Antibiotic resistance in bacterial infections is a growing global health concern, with β-lactamases playing a pivotal role. Understanding the origin and spread of β-lactamases can provide insights for developing novel drugs and therapies. However, complexity arises by the fact that they can affect their own selective advantage through antibiotic degradation, which also shields nearby susceptible bacteria. To address this eco-evolutionary feedback, we investigated the interplay between feedbacks and the spread of bacterial strains expressing different β-lactamase enzymes conferring varying levels of resistance to the cephalosporin cefotaxime (CTX). Specifically, we conducted head-to-head competitions between two related clinically relevant β-lactamase mutants, TEM-19 and TEM-52, and varied CTX concentration and cell density to explore the impact of these factors on competitive fitness and coexistence. Results revealed that at certain antibiotic treatment regimes, culture density, which modulated the eco-evolutionary feedback strength, determined whether resistance alleles could invade the population and replace or coexist with susceptible alleles. Furthermore, using a population dynamic model based solely on measured parameters, we investigated the underlying mechanisms of the observed dynamics. By validating our model with experimental data, we show that it can largely predict the equilibrium frequencies based on dose-dependent growth rates and antibiotic degradation activities of the individual strains. Our study emphasizes the importance of considering ecological feedbacks in the evolution of antibiotic resistance, even at the level of individual proteins, and highlights the potential of a predictive model applicable to clinically relevant conditions.