Article Text
PDF
Abstract
Objective The management and surveillance of glandular pre-cancerous lesions of the uterine cervix present distinct challenges compared with squamous lesions, primarily attributed to the lower effectiveness of diagnostic methods such as cytology or colposcopy. This study aimed to investigate the long-term safety of fertility-sparing treatment for adenocarcinoma in situ and microinvasive adenocarcinoma of the cervix, while identifying factors associated with recurrence, with a particular emphasis on the role of human papillomavirus (HPV) testing.
Methods We retrospectively reviewed data from all patients with histopathologically confirmed adenocarcinoma in situ or microinvasive cervical adenocarcinoma who received treatment at a single center between 2002 and 2023. The study involved the examination of consecutive surgical specimens and the follow-up details. Factors associated with recurrence were assessed in a subgroup of patients with available long-term follow-up data (at least 6 months).
Results In total, 143 patients (112 with adenocarcinoma in situ and 31 with adenocarcinoma) were included in the analysis. Among the 86 patients who underwent fertility-sparing treatment, the recurrence rate was 9% (12% for adenocarcinoma in situ and 4% for adenocarcinoma) during a median follow-up period of 56.6 months (range 7–179). No patients who were HPV negative experienced recurrence during the follow-up period. In contrast, among patients who were HPV positive, the recurrence rate was 38%. Additionally, HPV 16/18 positivity displayed a notable association with a higher risk of recurrence compared with the other high-risk genotypes, although this difference did not reach statistical significance (83% vs 10%; p=0.083, log-rank).
Conclusion Our retrospective study demonstrated a significant association between the risk of recurrence and HPV status during the follow-up period. Consequently, long-term follow-up utilizing HPV testing and genotyping appears to be a secure alternative to a hysterectomy.
- Adenocarcinoma
- Gynecologic Surgical Procedures
- Uterine Cervical Neoplasms
Data availability statement
Data are available upon reasonable request.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Cervical glandular lesions (adenocarcinoma in situ and early stage cervical adenocarcinoma) commonly affect young women and can significantly disrupt their reproductive plans. The potential for overlooking recurrence is substantial, as isolated dysplastic foci may persist in situ even following a cone biopsy with apparently clear resection margins. Given the inherent limitations of diagnostic methods during post-treatment monitoring, many patients are often advised to consider an expeditious hysterectomy as a precautionary measure.
WHAT THIS STUDY ADDS
We aimed to investigate the long-term safety of follow-up after fertility-sparing treatment for glandular dysplasia of the uterine cervix. We evaluated the effectiveness of both cytology and human papillomavirus (HPV) testing, and determined the optimal duration for surveillance among women who chose not to undergo a hysterectomy.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
We discovered that incorporating HPV testing and genotyping into the follow-up provides a secure alternative to a hysterectomy. Nonetheless, patients must adhere to consistent follow-up assessments spanning an extended timeframe.
Introduction
Pre-invasive glandular lesions of the cervix (adenocarcinoma in situ) frequently affect young patients and can significantly impact their fertility plans.1 Furthermore, compared with squamous lesions, the diagnosis, treatment, and follow-up of glandular lesions require a specific approach. Given that the majority of these lesions are localized in the endocervix, the sensitivity of cytology and colposcopy is limited.2 Furthermore, patients with positive margins and/or positive endocervical curettage after excisional surgery are at a significantly higher risk of residual or recurrent disease, necessitating re-excision.3 This approach is justified by the discontinuous spread of the dysplastic glandular epithelium, making it possible to miss part of the lesion during an incomplete resection.4
The higher risk of recurrence/persistence after a fertility-sparing treatment is reflected in the current American Society for Colposcopy and Cervical Pathology Risk-Based Management Consensus Guidelines.5 For women with adenocarcinoma in situ and a preserved uterus, careful long-term surveillance is recommended. A hysterectomy is often regarded, by many national standards, as the definitive treatment for glandular lesions in patients with completed reproductive plans. However, even after a hysterectomy, continued follow-up for an additional 25 years after the initial diagnosis is advised due to the risk of vaginal dysplasia.6
In comparison to adenocarcinoma in situ, there are limited data available on the prognosis of conservatively treated microinvasive adenocarcinoma. Recently published data from the SCCAN project demonstrated a very low risk of recurrence, suggesting similarities between these lesions and adenocarcinoma in situ.7 The aim of our study was to identify factors associated with recurrence in patients after fertility-sparing treatment for adenocarcinoma in situ and microinvasive adenocarcinoma, aiming to optimize the follow-up strategy and eliminate unnecessary treatment.
Methods
Patients
We conducted a retrospective study analyzing data from all patients with adenocarcinoma in situ or Fédération Internationale de Gynécologie et d'Obstétrique (FIGO) 2018 stage IA cervical adenocarcinoma consecutively treated at a single site, the Oncogynecology Center, Department of Obstetrics, Gynecology, and Neonatology, General University Hospital in Prague.
Patients were included if they met the following criteria: (1) histopathologically confirmed adenocarcinoma in situ or FIGO stage IA invasive adenocarcinoma; (2) primary management by conization or trachelectomy; (3) no previous diagnosis of invasive cervical cancer. Patients with concomitant invasive squamous lesions were excluded from the study. The study design is illustrated in Figure 1. The protocol was approved by the Institutional Review Board of the General University Hospital in Prague, Czech Republic in 2023. Due to the retrospective nature of the study, the need for informed consent signed by the patient was waived by the Institutional Review Board.
Study scheme; AIS: adenocarcinoma in situ; FIGO: Fédération Internationale de Gynécologie et d'Obstétrique.
Treatment
Patients were referred to our center based on abnormal screening cytology results. A specialist colposcopist performed colposcopic examinations prior to surgery. Excisional treatment was performed according to the topography of the lesion and the surgeon’s preferences either as electrosurgical conization using Fisher Cone Biopsy Excisor or Needle Excision of Transformation Zone; cold knife conization, or a simple vaginal trachelectomy. Endocervical curettage was performed as part of the procedure in all patients. Surgical lymph node staging (sentinel lymph node (SLN) biopsy and/or pelvic lymph node dissection) was completed in patients with adenocarcinoma FIGO stage IA1 with lymphovascular space invasion (LVSI) or stage IA2.
In cases where dysplasia was present in the resection margin or material from endocervical curettage (intraepithelial or invasive cancer), a repeated excision of the cervix or a hysterectomy was performed (see Figure 1). Patients with a preserved uterus, who had follow-up data available beyond 6 months after the excision procedure, were included in the long-term follow-up analysis, forming the ‘Fertility Sparing Cohort’.
Pathological Evaluation
Cone biopsy specimens were sectioned in 2–3 mm intervals radially until no tissue was left. Sections were stained with haematoxylin-eosin. Immunohistochemical analysis using the p16 antibody (clone JC2, dilution 1:200, Diagnostic Biosystems) was performed on selected blocks with adenocarcinoma in situ or adenocarcinoma. The p16 expression was classified as block positive (diffuse staining of tumor cells in the nuclear and/or cytoplasmic compartment), or negative (focal/patchy or absent staining). Histopathological classification followed the current WHO classification of female genital tumors.8
Follow-Up
Patients were scheduled for follow-up every 6 months after the fertility-sparing procedure. Each visit included an expert colposcopic examination, conventional cytology smear, and human papillomavirus (HPV) DNA test (Cobas 4800, Roche Molecular Systems, Pleasanton, USA). Patients with negative HPV tests and two consecutive negative cytology smears were referred back for routine screening. Subsequently, they were further followed according to national recommendations, which consist of an annual cervical smear. All women were advised to undergo a hysterectomy after completing their reproductive plans, in line with the national guidelines. Patients who received follow-up care outside our institution were interviewed by telephone. Recurrence was defined as the detection of adenocarcinoma in situ or adenocarcinoma (applicable to both adenocarcinoma in situ and adenocarcinoma patient groups) in a specimen from a repeated cone biopsy or hysterectomy performed during the follow-up.
Data Analysis
Standard measures of summary statistics were used to describe the primary data, including the relative and absolute frequencies, and the arithmetic mean with the SD. The Fisher exact test was used to compare frequency counts of demographic and clinical categorical variables. Continuous variables were compared using the Mann-Whitney U-test, a non-parametric alternative to an independent samples t-test. Disease-free survival was modeled using the Kaplan–Meier method with a log-rank test. It was defined as the time interval between the first conization and the date of disease recurrence. Receiver operating curves and the area under the curve (AUC) were used to compare the efficacy of the follow-up examinations (cytology vs HPV testing) to identify occurrence of recurrence. Receiver operating curve analysis was further extended by calculating the sensitivity, specificity, as well as positive and negative predictive values. The statistical significance level was set to α=0.05 for all statistical tests.
Results
General Cohort
The analysis included a total of 143 patients, with 112 patients having adenocarcinoma in situ and 31 patients with FIGO stage IA adenocarcinoma. The characteristics of the overall cohort and subgroups of patients with adenocarcinoma in situ and adenocarcinoma are summarized in Table 1. Treatment predominantly involved the Fisher Cone Biopsy Excisor (76%) or electrofrequency needle (21%), with the cold knife conization and simple vaginal trachelectomy performed in 2% and 1% of cases, respectively.
Patient characteristics
One hundred and four patients underwent a single conization, and 39 women underwent multiple cone biopsies to attain a clear resection margin. Twelve patients opted for a hysterectomy instead of a repeated conization. Free resection margins were finally achieved in all patients in our cohort. The general cohort comprised 25 patients with FIGO stage IA1 and six patients with stage IA2 adenocarcinoma. Five patients with adenocarcinoma underwent systematic pelvic lymph node dissection, while 10 underwent SLN biopsy only. All patients were found to be node negative in the final histology. Adenocarcinoma in situ lesions were more frequently associated with concomitant squamous cell lesions (cervical intraepithelial neoplasia) compared with adenocarcinoma (66% vs 36%; p=0.003).
Of the 33 patients who underwent a hysterectomy within the first 6 months after conization with free resection margins, 11 (33%) had adenocarcinoma in situ detected in the hysterectomy specimen (termed ‘skip lesion’). Among the 69 patients who underwent a hysterectomy in the general cohort (see Figure 1), 45 patients had follow-up data available for more than 2 years. During this period, we documented no cases of recurrence.
Fertility-sparing Cohort
The fertility-sparing cohort consisted of 86 patients with a preserved uterus and follow-up data available for a period longer than 6 months. During the follow-up, there was no significant difference in HPV status between the adenocarcinoma and adenocarcinoma in situ subgroups (18% vs 20% in HPV-positive adenocarcinoma in situ and adenocarcinoma subgroups, respectively; p=0.32). Patients had their first HPV-positive test, on average, 28.9 months after conization. There was no significant difference between patients with HPV 16/18 and those with HPV non 16/18 positivity (35.6 vs 25.8, p=0.86 respectively).
There was no significant difference in disease-free survival between patients initially diagnosed with adenocarcinoma in situ or adenocarcinoma (p=0.24, log-rank test) during a median follow-up period of 56.6 months (range 7–179) (see Figure 2). Among the eight patients who developed recurrence (9%), one patient who initially presented with adenocarcinoma experienced a recurrence of adenocarcinoma in situ, and of the remaining seven patients with recurrences who originally presented with adenocarcinoma in situ, two women developed adenocarcinoma and five developed adenocarcinoma in situ.
Kaplan-Meier estimates of disease-free survival in patients after fertility sparing treatment. (a) Adenocarcinoma in situ and microinvasive carcinoma of the uterine cervix; (b) According to the HPV status in the follow-up.
During the follow-up, 16 patients had abnormal cytology smear results, with 14 having low-grade (atypical squamous cells of undetermined significance, low-grade squamous intraepithelial lesion, or atypical glandular cells not otherwise specified) and two having high-grade (atypical squamous cells—cannot exclude high-grade squamous intraepithelial lesion (HSIL) or high-grade squamous intraepithelial lesion) cytology. Both patients with high-grade cytology results experienced recurrence.
None of the patients who tested HPV negative experienced recurrence during the follow-up. Among the patients with recurrence, five (63%) were HPV 16/18 positive, one (13%) was positive for other high-risk HPV types, and the HPV status of the remaining two (25%) was unknown. Both cases of invasive recurrence (25%) were associated with HPV 18 positivity during the follow-up period. Five out of six patients (83%) who tested positive for HPV 16/18 and one out of 10 patients (10%) who tested positive for HPV non 16/18 experienced recurrence during follow-up. HPV 16/18 positivity was consistently observed in all six patients during multiple subsequent visits, while nine out of 10 patients who tested positive for HPV non 16/18 became negative in the subsequent visit (indicating transient infection).
HPV positivity during the follow-up and the presence of concomitant squamous lesions were the only variables associated with an increased risk of recurrence (p<0.001 and p=0.031; log-rank test). On the other hand, neither a repeated cone biopsy (p=0.97, log-rank test) nor age (p=0.42, p=0.43, and p=0.14 for ages above 40, 35, and 30 years, respectively, log-rank test) were identified as risk factors for developing recurrence. HPV 16/18 positivity showed a stronger association with the likelihood of recurrence than other high-risk genotypes (p<0.001 and p=0.051, respectively, log-rank test, see Figure 2), but the difference in disease-free survival between these two categories was not statistically significant (p=0.08, log-rank test). Patients treated conservatively experienced more recurrences than those who underwent a hysterectomy (recurrence rate 9% vs 0%, p=0.035, log-rank test). This trend was evident in patients with a positive HPV test (p=0.001, log-rank test) and particularly notable in patients with HPV 16/18 genotypes (p<0.001, log-rank test).
Table 2 summarizes the efficacy of the tests used in follow-up. The sensitivity, specificity, and parameters of the receiver operating curve for normal versus abnormal cytology were 43%, 78%, 0.606, p=0.36; low grade (including normal) versus high-grade cytology were 29%, 100%, 0.643, p=0.39; HPV positive versus HPV negative were 100%, 82%, 0.911, p=0.001; and HPV 16/18 versus other HR types+HPV negative were 83%, 98%, 0.909, p=0.001. Receiver operating curves are visualized in Figure 3.
Receiver operating curve analysis of tests used in the follow-up of the patients after fertility sparing treatment of adenocarcinoma in situ and microinvasive adenocarcinoma; (a) HPV testing: positive versus negative; (b) HPV testing: 16/18 versus other genotypes and negative; (c) Cytology: high-grade versus low-grade (including normal); (d) Cytology: abnormal versus normal; AUC: area under the receiver operating curve.
Efficacy of follow-up methods
Discussion
Summary of Main Results
Our study provides evidence supporting the long-term safety of the fertility-sparing treatment in patients with adenocarcinoma in situ or microinvasive cervical adenocarcinoma. During the median follow-up period of 57 months, we observed a recurrence rate of only 9%. We found no significant difference in the prognosis of patients with conservatively treated adenocarcinoma in situ or microinvasive adenocarcinoma. The HPV test emerged as an important tool during the follow-up, exhibiting high sensitivity as a triage method to identify patients requiring more intensive monitoring or additional interventions. Conversely, repeated documentation of HPV negativity strongly correlated with a favorable prognosis in our cohort.
Results in the Context of Published Literature
HPV positivity during the follow-up after the fertility-sparing treatment is the strongest predictor associated with recurrence of glandular lesions.9 Follow-up based on HPV testing has higher sensitivity and specificity than cytology. However, the best results are reached with co-testing.10 We found that the HPV test outperformed cytology, which displayed poor sensitivity even when using the atypical squamous cells of undetermined significance (ASC-US) threshold (83% and 98% vs 43% and 78%, see Table 2). Furthermore, there were no instances of recurrence among those patients in our cohort who tested HPV negative. In contrast, five out of six patients with HPV 16/18 positivity experienced recurrence, with both invasive recurrences being associated with HPV 18.
While all six patients who tested HPV 16/18 positive had persistent infections during the follow-up period, the majority (nine of 10) of non 16/18 infections were transient. Transient HPV infections were not associated with recurrence (p=0.002). These results strongly suggest that HPV testing and genotyping can serve as a highly effective tool for risk triage during the follow-up after fertility-preserving treatment of glandular lesions. The relatively low recurrence rate of 9% strongly contrasts with a high proportion of patients exhibiting residual adenocarcinoma in situ in the specimen from a second excisional procedure or hysterectomy, known as ‘skip lesion’. The literature estimates this proportion to be 20%.4 In our series, we found 11 (33%) such patients out of 33 who underwent a hysterectomy within the 6 months following conization with free margins. Since no patient had a positive margin in the fertility-sparing cohort (see Figure 1) and no repeated conization was performed on patients with clear margins, it can be assumed that the proportion of patients with a skip lesion was initially comparable (one third). However, this fact did not translate into a higher recurrence rate. We can only speculate that most of these residual lesions likely regressed spontaneously, and the remaining ones were effectively identified through HPV testing. We found no significant difference in the recurrence rates between patients with adenocarcinoma in situ and microinvasive adenocarcinoma (11.5% vs 4.0%, p=0.24, log-rank test). Notably, only one patient with adenocarcinoma experienced recurrence and, interestingly, the recurrence was classified as adenocarcinoma in situ. The low recurrence rate of microinvasive adenocarcinoma was consistent with the findings of the recent SCCAN trial 7.
In accordance with the guidelines of The American Society for Colposcopy and Cervical Pathology, long-term follow-up involving co-testing and colposcopy with endocervical sampling is recommended after fertility-sparing treatment for adenocarcinoma in situ. Recurrences are most commonly observed within the 3 years after the first cervical biopsy 9 and, therefore, co-testing should be performed at 6 month intervals for 36 months, followed by annual screening for the subsequent 2 years.6 However, recent data highlight that all women treated for cervical intraepithelial neoplasia continue to face a persistent risk of developing HPV-dependent cancer, possibly more than 20 years after the primary surgery .11Our data support this finding, as one patient (13%) experienced a recurrence 135 months after the initial cone biopsy. All the remaining seven recurrences (88%) occurred within the first 5 years (21–62 months, mean 40.6±14.3).
Surveillance for at least 25 years after diagnosis is recommended for patients after hysterectomy.6 This prolonged follow-up is justified by the increased risk of developing dysplasia in the vaginal stump. However, it is important to note that our study did not confirm this association as we found no cases of recurrence after hysterectomy in our cohort of 45 patients (see Figure 1).
Strengths and weaknesses
The primary limitations of our study stem from its retrospective design and limited number of patients in the fertility-sparing cohort. This, together with the low recurrence rate, results in a relatively small absolute number of recurrences, which may affect the statistical calculations. The most notable example is the log-rank comparison of recurrences in patients with HPV 16/18 versus HPV non 16/18 positivity, which was not statistically significant (p=0.083) despite a substantial difference in the recurrence rates (83% vs 10%). However, we believe that these trends would remain consistent with a larger sample size, as our findings are not in contradiction with the cited literature. Nevertheless, it is important to highlight that all patients with adenocarcinoma in situ and adenocarcinoma treated within the designated timeframe at a single center were incorporated, leading to a uniformly treated group devoid of selection bias.
, which was not statistically significant (p=0.083) despite a substantial difference in the recurrence rates (83% vs 10%). However, we believe that these trends would remain consistent with a larger sample size, as our findings are not in contradiction with the cited literature. Nevertheless, it is important to highlight that all patients with adenocarcinoma in-situ and adenocarcinoma treated within the designated timeframe at a single center were incorporated, leading to a uniformly treated group devoid of selection bias.
Within the first 6 months, 33 patients opted for a hysterectomy after conization with clear margins (see Figure 1). These patients were excluded from the fertility-sparing cohort not due to the study’s protocol, but based on their personal preferences. This fact could potentially impact the composition of the fertility-sparing cohort. However, this allowed us to investigate the prevalence of residual foci of adenocarcinoma in situ following conization with free resection margins (‘skip lesions’). Another weakness is the inadequate data on HPV status and genotype before the treatment. Therefore, we were unable to determine whether the recurrences were associated with HPV persistence or newly acquired infections. Such an analysis could potentially lead to a more effective triage of patients, guiding them towards the recommendation of undergoing a hysterectomy.
Implications for Practice and Future Research
Our data support the potential for long-term conservative follow-up in patients after fertility-sparing treatment for both adenocarcinoma in situ and adenocarcinoma. However, hysterectomy as definitive treatment appears to be safer when HPV status is not taken into account (p=0.035). This underscores the importance of HPV testing and genotyping in the follow-up after conservative treatment, as it helps to effectively classify women according to risk of recurrence (p=0.001 for HPV positive and p<0.001 for HPV 16/18 positive). In the case of HPV negativity, patients can be allowed to pursue their reproductive plans without the need for early referral for hysterectomy.
Conclusions
In summary, our study reaffirms the low recurrence rate of adenocarcinoma in situ and microinvasive adenocarcinoma after fertility-sparing treatment. We have demonstrated the crucial role of HPV testing as the most effective tool for follow-up monitoring. The minimal risk of recurrence in patients who were HPV negative indicates that long-term follow-up can be considered as a viable alternative to hysterectomy, even in women with completed reproductive plans. However, it is imperative to counsel these patients about the importance of regular follow-up examinations to ensure early detection of any potential issues. Although most recurrences tend to occur within the first 5 years, it is essential to maintain vigilance beyond this period, as the risk of recurrence remains elevated. Persistent HPV 16/18 infection strongly supports the consideration of hysterectomy or more frequent surveillance. Furthermore, our study underscores the significance of HPV 18 positivity during follow-ups, as it may be associated with the development of invasive recurrences.
Data availability statement
Data are available upon reasonable request.
Ethics statements
Patient consent for publication
Acknowledgments
Supported by MH CZ-DRO-VFN00064165 and MH CZ-DRO-VFN64165. We extend our special thanks to Mr Kendall Z.H.K. for providing editorial support.
Footnotes
Contributors LD: guarantor, study design, collected data, manuscript writing; PF: collected data; KN: pathology consultation, manuscript writing; DC: manuscript writing, scientific advisor; MS: collected data; TZ: collected data, scientific advisor; MK: statistics advisor; LF: collected data; TF: collected data; DB: collected data; JS: guarantor, manuscript writing, collected data, scientific advisor.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.