Self-collected HPV Testing Improves Participation in Cervical Cancer - PowerPoint PPT Presentation

Self-collected HPV Testing Improves Participation in Cervical Cancer Screening: Systematic Review and Meta-analysis C . S A R A I R A C E Y P H D S T U D E N T D A L L A L A N A S C H O O L O F P U B L I C H E A LT H U N I V E R S I T Y O F T O R O N T O C O - A U T H O R : D I A N A W I T H R O W W O R L D C A N C E R C O N G R E S S A U G U S T 2 8 T H , 2 0 1 2 M O N T R E A L , Q U E B E C

Cervical Cancer and Cancer Screening Human papillomavirus • (HPV) is a necessary cause of cervical cancer Progressive disease • Screening programs • have dramatically reduced the incidence and mortality of cervical cancer through Participation in cervical cancer screening stagnated in • Pap testing many countries with implemented screening programs ~70% of eligible women in Ontario (target 85%) • How can we engage women to participate in cervical • cancer screening?

HPV Testing & Cervical Cancer Screening  Increasingly used as primary (or co-test) for cervical cancer screening  For detecting CIN2/3+ HPV testing: ! More sensitive (94.6% vs. 55.4%) than Pap testing ! Less specific (94.1% vs.96.8%) than Pap testing  Use in 30+year old women  Improves the utility of Pap testing  Self-collected samples are comparable as physician collected samples ! Self-collection is acceptable to women (Mayrand et al. 2007)

Barriers to Cervical Cancer Screening ! Clinic level barriers  Lack of access to primary care  Inconvenient clinic hours  Lack of transportation to clinic ! Test level barriers  Physical discomfort of Pap test  Embarrassment ! Individual level barriers  Cultural/religious values  Issues with intimate site contact

Objective ! Examine the extent to which providing self- sampled HPV testing increases participation in cervical cancer screening  Compared to Pap testing  Among women who are inadequately screened  In countries with implemented cervical cancer screening programs

Methods – Systematic Review Embase Medline HPV Infection Self- HPV sampling Testing Cervical Cancer

Methods – Systematic Review Embase Medline HPV Infection Self- HPV sampling Testing HPV self-sampling vs. Excluded: Cervical Cancer Pap testing Participation � Editorials, # Control group (Pap commentaries, testing invitation) unpublished, conference # Group allocation stated abstracts Title and abstracted " 168 articles # Inadequately screened � Duplicate datasets women � Ecological study designs � Study origin developing Full text " 17 articles country Reviewed " 10 articles

Review Results HPV Study Arm Control Study Arm Compliance Compliance Number of HPV study of HPV arm Control of Pap Arm Relative Study Location Participants Arm n(%) Pap Arm n(%) Compliance 95% CI 7,7870 Gok et al. 2012 Netherlands 26,409 25,561 261 17 (6.5%) 4.727 (2.984 - 7.488)* (30.8%) Szarewski et al. 2011 UK 3000 1,500 153 (10.2%) 1,500 68 (4.5%) 2.250 (1.701 - 2.967)* Giorgi et al. 2011 Italy 1235 616 121 (19.6%) 619 86 (13.9%) 1.414 (1.098 - 1.821)* Wikstrom et al. 2011 Sweden 4060 2,000 779 (39.0%) 2,060 188 (9.1%) 4.268 (3.685 - 4.943)* Virtanen et al. 2011 Finland 8699 2,397 756 (31.5%) 6,302 1,631 (25.9%) 1.219 (1.134 - 1.310*) Castle et al. 2011 US 119 77 62 (80.5%) 42 17 (40.5%) 1.989 (1.357 - 2.917)* Lazcano-Ponce et al. 2011 Mexico 25,061 9,371 9,202 (98.2%) 12,731 11,054 (86.8%) 1.131 (1.123 - 1.139)* Piana et al. 2011 France 9.334 3,552 939 (26.4%) 4,305 311 (7.2%) 3.659 (3.245 - 4.127)* Gok et al. 2010 Netherlands 27,163 26,886 7,455 (27.7%) 277 46 (16.6%) 1.670 (1.282 - 2.175)* Bias et al. 2007 Netherlands 2624 2,352 806 (34.3%) 272 48 (17.6%) 1.942 (1.493 - 2.525)* *Statistically significant <0.01

Review Results HPV Study Arm Control Study Arm Compliance Compliance Number of HPV study of HPV arm Control of Pap Arm Relative Study Location Participants Arm n(%) Pap Arm n(%) Compliance 95% CI 7,7870 Gok et al. 2012 Netherlands 26,409 25,561 261 17 (6.5%) 4.727 (2.984 - 7.488)* (30.8%) Szarewski et al. 2011 UK 3000 1,500 153 (10.2%) 1,500 68 (4.5%) 2.250 (1.701 - 2.967)* Giorgi et al. 2011 Italy 1235 616 121 (19.6%) 619 86 (13.9%) 1.414 (1.098 - 1.821)* Wikstrom et al. 2011 Sweden 4060 2,000 779 (39.0%) 2,060 188 (9.1%) 4.268 (3.685 - 4.943)* Virtanen et al. 2011 Finland 8699 2,397 756 (31.5%) 6,302 1,631 (25.9%) 1.219 (1.134 - 1.310*) Castle et al. 2011 US 119 77 62 (80.5%) 42 17 (40.5%) 1.989 (1.357 - 2.917)* Lazcano-Ponce et al. 2011 Mexico 25,061 9,371 9,202 (98.2%) 12,731 11,054 (86.8%) 1.131 (1.123 - 1.139)* Piana et al. 2011 France 9.334 3,552 939 (26.4%) 4,305 311 (7.2%) 3.659 (3.245 - 4.127)* Gok et al. 2010 Netherlands 27,163 26,886 7,455 (27.7%) 277 46 (16.6%) 1.670 (1.282 - 2.175)* Bias et al. 2007 Netherlands 2624 2,352 806 (34.3%) 272 48 (17.6%) 1.942 (1.493 - 2.525)* *Statistically significant <0.01

Review Results HPV Study Arm Control Study Arm Compliance Compliance Number of HPV study of HPV arm Control of Pap Arm Relative Study Location Participants Arm n(%) Pap Arm n(%) Compliance 95% CI 7,7870 Gok et al. 2012 Netherlands 26,409 25,561 261 17 (6.5%) 4.727 (2.984 - 7.488)* (30.8%) Szarewski et al. 2011 UK 3000 1,500 153 (10.2%) 1,500 68 (4.5%) 2.250 (1.701 - 2.967)* Giorgi et al. 2011 Italy 1235 616 121 (19.6%) 619 86 (13.9%) 1.414 (1.098 - 1.821)* Wikstrom et al. 2011 Sweden 4060 2,000 779 (39.0%) 2,060 188 (9.1%) 4.268 (3.685 - 4.943)* Virtanen et al. 2011 Finland 8699 2,397 756 (31.5%) 6,302 1,631 (25.9%) 1.219 (1.134 - 1.310*) Castle et al. 2011 US 119 77 62 (80.5%) 42 17 (40.5%) 1.989 (1.357 - 2.917)* Lazcano-Ponce et al. 2011 Mexico 25,061 9,371 9,202 (98.2%) 12,731 11,054 (86.8%) 1.131 (1.123 - 1.139)* Piana et al. 2011 France 9.334 3,552 939 (26.4%) 4,305 311 (7.2%) 3.659 (3.245 - 4.127)* Gok et al. 2010 Netherlands 27,163 26,886 7,455 (27.7%) 277 46 (16.6%) 1.670 (1.282 - 2.175)* Bias et al. 2007 Netherlands 2624 2,352 806 (34.3%) 272 48 (17.6%) 1.942 (1.493 - 2.525)* *Statistically significant <0.01

Review Results HPV Study Arm Control Study Arm Compliance Compliance Number of HPV study of HPV arm Control of Pap Arm Relative Study Location Participants Arm n(%) Pap Arm n(%) Compliance 95% CI 7,7870 Gok et al. 2012 Netherlands 26,409 25,561 261 17 (6.5%) 4.727 (2.984 - 7.488)* (30.8%) Szarewski et al. 2011 UK 3000 1,500 153 (10.2%) 1,500 68 (4.5%) 2.250 (1.701 - 2.967)* Giorgi et al. 2011 Italy 1235 616 121 (19.6%) 619 86 (13.9%) 1.414 (1.098 - 1.821)* Wikstrom et al. 2011 Sweden 4060 2,000 779 (39.0%) 2,060 188 (9.1%) 4.268 (3.685 - 4.943)* Virtanen et al. 2011 Finland 8699 2,397 756 (31.5%) 6,302 1,631 (25.9%) 1.219 (1.134 - 1.310*) Castle et al. 2011 US 119 77 62 (80.5%) 42 17 (40.5%) 1.989 (1.357 - 2.917)* Lazcano-Ponce et al. 2011 Mexico 25,061 9,371 9,202 (98.2%) 12,731 11,054 (86.8%) 1.131 (1.123 - 1.139)* Piana et al. 2011 France 9.334 3,552 939 (26.4%) 4,305 311 (7.2%) 3.659 (3.245 - 4.127)* Gok et al. 2010 Netherlands 27,163 26,886 7,455 (27.7%) 277 46 (16.6%) 1.670 (1.282 - 2.175)* Bias et al. 2007 Netherlands 2624 2,352 806 (34.3%) 272 48 (17.6%) 1.942 (1.493 - 2.525)* *Statistically significant <0.01

RR (95% CI) 4.73 (2.98, 7.49) 4.73 (2.98, 7.49) Meta-analysis 2.25 (1.71, 2.97) 2.25 (1.71, 2.97) Results 1.41 (1.10, 1.82) 1.41 (1.10, 1.82) All 10 studies were used • 4.27 (3.68, 4.94) 4.27 (3.68, 4.94) to calculate combined 1.22 (1.13, 1.31) 1.22 (1.13, 1.31) measure 1.99 (1.36, 2.92) 1.99 (1.36, 2.92) Random Effects Model • 1.13 (1.12, 1.14) 1.13 (1.12, 1.14) 3.66 (3.24, 4.13) 3.66 (3.24, 4.13) I 2 statistic for • heterogeneity 1.67 (1.28, 2.18) 1.67 (1.28, 2.18) 1.94 (1.49, 2.53) 1.94 (1.49, 2.53) Sensitivity analysis was • 2.14 (1.30 – 3.52) .000) 2.14 (1.30, 3.52) 2.14 (1.30, 3.52) conducted to remove the Overall (I-squared = 99.5%, p = 0.000) two non-European effects analysis NOTE: Weights are from random effects analysis studies .1 1 1 10 Relative Risk Combined relative compliance = 2.14 (95%CI 1.30 – 3.52)  Sensitivity analysis  Combined relative compliance = 2.34 (1.47 – 3.70 95% CI) !

Conclusions  Overall findings: ! Providing HPV self-testing to inadequately screened women could improve participation 2- fold in cervical cancer screening compared to offering Pap tests ! Majority of studies were conducted in urban European settings ! Provision of HPV self-testing requires investment in infrastructure to ensure appropriate follow-up and care are available

Acknowledgements Co-author: Diana Withrow Dissertation Committee: Dr. Dionne Gesink, Supervisor Dr. Ann Burchell Dr. Tom Wong Funding: University of Toronto