Colorectal carcinoma (CRC) is one of the most common cancer diseases in the Western countries moreover in Europe it is the second most common cause of cancer-related death. Since PI3K/Akt signaling pathway is activated in 60-70 % of human colon cancers, its inhibitors have been suggested as potential therapeutic agents. One of the most promising drugs, perifosine, is a synthetic alkylphospholipid that targets signal transduction pathways primarily at the cell membrane. In in vitro studies, perifosine has shown excellent efficacy to cancer cells in inducing apoptosis and inhibiting proliferation. Nevertheless, the data from clinical studies are mostly disappointing, showing no significant benefit in overall survival of patients. Therefore, in our work we reveal the reasons for perifosine failure in clinical studies, focusing on the specific features of tumor microenvironment and the problem of poor drug penetration into tumor tissue. Increased protein expression of extracellular matrix, increased rigidity, gradient of oxygen and nutrients and especially severe acidosis and hypoxia are characteristic for tumor environment. Acidosis and hypoxia can seriously influence drug efficacy in in vivo conditions. Therefore, our tests were performed on monolayers of cancer cells exposed to lactic acidosis, hypoxia or combination of them and on 3D culture models (spheroids), which are intended to faithfully reproduce the real tumor microenvironment. To accurate methodology for drug cytotoxicity tests in 3D models, we have suggested workflows for the colocalization of the variable drug level areas with the regions of live/dead or apoptotic cells inside the spheroid. Penetration and localization of drugs in 3D models of tumor tissue is done by Matrix-Assisted Laser Desorption/Ionization Mass Spectrometric Imaging (MALDI-MSI) and the cellular viability is detected by either immunohistochemistry or viability probes with subsequent Laser Scanning Confocal Microscopy (LSCM). Generation of section mask and equidistant homocentric curves, counting the MALDI MSI and LSCM signal along them and assessment of correlation coefficients allows final coregistration. In our experiments, we show that acidosis and acidosis combined with hypoxia diminish perifosine efficacy to monolayers as well as spheroids models of cancer tissue. In addition, in monolayers exposed to acidosis and acidosis combined with hypoxia, perifosine intracellular level is lower than in normal pH. In 3D models, perifosine does not penetrate deeper than 100-150 µm from the periphery of the spheroid, which diameter is 1 000 µm. These facts might contribute to resistance to this drug. We also proved that in the areas inside the spheroid, into which the perifosine does not reach, the cells are still viable. As an improvement of perifosine efficiency we have proposed two strategies, alkalinization of tumor microenvironment and combination therapy with inhibitor of the anti-apoptotic proteins of the BCL-2 family, ABT 737.