invasive cervical neoplasm. J Virol 65: 606-612, 1991 43. Stoler MH, Rhodes CR, Whitbeck A, et al: Human papillomavirus type 16 and 18 gene expression in cervical neoplasias. Hum Pathol 23: 117-128, 1992[Medline] 44. Park JS, Hwang ES, Park SN, et al: Physical status and expression of HPV genes in cervical cancers. Gynecol Oncol 65: 121-129, 1997[Medline]
45.
Werness BA, Levine AJ, Howley PM: Association of human papillomavirus types 16 and 18 E6 proteins with p53. Science 248: 76-79, 1990 46. Arends MJ, Wyllie AH, Bird CC: Human papillomavirus type 18 is associated with less apoptosis in fibroblast tumours than human papillomavirus type 16. Br J Cancer 72: 646-649, 1995[Medline] 47. Kurman RJ, Schiffman MH, Lancaster WD, et al: Analysis of individual human papillomavirus types in cervical neoplasia: A possible role for type 18 in rapid progression. Am J Obstet Gynecol 159: 293-296, 1988[Medline] 48. Ambros RA, Park JS, Shah KV, et al: Evaluation of histologic, morphometric, and immunohistochemical criteria in the differential diagnosis of small cell carcinomas of the cervix with particular reference to human papillomavirus types 16 and 18. Mod Pathol 4: 586-593, 1991[Medline] 49. Arends MJ, Donaldson YK, Duvall E, et al: Human papillomavirus type 18 associates with more advanced cervical neoplasia than human papillomavirus type 16. Hum Pathol 24: 432-437, 1993[Medline]
50.
Ho L, Chan SY, Burk RD, et al: The genetic drift of human papillomavirus type 16 is a means of reconstructing prehistoric viral spread and the movement of ancient human populations. J Virol 67: 6413-6423, 1993
51.
Xi LF, Koutsky LA, Galloway DA, et al: Genomic variation of human papillomavirus type 16 and risk for high grade cervical intraepithelial neoplasia. J Natl Cancer Inst 89: 796-802, 1997 Submitted March 15, 2000; accepted December 13, 2000.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2001 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
12) with HPV 18–related tumors, suggested associations between HPV 18–related tumors and poor outcome.
