Current Obstetrics and Gynecology Reports

, Volume 2, Issue 2, pp 86–93

Cervical Adenocarcinoma in Situ: Update and Management

  • Stephan Polterauer
  • Alexander Reinthaller
  • Reinhard Horvat
  • Elmar Joura
  • Christoph Grimm
Management of HPV and Associated Cervical Lesions (B Barrette, Section Editor)

DOI: 10.1007/s13669-013-0039-6

Cite this article as:
Polterauer, S., Reinthaller, A., Horvat, R. et al. Curr Obstet Gynecol Rep (2013) 2: 86. doi:10.1007/s13669-013-0039-6

Abstract

Adenocarcinoma in situ (AIS) of the uterine cervix is caused by infection with high-risk human papillomavirus and is the recognized precursor of invasive adenocarcinoma of the cervix. Because most AIS lesions are caused by HPV 16/18 infection, prophylactic HPV vaccination is an important step toward prevention of AIS, potentially reducing the incidence of invasive adenocarcinoma. Nonetheless, at the moment the incidence of AIS and invasive adenocarcinoma continues to increase, especially among young women when fertility preservation is an issue. Both diagnosis and treatment of AIS is challenging, because AIS lesions frequently extend into the endocervical canal, making detection and complete excision difficult. Hysterectomy remains the standard treatment for AIS. Selected patients, who wish to preserve fertility, with clear margins and negative ECC after initial conization are potential candidates for conservative treatment. If margins are involved after initial conization or ECC results are positive, the risk of residual or recurrent AIS and invasive adenocarcinoma of the cervix is considerably high. In these women, repeat surgery should be performed. For women, who do not undergo hysterectomy, long-term follow-up is recommended.

Keywords

ACISAISAtypical glandular cellsCervical adenocarcinoma in situEndocervical canalManagement of HPVHuman papillomavirusHPV

Introduction

Cervical adenocarcinoma in situ (AIS) arises in the glandular epithelium of the uterine cervix and is the recognized precursor to invasive adenocarcinoma [1•]. The incidence rate of AIS is much lower compared with cervical intraepithelial neoplasia (CIN) [2]. In contrast to CIN and squamous cervical cancer (SCC), which declined during the past decades, the incidence of AIS and invasive adenocarcinoma continues to increase, especially among young women [2]. Multicentricity with foci high in the endocervical canal and skip lesions are common findings in AIS. Adequate therapy can prevent occurrence of invasive adenocarcinoma in many cases. This article reviews current data and recommendations for the management of patients diagnosed with AIS.

Pathogenesis

AIS lesions usually originate in the squamocolumnar junction of the transformation zone and may extend into the endocervical canal [3]. Infections with human papillomavirus (HPV) 16/18 account for approximately 90 % of AIS and invasive adenocarcinomas [1•]. Interestingly, infection with multiple HPV types can be found in approximately 40 % of lesions [1•, 4]. It has been postulated that oncogenic virus types infect reserve cells of the transformation zone that are committed to glandular differentiation, eventually leading to the proliferation of atypical glandular cells (AGC). In addition, long-term oral contraceptive (OC) use may contribute to the pathogenesis of AIS in some women [4]. It has been postulated that OCs enhances HPV exposure to endocervical cells by causing ectopy and/or directly promoting carcinogenesis.

Incidence

The incidence rate of AIS is 1.25 per 10,000 women-years and therefore is much less common compared with cervical intraepithelial neoplasia (CIN) [2]. Of note, the incidence of AIS and invasive adenocarcinoma has increased within past decades, especially among young women with a mean age of approximately 35–40 years [2]. This is in contrast to the incidence of CIN, which has decreased since the implementation of cervical cancer screening and prophylactic vaccination [57]. It is likely that the increasing AIS incidence can be attributed in part to screening, reflecting increased recognition and, thus, detection of AIS lesions that previously were undiagnosed or were not categorized as such [8]. Nonetheless, life-style factors elevating estrogen levels, such as long-term OS use, have been postulated to increase the incidence of AIS [4].

Prevention

Vaccination

The finding that adenocarcinoma of the cervix is predominantly HPV-16- and HPV-18-related and that HPV vaccination has significantly prevented cervical adenocarcinoma in situ in clinical trials indicates that vaccination is the most promising measure to reduce the incidence of adenocarcinoma of the cervix and its precursor AIS [9, 10]. The quadrivalent vaccine against HPV types 6, 11, 16, and 18 showed 100 % vaccine efficacy for HPV 16/18-related AIS and 27 % vaccine efficacy for AIS lesions related to other HPV types [11]. Data from the Victorian Cervical Cytology Registry between 2003 and 2009 recorded a decrease in the incidence of histopathologically defined high-grade CIN and AIS lesions after the introduction of the Australian HPV vaccination program [9]. Because most AIS lesions are HPV16/18-related, prophylactic HPV vaccination might reduce future incidence of invasive adenocarcinoma. Although HPV vaccination is an important step toward cervical cancer prevention, it does not degrade the need for routine cervical cancer screening [1•, 12].

Screening - Cytology

As patients with squamous lesions, patients with AIS are typically asymptomatic, In rare cases vaginal bleeding can be present before diagnosis. Therefore, AIS is usually suspected based on pathological findings in the setting of cervical cancer screening and both AGC and abnormal squamous cells (ASC) can be indicative for AIS. AGC are found in approximately 5 % of all abnormal smears with a mean reporting rate of only 0.4 % of all smears in the United States in 2003 [13, 14]. AGC describe cellular changes in glandular cells that exceed those expected in a benign reactive or reparative process yet are not abnormal enough to be clearly neoplastic. AGC on cervical cytology usually originate from the glandular epithelium of the endocervix or endometrium. Although a variety of glandular lesions, including malignancies, are associated with AGC, CIN is the most common significant finding identified in women with AGC [15]. AGC represent a much greater risk for cervical precancer and cancer than ASC or low-grade squamous intraepithelial lesion (LSIL). Both high-grade squamous and glandular cervical epithelial abnormalities, as well as cervical and endometrial cancers, may be found in the follow-up to an AGC Pap result, and many of these serious lesions are found at much higher rates than with any other Pap reading other than HSIL and cancer. In most areas with well-established cervical screening programs, there has been no obvious reduction in the incidence of or mortality due to adenocarcinoma, despite substantial decreases in the incidence of and mortality due to squamous carcinoma [16]. The location and distribution of neoplastic lesions in the endocervical canal, as well as the difficulty in traditional cytologic assessment of glandular cells, lead to the high false-negative rate of conventional cytology and the limited colposcopic visualization of these lesions, altogether contributing to the challenges in diagnosing AIS [1•, 16, 17]. Recent studies suggest that additional methods, such as immunocytochemical staining for both p16INK4a and Ki-67 or immunohistochemical staining of insulin-like growth factor-II mRNA-binding protein 3 expression, seem to be useful methods for the diagnosis of cervical AIS and invasive adenocarcinoma [18, 19]. A recent study showed that primary HPV screening in combination with cytology triage was more sensitive than conventional cytology in detecting AIS lesions [20].

Initial Workup of Patients

Either glandular or squamous cytologic abnormalities can indicate presence of AIS. Cytologically, women with AGC require further evaluation for premalignant conditions of the cervix, uterus, and rarely, ovary. In 2006, the American Society for Colposcopy and Cervical Pathology (ASCCP) recommended that all women with a diagnosis of AGC of any subcategory, or AIS, should be evaluated by colposcopic examination and endocervical curettage (ECC) [21•, 22]. In addition, endometrial sampling is recommended in combination with colposcopy and colposcopically guided biopsies (CGB) of all suspicious lesions and endocervical sampling in women aged 35 years and older with all subcategories of AGC and AIS. Endometrial sampling also is recommended for women younger than age 35 years with clinical symptoms, such as vaginal bleeding, suggesting they may be at risk for neoplastic endometrial lesions [21•]. Pelvic examination and vaginal ultrasound are useful in the diagnostic workup. Due to the well-established link between HPV 16/18 and AIS, the ASCCP recommends HPV DNA testing if it has not already been performed at the initial screening [22].

Histopathology

The histologic diagnosis of AIS is based on results of CGB, ECC, endometrial sampling, and/or cone biopsy. AIS consistently has been characterized by typical histologic features, such as preservation of normal glandular architecture coupled with an alteration involving parts or all of the surface and/or glandular epithelium, nuclear enlargement, coarse chromatin, small single or multiple nucleoli, increased mitotic activity, and variable stratification of nuclei and absence of invasion (Fig. 1) [23]. A variety of subtypes of AIS have been described based on cytoplasmic characteristics, including endocervical, intestinal, endometrioid, and mixed adenosquamous types [3]. AIS often is multicentric and lesions can reach high up the endocervical canal. Skip lesions are a common finding and can be missed on punch or cone biopsy even in the presence of negative resection margins [24, 25].
https://static-content.springer.com/image/art%3A10.1007%2Fs13669-013-0039-6/MediaObjects/13669_2013_39_Fig1_HTML.gif
Fig. 1

Adenocarcinoma in situ of the uterine cervix (H&E). Typical features, such as increased mitotic figures, nuclear enlargement, coarse chromatin, small size, and variable stratification of nuclei, are present

Biology

Reasonable evidence supports the existence of AIS as a precursor lesion for invasive adenocarcinoma. Unlike SCC, earlier neoplastic precursors to AIS and adenocarcinoma are not well characterized. AIS is diagnosed in the population 10 to 20 years younger than those with invasive adenocarcinoma, with a mean age of approximately 35–40 years [3, 5, 26]. The natural history of AIS and adenocarcinoma is not as well defined as CIN and SCC largely related to their lower incidences. No studies have addressed progression rate from AIS to invasive carcinoma, time to malignant transformation, or spontaneous regression rate [3].

Management

Cervical AIS remains a therapeutic challenge, because a large proportion of patients confronted with this diagnosis wants to preserve fertility. Thus, treatment needs to be planned based on histologic findings at the time of conization considering important factors, such as the patient’s wish to preserve fertility. Careful counseling of patients opting for fertility preservation is crucial, because these patients need to be willing to accept a higher risk of recurrent AIS and invasive adenocarcinoma. Figure 2 provides an algorithm for the management of patients diagnosed with cervical AIS.
https://static-content.springer.com/image/art%3A10.1007%2Fs13669-013-0039-6/MediaObjects/13669_2013_39_Fig2_HTML.gif
Fig. 2

Treatment algorithm for patients diagnosed with cervical adenocarcinoma in situ. CGB = colposcopy-guided biopsy; ECC = endocervical curettage; *using HPV test, PAP, colposcopy ± CGB, and ECC, if test reveal positive results reexcision should be performed

Conization

If AIS is diagnosed by CGB or ECC an excisional biopsy is required in all women before any subsequent management decisions [21•, 27]. With respect to the type of conization, different modalities are acceptable. Some institutions would rather perform loop electrosurgical excision procedures (LEEP), whereas others opt for cold knife conization (CKC) or laser conization. Several studies have investigated the influence of type of conization on margin status and results are controversial. Overall, the rate of incomplete resection varies between 27 % and 51 % [25, 27, 28•, 29]. In most studies, cold knife conization is found to be the superior technique to achieve negative margins in the cone specimen. A lower rate of involved resection margins and even a decreased risk of recurrence after CKC compared with LEEP has been reported previously [25, 30, 31]. The higher rate of negative resection margins after CKC that has been reported in these (retrospective) studies might partially be explained by the fact that LEEP was used more frequently as a diagnostic procedure before hysterectomy compared with CKC being an intended therapeutic procedure. Other studies report equivalent rates of negative margins and recurrence after LEEP and CKC in selected young patients aged <30 years who desire future fertility [32•]. The authors suggest that patients with small colposcopic lesions and/or a small cervix should be treated with LEEP and other patients treated with CKC. Studies comparing laser cone biopsy with CKC show comparably low rates of involved resection margins [29, 33].

There are several potential arguments for CKC to be the preferred method in the treatment of AIS. Surgical margins show less thermal artifacts after CKC and typically more tissue is removed by CKC compared with laser and electrosurgical conization. In addition, LEEP and laser cone biopsies often are fragmented making evaluation of margins difficult. This occurs typically with larger lesions, in which several loop passes have to be performed to excise the lesion completely [24]. We recommend performing CKC or large loop excision of the transformation zone (LLETZ) when AIS is diagnosed by CGB or ECC, for large AIS lesions, and especially when conization is performed as a definitive treatment in patients desiring to preserve fertility. Nevertheless, in many cases AIS diagnosis will be made upon histology of the cone biopsy specimen after standard LEEP has been performed in the absence of suspicion for glandular lesion before conization.

Subsequent to cone biopsy typically ECC is performed from above the endocervical resection margin. ECC is a useful tool to predict residual AIS and identify candidates who require a second excision. ECC results provide additional information that can be used for counseling women considering fertility preservation [27, 34•].

Fertility-Sparing Treatment

For selected women who wish to preserve fertility, conservative management, i.e., conization followed by surveillance, might be an acceptable option [21•]. Of note, it is crucial that these women are counseled adequately and informed of the risk for recurrent AIS and adenocarcinoma. Where conservative management is to be undertaken, several criteria have to be met. The cone biopsy specimen must be thoroughly sectioned and examined by a gynecological pathologist, and the patient should remain under specialist supervision during the period of follow-up. The patient needs to understand the importance of continuing long-term cytologic surveillance and must be compliant with the follow-up program [29]. Conservative management following conization with negative margins and negative ECC appears relatively safe for patients with AIS desiring fertility. The overall risk of recurrent AIS appears to be approximately 5–10 % [24, 35••]. The overall risk of invasive adenocarcinoma appears to be approximately 0.1-5.2 % [35••]. After childbearing is completed, definitive hysterectomy should be recommended. Long-term follow-up for women who do not want to take the surgical risks of hysterectomy seems acceptable especially when follow-up test results were negative for a longer period of time. We recommend careful postconization follow-up for patients with clear margins every 6 months and after two rounds of negative test results, annual screening may be resumed. In accordance to follow-up guidelines after treatment for squamous lesions, annual screening should be continued at least 20 years or until age 65 years [21•]. When resection margins are involved or in case of positive ECC at time of initial conization, reexcision should be considered to increase the likelihood of complete excision of AIS and to reduce the risk of recurrence [12, 47]. Nevertheless in the 2006 guidelines, the ASCCP stated that reevaluation after 6 months using a combination of cervical cytology, HPV DNA testing, and colposcopy with endocervical sampling is an acceptable alternative to reexcision for patients who want to avoid reexcision [21•]. If tests reveal negative results, surveillance is an option and careful long-term follow-up is recommended. In case of positive test results reexcision should be performed. We recommend hysterectomy rather than surveillance for women with a positive conization margin following two or more conizations performed.

AIS Diagnosed in Pregnancy

According to ASCCP guidelines, pregnant women with any type of abnormal cervical cytology should be referred for colposcopy and CGB [21•]. Once invasive cancer is excluded, conservative management of squamous lesions in pregnancy is considered safe. Pregnant women are rarely diagnosed with AIS. Diagnostic conization is recommended during pregnancy to exclude invasive disease and when presence of invasive disease affects treatment planning. However, optimal management of AGC and AIS during pregnancy is not well established and no randomized, controlled trials have been performed. Some institutions recommend conization for all patients diagnosed with AIS in pregnancy, because histological verification of AIS by punch biopsy is not reliable and the postpartum regression rate cannot be determined precisely [3638]. When conization is performed during pregnancy, the optimal timeframe appears to be the second trimester, preferably between 14 to 20 weeks of gestation to avoid contractions or hemorrhage. The largest report comprises 11 patients diagnosed with AIS during pregnancy [36]. Of these, five patients underwent uncomplicated CKC at 14 to 19 weeks gestation. One of the six patients who underwent postpartum CKC required radical hysterectomy for undiagnosed stage IB1 cervical adenocarcinoma. All patients delivered at term, and the authors concluded that management of cervical AIS during pregnancy by early second trimester CKC is safe for mother and fetus.

Hysterectomy

Hysterectomy remains the preferred standard treatment for women diagnosed with AIS [21•]. AIS is frequently present in multifocal areas and as well as skip lesions that are not contiguous and can be several millimeters away from areas free of AIS. Therefore, negative margins on a diagnostic excisional specimen do not necessarily mean that the lesion has been completely excised. Because of these considerations, hysterectomy continues to be the treatment of choice for AIS in women who have completed child-bearing. However, AIS often occurs in women who wish to maintain their fertility. A number of studies have now clearly demonstrated that an excisional procedure is curative in the majority these patients. When hysterectomy is performed, all approaches—vaginal, minimally invasive and open—are possible options. Nevertheless, morcellement of the uterus has to be avoided. Total extrafascial hysterectomy is recommended; oophorectomy does not seem to be necessary in premenopausal women with AIS. Conization with frozen-section analysis was suggested as an effective technique for intraoperative triage of patients to immediate simple or radical hysterectomy when invasive disease is suspected before conization [39]. When invasive adenocarcinoma is diagnosed intraoperatively, pelvic lymph node dissection should be performed by a gynecologic oncologist according to international guidelines [40]. Bilateral salpingo-oophorectomy should be discussed because of the risk of ovarian metastases, which is higher in women with cervical adenocarcinoma compared with SCC. A recent study reported outcomes of women who were treated with fertility-sparing radical abdominal trachelectomy with early-stage adenocarcinoma of the cervix. Oncologic safety was demonstrated, although one quarter of all selected cases required hysterectomy or postoperative chemoradiation [41].

Posttreatment Follow-Up

Long-term follow-up is mandatory for women who do not undergo hysterectomy. Well-established follow-up guidelines for patients after conization for AIS are missing. For patients who were treated with conization for squamous lesions (CIN2+), the ASCCP recommends follow-up using a combination of cytology and colposcopy at 6-month intervals [21•]. Colposcopy with CGB of suspicious lesions and endocervical sampling is recommended for women who are HPV DNA-positive or have a repeat pathologic cytology result. If the HPV DNA test is negative or if two consecutive repeat cytology tests are negative, routine screening for at least 20 years commencing at 12-month intervals is recommended [21•]. Recent studies showed that late recurrences of AIS and invasive cancer can occur after conservative management for AIS [28•]. Therefore, extended follow-up is crucial to identify recurrent disease. A recent study closely followed 119 women after conservative treatment for AIS. Additional therapeutic procedures were necessary in 58 % during a mean follow-up of 40.9 months. At study conclusion, 5.9 % showed persistent disease, whereas 6.7 % had progressed to invasive adenocarcinoma. Positive HPV test results were shown to be the strongest predictive factor for disease recurrence (odds ratio [OR] = 3.74) independent of margin status [42].

Risk of Residual AIS After Conization

Initial conization margin status seems to be the critical factor in predicting the success of conservative management of cervical AIS. Patients with positive margins after conization are significantly more likely to have residual AIS. However, even if negative margin status is achieved, there is a definite risk of recurrent AIS. The meta-analysis by Salani et al. included a total of 1,278 patients. A repeated excisional procedure was performed in 607 patients. In total, 54 of 266 patients (20.3 %) with negative margins showed residual AIS in the secondary specimen, whereas 180 of 341 patients (52.8 %) were found to have residual disease when resection margins were positive. ECC is typically performed subsequent to cone biopsy. In contrast to CIN, ECC performed during conization frequently shows positive results. Recent studies have described that ECC results are positive in 3-42 % of AIS cases performed during conization [27, 31, 43, 44]. ECC results seem to be an additional useful important predictor for residual disease after initial conization. The positive predictive value of intraoperative ECC for residual AIS ranges between 76.9 % and 100 % [27, 31, 43, 44].

Risk of Invasive Adenocarcinoma After Conization

Invasive adenocarcinoma is a relatively rare finding at definitive treatment after conization when resection margins are clear. In their meta-analysis, Salani et al. analyzed 607 patients who underwent a secondary procedure (repeat conization or hysterectomy) after the initial conization. In total, 3.5 % of patients were found to have invasive adenocarcinoma in the hysterectomy specimen. The diagnosis of adenocarcinoma was made in 17 of 266 (6.3 %) patients with positive resection margins at initial conization compared with 4 of 341 (1.5 %) patients with negative margins on the initial excision [35••]. Including all 1,278 patients and all treatment strategies, 5.2 % of patients with positive and 0.7 % with clear margins were ultimately found to harbor invasive adenocarcinoma of the cervix [35••]. Another recent report was designed to identify patients who can be safely managed with an extrafascial hysterectomy based on predictors of invasion in the conization specimen when resection margins were involved. Data of patients who had definitive surgical management for cervical AIS following conization with positive margins and/or positive endocervical curettage were identified. The risk of invasive adenocarcinoma (stage IA1-IB1) was 12 % in a cohort of women with cervical AIS at initial conization [34•]. In the setting of positive margins after conization, AIS patients with positive ECC and pathologic suspicion of invasion at time of conization had a 33 % risk for invasive disease, whereas in absence of these two features, none of the patients had invasive disease at hysterectomy [34•].

Risk of Recurrent AIS or Invasive Adenocarcinoma After Conservative Management

Salani et al. identified 671 patients who were treated conservatively after initial conization. A total of 15 of 573 patients (2.6 %) with negative margins compared with 19 of 98 patients (19.4 %) with involved margins at initial conization were noted to have recurrence within the follow-up period. Of the 671 patients, 8 (1.2 %) were found to develop invasive adenocarcinoma—6 (0.9 %) and 2 (0.3 %) with positive and negative resection margins after initial conization, respectively [35••]. A smaller European multicenter study observed 166 women after conservative treatment for AIS without immediate reexcision [28•]. Of these, 69 patients had secondary, delayed surgical therapy within 36 months after initial therapy revealing no residual disease in 78.2 %, AIS in 10.1 %, and invasive adenocarcinoma in 11.6 % of patients [28•].

In addition to margin status, HPV status after conization seems to be the most important predictive factor for recurrent disease. In a large study, positive HPV test results after conization were shown to be the strongest predictive factor for disease recurrence (OR = 3.74) and was independently of margin status [42].

Risk of Recurrence After Hysterectomy

Local recurrences are rarely diagnosed in patients with AIS after hysterectomy [45, 46]. There are no clear guidelines for post hysterectomy follow-up for women diagnosed with AIS. According to guidelines for patients treated with hysterectomy for CIN2 or CIN3, it seems reasonable to recommend follow-up with cytology and HPV every 6 months until two negative results are obtained [12, 47]. After two negative results screening with cytology alone, every 1–3 years for 20 years after the initial posttreatment surveillance period seems reasonable [12, 47]. The role of HPV testing in this situation has not been clarified but might be a useful triage method.

Conclusions

Cervical AIS remains a therapeutic challenge to the gynecologist, particularly with regard to ascertaining which patients may be managed conservatively to maintain fertility. Hysterectomy remains the standard treatment for AIS. Selected patients who wish to preserve fertility can be alternatively managed with conservative treatment (i.e., conization). CKC, LEEP/LLETZ, and laser excision are appropriate procedures for cone biopsy. Young patients with clear margins and negative ECC after initial conization are potential candidates for conservative treatment. When margins are involved after initial conization or ECC results are positive, the risk of residual or recurrent AIS and invasive adenocarcinoma of the cervix are considerably high. In these patients with positive resection margins or positive ECC, repeat cone biopsy should be performed to increase the chance of complete resection. Long-term follow-up is recommended for women who do not undergo hysterectomy. Because most AIS lesions are caused by HPV 16/18 infection, prophylactic HPV vaccination is an important step toward prevention of AIS and should reduce the incidence of invasive adenocarcinoma.

Acknowledgments

This study was supported by the Karl Landsteiner Institute or General Gynecology and Experimental Gynecologic Oncology, Vienna, Austria.

Conflict of Interest

Stephan Polterauer declares he has no conflict of interest.

Alexander Reinthaller declares that he has no conflict of interest.

Reinhard Horvat declares that he has no conflict of interest.

Elmar Joura is a board member and consultant for Merck. He has received payment and/or honoraria from Merck, Sanofi Pasteur MSD, GlaxoSmithKline, and Roche Diagnostics.

Christoph Grimm declares that he has no conflict of interest.

Copyright information

© Springer Science+Business Media New York 2013

Authors and Affiliations

  • Stephan Polterauer
    • 1
  • Alexander Reinthaller
    • 1
    • 2
  • Reinhard Horvat
    • 3
  • Elmar Joura
    • 1
  • Christoph Grimm
    • 1
  1. 1.Department of Obstetrics and Gynecology, Division of General Gynecology and Gynecologic Oncology, Comprehensive Cancer Center Vienna, Gynecologic Cancer UnitMedical University Vienna, AustriaViennaAustria
  2. 2.Karl Landsteiner Institute for General Gynecology and Experimental Gynecologic OncologyViennaAustria
  3. 3.Department of Pathology, Division Gynecologic PathologyMedical University of ViennaViennaAustria