The estimation of the treatment effect in the two-sample problem with right censoring is of interest in survival analysis. In this article we consider both the location shift model and the scale change model. We establish the large-sample properties of a generalized Hodges-Lehmann type estimator. The strong consistency is established under the minimal possible conditions. The asymptotic normality is also obtained without imposing any conditions on the censoring mechanisms. As a by-product, we also establish a result for the oscillation behavior of the Kaplan-Meier process, which extends the Bahadur result for the empirical process to the censored case.

A generalized Hodges-Lehmann type estimator for the treatment effect in the two-sample problem with right censoring, is proposed based on an inverse-quantile-type idea using truncated versions of the Kaplan-Meier estimators over the subspace where they are consistent. Its strong consistency and asymptotic normality can be obtained, under no conditions on the uninformative censorings, and the resulting variance is easily estimable from the data. In simulation studies the proposed estimator is superior to existing procedures in the presence of heavy unequal censoring.

▪ Objective: To evaluate the safety and immunologic and antiviral effects of combination therapy with zidovudine and dideoxycytidine (ddC) in patients with advanced human immunodeficiency virus type 1 (HIV) infection. ▪ Design: A phase I/II open-label, dose-ranging study. ▪ Setting: Two AIDS Clinical Trials Group units. ▪ Patients: Patients (56) with advanced HIV disease. ▪ Interventions: Patients were randomly assigned to one of three paired regimens of zidovudine and ddC. We evaluated six dosing regimens, each involving oral administration of the study drugs at 8-hour intervals. ▪ Measurements: Pharmacokinetics, toxicity, CD4 counts, p24 antigenemia and clinical end points. ▪ Main Results: The median follow-up period was 40.6 weeks (range, 0.3 to 70 weeks). Neither drug affected the pharmacokinetic profile of the other. Episodes of serious hematologic toxicity were infrequent, occurring in only 17.9% of patients, and did not differ among the regimens (P = 0.15). Severe sensory peripheral neuropathy occurred in two patients (one patient each in regimens 1 and 4). One patient receiving regimen 4 died. The mean maximal increase in CD4 counts exceeded 109 cells/mm3, and 69% of patients receiving combinations containing 300 or 600 mg of zidovudine daily had an increase in CD4 counts of 50 cells/mm3 or greater. Regimens containing 600 mg of zidovudine daily (regimens 2 and 5) were also more likely to result in persistent increases in CD4 counts above pretreatment values than were the two lowest dose regimens (P = 0.003). The decline in CD4 counts was more rapid, and the suppression of the p24 antigenemia was less rapid and less sustained in patients receiving the lowest zidovudine dose alone (regimen 6). The addition of ddC to regimen 6 (regimen 3) resulted in a slower decline in the CD4 counts (P = 0.06). ▪ Conclusions: Combination therapy with zidovudine and ddC at the doses tested was well tolerated and did not result in toxicity. A daily oral dose of 150 mg of zidovudine appeared to produce a suboptimal effect on p24 antigenemia and CD4 counts. Combination therapy with ddC and higher doses of zidovudine produced greater and more persistent effects in patients with advanced HIV infection compared with other study regimens and with the results of previous trials of zidovudine monotherapy.