Evidence Based Evaluation of Anal Dysplasia Screening : Ready for Prime Time?

1. Evidence Based Evaluation of Anal Dysplasia Screening: Ready for Prime Time? Wm. Christopher Mathews, MD San Diego AETC, UCSD Owen Clinic

2. Clinical Case • 50 year old asymptomatic physician with HIV infection presented for routine care in May 1999 • CD4=350, HIV viral load 35,000 • Physical exam normal except for 3 cm irregular hard anal mass • Biopsy: invasive squamous cell carcinoma

3. Clinical Case -2 • Resection had positive margins • He was treated with radiotherapy and mitomycin C + 5FU • Severe disabling radiation proctitis • Biopsy at end of treatment showed residual tumor • Abdominal perineal resection in 11/99 • Small bowel obstructionileocolic anastasmosis (3/00) • Bilateral hydronephrosis and renal failure • Declined intervention • Viral load <50 prior to withdrawal of therapy

4. Audience Response Questions • Is anal dysplasia screening with Pap smears being routinely done in your primary clinical site? • Yes • No • Is high resolution anoscopy (HRA) available to patients receiving care at your primary clinical site • Yes, on site • Yes, by outside referral • Not available

5. Evidence-based screening: What kind of evidence is needed? • How important is the health condition to be sought in terms of frequency, morbidity, and mortality? • How good is the screening test in terms of accuracy, safety, simplicity, acceptability (to patients and providers), labeling effects, and financial costs? • How strong is the evidence that the outcome will improve if treatment is given after screening rather than at the time the patient presents with symptoms? (Fletcher, S. ACP Journal Club. 1998; 128:A12)

6. Lead-time Bias in Screening (http://bmj.com/epidem/epid.a.html)

7. How important is the health condition to be sought in terms of frequency, morbidity, and mortality?

8. Epidemiology • US Incidence of cervical cancer: 8 / 100,000 (1) • Incidence of anal carcinoma in men with history of anal receptive intercourse: 35 / 100,000 (2) • Current incidence of anal carcinoma similar to that of cervical CA prior to routine PAP screening • Anal CA among HIV + MSM about twice the incidence among HIV – MSM (3) (1) Qaulters et al, 1992. (2) Daling et al, 1987. (3)Goedert et al, 1998

10. Cervical CA as Model for Anal CA • Similar histology • Frequently arise in transformation zone (4) • Both strongly associated with oncogenic strains of HPV (5) • Both associated with squamous intraepithelial lesions (SIL) • Cervical HSIL Cervical CA • Anal HSIL suspected  Anal CA (4) Palefsky, AIDS, 1994. (5) Frisch et al, NEJM, 1997

11. HPV Types and Anal Dysplasia • HPV is double stranded DNA virus (>100 subtypes) • Low risk types (6, 11) associated with condyloma and LSIL • Intermediate risk types (31, 33, 35,45, 51, 52, 56) • High risk types (16, 18) • Present in 64% of invasive cervical CA (6) (6) Bosch et al, JNCI, 1995

12. Frisch et al. J Natl Cancer Inst 2000;92:1500–10

13. Relative risks (RRs) of developing cervical cancer (invasive or in situ), anal cancer (invasive or in situ, males only), Kaposi's sarcoma, or non-Hodgkin's lymphoma in the 4-27 months after the AIDS period, according to the CD4+ T-lymphocyte count within {+/-}1 month of AIDS onset Frisch, M. et al. J Natl Cancer Inst 2000;92:1500-1510

14. Four-year incidence of anal HSIL Year Chin-Hong et al. CID 2002;35:1127-34

15. Chin-Hong et al. CID 2002;35:1127-34 Immune suppression Genetic changes

16. HAART & HIV-associated Anal Cancer • Cohort of 8640 HIV seropositive patients • Overall incidence anal CA: 60/100,000 p-yrs • 120 times higher than age and gender matched controls • Incidence by time period • Pre-HAART 35/100,000 (95% CI: 15-72) • Post-HAART 92/100,000 (95% CI: 52-149) Bower et al. JAIDS 2004;37:1563-1565

17. Palefsky et al. AIDS 2005;19:1407-1414

18. 24-Month Survival and Adjusted Death Hazards (vs. persons having anal cancer without AIDS) Biggar et al. JAIDS 2005;39:293-299

19. Audience Response Questions • How strong is the evidence that invasive anal cancer is an important enough health condition to justify routine screening of HIV infected MSM? • Very strong • Moderately strong • Neither strong nor weak • Moderately weak • Very weak

20. Audience Response Questions • How strong is the evidence that invasive anal cancer is an important enough health condition to justify routine screening of HIV infected women? • Very strong • Moderately strong • Neither strong nor weak • Moderately weak • Very weak

21. I. How good is the screening test in terms of accuracy, safety, simplicity, acceptability (to patients and providers), labeling effects, and financial costs?

22. Bethesda Staging System(2001): CIN/AIN • Atypical squamous cells • Of undetermined significance (ASCUS-US) • Cannot exclude HSIL (ASC-H) • Squamous intraepithelial lesion (SIL) • Low grade SIL (LSIL) • Mild dysplasia/CIN 1 (HPV cellular changes) • High grade SIL (HSIL) • Moderate dysplasia/CIN2 • Severe dysplasia/ CIS / CIN 3 • Squamous cell carcinoma (Wright et al. JAMA 2002;287:2120-2129)

26. Chin-Hong et al. CID 2002:35:1127-34

27. Who to screen? • HIV+ and HIV- MSM • HIV+ women with history of • Anal receptive intercourse • Anogenital warts or HPV infection • Cervical dysplasia • Consider screening all HIV+ men and women

28. ASIL Screening Procedures • Ascertain risk ractors for ASIL • HIV status and degree of immune suppression • History of • Anogenital warts • Anal receptive intercourse • Prior ASIL or CSIL • Symptoms: discharge, pain, bleeding • Tobacco use • Ascertain anal STD risk

29. ASIL Screening Procedures • Examine perianal area, perineum, and genitalia, including inguinal nodes • Obtain PAP smear • before digital rectal exam • No prior douching or enemas • Use dacron, not cotton swab, moistened in tap water • Insert swab 1.5-2 inches • Rotate against anal wall in spiral fashion for 10 seconds while slowly withdrawing • Roll swab across labeled slide and dip in fixative • Perform digital rectal exam

30. High Resolution Anoscopy • Procedure • Informed consent with patient education materials • History and risk factor assessment • Examination of perianal, perineal, and genital regions • Obtain PAP and cultures (if indicated) • Digital rectal exam with lidocaine/water based lubricant mixture • Insert anoscope and through it insert 4X4 gauze soaked in 3% vinegar & rolled around a cotton swab for 1-2 minutes • Reinsert anoscope and examine with coloposcope

31. High Resolution Anoscopy • Lesions first examined after 3% acetic acid application • Suspicious lesions (acetowhite, punctation, atypical vessels, ulcerations) should be biopsied (baby Tischler forceps) • Lugol’s iodine can be applied • Dysplastic lesions turn mustard or light yellow instead of mahogany brown • Counsel regarding bleeding, pain, signs of infection • Follow-up appointment in 1-2 weeks

32. Monitoring after HRA • If PAP HSIL but biopsy not concordant • Repeat PAP and HRA in 3 months • If biopsy HSIL/severe dysplasia or CIS and patient remains untreated • Repeat HRA every 3-4 months • If PAP and HRA concordant LSIL/mild-moderate dysplasia • Repeat HRA in 6-12 months

33. Anal Canal before 3% Acetic Acid (Jay N et al. Dis Colon Rectum 1997;40:923)

34. Anal Transition Zone after Acetic Acid (x40) (Jay N et al. Dis Colon Rectum 1997;40:923)

35. After 3% Acetic Acid (25x)  indicates HGSIL area on biopsy (Jay N et al. Dis Colon Rectum 1997;40:923)

36. HGSIL with Punctation (X40) (Jay N et al. Dis Colon Rectum 1997;40:923)

37. Coarse Mosaicism & Punctation (x40) (Jay N et al. Dis Colon Rectum 1997;40:923)

38. Wart-like HGSIL (x16) (Jay N et al. Dis Colon Rectum 1997;40:923)

39. Flat LGSIL (x16)  Indicates granular surface (Jay N et al. Dis Colon Rectum 1997;40:923)

40. Anal Colposcopic View after Acetic Acid and Lugol’s Iodine (1) Healthy Lugol’s +; (2) LSIL Lugol’s +; (3) HSIL Lugol’s -

41. UCSD Owen Clinic

42. Slippage in Anal Cytology Technique

43. Measures of Agreement • Absolute agreement • Cohen’s kappa • Measure of chance-corrected agreement • How to interpret • 0.93-1.00 Excellent agreement • 0.81-0.92 Very good agreement • 0.61-0.80 Good agreement • 0.41-0.60 Fair agreement • 0.21-0.40 Slight agreement • 0.01-0.20 Poor agreement • ≤0.00 No agreement (Byrt T. Epidemiology 1996;7:561)

44. Overall Reproducibility of Cytologic Diagnosis

45. Variability in Cyto-Histopathologic Agreement among Examiners

46. Practice Effect for HRA Cyto-Histopathologic Agreement

47. The Fuzzy Gold Standard Issue • Because of sampling variability, a histopathologic diagnosis based on biopsy at HRA is not necessarily a criterion or gold standard diagnosis • No equivalent of the cervical LEEP in HRA • Consider patients with HSIL PAPs and biopsies showing lower grade disease • Is the PAP wrong? • Was the high grade lesion missed at HRA?

48. Agreement between PAP and Concurrent Biopsy

49. Agreement between Colposcopic Visual Impression and Biopsy BIOPSY Colpo Impression

50. Prevalence of AIN III or CIS at Biopsy by Simultaneous Cytologic Diagnosis (n=154)