The original Wakefield paper that people used to cite from the Lancet about the autism-vaccine link always confused the hell out of me as a scientist.
The autism link was speculative at best even as written. If I recall correctly, it was based on data from the paper about how vaccines might affect gut health and then how gut health in turn might have something to do with triggering autism. It was speculative and while there was some initial data presented indicating a correllation, there was no clear pathway and it was presented as a hypothesis that should be looked at in the future, not as a definitive conclusion on a causal pathway for autism expression.
Hypotheses get posited and debunked all the time, sometimes some zany stuff.
His co-authors gradually withdrew support for the major findings of the paper, and it turned out Wakefield either falsified or cherry picked the data that led him to speculate on that link in the first place. He was stripped of his medical license and not a single major follow up study has ever come close to replicating his results.
You can usually find an academic paper or two, often in a reputable journal, saying almost anything you'd like. The question isn't "Can I find a paper or two linking mountain biking with brain cancer?" or "Might climate change be natural instead of man-made?" The answer is almost definitely yes. The questions is, "What do the bulk of scientific papers looking at the relation between XYZ and various cancer types show, and how statistically significant are those results?" Just like polling, sometimes you get wonky samples and erroneous results without bad faith.
The shame of it is that environmental triggers for expressing latent autism are poorly understood, and the medical research community spent an inordinate amount of time trying to replicate and then debunking Wakefield's study instead of exploring other causes.