Lepton flavor non-universality in $b \to s \ell^+ \ell^-$ processes

We explore a scenario of New Physics entering the description of $B\to K^{(\ast )} \mu \mu$ decay through couplings to the operators ${\cal O}_{9,10}^\prime$, satisfying $C_9^\prime = - C_{10}^\prime$. From the current data on ${\cal B}(B_s\to \mu\mu)$ and ${\cal B}(B\to K \mu\mu)_{[15,22] \,\mathrm{GeV}^2}$, we obtain constraints on $\mathrm{Re} C_{10}^\prime$ and $\mathrm{Im} C_{10}^\prime$ which we then assume to be lepton specific, and find $R_K= {\cal B}(B\to K \mu\mu)/ {\cal B}(B\to K ee)_{[1,6]\,\mathrm{GeV}^2}=0.88(8)$, consistent with recent value measured at LHCb. A specific realization of this scenario is the one with a scalar leptoquark state $\Delta$, in which $C_{10}^\prime$ is related to the mass of $\Delta$ and its Yukawa couplings. We then show that this scenario does not make any significant impact on $B_s-\overline B_s$ mixing amplitude nor to ${\cal B}(B\to K\nu\bar \nu)$. Instead, it can modify $R_{K^\ast}= {\cal B}(B\to K^\ast \mu\mu)/ {\cal B}(B\to K^\ast ee)_{[1,6]\,\mathrm{GeV}^2}$, which will soon be experimentally measured and we find it to be $R_{K^\ast}= 1.11(8)$, while $R_{K^\ast}/R_K= 1.27(19)$. A similar ratio of forward-backward asymmetries also becomes lower than in the Standard Model.