Longitudinal twin analysis with at most three time points

Hello, I'm working on a twin database (monozigotic and dyzigotic) with 537 twin pairs, of which 311 are present in only one wave, 139 have two time points and 77 three data points.

I fitted the saturated and ACE models on the 311 with one time point only, but I'm wondering how to use the longitudinal aspect in the 216 twin pairs, as I did with linear mixed models in which I added random components.

Can you suggest a way (and/or also a possibile script) for this analysis?

Many thanks in advance

Ugo Lancia