The Wikipedia lists Frank's sign as
Frank's sign is a diagonal crease in the ear lobe extending from the tragus across the lobule to the rear edge of the auricle. The sign is named after Sanders T. Frank. […] It has been hypothesised that Frank's sign is indicative of cardiovascular disease and/or diabetes. Some studies have described Frank's sign as a marker of cardiovascular disease but not linked to the severity of the condition.
But going after the references provided there leads into inconclusive areas.
Frank's sign and coronary disease for example concludes just statistically
Frank's sign is therefore considered as a marker of the coronary disease, independent of risk factors but frequently associated with them. If its absence does not permit in any way to exclude the diagnosis of coronary disease, its presence corresponds in three quarters of the cases to an established coronary disease within a symptomatic population
Similarly unsatisfactory are the results dug up from reference material (Stanford Medicine: Frank's Sign - Diagonal earlobe crease (DELC)) as well as trying to follow this nice agglomeration of thematic papers (Joe Cannon: "Earlobe Crease And Heart Disease: Is It Real? Review Of Evidence").
While I would be more interested in the most likely explanation for its existence and relevance: Is it therefore even correct to consider this phenomenon as a purely correlative marker sign or is it as a whole so spurious as to be better left ignored?