9. Management of histologically confirmed low-grade squamous abnormalities

According to the two-tiered nomenclature for cervical histology recommended by the Lower Anogenital Squamous Terminology (LAST) Standardization Project[1] and adopted by the Royal College of Pathologists of Australasia, non-invasive human papillomavirus (HPV)-associated squamous lesions are classified as follows (see Chapter 3. Terminology):

• LSIL: low-grade squamous intraepithelial lesion
• HSIL: high-grade squamous intraepithelial lesion

Current pre-renewal National Cervical Screening Program (NCSP) guidelines do not recommend treatment for histologically confirmed low-grade squamous intraepithelial lesions (LSIL); cervical intraepithelial neoplasia grade one (CIN1) or a lesser lesion detected on biopsy. These lesions are considered to be an expression of a productive HPV infection.[2] The 2005 national guideline for the management of screen-detected cervical abnormalities[2] recommended that women with histologically confirmed low-grade squamous abnormalities undergo repeat conventional cytology (Pap test) at 12 and 24 months, and return to routine screening if both tests are negative or continue annual cytology until two consecutive tests are negative.

Within the pre-renewal NCSP, it is common for a woman’s cytology status to fluctuate between low-grade abnormality and negative,[2] probably due to HPV infection, clearance and reinfection.[2] Within programs based on primary HPV screening, clinical decisions can be informed by known HPV status as well as cytology.

Within the pre-renewal NCSP, the HPV status of individual women with histologically confirmed low-grade squamous abnormalities is not routinely available and recommendations are based on the assumption that it is unknown. Under the renewed NCSP, in contrast, the clinical significance of histological, colposcopic and cytological findings in women with screen-detected abnormalities is assessed in the context of known positive HPV status.

Back to top

Systematic review evidence

A systematic review was performed to identify studies evaluating efficacy and safety outcomes of management options for women with a positive oncogenic HPV-test result, a colposcopic impression of LSIL and histologically confirmed CIN1 or less on biopsy, whether concordant or discordant with referral liquid-based cytology (LBC):

• For those with a cytology finding of negative, possible low-grade squamous epithelial lesion or low-grade squamous epithelial lesion (pLSIL/LSIL), repeat HPV test at 12 months was compared with the combination of repeat LBC and HPV test (co-testing) at 12 months.
• For those with a cytology finding of possible or high-grade squamous epithelial lesion (pHSIL/HSIL), diagnostic excision was compared with co-testing at 12 months.

No randomised or pseudorandomised controlled trials were identified that compared:

• HPV testing alone with co-testing as follow-up for women with a positive HPV test result with a reflex LBC finding of negative or pLSIL/LSIL
• excisional treatment with follow-up by co-testing in women with a reflex cytology finding of pHSIL/HSIL.

The search strategies, inclusion/exclusion criteria and findings are described in detail in the Technical report.

General literature review evidence

In the absence of any direct evidence from the systematic review, a general review of the literature was performed to inform consensus-based recommendations. The review focused on the management of women with:

• a positive oncogenic HPV test result
• a colposcopic impression of LSIL
• histologically confirmed CIN1 or less on biopsy
• referral cytology report of negative, or prediction of pLSIL/LSIL or pHSIL/HSIL.

Several studies examined outcomes following histologically confirmed CIN1 or less on biopsy.[2][3][4][5][6][7][8][9] The most relevant findings are summarised below:

• In a prospective cohort study using data from the Kaiser Permanente Northern California health system,[4] the crude rate of CIN3+ was 0.7% following a single negative follow-up smear, 0.2% following a single negative follow-up HPV test and 0.1% following a negative follow-up co-test in women with HPV-positive ASC-US or any LSIL and less than CIN2 on colposcopy/biopsy over a maximum of 7 years of follow-up.
• In a prospective cohort study (ASCUS-LSIL Triage Study),[7] 45% of women with baseline ASCUS/LSIL and less than CIN2 on colposcopy/biopsy, a follow-up finding of HSIL and a positive HPV test, developed CIN3+ within 2-year follow-up. (In this cohort, ASCUS may have included ASC–H.)
• One retrospective cohort study[8] of women who underwent loop electrosurgical excision procedure (LEEP) after referral cytological prediction of pHSIL/HSIL reported histologically confirmed CIN3+ in 41% of those with normal histology on the initial biopsy and 64% of those with CIN1 at initial biopsy. However, HPV status was not known.

Overall, available evidence suggests the following conclusions:

• For women with a positive oncogenic HPV test result, ASC-US and histologically confirmed CIN1 or a lower-grade lesion, the negative predictive value of HPV follow-up testing alone will be greater than that of cytology alone, but less than co-testing.
• For women with a positive oncogenic HPV test result and initial referral cytology of LSIL, the finding of HSIL at follow-up cytology indicates a substantial risk of future CIN3+.
• Diagnostic excision of the transformation zone (TZ) appears to be the optimal approach for women with a positive oncogenic HPV test result, a referral cytological prediction of pHSIL/HSIL, a colposcopic impression of LSIL, and histologically confirmed CIN1 or a lower-grade lesion, especially if pHSIL/HSIL is confirmed on cytopathology review.[10] The PPV is variable and depends on the referral cytology prediction. One study reported that, where referral cytology predicted a high-grade abnormality, then the PPV for identifying HSIL on biopsy was 73%. If the referral cytology was low grade, the PPV was 48%, demonstrating a clear relationship between the PPV of colposcopic impression to the referral cytology prediction.[11]

A summary of the literature considered can be found in the Technical report.

Back to top

When there is discordance between the LBC report and histopathology, review of both cytology and histopathology should be carried out to inform management decisions.

Flowchart 9.1. Histological LSIL following colposcopy for LBC prediction of HSIL

Despite the recommendations contained in the pre-renewal NCSP guidelines, many women have continued to have unnecessary treatment for LSIL (≤ CIN1). These guidelines reiterate the previous advice, and colposcopists are advised not to treat these women unless there are exceptional circumstances. Compliance with these recommendations will benefit women by avoiding unnecessary treatment and consequent harms.

Cytological review for discordant results, as recommended in these guidelines, will benefit women by preventing over investigation and unnecessary treatment (see Chapter 5. Benefits, harms and cost-effectiveness of cervical screening in the renewed National Cervical Screening Program (NCSP)).

Clinical practice

These recommendations represent minimal change to those that apply to the pre-renewal NCSP.[2]

Timely expert review of cytology and histology is recommended for women with low-grade histology results that are discordant with preceding high-grade cytology findings, before proceeding to any diagnostic treatment. Implementation of this recommendation may increase the workload of pathologists, laboratories and colposcopists. Clinicians may need to spend extra time reviewing results and providing advice to women.

The major change in clinical practice is that conventional cytology is replaced by HPV testing and reflex LBC for the follow-up of women with histologically confirmed LSIL (≤ CIN1). The high negative predictive value of HPV testing should allow a significant proportion of women to return to routine cervical screening earlier than was possible under the pre-renewed NCSP.

Resourcing

Pathology review of discordant results may increase the workload of pathology services, leading to delay in service provision and increased distress to already anxious women. Ensuring that pathology services are adequately staffed (have enough pathologists and laboratory staff) may have cost implications.

Failure of colposcopists to comply with the recommendation not to offer treatment for women with LSIL (≤ CIN1) would lead to unnecessary costs.

Barriers to implementation

These recommendations represent minimal change to those that apply to the pre-renewal NCSP.[2] Accordingly, no significant barriers to implementation are anticipated. It is recommended that LSIL (≤ CIN1) should not be treated. Under the pre-renewed NCSP, clinical practice has not fully complied with this recommendation. It is estimated that approximately 30% of women with LSIL (≤ CIN1) undergo excisional treatment. This decision may be due to physician or patient anxiety, especially when the finding is persistent. Under the renewed NCSP, colposcopists may continue to have concerns about conservative management of lowgrade abnormalities, despite evidence supporting this approach. Failure to implement this recommendation would result in unnecessary treatment and consequent harms.

Colposcopists may proceed to diagnostic excisional procedures for apparent discordant pathology results, without arranging for expert pathology review to confirm the findings. Failure to implement the recommendation for cytological and histological review may lead to unnecessary treatment.

Unresolved issues

No unresolved issues have been identified.

The safety of this approach will be monitored by the Quality and Safety Monitoring Committee of the NCSP.

Future research priorities

Prospective studies are needed to measure compliance with recommendations, especially the recommendation against routine treatment for histologically confirmed LSIL (≤ CIN1), and the recommendation for pathology review of discordant results.

Clinical trials are needed to determine the optimal management, and duration of observation, for persistent or fluctuating LSIL (≤ CIN1).