"Molecules were prioritized by the statistical significance of the interaction between the human and viral proteins;"

The BET protein interaction with viral protein E was apparently the highest statistically ranked interaction.  

"...by their status as approved drugs, investigational new drugs (INDs, “clinical” in Table 1a,b), or as preclinical candidates; by their apparent selectivity; and by their availability (for purchase availability notes, see Supplemental Tables 3 and 4)."

Based on Apabetalone being Phase 3 stage, this might have helped. But this still doesn't explain why so many other BET inhibitors were left off the list. They use the term RVX-208 instead of apabetalone. If they would have used apabetalone instead, it would have probably been listed as number 1 simply because drugs in the table seem to be listed alphabetically (after first organizing by statistical significance of the viral protein interaction). As I said before, arbitrary.

"Chemoinformatic searches yielded 15 approved drugs, four investigational new drugs (clinical), and 18 pre-clinical candidates (Table 1a), while specialist knowledge revealed 12 approved drugs, 10 investigational new drugs (clinical), and 10 preclinical candidates (Table 1b)"

That sums up to 69. Likely that those chemoinformatic searches and specialist generated lists were not comprehensive. But at least for the BET inhibitor list, apabetalone is by far the most clinically advanced and it makes sense to pursue it.

BDAZ