Comment on "Nonuniversality of Transverse Momentum Dependent Parton Distributions at Small x"

We point out that the analysis in arXiv:1003.0482 actually verifies the universality of transverse momentum dependent quark distributions at small $x$, which supports the observation in our earlier work arXiv:0904.4150. Once the gluon exchanges responsible for the additional infrared divergences are factorized and summed into an exponential factor, it can be expressed as a matrix element of Wilson line operators, and treated as a nonperturbative input to the $k_T$ factorization of $pA$ collisions. The same exponential factor has been extracted from $pp$ collisions in arXiv:0904.4150, rendering possible experimental constraints on its behavior from $pp$ processes, and predictions for $pA$ processes.