Well...TI has a different modulus of elasticity than steel (among other things). Fatigue life depends on many parameters so I can only assume that "TI springs last 5 times longer than steel" just means that whoever designed them weighted their design towards a bit more fatigue life. Hard to make blanket comparisons.
Now that you've posted this stuff, I'm realizing that this component's task is different than I initally thought. It isn't so much a loaded suspension component as it is a position/location/movement component. That is, one could substitute a pivot-shackle (no "spring") and realize the same result. Actually, I imagine one could indeed somehow delete the TI leaf and sub some sort of internal locator for the rear suspension without altering the ride characteristics. The current design is pretty simple though and well thought-out.
If I were a weight-weenie, I'd be wondering if the leaf mounts (nuts, bolts, etc) could be deleted and a lighter system implemented. Tension/compression stress on that leaf are probably quite light (keeping the top of the rear wheel fork in place). I guess if I were a complete weight-weenie, I'd be deleting the suspension altogether...
Not gonna happen here. I want the suspension for all the designed reasons. I had to put off my planned Silvio purchase as we're in the midst of buying a new home...and I haven't sold my Catrike yet. Ah well...
edit: I neglected braking forces (partially due to my ignorance regarding the connection made thru the elastomers) in my thought process. If the braking forces are translated thru the elastomer mount to the seat frame, the spring may 'see' only minor compressive forces. However, if the elastomer does not wholly locate the top of the rear fork, braking forces are reacted by the TI spring via compression. I can see where the combination of a torsion load and braking might severely stress the spring at the point load. If the spring is reversing itself (during loading and unloading cycles, say, over a rough road) and braking is applied, I can see where one might possibly exceed the yield strength of the spring over time...fatigue cracking, etc.