Background : The safety of newly approved drugs must be assessed using postmarketing surveillance data. One of the difficulties in assessing the hazard rates of adverse events induced by the anti-cancer drug TS-1 was that the time to event was not exactly identified due to the interval censoring. Most patients were outpatients who underwent clinical laboratory tests almost periodically at 1- or 2-week intervals and therefore, the occurrence of an adverse event was confirmed at the time of testing days after the event occurrence.
Objective : The purpose of this study was to propose a new model of hazard functions for each of 4 items of adverse event induced by TS-1 using post-marketing surveillance data considering the interval censoring.
Methods : The data obtained from 3, 294 patients with gastric cancer who received an initial 4-week course of therapy with TS-1 administered orally twice a day, followed by a 4-week second course with a 2-week no-treatment period after the initial course, were used to estimate hazard functions. Four items of adverse event--hemoglobin level (HB), white blood cell (WBC), neutrophil (NEUT) and platelet counts (PLT) --were graded, respectively, using the criteria established by the Japan Society of Clinical Oncology. Slip-mixed log-logistic and slip-mixed Weibull models were proposed as candidate models for estimating hazard functions. The goodness of fit of the two candidate models was evaluated by applying them to the above-mentioned data. The hazard functions for each of 4 items were assessed using the model with the better fit.
Results : The initial occurrence of adverse event was shown to follow the slip-mixed log-logistic model for each of 4 items. Although most events occurred early on in the initial course of therapy, a small peak in HB was also observed in the second course, while no such peak appeared for the other items.

抄録全体を表示

Pharmacoepidemiological studies are often conducted to evaluate the association between the event with low incidence and exposure with low prevalence. To overcome the difficulty due to low event incidence and low exposure prevalence, White proposed a two-stage design in 1982 and Cain and Breslow further developed this design.
In this article, the case-cohort and nested case-control designs are reviewed in contrast with the two-stage design. In addition, the usefulness of the case-cohort design in pharmacoepidmiology is assessed in comparison with that of the nested case-control design. In the nested case-control study, the control subjects are selected only after cases occur irrespective of whether the method of “risk set matching” or “unmatched density sampling” is used. Therefore, all of the events evaluated in the study must be clearly defined in advance. On the other hand, in a prospective case-cohort study, a single subcohort, selected independent of cases, is used to analyze multiple outcomes. Owing to this feature, the case-cohort design may be useful to study unknown adverse events that have not been specified as a target event prior to the study but are recognized as a problem that requires in-depth evaluation during or after the study is conducted. Weaknesses and limitations of the case-cohort design as compared to the nested case-control design are also discussed.
Data yielded by simulations for hypothetical case-cohort studies are analyzed by the SAS PHREG procedure with the robust variance estimation. The program used and results of simulations are presented.
The case-cohort study design may be useful for various pharmacoepidemiological studies in Japan where no large medical database is available.

抄録全体を表示