yeah, you have to be careful with the studies that equate stuff in cell culture with stuff in real organisms.
we have metrics for "chemical X probabalistically causes cancer Y in tissue Z".
you can play games with the ROA and the probability. but the type of chemical (detergent, etc) is a non sequitor. if you expose cells to a detergent in high enough concentration, they'll lyse.
lysed cells are an ENTIRELY distinct "type" of cell death than the kind of cell death signature that you would see if the cell were cancerous. the author of that study didn't do any flow cytometry on his 293 T samples by the looks of it, but if he had, the differences would be enlightening. compare someone who has died in a plane crash to someone who died of ebola. you'd know that one was not the other very easily even if you couldn't attribute a cause.
anyhow, i looked at that paper you linked. the other big problem with the monsanto chemical -- which was not even mentioned in the lawsuit (!) was that it is an endocrine disruptor even in situations where it is not toxic. in other words, it's terrible for you and will distort your body's hormone chemistry in ways we probably don't even understand yet. that's just as big of a story as the chemical being legally confirmed as carcinogenic.