International Journal of Gynecological Cancer:
Prognosis of Adenosquamous Carcinoma Compared With Adenocarcinoma in Uterine Cervical Cancer: A Systematic Review and Meta-Analysis of Observational Studies
Lee, Jung-Yun MD*; Lee, Chulmin MD, PhD†; Hahn, SeoKyung PhD‡; Kim, Min A. MD, PhD§; Kim, Hee Seung MD*; Chung, Hyun Hoon MD, PhD*; Kim, Jae-Weon MD, PhD*; Park, Noh Hyun MD, PhD*; Song, Yong-Sang MD, PhD*
*Department of Obstetrics and Gynecology, College of Medicine, Seoul National University; †Department of Obstetrics and Gynecology, Inje University Sanggye Paik Hospital; and ‡Departments of Medicine, and §Pathology, Seoul National University College of Medicine, Seoul, Korea.
Address correspondence and reprint requests to Jae-Weon Kim, MD, PhD, Department of Obstetrics and Gynecology, College of Medicine, Seoul National University, 101 Daehak-ro, Jongno-gu, 110-744 Seoul, Korea. E-mail: email@example.com.
This work was supported by Research Resettlement Fund for the new faculty of Seoul National University.
The authors declare no conflicts of interest.
Supplemental digital content is available for this article. Direct URL citation appears in the printed text and is provided in the HTML and PDF versions of this article on the journal’s Web site (www.ijgc.net).
Received July 30, 2013
Received in revised form November 11, 2013
Accepted November 11, 2013
Objective: The aim of this study was to compare the survival outcomes of adenosquamous carcinoma (ASC) and adenocarcinoma (AC) of the cervix.
Methods: We searched PubMed and Embase for observational studies that compared the outcomes of 2 histologic subtypes. Hazards ratios (HRs) with 95% confidence intervals (CIs) were calculated with a fixed effects model.
Results: A total of 17 studies were included in the analyses. Patients with ASC were associated significantly with poorer overall survival (death HR, 1.27; 95% CI, 1.12–1.43; I2 = 0%) and recurrence-free survival (recurrence HR, 1.43; 95% CI, 1.05–1.95; I2 = 19.4%) than those with AC. For clinical stages I and II in particular, ASC predicted significantly poorer outcomes compared with AC (death HR, 1.41; 95% CI, 1.17–1.70; I2 = 0%).
Conclusions: This meta-analysis suggests that ASC may have poorer outcomes compared with AC of the cervix.
Cervical cancer is the second most common cause of cancer deaths in women worldwide.1 Adenocarcinoma (AC) and adenosquamous carcinoma (ASC) of the cervix are relatively uncommon histologic subtypes of cervical cancer compared with squamous cell carcinoma (SCC). However, the incidence of cervical AC is increasing and accounts for approximately 20% of all cervical cancers in developed countries.2,3 In contrast to a marked decrease in the incidence of SCC, AC has become more prevalent over the past 3 decades, indicating that cervical cytology screening is less effective for AC.4,5
The current guidelines for cervical cancer include the management of SCC, AC, and ASC.6 Histologic subtypes of AC and ASC are typically treated in the same manner as SCC. However, controversy remains over whether histologic subtypes are independent prognostic factors for patients with cervical cancer.
Several studies have shown that cervical AC has a poorer prognosis than squamous carcinoma.7,8 Some of these studies did not make a distinction between ASC and AC and included ASC in subtypes of AC.9–11 However, the prognostic role of ASC histologic diagnosis has not yet been determined. Adenosquamous carcinoma is composed of a mixture of malignant glandular and squamous components. Previous researchers have suggested that patients with ASC have a poorer prognosis and distinct clinical characteristics compared with those with AC,12–15 whereas others did not find a difference in outcomes between patients with ASC and AC.16–28 This inconsistency might be attributable to the comparative rarity of ASC. This study was therefore designed to evaluate the survival difference between ASC and AC by performing a systematic review and meta-analysis of observational studies.
We identified observational studies that compared the outcomes between ASC and AC in patients with cervical cancer by performing a literature search of reports in PubMed and Embase from their respective inceptions to April 2013. Our overall search strategy included terms for cervical cancer (cervical cancer, cervical carcinoma, cervix, and cervical), histologic diagnosis (adenosquamous and adenosquamous carcinoma), and outcomes (prognosis and survival). We followed the Meta-analysis of Observational Studies in Epidemiology and reviewed the reference lists of relevant articles to identify additional studies.
We screened the eligibility of all studies retrieved from the databases independently based on predetermined selection criteria. Studies that compared outcomes such as recurrence-free survival (RFS) and overall survival (OS) between ASC and AC were selected for inclusion. We excluded studies not published in the English-language medium and those with incomplete data on survival outcomes.
We extracted the following data from each publication independently: first author, year of publication, inclusion period, study region, clinical stage, number of patients, mode of primary treatment, and hazards ratio (HR) of recurrence or death.
The aim of this meta-analysis was to compare the RFS and OS of ASC and AC. The most appropriate statistic for the meta-analyses of time-to-event outcomes and survival is HR. Where available, log HR and its variance were extracted directly from each study. However, the most appropriate summary statistics are typically not presented in the eligible studies. When these statistical variables were not given in an article, we estimated them using the available data using methods reported by Tierney et al.29 A survival curve was read by Plot Digitizer (version 2.5.1), which was downloaded from http://autotrace.sourceforge.net. Although there was no substantial heterogeneity between the studies, we reported the pooled HR with 95% confidence interval (CI) on the basis of the fixed effects models. We examined heterogeneity in results across studies using the Higgins I2 value, which measures the percentage of total variance in the summary estimate resulting from between-study heterogeneity.30,31 For the I2 value, a value of greater than 50% is considered a measure of substantial heterogeneity. To examine the potential publication bias, results were analyzed using the Egger test and Begg funnel plot test.32 All statistical analyses were performed using Stata 11.0 (StataCorp, College Station, TX).
Study Selection and Characteristics
Figure 1 shows a flow diagram of how the relevant studies were selected. The search strategy identified 918 references and a total of 686 studies remained after duplicate articles had been excluded. After excluding articles that were clearly irrelevant from the titles and abstracts, 22 articles were identified for full-text review. Of these studies, 6 were excluded because of incomplete data on survival outcomes13,16,19,33–35 and 1 was included with bibliographic search.36 After this process, a total of 17 studies remained.12,14,15,17,18,20–28,36–38 A total of 4805 patients (ASC, 1295; AC, 3510) could be identified from the 17 studies. The main characteristics of the 17 eligible articles are shown in Table 1. Overall, OS was obtained from 15 articles. Of these studies, 13 evaluated the survival outcomes of early-stage disease (clinical stages I–II) and just 3 studies compared survival based on histologic subtypes solely in advanced-stage disease (clinical stages III–IV). Recurrence-free survival was extracted from 8 studies.
As shown in Figure 2, overall, patients with ASC were associated with poor OS (death HR, 1.27; 95% CI, 1.12–1.43) without substantial heterogeneity between studies (I2 = 0%; P = 0.672) from the fixed effects model. The estimates from the random effects model were the same as those from the fixed effects model. Table 2 shows the results of the subgroup meta-analyses. When we performed subgroup analysis by clinical stage, the inferiority of OS for patients with ASC was more prominent in early-stage disease (I–II; death HR, 1.41; 95% CI, 1.17–1.70; I2 = 0%). However, there was no significant difference between the 2 histologic subtypes in advanced disease stages (III–IV; death HR, 0.99; 95% CI, 0.81–1.21; I2 = 6.0%). Surgery was the primary mode of treatment of 5 of the studies. In these studies, patients with ASC did not show significant death HR compared with AC (death HR, 1.54; 95% CI, 0.91–2.62; I2 = 0%). After subgroup analysis by study region, patients with ASC had a higher risk of death than those with AC in western countries (death HR, 1.26; 95% CI, 1.11–1.44; I2 = 27%).
In addition, a meta-analysis of 8 studies demonstrated that patients with ASC had a higher risk of recurrence than those with AC (recurrence HR, 1.43; 95% CI, 1.05–1.95; I2 = 19.4%; Table 2. Sensitivity analyses were performed to evaluate the robustness of the death HR results. The poor prognosis of ASC compared with AC remained significant when any 1 study was removed from the analysis (Supplementary Figure 1, available at http://links.lww.com/IGC/A196).
There was no significant publication bias in either Begg funnel plot test (P = 0.259) or the Egger test (P = 0.141) in studies reporting death HR.
This meta-analysis showed significantly increased HRs for recurrence and death in patients with ASC compared with those with AC. There was no substantial heterogeneity between studies included in the meta-analysis.
The prognostic significance of ASC histologic diagnosis has been limited by its inclusion as a subtype of AC in many of the studies that evaluate prognosis and outcomes of cervical cancer.39,40 Because Cherry et al41 first reported the poor prognosis of ASC compared with pure AC, a number of studies have evaluated the prognostic significance of ASC. However, the prognostic significance of ASC histologic diagnosis has not yet been determined owing largely to the comparative rarity of this disease. The incidence of ASC is around 6% in uterine cervical cancer,37 which results in difficulties in identifying a sufficient number of cases when considering single institutions.
After a systematic review, we identified 17 relevant articles that compared the outcomes of ASC and AC from 1975 to 2013. The summary estimates of this meta-analysis provide the interesting result of poorer outcomes for patients with ASC in recurrence and death, although most individual studies failed to show a significant difference between the 2 histologic subtypes. This suggests that ASC may be a different disease entity from AC.
Mechanisms related to the poor outcomes of ASC have not yet been extensively evaluated. One potential mechanism may be attributed to the high-risk factors frequently observed in ASC. It is worth noting that a poorer tumor grade and higher prevalence of lymphovascular space invasion (LVSI) have been found more often in ASC than in AC.12,27,28,37 However, whether the poor prognosis of ASC remains significant after adjusting for these high-risk factors cannot be determined in this meta-analysis because most of the eligible studies presented univariate HRs from survival curves.
The study patients with ASC had a higher HR than those with AC, especially in early-stage disease. Radical surgery was the treatment of choice for most patients with early-stage cervical cancers. Considering the poor outcomes of ASC, even in early-stage disease, we propose more aggressive guidelines for adjuvant therapy in future clinical trials.
To the best of our knowledge, this is the first meta-analysis to evaluate the prognostic significance of ASC in cervical cancer. However, our study has several potential limitations; these are as follows: first, approximately one third of patient data comes from a single study using the SEER database; second, the performance of adjuvant treatment and type of primary treatment had not been standardized in the studies included in our analysis; third, although most of the studies included in the meta-analysis were conducted before concurrent chemoradiation (CCRT) was accepted as a standard treatment, the prognostic significance of ASC in patients with early-stage cervical cancer treated with radical hysterectomy and CCRT is yet to be determined; and lastly, the interpretation of the literature might be confounded by a failure of investigators to adopt uniform criteria for diagnosis of ASC.
According to the World Health Organization classification, the definition of ASC is composed of a mixture of malignant glandular and squamous epithelium. So, the histologic diagnosis elements are scattered mucin-producing atypical cells in otherwise looking SCC. If the SCC is very poorly differentiated, the glandular elements are identified only by the use of mucin stains. Adenosquamous carcinoma should be distinguished from endometrioid AC with squamous differentiation of the cervix. In ASC, the epithelial component is mostly endocervical type, which is mucin-secreting cells, whereas endometrial cells have no or little intracellular mucin production.42 In addition, glassy cell carcinoma has been recognized as a poorly differentiated form of ASC with an aggressive course and a poor prognosis.43,44 Individual cells have abundant eosinophilic, granular, ground-glass cytoplasm, large round to oval nuclei, and prominent nucleoli. Although glassy cell carcinoma has been classified as a variant of ASC according to World Health Organization classification, most previous studies did not clarify the inclusion of glassy cell carcinoma to ASC. Future studies should endeavor to be specific regarding the exact inclusion criteria being used.
In conclusion, we found that patients with early-stage ASC of the cervix may have worse survival outcomes than those with AC. The current management for ASC is virtually the same as that for AC or SCC. As such, the findings of this study suggest that a new strategy is warranted for the treatment of ASC. Further large-scale studies using multicenter databases or national cancer registries are required to confirm our results. Importantly, only a limited number of studies have been performed using data from after CCRT was accepted as a standard treatment option, meaning that the prognostic significance of ASC should be determined in using data from the CCRT era.
1. Jemal A, Bray F, Center MM, et al. Global cancer statistics. CA Cancer J Clin. 2011; 61: 69–90.
2. Wang SS, Sherman ME, Hildesheim A, et al. Cervical adenocarcinoma and squamous cell carcinoma incidence trends among white women and black women in the United States for 1976–2000. Cancer. 2004; 100: 1035–1044.
3. Bray F, Carstensen B, Møller H, et al. Incidence trends of adenocarcinoma of the cervix in 13 European countries. Cancer Epidemiol Biomarkers Prev. 2005; 14: 2191–2199.
4. Sasieni P, Castanon A, Cuzick J. Screening and adenocarcinoma of the cervix. Int J Cancer. 2009; 125: 525–529.
5. Castellsagué X, Díaz M, de Sanjosé S, et al. Worldwide human papillomavirus etiology of cervical adenocarcinoma and its cofactors: implications for screening and prevention. J Natl Cancer Inst. 2006; 98: 303–315.
7. Kato T, Watari H, Takeda M, et al. Multivariate prognostic analysis of adenocarcinoma of the uterine cervix treated with radical hysterectomy and systematic lymphadenectomy. J Gynecol Oncol. 2013; 24: 222–228.
8. Lee YY, Choi CH, Kim TJ, et al. A comparison of pure adenocarcinoma and squamous cell carcinoma of the cervix after radical hysterectomy in stage IB-IIA. Gynecol Oncol. 2011; 120: 439–443.
9. Eifel PJ, Burke TW, Morris M, et al. Adenocarcinoma as an independent risk factor for disease recurrence in patients with stage IB cervical carcinoma. Gynecol Oncol. 1995; 59: 38–44.
10. Nakanishi T, Ishikawa H, Suzuki Y, et al. A comparison of prognoses of pathologic stage Ib adenocarcinoma and squamous cell carcinoma of the uterine cervix. Gynecol Oncol. 2000; 79: 289–293.
11. Landoni F, Maneo A, Colombo A, et al. Randomised study of radical surgery versus radiotherapy for stage Ib-IIa cervical cancer. Lancet. 1997; 350: 535–540.
12. Farley JH, Hickey KW, Carlson JW, et al. Adenosquamous histology predicts a poor outcome for patients with advanced-stage, but not early-stage, cervical carcinoma. Cancer. 2003; 97: 2196–2202.
13. Gallup DG, Harper RH, Stock RJ. Poor prognosis in patients with adenosquamous cell carcinoma of the cervix. Obstet Gynecol. 1985; 65: 416–422.
14. Lea JS, Coleman RL, Garner EO, et al. Adenosquamous histology predicts poor outcome in low-risk stage IB1 cervical adenocarcinoma. Gynecol Oncol. 2003; 91: 558–562.
15. Look KY, Brunetto VL, Clarke-Pearson DL, et al. An analysis of cell type in patients with surgically staged stage IB carcinoma of the cervix: a Gynecologic Oncology Group study. Gynecol Oncol. 1996; 63: 304–311.
16. Korhonen MO. Adenocarcinoma of the uterine cervix. Prognosis and prognostic significance of histology. Cancer. 1984; 53: 1760–1763.
17. Hopkins MP, Schmidt RW, Roberts JA, et al. Gland cell carcinoma (adenocarcinoma) of the cervix. Obstet Gynecol. 1988; 72: 789–795.
18. Kilgore LC, Soong SJ, Gore H, et al. Analysis of prognostic features in adenocarcinoma of the cervix. Gynecol Oncol. 1988; 31: 137–153.
19. Angel C, DuBeshter B, Lin JY. Clinical presentation and management of stage I cervical adenocarcinoma: a 25 year experience. Gynecol Oncol. 1992; 44: 71–78.
20. Harrison TA, Sevin BU, Koechli O, et al. Adenosquamous carcinoma of the cervix: prognosis in early stage disease treated by radical hysterectomy. Gynecol Oncol. 1993; 50: 310–315.
21. Helm CW, Kinney WK, Keeney G, et al. A matched study of surgically treated stage IB adenosquamous carcinoma and adenocarcinoma of the uterine cervix. Int J Gynecol Cancer. 1993; 3: 245–249.
22. Shingleton HM, Bell MC, Fremgen A, et al. Is there really a difference in survival of women with squamous cell carcinoma, adenocarcinoma, and adenosquamous cell carcinoma of the cervix? Cancer. 1995; 76: 1948–1955.
23. Yasuda S, Kojima A, Maeno Y, et al. Poor prognosis of patients with stage Ib1 adenosquamous cell carcinoma of the uterine cervix with pelvic lymphnode metastasis. Kobe J Med Sci. 2006; 52: 9–15.
24. dos Reis R, Frumovitz M, Milam MR, et al. Adenosquamous carcinoma versus adenocarcinoma in early-stage cervical cancer patients undergoing radical hysterectomy: an outcomes analysis. Gynecol Oncol. 2007; 107: 458–463.
25. Meng YH, Li S, Hu T, et al. Clinical analysis of 132 cases of cervical adenosquamous carcinoma and cervical adenocarcinoma. Chin J Cancer. 2010; 29: 15–19.
26. Rudtanasudjatum K, Charoenkwan K, Khunamornpong S, et al. Impact of histology on prognosis of patients with early-stage cervical cancer treated with radical surgery. Int J Gynaecol Obstet. 2011; 115: 183–187.
27. Chen JL, Cheng JC, Kuo SH, et al. Outcome analysis of cervical adenosquamous carcinoma compared with adenocarcinoma. Acta Obstet Gynecol Scand. 2012; 91: 1158–1166.
28. Mabuchi S, Okazawa M, Kinose Y, et al. Comparison of the prognoses of FIGO stage I to stage II adenosquamous carcinoma and adenocarcinoma of the uterine cervix treated with radical hysterectomy. Int J Gynecol Cancer. 2012; 22: 1389–1397.
29. Tierney JF, Stewart LA, Ghersi D, et al. Practical methods for incorporating summary time-to-event data into meta-analysis. Trials. 2007; 8: 16.
30. Lau J, Ioannidis JP, Schmid CH. Quantitative synthesis in systematic reviews. Ann Intern Med. 1997; 127: 820–826.
31. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med. 2002; 21: 1539–1558.
32. Egger M, Davey Smith G, Schneider M, et al. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997; 315: 629–634.
33. Pekin T, Kavak Z, Yildizhan B, et al. Prognosis and treatment of primary adenocarcinoma and adenosquamous cell carcinoma of the uterine cervix. Eur J Gynaecol Oncol. 2001; 22: 160–163.
34. Flores-Luna L, Salazar-Martinez E, Escudero-De los Rios P, et al. Prognostic factors related to cervical cancer survival in Mexican women. Int J Gynaecol Obstet. 2001; 75: 33–42.
35. Costa MJ, McIlnay KR, Trelford J. Cervical carcinoma with glandular differentiation: histological evaluation predicts disease recurrence in clinical stage I or II patients. Hum Pathol. 1995; 26: 829–837.
0. For the complete list of references, please contact Jae-Weon Kim, MD. Email: firstname.lastname@example.org.
Cervical cancer; Adenosquamous carcinoma; Adenocarcinoma; Survival; Prognosis; Meta-analysis
Supplemental Digital Content
© 2014 by the International Gynecologic Cancer Society and the European Society of Gynaecological Oncology.
Highlight selected keywords in the article text.