The very last report in the FR24 data chain lacks any altitude data. For it to have been processed, a recognisable message must have been received by the servers, otherwise the message before would be the last data we would see. What strikes me as noteworthy about this is that the most obvious sequence of events I can think of where this data output would be recorded, would occur when the transponder was deliberately cycled from Mode C (S) through Mode A (no alt) on the way to STBY (Off).
As I understand, FR24 saves one report per minute, even if reports come in more often than that.
Assuming that your premise is correct (this is the first time I hear about this), what are the odds that the last report saved by FR24 just happens to be the one when the transponder was passing through mode A?