TY - JOUR
T1 - Prevalence and genotype distribution of human papillomavirus in cervical adenocarcinoma (usual type and variants)
T2 - A systematic review and meta-analysis
AU - Reynders, Celia
AU - Lerho, Thomas
AU - Goebel, Emily A.
AU - Crum, Christopher P.
AU - Vandenput, Sandrina
AU - Beaudart, Charlotte
AU - Herfs, Michael
N1 - Funding Information:
The authors thank their clinical colleagues at the University Hospital of Liege (Belgium) and at Brigham and Women's Hospital (Boston, USA) for helpful discussions. This work was supported in part by the FNRS (CDR J.0088.21 [Michael Herfs]), the University of Liege [Crédits Sectoriels de Recherche en Sciences de la Santé 2021–2023 (Michael Herfs)], the Télévie (PDR Televie 7.8507.19 [Michael Herfs]), and the Leon Fredericq Foundation (Michael Herfs). Celia Reynders and Thomas Lerho are Televie fellows. Michael Herfs is a Research Associate at the Belgian Fund for Scientific Research (FNRS).
Funding Information:
The authors thank their clinical colleagues at the University Hospital of Liege (Belgium) and at Brigham and Women's Hospital (Boston, USA) for helpful discussions. This work was supported in part by the FNRS (CDR J.0088.21 [Michael Herfs]), the University of Liege [Crédits Sectoriels de Recherche en Sciences de la Santé 2021–2023 (Michael Herfs)], the Télévie (PDR Televie 7.8507.19 [Michael Herfs]), and the Leon Fredericq Foundation (Michael Herfs). Celia Reynders and Thomas Lerho are Televie fellows. Michael Herfs is a Research Associate at the Belgian Fund for Scientific Research (FNRS).
Publisher Copyright:
© 2023 Wiley Periodicals LLC.
PY - 2023/10
Y1 - 2023/10
N2 - Cervical glandular neoplasms represent a heterogeneous group of tumors for which a comprehensive overview of the involvement of high-risk human papillomaviruses (HPV) in pathogenesis is still lacking. We first searched MEDLINE (PubMed), Embase, and Scopus databases (until October 2022), and systematically reviewed available literature. We then quantitatively estimated both pooled and genotype-specific prevalence of HPV DNA as well as the influence of various factors (e.g., geographical region, histological subtype, tissue/sample type) on computed effect size by means of random effects meta-analysis. In total, 379 studies comprising 17 129 cases of cervical adenocarcinoma were identified. The pooled HPV prevalence was 78.4% (95% confidence interval [95% CI]: 76.2–80.3) with a significant between-study heterogeneity (I2 = 79.4%, Q test p < 0.0001). Subgroup analyses indicated that the effect size differed substantially by geographical region (from 72.5% [95% CI: 68.7–76.1] in Asia to 86.8% [95% CI: 82.2–90.3] in Oceania) (p < 0.0001) and histological subtype of cancer (from 9.8% [95% CI: 5.5–17] in gastric-type to 85% [95% CI: 79.6–89.2] in usual-type cervical adenocarcinoma) (p < 0.0001). HPV16 and HPV18 were by far the most frequently detected viral strains with specific prevalence of 49.8% (95% CI: 46.9–52.6) and 45.3% (95% CI: 42.8–47.8), respectively. When stratified by continent or histologic variant, these genotype-specific results varied in a relatively limited manner. Altogether, these findings support that all histological subtypes of cervical adenocarcinoma are etiologically linked to high-risk HPV but to varying degrees. Therefore, a dual-criteria classification taking into account accurately both morphological and virological aspects could be an interesting evolution of the current binary World Health Organization classification, better reflecting the pathogenic diversity of the disease.
AB - Cervical glandular neoplasms represent a heterogeneous group of tumors for which a comprehensive overview of the involvement of high-risk human papillomaviruses (HPV) in pathogenesis is still lacking. We first searched MEDLINE (PubMed), Embase, and Scopus databases (until October 2022), and systematically reviewed available literature. We then quantitatively estimated both pooled and genotype-specific prevalence of HPV DNA as well as the influence of various factors (e.g., geographical region, histological subtype, tissue/sample type) on computed effect size by means of random effects meta-analysis. In total, 379 studies comprising 17 129 cases of cervical adenocarcinoma were identified. The pooled HPV prevalence was 78.4% (95% confidence interval [95% CI]: 76.2–80.3) with a significant between-study heterogeneity (I2 = 79.4%, Q test p < 0.0001). Subgroup analyses indicated that the effect size differed substantially by geographical region (from 72.5% [95% CI: 68.7–76.1] in Asia to 86.8% [95% CI: 82.2–90.3] in Oceania) (p < 0.0001) and histological subtype of cancer (from 9.8% [95% CI: 5.5–17] in gastric-type to 85% [95% CI: 79.6–89.2] in usual-type cervical adenocarcinoma) (p < 0.0001). HPV16 and HPV18 were by far the most frequently detected viral strains with specific prevalence of 49.8% (95% CI: 46.9–52.6) and 45.3% (95% CI: 42.8–47.8), respectively. When stratified by continent or histologic variant, these genotype-specific results varied in a relatively limited manner. Altogether, these findings support that all histological subtypes of cervical adenocarcinoma are etiologically linked to high-risk HPV but to varying degrees. Therefore, a dual-criteria classification taking into account accurately both morphological and virological aspects could be an interesting evolution of the current binary World Health Organization classification, better reflecting the pathogenic diversity of the disease.
KW - cervical adenocarcinoma
KW - epidemiology
KW - HPV
KW - meta-analysis
UR - http://www.scopus.com/inward/record.url?scp=85174853463&partnerID=8YFLogxK
U2 - 10.1002/jmv.29190
DO - 10.1002/jmv.29190
M3 - Article
AN - SCOPUS:85174853463
SN - 0146-6615
VL - 95
JO - Journal of medical virology
JF - Journal of medical virology
IS - 10
M1 - e29190
ER -
