Primary human papillomavirus testing versus CO-testing: clinical outcomes in populations with different disease prevalence.

Implementation of primary human papillomavirus (HPV) testing has been slow in the US perhaps due to concerns of decreased sensitivity compared with concurrent HPV and cytology testing (cotesting). We used the National Breast and Cervical Cancer Early Detection Program (NBCCEDP) and the Kaiser Permanente of Northern California (KPNC) cohort to quantify potential trade-offs with primary HPV compared with cotesting in four US populations with differing precancer/cancer prevalence. In all settings, cotesting required more lab tests and more colposcopies compared to primary HPV testing. Additional cervical intraepithelial neoplasia grade 3 or cancer (CIN3+) immediately detected from cotesting vs primary HPV decreased with decreasing population-average CIN3+ prevalence, from 71 per 100,000 screened among never/rarely screened individuals in the NBCCEDP (prevalence = 1,212 per 100,000) to 4 per 100,000 screened among individuals with prior HPV-negative results in KPNC (prevalence = 86 per 100,000). These data suggest that cotesting confer an unfavorable benefit-to-harm ratio over primary HPV testing.

Estimating human papillomavirus vaccine efficacy from a single-arm trial: proof-of-principle in the Costa Rica Vaccine Trial.

BACKGROUND: The World Health Organization recommends a 1- or 2-dose human papillomavirus (HPV) vaccination schedule for females aged 9 to 20 years. Studies confirming the efficacy of a single dose and vaccine modifications are needed, but randomized controlled trials are costly and face logistical and ethical challenges. We propose a resource-efficient single-arm trial design that uses untargeted and unaffected HPV types as controls. METHODS: We estimated HPV vaccine efficacy (VE) from a single arm by comparing 2 ratios: the ratio of the rate of persistent incident infection with vaccine-targeted HPV 16 and 18 (HPV 16/18) and cross-protected types HPV 31, 33, and 45 (HPV 31/33/45) to vaccine-unaffected types HPV 35, 39, 51, 52, 56, 58, 59, and 66 (HPV 35/39/51/52/56/58/59/66) vs the ratio of prevalence of these types at the time of trial enrollment. We compare VE estimates using only data from the bivalent HPV 16/18 vaccine arm of the Costa Rica Vaccine Trial with published VE estimates that used both the vaccine and control arms. RESULTS: Our single-arm approach among 3727 women yielded VE estimates against persistent HPV 16/18 infections similar to published 2-arm estimates from the trial (according-to-protocol cohort: 91.0% , 95% CI = 82.9% to 95.3% [single-arm] vs 90.9% , 95% CI = 82.0% to 95.9% [2-arm]; intention-to-treat cohort: 41.7%, 95% CI = 32.4% to 49.8% [single-arm] vs 49.0% , 95% CI = 38.1% to 58.1% [2-arm]). VE estimates were also similar in analytic subgroups (number of doses received; baseline HPV serology status). CONCLUSIONS: We demonstrate that a single-arm design yields valid VE estimates with similar precision to a randomized controlled trial. Single-arm studies can reduce the sample size and costs of future HPV vaccine trials while avoiding concerns related to unvaccinated control groups. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT00128661.

Multistate models for the natural history of cancer progression.

BACKGROUND: Multistate models can be effectively used to characterise the natural history of cancer. Inference from such models has previously been useful for setting screening policies. METHODS: We introduce the basic elements of multistate models and the challenges of applying these models to cancer data. Through simulation studies, we examine (1) the impact of assuming time-homogeneous Markov transition intensities when the intensities depend on the time since entry to the current state (i.e., the process is time-inhomogenous semi-Markov) and (2) the effect on precancer risk estimation when observation times depend on an unmodelled intermediate disease state. RESULTS: In the settings we examined, we found that misspecifying a time-inhomogenous semi-Markov process as a time-homogeneous Markov process resulted in biased estimates of the mean sojourn times. When screen-detection of the intermediate disease leads to more frequent future screening assessments, there was minimal bias induced compared to when screen-detection of the intermediate disease leads to less frequent screening. CONCLUSIONS: Multistate models are useful for estimating parameters governing the process dynamics in cancer such as transition rates, sojourn time distributions, and absolute and relative risks. As with most statistical models, to avoid incorrect inference, care should be given to use the appropriate specifications and assumptions.