A pilot protocol using DNA to screen for human papillomavirus (HPV) at Jewish General Hospital in Montreal, Canada, may provide a “roadmap” for the future, say researchers. Three years after the pilot program launched, the researchers reported that the molecular screening provided “new insight” into risk management for HPV.
The hospital faced roadblocks during the HPV screening process including a shortage of cytotechnologists, a limited number of smears read (by contract, cytotechnologists are limited to 50 smears per 8-hour shift), and overscreening (eg, annual vs every 3 years). As a result, the Viral Testing Alone with Papanicolaou Triage for Screening Cervical Cancer in Routine Practice (VASCAR) project switched from annual Papanicolaou (Pap) cytologic testing to molecular HPV testing.
The primary goal was to evaluate the overall efficacy and safety of HPV testing as a primary screening modality, followed by cytologic triage for women who tested HPV positive. Women were screened with the Hybrid Capture-2 assay. Those with negative results were retested at 3-year intervals. Those with positive results were triaged with conventional cytologic methods. Women with positive Pap results were referred to colposcopy. Those with negative Pap smears were co-tested 1 year later. The results were compared with the method used in the 3 years before VASCAR started (≥ atypical squamous cells of undetermined significance threshold for colposcopy referral).
VASCAR included 26,193 women, of whom 1,646 tested positive for HPV. Of those, 752 (46%) had Pap smears taken and the results entered in the database. The Pap triage result was positive for 210 women (28%). A biopsy with or without endocervical curetting had been completed in 50 women.
By contrast, during the pre-VASCAR years, the Pap positivity rate was 3.2% (1,120 women). Of those, 503 (45%) had colposcopy.
Nearly half (46.8%) the women in the VASCAR era had been screened at least once in the previous 3-year period and had been considered lesion free. Yet the researchers say VASCAR increased the rate of colposcopy referrals by a statistically significant 33% after the adoption of the HPV+/Pap triage approach. This “relatively modest” increase in referrals led to a nearly 3-fold increase in the rate of detection of histologically ascertained high-grade squamous intraepithelial lesions (HSILs) or cancer. The increase in lesion detection was due to the > 2-fold increase in the lesion detection rate per colposcopy.
The researchers say VASCAR is the first North American community-based project to use molecular-based primary cervical cancer screening. Despite diagrams, oral presentations, and written recommendations, there was a fairly steep learning curve, and the protocol was violated in nearly 12% of women who were screened. Fortunately, no cancer was found in HSILs in those violator cases that were HPV positive. The protocol violations were all encountered in the first year and consisted of submitting conventional Pap smears rather than molecular HPV tests. Meta-analyses have shown that the Pap test fails to detect, on average, half of clinically significant precancers and cancers, the researchers note. Interpretation is subjective, and if the reader is tired or distracted, cellular abnormalities may be missed.
However, the researchers found a serious pitfall, they say, in the “considerable delay” in the cytologic follow-up evaluation of the 7% HPV-positive women, mainly because of the need to collect specimens for Pap tests from the women rather than co-collecting samples at initial testing. In other words, the patients all had to be recalled. Only 46% were Pap triaged in the first 3 years of initial testing (the first round). That “unexpected and frustrating” situation will be remedied, they believe, when the public health system switches from subsequent collection of cervical samples to co-collection of liquid-based cytologic samples, thus reducing the process to 1 visit.
Source
Louvanto K, Chevarie-Davis M, Ramanakumar AV, Franco EL, Ferenczy A. Am J Obstet Gynecol. 2014;210(5):474.e1-7.
doi: 10.1016/j.ajog.2013.12.033.