shades of grey“Furthermore, it is important to emphasize that correlation between a biomarker and clinical endpoint is necessary but not sufficient to support that a drug effect on the biomarker will reliably predict a drug effect on the clinical endpoint,” the documents continued. “Notably, even the presence of strong correlation between the biomarker and the clinical endpoint within both the placebo and drug arms of a clinical trial may not be sufficient.”
....but money talks
https://endpts.com/fda-indicates-willingness-to-approve-biogen-als-drug-despite-failed-phiii-study/