I measured the Y arm finesse in the different alignments: one is the alignment recorded during PRFPMI lock yesterday and the other is the one that the beam on both ITM and ETM is roughtly centered by using camera.
Figure 1 shows the decay and fiting curve when the mirror alignment was former.
Figure 2 shows the decay and fiting curve when the mirror alignment was latter.

Though there are strange values in both measurement, the finesse value with the latter alignment seems slightly better.
During the measurement, I checked the beam spot on ITMY and it seems shifted several cm horizontally, so this finesse different might come from the clipping of the beam by recoil mass but I'm not so sure.

Anyway, the finesse value measured with the latter alignment is over 1300, finesse degradation seems not happened.

Note:

Somehow, fitting sometimes failed and finese value becomes negative (and absolute value is very large).
So, it might be better to set some limit so that finesse becomes positive.