i test your tle on my observation around that date , and the result is better than before remove unnecessary dot
before:
1 27607U 02058C 26115.38765514 .00001017 00000-0 13739-3 0 9991
2 27607 64.5516 290.8647 0075584 272.1087 266.7100 14.82926048256607
after:
1 27607U 02058C 26115.38765722 .00000000 00000-0 13739-3 0 01
2 27607 64.5516 291.6288 0075584 272.1087 268.8242 14.82926048 09
so as @jazzpi post , the mean anomaly value is still ok, and not always close with the reference. mean anomaly between 0-360 degree is valid.