FR: `fixed_design_xx()` with fixed enrollment rates and fixed minimum follow-up

[DanChaltiel]

Hi,

I am trying to reproduce a gsDesign::nSurv() calculation using gsDesign2, for a fixed design with stratified survival assumptions.

In gsDesign::nSurv(), the documentation states that both of the following are supported:

• Lachin and Foulkes (1986): fixed trial duration, enrollment rates scaled to achieve the desired power
• Kim and Tsiatis (1990): fixed enrollment rates with either fixed enrollment duration or fixed minimum follow-up

For example, in nSurv(), I can use T = NULL and minfup = 36 to solve for the required enrollment duration, given fixed enrollment rates.

I expected gsDesign2::fixed_design_lf() to possibly support the same mode, since it implements the Lachin-Foulkes method and seems to call gsDesign::nSurv() internally.

However, from the current fixed_design_lf() interface, I only see study_duration, not minfup, and I do not see an equivalent of T = NULL. My understanding is that fixed_design_lf() currently supports the fixed study duration setting, where enrollment rates are rescaled to achieve the target power, but not the fixed enrollment rate + fixed minimum follow-up setting.

Could you confirm whether this interpretation is correct?

If this mode is currently not supported in gsDesign2, is there a recommended workaround? Is support for the nSurv(T = NULL, minfup = ...) use case planned for fixed_design_lf() or another function?

Thanks!