The prophylactic efficacy of poly(ICLC) ( Poly(ICLC), the stabilized, synthetic, double-stranded polyriboinosinic-polyribocytidylic acid, is an effective in vivo immunomodulator and interferon inducer (2,7,9,14). In primates and rodents, poly(ICLC) has proved useful in the prophylaxis of several viral infections (1,3, 5,8,11,12). Poly(ICLC) confers an antiviral state that persists for several days by stimulation of the immune response (2, 14) and induction of interferon production (7,8). Interferon and the cellular immune response are both believed to play a role in altering the pathogenesis of viral infections.The objectives of this study were to identify the optimal prophylactic treatment schedule of poly(ICLC) in Rift Valley fever virus (RVFV)-infected mice, delineate the optimal timing between the first and additional doses of drug, and determine the minimal effective dose as a guide for a possible human study. mented with 16 mM HEPES [N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid], 7.5% heat-inactivated fetal bovine serum, and 5 ,ug of gentamicin per ml). Cells were further incubated at 37°C for 96 h and stained with 0.1% crystal violet to make the plaques visible. In the in vivo efficacy studies, mice were injected subcutaneously in the left groin with 0.1 ml (250 PFU) of virus.Statistical model. Determination of the optimal treatment schedule was done by ranking the treatment efficacy with the Cox proportional-hazard model, which defines efficacy in terms of the incremental relative risk of death compared with the relative risk of death of the standard day 0 treatment (4). The Cox model uses information from the entire survival curve to estimate the hazard of mortality as a function of the treatment schedule. The mathematical basis of the model is given by the equation H(T,Z) = g(7) exp(bZ), where T is time and Z is the indicator variable for schedule, H(T,Z) is the hazard (instantaneous rate of death) at time T for an animal in schedule Z, g (7) is the underlying hazard function for the standard schedule (Z = 0), and b is the regression coefficient for schedule Z.The factor exp(bZ) is the incremental relative risk of death (relative to the standard, defined as Z = 0; in our case, treatment and challenge at day 0). A test of b = 0 determines whether the incremental relative risk of death is significantly different from that of day 0 standard treatment, either increasing the risk (b > 0) or decreasing the risk (b < 0) relative to the standard. Special statistical methods were used to fit the model by using software from the Biomedical Data Processing computer package (4).
RESULTSDetermination of optimal prophylactic treatment schedule. In the exploratory prophylactic study, eight or seven doses of 20 ,g of poly(ICLC) per mouse were administered from day -4 until day +10 or from day -1 until day +10, respectively ( Fig. 1). Although these multiple-dose treatment schedules yielded 100% and 90% long-term survivors, respectively, practical purposes dictated the determination of the least number of inject...