Markov model

• Short-term (8 days) comparative efficacy between fostemsavir and placebo from the BRIGHTE trial
• Long-term comparative efficacy of fostemsavir plus OBT vs. OBT alone from a MAIC to the VIKING-3 study population

• The comparative clinical effectiveness of adding fostemsavir to OBT is uncertain due to the short observation period of the BRIGHTE trial (8 days). Additionally, long-term comparative effects were estimated through a MAIC with methodological limitations that make the magnitude of benefit of fostemsavir highly uncertain.
• The method used to model the natural history of patients with HIV based on CD4+ count lacked transparency and could not be validated. This added additional uncertainty to the estimated long-term clinical effectiveness estimates.
• Nearly all incremental QALYs were estimated through extrapolation, but no evidence was available to quantify the durability of fostemsavir’s effect on CD4+ count over time.
• The sponsor’s pharmacoeconomic model assumed that CD4+ count and viral load were independent, with equal transition probabilities between CD4+-based health states irrespective of viral load. This assumption was not supported by evidence or clinical expertise.
• Uncertainty around multiple inputs in the model was based on arbitrary values rather than evidence from the trial, the MAIC, or the literature. Consequently, the uncertainty has not been effectively captured in the model.
• Parameter uncertainty within the model appears to introduce an asymmetric bias in estimated costs and QALYs. This asymmetry creates a notable discrepancy between deterministic and probabilistic results that favoured OBT alone.

• Given the limitations identified within the sponsor’s economic analysis, CADTH was not able to use the model to provide a more reliable estimate of the cost-effectiveness of fostemsavir. The sponsor’s submitted results produced an ICER of $469,086 per QALY gained (incremental cost = $315,607; incremental QALYs = 0.673)
• Based on the sponsor’s analysis, a 94% price reduction would be required for fostemsavir plus OBT to be considered cost-effective at a WTP threshold of $50,000 per QALY gained compared to OBT alone. Even with this price reduction, the probabilistic results suggest a 36% probability that fostemsavir would not be cost-effective at any WTP threshold, due to high uncertainty around the predicted QALYs.