Clarifying the Equivocal Diagnosis of Cervical Intraepithelial Neoplasia 2: Still a Work in Progress

To the Editor:

Kremer et al¹ reported on a clinical trial (CONCERVE Study) to monitor women diagnosed with biopsy-proven cervical intraepithelial neoplasia (CIN) grade 2 (CIN2) or 3 (CIN3; CIN2/3), rather than immediate treatment of the abnormality by surgical excision. The aim of the study was to reflect clinical practice to gain insight into the real-world impact of a wait-and-see policy for CIN2/3. They found a high percentage of CIN2/3 regressed clinically, defined as histologically confirmed absence of CIN2 or worse (CIN2+) or an human papillomavirus (HPV)–negative clinician-collected sample with normal cytology in case histology was absent and no histologic diagnosis of CIN2+ during further follow-up. CIN2/3 was more likely to regress if it was negative for FAM19A4/miR124-2 methylation compared with those who were positive (74.7% v 51.4%, respectively, P = .014), leading the authors to suggest that FAM19A4/miR124-2 methylation testing could support a wait-and-see policy in women with CIN2/3.

However, conflation of CIN2 and CIN3, study biases, and their definition of regression may lead to misinterpretation of their results. First, it is well established that CIN2 and CIN3 carry very different risks of invasive cancer. CIN2 is a poorly reproducible category that includes manifestations of HPV infections and some early abnormalities that is known to be highly regressive, especially in young women.² Meanwhile, CIN3 is considered cervical precancer and presents a high risk of developing into invasive cervical cancer if untreated.³ Indeed, there is broad consensus that CIN3+ (v CIN2+) should be used as primary end point for studies of natural history, screening, and management.⁴

Guidelines in the United States⁵ and in many other countries^6,7 recommend different management of CIN2 and CIN3.⁸ Although CIN2 can be monitored, especially in women still considering bearing children because of the low risk of cancer and elevated risk of preterm delivery related to excisional treatment,⁹ CIN3—because it is higher risk of cancer—is almost universally treated. Therefore, combining the two diagnoses does not reflect current real-world practice. Moreover, because of the restrictive inclusion criteria on the basis of lesion size and history of CIN1 or greater, the study included a disproportionate number of CIN2 versus CIN3 than that occurs in clinical practice, which will result in higher than expected regression in a CIN2/3 cohort.

The trial report did not provide stratification on the initial diagnosis (CIN2 v CIN3) to determine whether clinical regression was the same in the two groups and whether FAM19A4/miR124-2 methylation was similarly associated with regression for the two diagnoses. It is possible that all or most of the observed regression occurred in those diagnosed with CIN2. Some of these authors reported that FAM19A4/miR124-2 methylation was more commonly found in CIN3 than CIN2.¹⁰ Notably, a considerable number (approximately 25%) of methylation-negative cases did not regress. Without knowing how many CIN3 are in that group, the level of reassurance provided by a negative methylation test cannot be appreciated.

Their regression definition may contribute to misinterpretation of these results. Given that CIN2 is a poorly reproducible diagnosis¹¹ and considered a mix of CIN3 and CIN1, transitions from CIN2 to CIN1 and/or HPV infection may not be truly regression. Indeed, in CONCERVE, lower-risk CIN2, those with antecedent low-grade (v high-grade) cytology, and smaller abnormalities were the most likely to regress.¹ Thus, regression in those cases might be a misclassified HPV infection, rather than precancer, that ultimately resolved on its own.

CONCERVE and other studies highlight the need for uniform management guidelines and standardization of end points for evaluation of interventions. Women diagnosed with CIN2 by biopsy during regular, routine screening (eg, every 5 years by HPV) and still considering childbearing could, with informed decision making, give their abnormality an opportunity to resolve without intervention. Markers such as p16^INK4a immunohistochemistry, FAM19A4/miR124-2 methylation, severity of cytology, and cancer risk associated with specific HPV types, might further inform the decision to wait and watch versus immediate treatment.^1,10,12 On the other hand, if there are mitigating circumstances such as a likelihood of loss to clinical follow-up, immediate treatment may be preferred. Combining CIN2 with CIN3 and managing them similarly obscures important differences in risk and muddles the informed‐decision process.

Correct end points are critical for interpreting the effectiveness of interventions or, in this case, the de-escalation of interventions. Numerous studies, including those evaluating novel therapies, have included a CIN2-to-CIN1 transition in their definition of regression, despite the growing body of literature that this interpretation is problematic. Even worse, those women who have this CIN2-to-CIN1 transition may still have type-specific HPV persistence, a necessary step in cervical carcinogenesis. Thus, not only does this result in an overestimation of the effectiveness on an intervention but in real-world practice, such women may be falsely reassured about their cancer risk.

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Clarifying the Equivocal Diagnosis of Cervical Intraepithelial Neoplasia 2: Still a Work in Progress

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).

No potential conflicts of interest were reported.