Massimo Confortini

Efficacy of human papillomavirus testing for the detection of invasive cervical cancers and cervical intraepithelial neoplasia: a randomised controlled trial

BACKGROUND Human papillomavirus (HPV) testing is known to be more sensitive, but less specific than cytology for detecting cervical intraepithelial neoplasia (CIN). We assessed the efficacy of cervical-cancer screening policies that are based on HPV testing. METHODS Between March, 2004, and December, 2004, in two separate recruitment phases, women aged 25-60 years were randomly assigned to conventional cytology or to HPV testing in combination with liquid-based cytology (first phase) or alone (second phase). Randomisation was done by computer in two screening centres and by sequential opening of numbered sealed envelopes in the remaining seven centres. During phase one, women who were HPV-positive and aged 35-60 years were referred to colposcopy, whereas women aged 25-34 years were referred to colposcopy only if cytology was also abnormal or HPV testing was persistently positive. During phase two, women in the HPV group were referred for colposcopy if the HPV test was positive. Two rounds of screening occurred in each phase, and all women had cytology testing only at the second round. The primary endpoint was the detection of grade 2 and 3 CIN, and of invasive cervical cancers during the first and second screening rounds. Analysis was done by intention to screen. This trial is registered, number ISRCTN81678807. FINDINGS In total for both phases, 47,001 women were randomly assigned to the cytology group and 47,369 to HPV testing. 33,851 women from the cytology group and 32,998 from the HPV-testing group had a second round of screening. We also retrieved the histological diagnoses from screening done elsewhere. The detection of invasive cervical cancers was similar for the two groups in the first round of screening (nine in the cytology group vs seven in the HPV group, p=0.62); no cases were detected in the HPV group during round two, compared with nine in the cytology group (p=0.004). Overall, in the two rounds of screening, 18 invasive cancers were detected in the cytology group versus seven in the HPV group (p=0.028). Among women aged 35-60 years, at round one the relative detection (HPV vs cytology) was 2.00 (95% CI 1.44-2.77) for CIN2, 2.08 (1.47-2.95) for CIN3, and 2.03 (1.60-2.57) for CIN2 and 3 together. At round two the relative detection was 0.54 (0.23-1.28) for CIN2, 0.48 (0.21-1.11) for CIN3, and 0.51 (0.28-0.93) for CIN2 and 3 together. Among women aged 25-34 years, there was significant heterogeneity between phases in the relative detection of CIN3. At round one the relative detection was 0.93 (0.52-1.64) in phase one and 3.91 (2.02-7.57) in phase two. At round two the relative detection was 1.34 (0.46-3.84) in phase one and 0.20 (0.04-0.93) in phase two. Pooling both phases, the detection ratio of CIN2 for women aged 25-34 years was 4.09 (2.24-7.48) at round one and 0.64 (0.23-1.27) at round two. INTERPRETATION HPV-based screening is more effective than cytology in preventing invasive cervical cancer, by detecting persistent high-grade lesions earlier and providing a longer low-risk period. However, in younger women, HPV screening leads to over-diagnosis of regressive CIN2. FUNDING European Union, Italian Ministry of Health, Regional Health Administrations of Piemonte, Tuscany, Veneto and Emilia-Romagna, and Public Health Agency of Lazio.

Results at Recruitment From a Randomized Controlled Trial Comparing Human Papillomavirus Testing Alone With Conventional Cytology as the Primary Cervical Cancer Screening Test

BACKGROUND In the first recruitment phase of a randomized trial of cervical cancer screening methods (New Technologies for Cervical Cancer Screening [NTCC] study), we compared screening with conventional cytology with screening by human papillomavirus (HPV) testing in combination with liquid-based cytology. HPV-positive women were directly referred to colposcopy if aged 35 or older; if younger, they were retested after 1 year. METHODS In the second recruitment phase of NTCC, we randomly assigned women to conventional cytology (n = 24,661) with referral to colposcopy if cytology indicated atypical squamous cells of undetermined significance or more severe abnormality or to testing for high-risk HPV DNA alone by Hybrid Capture 2 (n = 24,535) with referral to colposcopy if the test was positive at a concentration of HPV DNA 1 pg/mL or greater. For the main endpoint of the study, histologic detection of cervical intraepithelial neoplasia of grade 2 or more (CIN2+), we calculated and compared sensitivity and positive predictive value (PPV) of the two screening methods using HPV DNA cutoffs of 1 pg/mL and 2 pg/mL. All statistical tests were two-sided. RESULTS For women aged 35-60 years, the relative sensitivity of HPV testing for detection of CIN2+ at a cutoff of 1 pg/mL vs conventional cytology was 1.92 (95% CI = 1.28 to 2.87) and the relative PPV was 0.80 (95% CI = 0.55 to 1.18). At a cutoff of 2 pg/mL HPV DNA, the relative sensitivity was 1.81 (95% CI = 1.20 to 2.72) and the relative PPV was 0.99 (95% CI = 0.67 to 1.46). In this age group, there was no evidence of heterogeneity between study phases. Among women aged 25-34 years, the relative sensitivity for detection of CIN2+ of HPV testing at a cutoff of 1 pg/mL vs cytology was 3.50 (95% CI = 2.11 to 5.82), statistically significantly larger (P = .019) than that observed in phase 1 at this age (1.58; 95% CI = 1.03 to 2.44). CONCLUSIONS For women aged 35-60 years, HPV testing with a cutoff of 2 pg/mL achieves a substantial gain in sensitivity over cytology with only a small reduction in PPV. Among women aged 25-34 years, the large relative sensitivity of HPV testing compared with conventional cytology and the difference between relative sensitivity during phases 1 and 2 suggests that there is frequent regression of CIN2+ that are detected by direct referral of younger HPV-positive women to colposcopy. Thus, triage test or repeat testing is needed if HPV is to be used for primary testing in this context.

Accuracy of liquid based versus conventional cytology: overall results of new technologies for cervical cancer screening randomised controlled trial

Objective To compare the accuracy of conventional cytology with liquid based cytology for primary screening of cervical cancer. Design Randomised controlled trial. Setting Nine screening programmes in Italy. Participants Women aged 25-60 attending for a new screening round: 22 466 were assigned to the conventional arm and 22 708 were assigned to the experimental arm. Interventions Conventional cytology compared with liquid based cytology and testing for human papillomavirus. Main outcome measure Relative sensitivity for cervical intraepithelial neoplasia of grade 2 or more at blindly reviewed histology, with atypical cells of undetermined significance or more severe cytology considered a positive result. Results In an intention to screen analysis liquid based cytology showed no significant increase in sensitivity for cervical intraepithelial neoplasia of grade 2 or more (relative sensitivity 1.17, 95% confidence interval 0.87 to 1.56) whereas the positive predictive value was reduced (relative positive predictive value v conventional cytology 0.58, 0.44 to 0.77). Liquid based cytology detected more lesions of grade 1 or more (relative sensitivity 1.68, 1.40 to 2.02), with a larger increase among women aged 25-34 (P for heterogeneity 0.0006), but did not detect more lesions of grade 3 or more (relative sensitivity 0.84, 0.56 to 1.25). Results were similar when only low grade intraepithelial lesions or more severe cytology were considered a positive result. No evidence was found of heterogeneity between centres or of improvement with increasing time from start of the study. The relative frequency of women with at least one unsatisfactory result was lower with liquid based cytology (0.62, 0.56 to 0.69). Conclusion Liquid based cytology showed no statistically significant difference in sensitivity to conventional cytology for detection of cervical intraepithelial neoplasia of grade 2 or more. More positive results were found, however, leading to a lower positive predictive value. A large reduction in unsatisfactory smears was evident. Trial registration Current Controlled Trials ISRCTN81678807.

Use of p16-INK4A overexpression to increase the specificity of human papillomavirus testing: a nested substudy of the NTCC randomised controlled trial

BACKGROUND Human-papillomavirus (HPV) testing is more sensitive, but less specific, than conventional cytology for detecting high-grade cervical intraepithelial neoplasia (CIN). We assessed whether HPV testing with triage by p16-INK4A overexpression can increase specificity while maintaining sensitivity. METHODS HPV-positive women were enrolled between June 10, 2003, and Dec 31, 2004 in a multicentre randomised controlled trial, which compared stand-alone HPV testing by Hybrid Capture 2 (experimental group) with conventional cytology, were referred for colposcopy. In seven of nine centres, cytospin preparations from these women were tested for p16-INK4A overexpression by immunostaining. The sensitivity and specificity for CIN grade 2 or more, determined at blind review of histology, were calculated for these women. We also estimated the relative sensitivity and relative referral to colposcopy that would have been obtained by HPV testing with p16-INK4A triage compared with conventional cytology. This trial is registered as a Standard Randomised Controlled Trial, number ISRCTN81678807. FINDINGS 24 661 women were randomly assigned to the experimental group. 1137 women (74% of those undergoing colposcopy in relevant centres), including 50 with CIN2 and 42 with CIN3 or cancer, had valid p16-INK4A immunostaining. For the endpoint of CIN2+, sensitivity and specificity of p16-INK4A (deemed positive with any number of stained cells-except endocervical, metaplastic, and atrophic cells if morphologically normal) in HPV-positive women of any age were 88% (81 of 92; 95% CI 80-94) and 61% (633 of 1045; 57-64), respectively. In the 35-60-year age group, the relative sensitivity of HPV testing and p16-INK4A triage versus conventional cytology for CIN2+ was 1.53 (95% CI 1.15-2.02) and relative referral was 1.08 (0.96-1.21). In the 25-34-year age group, relative sensitivity was 3.01 (1.82-5.17) and relative referral was 1.15 (0.96-1.37). In the latter age group, when 5% or more stained cells were deemed positive, the corresponding values were 2.06 (1.20-3.68) and 0.58 (0.46-0.73), respectively. INTERPRETATION HPV testing with p16-INK4A triage produces a significant increase in sensitivity compared with conventional cytology, with no substantial increase in referral to colposcopy.

Risk of high-grade cervical intraepithelial neoplasia during follow-up in HPV-positive women according to baseline p16-INK4A results: a prospective analysis of a nested substudy of the NTCC randomised controlled trial

BACKGROUND Immunostaining for p16-INK4A (henceforth p16) is a sensitive and specific method for detection of high-grade cervical intraepithelial neoplasia (CIN) in women infected with human papillomavirus (HPV), but longitudinal data have not been obtained. We investigated the relation between p16 status and risk of CIN during 3 years of follow-up. METHODS Women aged 25-60 years were enrolled between June 10, 2003, and Dec 31, 2004, in a multicentre randomised trial comparing HPV testing with cytology. HPV-positive women were referred for colposcopy and, in seven of nine centres, were tested for p16 overexpression by immunostaining. If no CIN was detected, these women were followed up at yearly intervals until clearance of HPV infection. The primary endpoint was histologically confirmed CIN of grade 2 or worse (CIN of grade 2 [CIN2], CIN of grade 3 [CIN3], or invasive cervical cancer) at recruitment or during follow-up. We calculated the absolute and relative risks by p16 status at recruitment. We also calculated the longitudinal sensitivity of p16 testing. Additionally, we assessed the relative sensitivity of an alternative strategy (referral to colposcopy and follow-up of only HPV-positive, p16-positive women) versus conventional cytology in two age groups. Percentages were weighted by the inverse of the tested fraction. The trial in which this study is nested is registered, number ISRCTN81678807. FINDINGS Of 1042 HPV-positive women who were tested for p16 with no CIN detected during the first round of screening, 944 (91%) had further HPV tests. 793 (84%) of these 944 were followed up until detection of CIN2 or worse, HPV infection clearance, or for at least 3 years. CIN2 or worse was detected during follow-up in more p16-positive women (31 of 365, 8·8% [95% CI 5·8-11·8]) than in p16-negative women (17 of 579, 3·7% [1·9-5·4]; relative risk [RR] 2·61 [95% CI 1·49-4·59]). RR was higher in women aged 35-60 years at recruitment (3·37 [1·39-8·15]) than in those aged 25-34 years (2·15 [1·00-4·61]), but age was not a significant modifier. CIN3 or worse was detected during follow-up in more p16-positive women (16 of 365, 4·4% [2·3-6·6]) than in p16-negative women (six of 579, 1·3% [0·2-2·3]; RR 3·90 [95% CI 1·57-9·68]). Longitudinal sensitivity of p16 testing for detection of CIN3 or worse during follow-up at all ages was 77·8% (95% CI 63·9-91·6). The relative sensitivity of the alternative strategy compared with conventional cytology was 2·08 (1·13-3·56) in women aged 35-60 years and 2·86 (1·28-5·36) in those aged 25-34 years. HPV-positive, p16-negative women aged 35-60 years had a higher cumulative risk of CIN3 or worse during recruitment or follow-up (2·0%, 95% CI 0·3-3·7) than did HPV-negative women (0·01%, 0-0·04) or those who were cytologically normal (0·04%, 0·02-0·09) at recruitment. INTERPRETATION p16 overexpression is a marker for CIN2 or worse or for development of CIN2 or worse within 3 years in HPV-positive women, especially those aged 35-60 years. HPV-positive, p16-positive women need immediate colposcopy and, if the assessment is negative, annual follow-up. Immediate colposcopy can be avoided in HPV-positive, p16-negative women, who can be safely managed with repeat screening after 2-3 year intervals. FUNDING European Union; Italian Ministry of Health; Regional Health Administrations of Piemonte, Tuscany, Veneto and Emilia Romagna; and Public Health Agency of Lazio Region.

The effect of self-sampled HPV testing on participation to cervical cancer screening in Italy: a randomised controlled trial (ISRCTN96071600)

Background:In Italy, cervical cancer screening programmes actively invite women aged 25–64 years. Programmes are hindered by low participation.Methods:A sample of non-responder women aged 35–64 years, belonging to three different programmes (in Rome, Florence and Teramo), was randomly split into four arms: two control groups received standard recall letters to perform either Pap-test (first group) or human papillomavirus (HPV) test (second group) at the clinic. A third arm was sent letters offering a self-sampler for HPV testing, to be requested by phone, whereas a fourth group was directly sent the self-samplers home.Results:Compliance with standard recall was 13.9% (N619). Offering HPV test at the clinic had a nonsignificant effect on compliance (N616, relative risk (RR)=1.08; 95% CI=0.82–1.41). Self-sampler at request had the poorest performance, 8.7% (N622, RR=0.62; 95% CI=0.45–0.86), whereas direct mailing of the self-sampler registered the highest compliance: 19.6% (N616, RR=1.41; 95% CI=1.10–1.82). This effect on compliance was observed only in urban areas, Florence and Rome (N438, RR=1.69; 95% CI=1.24–2.30), but not in Abruzzo (N178, RR=0.95; 95% CI=0.61–1.50), a prevalently rural area.Conclusions:Mailing self-samplers to non-responders may increase compliance as compared with delivering standard recall letters. Nevertheless, effectiveness is context specific and the strategy costs should be carefully considered.

Immunochemical faecal occult blood test: number of samples and positivity cutoff. What is the best strategy for colorectal cancer screening?

Immunochemical faecal occult blood tests have shown a greater sensitivity than guaiac test in colorectal cancer screening, but optimal number of samples and cutoff have still to be defined. The aim of this multicentric study was to evaluate the performance of immunochemical-based screening strategies according to different positivity thresholds (80, 100, 120 ng ml−1) and single vs double sampling (one, at least one, or both positive samples) using 1-day sample with cutoff at 100 ng ml−1 as the reference strategy. A total of 20 596 subjects aged 50–69 years were enrolled from Italian population-based screening programmes. Positivity rate was 4.5% for reference strategy and 8.0 and 2.0% for the most sensitive and the most specific strategy, respectively. Cancer detection rate of reference strategy was 2.8‰, and ranged between 2.1 and 3.4‰ in other strategies; reference strategy detected 15.6‰ advanced adenomas (range=10.0–22.5‰). The number needed to scope to find a cancer or an advanced adenoma was lower than 2 (1.5–1.7) for the most specific strategies, whereas it was 2.4–2.7, according to different thresholds, for the most sensitive ones. Different strategies seem to have a greater impact on adenomas rather than on cancer detection rate. The study provides information when deciding screening protocols and to adapt them to local resources.

Influence of seasonal variations in ambient temperatures on performance of immunochemical faecal occult blood test for colorectal cancer screening: observational study from the Florence district

Background Faecal occult blood testing (FOBT) in population screening has proved to be effective in reducing mortality from colorectal cancer. In Italy a latex agglutination FOBT has been adopted for a single-sample screening programme. The aim of this study was to examine the performance of FOBTs in the Florence screening programme over several seasons to evaluate the impact of variations in ambient temperature on the performance of the screening test. Methods Measured haemoglobin (Hb) concentrations were aggregated into seasons with their average ambient temperature (AAT). Using logistic regression, the AAT over the period preceding the test measurement was analysed. This period included the time between faecal sampling and return of the test sample (mean 7 days) and the time in the laboratory refrigerator before analysis (mean 4 days). The AAT from days 5–11 before analysis of the test sample was considered a determinant of test positivity. The Kruskal–Wallis rank test was used to evaluate the significance of seasonal and/or AAT-related differences in Hb concentration. A logistic regression model adjusted for sex, age, season and screening episode (first or repeated examination) was constructed. Results 199 654 FOBT results were examined. Mean FOBT seasonal Hb concentrations (ng/ml) were: spring 27.6 (95% CI 26.2 to 29.1); summer 25.2 (95% CI 23.1 to 27.3); autumn 29.2 (95% CI 27.7 to 30.6); winter 29.5 (95% CI 27.9 to 31.1). Logistic regression showed that there was a 17% lower probability of the FOBT being positive in summer than in winter. The results of the logistic regression showed that an increase in temperature of 1°C produced a 0.7% reduction in probability of a FOBT being positive. In the summer the probability of detecting a cancer or an advanced adenoma was about 13% lower than in the winter. Conclusions This study showed that there is a significant fall in Hb concentration at higher ambient temperatures. These results will have important implications for the organisation of immunochemical FOBT-based screening programmes, particularly in countries with high ambient temperatures.

Reproducibility of HPV DNA Testing by Hybrid Capture 2 in a Screening Setting Intralaboratory and Interlaboratory Quality Control in Seven Laboratories Participating in the Same Clinical Trial

Within a large Italian randomized trial on new technologies for cervical cancer screening involving 7 laboratories with different levels of experience, an intralaboratory and interlaboratory quality control program for human papillomavirus (HPV) DNA testing by Hybrid Capture 2 (HC2; Digene, Gaithersburg, MD) was implemented. To monitor the hybridization and detection steps, target samples containing purified, concentration-defined, HPV DNA were introduced in each test run. Only 3 of 1,024 showed a mistake in a positive vs negative classification with a 1 relative light unit (RLU)/positive control specimen (PC) ratio cutoff. To monitor the preanalytic steps (particularly denaturation), blinded specimens (33 collected in PreservCyt (Cytyc, Boxborough, MA) and 36 in Specimen Transport Medium (STM, Digene) were centrally prepared, divided into aliquots, and sent to each laboratory. The multiple-rater scores for negative (<1 RLU/PC), low-positive (1 to <11 RLU/PC), and high-positive (> or =11 RLU/PC) samples, respectively, were 0.91, 0.60, and 0.69 with PreservCyt and 0.93, 0.87, and 0.90 with STM. Our data showed high reliability and reproducibility with HC2, with values higher for STM than ThinPrep (Cytyc) samples.

Reproducibility of HPV DNA Testing by Hybrid Capture 2 in a Screening Setting

Within a large Italian randomized trial on new technologies for cervical canc+er screening involving 7 laboratories with different levels of experience, an intralaboratory and interlaboratory quality control program for human papillomavirus (HPV) DNA testing by Hybrid Capture 2 (HC2; Digene, Gaithersburg, MD) was implemented. To monitor the hybridization and detection steps, target samples containing purified, concentration-defined, HPV DNA were introduced in each test run. Only 3 of 1,024 showed a mistake in a positive vs negative classification with a 1 relative light unit (RLU)/positive control specimen (PC) ratio cutoff. To monitor the preanalytic steps (particularly denaturation), blinded specimens (33 collected in PreservCyt [Cytyc, Boxborough, MA] and 36 in Specimen Transport Medium [STM, Digene]) were centrally prepared, divided into aliquots, and sent to each laboratory. The multiple-rater κ scores for negative (<1 RLU/PC), low-positive (1 to <11 RLU/PC), and high-positive (≥11 RLU/PC) samples, respectively, were 0.91, 0.60, and 0.69 with PreservCyt and 0.93, 0.87, and 0.90 with STM. Our data showed high reliability and reproducibility with HC2, with κ values higher for STM than ThinPrep (Cytyc) samples.