Absence or Presence of High-Grade Squamous Intraepithelial Lesion in Cervical Conization Specimens:  A Clinicopathologic Study of 540 Cases

Abstract

Objectives: To explore the implications of cervical conization specimens lacking the targeted high-grade squamous intraepithelial lesions (negative cone).

Methods: We studied 540 conization procedures: 400 positive cones and 140 negative cones. Clinicopathologic features and 2-year follow-up results were reported.

Results: Negative cones comprised 22% of procedures triggered by CIN2 or higher biopsies. Procedures triggered by cytology produced much higher percentages of negative cones (37% high-grade squamous intraepithelial lesion [HSIL], 46% atypical squamous cells—cannot exclude high-grade squamous intraepithelial lesion [ASC-H], and 76% low-grade squamous intraepithelial lesion—cannot exclude high-grade squamous intraepithelial lesion [LSIL-H]). Upon reviewing negative excision-triggering biopsy and cytology, we downgraded 24 (24%) CIN2 biopsies, three (14%) HSIL, five (83%) ASC-H, and 12 (92%) LSIL-H. One-third of our negative cones can be attributed to overdiagnosis either on biopsy or cytology. Patients with negative cones were older and had smaller excisions, negative colposcopic findings, and negative/equivocal high-risk human papillomavirus (HR-HPV). Within 2 years, 35 (25%) women with negative cones experienced ASCUS or LSIL. Only one (0.7%) recurred as CIN3, a significantly lower percentage than women with positive cones (13%).

Conclusions: We advocate careful review of all excision-triggering biopsy and cytology, especially in cases of LSIL-H. Patients with negative cones should be surveyed with cytology and HR-HPV testing.

Upon completion of this activity you will be able to:

• list possible causes of negative cervical conization specimens.

• describe the clinicopathologic features of women with negative conization results.

The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 Credit™ per article. Physicians should claim only the credit commensurate with the extent of their participation in the activ­ity. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module.

The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose.

Exam is located at www.ascp.org/ajcpcme.

Cervical high-grade squamous intraepithelial lesions progress to invasive cancer at a rate of 0.5% to 1% per year.^1,2 Conization using the loop electrosurgical excision procedure (LEEP) or cold knife can effectively prevent this progression.^3,4 It is a challenge for clinicians to completely remove target lesions while attempting to minimize damage to cervical integrity.^5,6 Histologic examination of the excised specimens not only confirms the diagnosis but provides prognostic factors that help guide subsequent management. Occasionally, we encounter excised specimens with only a low-grade lesion or normal epithelium that lacks their target high-grade lesion (negative cone). Several studies have acknowledged the ambiguous implication of such specimens, reporting its incidence as 10% to 39%.^7-12 Facing a negative result, clinicians often question whether the preceding cytology or biopsy specimen was correctly diagnosed, whether the excision may have missed the target lesion, and how these patients should be managed.

Such clinical concerns prompted us to study a total of 540 conization procedures triggered by either cervical intraepithelial neoplasia grade 2 or higher (CIN2+) biopsy or cytology diagnoses of high-grade squamous intraepithelial lesion (HSIL), atypical squamous cells—cannot exclude high-grade squamous intraepithelial lesion (ASC-H), or low-grade squamous intraepithelial lesion (LSIL)—cannot exclude high-grade squamous intraepithelial lesion (LSIL-H). We compared multiple clinicopathologic features among patients with negative vs positive cones, including age, size of specimen, preceding cytology and biopsy diagnoses, high-risk human papillomaviruses (HR-HPVs), and colposcopic findings. Then we followed all patients for 2 years with Papanicolaou (Pap) smear and human papillomavirus (HPV) tests every 6 months, aiming to compare disease recurrence or progression for these two groups.

Materials and Methods

After institutional review board approval was obtained, the UMass pathology database was searched for cervical LEEP or cold-knife cone specimens from January 2009 through December 2012. Patients with less than 2 years of follow-up, prior excisional procedures, or cervical glandular lesions were excluded. Most of the preceding cytology, biopsy, and HPV tests (90%) were performed at our facility, with the rest obtained from patient medical records. Cytology diagnoses were reported based on the Bethesda System.¹³ Biopsy diagnoses were reported based on the Lower Anogenital Squamous Terminology project recommendations¹⁴: low-grade intraepithelial squamous lesion (CIN1) and high-grade intraepithelial squamous lesion (CIN2 or CIN3). The size of the excised specimen was recorded as the sum of all tissue fragments. All of the tissue fragments were submitted for histologic examination. Routinely, four to six H&E levels were sectioned from each paraffin block. All negative cone cases were examined on multiple levels through tissue blocks. Immunohistochemistry stain p16 was performed on one to five selected sections from every case. At least two pathologists reviewed the slides and agreed on the absence of high-grade lesions. Resection margins were recorded as positive or negative for high-grade lesions. Some authors (V.W., Y.L., C.L.O. and A.K.) reviewed all negative cone cases, and others (Y.L. and C.L.O.) reviewed all biopsy specimens and Pap smears before negative cones.

HR-HPV test results within 6 months before the excisional procedure were retrieved from medical records. The Digene Hybrid Capture 2 Test (Qiagen, Gaithersburg, MD) was used to detect the 13 common HR-HPV types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68). Based on the manufacturer’s recommendations, results were reported as positive (a relative light unit/cutoff ratio ≥2.5), negative (ratio <1.0), or equivocal (ratio between 1.0 and 2.5).

Gynecologists at our hospital recorded their colposcopy impression at the time of excisional procedure according to American Society of Colposcopy and Cervical Pathology (ASCCP) recommendations, whereby (1) a colposcopic examination is considered adequate when the entire transformation zone is visualized, and (2) cervical lesions are graded based on multiple parameters (reaction to acetic acid, blood vessel pattern, lesion border, etc) and are categorized as inadequate, low grade or less, or high grade.

The Pearson χ² test and Student t test were used for comparing each set of categorical variables in the study. A P value less than .05 was considered statistically significant.

Results

During a 4-year period, a total of 540 women underwent LEEP or cold-knife cone procedures at our facility following diagnoses of either CIN2 + biopsy or cytology diagnosis of HSIL, ASC-H, or LSIL-H. CIN2 + lesions were present in 400 (74%) cone specimens but absent in 140 (26%) specimens. All negative cone specimens were confirmed by examining multiple levels through tissue blocks and p16 immunohistochemistry stain on one to five selected sections.

Lesions in positive cones included CIN2 (n = 224), CIN2 and CIN3 (n = 10), CIN 3 (n = 159), and invasive squamous cell carcinoma (n = 7). Lesions in negative cones included CIN1 (n = 122) or normal squamous epithelium (n = 18).

Excisional procedures were triggered by various biopsy or cytology results Table 1. Among 453 excisions triggered by CIN2 + biopsy specimens, 100 (22%) resulted in negative specimens. A total of 87 cases were triggered by cytology diagnosis alone (HSIL, ASC-H, and LSIL-H). Their concurrent biopsy diagnoses were CIN1 or negative for dysplasia. Among procedures triggered by HSIL (n = 57) or ASC-H (n = 13), the number of negative cones was 21 (37%) and six (46%), respectively. Of 17 excisions triggered by LSIL-H, 13 (76%) produced negative cone specimens (P < .001).

Table 1

Negative Cone Specimens Triggered by Various Biopsy or Cytology Diagnoses

Conization Procedure Triggered by	No. of Cases (n = 540)	Negative Specimen (n = 140), No. (%)
CIN2 + biopsy	453	100 (22)
HSIL cytology	57	21 (37)
ASC-H cytology	13	6 (46)
LSIL-H cytology	17	13 (76a)

CIN2+, cervical intraepithelial neoplasia grade 2 or higher; HSIL, high-grade squamous intraepithelial lesion; ASC-H, atypical squamous cells—cannot exclude high-grade squamous intraepithelial lesion; LSIL-H, low-grade squamous intraepithelial lesion—cannot exclude high-grade squamous intraepithelial lesion.

^aThe difference between LSIL-H and the other three groups is significant at P < .001.

Two authors (Y.L. and C.L.O.) reviewed all inciting biopsy specimens and Pap smears for the 140 negative cones, including 100 CIN2 + biopsy specimens, 21 HSIL cytology cases, six ASC-H cases, and 13 LSIL-H cases. In our opinion, 24 (24%) of the biopsy specimens that were initially called CIN2 + should be downgraded to CIN1 (n = 21) or negative (reactive changes, n = 3). In addition, we downgraded 20 (50%) Pap smears to LSIL or atypical squamous cells of undetermined significance (ASCUS), including three (14%) cases of HSIL, five (83%) cases of ASC-H, and 12 (92%) cases of LSIL-H.

Multiple clinicopathologic features were compared among patients with negative or positive cone specimens Table 2. The mean age was 36 years (range, 17-62 years) for the negative group, whereas it was 32 years (range, 19-68 years) for the positive group (P < .001). The mean specimen size was 2.7 cm (range, 0.6-11.2 cm) for the negative group vs 3.8 cm (range, 0.6-15 cm) for the positive group (P < .001). Preexcision HR-HPV status was known for 100 women in the negative group and 334 in the positive group. A negative or equivocal HR-HPV result was reported in 16 (16%) women in the negative group vs 17 (5%) in the positive group (P < .001). Colposcopic findings were available for 90 women in the negative group and 225 in the positive group. In the negative group, five (6%) were inadequate, 38 (42%) low grade or less, and 47 (52%) high grade. In the positive group, four (2%) were inadequate, 18 (8%) low grade or less, and 203 (90%) high grade. Compared with the negative group, the positive group had a significantly higher percentage of visible high-grade lesions on colposcopic examination (P < .001).

Table 2

Comparison of Clinicopathologic Features Between Patients With Negative or Positive Cone Specimens

Characteristic	Negative Cone	Positive Cone	P Valuea
Age, y	n = 140	n = 400
Range	17-62	19-68
Mean (SD)	36 (10)	32 (9)	<.001
Specimen size, cm	n = 140	n = 400
Range	0.6-11.2	0.6-15
Mean (SD)	2.7 (1.6)	3.8 (2.9)	<.001
HR-HPV status, No. (%)	n = 100	n = 334
Negative or equivocal	16 (16)	17 (5)
Positive	84 (84)	317 (95)	<.001
Colposcopy findings, No. (%)	n = 90	n = 225
Inadequate	5 (6)	4 (2)
≤ Low grade	38 (42)	18 (8)
High grade	47 (52)	203 (90)	<.001

HR-HPV, high-risk human papillomavirus.

^aThe difference between negative and positive cone groups is significant at P < .001.

All patients were monitored by cervical cytology at 6-month intervals following the conization Table 3. Of 140 women with negative cones, 35 (25%) experienced abnormal cytology within 2 years, including ASCUS (n = 23), LSIL (n = 11), and HSIL (n = 1, CIN3 as confirmed by biopsy). Women in the positive group were divided into two subgroups based on resection margins. Among 223 women with uninvolved margins, 55 (25%) developed abnormal cytology, including ASCUS (n = 20), LSIL (n = 24), and HSIL (n = 11, as confirmed by biopsy). Of 177 women with involved margins, 73 (41%) developed abnormal cytology, including ASCUS (n = 16), LSIL (n = 17), and HSIL (n = 40, as confirmed by biopsy). The recurrence rate of high-grade lesions was significantly lower among women with negative cones than for women with positive cones regardless of margin status (P < .001). Subsequent to the conization procedure, persistent positive HR-HPV was detected in 19 (23%) of the negative group vs 77 (24%) of the positive group.

Table 3

Two-Year Follow-Up Results for Patients With Negative or Positive Cone Specimens

		Abnormal Cytology, No. (%)
Characteristic	HR-HPV Conversion, %	Overall Incidence	ASCUS	LSIL	HSIL
Negative cone (n = 140)	77	35 (25)	23 (16)	11 (8)	1 (0.7)
Positive cone (n = 400)	76
Uninvolved margins (n = 223)		55 (25)	20 (9)	24 (11)	11 (5)
Involved margins (n = 177)		73 (41)	16 (9)	17 (10)	40 (22)

ASCUS, atypical squamous cells of undetermined significance; HR-HPV, high-risk human papillomavirus; HSIL, high-grade squamous intraepithelial lesion; LSIL-H, low-grade squamous intraepithelial lesion.

Discussion

We retrospectively studied a total of 540 cervical conization specimens, including 400 (74%) positive cones and 140 (26%) negative cones. Among 453 CIN2 + biopsy-triggered excisional procedures, 100 (22%) produced negative specimens. Explanations for this phenomenon include the possibility that (1) there was excisional error, (2) biopsy was overdiagnosed, (3) lesions were completely removed by biopsy, or (4) lesions spontaneously regressed. On the basis of our investigation, we believe that one case in our series was due to excisional error, 24 to overdiagnosis on biopsy, and the remaining 75 cases most likely due to the last two possibilities.

Conization procedures triggered by cytology diagnosis produced a much higher percentage of negative specimens (37% HSIL, 46% ASC-H, and 76% LSIL-H). Current ASCCP guidelines provide three options for women with HSIL or ASC-H: (1) cotesting at 12 and 24 months, (2) diagnostic excisional procedure, or (3) review of cytologic, histologic, and colposcopic findings. We reviewed the 40 Pap smears before the negative cones and downgraded 20 (50%) to LSIL or ASCUS, including three (14%) cases of HSIL, five (83%) cases of ASC-H, and 12 (92%) cases of LSIL-H. Theoretically, these 20 patients would have avoided excisional procedures if review had been done beforehand. None developed HSIL during the 2-year follow-up period.

Several studies have drawn attention to diagnostic errors in the cytology or biopsy specimen before negative conization. Thompson et al¹⁵ reported that false-positive reporting of Pap tests or tissue specimens resulted in 11 unnecessary conizations among 95 cases. Carrigg et al¹⁶ reported that 19 (28.3%) presurgical specimens from 15 patients were overcalled initially, including eight Pap smears, 10 biopsy specimens, and one endocervical curettage. These findings suggest that limiting diagnostic errors in presurgical materials is the first step in reducing unnecessary excisional procedures.

We put special emphasis on excision-triggering cytology diagnoses, especially LSIL-H. Diagnosis of LSIL-H is generally given to cases with findings of LSIL and ASC-H. In recent years, LSIL-H has been increasingly used in many laboratories despite the fact that it is not officially accepted in the Bethesda System.¹⁷ In our practice, the usage of LSIL-H rose from 5.8% of total squamous intraepithelial lesions in 2005 to 12% in 2010.¹⁸ Owens et al¹⁹ reported the risk of CIN2 + lesions as 40% for LSIL-H, an intermediate point between LSIL (10.8%) and HSIL (65.5%). There is yet no clear guideline for LSIL-H management. Based on our results, immediate excision for HSIL and ASC-H is justified, but not for LSIL-H. Precedents for this approach exist in many practices.^20-22 Among them, 13 (76%) produced negative specimens. None of these women developed high-grade lesions within 2 years of follow-up. We caution that although the designation LSIL-H is on the rise, it has significantly less predictive power than HSIL and ASC-H. To avoid overtreatment, LSIL-H in isolation should not be considered a signal for conization.

Several studies have reported that negative cones are associated with negative or low viral load in the preconization HR-HPV test, infection of HPV other than type 16, and CIN2 on biopsy.^23,24 We observed similar results: 61% of our negative specimens followed CIN2 on biopsy and 16% tested negative or equivocal for HR-HPV vs 5% of the positive group (P < .001). In addition, we noted that these women were older (mean age, 36 vs 32 years, P < .001) and had smaller excisions (mean size, 2.7 vs 3.8 cm, P < .001). Colposcopic examinations were more likely to be inadequate (6% vs 2%) and to reveal low-grade lesions (42% vs 8%, P < .001). Overall, these clinical characteristics imply a less severe or smaller volume of cervical lesions that have been removed altogether by biopsy or else spontaneously regressed.

Within 2 years, 25% of the negative group in our study encountered abnormal cytology, primarily ASCUS and LSIL. Only one (0.7%) woman had histologically confirmed CIN3 recurrence. We observed significantly higher HSIL recurrence among those with positive cone/uninvolved margin (5%) or positive cone/involved margin (22%). Livasy et al⁷ reported high recurrence after negative LEEPs (24%), primarily as high-grade lesions (13%). The authors suggested that blind excision or “piecemeal” manner can easily miss the target and leave high-grade lesions untreated. In the study from Carrigg et al,¹⁶ 30% of patients were found to have dysplasia (ungraded), and one patient was found to have invasive carcinoma on subsequent follow-up. The authors also considered sampling error and emphasized the need for careful surveillance. Compared with these studies, our recurrence rate was much less severe. This may be explained by the fact that excisional error was not the main source of negative specimens in our series. In addition, we observed similar rates of HR-HPV conversion among patients with negative or positive cones (77% vs 76%). Studies have shown that post-LEEP HPV conversion predicts the probability of negative cytologic follow-up as well as low risk of recurrence.^25-27

In conclusion, negative cone specimens do not imply a failed excisional procedure. We advocate careful review of all excision-triggering biopsy and cytology materials, especially the category of LSIL-H. In isolation, LSIL-H should not be taken as an indicator for conization. Given the good prognosis observed in our series, patients with negative cones should be surveyed with regular cytology and HR-HPV testing.

Acknowledgments

We thank John Stone for assistance in editing this article.