Comparative Oncologic Outcomes in High-Risk Human Papillomavirus-Positive and -Negative Cervical Intraepithelial Neoplasia

Introduction

The 2020 World Health Organization (WHO) classification of female genital tumors redefined cervical squamous cell carcinoma and adenocarcinoma by categorizing them into human papillomavirus (HPV)-associated and HPV-independent subtypes. Notably, precursor intraepithelial neoplasia was not incorporated into this classification, as isolated HPV-negative precursor lesions had not been conclusively documented until 2022 [1–3].

Persistent infection with high-risk human papillomavirus (HR-HPV) is a major contributing factor in the development of cervical cancer. While there are at least 14 oncogenic variants, types 16 and 18 are the most significant [4,5]. Although the mechanism of progression varies by HPV type, viral genome integration into host DNA represents a critical step in carcinogenesis. This integration triggers molecular alterations such as oncogene amplification, chromosomal rearrangements, and genomic instability [6].

HR-HPV genomes are detected in more than 95% of cervical cancer biopsies [6]. However, zur Hausen emphasized that this does not confirm HPV infection as the exclusive causative factor, underscoring the complexity of cervical carcinogenesis and the likely involvement of additional cofactors [7]. This is highlighted by the approximately 5% of cervical carcinomas that are reported to be HPV-negative, the exact etiology of which remains uncertain [8–10].

HR-HPV-negative intraepithelial lesions represent a clinically relevant paradox. While HPV is considered central to cervical carcinogenesis, approximately 10–15% of high-grade cervical dysplasia (cervical intraepithelial neoplasia grades 2 and 3, CIN2/3) cases are reported as HR-HPV-negative [11–14]. This negativity is frequently interpreted as false-negative results, as meta-analyses of primary HPV screening trials report sensitivities of 97–98% for high-grade dysplasia detection using Hybrid Capture 2 (HC2) and other clinically validated Polymerase Chain Reaction (PCR) assays [15]. However, HR-HPV negativity may reflect 2 distinct scenarios: truly HPV-independent lesions arising via rare pathways, or false-negative results. Causes for the latter include low viral load, infection with untested HPV types, viral clearance after malignant transformation has been initiated, poor sampling, or histopathologic errors [8,10,16–20]. Additional explanations for negative results include the “hit-and-run” hypothesis, whereby HPV initiates carcinogenesis but becomes undetectable by the time the lesion develops [21–24].

Distinguishing between a true HPV-negative lesion and a false-negative result is essential. This distinction can be established using several approaches: repeated high-sensitivity molecular testing, viral load assessment, inclusion of HPV types not covered by standard assays, and correlation with surrogate biomarkers such as p16 immunostaining [3,21,25,26]. Clarifying this distinction remains a key scientific priority for understanding the etiology and clinical implications of these cases.

The natural history and clinical significance of HR-HPV-negative high-grade intraepithelial lesions remain poorly defined [11]. This uncertainty is compounded by conflicting observations: some initially HR-HPV-negative lesions reveal HPV upon more sensitive re-analysis [21,27], while others persist as HPV-negative despite repeated testing [21]. Although these lesions lack standardized nomenclature, some studies suggest that HR-HPV-negative high-grade squamous intraepithelial lesions may exhibit distinct features, such as a peripheral location, subtle colposcopic signs, and potentially more favorable outcomes, while other reports document no significant demographic or clinical differences compared to their HPV-positive counterparts [11,12,27–29]. Reflecting this unresolved debate, current clinical guidelines recommend uniform management for high-grade lesions regardless of HPV status [30]. Although the behavior of HR-HPV-negative high-grade lesions is not well understood, it is clear that HR-HPV-negative cervical cancers are more aggressive, typically occur in older women, and have worse outcomes than HPV-positive cancers [8,31–35].

While primary HR-HPV testing increases the detection of high-grade dysplasia compared to cytology alone [36–38], and justifiably supports its integration into screening protocols [39], a significant challenge remains: high-grade lesions can test negative for HR-HPV [11,21,40]. These HR-HPV-negative cases challenge conventional screening and complicate diagnostic, colposcopic, and therapeutic decisions. Furthermore, they are inherently undetectable by primary HPV screening algorithms [41], creating a gap in our preventive strategy. From a public health perspective, clarifying outcomes of HR-HPV-negative high-grade intraepithelial lesions is vital for ensuring management guidelines remain evidence-based and universally applicable.

Therefore, this study aimed to compare recurrence and hysterectomy rates in a cohort of 712 women undergoing cervical conization for CIN3, based on their HR-HPV status.

Material and Methods

ETHICS APPROVAL:

The study was conducted in accordance with the ethical principles stated in the Declaration of Helsinki and approved by the institutional ethics committee (Akdeniz University Clinical Research Ethics Committee-KAEK-593, date: 02.08.2023). All patient data were anonymized prior to analysis, and unique study identifiers were used to maintain confidentiality. Only authorized personnel had access to the deidentified dataset. As a retrospective study, additional informed consent was not required, and the use of patient data was approved by the hospital’s institutional review board and management team.

STUDY DESIGN AND POPULATION:

Following ethics approval, we retrospectively analyzed the hospital records of patients diagnosed with CIN3, confirmed by histopathological assessment of colposcopically-directed biopsy, endocervical curettage (ECC), or cervical conization. Patients were treated and monitored at 2 tertiary healthcare institutions: the Department of Gynecological Oncology, Akdeniz University Faculty of Medicine, and the Department of Obstetrics and Gynecology, University of Health Sciences Antalya Training and Research Hospital, Turkey, between 2014 and 2023. Diagnoses of CIN3, defined according to the criteria of the Lower Anogenital Squamous Terminology (LAST) histologic classification, were established through histopathologic examination of hematoxylin and eosin-stained specimens [42]. These included samples obtained from ECC, biopsy, loop electrosurgical excision procedures (LEEP), and hysterectomy.

Inclusion criteria were: (1) histologically confirmed CIN3, (2) squamous cell lesions, (3) known HPV status at diagnosis, (4) regular follow-up attendance, and (5) complete accessible medical data. Exclusion criteria included CIN2 or lower-grade lesions, glandular lesions, prior hysterectomy, previous cervical procedures, pregnancy at diagnosis, unknown HPV status, coexisting gynecologic tumors, incomplete data, or irregular follow-up.

DATA COLLECTION:

Demographic, clinical, and obstetric characteristics at diagnosis, HPV status and HR-HPV types, Pap test results (ThinPrep-Hologic), histopathological findings from biopsies and endocervical specimens, margin status, surgical interventions, recurrence, and indications for hysterectomy were systematically collected. Immunosuppressive conditions, including Human Immunodeficiency Virus (HIV infection), immunosuppressive therapy, hematologic malignancies, and organ transplantation, were recorded. Data were extracted from both paper-based and electronic medical records, with additional verification through centralized Ministry of Health records when necessary.

MEASURES TO ENSURE DATA RELIABILITY:

To ensure data reliability and minimize potential biases inherent to the retrospective design, we implemented several rigorous quality control measures. Standardized data collection forms were used across both participating centers to ensure uniformity, and all data were cross-checked by independent reviewers to verify accuracy. Furthermore, all laboratory procedures, including HPV testing, cytology, and histopathological analyses, were performed in accordance with standardized operating procedures, and internal quality control measures were applied to each batch of samples, serving as a critical methodological control to ensure the analytical validity and objectivity of the results. We employed predefined protocols for handling missing data and conducted all outcome analyses in a blinded manner with respect to HPV status to prevent assessor bias. These steps strengthened the validity and reproducibility of our findings.

HPV TESTING:

Cervical samples were collected prior to any surgical intervention using standard cytobrush techniques with a Cervex-Brush^® Combi and preserved in SurePath™ liquid-based cytology medium (Becton Dickinson, Franklin Lakes, NJ, USA). The brushes were immediately immersed in the preservative fluid, mixed, and stored at 4°C until processing. All samples were processed using standard laboratory equipment, including a refrigerated centrifuge (Eppendorf 5810R, Germany), automated DNA extraction kit (Qiagen QIAamp DNA Mini Kit, Hilden, Germany), and micropipettes with sterile filtered tips (Eppendorf, Germany).

Each patient underwent a single HPV test using the Cobas 4800 HPV test (Roche HPV assay), which demonstrated a sensitivity of 93.5% (82.5–99.8) and a specificity of 69.3% (66.9–71.5) [8,43]. This testing was uniformly conducted in both participating institutions. The Cobas 4800 HPV test is designed to detect 14 HR-HPV types, as classified by the International Agency for Research on Cancer (IARC). These include HR-HPV genotypes 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68 [44].

This fully automated, qualitative in vitro assay employs real-time PCR and nucleic acid hybridization to amplify target DNA according to the manufacturer’s protocol for the Cobas 4800 HPV test (Roche Diagnostics), ensuring that samples are processed within the recommended time frame to preserve DNA integrity. PCR cycling conditions included initial denaturation at 95°C for 10 minutes, followed by 40 cycles of 95°C for 15 seconds and 60°C for 1 minute, in accordance with the manufacturer’s instructions. This approach enables the detection of all 14 high-risk HPV types in a single analysis, providing pooled results for 12 high-risk genotypes and individual results for the most oncogenic genotypes, HPV 16 and 18, based on predefined and validated cycle threshold cut-off values, and objectively assigning a positive or negative result [45]. Internal controls for cellular adequacy and PCR amplification were incorporated into the test procedure to ensure assay validity in accordance with the manufacturer’s specifications. The raw data and final results were exported into the laboratory information system without manual alteration to preserve objectivity.

For the purposes of this study, results were categorized for statistical analysis as follows: samples negative for all 14 high-risk HPV types were classified as HPV-negative; samples positive for HPV 16, regardless of the status of other channels, were classified as HPV 16 positive; samples positive for HPV 18 but negative for HPV 16 were classified as HPV 18 positive; and samples positive for the pooled 12 high-risk HPV types but negative for both HPV 16 and 18 were classified as other HR-HPV-positive. If the initial HPV test result was negative, no further HPV testing was performed.

All diagnostic procedures, including HPV testing, liquid-based cytology, colposcopies, and histopathological biopsies, were performed by certified gynecologic oncologists and accredited colposcopists. The surgeons were informed of the HPV test results prior to performing any interventions.

POSTOPERATIVE MANAGEMENT AND FOLLOW-UP:

The follow-up of the patients was conducted in accordance with the recommendations outlined in the American Society for Colposcopy and Cervical Pathology (ASCCP) guidelines that were in effect at the time of patients’ treatment [30,46]. Following LEEP, patients with positive pathological margins underwent colposcopic evaluation 3 months after conization, while those with negative margins were assessed 6 months post-conization. Patients were advised to undergo co-testing (cytology and HPV testing) at six-month intervals for the first 2 years. Colposcopy was recommended based on the results of these consecutive tests and clinical findings during this period. Annual follow-up examinations were recommended thereafter. Patients were followed from the date of conization until the first occurrence of any of the following events: a diagnosis of recurrent CIN2 or higher lesion (CIN2+), the last follow-up visit, or hysterectomy.

OUTCOME DEFINITIONS:

Recurrence was defined as a new histologically confirmed diagnosis of CIN2+ with specific diagnostic criteria stratified according to surgical margin status following initial conization. For patients who had previously achieved negative surgical margins, recurrence required confirmation via cervical multisite biopsy and/or ECC. In cases with positive surgical margins, a diagnosis of recurrence was only established when CIN2+ were detected following at least one documented negative co-test result or normal colposcopic examination. Lesions classified as less than CIN2 were not considered indicative of recurrent disease.

Patients were excluded from the multivariable logistic regression analysis for recurrence if they met any of the following criteria: (1) no follow-up after conization, (2) did not undergo conization, (3) diagnosis of cervical cancer via cervical biopsy or cone specimens, or (4) immediate hysterectomy following conization. For the analysis evaluating factors associated with hysterectomy, patients who underwent hysterectomy for reasons unrelated to cervical intraepithelial lesions were also excluded. In the multivariable models, potential confounders were selected based on their established clinical relevance in the literature and statistical significance in univariable analyses. Specifically, variables included were age, parity, menopausal status, immunosuppression, comorbidity, smoking status, HPV status and types, and surgical margin positivity, as these are known to influence recurrence and treatment outcomes.

STATISTICAL ANALYSIS:

Statistical analysis was performed using IBM SPSS Statistics for Windows version 27.0 (IBM Corp., Armonk, NY). Data were summarized using frequency tables and descriptive statistics. Continuous variables were assessed for normality with the Shapiro-Wilk test and are presented as mean ± standard deviation, median, or n (%), as appropriate. The primary outcome variable for the recurrence and hysterectomy analyses was a binary measure (yes/no), defined as either a newly histologically confirmed CIN2+ lesion or undergoing hysterectomy, respectively. The primary exposure variable was HPV status (positive/negative). Based on the distribution, group comparisons were made using the Independent Samples t-test (for parametric data) or the Mann-Whitney U test (for non-parametric data). Associations between categorical variables were analyzed with Pearson’s chi-square test. To identify independent risk factors for recurrence and hysterectomy, we performed binary logistic regression using the Backward Likelihood Ratio (LR) method. Variables found significant in univariable analyses were initially included in the multivariable models. The model’s goodness-of-fit was assessed using the Hosmer-Lemeshow test. Results are presented as adjusted odds ratios (OR) with 95% confidence intervals (CI). Recurrence-free survival was estimated using the Kaplan-Meier method, with the time-to-event defined as the duration from the date of conization to the date of histologically confirmed recurrence. Patients without recurrence were censored at their last follow-up date. Differences in survival curves between HPV groups were compared using the Log-rank test. A p value <0.05 was considered statistically significant.

Patients with missing or incomplete data for key variables were excluded from the relevant analyses, and no data imputation was performed due to the study’s retrospective design and the low proportion of missing data.

Results

STUDY POPULATION AND BASELINE CHARACTERISTICS:

In total, retrospective data from 1540 women diagnosed with high-grade cervical dysplasia (CIN2 and CIN3) and treated at 2 tertiary referral centers between 2014 and 2023 were analyzed. Of these, 828 patients (53.8%) were excluded due to unmet inclusion criteria, resulting in an analytical cohort of 712 women (46.2%). The study population comprised 64 women (9%) with CIN3 who tested negative for HR-HPV and 648 women (91%) who tested positive, as illustrated in Figure 1.

Table 1 presents the demographic and clinical baseline characteristics of patients with CIN3 according to HR-HPV infection status (HPV-positive vs HPV-negative), with no significant differences observed between the groups (p>0.05). HR-HPV-positive patients were more often referred via routine screening, while HR-HPV-negative patients were mainly referred due to abnormal cytology or symptoms (p<0.05). p16 positivity was higher in HR-HPV-positive CIN3 compared to HR-HPV-negative CIN3 (p=0.001), although some discordance was observed. Cervical cancer was rare in both groups (2% vs 1.5%), with comparable risk (p>0.05).

LOGISTIC REGRESSION AND SURVIVAL ANALYSIS:

A total of 532 patients from the study population were included in the regression analysis conducted for recurrence assessment. The following cases were excluded from the regression analysis: patients who discontinued follow-up after conization (n=118), those who did not undergo conization (n=16), individuals diagnosed with cervical cancer via histopathological evaluation of cervical biopsy or cone specimens (n=14), and patients who underwent immediate hysterectomy following conization (n=32). The median follow-up for patients without recurrence was 19.2 months, while the median time to recurrence was 13.3 months. Recurrence occurred in 12.1% of HR-HPV-positive and 6.2% of HR-HPV-negative patients, with no significant difference in recurrence rates or time to recurrence (median 16.7 vs 23.4 months, respectively) (P>0.05). Figure 2 shows cumulative recurrence probability in patients with CIN3 according to HPV status. Recurrence rates were analyzed according to combined HPV and p16 status in subgroups. Among HR-HPV-positive patients, 26 of the p16-positive and 1 of the p16-negative cases experienced recurrence. In the HR-HPV-negative group, 1 of the p16-positive and 2 of the p16-negative cases recurred. Statistical analysis demonstrated no significant differences between these subgroups (χ2=5.23, df=3, P=0.156), suggesting that recurrence was independent of HPV or p16 expression.

Univariate analysis revealed no statistically significant correlation between HR-HPV status (positive/negative) and recurrence risk. Although age and reproductive history appeared to correlate with recurrence risk (Table 2), multivariate analysis revealed no statistically significant association with any of these factors.

A total of 561 patients from the study population were included in the regression analysis performed to evaluate hysterectomy outcomes. The following cases were excluded from the regression analysis: patients who discontinued follow-up after conization (n=118): those who did not undergo conization (n=16), individuals diagnosed with cervical cancer via histopathological evaluation of cervical biopsy or cone specimens (n=14), and patients who underwent hysterectomy for reasons unrelated to cervical intraepithelial lesion (n=3). Hysterectomy was performed in 11.4% (n=64) of patients, most commonly due to ≥CIN2+ at the surgical margin, with no significant difference between HR-HPV-positive and HR-HPV-negative groups (11.7% vs 8.7%; P>0.05) (Table 3). The median follow-up duration was 18.6 months for patients who did not undergo hysterectomy, compared to 3.4 months for those who underwent the procedure. Age and menopausal status were the strongest predictors of hysterectomy, with a significant interaction effect (Table 4).

HYPOTHESIS-TESTING OUTCOMES:

The study findings directly support the primary research objectives. Baseline clinicopathological characteristics and cervical cancer risk were comparable between HR-HPV-positive and HR-HPV-negative women, confirming equivalence at study entry. Recurrence analysis demonstrated no significant difference in recurrence rates between HR-HPV-positive and HR-HPV-negative patients, indicating that HPV status alone may not be a strong predictor of recurrence. Similarly, hysterectomy outcomes were comparable between the 2 groups, with patient age and menopausal status emerging as more influential predictors. Collectively, these results validate the study hypotheses regarding the impact of HR-HPV status on clinical outcomes in women with CIN3.

Discussion

In this study, we evaluated recurrence and hysterectomy rates in women with CIN3, comparing HR-HPV-positive and HR-HPV-negative cases. HR-HPV-negative CIN3 accounted for 9% of our cohort, and we observed no significant differences in recurrence, hysterectomy incidence, cervical cancer detection, or time to recurrence between the 2 groups.

Regauer et al provided the first comprehensive histomorphologic and immunohistochemical characterization of 3 HPV-negative cervical precursor lesions without associated invasive cancer [47]. Their study established HPV negativity through multiple verification methods, including the absence of both DNA and E6/E7 mRNA, supplemented by whole-genome sequencing. In contrast, our retrospective study reflects real-world clinical practice by utilizing a single HR-HPV test without specimen re-analysis or pathology review. This approach aligns with current guidelines, which do not recommend repeating HPV testing or pathology review when colposcopic examination is indicated for confirmed HR-HPV-negative patients [30]. Nevertheless, archived cervical specimens from prospective validation studies remain valuable for subsequent re-analysis. Recent molecular studies have further challenged initial classifications by detecting HPV DNA in a significant proportion of lesions originally deemed HPV-negative [3,21,26,27,48].

The critical role of accurate HR-HPV diagnosis in cervical cancer elimination is well-established, as it directly determines the efficacy of population-based screening and the surveillance of vaccination impact [49]. However, the existence of HR-HPV-negative lesions presents a significant challenge to this strategy. Multiple studies confirm that a notable subset of patients with a subsequent CIN2+ diagnosis had prior negative screening results. The reported prevalence of these false-negative cases demonstrates a clear time-dependent increase, with large-cohort studies showing rates of 2.4% to 4.1% for tests within 12 months of a CIN3 diagnosis, rising substantially to 19.9% to 29.4% when the screening occurred more than 12 months prior to diagnosis [50–52]. This trend is further supported by studies with shorter screening-to-diagnosis intervals, which report rates aligning closely with our findings, ranging from 4.5% to 12.3% [53–57]. Another study also found that 22.7% of patients with high-grade pre-cancerous lesions were HR-HPV-negative [58], showing this is a real problem, not just a rare exception.

Current evidence reveals the heterogeneous nature of these lesions; some are true HPV-independent precancers [47], while others are attributed to false-negative results or diagnostic misinterpretations [2,3]. This diagnostic uncertainty fundamentally challenges conventional screening and management approaches. Consequently, understanding their precise origin and clinical implications becomes imperative. A notable initiative addressing this need was Sweden’s 2019 national campaign, led by its HPV reference laboratory, which systematically re-analyzed HR-HPV-negative high-grade lesions [21], showing that many initially negative samples were ultimately false negatives, highlighting the critical importance of verification re-testing to establish true HPV status and determine the correct etiological basis of such lesions.

CIN3 development is strongly associated with persistent HPV infection, typically detectable years before diagnosis. However, the inherent limitations in sensitivity and specificity of HPV testing [28] suggest that most HR-HPV-negative cases likely are false-negative results rather than true biological negativity. This interpretation is supported by the comparable oncologic outcomes observed between HR-HPV-negative and -positive CIN3 cases in our study, indicating a retained malignant potential regardless of HPV status. Several factors may explain this clinical observation: possible false-negative results, rare HPV-independent oncogenic pathways, viral clearance after malignant transformation, or new/reactivated HPV infections during follow-up [59,60]. Furthermore, once CIN3 becomes established, disease behavior may be predominantly governed by host factors rather than viral etiology [61], which could also account for the absence of significant intergroup differences.

False-negative HR-HPV results can occur when tests fail to detect all relevant HPV genotypes. This was demonstrated in an ATHENA trial sub-analysis, where all initially HR-HPV-negative cases tested positive upon re-testing, revealing infections caused by genotypes undetectable by standard platforms [3]. Therefore, expanding the genotype coverage of modern HPV tests could significantly reduce these false negatives. Clinically, it is crucial to recognize that HR-HPV-negative CIN3 cases still carry substantial recurrence risk and malignant potential. This underscores the need for consistent, careful follow-up and management regardless of HPV status.

Immunohistochemical (IHC) analysis in this cohort revealed p16 immunopositivity, a clinically validated yet imperfect surrogate marker of HPV oncogenic activity, in 34% of cases initially classified as HR-HPV-negative. This discordance suggests either false-negative HPV testing or histopathological misinterpretation. Our findings align with Castle et al’s observations, where approximately one-third of CIN2+ lesions with initially negative HR-HPV results were positive upon confirmatory testing [12]. This consistency underscores the importance of integrating cytology and colposcopic findings into diagnostic assessments, rather than relying solely on HPV or p16 testing alone.

In our cohort, recurrence rates were similar across subgroups stratified by HPV and p16 status, indicating that p16 discordance does not significantly influence short-term outcomes, although subgroup sizes were limited. However, for women with both HR-HPV-negative and p16-negative biopsies, the diagnosis of CIN3 should be carefully reconsidered to avoid unnecessary LEEP procedures. Such interventions carry a risk of iatrogenic preterm delivery without providing cancer prevention benefits in these particular cases [62,63]. Finally, comprehensive p16 immunohistochemistry data for CIN3 cases were unavailable across our full study cohort, necessitating cautious interpretation of biomarker correlations in this subgroup.

In the present study, the proportion of HR-HPV-negative patients (9%) was lower than the 15% reported by Bogani et al, but aligned with other published rates [11,40]. While Bogani et al observed an 8-fold higher recurrence risk in HR-HPV-positive patients, our findings revealed no such difference, likely due to variations in study design and cohort characteristics. Unlike Bogani’s study, which combined CIN2 and CIN3 lesions, our analysis focused exclusively on histologically confirmed CIN3 cases, yielding a more homogeneous cohort with greater histopathological severity. Methodological factors such as HPV testing techniques, follow-up duration, patient selection strategy, and referral patterns may also account for the discrepant results. The concept, that HPV status does not consistently predict risk across patient populations is supported by Katki et al, who found similar cervical cancer risks in HR-HPV-positive and HR-HPV-negative women with high-grade lesions, except for cases involving atypical glandular cells (AGC) [64]. Conversely, Castle et al’s meta-analysis indicated that HR-HPV-negative CIN2+ lesions generally exhibit milder colposcopic findings, a higher prevalence of low-risk HPV types, and a lower carcinoma risk than their HR-HPV-positive counterparts [12]. The absence of such differences in our study may be related to the shorter follow-up period and the exclusive inclusion of CIN3 cases, which could have limited the detection of late recurrences.

The superior oncologic outcomes in HR-HPV-negative patients may be explained by several biological and diagnostic factors. First, some high-grade lesions may originate from non-oncogenic HPV subtypes or high-risk variants with significant genomic sequence divergence [18,65]. Second, deletions in the HPV genome, particularly the L1 capsid region during viral integration into host DNA, can disrupt episomal maintenance and reduce viral persistence, thereby lowering recurrence risk [10]. Third, undetectable HPV viremia at diagnosis owing to viral loads enables host immune eradication prior to neoplastic progression [66]. Fourth, the diagnostic sensitivity of HPV testing depends on multiple pre-analytical and analytical variables, including test selection, implementation quality, and laboratory proficiency [67]. Finally, some cases may represent diagnostic challenges where benign epithelial changes mimicking high-grade lesions were initially misclassified as CIN3, particularly in the absence of definitive HPV association [68].

Prevailing views suggest that truly HR-HPV-negative high-grade lesions are rare and often misclassified as cervical intraepithelial neoplasia grade 1 (CIN1) [21,28]. However, our data demonstrate that HR-HPV-negative CIN3 should not be dismissed, as recurrence and cancer detection rates remain comparable to HR-HPV-positive CIN3. Persistent HPV infection is a necessary biological precursor for clinically significant CIN3 [69]. By contrast, transient HPV infections typically maintain episomal viral DNA and are biologically associated with LSIL, which frequently undergo spontaneous regression [70].

The principal strength of this study lies in its exclusive focus on patients with histologically confirmed CIN3 – an unequivocal pre-cancerous lesion. By analyzing a pure CIN3 cohort, we established a homogeneous population that eliminates the diagnostic variability and biological heterogeneity inherent in studies combining CIN2 and CIN3. This methodological approach enhances the reliability of our findings regarding CIN3’s clinical behavior. The exclusion of CIN2 lesions was particularly important given their heterogeneous natural history, which often resembles CIN1 regression rather than CIN3 progression. By excluding CIN2/3 cases, we avoided diagnostic uncertainty while focusing on lesions with consistently higher clinical relevance and invasive cervical cancer risk compared to the more variable CIN2 category.

The retrospective design of this study imposes several inherent limitations. First, potential selection bias may have affected cohort composition due to its reliance on predefined inclusion criteria and available medical records. Second, although standardized protocols were followed, variations in sample collection, handling, and testing procedures inherent to routine clinical practice may have introduced inconsistencies. These pre-analytical and analytical factors, combined with the known sensitivity and specificity limitations of HPV testing, cervical cytology, and colposcopy, are additional methodological constraints. Furthermore, surgical procedures were performed by multiple surgeons with varying levels of experience in gynecologic oncology rather than by a single operator, potentially contributing to outcome heterogeneity. Finally, the observational nature of retrospective data collection precludes definitive causal inferences. These limitations warrant careful consideration when interpreting results and generalizing findings to broader populations.

Our analysis acknowledges 2 potentially unaccounted confounders: HPV vaccination status and pharmacologic agents that might influence HPV clearance. Incorporating these variables into multivariate models could have strengthened the validity of our oncologic outcome assessments. Additionally, while established evidence indicates varying recurrence risks based on anatomical margin location (endocervical versus ectocervical), our study lacks this stratification. However, a recent meta-analysis demonstrated HR-HPV testing provides superior predictive accuracy for recurrence compared to margin status evaluation [71], supporting our methodological focus on virological rather than anatomical factors.

A principal limitation of this study is the incomplete characterization of HR-HPV-negative CIN3’s natural history, constrained by both the relatively small cohort of HPV-negative precursor lesions, limiting statistical power for subgroup analyses, and the comparatively short follow-up duration. Furthermore, the absence of centralized pathology review may have affected the consistency and accuracy of histopathological interpretation. Without such re-evaluation, distinguishing truly HR-HPV-negative cases from potential false-positive histology is challenging. If a significant proportion of the HR-HPV-negative cohort had represented misclassified lesions, key outcome measures would likely have differed significantly from HR-HPV-positive cases, an effect not observed in our analysis. This underscores the importance of standardized, centralized pathology assessment in future studies to validate and refine these findings

These limitations highlight the need for prospective validation studies investigating HPV status and diagnostic modalities with enhanced specificity to accurately predict post-treatment oncologic outcomes, particularly when p16 overexpression conflicts with HPV test results.

Conclusions

HR-HPV-negative CIN3 is a clinically significant entity, comprising approximately 9% of cases, and demonstrates comparable risk of recurrence and hysterectomy as HR-HPV-positive CIN3. These findings indicate that HR-HPV-negative patients require follow-up vigilance equivalent to that for HR-HPV-positive patients. Further prospective studies are warranted to confirm these observations.