Treatment options for HER-2-positive metastatic breast cancer (mBC) patients have expanded markedly since trastuzumab approval in 1998. Several other regimens are now available, including pertuzumab plus trastuzumab plus docetaxel, T-DM1, capecitabine plus lapatinib, and trastuzumab plus lapatinib. This study assesses the cost-effectiveness of four treatment sequences for HER-2-positive mBC according to the Taiwanese National Health Insurance Administration (TNHIA). Costs (U.S. Dollars) and effectiveness (quality-adjusted life years) of four treatment sequences for HER-2-positive mBC patients were examined using a Markov model over a lifetime horizon. Transition probabilities, disease progression, and probability of adverse events and survival were derived from clinical trial data. Costs and health utilities were estimated from TNHIA, Taipei Medical University Hospital, and the literature. Deterministic, probabilistic sensitivity analyses and a scenario analysis examined parameter uncertainty and accounted for drug wastage in dosage and cost calculations. Sequence 3 (1st line: trastuzumab plus docetaxel; 2nd line: T-DM1; 3rd line: trastuzumab plus lapatinib) was the most cost-effective sequence followed by sequence 1 (1st line: pertuzumab plus trastuzumab plus docetaxel; 2nd line: T-DM1; 3rd line: capecitabine plus lapatinib), and sequence 4 (1st line: trastuzumab plus docetaxel; 2nd line: trastuzumab plus lapatinib; 3rd line: trastuzumab plus capecitabine), respectively. The model was sensitive to costs and transition probabilities, but not particularly sensitive to the wastage assumption. From the perspective of the TNHIA, trastuzumab plus docetaxel as 1st line followed by T-DM1 and trastuzumab plus lapatinib as 2nd and 3rd line represents the most cost-effective strategy among the four sequences considered for treating HER-2-positive mBC patients.

Concerns about the effects of propranolol on the central nervous system (CNS) in the infantile hemangioma (IH) population have been raised. We conducted a meta-analysis of the CNS and sleep-related effects of oral propranolol in IH patients. PubMed, Embase, Cochrance, Web of Science, and Clinicaltrials.gov were searched for relevant studies. We included clinical trials that compared oral propranolol with other treatments among IH patients under 6 years old and monitored and reported any adverse events. Study characteristics, types and number of adverse events were abstracted. Cochrane Collaboration Risk of Bias Tool was used to assess risk of bias. Our main outcomes were CNS and sleep-related effects. Random-effects models were used to estimate the pooled risk ratio. We did not observe statistically significant associations between oral propranolol and CNS or sleep-related effects. Oral propranolol appeared to have a safer profile of CNS effects than corticosteroids (RR = 0.27, 95% CI 0.02–3.00), but had an increased risk versus non-corticosteroids (for CNS effect, RR = 1.40, 95% CI 0.86–2.27; for sleep-related effects, RR = 1.63, 95% CI 0.88–3.03). Despite no statistically significant associations, there were suggestive findings of increased CNS effects and sleep-related risk of propranolol versus non-corticosteroids. In practice, CNS and sleep-related events should be monitored more closely among IH patients treated with oral propranolol.

Generic entry of newer anticoagulants is expected to decrease the costs of atrial fibrillation management. However, when making switches between brand and generic medications, bioequivalence failures are possible. The objectives of this study were to predict and compare the lifetime cost-effectiveness of brand dabigatran with hypothetical future generics. Markov micro-simulations were modified to predict the lifetime costs and quality-adjusted life years of patients on either brand or generic dabigatran from a U.S. private payer perspective. Event rates for generics were predicted using previously developed pharmacokinetic-pharmacodynamic models. The analyses showed that generic dabigatran with lower-than-brand systemic exposure was dominant. Meanwhile, generic dabigatran with extremely high systemic exposure was not cost-effective compared to the brand reference. Cost-effectiveness of generic medications cannot always be assumed as shown in this example. Combined use of pharmacometric and pharmacoeconomic models can assist in decision making between brand and generic pharmacotherapies.