Leeson

2. Colposcopic treatment standards
Simon Leeson
Consultant Gynaecologist and Oncologist
Honorary Senior Lecturer
Betsi Cadwaladr University Health Board
Canolfan Lawfeddygol Canser Gynaecolegol Gogledd Cymru
North Wales Gynaecological Cancer Surgical Centre

3. Aims
• To provide recommendations for the use of colposcopy to treat
precancerous cervical lesions.
• To provide recommendations for follow-up to assess effectiveness of
treatment.

4. Objectives
• To assess current standards for colposcopic treatment in Europe.
• To evaluate short-term follow-up standards and any variation in
context of national populations/ screening programmes.
• Assess how agreed standards implemented/ measured.
• Record improvements of care as consequence of quality measures.

5. Introduction
• Need data for quality assurance. Data to be collected prospectively,
validated/ be exchangeable. In order to allow comparison/
benchmarking with others, same data items need to be collected.
• It follows that basic organisational issues essential for QA to occur
include:
agreed minimum database
standardised data
organisation of data collection/ exchange i.e. who is responsible for data
collection
collation/ analysis…

6. Introduction
• Need data for quality assurance. Data to be collected prospectively,
validated/ be exchangeable. In order to allow comparison/
benchmarking with others, same data items need to be collected.
• It follows that basic organisational issues essential for QA to occur
include:
agreed minimum database
standardised data
organisation of data collection/ exchange i.e. who is responsible for data
collection
collation/ analysis…

7. In 2012 EFC agreed 6 quality measures by Delphi process1.
• % excisional treatments/ conisations containing CIN2+ (85%)
• % cases having a colposcopic examination prior to treatment for
abnormal cervical cytology (100%)
• % excised lesions/ conisations with clear margins (80%)
• documentation of whether the squamocolumnar junction has been
seen or not (100%)
• number of colposcopies personally performed each year for a low-
grade/ minor abnormality on cervical cytology (>50)
• number of colposcopies personally performed each year for a high-
grade/ major abnormality on cervical cytology (>50).
1Moss et al. 2013

8. Previously published data
• Audit of all invasive cervical cancer cases in Sweden 1999–2001
<67yr/ prev abnormal cytology in prior 0.5–6.5yrs (n=143 cases) as
population-based case control study (Silfverdal et al. 2011). Histology
taken from 55 cases of which 24 treated. Ablative therapy increased
OR compared with excisional (3.82; 1.01–14.4), laser conisation
decreased OR compared with LLETZ (0.06; 0.01–0.36).
• ‘This underlines the need of proficiency training of the clinician in
surgical techniques as well as of careful follow-up after treatment.
Larger studies are warranted to further evaluate the effectiveness of
“see and treat”, LLETZ and ablative treatment in preventing invasive
disease…’

9. • Population based cohort study (1958-2002) - women with previous CIN3
had increased risk invasive cervical cancer compared with general female
population (standardised incidence ratio 2.34, 95% CI 2.18-2.50) but risk
still increased after 25yrs. Risk accentuated in women aged >50 (Strander
et al. 2007).
• These 2 papers show volume of tissue removed/ excision of TZ appears to
equate to risk of cancer.
• ‘Data suggest an increased risk of invasive cancer after long term follow-up
as treatment of CIN has become less radical over time’ (Strander and
Adolfsson, 2014).

10. Brussels satellite meeting 6-7.11.15
Marc Arbyn - Excision margins
• Whatever excisional treatment procedure, overall 23% median +ve
margins. Meta-analysis of margin status for prediction of failure of
treatment sens of 64%, spec of 84%, +ve PPV 22% for CIN2+. If -ve
3.2% (cNPV). RR 6.6 and so relevant QI.
• Association between margin status/ treatment failure. Also
association with excision/ pre-term labour. No paper looks at both
obstetric/ cancer outcomes combined.
• Strander et al, (2014) showed that compared to 1958-70 increase in
treatment failure after 1970 with ‘excess RR’ of new cancer cases
from 2001-8 (1.94 (CI 1.45-2.59)). Less hysterectomies/ cones +small
loops may be to blame following concern on poor obstetric outcome.

11. • ‘It is worrying that Strander and colleagues found that women who
received local treatment more recently were at greater risk of
developing cervical/ vaginal cancer’ (Arbyn et al, 2014).
• Summary: margin status could be used for quality of treatment but
other QIs could be relevant.

12. • Loop length/ numbers of treatments, HPV testing, age could most
accurately predict risk of failure but:
for treatment length/ risk of preterm labour - if <10mm no risk of preterm
delivery. If >10mm RR 2. Increasing excision length associated with significant
increase in risk of preterm delivery, with estimated 6% increase in risk/
additional mm of tissue excised (OR 1.06, 95% CI 1.03-1.09).
having had 2+ LLETZs increased risk almost x4 for subsequent preterm delivery
compared with no LLETZ before delivery, and almost x2 risk compared with 1
LLETZ before delivery (Noehr et al, 2009).
severity of cone histology/ time since LLETZ not associated with risk of preterm
delivery.
series of 3,426 LLETZ procedures, women aged >50 with CIN at margins of
excision constituted minority high risk group. CIN3 extending to lateral/ deep
margins should be offered retreatment rather than surveillance (Flannelly et al,
2001).

13. • Some treatments performed without colposcopy. In France, 76% not
performed under colposcopic vision. Similar situation in Spain.

14. 1Bornstein et al. 2012

15. Loop length

16. Type 1
• Completely ectocervical
• Fully visible
• Small or large
Transformation Zone Classification

18. Type 2
• Has endocervical
component
• Fully visible
• May have ectocervical
component, which may be
small or large
Transformation Zone Classification

20. Transformation Zone Classification
Type 3
• Has endocervical
component
• Not fully visible
• May have ectocervical
component, which may be
small or large

22. 8.4.3 Length of excision
• goal of excision is to remove all abnormal epithelium.
Type I cervical TZ:
• for treating ectocervical lesions, excisional techniques
should remove tissue to length of >7mm (95%), though aim should be
to remove <10mm in women of reproductive age.
Type II cervical TZ:
• excisional techniques should remove tissue to length of 10-15mm,
depending on position of SCJ within endocervical canal.
Type III cervical TZ:
• excisional techniques should remove tissue to length of 15-25mm.

23. Evidence
Histological assessment of depth of crypt involvement by CIN3 has
shown mean depth of 1-2mm with max. 5.22mm/ mean +3 SDs
(99.7%) of 3.80mm.
Incomplete excision, especially of endocervical margin, is important
adverse factor for recurrence.
In women <35, excisions >10mm in length not associated with
improved recurrence rates. Increased risk of preterm delivery after
LLETZ >10mm in length.
In 1 study loop excisions >12mm in length associated with x3 increase
in preterm delivery1, however, in recent nested case control study
within NHSCSP, absolute risk of preterm delivery 8% for excisions 10-
14mm, rising to 18% for excisions >20mm in length2.
1Khalid et al. 2012
2Castanon et al. PaCT Study Group. 2014

24. QIs modified after discussion with delegates:
• Women who undergo excisional treatments should have a definitive
histology of CIN2+ (>85%). Definitive histology is the highest grade
from any diagnostic or therapeutic biopsies.
• A colposcopic examination should be performed prior to treatment
for an abnormal cervical screening test (100%).
• Colposcopy TZ type (1,2 or 3) should be documented (100%).
But for further debate:
• Excisional treatment specimens should have clear margins (80%)
Consider which margin for margin status (ecto/ endo or both).
EFC Brussels Nov, 2015

25. UK (%) Germany
(%)
Italy (%)
% excisional treatments/ conisations
containing CIN2+ (85%)
% cases having a colposcopic
examination prior to treatment for
abnormal cervical cytology (100%)
% excised lesions/ conisations with clear
margins (80%)
Documentation of whether the
squamocolumnar junction has been seen
or not (100%)

26. UK (%) Germany
(%)
Italy (%)
% excisional treatments/ conisations
containing CIN2+ (85%)
91 83 68
% cases having a colposcopic
examination prior to treatment for
abnormal cervical cytology (100%)
100 94 98
% excised lesions/ conisations with clear
margins (80%)
25 73
Documentation of whether the
squamocolumnar junction has been seen
or not (100%)
93 95 99
>80% clear margins in excised lesions/conizations not considered useful
by some German clinics and not documented.1
1Luyten et al, 2015

28. HPV test of cure (ToC)
• Meta-analysis of 5 studies including 1032 women treated by excision
of CIN2/3
• Sensitivity for HCII to detect CIN2+ 90.7% vs. 76.6% for cytology
(threshold ASCUS)
• Specificity of 74.6% vs. 89.7% (Chan et al, 2009)
• Incidence recurrent CIN2+ 6.6%
~15-23% increase in detection recurrent HG-CIN (Arbyn et al, 2006)
• Benefit of –ve HPV test is rapid return to 3yrly recall.

29. Algorithm for Test of Cure
Treatment
6 month follow up
Cytology Cytology
High grade dyskaryosis or worse Normal, borderline or low grade dyskaryosis
Colposcopy HPV +ve HPV –ve
Approx 14-28% of all treatments
Treat, or if N (following ToC) 3 year recall
for 3 year recall
2.9% CIN2+ @ 3 years*
*Kitchener et al, 2008

31. NHS
Cancer Screening Programmes
February, 2015
Test +ve rate 12.9%
PPV HPV testing 33.3%
PPV cytology 58.4%

32. NHS
Cancer Screening Programmes
February, 2015
HPV referral rate 4.3%
Cytology referral rate
4.1%

34. HPV test +ve for >25 yrs
16/18 or cytology +ve 16/18 or cytology -ve
HPV test @
12 months
Post vaccinated era
3 year recall
HPV +ve and reflex cytology HPV -ve
Margin status –
endo/ ecto/ lateral
Excision length
If treatment for CIN

35. HPV test +ve for >25 yrs
16/18 or cytology +ve 16/18 or cytology -ve
HPV test @
12 months
Post vaccinated era
3 year recall
HPV +ve and reflex cytology HPV -ve
Margin status –
endo/ ecto/ lateral
Excision length
If treatment for CIN
Modify loop length
for TZ type

36. Conclusions
• Decide if we need additional treatment standards despite Delphi.
record conisation length (if treatment excisional)
record if deep/lateral margin involved
record HPV or cytology (threshold ASCUS) +ve rate at 6-8 months
post treatment
and record recurrent CIN2+ rate within 12/24 months of treatment.
• Women who undergo excisional treatments should have a definitive histology of
CIN2+ (>85%). Definitive histology is the highest grade from any diagnostic or
therapeutic biopsies.
• A colposcopic examination should be performed prior to treatment for abnormal
cervical screening test (100%).
• Colposcopy TZ type (1,2 or 3) should be documented (100%).
• Excisional treatment specimens should have clear margins (80%).

38. The 2017 Annual Scientific Meeting of the BSCCP
Save the Date!
www.bsccpconference.co.uk
Wednesday 3rd to Friday 5th May 2017
Cardiff City Football Stadium, Cardiff

39. Canolfan Lawfeddygol Canser Gynaecolegol Gogledd Cymru
North Wales Gynaecological Cancer Surgical Centre