Exploring conservative management for cervical intraepithelial neoplasia grade 2 in organised cervical cancer screening programmes: a multicentre study in Italy

Introduction

Aims of the organised population-based screening are the detection and treatment of the high-grade lesions CIN2 and CIN3 (cervical intraepithelial neoplasia grades 2 and 3), and early detection of invasive cancers. CIN2 has a higher spontaneous regression rate (40%–60%, depending on age) and a lower progression capacity (around 15%) than CIN3 (30%–40% for both).1 Excisional treatment (recommended for both lesions)2 is highly effective in curing the lesions, but can pose risks such as preterm birth, premature rupture of membranes, low birth weight.3 4 In the last decade, a few countries have developed specific guidelines for the conservative management of CIN25–8; young age (below 25–30 years) is among the most used criteria for selecting suitable cases, as regression probability is higher in younger than older women,5 the risk of progression after treatment is associated to older age at initial excision,9 and women of childbearing age (15–45 years) benefit most. Active surveillance is performed also in countries lacking guidelines6 7 10; a recent Dutch nationwide survey disclosed that 41% of the women with a histological diagnosis of CIN2 did not undergo excision within 3 months after biopsy.11 Lack of compliance to the follow-up visits and the long-term risk of cervical cancer might constitute barriers to the adoption of active surveillance.7 In order to provide guidelines and recommendations, several studies are ongoing to evaluate the clinical outcomes of untreated CIN2 under active surveillance.12 13

Besides searching for factors and biomarkers14–16 able to correctly stratify (at the time of diagnosis) the CIN2 lesions at higher probability of regression (or progression), it is very important also to understand the women’s acceptability and the organisational impact on screening programmes.7 Women’s attitudes have been less frequently investigated; in Australia, active surveillance was preferred to surgery by 79% of women aged 25–40 years who were presented a hypothetical scenario of CIN2 diagnosis.17

In Italy, organised cervical cancer screening programmes started in the late 1990s, are performed according to national recommendations, by call–recall invitation of all women aged 25–64 years, and are operated at a regional level. HPV (human papillomavirus)-based testing is being implemented for women aged 30–64 years. Women negative for high-risk HPV (hrHPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, 68) are called for a new screening round 5 years later. In case of hrHPV positivity, cytology triage is performed on the same liquid-based sample. Women with cytological alterations (atypical squamous cells of undetermined significance or worse, ASC-US+) are referred to immediate colposcopy; those with normal cytology undergo 1 year recall for hrHPV testing repeat, with colposcopy in case of persistent hrHPV positivity, and return to screening in case of viral clearance. Women 25–29 years-old undergo cytology testing every 3 years, with immediate colposcopy for ASC-US+.18–20 The aim of this paper is to describe the acceptability of conservative management by women with a CIN2 lesion for its implementation within organised cervical cancer screening programmes, by analysing the reasons of exclusion from or of refusal to participate to our study, and compliance to follow-up.

Methods

Results

During the enrolment phase (April 2019–October 2021) of this prospective multicentre study, a total of 640 cases of CIN2 lesions were diagnosed among the women attending the four screening programmes, which contributed different numbers of cases to the study (range: 40–182, median: 48.5). The mean and median ages of the entire cohort were 37 and 35 years, respectively.

An overview of the entire cohort is provided in figure 1. The distribution of cases by age class, according to eligibility and acceptance to enter the study, is summarised in table 1.

Request permission

Overall, according to the predefined criteria, 228 (35.6%) women (median age 46) were excluded and 412 (64.4%) were eligible, with a statistically significant difference according to the age group (p<0.001); 93 eligible women refused to enter the study (median age 34). The reasons of exclusion and refusal to enter the study are reported in table 2.

Out of the 228 excluded women, 122 (53.5%) were older than 46 years, 90 (39.5%) had gynaecological characteristics deemed as ‘high-risk’ (ie, not visible TZ, previous high-grade lesion, pluricentric lesions, not completely visible lesion), while other reasons (ongoing pregnancy, other gynaecologic reasons, impossibility to comply with the study protocol, selection error) accounted for 7% (16 cases) only. The 45 years age limit, therefore, ranked first (53% of the cases) among the exclusion criteria, followed by a previous high-grade lesion (17.5%). A closer look at the characteristics of the women older than 45 years disclosed that 80.3% had age as the only exclusion criterion, while also other exclusion criteria were registered in 24 (19.7%) cases; in particular, a not visible TZ and/or lesion were recorded in 17 (13.9%) cases, and a previous high-grade lesion in 7 (5.7%) cases.

Considering the whole cohort, the distribution by age class of the gynaecological reasons of exclusion was as follows. The squamo-columnar junction was not visible only in women older than 34 years, with a frequency of 8% (12/152) among women 35–55 years old and of 27.3% (9/33) after that age. A CIN2+ lesion had been previously diagnosed in 47 (7.3%) women, without significant differences by age. An endocervical lesion was recorded in 26 women, with the highest frequency (18/63, 28.5%) among those aged 35–45 years. Pluricentric lesions were found in 20 cases, with a decreasing frequency by age, from 38% (8/21) in the 25–29 age class to 0% in women older than 45 years.

Among the 93 eligible women not included in the study, the refusal was mostly motivated by personal reasons (66/93; 71%), followed by external causes (16/93; 17%). The most frequent personal reasons have been desire to become pregnant, anxiety, willingness to be managed by the personal gynaecologist, lack of interest in study participation, refusal to undergo the additional sampling for biomarkers’ evaluation, difficulty in complying with the follow-up visits, fear due to the COVID-19 pandemic and serious concomitant extragenital morbidity. Residence elsewhere and language problems were the most frequently reported external causes. Willingness to be treated immediately within the screening programme accounted for less than 5% (20/412; 8 refused participation, 12 refused continuing active surveillance during follow-up). No feedback after the counselling visit occurred for 11 (3%) patients. Acceptance was (not significantly) higher among 25–34 years old than among older women.

The enrolment period, initially forecasted to last 12 months, was extended to 30 months, as a consequence of the COVID-19 pandemic that caused a 2 months’ suspension (during March and April 2020) of the invitations to a new screening round, followed by a very slow recovery of the participation rates. The final cohort of cases included in the study is made up of 319 women, with mean and median ages of 34 and 33 years, respectively, and homogeneously distributed among age groups (p=0.8105). The follow-up ended in October 2023.

During the scheduled 24 months of follow-up, 125 (39%) of the enrolled women exited the study for personal reasons, the majority (77/125, 62%) after the 6/12 months’ visit. These women were recalled to solicit their attendance; a large number declared they had referred to other gynaecological facilities, often accounting the COVID-19 pandemic as the cause. We compared these non-compliant women to the entire cohort. It emerged that 67% of the non-adherent women belonged to the three centres which cumulatively enrolled 43% of the cohort, while no difference by age was observed. We are actively searching additional information on these women, to understand their clinical outcome up to 24 months from enrolment (and correlate it to the biomarkers’ results at baseline). Willingness to be treated for reasons other than protocol was expressed by 12 (4%) women (3 at baseline, 4 at 6 months, 3 at 12 months and 2 at 18 months, respectively); the histological diagnosis was CIN2 in 6, CIN3 in 3 and negative in 3 cases, respectively. The remaining 182 (57%) women fully adhered to the protocol.

In terms of workflow for the screening programme, the only additional activity performed is the sampling for the biomarkers’ evaluation at the counselling visit, while all the follow-up visits have the same timing in use after excisional treatment. In terms of additional costs for the execution of the biomarkers’ analyses, only the partial HPV 16/18 genotyping is actually routinely available, while all the others are not in routine use.

Discussion

To evaluate the acceptability of CIN2 conservative management by the women and the impact of active surveillance on the screening programme, we analysed the age distribution of the study predefined criteria, the reasons of refusal to enter the study by eligible women and the adherence to the study protocol of those enrolled. This analysis was conducted within a multicentre prospective trial that involved 640 women with a diagnosis of CIN2 detected within organised cervical cancer screening programmes. On the basis of our predefined inclusion and exclusion criteria, that determined eligibility, and the refusal by eligible women, half of the cases could finally enter the study.

Almost one fifth (122/640, 19%) of the women were excluded because they were older than 46 years. This age limit was set by considering the benefit of avoiding the surgical treatment in women of childbearing age and the lower rate of spontaneous regression in older women. Indeed, most studies on the conservative management of CIN2 have been conducted in <45 years old women,6 9 12 14 17 and less frequently16 22 in women of all ages. The CIN2 regression rates recorded in these studies differed by gynaecological or viral (ie, HPV genotype) characteristics, but were not related to the woman’s age. On the other hand, age limits have an impact on the proportion of women to whom active surveillance for CIN2 would finally be applied.

Among the women eligible to the study, 77.4% accepted the wait-and-see strategy. This figure is very close to the 79% acceptance rate observed among 1638 women residing in Australia given a diagnosis of CIN2 in a hypothetical scenario23 and confronted with the choice between immediate surgical treatment or active surveillance. These women were randomised according to alternate terminology used to describe the regression of the lesion and the effects of the surgical procedure. Both the degree of understanding of the information provided and the perception of benefits and risks of surgery (overestimated and underestimated, respectively) had an influence on the final decision; this highlights the importance of communication and the need to assess patient understanding to close the loop of communication. It has been reported that a CIN2 diagnosis, as well as the indication to undergo a surgical procedure, can affect the psychosocial well-being of the affected woman. A cross-sectional study investigating the prevalence of depression/anxiety and the health-related quality of life in women with a CIN2 diagnosis who underwent either surgery or conservative management found that the conservative subgroup exhibited higher scores for health-related quality of life compared with women undergoing surgery, and that all CIN2 affected women showed a higher prevalence of depression/anxiety than those of the healthy group.24

To safely implement active surveillance and avoiding the risk of missing a progressive lesion, compliance is a critical aspect. In our study, full adherence to the 24 months’ study protocol was recorded for less than 60% of the women. Nonetheless, we registered a higher adhesion (74%) in the largest centre than in the others (where the small numbers might have partly influenced the figures), and a 70% overall rate at 12 months. Indeed, since in a recent large study22 90% of regression and progression outcomes were recorded within 12 months of follow-up, this represents a positive finding. To effectively evaluate the likelihood of patients’ compliance and their education about the importance of monitoring visits, training on communication should be provided to the dedicated healthcare personnel.

The follow-up protocol in use in our study for the CIN2 active surveillance has a number of visits comparable to that of the women treated as per routine practice. Therefore, no immediate impact on the screening programme’s organisation occurred. After completion of the follow-up, the overall impact determined by the clinical outcome will be evaluated; women with a regressive CIN2 lesion would have spared the surgical treatment, while women who underwent delayed treatment would have had an increased number of visits. Moreover, the analysis of the timing of lesions’ regression and the results of the biomarkers under study will allow the definition of protocols capable to balance the number of visits in relation to the woman’s characteristics and the biomarkers’ results, possibly taking into account also lifestyle factors that could reflect their cervical precancer risk.25 To evaluate the cost-effectiveness of implementing a protocol for CIN2 active surveillance within an organised cervical cancer screening, the final data of our study will be analysed along with other similar studies, in order to formulate a national guideline. A recent analysis conducted in Germany on the healthcare costs for women 18–45 years old with CIN diagnosis, over a 2-year follow-up period, disclosed higher costs for those undergoing treatment (€ 1020) than for those under active surveillance (€ 328).26

In the future, the epidemiology of HPV infection and preneoplastic and invasive lesions will be modified by the combined protective effect of HPV vaccination in adolescent girls and cervical screening by primary HPV testing27 28; in particular, the women vaccinated before age 15 will not develop CIN2 lesions caused by HPV types prevented by vaccination (mainly HPV16 and HPV18). In our study, we could not evaluate whether a correlation between the vaccination status and the acceptance of conservative management could possibly exist since the vast majority (97.5%) of the women involved in our study were born before 1996 and only 1 of the 102 for whom the information is available had been vaccinated at 15 years of age.

Conclusions

Women are willing to accept a conservative management for CIN2 lesions within cervical screening programmes. Among eligible not enrolled women, refusal for personal reasons and willingness to be treated accounted for 71% and 9%, respectively. On the basis of our experience, careful selection of the patients with CIN2 lesions (with attention to communication for their commitment to adherence to follow-up) and specific training to all the healthcare operators (within and outside the screening programme) are key aspects to ensure a safe active surveillance.