I once had a set of fast race wheels that I really liked- they were american classic hubs, tubular matrix iso aero rims with bladed spokes laced radially except crossed on the rear drive side. 1994 US amature road championships were held in a downtown Seattle circiut that featured some very steep decents and I was riding the radially laced wheels. I had to brake hard on a downhill to make a sharp turn at the bottom of the hill. There was a lot of force being generated by braking and I had a slight shudder of my front brake then BAM, a section of the front hub flange broke loose, wheel went massivly out of true and several spokes were slapping the front fork. I somehow managed to keep the bike upright and slow down enough to make the corner without taking out 1/2 the pack and then quickly got a neutral wheel change and chased back onto the main pack on a long climb.
As I later analyzed the wheel failure, I concluded that the spoke tension of radially lacing is enourmously high during hard rim braking, probably much higher tension than can be generated at the rear wheel while sprinting hard. In order for braking forces to be transmitted from the rim via the spokes and to the hub and the forks, the rim must rotate to lag slightly behind in relation to the hub. This rotation is accomplished by stretching of the radial spokes. The tension forces involved in producing windup are much higher than any cross lacing pattern where 1/2 of the spokes are already angled so can transmitt braking force without windup.
I got a new hub and re-built the wheel as a 1X patten and have not sufferend any similar failures.