If you are going to try to cite "facts" include the source material.
https://onlinelibrary.wiley.com/doi/abs/10.1111/1468-0009.12403States that the FDA went against the panel vote 22% of the time between the years 2008 and 2015.
However of that 22% the FDA rejected a panel approved drug 75% of time and approved a panel rejected drug 25% of the time.
So the headline figure is closer to 16% of the time the FDA rejected a previously positive panel vote.
But of course it's more nuanced than that, 50% of the overturns occurred in the first 2 years of that period and only 10% in the last 2 years of that 8 year period. So over the course of 8 years the FDA became 5 times less likely to overturn a decision. So they went from overturning around 40% in 2008-9 to 8% in 2014-15.
And of course it's even more nuanced than that, as the article explains "Discordance was associated with lower degrees of advisory committee consensus and was more likely for agency actions focused on medical product safety than for novel approvals"
In other words, you had a far higher chance of being overturned the panel was very divided on the matter and the vote was close. 9-1 isn't close and having heard the panel discussion, there was really only one strong dissenter and he was unconcerned about safety or efficacy, he was wanting to isolate the specific "magic ingredient" in cells produced by MSB to ensure that the results would be reproducible across all manufactured doses and therefore wanted more trials with different markers being isolated - he seemed to be ignoring the ethics of subjecting some children to a painful death. This leads to the other nuance in the stats, votes tend to be overturned on safety reasons and there was a strong consensus of there being no safety issues and this is further diluted by the fact that "safety" seems less of an issue when the patient will be dead in 20-100 days.
So to recap:
By 2015 8% of panel decisions were being overturned and of those only 6% went from Panel Yes to FDA No.
Of those 6% the FDA was disproportionately inclined to reject when the Panel decision was close rather than a clear majority and when they had definite safety concerns, neither of which apply to Ryoncil.
(20min delay)