Nothing will cause a door to fail more quickly than being rigged too tightly closed. Every gear cycle sets up stresses around the actuator rod end attachment. Ultimately, the mount is torn from the parent material. This may have been a factor in your instance. Do you remember from your walkaround whether or not the door appeared to be underflush ? Did it appear to be sitting proud perhaps ? In this case, it's not impossible that repetitive exposures to slipstream could have played a part in weakening the structure. As Ghengis has said, composite stress is VERY difficult to detect until such times as you have a failure.
I have seen instances of gear strut doors, also wheel well doors on widebodies fail because of being rigged too firmly closed.
Like Ghengis, I'm not familiar with the type but am surprised that you would have composite doors.