ReviewAre self-collected samples comparable to physician-collected cervical specimens for human papillomavirus DNA testing? A systematic review and meta-analysis
Introduction
Human papillomavirus (HPV) is now well established as the necessary cause of cervical cancer [1]. Implementation of efficient strategies to study, prevent and control HPV genital tract infection, including vaccination, requires a simple and easy method for HPV detection [2]. Self-sampling has several advantages compared with physician-collected samples for detection of HPV genital infection. Self-sampling is a less costly and non-invasive collection procedure. Self-collected samples can be more easily obtained in settings with limited resources or in populations difficult to reach. Numerous studies have reported lately that self-obtained samples of the anogenital tract in women were accurate and suitable for DNA testing [3], [4], [5], [6], [7]. Currently, there is no consensus on which sampling method is the most effective for HPV DNA testing. However, a sample collected by a health care professional during a speculum examination is generally considered as the “gold standard”.
In 2005, a meta-analysis of 12 studies revealed that self-sampling had a sensitivity ≥ 70% and a specificity of 88% compared to physician-sampling to detect the presence of HPV infection [8]. However, the authors did not discriminate between results obtained with high-risk (HR) and low-risk (LR) HPV types. The prevalence of LR- and HR-HPV types depends on grade of anogenital disease and sampling site. Since the publication of that meta-analysis, relevant studies comparing self- versus physician-obtained sampling have been published [3], [9], [10].
The aim of this systematic review was to update the previously published meta-analysis, by including all available relevant epidemiological studies on the accuracy of self- and physician-obtained samples to identify women infected by HPV in the anogenital tract. We focused on detectability as the outcome of interest, i.e., any positive test, whether originating from self- or from physician-sampling. Our intent was on virological detection only, not on comparing collection methods as per their ability to distinguish HPV testing performances for detecting cervical precancerous lesions.
Section snippets
Search strategy
A systematic literature search of MEDLINE from 1966 through November 2005 and of EMBASE from 1990 through November 2005 was conducted without language restriction entering “human papillomavirus” or “HPV” and “self-” or “patient-” or “auto-” or “physician-” or “clinician-” or “health professional-” combined with “collection” or “obtained” or “sampling” or “testing” as both medical subject heading (MeSH) terms and text words. In addition, bibliographies of studies referenced in a previous
Literature search and study selection
Of the 450 reports that were screened, 66 were relevant for the current study. Reasons for exclusion of 48 of these studies included: absence of comparison with a physician-obtained sample or evaluation of the acceptability of self-sampling only (n = 43), lack of frequency data to construct 2 × 2 tables (n = 4) [12], [13], [14], [15], and duplication of previously published data (n = 1) [16]. The remaining informative 18 reports (5441 participants) were published between 1992 and November 2005 (Table 1
Discussion
In this systematic review, the overall detection of HPV genital infection irrespective of HPV type category was equivalent for self-and physician-obtained samples. Although the presence of lesions may modify the ability to detect HPV, the agreement between self- and physician-obtained samplings was good regardless of the disease prevalence in the tested population. As LR-HPV infect more often the vagina and vulva than HR-HPV, they may be more easily collected by self-sampling and explain the
Acknowledgments
We thank Dr Felippe Lorenzato and Dr Diane Harper for contributing additional data from their studies.
References (28)
- et al.
Randomized clinical trial of PCR-determined human papillomavirus detection methods: self-sampling versus clinician-directed biologic concordance and women's preferences
Am. J. Obstet. Gynecol.
(2002) - et al.
Comparison of type-specific human papillomavirus data from self and clinician directed sampling
Gynecol. Oncol.
(2005) - et al.
Cross-sectional study of patient- and physician-collected cervical cytology and human papillomavirus
Obstet. Gynecol.
(2003) - et al.
Primary cervical cancer screening by self-sampling of human papillomavirus DNA in internal medicine outpatient clinics
Ann. Oncol.
(2004) Self-testing for human papillomavirus using a vaginal swab: placing prevention of cervical cancer in the patient's hands
Ann. Oncol.
(2004)- et al.
Human papillomavirus detection for cervical cancer prevention with polymerase chain reaction in self-collected samples
Am. J. Obstet. Gynecol.
(2002) - et al.
Screening for cervical neoplasia by self-assessment for human papillomavirus DNA
Lancet
(1999) - et al.
Self-administered home cervicovaginal lavage: a novel tool for the clinical–epidemiologic investigation of genital human papillomavirus infections
Am. J. Obstet. Gynecol.
(1992) Human papillomavirus is a necessary cause of invasive cervical cancer worldwide
J. Pathol.
(1999)- et al.
Chapter 12: Human papillomavirus technologies
J. Natl. Cancer Inst. Monogr.
(2003)