The COVID-19 pandemic has challenged health systems worldwide by decreasing their reserves and effectiveness. In this changing landscape, the urge for reallocation of financial and human resources represents a top priority. In screening, effectiveness and efficiency are most relevant. In the quest against cervical cancer, numerous molecular ancillary techniques detecting HPV DNA or mRNA or other related biomarkers complement morphological assessment by the Papanicolaou test. However, no technique is perfect as sensitivity increases at the cost of specificity. Various approaches try to resolve this issue by incorporating several examination results, such as artificial intelligence are proposed. In this study, 1,258 cases with a complete result dataset for cytology, HPV DNA, HPV mRNA, and p16 were used to evaluate the performance of a self-organizing map (SOM), an unsupervised artificial neural network. The results of the SOM application were encouraging since it is capable of producing maps discriminating the necessary tests and has improved performance.
TopIntroduction
Cervical cancer (CC) is the third most common cancer and the fourth leading cause of cancer death in females worldwide (Jemal, Bray, Center et al., 2011). More than 85% of these cases and deaths are in developing countries, predominantly due to the absence of cervical screening infrastructures to detect and treat precancerous lesions and early-stage cervical cancer. Despite the advances in screening, cervical cancer remains a serious public health problem even in developed countries, in part due to the high percentage of detection failures (Leyden, Manos, Geiger et al., 2005).
CC is known to be caused almost exclusively by human papillomavirus (HPV) infection, the most typical sexually transmitted infection worldwide. About 100 types of HPV viruses that can infect humans have been identified. Among them, 15 possess a high-risk oncogenic potential and might cause CC. Improved understanding of HPV infection and the natural history of cervical neoplasia have resulted in the implementation of the HPV DNA test along with the Papanicolaou test (Pap-test) or as a standalone test (US Preventive Services Task Force, 2018).
Nowadays, several ancillary techniques for CC screening are available. The most established are HPV DNA typing and mRNA identification of the viral E6/E7 oncogenes linked to oncogenic activation. Notably, mRNA typing with nucleic acid-based amplification (NASBA) (Smits, van Gemen, Schukkink et al., 1995) and mRNA-Flow-FISH techniques in screening programs produced excellent results with enhanced PPV; thus reducing unnecessary recalls and referrals to colposcopy (Coquillard, Palao, & Patterson, 2011; Kottaridi, Tsiodras, Spathis et al., 2011; Narimatsu & Patterson, 2005; Sorbye, Fismen, Gutteberg et al., 2010; Trope, Sjoborg, Eskild et al., 2009). At that time, it was evident that the immunocytochemical detection of p16 could increase the diagnostic accuracy of the Pap-test (Tsoumpou, Arbyn, Kyrgiou et al., 2009; Valasoulis, Tsoumpou, Founta et al., 2011); a growing body of literature shows promising results for this biomarker, either as a standalone or combined with Ki-67 assessment.
Several published studies have so far attempted to clarify the role of each technique as a candidate unique test to substitute the Pap test (Benevolo, Vocaturo, Caraceni et al., 2011; Carozzi, 2007; Coquillard, Palao, & Patterson, 2011; Cuschieri & Wentzensen, 2008; Cuzick, Arbyn, Sankaranarayanan et al., 2008; Denton, Bergeron, Klement et al., 2010; Kottaridi, Tsiodras, Spathis, Chranioti, Pappas, Kassanos, Panayiotides, & Karakitsos, 2011; Mathew & George, 2009; Mayrand, Duarte-Franco, Rodrigues et al., 2007; Narimatsu & Patterson, 2005; Naucler, Ryd, Tornberg et al., 2009; Smits, van Gemen, Schukkink, van der Velden, Tjong, Jebbink, & ter Schegget, 1995; Sorbye, Fismen, Gutteberg, & Mortensen, 2010; Trope, Sjoborg, Eskild, Cuschieri, Eriksen, Thoresen, Steinbakk, Laurak, Jonassen, Westerhagen, Jacobsen, & Lie, 2009; Tsoumpou, Arbyn, Kyrgiou, Wentzensen, Koliopoulos, Martin-Hirsch, Malamou-Mitsi, & Paraskevaidis, 2009; Valasoulis, Tsoumpou, Founta, Kyrgiou, Dalkalitsis, Nasioutziki, Kassanos, Paraskevaidis, & Karakitsos, 2011). As a concluding remark, the performance of the methods mentioned above differs significantly, affected by the disease incidence and the prevalence of HPV infection in the population study group; therefore, the application of a single method, despite apparently improving key performance indicators, might not reliably determine an individual women's risk of harboring cervical intraepithelial neoplasia (CIN). However, most published studies (Cuzick, Arbyn, Sankaranarayanan, Tsu, Ronco, Mayrand, Dillner, & Meijer, 2008; Karakitsos, Pouliakis, Meristoudis et al., 2011; Mathew & George, 2009; Mayrand, Duarte-Franco, Rodrigues, Walter, Hanley, Ferenczy, Ratnam, Coutlee, & Franco, 2007; Naucler, Ryd, Tornberg, Strand, Wadell, Elfgren, Radberg, Strander, Forslund, Hansson, Hagmar, Johansson, Rylander, & Dillner, 2009; Schiffman, Kinney, Cheung et al., 2018; Spathis, Kottaridi, Chranioti et al., 2012) conclude that the sensitivity of cytology combined with the HPV DNA test is higher than the sensitivity of each method.