Search tips
Search criteria

Results 1-25 (60)

Clipboard (0)

Select a Filter Below

more »
Year of Publication
1.  Reassurance Against Future Risk of Precancer and Cancer Conferred by a Negative Human Papillomavirus Test 
Primary human papillomavirus (HPV) testing (without concurrent Pap tests) every 3 years is under consideration in the United States as an alternative to the two recommended cervical cancer screening strategies: primary Pap testing every 3 years, or concurrent Pap and HPV testing (“cotesting”) every 5 years. Using logistic regression and Weibull survival models, we estimated and compared risks of cancer and cervical intraepithelial neoplasia grade 3 or worse (CIN3+) for the three strategies among 1011092 women aged 30 to 64 years testing HPV-negative and/or Pap-negative in routine screening at Kaiser Permanente Northern California since 2003. All statistical tests were two sided. Three-year risks following an HPV-negative result were lower than 3-year risks following a Pap-negative result (CIN3+ = 0.069% vs 0.19%, P < .0001; Cancer = 0.011% vs 0.020%, P < .0001) and 5-year risks following an HPV-negative/Pap-negative cotest (CIN3+ = 0.069% vs 0.11%, P < .0001; Cancer = 0.011% vs 0.014%, P = .21). These findings suggest that primary HPV testing merits consideration as another alternative for cervical screening.
PMCID: PMC4111283  PMID: 25038467
2.  Circulating Inflammation Markers and Prospective Risk for Lung Cancer 
Despite growing recognition of an etiologic role for inflammation in lung carcinogenesis, few prospective epidemiologic studies have comprehensively investigated the association of circulating inflammation markers with lung cancer.
We conducted a nested case–control study (n = 526 lung cancer patients and n = 592 control subjects) within the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial. Control subjects were matched to lung cancer case patients on age, sex, follow-up time (median = 2.9 years), randomization year, and smoking (pack-years and time since quitting). Serum levels of 77 inflammation markers were measured using a Luminex bead-based assay. Conditional logistic regression and weighted Cox models were used to estimate odds ratios (ORs) and cumulative risks, respectively.
Of 68 evaluable markers, 11 were statistically significantly associated with lung cancer risk (P trend across marker categories < .05), including acute-phase proteins (C-reactive protein [CRP], serum amyloid A [SAA]), proinflammatory cytokines (soluble tumor necrosis factor receptor 2 [sTNFRII]), anti-inflammatory cytokines (interleukin 1 receptor antagonist [IL-1RA]), lymphoid differentiation cytokines (interleukin 7 [IL-7]), growth factors (transforming growth factor alpha [TGF-A]), and chemokines (epithelial neutrophil-activating peptide 78 [ENA 78/CXCL5], monokine induced by gamma interferon [MIG/CXCL9], B cell–attracting chemokine 1 [BCA-1/CXCL13], thymus activation regulated chemokine [TARC/CCL17], macrophage-derived chemokine [MDC/CCL22]). Elevated marker levels were associated with increased lung cancer risk, with odds ratios comparing the highest vs the lowest group ranging from 1.47 (IL-7) to 2.27 (CRP). For IL-1RA, elevated levels were associated with decreased lung cancer risk (OR = 0.71; 95% confidence interval = 0.51 to 1.00). Associations did not differ by smoking, lung cancer histology, or latency. A cross-validated inflammation score using four independent markers (CRP, BCA-1/CXCL13, MDC/CCL22, and IL-1RA) provided good separation in 10-year lung cancer cumulative risks among former smokers (quartile [Q] 1 = 1.1% vs Q4 = 3.1%) and current smokers (Q1 = 2.3% vs Q4 = 7.9%) even after adjustment for smoking.
Some circulating inflammation marker levels are associated with prospective lung cancer risk.
PMCID: PMC3888091  PMID: 24249745
3.  Prevalence of and Risk Factors for Oral Human Papillomavirus Among Young Women in Costa Rica 
The Journal of Infectious Diseases  2013;208(10):1643-1652.
Background. Little is known about the epidemiology of oral human papillomavirus (HPV) in Latin America.
Methods. Women (N = 5838) aged 22–29 in the control and vaccine arms of an HPV-16/18 vaccine trial in Costa Rica had oral, cervical, and anal specimens collected. Samples were tested for alpha mucosal HPV types (SPF10/LiPA25 version 1); a subset of oral samples (n = 500) was tested for cutaneous HPV types in the genera alpha, beta, gamma, mu, and nu.
Results. In the control arm (n = 2926), 1.9% of women had an oral alpha mucosal HPV detected, 1.3% had carcinogenic HPV, and 0.4% had HPV-16; similar patterns for non-16/18 HPV types were observed in the vaccine arm. Independent risk factors for any oral alpha mucosal HPV among women in the control arm included marital status (adjusted odds ratio [AOR], 3.2; 95% confidence interval [CI], 1.8–5.7 for single compared to married/living as married), number of sexual partners (AOR, 2.4; 95% CI, 1.0–6.1 for ≥4 partners compared to 0–1 partners), chronic sinusitis (AOR, 3.1; 95% CI, 1.5–6.7), and cervical HPV infection (AOR, 2.6; 95% CI, 1.4–4.6). Detection of beta HPV was common (18.6%) and not associated with sexual activity.
Conclusions. Unlike cutaneous HPV types, alpha mucosal HPV types were uncommon in the oral region and were predominately associated with sexual behavior.
Clinical Trials Registration. NCT00128661.
PMCID: PMC3805238  PMID: 24014882
human papillomavirus vaccine; HPV; oropharynx cancer; Costa Rica; Guanacaste; oral HPV DNA
4.  A migrant study of pubertal timing and tempo in British-Bangladeshi girls at varying risk for breast cancer 
Earlier menarche is related to subsequent breast cancer risk, yet international differences in the age and tempo of other pubertal milestones and their relationships with body mass index (BMI) are not firmly established in populations at differing risk for breast cancer. We compared age and tempo of adrenarche, thelarche, pubarche, and menarche in a migrant study of Bangladeshi girls to the United Kingdom (UK) and assessed whether differences by migration were explained by differences in BMI.
Included were groups of Bangladeshi (n =168), British-Bangladeshi (n =174) and white British (n =54) girls, aged 5 to 16 years. Interviewer-administered questionnaires obtained pubertal staging; height and weight were measured. Salivary dehydroepiandrosterone-sulfate concentrations >400 pg/ml defined adrenarche. Median ages of pubertal milestones and hazard ratios (HR) with 95% confidence intervals (CI) were estimated from Weibull survival models.
In all three groups, adrenarche occurred earliest, followed by thelarche, pubarche, and finally menarche. Neither median age at adrenarche (Bangladeshi = 7.2, British-Bangladeshi = 7.4, white British = 7.1; P-trend = 0.70) nor at menarche (Bangladeshi = 12.5, British-Bangladeshi = 12.1, white British = 12.6; P-trend = 0.70) differed across groups. In contrast, median age at thelarche (Bangladeshi = 10.7, British-Bangladeshi = 9.6, white British = 8.7; P-trend <0.01) occurred earlier among girls living in the UK. Compared with Bangladeshi girls, HRs (95% CI) for earlier thelarche were 1.6 (1.1 to 2.4) for British-Bangladeshi girls and 2.6 (1.5 to 4.4) for white British girls (P-trend <0.01), but were attenuated after adjustment for BMI (British-Bangladeshi = 1.1 (0.7 to 1.8), white British = 1.7(1.0 to 3.1); P-trend =0.20).
Thelarche occurred earlier, but puberty progressed slower with increasing exposure to the UK environment; differences were partially explained by greater BMI. The growth environment might account for much of the ethnic differences in pubertal development observed across and within countries.
PMCID: PMC4303203  PMID: 25398700
Long-term breast cancer trends in incidence in the United States (US) show rising ER positive rates and falling ER negative rates. We hypothesized that these divergent trends reflect etiologic heterogeneity and that comparable trends should be observed in other countries with similar risk factor profiles. We, therefore, analyzed invasive female breast cancers in Denmark, a country with similar risk factors as the US. We summarized the overall trend in age-standardized rates with the estimated annual percentage change (EAPC) statistic (1993–2010) and used age-period-cohort models to estimate age-specific EAPCs, cohort rate ratios (CRRs), and projections for future time periods (2011–2018). In Denmark, the overall rate of ER positive cancers rose between 1993 and 2010 by 3·0%/year (95% CI: 2·8 to 3·3%/yr) while the overall rate of ER negative cancers fell by 2·1%/year (95% CI: −2·5 to −1·6%/yr). The ER positive rate increased fastest among postmenopausal women and the ER negative rate decreased fastest among premenopausal women, reflecting that cohorts born after 1944 were at relatively higher risk of ER positive tumors and lower risk of ER negative tumors. If current trends continue, ER positive cancers will increase at least 13% by 2018 in Denmark, ER negative cancers will fall 15% by 2018, and breast cancer overall will increase at least 7% by 2018. Divergent ER-specific trends are consistent with distinct etiologic pathways. If trends in known risk factors are responsible, the Danish and US experience may foreshadow a common pattern worldwide.
PMCID: PMC3749265  PMID: 23616071
Breast cancer; Estrogen receptor; Epidemiology; Age-period-cohort models
6.  Childhood Environment Influences Adrenarcheal Timing among First-Generation Bangladeshi Migrant Girls to the UK 
PLoS ONE  2014;9(10):e109200.
Adrenarche is a key early life event that marks middle childhood at approximately 7 years of age. Earlier work with British-Bangladeshi migrant women suggested that environmental conditions experienced before adrenarche influence adult reproductive function. We therefore investigated whether Bangladeshi children who migrate to the United Kingdom (UK) reach adrenarche earlier than non-migrants in Bangladesh or the United Kingdom.
Methods and Findings
Healthy girls, aged 5–16 years, were recruited from schools in Sylhet, Bangladesh and London, England comprising four groups: Sylhetis (n = 165), first-generation migrants to the United Kingdom (n = 42), second-generation girls (n = 162), and British girls of European origin (n = 50). Anthropometric measurements were collected together with questionnaire data for migration and socioeconomic characteristics. Saliva samples were assayed for dehydroepiandrosterone (DHEAS) using enzyme-linked immunosorbent assays. Multiple linear regressions tested for group differences in anthropometric and socioeconomic variables and DHEAS levels. Median ages at adrenarche (DHEAS>400 pg/ml) were estimated using Weibull regression models for parametric survival analysis. Hazard ratios for reaching adrenarche earlier and 95% confidence intervals (CI), both unadjusted and adjusted for anthropometric variables, were estimated from the survival analyses. First-generation migrants had a median age at adrenarche (5.3 years) that was significantly earlier than Sylheti (7.2), second-generation (7.4), and European (7.1) girls. In univariate analyses, first-generation girls reached adrenarche significantly earlier than Sylhetis [HR (CI): 2.8 (1.4–5.5]. In multivariate models, first generation girls still reached adrenarche earlier than Sylhetis after adjusting for height [HR(CI): 1.9 (0.9–4.1)] and weight [HR(CI):1.7 (0.8–3.8)], but these results were attenuated.
We suggest that rapid catch-up growth experienced by first generation girls during early childhood may explain their advanced adrenarche. The environmental conditions leading to an earlier adrenarche, as well as the health implications of this early transition, merit further exploration.
PMCID: PMC4195659  PMID: 25309977
7.  Cross-protective vaccine efficacy of the bivalent HPV vaccine against HPV31 is associated with humoral immune responses 
Human Vaccines & Immunotherapeutics  2013;9(7):1399-1406.
Background: We investigated the role of antibody responses as potential mechanism for the cross-protective vaccine-efficacies (VE) observed from randomized clinical trials of the HPV16/18 bivalent vaccine.
Results: HPV31 cases had lower HPV16 antibody levels than controls (OR4th quartile compared with 1st quartile = 0.63; 95%CI: 0.36–1.08; p-trend = 0.03). HPV31 cases were also less likely to have detectable HPV31 neutralization, and HPV16 avidity than controls. No statistically significant differences by HPV18 antibody or HPV45 neutralization were observed among HPV45 cases and controls. Protection against HPV58 was not associated with any of the markers, confirming the specificity of our findings.
Methods: Samples are from three-dose HPV vaccine recipients from the Costa Rica HPV16/18 vaccine trial. Women with a new HPV31, HPV45, or HPV58 infections over four years of follow-up were compared with randomly selected control women—with no new infection with HPV31/45/58—with respect to HPV16 and HPV18 antibody, HPV31, HPV45, and HPV58 neutralization, and HPV16 avidity.
Conclusions: High HPV16 levels and avidity, and the ability to neutralize HPV31 were associated with protection against newly detected HPV31 infections, suggesting that the partial VE demonstrated for HPV31 is likely to be mediated at least in part through antibodies induced by HPV16/18 vaccination.
PMCID: PMC3974884  PMID: 23571174
HPV vaccine; humoral; immune response; cross-protection; mechanisms for protection
8.  Comparison of Algorithm-based Estimates of Occupational Diesel Exhaust Exposure to Those of Multiple Independent Raters in a Population-based Case–Control Study 
Annals of Occupational Hygiene  2012;57(4):470-481.
Algorithm-based exposure assessments based on patterns in questionnaire responses and professional judgment can readily apply transparent exposure decision rules to thousands of jobs quickly. However, we need to better understand how algorithms compare to a one-by-one job review by an exposure assessor. We compared algorithm-based estimates of diesel exhaust exposure to those of three independent raters within the New England Bladder Cancer Study, a population-based case–control study, and identified conditions under which disparities occurred in the assessments of the algorithm and the raters.
Occupational diesel exhaust exposure was assessed previously using an algorithm and a single rater for all 14 983 jobs reported by 2631 study participants during personal interviews conducted from 2001 to 2004. Two additional raters independently assessed a random subset of 324 jobs that were selected based on strata defined by the cross-tabulations of the algorithm and the first rater’s probability assessments for each job, oversampling their disagreements. The algorithm and each rater assessed the probability, intensity and frequency of occupational diesel exhaust exposure, as well as a confidence rating for each metric. Agreement among the raters, their aggregate rating (average of the three raters’ ratings) and the algorithm were evaluated using proportion of agreement, kappa and weighted kappa (κw). Agreement analyses on the subset used inverse probability weighting to extrapolate the subset to estimate agreement for all jobs. Classification and Regression Tree (CART) models were used to identify patterns in questionnaire responses that predicted disparities in exposure status (i.e., unexposed versus exposed) between the first rater and the algorithm-based estimates.
For the probability, intensity and frequency exposure metrics, moderate to moderately high agreement was observed among raters (κw = 0.50–0.76) and between the algorithm and the individual raters (κw = 0.58–0.81). For these metrics, the algorithm estimates had consistently higher agreement with the aggregate rating (κw = 0.82) than with the individual raters. For all metrics, the agreement between the algorithm and the aggregate ratings was highest for the unexposed category (90–93%) and was poor to moderate for the exposed categories (9–64%). Lower agreement was observed for jobs with a start year <1965 versus ≥1965. For the confidence metrics, the agreement was poor to moderate among raters (κw = 0.17–0.45) and between the algorithm and the individual raters (κw = 0.24–0.61). CART models identified patterns in the questionnaire responses that predicted a fair-to-moderate (33–89%) proportion of the disagreements between the raters’ and the algorithm estimates.
The agreement between any two raters was similar to the agreement between an algorithm-based approach and individual raters, providing additional support for using the more efficient and transparent algorithm-based approach. CART models identified some patterns in disagreements between the first rater and the algorithm. Given the absence of a gold standard for estimating exposure, these patterns can be reviewed by a team of exposure assessors to determine whether the algorithm should be revised for future studies.
PMCID: PMC3629988  PMID: 23184256
case–control; diesel exhaust; expert judgement; exposure assessment
9.  A general binomial regression model to estimate standardized risk differences from binary response data 
Statistics in medicine  2012;32(5):808-821.
Estimates of absolute risks and risk differences are necessary for evaluating the clinical and population impact of biomedical research findings. We have developed a linear-expit regression model (LEXPIT) to incorporate linear and nonlinear risk effects to estimate absolute risk from studies of a binary outcome. The LEXPIT is a generalization of both the binomial linear and logistic regression models. The coefficients of the LEXPIT linear terms estimate adjusted risk differences, while the exponentiated nonlinear terms estimate residual odds ratios. The LEXPIT could be particularly useful for epidemiological studies of risk association, where adjustment for multiple confounding variables is common. We present a constrained maximum likelihood estimation algorithm that ensures the feasibility of risk estimates of the LEXPIT model and describe procedures for defining the feasible region of the parameter space, judging convergence, and evaluating boundary cases. Simulations demonstrate that the methodology is computationally robust and yields feasible, consistent estimators. We applied the LEXPIT model to estimate the absolute five-year risk of cervical precancer or cancer associated with different Pap and human papillomavirus test results in 167,171 women undergoing screening at Kaiser Permanente Northern California. The LEXPIT model found an increased risk due to abnormal Pap test in HPV-negative that was not detected with logistic regression. Our R package blm provides free and easy-to-use software for fitting the LEXPIT model.
PMCID: PMC3982929  PMID: 22865328
Absolute risk; Binomial regression; Constrained maximization; Logistic regression; Risk difference
10.  Five-year risk of recurrence following treatment of CIN2, CIN3, or AIS: performance of HPV and Pap cotesting in post-treatment management 
Journal of lower genital tract disease  2013;17(5 0 1):S78-S84.
After excisional treatment, CIN2+ can recur. It is not clear how many negative post-treatment Pap or cotest results are needed to assure adequate safety against CIN2+, prior to return to extended retesting intervals.
We observed 5-year risks of CIN2+, and outcomes for 3 follow-up management strategies after treatment (Pap alone, HPV alone, and cotesting) for 3273 women aged 25 and older who were treated for CIN2, CIN3, or AIS between 2003–2010 at Kaiser Permanente Northern California.
Five-year risks of recurrent CIN2+ after treatment varied both by antecedent screening test result and the histology of the treated lesion. The risk ranged from 5% for CIN2 preceded by HPV-positive/ASC-US or LSIL to 16% for CIN3/AIS preceded by AGC/ASC-H/HSIL+ (p<0.0001). However, after post-treatment negative tests, risks were lowered and similar regardless of antecedent screening test and histology of treated disease. The 5-year recurrent CIN2+ risk associated with a negative post-treatment cotest (2.4%) was lower than that of an HPV-negative test alone (3.7%, p=0.3) or Pap-negative result alone (4.2%, p=0.1). Two negative post-treatment tests of each kind conferred slightly lower 5-year CIN2+ risk than one (Pap-negative: 2.7% vs. 4.2%, p=0.2; HPV-negative: 2.7% vs. 3.7%, p=0.7; HPV-negative/Pap-negative: 1.5% vs. 2.4%, p=0.8). The 5-year CIN2+ risk associated with 2 negative cotests of 1.5% (95%CI 0.3% to 7.2%) approached the 0.68% risk associated with a negative Pap test during routine screening.
Women with antecedent AGC/ASC-H/HSIL+ Pap results or those treated with CIN3/AIS had a substantial risk of developing CIN2+ after treatment. Based on the principle of “benchmarking to implicit risk thresholds”, after negative test results following treatment, no subgroup of women achieved risk sufficiently low risk to return to 5-year routine screening. However, negative cotests after treatment provided more reassurance against recurrent CIN2+ than either negative Pap tests or HPV tests alone.
PMCID: PMC3616418  PMID: 23519309
Human Papillomavirus (HPV); cancer prevention; Pap; cervical intraepithelialneoplasia (CIN); Hybrid Capture 2 (HC2); post-treatment; test of cure
11.  Benchmarking CIN3+ risk as the basis for incorporating HPV and Pap cotesting into cervical screening and management guidelines 
Journal of lower genital tract disease  2013;17(5 0 1):S28-S35.
In 2012, the United States Preventive Services Task Force (USPSTF) and a consensus of 25 organizations endorsed concurrent cytology and HPV testing (“cotesting”) for cervical cancer screening. Past screening and management guidelines were implicitly based on risks defined by Pap-alone, without consideration of HPV test results. To promote management that is consistent with accepted practice, new guidelines incorporating cotesting should aim to achieve equal management of women at equal risk of cervical intraepithelial neoplasia grade 3 and cancer (CIN3+).
We estimated cumulative 5-year risks of CIN3+ for 965,360 women aged 30–64 undergoing cotesting at Kaiser Permanente Northern California 2003–2010. We calculated the implicit risk thresholds for Pap-alone and applied them for new management guidance on HPV and Pap cotesting, citing 2 examples: HPV-positive/ASC-US and HPV-negative/Pap-negative. We call this guidance process “benchmarking”.
LSIL, for which immediate colposcopy is prescribed, carries 5-year CIN3+ risk of 5.2%, suggesting that test results with similar risks should be managed with colposcopy. Similarly, ASC-US (2.6% risk) is managed with 6–12 month follow-up and Pap-negative (0.26% risk) is managed with 3-year follow-up. The 5-year CIN3+ risk for women with HPV-positive/ASC-US was 6.8% (95%CI 6.2% to 7.6%). This is greater than the 5.2% risk implicitly leading to referral to colposcopy, consistent with current management recommendations that HPV-positive/ASC-US should be referred for immediate colposcopy. The 5-year CIN3+ risk for women with HPV-negative/Pap-negative was 0.08% (95%CI 0.07% to 0.09%), far below the 0.26% implicitly required for a 3-year return and justifying a longer (e.g., 5-year) return.
Using the principle of “equal management of equal risks,” benchmarking to implicit risk thresholds based on Pap-alone can be used to achieve safe and consistent incorporation of cotesting.
PMCID: PMC3616419  PMID: 23519302
Human Papillomavirus (HPV); cancer prevention; Pap; cervical intraepithelial neoplasia (CIN); Hybrid Capture 2 (HC2)
12.  Five-year risk of CIN3+ and cervical cancer for women who test Pap-negative, but are HPV-positive 
Journal of lower genital tract disease  2013;17(5 0 1):S56-S63.
Current US guidelines for cotesting recommend that the large numbers of women who test Pap-negative, but HPV-positive, return in 1 year, and those who remain HPV-positive or have LSIL (or worse) Pap be referred for colposcopy. However, the performance of these guidelines in routine clinical practice has not been evaluated.
We estimated cumulative 5-year risks of CIN3+ for 32,374 women aged 30–64 with HPV-positive/Pap-negative cotest results at Kaiser Permanente Northern California during 2003–2010.
The 5-year CIN3+ risk following an HPV-positive/Pap-negative cotest result, which was found in 3.6% of women, was 4.5% (95%CI 4.2%–4.8%). The 5-year cancer risk was 0.34% (95%CI 0.26% to 0.45%) and half of the cases were adenocarcinoma. Overall, 47% of the women remained HPV-positive upon return (median 415 days after baseline), a percentage that varied little over ages 30–64. At the return following a baseline HPV-positive/Pap-negative result, almost every repeat cotest result predicted greater subsequent 5-year CIN3+ risk than the same cotest result had at baseline (HPV-positive/LSIL: 9.2% vs. 6.1%, p=0.01; HPV-positive/ASC-US: 7.9% vs. 6.8%, p=0.2; HPV-positive/Pap-negative: 7.4% vs. 4.5%, p<0.0001; HPV-negative/LSIL: 1.7% vs. 2.0%, p=0.8; HPV-negative/ASC-US: 2.9% vs. 0.43%, p=0.0005; HPV-negative/Pap-negative: 0.93% vs. 0.08%, p<0.0001).
Using the principle of “equal management of equal risks”, HPV-positive/Pap-negative women had subsequent CIN3+ risk consistent with risk thresholds for a 1-year return. However, upon returning in approximately 1 year, about one-half of women will be referred for colposcopy due to continued HPV positivity or Pap abnormality. Clinicians should keep in mind that cotest results at the return following a baseline HPV-positive/Pap-negative finding are riskier than the same baseline cotest results in the general population, supporting intensified clinical management of return cotests.
PMCID: PMC3616446  PMID: 23519306
Human Papillomavirus (HPV); cancer prevention; Pap; cervical intraepithelial neoplasia (CIN); Hybrid Capture 2 (HC2); prospective cohort
13.  Five-year risk of CIN3+ and cervical cancer for women with HPV-positive and HPV-negative high-grade Pap results 
Journal of lower genital tract disease  2013;17(5 0 1):S50-S55.
High-grade Pap results (e.g., AGC, ASC-H, and HSIL) predict high cancer risks, resulting in referral to colposcopy without HPV triage. However, new guidelines recommending cotesting for women 30 and older imply that clinicians will often receive the HPV test result concurrently for high-grade Pap results. We examined whether HPV testing provides useful risk stratification in this context.
From a cohort of 965,360 women age 30–64 undergoing cotesting at Kaiser Permanente Northern California, we estimated 5-year risks of cervical cancer and CIN3+ for AGC (2,074 women), ASC-H (1,647 women), and HSIL (2,019 women) according to HPV test results.
HPV positivity of AGC Pap results was 25% and decreased with age (30–34: 44% vs. 60–64: 17%, p<0.0001), while HPV-positivity of ASC-H and HSIL were much higher (71% and 94% respectively) and decreased less with age. Even for these high-grade Pap results, 5-year CIN3+ risks differed substantially between HPV-positive and HPV-negative women (AGC: 33% vs. 0.93%, p<0.0001; ASC-H: 25% vs. 3.5%, p<0.0001; HSIL: 49% vs. 30%, p=0.006). However, except for AGC, cervical cancer risks differed less between HPV-positive and HPV-negative women (AGC: 9.0% vs. 0.37%, p<0.0001; ASC-H: 2.5% vs. 2.1%, p=0.8; HSIL: 6.6% vs. 6.8%, p=0.7).
The risks of CIN3+ for HPV-negative high-grade Pap results were lower than for HPV-positive high-grade Pap results, especially for AGC. However, by the principle of “equal management of equal risks”, all HPV-negative high-grade Pap results had cancer risks high enough to warrant colposcopy, confirming that there is no current role for HPV triage of high-grade Pap results.
PMCID: PMC3616447  PMID: 23519305
Human Papillomavirus (HPV); cancer prevention; baseline; cervical intraepithelial neoplasia (CIN); Hybrid Capture 2 (HC2); screening guidelines; HSIL; AGC; ASC-H
14.  Estimating 5-year risk of CIN3+ to guide the management of women aged 21–24 
Journal of lower genital tract disease  2013;17(5 0 1):S64-S68.
Current US national guidelines recommend beginning screening at age 21 using Pap tests only, with cotesting starting at age 30. To inform the management of Pap test abnormalities among women aged 21–24, who have extremely low cancer risks, we compared risks of CIN3+ for women aged 21–24 versus 25–29 or 30–64.
We estimated 5-year risks of CIN3+ for Pap test results, with HPV triage of ASC-US, for 133,947 women aged 21–24, compared with 135,382 women age 25–29 and 965,360 women age 30–64, between 2003–2010 at Kaiser Permanente Northern California.
There were 3 cancers diagnosed during follow-up in women aged 21–24. Following high-grade Pap results (0.6% of Pap results), 5-year CIN3+ risks for women aged 21–24 were comparable to those aged 25–29 and 30–64 (AGC: 6.9% vs. 14% vs. 8.5%, p=0.8; ASC-H: 16% vs. 24% vs. 18%, p=0.8; HSIL: 28% vs. 28% vs. 47%, p=0.4). Following LSIL, 5-year CIN3+ risk was lower for ages 21–24 (3.0%) than ages 25–29 (5.0%, p=0.01) or ages 30–64 (5.2%, p=0.0002). Although 5-year CIN3+ risk for HPV-negative/ASC-US was similar across all 3 groups (0.57% vs. 0.59% vs. 0.43%, p=1), risk for HPV-positive/ASC-US was lower for age 21–24 (4.4%) than 25–29 (7.1%, p<0.0001) or 30–64 (6.8%, p<0.0001).
Women aged 21–24 had almost zero cancer risk, and positive Pap test results predicted low CIN3+ risk except for the 0.6% of Pap results that were high-grade. The generally low risk supports conservative management of women aged 21–24.
PMCID: PMC3616448  PMID: 23519307
Human Papillomavirus (HPV); cancer prevention; Pap; cervical intraepithelial neoplasia (CIN); Hybrid Capture 2 (HC2); young women
15.  Follow-up testing post-colposcopy: Five-year risk of CIN2+ after a colposcopic diagnosis of CIN1 or less 
Journal of lower genital tract disease  2013;17(5 0 1):S69-S77.
The majority of women referred for colposcopy are not diagnosed with CIN2+ but, nonetheless, are typically asked to return much sooner than their next routine screening interval in 3-5 years. An important question is how many subsequent negative Pap results, or negative Pap and HPV cotest results, are needed prior to returning to an extended retesting interval.
We estimated 5-year risks of CIN2+ for 3 follow-up management strategies after colposcopy (Pap-alone, HPV-alone and cotesting) for 20,319 women aged 25 and older screened from 2003-2010 at Kaiser Permanente Northern California who were referred for colposcopy but for whom CIN2+ was not initially diagnosed (i.e., “Women with CIN1/negative colposcopy”).
Screening results immediately antecedent to CIN1/negative colposcopy influenced subsequent 5-year CIN2+ risk: women with an antecedent HPV-positive/ASC-US or LSIL Pap had a lower risk (10%) than those with antecedent ASC-H (16%, p<0.0001) or HSIL+ (24%, p<0.0001). For women with an antecedent HPV-positive/ASC-US or LSIL, a single negative cotest approximately 1 year post-colposcopy predicted lower subsequent 5-year risk of CIN2+ (1.1%) than 2 sequential negative HPV tests (1.8%, p=0.3) or 2 sequential negative Pap results (4.0%, p<0.0001). For those with an antecedent ASC-H or HSIL+ Pap, 1 negative cotest after 1 year predicted lower subsequent 5-year risk of CIN2+ (2.2%) than 1 negative HPV test (4.4%, p=0.4) or 1 negative Pap (7.0%, p=0.06); insufficient data existed to calculate risk after sequential negative cotests for women with high grade antecedent cytology.
After CIN1/negative colposcopy followed by negative post-colposcopy tests, women did not achieve sufficiently low CIN2+ risk to return to 5-year routine screening. For women with antecedent HPV-positive/ASC-US or LSIL, a single negative post-colposcopy cotest reduced risk to a level consistent with a 3-year return. For women with antecedent ASC-H or HSIL+, no single negative test result sufficed to reduce risk to a level consistent with a 3-year return.
For women with CIN1/negative colposcopy and antecedent HPV-positive/ASC-US or LSIL, a single negative post-colposcopy cotest reduced risk to a level consistent with a 3-year return.
PMCID: PMC3616505  PMID: 23519308
Human Papillomavirus (HPV); cancer prevention; Pap; cervical intraepithelial neoplasia (CIN); Hybrid Capture 2 (HC2); colposcopy
16.  Five-year risk of CIN3+ and cervical cancer for women with HPV testing of ASC-US Pap results 
Journal of lower genital tract disease  2013;17(5 0 1):S36-S42.
New screening guidelines recommend that HPV-negative/ASC-US results be considered as equivalent to HPV-negative/Pap-negative results, leading to rescreening in 5 years. However, despite ample research data, the routine clinical performance of HPV testing of women with ASC-US has not been adequately documented.
We estimated 5-year risks of CIN3+ and cancer for 2 groups between 2003-2010 at Kaiser Permanente Northern California: 27,050 women aged 30-64 who underwent HPV and Pap cotesting and had an ASC-US Pap result, and 12,209 women aged 25-29 who underwent HPV triage of ASC-US.
Five-year risks of CIN3+ and of cancer for women aged 30-64 testing HPV-negative/ASC-US and for 923,152 women testing Pap-negative alone were similar although statistically distinguishable (CIN3+: 0.43% vs. 0.26% (p=0.001); Cancer: 0.050% vs. 0.025% (p=0.1, respectively)). The cancer risk increase for HPV-negative/ASC-US versus Pap-negative alone was confined to women aged 60-64 (0.26% vs. 0.035%, p=0.3). Five-year risks of CIN3+ and of cancer for women with HPV-negative/ASC-US were substantially higher than those for women testing HPV-negative/Pap-negative (CIN3+: 0.43% vs. 0.08% (p<0.0001); Cancer: 0.050% vs. 0.011% (p=0.003, respectively)). For women aged 30-64 testing HPV-positive/ASC-US, 5-year risks of CIN3+ and cancer were slightly higher than for the 9,374 women with LSIL (CIN3+: 6.8 % vs. 5.2% (p=0.0007); Cancer: 0.41% vs. 0.16% (p=0.04)). Similar patterns were seen for women aged 25-29.
Women with HPV-negative/ASC-US had similar risk as women testing Pap-negative alone, but had higher risk than women testing HPV-negative/Pap-negative. Based on the principle of “equal management of equal risks”, our findings support equal management of women with HPV-negative/ASC-US and those with Pap-negative alone, except for exiting women from screening because cancer risks at ages 60-64 may be higher for HPV-negative/ASC-US. Our findings also support managing HPV-positive/ASC-US and LSIL similarly.
Women testing HPV-negative/ASC-US have similar risk of CIN3+ or cancer as women testing Pap-negative alone, but have higher risk than women testing HPV-negative/Pap-negative.
PMCID: PMC3616508  PMID: 23519303
Human Papillomavirus (HPV); cancer prevention; Pap; cervical intraepithelial neoplasia (CIN); Hybrid Capture 2 (HC2); ASC-US
17.  Five-year risks of CIN2+ and CIN3+ for women with HPV-positive and HPV-negative LSIL Pap results 
Journal of lower genital tract disease  2013;17(5 0 1):S43-S49.
LSIL Pap results do not typically lead to HPV testing. HPV triage is not cost-effective because most cases are HPV-positive. However, under new national guidelines recommending cotesting for women ages 30–64, clinicians will increasingly receive the HPV test result for LSIL Pap results. Some authors have suggested that HPV triage might be effective at older ages, when the percentage of HPV positivity among women with LSIL declines.
We estimated 5-year risks of CIN2+ and CIN3+ for 9,033 women aged 30–64 who had both an HPV test and an LSIL Pap result.
HPV-positivity of LSIL decreased only slightly with age (30–34: 88% vs. 60–64: 72%, p<0.0001). The 5-year risks of CIN2+ and CIN3+ for women aged 30–64 testing HPV-positive/LSIL were larger than those for women testing HPV-negative/LSIL (CIN2+: 19% vs. 5.1%, p<0.0001; CIN3+: 6.1% vs. 2.0%, p<0.0001). The 5-year CIN3+ risk in HPV-negative/LSIL women was similar to that for women with ASC-US Pap without knowledge of HPV test results (2.0% vs. 2.6%, p=0.4).
HPV-negative/LSIL posed lower risk than other Pap results that guidelines currently recommend for referral to immediate colposcopy. By the principle of “equal management of equal risks,” women with HPV-negative/LSIL might reasonably be managed similarly to those with ASC-US Pap results without knowledge of HPV testing, i.e., re-testing at 6–12 months, rather than immediate colposcopy. Although the HPV test result for LSIL Pap results provides actionable information to clinicians who screen with cotesting, the high HPV-positivity of LSIL at even the oldest ages suggests the lack of cost-effectiveness of HPV triage of LSIL for clinicians who do not use routine cotesting.
PMCID: PMC3637971  PMID: 23519304
Human Papillomavirus (HPV); cancer prevention; cytology; cervical intraepithelial neoplasia (CIN); Hybrid Capture 2 (HC2); LSIL
18.  Selection Criteria for Lung-Cancer Screening 
The New England journal of medicine  2013;368(8):728-736.
The National Lung Screening Trial (NLST) used risk factors for lung cancer (e.g., ≥30 pack-years of smoking and <15 years since quitting) as selection criteria for lung-cancer screening. Use of an accurate model that incorporates additional risk factors to select persons for screening may identify more persons who have lung cancer or in whom lung cancer will develop.
We modified the 2011 lung-cancer risk-prediction model from our Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial to ensure applicability to NLST data; risk was the probability of a diagnosis of lung cancer during the 6-year study period. We developed and validated the model (PLCOM2012) with data from the 80,375 persons in the PLCO control and intervention groups who had ever smoked. Discrimination (area under the receiver-operating-characteristic curve [AUC]) and calibration were assessed. In the validation data set, 14,144 of 37,332 persons (37.9%) met NLST criteria. For comparison, 14,144 highest-risk persons were considered positive (eligible for screening) according to PLCOM2012 criteria. We compared the accuracy of PLCOM2012 criteria with NLST criteria to detect lung cancer. Cox models were used to evaluate whether the reduction in mortality among 53,202 persons undergoing low-dose computed tomographic screening in the NLST differed according to risk.
The AUC was 0.803 in the development data set and 0.797 in the validation data set. As compared with NLST criteria, PLCOM2012 criteria had improved sensitivity (83.0% vs. 71.1%, P<0.001) and positive predictive value (4.0% vs. 3.4%, P = 0.01), without loss of specificity (62.9% and. 62.7%, respectively; P = 0.54); 41.3% fewer lung cancers were missed. The NLST screening effect did not vary according to PLCOM2012 risk (P = 0.61 for interaction).
The use of the PLCOM2012 model was more sensitive than the NLST criteria for lung-cancer detection.
PMCID: PMC3929969  PMID: 23425165
19.  Targeting of Low-Dose CT Screening According to the Risk of Lung-Cancer Death 
The New England journal of medicine  2013;369(3):245-254.
In the National Lung Screening Trial (NLST), screening with low-dose computed tomography (CT) resulted in a 20% reduction in lung-cancer mortality among participants between the ages of 55 and 74 years with a minimum of 30 pack-years of smoking and no more than 15 years since quitting. It is not known whether the benefits and potential harms of such screening vary according to lung-cancer risk.
We assessed the variation in efficacy, the number of false positive results, and the number of lung-cancer deaths prevented among 26,604 participants in the NLST who underwent low-dose CT screening, as compared with the 26,554 participants who underwent chest radiography, according to the quintile of 5-year risk of lung-cancer death (ranging from 0.15 to 0.55% in the lowest-risk group [quintile 1] to more than 2.00% in the highest-risk group [quintile 5]).
The number of lung-cancer deaths per 10,000 person-years that were prevented in the CT-screening group, as compared with the radiography group, increased according to risk quintile (0.2 in quintile 1, 3.5 in quintile 2, 5.1 in quintile 3, 11.0 in quintile 4, and 12.0 in quintile 5; P = 0.01 for trend). Across risk quintiles, there were significant decreasing trends in the number of participants with false positive results per screening-prevented lung-cancer death (1648 in quintile 1, 181 in quintile 2, 147 in quintile 3, 64 in quintile 4, and 65 in quintile 5). The 60% of participants at highest risk for lung-cancer death (quintiles 3 through 5) accounted for 88% of the screening-prevented lung-cancer deaths and for 64% of participants with false positive results. The 20% of participants at lowest risk (quintile 1) accounted for only 1% of prevented lung-cancer deaths.
Screening with low-dose CT prevented the greatest number of deaths from lung cancer among participants who were at highest risk and prevented very few deaths among those at lowest risk. These findings provide empirical support for risk-based targeting of smokers for such screening. (Funded by the National Cancer Institute.)
PMCID: PMC3783654  PMID: 23863051
20.  A regression model for risk difference estimation in population-based case–control studies clarifies gender differences in lung cancer risk of smokers and never smokers 
Additive risk models are necessary for understanding the joint effects of exposures on individual and population disease risk. Yet technical challenges have limited the consideration of additive risk models in case–control studies.
Using a flexible risk regression model that allows additive and multiplicative components to estimate absolute risks and risk differences, we report a new analysis of data from the population-based case–control Environment And Genetics in Lung cancer Etiology study, conducted in Northern Italy between 2002–2005. The analysis provides estimates of the gender-specific absolute risk (cumulative risk) for non-smoking- and smoking-associated lung cancer, adjusted for demographic, occupational, and smoking history variables.
In the multiple-variable lexpit regression, the adjusted 3-year absolute risk of lung cancer in never smokers was 4.6 per 100,000 persons higher in women than men. However, the absolute increase in 3-year risk of lung cancer for every 10 additional pack-years smoked was less for women than men, 13.6 versus 52.9 per 100,000 persons.
In a Northern Italian population, the absolute risk of lung cancer among never smokers is higher in women than men but among smokers is lower in women than men. Lexpit regression is a novel approach to additive-multiplicative risk modeling that can contribute to clearer interpretation of population-based case–control studies.
PMCID: PMC3840559  PMID: 24252624
Additive risk; Absolute risk; Case–control study; EAGLE; Lung cancer; Risk assessment; Sex factors; Smoking
21.  Evaluating breast cancer risk projections for Hispanic women 
Breast cancer research and treatment  2011;132(1):10.1007/s10549-011-1900-9.
For Hispanic women, the Breast Cancer Risk Assessment Tool (BCRAT; “Gail Model”) combines 1990–1996 breast cancer incidence for Hispanic women with relative risks for breast cancer risk factors from non-Hispanic white (NHW) women. BCRAT risk projections have never been comprehensively evaluated for Hispanic women. We compared the relative risks and calibration of BCRAT risk projections for 6,353 Hispanic to 128,976 NHW postmenopausal participants aged 50 and older in the Women’s Health Initiative (WHI). Calibration was assessed by the ratio of the number of breast cancers observed with that expected by the BCRAT (O/E). We re-evaluated calibration for an updated BCRAT that combined BCRAT relative risks with 1993–2007 breast cancer incidence that is contemporaneous with the WHI. Cox regression was used to estimate relative risks. Discriminatory accuracy was assessed using the concordance statistic (AUC). In the WHI Main Study, the BCRAT underestimated the number of breast cancers by 18% in both Hispanics (O/E = 1.18, P = 0.06) and NHWs (O/E = 1.18, P < 0.001). Updating the BCRAT improved calibration for Hispanic women (O/E = 1.08, P = 0.4) and NHW women (O/E = 0.98, P = 0.2). For Hispanic women, relative risks for number of breast biopsies (1.71 vs. 1.27, P = 0.03) and age at first birth (0.97 vs. 1.24, P = 0.02) differed between the WHI and BCRAT. The AUC was higher for Hispanic women than NHW women (0.63 vs. 0.58, P = 0.03). Updating the BCRAT with contemporaneous breast cancer incidence rates improved calibration in the WHI. The modest discriminatory accuracy of the BCRAT for Hispanic women might improve by using risk factor relative risks specific to Hispanic women.
PMCID: PMC3827770  PMID: 22147080
Hispanic; Breast cancer; Risk prediction; Risk assessment; BCRAT
22.  Prospective study of genomic hypomethylation of leukocyte DNA and colorectal cancer risk 
Systematic genome-wide reductions of methylated cytosine (5-mC) levels have been observed in colorectal cancer tissue and are suspected to play a role in carcinogenesis, possibly as a consequence of inadequate folate intake. Reduced 5-mC levels in peripheral blood leukocytes have been associated with increased risk of colorectal cancer and adenoma in cross-sectional studies.
To minimize disease- and/or treatment-related effects, we studied leukocyte 5-mC levels in prospectively collected blood specimens of 370 cases and 493 controls who were cancer-free at blood collection from the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial., Leukocyte 5-mC level was determined by an HPLC/Tandem Mass Spectrometry method and expressed as the relative amount of methyl- to total cytosine residues, or %5-mC. We estimated the association between colorectal cancer risk and %5-mC categories by computing odds ratios (ORs) and 95% confidence intervals (CIs) through logistic regression modeling.
We observed no dose-dependent association between colorectal cancer and %5-mC categories (lowest tertile vs. highest: OR=1.14, 95% CI=0.80–1.63; P trend=0.51). However, among subjects whose 5-mC levels were at the highest tertile, we observed an inverse association between natural folate intake and colorectal cancer (highest tertile of natural folate vs. lowest: OR=0.35, 95% CI=0.17–0.71; P trend=0.003; P interaction=0.003).
This prospective investigation show no clear association between leukocyte 5-mC level and subsequent colorectal cancer risk, but a suggestive risk modification between 5-mC level and natural folate intake.
Adequate folate status may protect against colorectal carcinogenesis through mechanisms involving adequate DNA methylation in the genome.
PMCID: PMC3493855  PMID: 23001241
5-mC; PLCO; folate; colorectal
23.  Estimating the agreement and diagnostic accuracy of two diagnostic tests when one test is conducted on only a subsample of specimens 
Statistics in medicine  2011;31(5):10.1002/sim.4422.
We focus on the efficient usage of specimen repositories for the evaluation of new diagnostic tests and for comparing new tests with existing tests. Typically, all pre-existing diagnostic tests will already have been conducted on all specimens. However, we propose retesting only a judicious subsample of the specimens by the new diagnostic test. Subsampling minimizes study costs and specimen consumption yet estimates of agreement or diagnostic accuracy potentially retain adequate statistical efficiency. We introduce methods to estimate agreement statistics and conduct symmetry tests when the second test is conducted on only a subsample and no gold standard exists. The methods treat the subsample as a stratified two-phase sample and use inverse-probability weighting (IPW). Strata can be any information available on all specimens and can be used to oversample the most informative specimens. The verification bias framework applies if the test conducted on only the subsample is a gold standard. We also present IPW-based estimators of diagnostic accuracy that take advantage of stratification. We present three examples demonstrating that adequate statistical efficiency can be achieved under subsampling while greatly reducing the number of specimens requiring re-testing. Naively using standard estimators that ignore subsampling can lead to drastically misleading estimates. Through simulation, we assess the finite-sample properties of our estimators and consider other possible sampling designs for our examples that could have further improved statistical efficiency. To help promote subsampling designs, our R package CompareTests computes all of our agreement and diagnostic accuracy statistics.
PMCID: PMC3809897  PMID: 22139832
Verification Bias; Symmetry Test; Kappa; Two-Phase Design; HPV; sensitivity; specificity; Gold Standard
24.  Comparison of two expert-based assessments of diesel exhaust exposure in a case-control study: Programmable decision rules versus expert review of individual jobs 
Professional judgment is necessary to assess occupational exposure in population-based case-control studies; however, the assessments lack transparency and are time-consuming to perform. To improve transparency and efficiency, we systematically applied decision rules to the questionnaire responses to assess diesel exhaust exposure in the New England Bladder Cancer Study, a population-based case-control study.
2,631 participants reported 14,983 jobs; 2,749 jobs were administered questionnaires (‘modules’) with diesel-relevant questions. We applied decision rules to assign exposure metrics based solely on the occupational history responses (OH estimates) and based on the module responses (module estimates); we combined the separate OH and module estimates (OH/module estimates). Each job was also reviewed one at a time to assign exposure (one-by-one review estimates). We evaluated the agreement between the OH, OH/module, and one-by-one review estimates.
The proportion of exposed jobs was 20–25% for all jobs, depending on approach, and 54–60% for jobs with diesel-relevant modules. The OH/module and one-by-one review had moderately high agreement for all jobs (κw=0.68–0.81) and for jobs with diesel-relevant modules (κw=0.62–0.78) for the probability, intensity, and frequency metrics. For exposed subjects, the Spearman correlation statistic was 0.72 between the cumulative OH/module and one-by-one review estimates.
The agreement seen here may represent an upper level of agreement because the algorithm and one-by-one review estimates were not fully independent. This study shows that applying decision-based rules can reproduce a one-by-one review, increase transparency and efficiency, and provide a mechanism to replicate exposure decisions in other studies.
PMCID: PMC3439531  PMID: 22843440
25.  Prevalence of and Risk Factors for Anal Human Papillomavirus Infection Among Young Healthy Women in Costa Rica 
The Journal of Infectious Diseases  2012;206(7):1103-1110.
Background. Anal cancer is caused by human papillomavirus (HPV), yet little is known about anal HPV infection among healthy young women.
Methods. A total of 2017 sexually active women in the control arm of an HPV-16/18 vaccine trial had a single anal specimen collected by a clinician at the 4-year study visit. Samples were tested for HPV by SPF10 PCR/DEIA/LiPA25, version 1.
Results. A total of 4% of women had HPV-16, 22% had oncogenic HPV, and 31% had any HPV detected in an anal specimen. The prevalence of anal HPV was higher among women who reported anal intercourse, compared with those who did not (43.4% vs 28.4%; P < .001). Among women who reported anal intercourse, cervical HPV (adjusted odds ratio [aOR], 5.3 [95% confidence interval {CI}, 3.4–8.2]), number of sex partners (aOR, 2.2 [95% CI, 1.1–4.6] for ≥4 partners), and number of anal intercourse partners (aOR, 1.9 [95% CI, 1.1–3.3] for ≥2 partners) were independent risk factors for anal HPV detection. Among women who reported no anal intercourse, cervical HPV (aOR, 4.7 [95% CI, 3.7–5.9]), number of sex partners (aOR, 2.4 [95% CI, 1.7–3.4] for ≥4 partners), and report of anal fissures (aOR, 2.3 [95% CI, 1.1–4.8]) were associated with an increased odds of anal HPV detection.
Conclusion. Anal HPV is common among young women, even those who report no anal sex, and was associated with cervical HPV infection. Anal fissures in women who report never having had anal intercourse may facilitate HPV exposure.
Clinical Trials Registration. NCT00128661.
PMCID: PMC3499109  PMID: 22850119

Results 1-25 (60)