I disagree with you. If the risk is 100% as it is with some ALS-causing gene mutations, as well as one that runs in my (in-law) family, it's does indeed alone explain risk. At least for the disease in my family, timing of onset is relatively consistent too (+/- 5 years or something). And maybe more importantly, the pathology is such that internal damage is happening for years, probably decades, before the first outward symptoms appear. If "family member noticed a tiny little problem that maybe was nothing, maybe the first symptom of the disease, but was concerned enough to go to the doctor" is the criteria for diagnosing the disease, and starting treatment, that's way, way more subjective and silly than looking for a specific gene mutation that causes the underlying pathology.
In the liability threshold model for disease, complex traits can certainly have monogenic variants that essentially guarantee that disease will occur. For example, homozygous FH (LDLR) will cause clinical disease onset in infancy or early childhood. But most people with FH don’t have homozygous FH.
In fact, we know that ALS is a complex disease because we know that common genetic variants influence ALS risk, but that doesn’t mean that every monogenic ALS variant is going to have its penetrance and onset modified to the same degree. So by saying that ALS is a complex disease I don't mean to take away from the fact that for some families, their causal variant may have a consistent effect with only minimal variation from the genetic background.
I’m fully supportive of preventive therapy. I think that is the actual future of medicine. And so I am a big fan of the idea of using genetics to identify a population at risk, doing individual biochemistry (or whatever endophenotype people use - perhaps imaging or other measurements) to see if the pathophysiologic process is occurring, and treating to prevent disease.
