Evaluation and Optimization of the Clinical Accuracy of Hybribio's 14 High-Risk HPV with 16/18 Genotyping Assay within the VALGENT-3 Framework

Impact of high-performance human papillomavirus testing to improve cervical cancer screening in China: a prospective population-based multicentre cohort study

OBJECTIVES

The aim of the study was to evaluate the clinical performance of HBRT-H14, a real-time PCR-based assay that separates human papillomavirus (HPV) 16 and HPV18 from 12 other high-risk (HR) HPV types, in population according to Chinese guideline.

METHODS

A total of 9829 eligible women aged 21-64 years from Henan, Shanxi, and Guangdong provinces were performed by HBRT-H14 testing and liquid-based cytology (LBC) screening at baseline and followed up for 3-year. The sensitivity, specificity, positive predictive value (absolute risk), and negative predictive value of LBC diagnosis and HPV testing were calculated for cervical intraepithelial neoplasia grade 2 or worse (CIN2+) Lesions.

RESULTS

At baseline, 80 (0.81%) participants were diagnosed with CIN2+. HR-HPV with reflex LBC had a significantly higher sensitivity (78/80, 97.50% [95% CI, 91.34-99.31%] vs. 62/80, 77.50% [67.21-85.27%], McNemar's test p < 0.001), and a slightly lower specificity (8528/9749, 87.48% [86.80-88.12%] vs. 8900/9749, 91.29% [90.72-91.83%], McNemar's test p < 0.001) than LBC with reflex HR-HPV for CIN2+. 7832 (79.6%) participants completed 3-year follow-up and 172 (2.20%) participants were cumulatively diagnosed with CIN2+. Compared with LBC with reflex HR-HPV, HR-HPV with reflex LBC significantly increased the sensitivity (161/172, 93.60% [88.91-96.39%] vs. 87/172, 50.58% [43.18-57.96%], McNemar's test p < 0.001), but marginally decreased the specificity (6776/7660, 88.46% [87.72-89.16%] vs. 6933/7660, 90.51% [89.83-91.15], McNemar's test p < 0.001). In addition, the absolute 3-year risk of CIN2+ in HPV16/18-positive individuals was as high as 33% (80/238), whereas the risk in the HPV-negative population was only 0.16% (11/6787), much lower than those in the negative for intraepithelial lesion or malignancy population (1.21%, 85/7018). Moreover, similar results were found in women ≥30 years old.

DISCUSSION

The study has indicated that HBRT-14 has a reliable clinical performance for use in cervical screening. The validated HPV test would improve the quality of population screening.

Comparison of Hybribio-H13 and Hybrid Capture® 2 human papillomavirus tests for detection of CIN2+ and CIN3

INTRODUCTION

Low-cost, accurate high-risk HPV tests are needed for cervical cancer screening in limited-resource settings.

OBJECTIVE

To compare the performance of the low-cost Hybribio-H13 test with the Hybrid Capture® 2 to detect cervical intraepithelial neoplasia grade 2 or 3 (CIN2 and CIN3).

MATERIALS AND METHODS

Archived baseline samples tested by the Hybrid Capture® 2 from women of the ASCUS-COL trial, aged 20 to 69 years, with biopsy-colposcopy directed diagnosis of CIN2+ (n = 143), CIN3+ (n = 51), and < CIN2 (n = 632) were blindly tested by the Hybribio-H13 test.

RESULTS

The relative sensitivity of the Hybribio-H13 test versus the Hybrid Capture® 2 for detecting CIN2+ was 0.89 (90% CI = 0,80-0,98; NIT = 0,66), and for CIN3+ was 0,92 (90% CI = 0,85-0,98; NIT = 0,35). Relative specificity was 1.19 (90% CI = 1.05-1.33; NIT <0.00001). In the analysis restricted to women older than 30 years, the relative sensitivity of the Hybribio-H13 for CIN3+ was marginally below unity (ratio = 0.97; 90% CI = 0.95-0.99), and the specificity remained higher than the Hybrid Capture® 2 test.

CONCLUSION

The Hybribio-H13 test was as specific as the Hybrid Capture® 2 for detecting CIN2+ or CIN3+ but less sensitive. Considering these results and the young age of the population recruited for screening because of ASCUS cytology, we suggest our results warrant the evaluation of the Hybribio-H13 for screening cervical cancer, especially in the evaluated population.

A comparative analysis of cycle threshold (Ct) values from Cobas4800 and AmpFire HPV assay for triage of women with positive hrHPV results

BACKGROUND

To compare the triage performance of HPV viral loads reflected by cycle threshold values (CtV) from two different HPV testing assays: the PCR based Cobas4800 and the isothermal amplification based AmpFire assay.

METHODS

We used the data from a sub-study of The Chinese Multi-Center Screening Trial and analyzed the data of the cases positive in both Cobas4800 and AmpFire assays with recorded CtV. Spearman's correlation was applied to analyze the association between CtV from AmpFire and Cobas4800 assays, as well as the correlation between CtV and the histological lesion grades. The 50th percentile of CtV was used as the cutoff to construct triage algorithms for HPV-positive cases. McNemar's test was used to analyze the differences in sensitivity and specificity for detecting CIN2 + and CIN3 + in different triage algorithms.

RESULTS

Four hundred forty-six HPV positive women who had consistent HPV results from Cobas4800 and AmpFire in terms of the HPV genotype and reported Ct values were included in the analysis. The mean CtV of hrHPV tested by Cobas4800 and AmpFire were linear correlated. Direct association were showed between the severity of cervical lesions and the HPV viral loads reflected by CtV of hrHPV, HPV16, non-16/18 hrHPV and A9 group from both assays. HPV16/18 genotyping combined with low-CtV for non-16/18 hrHPV, especially A9 group, were demonstrated to be satisfactory in the sensitivity and specificity for detecting CIN2 + or CIN3 + .

CONCLUSION

Ct value represented a good triage marker in both PCR-based and isothermal amplification HPV detection.

Verification of the association of the cycle threshold (Ct) values from HPV testing on Cobas4800 with the histologic grades of cervical lesions using data from two population-based cervical cancer screening trials

Objective

To verify the association of high-risk human papillomavirus (hrHPV) viral load reflected by cycle threshold (Ct) values from HPV testing on Cobas4800 assay with the histologic grades of cervical lesions via analysis on the combined data from two cervical cancer screening trials and to explore the referability of Ct values in management of the abnormalities from cervical cancer primary screening.

Methods

We analyzed the data from Chinese Multi-Center Screening Trial (CHMUST) and BUJI Cervical Cancer Screening Study Project (BUJI Study). All data to be analyzed in this paper were related to provider-collected samples. One-way ANOVA was performed to compare the Ct values among different histological groups, and Kendall’s tau-b correlation was applied to examine the association between Ct values and cervical lesion grades. The stepwise incidence of CIN2+ and CIN3+ in every 100 HPV positive individuals were calculated according to the descending of the genotype specific Ct values. The highest Ct values related to CIN3+ incidence rate 4% (or 25%) were used as the cutoffs to distinguish low-Ct value cases from the high-Ct value ones.

Results

A total of 1376 women in CHUMUST and BUJI Study who were aged 30–59 and positive of hrHPV for provider-collected samples on Cobas4800 with complete data in terms of the relevant Ct values (CtV) and the histological diagnosis were included for analysis. Our data showed significant difference among different histological grades of cervical lesions in the CtV of hrHPV, HPV16-plus (positive of HPV16 only or HPV16 plus 18 and/or pooled 12-HPV), and pooled 12-HPV (P < 0.05). No significant difference was found among different grades of lesions in term of correlated CtV of HPV18-plus (positive of HPV18 only or HPV18 plus pooled 12-HPV) (P > 0.05). The CIN2+ or CIN3+ incidence per 100 positives significantly increased corresponding to the descending of the CtV of hrHPV, HPV16-plus and pooled 12-HPV. Compared with high-CtV groups (CtV > 33.2 for hrHPV, CtV > 29.6 for pooled 12-HPV), the relevant risks (RRs) of CIN2+ for hrHPV and pooled 12-HPV positive groups with low-CtV (CtV ≤ 33.2 and ≤ 29.6, respectively) were 3.2 (95%CI 2.18–4.80) and 2.3 (95%CI 1.50–3.45). Similarly, the RRs of CIN3+ for hrHPV and pooled 12-HPV positive groups with low-CtV were 6.5 (95%CI 2.83–14.80) and 2.7 (95%CI 1.15–6.39), respectively. The RRs of CIN2+ for medium- (30.3 < CtV ≤ 37.4) and low- (≤ 30.3) CtV HPV16-plus positives were 5.1 (95%CI 0.68–38.38) and 20.6 (95%CI 2.96–143.92) related to high-CtV (> 37.4) groups, and the CIN3+ incidence in low-CtV value group was nine-fold higher of that in medium-CtV ones [RRs, 9.0 (95%CI 2.89–28.10)]. In comparing with the algorithms of “HPV16-plus/18-plus + cytology ≥ ASCUS for pooled 12-HPV”, triage algorithm “HPV16-plus/18-plus + Ct value ≤ 33.2 for pooled 12-HPV” could achieve a comparable sensitivity of 93.2%.

Conclusion

HPV viral loads reflected by Ct values for hrHPV, HPV16-plus and pooled 12-HPV from Cobas4800 HPV testing were directly associated with the severity of cervical lesions. A lower HPV genotype-specific Ct value prompted a significantly high CIN3+ risk of 4% or higher in women positive of hrHPV, HPV16-plus or pooled 12-HPV, indicating that HPV viral load reflected by Ct values on Cobas4800 may be a promising risk indicator in management of abnormalities from primary cervical cancer screening.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13027-022-00440-4.

Clinical Performance of the Full Genotyping Agena MassARRAY HPV Assay Using SurePath Screening Samples within the VALGENT4 Framework

The clinical performance evaluation of the novel MassARRAY human papillomavirus (MA-HPV) assay was performed using Danish SurePath cervical cancer screening samples under the fourth Validation of HPV Genotyping Tests (VALGENT) framework. The MA-HPV assay is a mass array-based assay that individually detects 14 oncogenic HPV genotypes and five nononcogenic types. The MA-HPV assay was validated using the VALGENT4 panel, which constitutes 997 consecutive samples from a screening population in addition to 297 disease-enriched samples with abnormal cytology findings. The clinical accuracy of the MA-HPV assay for sensitivity and specificity was assessed relative to that of the general primer 5+/6+ PCR enzyme immunoassay (GP-EIA), by a noninferiority test. The type-specific concordance of the MA-HPV assay was assessed as well. The relative sensitivity of the MA-HPV assay for cervical intraepithelial neoplasia ≥2 or ≥3 was 1.02 (95% CI, 0.98-1.05) and 1.01 (95% CI, 0.99-1.04), respectively. The sensitivity of the MA-HPV was noninferior to that of the GP-EIA (P = 0.0001), whereas the specificity of the MA-HPV was inferior (0.89; 95% CI, 0.85-0.91; P > 0.99). The MA-HPV assay is a clinical sensitive assay with a lower clinical specificity compared with the GP-EIA. The assay in its current form seems more suited to play a role where specificity is of lesser importance but where high sensitivity is paramount.

2020 list of human papillomavirus assays suitable for primary cervical cancer screening

BACKGROUND

Only clinically validated HPV assays can be accepted in cervical cancer screening.

OBJECTIVES

To update the list of high-risk HPV assays that fulfil the 2009 international validation criteria (Meijer-2009).

DATA SOURCES

PubMed/Medline, Embase, Scopus, references from selected studies; published in January 2014 to August 2020.

STUDY ELIGIBILITY CRITERIA

HPV test validation studies and primary screening studies, involving testing with an index HPV test and a comparator HPV test with reporting of disease outcome (occurrence of histologically confirmed cervical precancer; CIN2+).

PARTICIPANTS

Women participating in cervical cancer screening.

INTERVENTIONS

Testing with an index and a comparator HPV test of clinician-collected cervical specimens and assessment of disease outcome (

METHODS

Assessment of relative clinical accuracy (including non-inferiority statistics index vs comparator assay) and test reproducibility in individual studies; random effects meta-analyses of the relative clinical sensitivity and specificity of index vs comparator tests.

RESULTS

Seven hrHPV DNA tests consistently fulfilled all validation criteria in multiple studies using predefined test positivity cut-offs (Abbott RealTime High Risk HPV, Anyplex II HPV HR Detection, BD Onclarity HPV Assay, Cobas 4800 HPV Test, HPV-Risk Assay, PapilloCheck HPV-Screening Test and Xpert HPV). Another assay (Alinity m HR HPV Assay) was fully validated in one validation study. The newer Cobas 6800 HPV Test, was validated in two studies against Cobas 4800. Other tests partially fulfilled the international validation criteria (Cervista HPV HR Test, EUROArray HPV, Hybribio's 14 High-Risk HPV, LMNX Genotyping Kit GP HPV, MALDI-TOF, RIATOL qPCR and a number of other in-house developed assays) since the non-inferior accuracy was reached after a posteriori cut-off optimization, inconsistent accuracy findings in different studies, and/or insufficient reproducibility assessment. The APTIMA HPV Assay targeting E6/E7 mRNA of hrHPV was fully validated in one formal validation study and showed slightly lower pooled sensitivity but higher specificity than the standard comparator tests in seven screening studies. However, the current international validation criteria relate to DNA assays. The additional requirement for longitudinal performance data required for non-DNA based HPV assays was not assessed in this review.

CONCLUSIONS

Eleven hrHPV DNA assays fulfil all requirements for use in cervical cancer screening using clinician-collected specimens.

Collapse

Key Words

• Cervical cancer
• Cervical cancer screening
• Diagnostic test accuracy: validation of tests
• Human papillomavirus
• Meta-analysis
• Systematic review

Collapse

MESH Headings

• Alphapapillomavirus
• Early Detection of Cancer
• Female
• Genotyping Techniques
• Humans
• Papillomaviridae/isolation & purification
• Papillomavirus Infections/diagnosis
• Reproducibility of Results
• Sensitivity and Specificity
• Uterine Cervical Neoplasms/diagnosis
• Uterine Cervical Neoplasms/virology

Collapse

Grants

• 001 World Health Organization

Collapse

Affiliation(s)

Marc Arbyn Unit of Cancer Epidemiology, Belgian Cancer Centre, Scientific Institute of Public Health, Brussels, Belgium; Department of Human Structure and Repair, Faculty of Medicine and Health Sciences, University Ghent, Ghent, Belgium.
Marie Simon Haute Autorité de Santé, Saint Denis, France
Eliana Peeters Unit of Cancer Epidemiology, Belgian Cancer Centre, Scientific Institute of Public Health, Brussels, Belgium
Lan Xu Unit of Cancer Epidemiology, Belgian Cancer Centre, Scientific Institute of Public Health, Brussels, Belgium; School of Public Health, Shanghai Jiao Tong University School of Medicine, Shanghai, China
Chris J L M Meijer Department of Pathology, Amsterdam University Medical Centre, location VUMC, Amsterdam, the Netherlands
Johannes Berkhof Department of Clinical Epidemiology and Biostatistics, VU University Medical Centre, Amsterdam, the Netherlands
Kate Cuschieri Scottish HPV Reference Laboratory, Royal Infirmary of Edinburgh, Edinburgh, UK
Jesper Bonde Molecular Pathology Laboratory, Department of Pathology, Copenhagen University Hospital, Hvidovre, Denmark
Anja Ostrbenk Vanlencak Institute of Microbiology and Immunology, Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia
Fang-Hui Zhao Department of Cancer Epidemiology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China
Remila Rezhake Unit of Cancer Epidemiology, Belgian Cancer Centre, Scientific Institute of Public Health, Brussels, Belgium; Department of Cancer Epidemiology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China; The 3rd Affiliated Teaching Hospital of Xinjiang Medical University (Affiliated Cancer Hospital), Urumqi, China
Murat Gultekin Hacettepe University Faculty of Medicine, Department of Obstetrics and Gynecology, Division of Gynaecological Oncology, Ankara, Turkey
Joakim Dillner Department of Laboratory Medicine, Karolinska Institutet, Stockholm, Sweden
Silvia de Sanjosé National Cancer Institute, Rockville, USA
Karen Canfell Cancer Research Division, Cancer Council NSW, Sydney, Australia; School of Public Health, University of Sydney, Sydney, Australia
Peter Hillemanns Departments of Gynaecology and Obstetrics, Hannover Medical School, Germany
Maribel Almonte International Agency for Research on Cancer, Lyon, France
Nicolas Wentzensen Division of Cancer Epidemiology and Genetics, National Cancer, Institute, Bethesda, MD, USA
Mario Poljak Institute of Microbiology and Immunology, Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia

Collapse

Comparison of the clinical and analytical performance of Alinity m HR HPV and cobas 4800 HPV assays in a population-based screening setting

BACKGROUND

The recently launched Abbott Alinity m HR HPV (Alinity) assay separately identifies high-risk human papillomavirus (hrHPV) genotypes HPV16, HPV18, and HPV45, and reports 11 other genotypes as two aggregates.

METHODS

Clinical and analytical performance of Alinity was compared with the cobas 4800 HPV assay on 4,334 women aged 20-64 years attending routine, population-based organized cervical cancer screening during 2009/2010. After 36 months, they were invited to participate in the second screening round (2012-2014) and later followed-up through centralized national cervical cancer screening registry.

RESULTS

In women 30 and older, the clinical sensitivity for cervical intraepithelial neoplasia grade 2+ (CIN2+) was 100.0% (95% CI, 88.2-100.0%) for Alinity and 100.0% (95% CI, 88.2-100.0%) for cobas, and for CIN3+ 100.0% (95% CI, 78.9-100.0%) for both assays. The clinical specificity for ≤ CIN1 in women 30 and older was 92.4% (95% CI, 91.4-93.3%) and 92.9% (95% CI, 91.9-93.7%), respectively. The assays demonstrated excellent overall agreement for hrHPV detection (97.9%) and genotype-specific agreement for HPV16 (99.6%), HPV18 (99.8%), and other hrHPV (98.1%). Overall positive agreement and positive agreements for HPV16, HPV18, and other hrHPV genotypes were 84.3%, 89.1%, 73.2%, and 82.3%. Based on a 5-year CIN3+ risk, slightly more HPV-positive women would require referral to immediate colposcopy after testing with Alinity vs. cobas (4.1% vs. 3.8%; p = 0.470), but significantly fewer Alinity-tested women would need a 6- to 12-month follow-up visit compared with those tested with cobas (5.0% vs. 8.6%; p < 0.0001).

CONCLUSIONS

Alinity and cobas have comparable clinical performance and showed excellent overall and genotype-specific agreement. The Alinity's extended genotyping ability could help predict the 5-year CIN3+ risk and cost-saving management of HPV-screen-positive women.

Clinical and analytical performance of the CLART HPV 4S assay with SurePath screening samples from the Danish cervical cancer screening program using the VALGENT framework

The CLART HPV4S (CLART4S) is a novel full genotyping assay, based on PCR/microarray technology. We assessed the clinical accuracy of the CLART4S assays under the fourth installment of the VALGENT framework. The VALGENT cohort comprised 998 consecutive cervical samples from women participating in the Danish screening programme enriched with 297 samples with abnormal cytology (100 ASCUS, 100 LSIL, 97 HSIL). The CLART4S assay detects 16 HPV genotypes individually: 14 oncogenic and two non-oncogenic HPV types. The GP5+/6+ PCR Enzyme-Immuno-Assay (GP-EIA) and GP5+/6+ PCR with Luminex genotyping (GP-LMNX) were used as comparator tests for clinical accuracy and HPV genotype concordance, respectively. The sensitivity for ≥ CIN2 for the CLART4S assay was 96.7 % (GP-EIA: 92.6 %) with a relative sensitivity of 1.04 (1.00-1.09). The sensitivity for ≥ CIN3 was 98.8 % (GP-EIA: 94.0 %), with relative sensitivity of 1.05 (1.00-1.10). The specificity for <CIN2 was 88.6 % (GP-EIA: 89.2 %) with a relative specificity of 0.99 (0.98-1.01). The CLART 4S was found to be non-inferior to that of GP-EIA for both sensitivity (p < 0.0001) and specificity (p = 0.0452). The overall oncogenic HPV concordance between CLART4S and GP-LMNX was high, however when looking at individual genotype agreement the concordance was more diverse, with the highest agreement found in the Screening population.

The effectiveness of human papillomavirus load, reflected by cycle threshold values, for the triage of HPV-positive self-samples in cervical cancer screening

OBJECTIVE

The performance of Cobas4800 cycle threshold value (Ct-value, reflecting viral load) combined with human papillomavirus (HPV) 16/18 genotyping was explored as a method of risk stratification to triage patients after primary HPV screening of self-collected samples.

METHODS

The Chinese Multi-site Screening Trial database was reviewed, with focus on self-collected samples, using the results of Cobas4800 HPV assay. Quartiles of Ct-values of each genotype were used for grouping and developing screening algorithms. Diagnostic accuracy for paired comparisons between algorithms was obtained using McNemar's test.

RESULTS

A total of 10,498 women were included. The Ct-values of HPV16 and other high-risk HPV were inversely correlated with the severity of cervical lesions (p < 0.001). Risks for cervical intraepithelial neoplasia (CIN2+/CIN3+) were significantly stratified by Ct-values from channels detecting HPV16 and other high-risk HPV types. "HPV with HPV16/18 and reflex Ct <33.7" (algorithm G) achieved a favorable sensitivity to "HPV with atypical squamous cells of undetermined significance or worse (≥ASCUS)" (81.9% vs. 70.1% for CIN2+, p < 0.001), a comparable sensitivity to "HPV with HPV16/18 reflex cytology ≥ASCUS" (81.9% vs. 81.3% for CIN2+, p > 0.05), and resulted in a slightly lower specificity than the latter two algorithms (92.6% vs. 97.4% and 95.4% respectively for CIN2+, p < 0.05). However, algorithm G achieved a comparable sensitivity to HPV testing alone for CIN3+, and reduced the colposcopy referral rate from 13.7% for HPV testing alone to 8.4%.

CONCLUSIONS

HPV viral loads reflected by Ct-values are associated with the severity of cervical lesions. Ct-values with an appropriate cut-off of 33.7, combined with HPV16/18 genotyping, represent a promising triage of HPV-positive women particularly for self-collected samples.