Danping Liu

Statistical approaches using longitudinal biomarkers for disease early detection: A comparison of methodologies

Early detection of clinical outcomes such as cancer may be predicted using longitudinal biomarker measurements. Tracking longitudinal biomarkers as a way to identify early disease onset may help to reduce mortality from diseases like ovarian cancer that are more treatable if detected early. Two disease risk prediction frameworks, the shared random effects model (SREM) and the pattern mixture model (PMM) could be used to assess longitudinal biomarkers on disease early detection. In this article, we studied the discrimination and calibration performances of SREM and PMM on disease early detection through an application to ovarian cancer, where early detection using the risk of ovarian cancer algorithm (ROCA) has been evaluated. Comparisons of the above three approaches were performed via analyses of the ovarian cancer data from the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial. Discrimination was evaluated by the time‐dependent receiver operating characteristic curve and its area, while calibration was assessed using calibration plot and the ratio of observed to expected number of diseased subjects. The out‐of‐sample performances were calculated via using leave‐one‐out cross‐validation, aiming to minimize potential model overfitting. A careful analysis of using the biomarker cancer antigen 125 for ovarian cancer early detection showed significantly improved discrimination performance of PMM as compared with SREM and ROCA, nevertheless all approaches were generally well calibrated. Robustness of all approaches was further investigated in extensive simulation studies. The improved performance of PMM relative to ROCA is in part due to the fact that the biomarker measurements were taken at a yearly interval, which is not frequent enough to reliably estimate the changepoint or the slope after changepoint in cases under ROCA.

Validation of TypeSeq2, a Next-Generation–Based Sequencing Assay for the Detection of 46 Human Papillomavirus Genotypes, at the US National Cancer Institute and Costa Rica Laboratories

Abstract Background Cervical cancer is caused by persistent infection with carcinogenic human papillomavirus (HPV) genotypes. Prophylactic HPV vaccines are highly efficacious in preventing the acquisition of HPV infection. HPV vaccine trials and epidemiologic studies based on virologic endpoints rely on valid and reproducible measurements of HPV. We evaluated the second version of TypeSeq (TS2), a next-generation, sequencing-based assay that detects 46 HPV genotypes, in a historical phase 3 clinical trial. Methods We used 1214 stored cervical samples from women enrolled in the Costa Rica HPV Vaccine Trial with available HPV results from Short PCR Fragment 10- Line Probe Assay 25 (SPF10-LiPA25). TS2 was first validated at the National Cancer Institute (NCI) and transferred to the laboratory in Costa Rica, where we conducted a second validation study. We compared TS2 results generated at each laboratory to the SPF10-LiPA25 results. Results Overall, each laboratory demonstrated high positive agreement for most carcinogenic and noncarcinogenic genotypes between TS2 and SPF10-LiPA25. Intralaboratory comparisons revealed very high agreement in repeated testing. Interlaboratory comparisons showed high agreement for most carcinogenic and noncarcinogenic types. Overall, there were no statistically significant differences in vaccine efficacy in the according-to-protocol cohort using TS2 (either in NCI or Costa Rica) or SPF10-LiPA25 (McNemar P values >.05). Costa Rica produced similar vaccine efficacy estimates as NCI for HPV16/18, HPV31/33/45, and HPV35/39/51/52/56/58/59 as NCI (P values ≥.36). Conclusions Compared to SPF10-LiPA25, a well-established standard for HPV genotyping, TS2 demonstrated high accuracy. Inter- and intralaboratory comparisons demonstrated that TS2 is valid and reproducible. TS2 can accurately classify the presence of HPV, which is essential in HPV vaccine trials evaluating virological endpoints. Clinical Trials Registration NCT00128661.