Interlaboratory variation in the performance of liquid-based cytology: Insights from the ATHENA trial

Abstract

Although it is recognized that cervical cytology is highly subjective, and that there is considerable interlaboratory variation in how slides are evaluated, little is known as to how this impacts the performance of cytology. In the ATHENA trial, liquid-based cytology specimens from 46,887 eligible women !21 years of age were evaluated at four large regional US laboratories, providing a unique opportunity to evaluate the impact of interlaboratory variations on the performance of cervical cytology. All women with abnormal cytology (atypical squamous cells of undetermined significance or higher) were referred to colposcopy, as were all high-risk human papillomavirus (hrHPV)-positive women !25 years of age and a random subset of those !25 years of age who were negative by both hrHPV testing and cytology. Sociodemographics, risk factors for cervical disease, and prevalence of cervical intraepithelial neoplasia (CIN) were similar across the laboratories. There were considerable differences among the laboratories both in overall cytological abnormal rates, ranging from 3.8 to 9.9%, and in sensitivity of cytology to detect CIN grade 2 or worse (CIN21), from 42.0 to 73.0%. In contrast, the hrHPV positivity rate varied only from 10.9 to 13.4%, and the sensitivity of hrHPV testing from 88.2 to 90.1%. These observations suggest that hrHPV testing without cytology should be considered as the initial method for cervical cancer screening.It is widely recognized that cervical cytology is highly subjective, and that this results in considerable interlaboratory variation in the rate of cytological abnormalities. A 2006 survey by the College of American Pathology (CAP) found that the median atypical squamous cells of undetermined significance (ASC-US) rate in different US cytology laboratories varied from 1.7% (lowest 5th percentile of laboratories) to 9.6% (upper 95th percentile of laboratories).1 The median ASC-US rate in the United States at that time was 4.9%. Relatively little is known about how this interlaboratory variation in the rate of cytological abnormalities impacts the performance of cytology for detecting cervical intraepithelial neoplasia (CIN) 2,3 lesions and cervical cancer (CIN21). This is not surprising since there continues to be considerable controversy in the cervical cancer screening community as to the performance of cervical cytology in a "real-world" setting. A metaanalysis published in 1995 reported that the sensitivity of conventional cervical cytology was 49% and the specificity was 62%.2 In 2000, Nanda et al. evaluated the performance of cervical cytology with another meta-analysis for the US Preventive Services Task Force that included studies utilizing liquid-based cytology.3 This meta-analysis found that in the least biased studies, the sensitivity of cytology varied from 30 to 87%, and specificity ranged from 86 to 100%. Such a wide range of performance estimates in different studies is not unexpected given the inherent difficulty of determining the underlying prevalence of high-grade cervical in...

Guía de cribado del cáncer de cuello de útero en España, 2014

“…El material obtenido se conserva inmediatamente tras su extracción en un medio líquido, normalmente de base alcohólica, que permite su almacenaje y transporte, y la extensión se realiza en el laboratorio. Los estudios publicados, incluyendo un metaanálisis, coinciden en que este tipo de citología disminuye los casos inadecuados para diagnóstico, en los que hay que repetir la toma de la muestra, acorta el tiempo de lectura al microscopio y ofrece un discreto aumento de la sensibilidad [76][77][78][79][80][81][82][83][84][85] . Un valor añadido de la citología en medio líquido es que no se utiliza todo el material para realizar el estudio citológico, y el material remanente conservado en el líquido de fijación durante semanas a temperatura ambiente permite realizar técnicas adicionales, moleculares como la determinación de VPH, o de inmunocitoquímica, como la detección de p16/KI67, o futuros nuevos marcadores, evitando así una nueva toma y, por tanto, una visita de la paciente, factor que debe tenerse en cuenta en el cálculo de coste-efectividad de esta técnica 79 .…”

Human Papillomavirus Genotyping Testing Practice in 2014: Results of a College of American Pathologists National Survey

“…Kaiser Permanente Northern California reported follow-up data on 11 The United States ATHENA (Addressing the Need for Advanced HPV Diagnostics) trial reported Roche cobas HPV rates. 12 Among 45 933 women aged 21 years or older undergoing routine cervical cancer screening, prevalence of HR-HPV (14 genotypes), HPV16, and HPV18 was 12.5%, 2.8%, and 1.0%, respectively, and the presence of HPV16 was the most important risk factor for CIN. The overall prevalence of HR-HPV for women with negative Pap test results was 9.8%, 12 lower than the results of our study, which showed positive results in approximately one-fifth of women with negative Pap test results (Table 7).…”

“…12 Among 45 933 women aged 21 years or older undergoing routine cervical cancer screening, prevalence of HR-HPV (14 genotypes), HPV16, and HPV18 was 12.5%, 2.8%, and 1.0%, respectively, and the presence of HPV16 was the most important risk factor for CIN. The overall prevalence of HR-HPV for women with negative Pap test results was 9.8%, 12 lower than the results of our study, which showed positive results in approximately one-fifth of women with negative Pap test results (Table 7). Among the 32 260 women aged 30 years or older with a negative Pap test result, the overall prevalence of HR-HPV detected with the cobas HPV Test was 6.7% and the overall prevalence rates for HPV16, HPV18, and 12 other HR-HPV genotypes were 1.0%, 0.5%, and 5.2%, respectively.…”

“…Several large prospective studies 11,12 have demonstrated that HPV16 status confers the greatest type-specific risk stratification for the development of cervical cancer. Kaiser Permanente Northern California reported follow-up data on 11 The United States ATHENA (Addressing the Need for Advanced HPV Diagnostics) trial reported Roche cobas HPV rates.…”

Is 58% sensitivity for detection of cervical intraepithelial neoplasia 3 and invasive cervical cancer optimal for cervical screening?

“…[13] Furthermore, data on the ASC/squamous intraepithelial lesion ratios of these laboratories place them in a laboratory performance category linked to suboptimal screening sensitivity. [14] Performance of these four US cytology laboratories[15] appears particularly suboptimal when compared to the performance of United Kingdom (UK) laboratories competing against hybrid capture two HPV testing in the first two rounds of the ARTISTIC trial. [16] The performance of the UK LBC laboratories was so strong that the UK investigators commented: “It is difficult to escape the conclusion that LBC was more sensitive in ARTISTIC than earlier conventional cytology”.…”