Difference between SimpleITK.Euler3DTransform and scipy.spatial.transform.Rotation.from_euler?

Using these two library functions:

• SimpleITK.Euler3DTransform
• scipy.spatial.transform.Rotation.from_euler

to create a simple rotation matrix from Euler Angles:

import numpy as np
import SimpleITK as sitk
from scipy.spatial.transform import Rotation
from math import pi

euler_angles = [pi / 10, pi / 18, pi / 36]

sitk_matrix = sitk.Euler3DTransform((0, 0, 0), *euler_angles).GetMatrix()
sitk_matrix = np.array(sitk_matrix).reshape((3,3))
print(np.array_str(sitk_matrix, precision=3, suppress_small=True))

order = 'XYZ' # Different results for any order in ['XYZ','XZY','YZX','YXZ','ZXY','ZYX','xyz','xzy','yzx','yxz','zxy','zyx']
scipy_matrix = Rotation.from_euler(order, euler_angles).as_matrix()
print(np.array_str(scipy_matrix, precision=3, suppress_small=True))

I get two different results:

[[ 0.976 -0.083  0.2  ]
 [ 0.139  0.947 -0.288]
 [-0.165  0.309  0.937]]
[[ 0.981 -0.086  0.174]
 [ 0.136  0.943 -0.304]
 [-0.138  0.322  0.937]]

Why? How can I compute the same matrix as SimpleITK using scipy?