Patient-derived organoids recently emerged as promising ex vivo 3D culture models recapitulating histological and molecular characteristics of original tissues, thus proteomic profiling of organoids could be valuable for function investigation and clinical translation. However, organoids are usually cultured in murine Matrigel (served as scaffolds and matrix), which brings an issue to separate organoids from Matrigel. Because of the complex compositions of Matrigel and thousands of identical peptides shared between Matrigel and organoids, insufficiently dissolved Matrigel could influence proteomic analysis of organoids in multiple ways. Thus, how to dissolve Matrigel matrix and recovery organoid cells efficiently is vital for sample preparation. Here, we comprehensively compared three popular Matrigel dissolving methods (cell recovery solution, dispase, and PBS–EDTA buffer) and investigated the effect of undissolved Matrigel proteins on proteomic profiles of organoids. By integrative analysis of label-free proteomes of Matrigel and stable isotope labeling by amino acids in cell culture proteomes of organoids collected by three methods, respectively, we found that dispase showed an optimal efficiency, with the highest peptide yield and the highest incorporation ratio of stable isotope labeling by amino acids in cell culture labels (97.1%), as well as with the least potential Matrigel contaminants. To help analysis of proteomic profiles of organoids collected by the other two methods, we identified 312 high-confidence Matrigel contaminants, which could be filtered out to attenuate Matrigel interference with minimal loss of biological information. Together, our study identifies bioinformatics and experimental approaches to eliminate interference of Matrigel contaminants efficiently, which will be valuable for basic and translational proteomic research using organoid models.