The tell-tale word in the abstract is that the biomarkers "link" to cancer. The data do not provide direct evidence of a causal process. The trouble is that it is very difficult to measure chemical exposure over a period of 50+ years, and unethical to experiment on people. The upshot is that studies like this are not particularly informative on their own. If you have different kinds of data all pointing in the same direction, then maybe you have a finding.