In Peskin and Schroeder exc. 4.3 we are asked to compute the differential cross section of $\Phi^1\Phi^2\rightarrow \Phi^1\Phi^2$ process for the linear sigma model: Compute, to leading order in $\lambda$, the differential cross section, in the center of mass frame, for the scattering process $\Phi^1\Phi^2 \rightarrow \Phi^1\Phi^2$.
Since all the particles involved have identical masses and we have two incoming and two outgoing particles, we can use (4.85): $$(d\sigma/d\Omega)_{CM} = \frac{|M|^2}{64\pi^2E_{CM}}.\tag{4.85}$$
Up to leading order in $\lambda$ we find $M=-2i\lambda$, since the Feynman rule for the vertex with particle types $i,j,k,l$ is $$-2i\lambda(\delta^{ij}\delta^{kl}+\delta^{il}\delta^{jk}+\delta^{ik}\delta^{jl}).$$ This would give a differential cross section of $$(d\sigma/d\Omega)_{CM} =\frac{\lambda^2}{16\pi^2E_{CM}}.$$ This is the answer that can be found in several sources (e.g. Zhong-Zhi Xianyu's solutions).
However, this only seems to take into account the interactions. As 4.85 is derived while ignoring the "trivial case" of forward scattering where no interaction takes place, I suspect a contribution due to forward scattering with no interaction is missing here. What is this contribution?