Small cell neuroendocrine carcinoma of the uterine cervix (SNEC) is a rare malignancy that comprises less than 5% of all cases of cervical cancers.1,2 Small cell neuroendocrine carcinoma of the uterine cervix is a highly aggressive tumor associated with decreased survival even in patients with early-stage SNEC and negative lymph node.3 Several studies have emphasized the poor prognosis of patients with SNEC.1–6 Five-year survival rates of patients with SNEC vary from 0% to 51%,5,7 and long-term survival can be accomplished only in patients with early-stage disease.1,5,8,9 Moreover, patients with SNEC are more likely to have early nodal and distant metastases compared with those with other more common histologic types of cervical cancer.1–4,10
Prognostic factors of patients with SNEC have been variably reported. The International Federation of Gynecology and Obstetrics (FIGO) stage,1,3,5–7,11–13 lymph node metastasis,1,7,11 and age at diagnosis3,14 have previously been reported as important prognostic factors. However, some studies demonstrated no significant prognostic value of lymph node involvement3,5,12 and age.5,13
Because of the rarity of SNEC, the prognostic factors of patients with SNEC are still controversial and most previous studies had rather small sample sizes of patients with SNEC.1,3,5,11,13 Only few large-scale analyses of patients with SNEC have been reported; however, these studies derived from multiple institutional bases or from case collection from cancer registry databases or reported data from the literature.3,7,12 This may result in some limitations such as the variation in treatment policy or the lack of a central pathology review. Therefore, a large-scale study based on a single institution’s experience and uniform pathology review and treatment guidelines is warranted to identify prognostic factors in patients with SNEC. In this study, we performed a retrospective review to determine the clinicopathologic factors associated with survival in patients with SNEC treated in a university hospital setting.
MATERIALS AND METHODS
After approval of the ethics review board, all patients with histologically confirmed SNEC, who were diagnosed and treated at Chiang Mai University Hospital between January 1995 and October 2011, were identified. The histologic slides were reviewed to confirm the diagnosis of SNEC based on a consensus agreement of 2 pathologists using the standard morphologic criteria.15 Immunohistochemical stains for neuroendocrine markers (chromogranin, synaptophysin, or CD56) were used to confirm the morphologic diagnosis in selected cases. The presence of non–neuroendocrine carcinomatous component (mixed type) was recorded. The exclusion of metastatic small cell carcinoma from the other primary site was based on a clinicopathologic correlation. The information on tumor homology diagnosis was obtained from both biopsy and hysterectomy specimens. The biopsy specimens were used in patients with advanced-stage disease who did not undergo surgery, whereas the hysterectomy specimens were used in surgically treated patients to confirm tumor homology diagnosis. Identification of lymphovascular space invasion (LVSI) was based on the hematoxylin and eosin (H&E) stained sections using the criteria by Roman et al,16 with exclusion of equivocal results. Pathologic data were obtained from pathology reports. In cases where the pathologic data was incomplete, histopathologic slides were re-examined by a pathologist.
The clinical data were extracted from medical and cancer registry records. The dates of patients’ deaths were obtained from medical records and/or the registry of the Thai Ministry of Interior and all were updated last January 2012. The outcomes were cancer-related death or death from other noncancer causes. The definition of cancer-specific survival (CSS) was the time from the date of treatment to the date of cancer-related death, last follow-up, or censoring. The variables included age at diagnosis, FIGO stage, tumor homology, tumor size, lymph node involvement, number of positive lymph nodes, LVSI, parametrial involvement, depth of stromal invasion, and treatment.
All patients were staged according to FIGO stage. Early stage was defined as FIGO stages I to IIA and advanced stage as stages IIB to IVB. For surgically treated patients, we also assessed the postoperative pathologic classification according to the International Union Against Cancer TNM classification of malignant tumors. Regarding the treatment guidelines in our institution, patients with early-stage diseases were usually treated with primary radical hysterectomy with pelvic lymphadenectomy. The pathologic findings of primary treatment were used as indicators for the following individualized adjuvant therapy: radiation therapy (RT), concurrent chemoradiation therapy (CCRT), and chemotherapy (CT). Patients with advanced-stage diseases were typically treated with RT with or without CT.
Statistical analysis was performed using Stata software version 11. Concerning the probabilities of competing events, we considered approaching competing risk regression analysis for evaluating the independent prognostic factors affecting survival. In fact, patients had a chance to experience a competing event either cancer-related deaths or death from other causes over the follow-up time. These 2 competing types of deaths could not be considered absolutely independent in some cases. Adjusted CSS curves of the patients were evaluated from cumulative incidence curves. Wald test was used to compare median survival time. Prognostic factors with P values of less than 0.25 in univariable analysis were further assessed in multivariable analysis. All tests were 2-tailed, and P values of less than 0.05 were considered statistically significant.
Clinicopathologic Characteristics of the Patients
Overall, 130 patients with histologically confirmed SNEC during the study period met the study criteria. The patients’ ages ranged from 26 to 80 years, and the mean (SD) age was 44.3 (10.9) years. Of these, 71 patients had FIGO stage I, 37 patients had FIGO stage II, 16 patients had FIGO stage III, and 6 patients had FIGO stage IV disease. The mixed histologic pattern in addition to the SNEC component was mostly associated with adenocarcinoma (21/28 patients; 75.0%). The remaining cases included squamous cell carcinoma (4/28 patients; 14.3%) and adenosquamous carcinoma (3/28 patients; 10.7%). Clinicopathologic characteristics of the entire group patients are shown in Table 1.
Among the 70 surgically treated patients, 68 (97.1%) patients had a radical hysterectomy with pelvic lymph node dissection and/or paraaortic lymph node dissection and 2 (2.9%) patients had a simple hysterectomy without lymph node dissection. Of those patients who had lymph node dissection, 14 (20.6%) patients had lymph node involvement. Of these 14 patients, only 1 had paraaortic lymph node involvement, whereas the rest of them had pelvic lymph node involvement. Most of these patients received adjuvant therapy, which were CT, RT, or CCRT (Table 2).
Survival and Prognostic Factors
Overall 5-year survival was 45.3%, and 5-year CSS was 47.5%. Five-year CSS for patients with FIGO stage I was 63%, 54% for FIGO stage IIA, 26% for FIGO stage IIB, and 0% for FIGO stages III and IV. The 5-year CSS for patients with early and advanced stages were 63% and 18% (P < 0.001), respectively. The median overall survival and median CSS were 47.8 (95% confidence interval [CI], 24.7–200.1) and 58.1 months (95% CI, 28.6–200.1), respectively. The factors significantly associated with differences in median survival time for entire group were age at diagnosis, FIGO stage, and primary treatment (Table 1), whereas age at diagnosis, FIGO stage, pathologic stage, depth of stromal invasion, and lymph node involvement were prognostic factors for differences in median survival time among surgically treated patients (Table 2). At the time of last follow-up, approximately half of the patients had died (63/130 patients; 48.5%). The proportion of cancer-related deaths and death from other causes between the early- and advanced-stage groups was equivalent (Table 3).
We also determined whether adjuvant CT would be beneficial in patients treated by surgery. Of the 70 surgically treated patients, 54.2% (38 patients) received adjuvant CT. Cisplatin and etoposide (PE regimen) were most commonly used for CT regimen, accounting for 75.9% of patients. Of the remainder, 15.5% received other cisplatin combinations and 8.6% received cisplatin alone. Patients who received adjuvant CT had better survival rates than those who had surgery alone, those who received adjuvant RT, and those receiving adjuvant CCRT, with 5-year survival rates of 74.4%, 55.6%, 53.3%, and 30.1%, respectively (P = 0.041).
For multivariable analysis, early- and advanced-stage groups were analyzed separately because some pathologic data were available only in patients undergoing surgery that was mostly performed in patients with early-stage disease. The eligible variables further evaluated in multivariable analyses in the early-stage group included age at diagnosis, FIGO stage, tumor size, lymph node involvement, depth of stromal invasion, and treatment, whereas variables in the advanced stage group included age at diagnosis, FIGO stage, tumor homology, and treatment. Recurrence and progression were adjusted in both early- and advanced-stage models for special focus on outcomes of deaths.
In the early-stage model, age older than 60 years at diagnosis and deep stromal invasion were significant prognostic factors for poorer survival with hazards ratio (HR) 4.9 (95% CI, 1.5–15.8; P = 0.007) and HR 2.9 (95% CI, 1.3–6.9; P = 0.011), respectively (Table 4, Fig. 1). Lymph node involvement, a strong prognostic factor in univariable analysis, failed to reach significance in multivariable analysis. In the advanced-stage model, age at diagnosis remained the important prognostic factor. Patients aged younger than 45 years and those older than 60 years had significantly poorer survival compared with those patients aged 45 to 60 years, with HR 3.4 (95% CI, 1.1-10.5; P = 0.035) and HR 9.9 (95% CI, 2.8-35.4; P < 0.001), respectively. International Federation of Gynecology and Obstetrics stage was also found to be a strong prognostic factor. Patients with FIGO stage IV had significantly poorer survival compared with patients with FIGO stage IIB (HR, 7.4; P = 0.024; Table 4, Fig. 2).
Small cell neuroendocrine carcinoma of the uterine cervix is an aggressive subtype of cervical cancer associated with poor prognosis. In this study, we analyzed 130 patients with SNEC from our single university hospital with uniform pathology review and treatment guidelines to estimate prognostic factors of survival. To our knowledge, this study had the largest number of the patients with SNEC from a single institution.
The median CSS from this study was 58.1 months, much longer than what was reported in an earlier study with a median CSS of 24.8 months.7 Moreover, the 5-year CSS in patients with early-stage disease in our study was also slightly better than what was reported in previous studies (63% and ranged from 37 % to 52%, respectively), whereas 5-year CSS in patients with advanced-stage disease was similar (18% in our study compared with 10% to 25 % in previous studies).7,12 The explanation of this better survival rate in patients with early-stage disease is probably correlated with the treatment modality that included radical hysterectomy and adjuvant CT that was also reported in previous studies.11,17
Age at diagnosis older than 60 years was an important prognostic factor regardless of stage at diagnosis. Among the patients with advanced-stage disease, age at diagnosis younger than 45 years was also associated with decreased survival compared with those aged 45 to 60 years. Several studies have also shown a trend of decreased survival with increasing age.7,9,12 One study revealed that patients aged 50 years and younger had better overall survival than those aged older than 50 years (P = 0.020).14 Another study reported a similar finding to ours, when compared with patients aged 75 year and older, patients aged 65 to 74 years and those aged 45 to 54 years had significantly lower risks of cancer death (HR, 0.34; P = 0.011 and HR, 0.40; P = 0.029, respectively), whereas patients aged 55 to 64 years and those younger than 45 years had similar risks of cancer death.3
Several studies have consistently revealed the impact of FIGO stage on survival. The more advanced stage is correlated with a higher risk of poorer survival.1,3,5,7,12,13,18,19 Although FIGO stage has been identified as a strong prognostic factor, all reference groups used in previous studies were either FIGO stage I or stage I to IIA (early stage) to compare the effects of FIGO stage on survival in multivariable analysis.3,5,7,12 The Surveillance, Epidemiology, and End Results (SEER) study showed that compared with stage I, patients with SNEC with stage IV had a 4.6 times higher risk of death (P < 0.001) and patients with stage III had a higher risk 2.4 times (P = 0.017).3 Our study showed that FIGO stage was a significant prognostic factor only in the advanced-stage group with a 7.4 times higher risk of cancer death among patients with FIGO stage IV compared with patients with FIGO stage IIB.
As shown in previous studies, deep stromal invasion tended to be associated with poor survival.7,13,19,20 In our study, deep stromal invasion was also a prognostic factor in patients with early-stage disease. Patients with deep stromal invasion had a 2.9 times higher risk of cancer death (P = 0.011). This result is consistent with a previous study reporting that patients with depth of stromal invasion of two thirds or greater had poorer survival than those with a lesser extent of invasion.21 Although other reports also showed a similar trend for decreased survival among patients with deep stromal invasion, no statistical significance was observed.7,13
Although many previous studies emphasized lymph node involvement as an important prognostic factor in patients with early-stage SNEC,1,8,11,17,19 lymph node metastasis failed to reach a statistical significance in our study. However, there was a trend of higher risk of cancer-related death among patients who had lymph node involvement compared with those who had not. Similarly, Wang et al7 reported that lymph node involvement was not a prognostic factor for patients who underwent primary surgery. However, lymph node involvement including that diagnosed by clinical/imaging investigation was predictive of the prognosis when patients of all stages were considered.
Because SNEC has an aggressive behavior, patients with SNEC had poorer survival than those with squamous cell carcinoma or adenocarinoma.3,4,8,22,23 Although the optimal treatment of patients with SNEC with early-stage disease has not been established,1,5,24 multimodalities of treatment is recommended.5,11,17 In our multivariable analysis, we found no significant difference in benefits of treatment modalities. However, when adjuvant therapy was solely considered among patients with early-stage disease who underwent surgery, we found that those who received adjuvant CT had significantly better survival than those who had surgery alone, those who received adjuvant RT, and those receiving CCRT (P = 0.041). Similarly, other studies have reported the benefits of adjuvant CT after radical hysterectomy in patients with early-stage disease.11,17 The beneficial effect of PE regimen of CT, the most commonly used regimen in our study, was also observed in previous studies,9,12,13 and this might be related to the relatively high survival rate of patients with SNEC in our series. In contrast, an earlier study found no significant difference in survivorship among patients with early-stage disease treated with different modalities (radical surgery vs surgery plus adjuvant CT vs surgery plus adjuvant CT and RT).19
This study had a large number of the patients with SNEC and had homogenous data, which strengthened the study. We also recognize a study limitation owing to its retrospective design. The medical records were sometimes incomplete, resulting in the clinical or imaging investigation of tumor involvement in the advanced stage could not be evaluated in this study.
In conclusion, age at diagnosis was an important prognostic factor in both groups of patients with early- and advanced-stage disease. Deep stromal invasion was a poor prognostic factor in patients with early stage disease and FIGO stage IV in the advanced-stage group. Adjuvant CT may be beneficial for surgically treated patients with early-stage disease.
The authors are pleased to acknowledge the National Research University Project under Thailand’s Office of the Higher Education Commission and the Graduate School of Chiang Mai University in Thailand for financial support.
The authors acknowledge Associate Professor Surapan Khunamornpong from the Department of Pathology in Chiang Mai University and Dr Mette Søgaard from the Department of Clinical Epidemiology in Aarhus University Hospital in Denmark for their opinion on this article.
1. Viswanathan AN, Deavers MT, Jhingran A, et al. Small cell neuroendocrine carcinoma of the cervix: outcome and patterns of recurrence. Gynecol Oncol. 2004; 93c: 27–33.
2. Crowder S, Tuller E. Small cell carcinoma of the female genital tract. Semin Oncol. 2007; 34: 57–63.
3. Chen J, Macdonald OK, Gaffney DK. Incidence, mortality, and prognostic factors of small cell carcinoma of the cervix. Obstet Gynecol. 2008; 111: 1394–1402.
4. Vinh-Hung V, Bourgain C, Vlastos G, et al. Prognostic value of histopathology and trends in cervical cancer: a SEER population study. BMC Cancer. 2007; 7: 164.
5. Chan JK, Loizzi V, Burger RA, et al. Prognostic factors in neuroendocrine small cell cervical carcinoma: a multivariate analysis. Cancer. 2003; 97: 568–574.
6. Huang CY, Chen YL, Chu TC, et al. Prognostic factors in women with early stage small cell carcinoma of the uterine cervix. Oncol Res. 2009; 18: 279–286.
7. Wang KL, Chang TC, Jung SM, et al. Primary treatment and prognostic factors of small cell neuroendocrine carcinoma of the uterine cervix: a Taiwanese Gynecologic Oncology Group study. Eur J Cancer. 2012; 48: 1484–1494.
8. Alfsen GC, Kristensen GB, Skovlund E, et al. Histologic subtype has minor importance for overall survival in patients with adenocarcinoma of the uterine cervix: a population-based study of prognostic factors in 505 patients with nonsquamous cell carcinomas of the cervix. Cancer. 2001; 92: 2471–2483.
9. Zivanovic O, Leitao MM Jr, Park KJ, et al. Small cell neuroendocrine carcinoma of the cervix: analysis of outcome, recurrence pattern and the impact of platinum-based combination chemotherapy. Gynecol Oncol. 2009; 112: 590–593.
10. Weed JC Jr, Graff AT, Shoup B, et al. Small cell undifferentiated (neuroendocrine) carcinoma of the uterine cervix. J Am Coll Surg. 2003; 197: 44–51.
11. Chang TC, Lai CH, Tseng CJ, et al. Prognostic factors in surgically treated small cell cervical carcinoma followed by adjuvant chemotherapy. Cancer. 1998; 83: 712–718.
12. Cohen JG, Kapp DS, Shin JY, et al. Small cell carcinoma of the cervix: treatment and survival outcomes of 188 patients. Am J Obstet Gynecol. 2011; 203: 347.e1–347.e6.
13. Lee JM, Lee KB, Nam JH, et al. Prognostic factors in FIGO stage IB-IIA small cell neuroendocrine carcinoma of the uterine cervix treated surgically: results of a multi-center retrospective Korean study. Ann Oncol. 2008; 19: 321–326.
14. Hoskins PJ, Swenerton KD, Pike JA, et al. Small-cell carcinoma of the cervix: fourteen years of experience at a single institution using a combined-modality regimen of involved-field irradiation and platinum-based combination chemotherapy. J Clin Oncol. 2003; 21: 3495–3501.
15. Albores-Saavedra J, Gersell D, Gilks CB, et al. Terminology of endocrine tumors of the uterine cervix: results of a workshop sponsored by the College of American Pathologists and the National Cancer Institute. Arch Pathol Lab Med. 1997; 121: 34–39.
16. Roman LD, Felix JC, Muderspach LI, et al. Influence of quantity of lymph-vascular space invasion on the risk of nodal metastases in women with early-stage squamous cancer of the cervix. Gynecol Oncol. 1998; 68: 220–225.
17. Boruta DM 2nd, Schorge JO, Duska LA, et al. Multimodality therapy in early-stage neuroendocrine carcinoma of the uterine cervix. Gynecol Oncol. 2001; 81: 82–87.
18. Lan-Fang L, Hai-Yan S, Zuo-Ming Y, et al. Small cell neuroendocrine carcinoma of the cervix: analysis of the prognosis and role of radiation therapy for 43 cases. Eur J Gynaecol Oncol. 2012; 33: 68–73.
19. Tian WJ, Zhang MQ, Shui RH. Prognostic factors and treatment comparison in early-stage small cell carcinoma of the uterine cervix. Oncol Lett. 2012; 3: 125–130.
20. Peng P, Ming W, Jiaxin Y, et al. Neuroendocrine tumor of the uterine cervix: a clinicopathologic study of 14 cases. Arch Gynecol Obstet. 2012; 286: 1247–1253.
21. Bermúdez A, Vighi S, García A, et al. Neuroendocrine cervical carcinoma: a diagnostic and therapeutic challenge. Gynecol Oncol. 2001; 82: 32–39.
22. McCusker ME, Cot TR, Clegg LX, et al. Endocrine tumors of the uterine cervix: incidence, demographics, and survival with comparison to squamous cell carcinoma. Gynecol Oncol. 2003; 88: 333–339.
23. Sevin BU, Method MW, Nadji M, et al. Efficacy of radical hysterectomy as treatment for patients with small cell carcinoma of the cervix. Cancer. 1996; 77: 1489–1493.
24. Wang KL, Yang YC, Wang TY, et al. Neuroendocrine carcinoma of the uterine cervix: a clinicopathologic retrospective study of 31 cases with prognostic implications. J Chemother. 2006; 18: 209–216.