Evaluation of log‐rank tests for infrequent observations from a multi‐state process, with application to HPV vaccine efficacy

Abstract

Genital infection by human papillomavirus (HPV) is a common sexually transmitted disease, with over 25 per cent prevalence among young women in the US. Infections are usually without symptoms and transient (or reversible), but a small proportion of infections persist and are believed to be responsible for nearly all cervical cancers and precursor lesions such as cervical intraepithelial neoplasia (CIN). Therefore, successful vaccines against persistent HPV infections could have a great impact in preventing cervical cancers. In trials being planned, ongoing, and recently completed, a log-rank or a similar test may be employed to assess a vaccine effect in comparison to placebo, with an infection 'event' defined to capture persistent but not transient infections. However, it is not clear how best to define such an event, because (1) diagnostic tests cannot distinguish a persistent from a transient infection, (2) participants are only examined periodically, and (3) there can be misclassification errors in the detection of infections. This paper evaluates several definitions of persistent infection that are based on periodically observed infection statuses by postulating a multi-state model for persistent and transient infections. The type I error and the power of tests on vaccine efficacy based on these operational definitions are then examined under various scenarios of how a vaccine might affect the infection-disease process. We find that none of the candidates performs satisfactorily, thus raising concerns that clinical trials based only on infection endpoints will not be reliable.

Evaluation of operational chronic infection endpoints for HCV vaccine trials

“…The first three endpoints are similar to the ones considered in [18] in the application to HPV. In a study with 3-month visit intervals, the endpoints +++ and ++…+ correspond to chronic infection marked by HCV persistence for at least 6 months found in the literature [17].…”

“…As an example, presence of HCV below what can be detected by the assay due to the assay lower limit of detection may contribute to the detection error. We make the following conditional independence assumption regarding the misclassification errors due to imperfect HCV tests, as assumed in [18]:…”

“…Although our interest is in testing for VE P , the potential efficacies against acquiring and in clearing transient infections, given by VE T and VE C , may distort the type I error of the test on VE P , as discussed in [18]. While the global null hypothesis, maintains the type I error given by the nominal size of the test, the composite null hypothesis of interest, that the vaccine has no effect on the transition intensity for developing a chronic infection may not.…”

“…The null hypothesis of interest H 0 is composite. When the null hypothesis is composite, the type I error can be treated as power [18]. The type I error for the composite null hypothesis of interest H 0 is the probability of obtaining a significant result when VE P =0 but at least one of δ or η is nonzero; that is, the type I error for the test on VE P depends on the values of VE T and VE C .…”

“…In this paper, we consider several operational definitions of HCV infection that may be used to indicate chronic infections, in the context of a vaccine trial. These definitions are operationally defined as testing positive for HCV detection in consecutive visits, similar to the definitions considered in the HPV setting [18]. We aim to evaluate various operational definitions of HCV chronic infection as endpoint candidates for a vaccine trial by examining the type I error rate and the power of the log-rank test using each endpoint candidate.…”