​Klinični pomen mejnih rezultatov triažnega testa za človeške papilomaviruse v državnem presejalnem programu ZORA

Kos, Jerneja (author), Kloboves Prevodnik, Veronika (mentor), Poljak, Mario (comentor)

Uvod: Test Hybrid Capture 2 High-Risk HPV DNA (HC2) (Qiagen, Hilden, Nemčija) se v Sloveniji več kot 10 let uporablja kot triažni test pri naslednjih indikacijah: atipične ploščate celice, neopredeljene (APC-N), ploščatocelična intraepitelijska lezija nizke stopnje pri ženskah, starejših od 35 let (PIL-NS ? 35), atipične žlezne celice, neopredeljene (AŽC-N), cervikalna intraepitelijska neoplazija stopnje 1 (CIN 1) in cervikalna intraepitelijska neoplazije stopnje dve in več (CIN 2+) po zdravljenju. Test na podlagi ugotovljene prisotnosti ali odsotnosti človeških papilomavirusov (HPV, angl. human papillomaviruses) razdeli ženske na tiste z večjim in manjšim tveganjem za razvoj raka materničnega vratu. Deluje na tehnologiji ojačanja signala, s katero določi prisotnost ali odsotnost 13 visoko tveganih genotipov HPV. Rezultati testiranja s HC2 se izkazujejo kot razmerje med relativnimi svetlobnimi enotami (angl. relative light units, RLU) in mejno vrednostjo (angl. cut-off, CO). Za mejo pozitivnosti je proizvajalec postavil vrednost 1,0 RLU/CO, torej so vzorci z rezultati, ki so večji ali enaki 1,0 RLU/CO, HPV-pozitivni, vzorci pod 1,0 RLU/CO pa HPV-negativni. Kot pri vsakem laboratorijskem testiranju je rezultat triažnega testa HPV v določenem številu vzorcev mejno pozitiven ali mejno negativen. Med vzorci z mejnimi rezultati so lažno negativni in lažno pozitivni rezultati, ki lahko povzročijo napačno klinično obravnavo testiranih žensk. Večina laboratorijskih testov ima zato jasno postavljena navodila za obravnavo vzorcev z mejnimi rezultati. Nasprotno pa proizvajalec testa HC2 nima jasno zapisanih navodil za nadaljnjo obravnavo vzorcev, odvzetih v transportni medij STM (angl. Specimen Transport Medium), pri katerih je rezultat testiranja mejen. Namen: Opredeliti klinični pomen mejnih rezultatov triažnega testa HPV (definiranih kot vrednosti med 0,4 in 4,0 RLU/CO), ugotoviti analitične razloge za njihov nastanek ter ovrednotiti različne postopke za njihovo najprimernejšo razrešitev. Metode: Raziskavo smo razdelili na tri dele. V prvi del smo vključili 594 žensk, ki so bile v okviru raziskave L3-5512 v skladu s slovenskimi smernicami napotene na kolposkopijo. Tam jim je ginekolog odvzel bris materničnega vratu za test HPV. V drugi del smo vključili 47.225 žensk iz državnega presejalnega programa ZORA, ki so imele med letoma 2011 in 2017 zabeležen bris materničnega vratu, na podlagi katerega so bile napotene na prvi triažni test HPV. V tretji del smo vključili 1.017 žensk iz državnega presejalnega programa ZORA, ki so bile med septembrom 2019 in junijem 2020 v skladu s slovenskimi smernicami napotene na triažni test HPV in pri katerih je bila izmerjena vrednost triažnega testa HC2 v mejnem območju 0,4–4,0 RLU/CO. Za vse ženske smo zbrali vrednosti RLU/CO rezultata testa HC2. Rezultate testa HC2 smo vrednotili glede na proizvajalčevo mejo pozitivnosti (1,0 RLU/CO) in glede na različna mejna območja v sivi coni od 0,4–4,0 RLU/CO. V prvem delu raziskave smo določiti delež žensk z mejnim rezultatom testa HC2, izračunali njihovo tveganje za CIN 2+, določili občutljivost in specifičnost testa HC2 v mejnem območju ter ocenili smiselnost ponavljanja vzorcev z mejnimi rezultati. V drugem delu raziskave smo rezultate iz prvega dela raziskave preverili na populaciji žensk iz državnega presejalnega programa ZORA. Dodatno smo določili kumulativno incidenco CIN 2+ za vse ženske, ki so bile na podlagi slovenskih smernic napotene na prvi triažni test HPV. Kumulativne incidence CIN 2+ smo izračunali za vsako posamezno indikacijo in za ženske s pozitivnim in negativnim rezultatom testa HPV. Kumulativne incidence CIN 2+ smo vrednotili glede na 2- in 20-odstotni prag tveganja, ki ženske razdelita v tri skupne. Ženske z enoletno kumulativno incidenco CIN 2+, višjo od 20 %, spadajo v skupino z visokim tveganjem in potrebujejo takojšnjo napotitev na kolposkopijo, ženske z enoletno kumulativno incidenco, nižjo od 2 %, spadajo v skupino z nizkim tveganjem in se vrnejo v redno presejanje in ženske z enoletno kumulativno incidenco CIN 2+ med 2% in 20 % spadajo v skupino s srednjim tveganjem in potrebujejo predčasen kontrolni pregled. Preverili smo tudi, ali imajo ženske z mejno negativnim in mejno pozitivnim rezultatom testa HC2 večjo ali manjšo kumulativno incidenco CIN 2+ kot ženske z jasno negativnim ali jasno pozitivnim rezultatom. V tretjem delu raziskave smo primerjali dva možna načina nadaljnje obravnave žensk z mejnim rezultatom triažnega testa HPV in določili analitične razloge za pojav mejnih rezultatov. Rezultati: Rezultati testiranja s testom HC2 so pokazali, da je bilo v povprečju 10 % vzorcev v mejnem območju 0,4?4,0 RLU/CO. Večina žensk z mejnim rezultatom testa HC2 je imela srednje tveganje za CIN 2+ med 2–20 %. Med ženskami z mejno negativnim rezultatom so imele le ženske z APC-N značilno večjo tri- do petletno kumulativno incidenco CIN 2+ od žensk z jasno negativnim rezultatom z APC-N. Vse ženske z mejno pozitivnim rezultatom so imele značilno manjšo tri- do petletno kumulativno incidenco CIN 2+ od žensk z jasno pozitivnim rezultatom. Spreminjanje meje pozitivnosti testa HC2 znotraj mejno negativnega območja 0,4?1,0 RLU/CO je izboljšalo občutljivost testa HC2 med 1,3 in 2,8 odstotne točke ob minimalni izgubi specifičnosti. Spreminjanje meje pozitivnosti testa HC2 znotraj mejno pozitivnega območja 1,0?2,0 RLU/CO je izboljšalo specifičnost testa HC2 med 2,0 in 3,2 odstotne točke ob minimalni izgubi občutljivosti. Nasprotno smo pri dvigu meje pozitivnosti na 4,0 RLU/CO opazili večji padec občutljivosti. Zato je dvig meje pozitivnosti testa HC2 za takojšnjo kolposkopijo z 1,0 na 2,0 RLU/CO varen za vse ženske, ki po enem letu ne presežejo 20-odstotnega praga tveganja za takojšnjo kolposkopijo. Pri ponovnem testiranju mejnih vzorcev s testom HC2 je v povprečju 87,7 % vzorcev ohranilo rezultat prvega testiranja. S ponovnim testiranjem mejnih vzorcev z genotipizacijskim testom HPV je ohranilo rezultat prvega testiranja 62,8 % vzorcev in 69,5 % mejnih vzorcev je vsebovalo enega od 14 visoko tveganih genotipov HPV. Pri 3,6 % žensk z mejnimi rezultati testa HC2 smo histološko potrdili CIN 2+. V mejnih vzorcih polovice žensk s CIN 2+ smo dokazali HPV16/18. S ponovnim testiranjem mejnih vzorcev s testom HC2 ali z genotipizacijskim testom HPV nismo odkrili značilno večjega števila dodatnih CIN 2+. Zaključki: Na podlagi rezultatov raziskave predlagamo naslednja priporočila za klinično obravnavo žensk z mejnimi rezultati testa HC2: 1. Meja pozitivnosti testa HC2 ostane pri trenutni meji 1,0 RLU/CO: a. ženske z mejno pozitivnimi rezultati testa HC2 z napotnimi indikacijami APC-N, PIL-NS ? 35 let, AŽC-N, CIN 1 in CIN 2+ po zdravljenju so obravnavane v skladu s slovenskimi smernicami za ženske s pozitivnim rezultatom triažnega testa HPV; b. v primeru ponavljajočih testov HC2 z mejno pozitivnimi rezultati priporočamo takojšnje testiranje preostanka vzorca z genotipizacijskim testom HPV, ki ima klinično validirano mejno vrednost testa; c. pri ženskah z mejno negativnim rezultatom testa HC2 v območju 0,7?1,0 RLU/CO z napotnimi indikacijami APC-N, PIL-NS ? 35 let, AŽC-N, CIN 1 in CIN 2+ po zdravljenju izvedemo takojšnje testiranje preostanka vzorca z genotipizacijskim testom HPV, ki ima klinično validirano mejno vrednost testa: i. ženske s pozivnim rezultatom testa na HPV16/18 se obravnavajo v skladu s slovenskimi smernicami za ženske s pozitivnim rezultatom triažnega testa HPV, ii. ženske z negativnim rezultatom testa na HPV16/18 se obravnavajo v skladu s slovenskimi smernicami za ženske z negativnim rezultatom triažnega testa HPV. Iz populacijskega vidika bi lahko razmislili o dvigu meje pozitivnosti testa HC2 za takojšnjo kolposkopijo na vrednost 2,0 RLU/CO, saj je pri tej vrednosti občutljivost testa HC2 še vedno dovolj visoka, hkrati pa s tem zmanjšamo napotitev žensk na nadaljnje diagnostične posege in posledično zmanjšamo obremenjenost žensk ter zdravstvenega sistema. To velja za vse ženske, z izjemo tistih z indikacijo CIN 2+ po zdravljenju, pri katerih je občutljivost testa HC2 občutno manjša v primerjavi z drugimi napotnimi indikacijami.

• Language: Slovenian
• Keywords: test HPV, Hybrid Capture 2, transportni medij STM, siva cona, mejni rezultati, kumulativna incidenca CIN2+
• Work type: Doctoral dissertation
• Organization: MF - Faculty of Medicine
• Year: 2022
• PID: 20.500.12556/RUL-142310
• COBISS.SI-ID: 142481923
• Publication date in RUL: 29.10.2022

Secondary language

• Language: English
• Title: Clinical relevance of borderline results of the triage test for human papillomaviruses in the national screening program ZORA

Abstract

Introduction: The Hybrid Capture 2 High-Risk HPV DNA Test (HC2) (Qiagen, Hilden, Germany) has been used in Slovenia for more than ten years as a triage test for the following indications: atypical squamous cells of undetermined significance (ASC-US), low-grade squamous intraepithelial lesions for women over 35 years old (LSIL 䁥 35), atypical glandular cells of undetermined significance (AGUS), cervical intraepithelial neoplasia grade 1 (CIN 1), and CIN grade 2 or worse (CIN 2+) following treatment. Based on the presence or absence of targeted human papillomaviruses (HPV), the test classifies women as having a higher or lower risk of developing cervical cancer. The test uses signal amplification technology to determine the presence or absence of 13 high-risk HPV genotypes. The HC2 test results are presented as relative light units/cut-off (RLU/CO) and the cut-off value for a positive result is 1.0 RLU/CO. Samples with results greater than or equal to 1.0 RLU/CO are considered HPV positive, while samples below 1.0 RLU/CO are HPV negative. As with any laboratory test, a certain number of tested samples receive borderline positive or borderline negative results. Among these borderline samples, there will be some false positive and false negative results, which can lead to the incorrect clinical management of the tested women. Most laboratory tests therefore have clearly defined instructions for handling samples with borderline results. However, the HC2 test does not have any clearly specified instructions for the further processing of borderline samples that were stored in the Specimen Transport Medium (STM) (Qiagen, Hilden, Germany). Aim: To determine the clinical significance of borderline HPV triage test results (defined as values between 0.4 and 4.0 RLU/CO), to determine the analytical causes of borderline results and to evaluate different management algorithms for women with borderline HPV triage test results. Material and methods: The study was divided into three parts. In the first part, we included 594 women who were referred for a colposcopy in accordance with the Slovenian Cervical Cancer Screening Guidelines as part of the L3-5512 study, where a gynaecologist took a cervical smear for an HPV test; in the second part, we included 47,225 women from the national cervical cancer screening program ZORA who were referred for their first HPV triage test based on a cervical smear collected between 2011 and 2017; in the third part, we included 1,017 women from the national cervical cancer screening program ZORA who were referred for an HPV triage test in accordance with the Slovenian Cervical Cancer Screening Guidelines between September 2019 and June 2020 and whose measured value in the HC2 triage test was within the 0.4 and 4.0 RLU/CO borderline range. For all participants, we collected the RLU/CO values of their HC2 test results. The results of the HC2 test were evaluated according to the manufacturer’s threshold of positivity (1.0 RLU/CO) and according to the different borderline ranges within the grey zone between 0.4 and 4.0 RLU/CO. In the first part of the study, we determined the proportion of women with borderline HC2 test results, calculated their risk of CIN 2+, determined the sensitivity and specificity of the HC2 test in the borderline range, and assessed the significance of retesting samples giving borderline HC2 test results. In the second part of the study, we verified all the results from the first part of the study using a population of women from the cervical cancer screening program. In addition, we determined the cumulative incidence of CIN 2+ for all the women who were referred for their first HPV triage test in accordance with the Slovenian Cervical Cancer Screening Guidelines. We calculated the cumulative incidence of CIN 2+ for each individual referral indication and for HPV-positive and HPV-negative women. Cumulative incidences of CIN 2+ were evaluated according to the 2 % and 20 % risk thresholds, which split the women into three groups. Women with a one-year cumulative incidence of CIN 2+ higher than 20 % belong in the high-risk group and need an immediate referral for a colposcopy, women with a one-year cumulative incidence lower than 2 % belong in the low-risk group and return to regular screening, and women with a one-year cumulative incidence of CIN 2+ between 2 % and 20 % belong to the medium-risk group and require early control. We also determined whether the women with borderline negative and borderline positive HC2 test results had a higher or lower cumulative incidence of CIN 2+ than those with clearly negative or clearly positive HC2 test results. In the third part of the study, we compared two possible protocols for the further management of samples with borderline HC2 test results and determined analytical causes for the borderline results. Results: The results of the HC2 triage test showed that on average, 10% of the samples were within the borderline range between 0.4 and 4.0 RLU/CO. Most women with a borderline HC2 test result had a medium risk of CIN 2+ between 2 % and 20 %. Among the borderline negative women, only women with ASC-US had a significantly higher three- to five-year cumulative incidence of CIN 2+ than the clearly negative women with ASC-US. All the borderline positive women had a significantly lower three- to five-year cumulative incidence of CIN 2+ than the clearly positive women. Changing the threshold of positivity of the HC2 test within the borderline negative range from 0.4 to 1.0 RLU/CO improved the sensitivity of the HC2 test from 1.3 to 2.8 percentage points with a minimal loss of specificity. Changing the threshold of positivity of the HC2 test within the borderline positive range from 1.0 to 2.0 RLU/CO improved the specificity of the HC2 test from 2.0 to 3.2 percentage points with a minimal loss of sensitivity. Raising the threshold of positivity to 4.0 RL/CO caused a greater decrease in sensitivity. Therefore, raising the threshold of positivity of the HC2 test for immediate colposcopy from 1.0 to 2.0 RLU/CO is safe for all women who do not exceed the 20% risk threshold for an immediate colposcopy after one year. After retesting the borderline samples with HC2, an average of 87.7 % of borderline samples kept the result from the first test. After retesting the borderline samples with an HPV genotyping test, 62.8 % of the borderline samples kept their first results. In 69.5 % of the borderline samples, we detected at least one of the 14 high-risk HPV genotypes. 3.6 % of the women with borderline HC2 test results had histologically confirmed CIN 2+. HPV16/18 was detected in borderline samples from half of the women with CIN 2+. Retesting the borderline samples with an HC2 test or an HPV genotyping test did not lead to a significantly higher detection of CIN 2+. Conclusion: Based on the results of our study, we propose the following recommendations for the clinical treatment of women with borderline HC2 test results: 1. The threshold of positivity of the HC2 test remains at the current limit of 1.0 RLU/CO a. borderline positive women with the referral indications ASC-US, LSIL 䁥 35 years, AGUS, CIN 1 and CIN 2+ after treatment are managed according to the Slovenian guidelines for women with a positive HPV triage test result b. in the case of repeated HC2 tests with borderline positive results, we recommend the immediate testing of the residual sample with an HPV genotyping test that has a clinically validated threshold c. for women with a borderline negative HC2 test result from 0.7 to 1.0 RLU/CO with the referral indications ASC-US, LSIL䁥35 years, AGUS, CIN 1 and CIN 2+ after treatment, we recommend the immediate testing of the residual sample with an HPV genotyping test that has a clinically validated threshold i. HPV16/18 positive women are managed according to the Slovenian guidelines for women with a positive HPV triage test result ii. HPV16/18 negative women are managed according to the Slovenian guidelines for women with a negative HPV triage test result From a population point of view, we should consider raising the threshold of the HC2 test for an immediate colposcopy to 2.0 RLU/CO, as the sensitivity of the HC2 test at this RLU/CO value is still sufficiently high. This would reduce the number of referrals of women for further diagnostic tests and consequently reduce the burden on the women and on the health system. This applies to all women, with the exception of women with a CIN 2+ indication after treatment, for which the sensitivity of the HC2 test is significantly lower compared to other referral indications.

• Keywords: HPV test, Hybrid Capture 2, Specimen Transport Medium, borderline results, cumulative incidence of CIN2+