The prevalence of human papillomavirus in ovarian cancer: a systematic review.

OBJECTIVE: We performed a systematic review and a meta-analysis to estimate the prevalence of human papillomavirus (HPV) in ovarian cancer. METHODS: A comprehensive search of the Cochrane Library, MEDLINE, CANCERLIT, LILACS, Grey literature and EMBASE was performed for articles published from January 1990 to March 2012. The following MeSH (Medical Subject Headings) terms were searched: "ovarian tumor" or "ovarian cancers" and "HPV" or "human papillomavirus." Included were case-control and cross-sectional studies, prospective or retrospective, that evaluated clinical ovarian cancer and provided a clear description of the use of in situ hybridization, Southern blot hybridization, and polymerase chain reaction. The statistical analysis was performed using REVMAN 5.0. RESULTS: In total, 24 primary studies were included in this meta-analysis. Studies from 11 countries on 3 continents contained data on HPV and ovarian cancer, including 889 subjects. Overall, the HPV prevalence in patients with ovarian cancer was 17.5 (95% confidence interval [CI], 15.0%-20.0%). Human papillomavirus prevalence ranged from 4.0% (95% CI, 1.7%-6.3%) in Europe to 31.4% (95% CI, 26.9%-35.9%) in Asia. An aggregate of 4 case-control studies from Asia showed an odds ratio of 2.48 (95% CI, 0.64-9.57). CONCLUSIONS: We found a high prevalence of HPV-positive DNA in ovarian cancer cases, but the role of HPV in ovarian cancer remains inconclusive. Further studies are needed to control case to answer this question.

[Meta analysis of the use of Bayesian networks in breast cancer diagnosis]. / Metanálise do uso de redes bayesianas no diagnóstico de câncer de mama.

The aim of this study was to determine the accuracy of Bayesian networks in supporting breast cancer diagnoses. Systematic review and meta-analysis were carried out, including articles and papers published between January 1990 and March 2013. We included prospective and retrospective cross-sectional studies of the accuracy of diagnoses of breast lesions (target conditions) made using Bayesian networks (index test). Four primary studies that included 1,223 breast lesions were analyzed, 89.52% (444/496) of the breast cancer cases and 6.33% (46/727) of the benign lesions were positive based on the Bayesian network analysis. The area under the curve (AUC) for the summary receiver operating characteristic curve (SROC) was 0.97, with a Q* value of 0.92. Using Bayesian networks to diagnose malignant lesions increased the pretest probability of a true positive from 40.03% to 90.05% and decreased the probability of a false negative to 6.44%. Therefore, our results demonstrated that Bayesian networks provide an accurate and non-invasive method to support breast cancer diagnosis.

Metanálise do uso de redes bayesianas no diagnóstico de câncer de mama / Meta analysis of the use of Bayesian networks in breast cancer diagnosis / Metaanálisis del uso de los modelos bayesianos en el diagnóstico de cáncer de mama

O objetivo deste estudo foi avaliar a acurácia das redes bayesianas no apoio ao diagnóstico de câncer de mama. Foram realizadas revisão sistemática e metanálise, que incluíram artigos e relatórios publicados entre Janeiro de 1990 e Março de 2013. Foram incluídos estudos transversais prospectivos e retrospectivos que avaliaram a acurácia do diagnóstico de lesões de mama (condição alvo) usando as redes bayesianas (teste em avaliação). Quatro estudos primários que incluíram 1.223 lesões de mama foram analisados, 89,52% (444/496) dos casos de câncer de mama e 6,33% (46/727) das lesões benignas foram positivas tendo-se como base a análise das redes bayesianas. A área dentro da curva SROC (característica de operação do receptor sumária) foi 0,97, com um valor Q* de 0,92. O uso de redes bayesianas no diagnóstico de lesões malignas aumentou a probabilidade pré-teste para um verdadeiro positivo de 40,03% para 90,05% e diminuiu a probabilidade de um falso negativo para 6,44%. Portanto, nossos resultados demonstraram que as redes bayesianas oferecem um método acurado e não invasivo no apoio ao diagnóstico de câncer de mama.

The aim of this study was to determine the accuracy of Bayesian networks in supporting breast cancer diagnoses. Systematic review and meta-analysis were carried out, including articles and papers published between January 1990 and March 2013. We included prospective and retrospective cross-sectional studies of the accuracy of diagnoses of breast lesions (target conditions) made using Bayesian networks (index test). Four primary studies that included 1,223 breast lesions were analyzed, 89.52% (444/496) of the breast cancer cases and 6.33% (46/727) of the benign lesions were positive based on the Bayesian network analysis. The area under the curve (AUC) for the summary receiver operating characteristic curve (SROC) was 0.97, with a Q* value of 0.92. Using Bayesian networks to diagnose malignant lesions increased the pretest probability of a true positive from 40.03% to 90.05% and decreased the probability of a false negative to 6.44%. Therefore, our results demonstrated that Bayesian networks provide an accurate and non-invasive method to support breast cancer diagnosis.

El objetivo de este estudio fue evaluar la exactitud de las redes bayesianas para apoyar el diagnóstico de cáncer de mama. Se realizó una revisión sistemática y un metaanálisis, que incluyeron artículos y estudios publicados entre enero de 1990 y marzo de 2013. Se incluyeron estudios transversales prospectivos y retrospectivos, que evaluaron la exactitud del diagnóstico de lesiones mamarias (condición de destino), utilizando redes bayesianas (prueba de evaluación). Se analizaron cuatro estudios que incluyeron 1.223 lesiones de mama primarias, un 89,52% (444/496) de los casos de cáncer de mama, y un 6,33% (46/727) de las lesiones benignas se tomaron como base de análisis de las redes bayesianas. El área bajo la curva SROC (característica operativa del receptor) fue de un 0,97, con un valor de Q* de un 0,92. El uso de las redes bayesianas en el diagnóstico de las lesiones malignas aumentó la probabilidad pre test de un verdadero positivo desde un 40,03% a un 90,05%, y la disminución de la probabilidad de un falso negativo de un 6,44%. Por lo tanto, nuestros resultados demuestran que las redes bayesianas ofrecen un método preciso y no invasivo en el apoyo del diagnóstico del cáncer mamario.