Motivation for the Huisken monotonicity formula

Question

In mean curvature flow one evolves an immersed manifold $F_0: M^n \rightarrow \mathbb{R}^{n+1}$ along it's mean curvature, i.e. one searches for a solution $F:M \times [0,T) \rightarrow \mathbb{R}^{n+1}$ of the following problem $$ \frac{\partial F}{\partial t} = H $$ with $F(\cdot,0) = F_0$.

Huisken proved the following famous formula in the study of mean curvature flow $$ \frac{d}{dt} \int_{M_t} \rho(x,t) d \mu_t = -\int_{M_t} \rho(x,t) \left| H+\frac{1}{2 \tau} F^{\perp} \right|^2 d \mu_t, $$ where $\rho(x,t) = \frac{1}{4 \pi(t_0-t))^{n/2}} \exp \left( - \frac{|x|^2}{4(t_0-t)}\right)$.

I would like to know how Huisken came up with this formula. Where there any previous results for which it was based on?