Triage of a positive HPV result
An important limitation of HPV testing is its inability to discriminate between transient and persistent infections. Referral of all HPV-positive cases to colposcopy would overburden the system with associated unnecessary procedures. Hence, a triage strategy is essential to identify clinically important infections that truly require colposcopic evaluation. The FIGURE illustrates the management of a primary HPV test result performed for screening.
HPV genotyping
One strategy for triaging a positive HPV test result is genotyping. HPV 16 and 18 have the highest risk of persistence and progression and merit immediate referral to colposcopy. In the ATHENA trial, CIN 3 was identified in 17.8% (95% CI, 14.8–20.7%) of HPV 16 positive women at baseline, and the CIR increased to 25.2% (95% CI, 21.7–28.7%) after 3 years. The 3-year CIR of CIN 3 was only 5.4% (95% CI, 4.5–6.3%) in women with HPV genotypes other than 16/18. HPV 18–positive women had a 3-year CIR that was intermediate between women with HPV 16 and women with the 12 other genotypes.6 Hence, HPV 16/18–positive cases should be referred for immediate colposcopy, and negative cases should be followed up with cytology and referred for colposcopy if the cytology is ASCUS or worse.31
In July 2020, extended genotyping was approved by the FDA with individual detection of HPV 31, 51, 52 (in addition to 16, 18, and 45) and pooled detection of 33/58, 35/39/68, and 56/59/66. One study found that individual genotypes HPV 16 and 31 carry baseline risk values for CIN 3+ (8.1% and 7.5%, respectively) that are above the 5-year risk threshold for referral to colposcopy following the ASCCP risk-based management guideline.32
Cytology
The higher specificity of cytology makes it an option for triaging HPV-positive cases, and current management guidelines recommend triage to both genotyping and cytology for all patients who are HPV positive, and especially if they are HPV positive but HPV 16/18 negative. Of note, cytology results remain more subjective than those of primary HPV testing, but the combination of initial HPV testing with reflex to cytology is a reasonable and cost effective next step.18 The VASCAR trial found higher colposcopy referrals in the HPV screening and cytology triage group compared with the cytology alone group (19.36 vs 14.54 per 1,000 women).33 The ATHENA trial investigated various triage strategies for HPV-positive cases and its impact on colposcopy referrals.6 Using HPV genotyping and reflex cytology, if HPV 16/18 was positive, colposcopy was advised, but if any of the other 12 HPV types were positive, reflex cytology was done. If reported as ASCUS or worse, colposcopy was performed; conversely, if it was normal, women were rescreened with cotesting after 1 year. Although this strategy led to a reduction in the number of colposcopies, referrals were still higher in the primary HPV arm (3,769 colposcopies per 294 cases) compared with cytology (1,934 colposcopies per 179 cases) or cotesting (3,097 colposcopies per 240 cases) in women aged 25 years.14
p16/Ki-67 Dual-Stain
Diffused p16 immunohistochemical staining, as opposed to focal staining, is associated with active HPV infection but can be present in low-grade as well as high-grade lesions.34 Ki-67 is a marker of cellular proliferation. Coexpression of p16 and Ki-67 indicates a loss of cell cycle regulation and is a hallmark of neoplastic transformation. When positive, these tests are supportive of active HPV infection and of a high-grade lesion. Incorporation of these stains to cytology alone provides additional objective reassurance to cytology, where there is much inter- and intra-observer variability. These stains can be done by laboratories using the stains alone or they can use the FDA-approved p16/Ki-67 Dual-Stain immunohistochemistry (DS), CINtec PLUS Cytology (Roche Diagnostics). However, DS is not yet formally incorporated into triage algorithms by national guidelines.
The IMPACT trial assessed the performance of DS compared with cytology in the triage of HPV-positive results, with or without HPV 16/18 genotyping.35 This was a prospective observational screening study of 35,263 women aged 25 to 65 years across 32 sites in the United States. Of the 4,927 HPV-positive patients with DS results, the sensitivity of DS for CIN 3+ was 91.9% (95% CI, 86.1%–95.4%) and 86.0% (95% CI, 77.5%–91.6%) in HPV 16/18–positive and in the 12 other genotypes, respectively. Using DS alone to triage HPV-positive results showed significantly higher sensitivity and specificity than HPV 16/18 genotyping with cytology triage of 12 “other” genotypes, and substantially higher sensitivity but lower specificity than using cytology alone. Of note, triage with DS alone would have referred significantly fewer women to colposcopy than HPV 16/18 genotyping with cytology triage for the 12 other genotypes (48.6% vs 56.0%; P< .0001).
Similarly, a retrospective analysis of the ATHENA trial cohort of HPV-positive results of 7,727 patients aged 25 years or older also demonstrated increased sensitivity of DS compared with cytology (74.9% vs 51.9%; P<.0001) and similar specificities (74.1% vs 75%; P = .3198).36 The European PALMS study, which included 27,349 women aged 18 years or older across 5 countries who underwent routine screening with HPV testing, cytology, and DS, confirmed these findings.37 The sensitivity of DS was higher than that of cytology (86.7% vs 68.5%; P<.001) for CIN 3+ with comparable specificities (95.2% vs 95.4%; P = .15).
Challenges and opportunities to improve access to primary HPV screening
The historical success of the Pap test in reducing the incidence of cervical cancer relied on individuals having access to the test. This remains true as screening transitions to primary HPV testing. Limitations of HPV-based screening include provider and patient knowledge; access to tests; cost; need for new laboratory infrastructure; need to leverage the electronic health record to record results, calculate a patient’s risk and determine next steps; and the need to re-educate patients and providers about this new model of care. The American Cancer Society and the Centers for Disease Control and Prevention are currently leading initiatives to help adopt primary HPV screening in the United States and to facilitate new care approaches.
Self-collection and independence from subjective cytology would further improve access. Multiple effectiveness studies and patient acceptability studies have shown that primary HPV screening via self-collection is effective, cost effective, and acceptable to women, especially among underscreened populations.38 Sensitivity is comparable to clinician-obtained samples with polymerase chain reaction–based HPV tests. Furthermore, newer molecular tests that detect methylated target host genes or methylated viral genome can be used to triage HPV-positive cases. Several host methylation markers that identify the specific host genes (for example, CADM1, MAL, and miR-124-2) have been shown to be more specific, reproducible, and can be used in self-collected samples as they are based on molecular methylation analysis.39 The ASCCP monitors these new developments and will incorporate promising tests and approaches once validated and FDA approved into the risk-based management guidelines. An erratum was recently published, and the risk-calculator is also available on the ASCCP website free of charge (https://app.asccp.org).40
In conclusion, transition to primary HPV testing from Pap cytology in cervical cancer screening has many challenges but also opportunities. Learning from the experience of countries that have already adopted primary HPV testing is crucial to successful implementation of this new screening paradigm.41 The evidence supporting primary HPV screening with its improved sensitivity is clear, and the existing triage options and innovations will continue to improve triage of patients with clinically important lesions as well as accessibility. With strong advocacy and sound implementation, the WHO goal of cervical cancer elimination and 70% of women being screened with a high-performance test by age 35 and again by age 45 is achievable. ●