Molecular prognostic factors in early-stage cervical adenocarcinoma

Introduction

According to the literature reports adenocarcinoma of the uterine cervix accounts for 15-25% of all the cervical cancer (CC) cases [1]. Unlike squamous cell carcinoma, in which case lower incidence rate in the last decades is accompanied by decrease in its mortality, the mortality of cervical adenocarcinoma (CAC) remains unaltered and may indicate more aggressive character of the disease [2].
Within the past 24 years in the USA, the proportion of CAC has increased to around 29% [3]. It is probably due to more and more frequent detection of precursor lesions and related decrease in the rate of squamous cell carcinomas. Cervical adenocarcinoma, compared with primary cervical cancer (PCC), has more histological subtypes of different lymph node metastases potential or tendency to relapse. It seems that the proportion of adenocarcinomas of the CC in the world has been increasing over the last decades, particularly in younger women [4, 5, 6, 7, 8, 9].
There is a shortage of data regarding immunohistochemical and molecular features and their prognostic relevance in CAC. The most of clinical results discuss clinicopathological factors only [10, 11, 12, 13, 14, 15, 16, 17, 18]. Aim of the present study was to evaluate molecular prognostic factors in early-stage cervical adenocarcinoma (esCAC) patients treated with primary surgery.

Material and methods

Between 1985 and 2000, 195 patients with CAC confirmed by biopsy, uterine cervix scrapings examination or resection tissue material examination were treated at the Gynaecology Department of Centre of Oncology, Maria Sklodowska-Curie Memorial Institute, Krakow Branch, Poland. In that number, 120 patients with esCAC were treated with primary surgery.
Results of combined treatment (surgery only or surgery + radiotherapy) in term of progression-free and overall survival were evaluated during follow-up visits. Disease progression or recurrence was defined as the presence of measurable lesions observed in medical examination (physical examination, particularly gynaecological) or imaging (computer tomography of abdomen and pelvis, X-ray examination of chest, US examination of abdomen and pelvis).
Available histological specimens from 82 CAC patients treated with primary surgery were reviewed by an experienced pathologist prior to immunohistochemistry (IHC) and flow cytometry (FCM) analysis for prognostic factors. Retrospective morphological assessment of tissue material was performed on sections prepared from paraffin-embedded histological specimens from the archives of the Department of Tumour Pathology, Centre of Oncology, Maria Sklodowska-Curie Memorial Institute, Krakow Branch, or borrowed from other pathology departments (hospitals of the Ma³opolska province). Due to technical issues, 23 of the specimens were not suitable for the planned testing and were excluded from detailed analysis. For the remaining 59, IHC and FCM analyses were performed on fixed paraffin-embedded sections of tumour tissue. Tumour tissue sections were routinely stained with hematoxylin and eosin followed by microscopic examination. Based on the microscopic evaluation, all tumours were classified according to the current World Health Organization (WHO) guidelines published in 2014 [19]. In all cases, tumour histology and grade were established as well as nuclear atypia grade of tumour cells.
All immunohistochemical analyses were performed on paraffin-embedded section of 5µ thickness. Diagnostic techniques applied are presented in Table I.
Flow cytometry (FCM) analysis of paraffin-embedded tumor tissue was performed using flow cytometer FACSCalibur equipped with argon laser (15 mW, 488 nm). DNA histogram analysis was performed with ModFit application. The analysis included at least 20 000 collected parameters.
DNA histograms were classified according to the criteria established by Cytometry Consensus Conference in 1992 [20]. DNA ploidy was described by DNA index (DI). Histogram of DI = 1.00 was classified as diploid, and of DI ≠ 1,00 as aneuploid. Tumour proliferation rate was expressed in S-phase fraction or proliferation index (S+G2M-phase fraction) [21].
Significant demographic, microscopic and clinical characteristics of 59 patients, whose specimens were analysed, were compared with the whole group of consecutive 120 patients with esCAC. Those included age, menopausal status, comorbidities, FIGO stage, surgical procedures, grading of the tumour, surgical margin, parametrial involvement, lymphovascular space invasion (LVSI), regional lymph node status. The analysis confirmed no differences in the distribution of statistically significant variables. Hence, the prognostic value of analysed factors can be translated into the general population of esCAC patients.
Treatment effectiveness was evaluated using overall 5-year survival. Survival probability was estimated using the Kaplan-Meier method [22].
The Peto log-rank test was used to evaluate statistical significance of the observed result differences [23]. Statistical significance level was set at p ≤ 0.05. Continuous variables, such as age or overall survival time, as well as discrete variables were tested, both original and adjusted, to establish any variability range significantly affecting the survival.
Influence of selected factors on patient survival times was assessed using Cox’s proportional hazard model [24].

Results

Comparison of demographic, microscopic and clinical characteristics of 59 esCAC patients included in the detailed IHC and FCM with the whole group of 120 patients with esCAC showed no differences in distribution of following: age, menopausal status, concomitant diseases, FIGO stage, surgical margin, parametrial and uterine infiltration, LVSI presence and lymp node status. There were a minor differences in distribution of surgery extend (patients with specimen analysed were more often treated with pelvic lymphadenectomy then the whole group of 120 patients – 57.6% vs. 45.8%, respectively; p = 0.027) and the grade of the CAC (analyzed group was more likely to have G2/G3 tumour then the whole group – 77% vs. 63.3%, respectively; p = 0.003)
Detailed results of IHC and FCM analyses performed for the 59 esCAC patient subgroup are given in Table II.
Mean DNA index was 1.2 (SD ±0.32), mean percentage of cells in S-phase (SPF) was 1.07 (SD ±10.54) and mean SG2M index was 14.58 (SD ±12.95).
Overall survival rate estimated using Kaplan-Meier method was 74.6% in the investigated group of 59 CAC patients and 73.3% in the entire group of 120 patients with esCAC. Kaplan-Meier chart for overall 5-year survival both groups is presented in Fig. 1.
Among the IHC and FCM features univariate analysis showed statistical significance of nm23-H1 gene expression and total S-phase fraction ≤ 11.9% (index S-TOT). Differences in treatment outcome in the 59 CAC patient group depending on S-TOT value are provided in Table III.
In multivariate analysis LVSI and parametrial involvement had significant, negative impact on survival (HR = 8.04, p < 0.003 and HR = 4.03, p < 0.017, respectively). However, none of the tested IHC and FCM features had any influence on overall 5-year survival.

Discussion

Group of 59 patients, with pathological specimens available, was subject to thorough microscopic evaluation and IHC and FCM analyses described above. In total, 18 different factors were studied using both methods. As mentioned above this group of patients were compared with the whole group of consecutive 120 patients with esCAC to confirmed no differences in the distribution of statistically significant, clinical characteristics.
The significance of nm23-H1 metastasis suppressor gene expression in CAC is not well determined. According to the literature reports, the gene expression is observed in 44-75% CAC patients. In the investigated group, as much as 96.6% of patients shows nm23-H1 gene expression; however, it is moderate (2+) or strong (3+) in 83% of them. Reports on the prognostic value of nm23-H1 gene expression are contradictory. Mandai et al. suggested already in 1995 that its expression is directly correlated with 10-year survival rate in univariate analysis, and when combined with c-erbB-2/Her2-neu gene expression the correlation is observed also in multivariate analysis. Similar conclusions were drawn by Huang et al. and Kristensen et al., on the other hand, present opposite findings and suggest inverse correlation between the gene expression and survival times, however only in univariate analysis. Moreover, against the suggested gene role, none of the authors did prove any suppressing effect of the gene on locoregional lymph node metastasis.
In the present study, nm23-H1 expression inversely correlated with survival times and was not connected with the rate of lymph node metastasis. Patients of strong nm23-H1 expression lived shorter than patients without the gene product expression and had 65.5% and 89.2% of 5-year survival rate, respectively [25, 26, 27].
S-phase fraction was the only FCM feature that in univariate analysis was distinctly related with survival. The patients with esCAC characterized by low S-TOT (≤ 11.9%) had over two times higher chances to survive 5 years than patients of S-TOT > 11.9%. Five-year survival rate was 42.9% and 78.9%, respectively. In the available literature, there were no reports to be found concerning FCM data in CAC. Other FCF features, that is DNA ploidy, DNA index and SG2M, had no prognostic value in the present study.
In the summary, it should be emphasized that, regardless of the univariate analysis results presented above, multivariate analysis of the 59-patient group did not show any statistical significance of IHC and FCM features for CAC patients prognosis.
The present analysis has several limitations that should be mentioned. Firstly, the investigated group of 59 patients, for which detailed IHC and FCM analyses were performed, had rather good prognosis and, thus, relatively low mortality in the studied period. Combined with relatively small size of the patient group, it lowered statistical impact of the results and made it more difficult to evaluate prognostic factors of distinctly lower influence on the patients’ prognosis.
Secondly, due to variable quality and quantity of the available histological material, it was impossible to repeat or clarify some of the analyses. For example, many specimens lacked tissue components necessary to run analyses, as was in case of 22-1-1 antigen.
Regardless the limitations, the present study seems to assess comprehensively molecular prognostic factors in CAC patients. Available reports present no such extended analysis as this one that would evaluate in detail immunohistochemical and cytofluorometric characteristics of CAC and their influence on patient prognosis. Multivariate analysis allowed to conclude that IHC and FCM features have little impact on patient prognosis in esCAC, and the most important seem to be clinical and histopathological characteristics.

The authors declare no conflict of interest.

References

1. Wright C, Ferenczy A, Kurman RJ. Carcinoma and Rother tumors of the cervix. In: Kurman RJ (ed.). Blaustein’s Pathology of the Female Genital tract, 4th ed. New York, Springer-Verlag, 1994; 279-326.
2. Sherman ME, Wang SS, Carreon J, et al. Mortality trends for cervical squamous and adenocarcinoma in the United States. Relation to incidence and survival. Cancer 2005; 103: 1258-1264.
3. Smith HO, Tiffany M, Qualls CR, et al. The rising incidence of adenocarcinoma relative to squamous cell carcinoma of the uterine cervix in the U.S.: A 24-year population-based study. Gynecol Oncol 2000; 78: 97-105.
4. Peters RK, Chao A, Mack TM, et al. Increased frequency of adenocarcinoma of the uterine cervix in young women in Los Angeles County. J Natl Cancer Inst 1986; 76: 423-428.
5. Schwartz SM, Weiss N. Increased incidence of adenocarcinoma of the cervix in young women in the United States. Am J Epidemiol 1986; 124: 1045-1047.
6. Anton-Culver H, Bloss JD, Bringman D, et al. Comparison of adenocarcinoma and squamous cell carcinoma of the uterine cervix: A population based epidemiologic study. Am J Obstet Gynecol 1992; 186: 1507-1514.
7. Hopkins P, Morley GW. A comparison of adenocarcinoma and squamous cell carcinoma of the cervix. Obstet Gynecol 1991; 7: 912-917.
8. Nieminen P, Kallio M, Hakama M. The effect of mass screening on incidence and mortality of squamous and adenocarcinoma of cervix uteri. Obstet Gynecol 1995; 85: 1017-1021.
9. Platz CE, Benda JA. Histology of cancer, incidence and prognosis: SEER population-based data, 1973-1987, female genital tract cancer. Cancer 1995; 75: 270-294.
10. Iversen T, Abeler V, Kjrostad KE. Factors influencing the treatment of patients with stage IA carcinoma of the cervix. Br J Obstet Gynaecol 1979; 86: 593-597.
11. Qizilbash AH. In situ and microinvasive adenocarcinoma of the uterine cervix: A clinical, cytologic and histologic study of 14 cases. Am I Clin Pathol 1975; 64: 155-170.
12. Schorge JO, Lee KR, Flynn CE, et al. Stage 1A1 cervical adenocarcinoma: Definition and treatment. Obstet Gynecol 1999; 93: 219-222.
13. Kaku T, Kamura T, Sakai K, et al. Early adenocarcinoma of the uterine cervix. Gynecol Oncol 1997; 65: 281-285.
14. Hopkins P, Peters WA, Anderson W, et al. Invasive cervical cancer treated initially by standard hysterectomy. Gynecol Oncol 1990; 36: 7-12.
15. McHale M, Le TD, Burger RA, et al. Fertility sparing treatment for in situ and early invasive adenocarcinoma of the cervix. Obstet Gynecol 2001; 98: 726-731.
16. Matsukuma K, Tsukamoto N, Kaku T, et al. Early adenocarcinoma of the uterine cervix: its histologic and immunohistologic study. Gynecol Oncol 1989; 35: 38-43.
17. Rollason TP, Cullimore J, Bradgate MG. A suggested columnar cell morphological equivalent of squamous carcinoma in situ with early stromal invasion. Int J Gynaecol Pathol 1989; 8: 230-236.
18. Kaspar HG, Dinh TV, Doherty MG, et al. Clinical implications of tumor volume measurement in stage I adenocarcinoma of the cervix. Obstet Gynecol 1993; 81: 296-300.
19. Kurman RJ, Carcangiu ML, Herrington CS, et al. (eds.). WHO Classification of Tumours of Female Reproductive Organs. IARC, Lyon 2014.
20. Shankey TW, Rabinovitch PS, Bagwell B, et al. Guidelines for implementation of clinical DNA cytometry. Cytometry 1993; 14: 472-477.
21. Hedley DW, Friedlander ML, Taylor IW, et al. Method for analysis of cellular DNA content of paraffin-embedded pathological material using flow cytometry. J Histochem Cytochem 1983; 31: 1333-1335.
22. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457.
23. Peto R, Pike MC, Armitage P, et al. Design and analysis of randomized clinical trials requiring prolonged observation of each patient. I. Introduction and design. Br J Cancer 1976; 34: 585-612.
24. Cox DR. Regression models and life-tables (with discussion). J R Stat Soc B 1972; 34: 187.
25. Mandai M, Konishi I, Komatsu T, et al. Mutation of the nm23 gene, loss of heterozygosity at the nm23 locus and K-ras mutation in ovarian carcinoma: correlation with tumour progression and nm23 gene expression. Br J Cancer 1995; 72: 691-695.
26. Huang Y, Cai S, Yu S. Relationship between nm23-H1 expression and lymph node metastasis and prognosis in cervical cancer. Zhonghua Fu Chan Ke Za Zhi 1997; 32: 718-721.
27. Kristensen GB, Holm R, Abeler VM, et al. Evaluation of the prognostic significance of nm23/NDP kinase protein expression in cervical carcinoma: an immunohistochemical study. Gynecol Oncol 1996; 61: 378-383.

Address for correspondence

Wiktor Szatkowski
Gynecologic Oncology Department
Maria Sk³odowska-Curie Memorial Cancer Center
and Institute of Oncology
Krakow Branch
Garncarska 11
31-115 Krakow, Poland
e-mail: vigor27@wp.pl