Post Colposcopy Management of ASC-US and LSIL Pap Tests

There is weak evidence supporting optimal follow-up of women with ASC-US or LSIL cytology found to have low grade disease or normal findings at initial colposcopy. Surveillance options include continued colposcopy, discharge with Pap testing, or HPV testing at 12 months. The investigators performed a pilot randomized controlled trial (RCT) comparing these 3 follow-up policies. Study objectives are to determine the feasibility of an RCT and to compare the incidence of \>/=HSIL in each of the arms by intention to treat principle.

1\. BACKGROUND In 2010 the estimated age standardized incidence and mortality rates for cancer of the uterine cervix in Canada were 7 and 2 per 100,000 women respectively. Squamous cell carcinoma is the most frequent type and arises from a premalignant squamous dysplasia called cervical intraepithelial neoplasia (CIN). There are 3 grades of CIN from I to III in order of increasing severity. A systematic review of 73 studies on the natural history of CIN, determined nearly 89% of CIN I regressed to normal or a benign lesion or persisted unchanged, and progression to CIN III occurred in 11% and to cancer in 1%. In contrast, CIN III and II had a greater malignant potential as 12% and 5% respectively progressed to cancer. The low incidence and mortality rates of cervical cancer are largely attributed to the availability of screening for CIN with the Pap test (Papanicolaou stained cervical smear) and the colposcopic management of cytologic abnormalities. While rates have plateaued in recent years, decreases are expected with additional improvements in Pap test performance and pre-colposcopy management of cytologic abnormalities. Changes to the post-colposcopy management of cytologic abnormalities could also improve the rates, but rigorous investigation of potential changes is lacking. 1.1 Management and outcome of cytological abnormalities The Bethesda System (TBS) is the recommended Pap test reporting terminology in Canada. TBS 2001 classifies results as negative for intraepithelial lesion or malignancy (NILM), or an epithelial cell abnormality. The latter includes 1) atypical squamous cells of undetermined significance (ASC-US), 2) atypical glandular cells (AGC), 3) low grade squamous intraepithelial lesion (LSIL), 4) atypical squamous cells, cannot exclude HSIL (ASC-H), 5) high-grade SIL (HSIL), 6) adenocarcinoma in situ (AIS), and 7) malignant. Management of cytological abnormalities may involve colposcopic examination. The colposcope magnifies the cervix and allows directed biopsy of any visualized mucosal abnormality (impression). A repeat Pap test and tissue sampling of the endocervical canal (endocervical curettage; ECC) can also be carried out. CIN II, III and cancer diagnosed at the first colposcopy exam are considered to be incident disease undersampled by the referral Pap test (incident \>/=CIN II) 3. CIN II or higher detected within the subsequent 24 months is categorized as progressive disease. Some investigators however, also consider it incident disease that escaped earlier colposcopic detection and refer to incidentally detected and progressive CIN II or higher as cumulative CIN II (CIN 2+). Approximately 1% of cytological results in the United States is an HSIL, 4.4 % is an ASCUS (atypical squamous cells of undetermined significance to include ASC-US and ASC-H) and 1.6% an LSIL. Canadian data is somewhat similar (Calgary Laboratory Services unpublished quality assurance data), Since upwards of 60-80% of HSIL is a CIN II/III colposcopic examination and treatment is appropriate so as to prevent cervical cancer. Uncertainty and controversy in regard to the outcome of ASC-US and LSIL existed until the completion of the ALTS (ASCUS and Low Grade Triage Study) randomized clinical trial. The trial determined 13% of the combined ASCUS and LSIL cohorts had an incident \>/=CIN II and an additional 8% progressed to CIN II/III. These results underscored the importance of managing low grade cytological abnormalities to prevent cervical cancer. 1.2 Pre-colposcopy management of ASC-US and LSIL Current management of ASC-US and LSIL in Canada is guided by the Clinical Practice Guidelines of the Society of Obstetricians and Gynecologists of Canada (SOGC) and the 2001 and 2006 guidelines of the American Society for Colposcopy and Cervical Pathology (ASCCP). Management options include referral to colposcopy after a single or double test result of ASC-US or LSIL or a single ASC-US result testing positive for HPV (Human Papilloma Virus) DNA (deoxyribonucleic acid). Management in Canada is mostly based on repeat Pap testing since HPV testing is not routinely available in every jurisdiction. For example, in Alberta, 2 years of repeat Pap testing carried out at 6-month intervals is the standard for an ASC-US or LSIL detected during routine screening and referral to colposcopy occurs when follow up Pap tests show persistence or progression. 1.3 Colposcopy management of ASC-US and LSIL Mucosal abnormalities of the cervix are biopsied and additional treatment is based on the histopathological results. Repeating the Pap test at colposcopy is of minimal value in the added detection of CIN II/III/cancer, and is not routinely performed. An ECC is recommended to sample disease in the endocervical canal. ASC-US and LSIL tests with incident \>/=CIN II are treated with ablation or excision, whereas those with \</=CIN 1 (benign/CIN I) are not treated. 1.4 Post-colposcopy management of ASC-US and LSIL The post colposcopy management of women referred with ASC-US or LSIL and who do not have incident \>/=CIN II is still uncertain. Management algorithms are available from the ASCCP despite the absence of strong evidence to support all of the recommendations. Amongst those with \</= CIN I at colposcopy, a repeat Pap test at 12 months is recommended if the women were referred with ASC-US and the HPV DNA status at colposcopy was unknown, and repeated at 6 and 12 months if they were referred with an HPV positive ASC-US or an LSIL. HPV DNA testing at 12 months is an option for those referred with an HPV positive ASC-US or an LSIL. Women would be referred back to colposcopy if the repeat Pap test was ASC-US or higher or the HPV DNA test was positive. The evidence was rated BII and BIII by the ASCCP which they defined as "moderate evidence for efficacy or only limited clinical benefit supports recommendation for use (B); based on evidence from at least one clinical trial without randomization, from cohort or case controlled analytic studies, or from multiple time series studies, or dramatic results from uncontrolled experiments (II); evidence from opinions of respected authorities based on clinical experience, descriptive studies, or reports of expert committees (III)". Post-colposcopy management of ASC-US and LSIL in Canada is very variable and likely reflects the lack of strong evidence supporting the existing guidelines. Based on 102 survey responses from 252 Canadian colposcopists, 43% recommended discharge from colposcopy for ASC-US/LSIL negative for CIN, while 53% recommended repeat colposcopy. For ASCUS/LSIL positive for CIN I, 13% recommended discharge to Pap test follow-up, 65% recommended repeat colposcopy, and 16% recommended treatment. The number of follow up colposcopy exams was variable but there was at least one at 6 months and more often a second at 12 months. The survey highlighted a greater reliance on colposcopy as a standard of follow up care and a reluctance to discharge women for follow up with routine Pap tests performed in the primary care setting. Notably, no participant reported utilizing HPV DNA testing. The high (48%) level of survey participation and individual participant's comments attested to the community's interest in scientifically evaluating the best post-colposcopy management strategy for this clinical scenario. 1.4 Post-colposcopy management of ASC-US and LSIL negative for incident CIN II/III The ALTS trial was principally designed to assess the pre-colposcopy management of women with ASCUS or LSIL. Later, some of the trial data was used to test repeat cytology and HPV testing as post-colposcopy management options. Repeat colposcopy examination could not be measured as a management option. HPV testing at 12 months was found to be 92% sensitive for progression to \>/=CIN II amongst those with \</=CIN I at the initial colposcopy and would refer 55% of women back to colposcopy. The sensitivity compared very well with the 88% obtained with 2 follow up ASCUS Pap tests which however would refer slightly more (64%) women back to colposcopy. The sensitivity of a single colposcopy exam for cumulative CIN II is generally considered to be approximately 80% but may be closer to 55% based on recent data from the ALTS trial. Thus post colposcopy follow up with a single HPV test at 12 months or repeat Pap testing at 6 and 12 months may be equivalent to or more sensitive than follow up with 1 or even 2 colposcopy exams. However, a randomized trial is needed to prove this hypothesis and provide the evidence needed to change the current colposcopy dominated practice in Canada. 2 RESEARCH DESIGN This is a pilot randomized controlled trial performed at the Holy Cross Colposcopy Clinic of the Calgary Health Zone for the purpose of testing a randomized clinical trial which will be designed to test above-stated objectives. The pilot will test the recruitment, randomization, operational feasibility, and clinical outcomes. The pilot data will be added to the future full proposal of the clinical trial. Potential additional studies stemming from the clinical trial will include the evaluation of molecular determinants of outcome, e.g. persistence of specific HPV types, viral load and viral integration, analyses of temporal changes in behavioral factors, and cost effectiveness of the 3 arms. 3 RESEARCH METHODOLOGY 3.1 Pilot sample, enrolment and randomization Usual colposcopic and pathology practices will be applied at the initial exam and a repeat Pap test will not be performed. After enrolment, baseline data will be extracted by the study team from the clinic charts, including date of the referral Pap test and result, date of the colposcopy exam, colposcopy impression, and cervical biopsy and ECC results. Those with \</=CIN I by impression and pathology will not be treated. These women will be randomized by RANDOMIZE.NET to one of the 3 follow up policies - colposcopy (considered standard of care), cytology or HPV testing (simulating discharge to community). A sticker will be added to the clinic charts describing the woman's policy. All cervical disease identified at these referrals back to colposcopy will be managed in accordance with standard protocols. 4 DATA MANAGEMENT AND ANALYSIS 4.1 Data management All data collected by the research team will be collected in a de-identified fashion in the REDCap Software managed by University of Calgary. Data discrepancies will be reviewed by an independent third person. Data forms will be checked prior to data entry, and range and logic checks carried out regularly on entered data to ensure the quality of the data. Any results that appear to be out of range or do not comply with the logic checks will be checked with the hospital or clinic case notes and the woman and her clinician. 4.2 Data analysis Patient demographics and characteristics will be reported using descriptive statistics, and comparisons between the three arms will be done using the Chi-square test or Fisher's exact test for categorical variables and the Kruskal-Wallis test for continuous variables. For the primary outcome, the investigators will report the proportion of high risk disease along with the 95% confidence intervals, and compared the arms using relative risks and 95% confidence intervals, with the Colposcopy arm as the reference group. Sensitivity, specificity, positive predictive value, and negative predictive value will be calculated in each arm. SAS 9.3 (SAS Institute Inc., Cary, NC, USA) will be used for analyses. Non-inferiority of repeat Pap and HPV testing follow-up policies strategies, relative to colposcopy is hypothesized. Inferiority will be defined as the smallest relative disadvantage in sensitivity that would be of practical clinical interest. Investigators assume this margin to be less than 1% based on informal discussions with colposcopists in Calgary and in the absence of relevant publications.