Drug resistance in cancer is a complex issue that involves multiple factors, including tumor burden, growth kinetics, tumor heterogeneity, physical barriers, the immune system, and undruggable cancer drivers. The authors propose four general solutions to drug resistance: earlier detection of tumors, adaptive monitoring during therapy, the addition of novel drugs and improved pharmacological principles, and the identification of cancer cell dependencies through high-throughput synthetic lethality screens, integration of clinico-genomic data, and computational modeling. These approaches can be synthesized for each tumor at any decision point to inform therapy choices. Tumor burden and growth kinetics are critical determinants of resistance. The Goldie–Coldman hypothesis suggests that the probability of drug-resistant clones in a tumor depends on mutation rate and tumor size. The Norton–Simon hypothesis explains cancer growth and regression after therapy, emphasizing the importance of dose-dense chemotherapy to prevent rapid regrowth. Tumor heterogeneity is a major cause of drug resistance, as cancer cells acquire genomic alterations through various mutational processes. This leads to parallel and convergent evolution, as well as spatial segregation of clones in primary and metastatic sites. Tumor heterogeneity is measured through genomic sequencing, but this approach has limitations in capturing true heterogeneity. Physical barriers, such as the blood-brain barrier, can prevent drugs from reaching certain areas of the body, leading to resistance. Anti-angiogenic agents can normalize vascular structure and function, facilitating drug delivery. However, resistance to these agents can still occur. The immune system and tumor microenvironment play a significant role in resistance. Immunosuppressive microenvironments, or 'immune deserts,' can hinder the effectiveness of checkpoint inhibitors. Techniques to turn 'cold' tumors into 'hot' ones by recruiting immune effectors are being explored. Undruggable genomic drivers, such as MYC, RAS, and TP53, remain challenging targets. New approaches, including miniproteins, allele-specific inhibitors, and small molecules that covalently bind to p53, are being developed to address these targets. Selective therapeutic pressure can lead to adaptive responses or acquired resistance. Adaptive responses can occur rapidly and may be responsible for short-lived clinical responses. Acquired resistance involves the emergence of new activating mutations, pathway alterations, or histological changes. Overcoming resistance requires strategies such as earlier detection, deeper responses, therapeutic monitoring, and exploitation of cancer dependencies. ctDNA testing offers a non-invasive method for detecting cancer and monitoring clonal evolution. It can also help identify high-risk groups for additional therapeutic intervention. Monitoring response and adaptive interventions are crucial for managing resistance. Functional imaging and ctDNA-based methods can provide real-time data on tumor heterogeneity and response to therapy. These approaches can help identify patients who may relapse and inform the next therapy at an earlier time point. Resistance to immunotherapy is different from resistance to targeted therapy. Mechanisms of resistance to immunotherapy include mutations in B2M, JAK1Drug resistance in cancer is a complex issue that involves multiple factors, including tumor burden, growth kinetics, tumor heterogeneity, physical barriers, the immune system, and undruggable cancer drivers. The authors propose four general solutions to drug resistance: earlier detection of tumors, adaptive monitoring during therapy, the addition of novel drugs and improved pharmacological principles, and the identification of cancer cell dependencies through high-throughput synthetic lethality screens, integration of clinico-genomic data, and computational modeling. These approaches can be synthesized for each tumor at any decision point to inform therapy choices. Tumor burden and growth kinetics are critical determinants of resistance. The Goldie–Coldman hypothesis suggests that the probability of drug-resistant clones in a tumor depends on mutation rate and tumor size. The Norton–Simon hypothesis explains cancer growth and regression after therapy, emphasizing the importance of dose-dense chemotherapy to prevent rapid regrowth. Tumor heterogeneity is a major cause of drug resistance, as cancer cells acquire genomic alterations through various mutational processes. This leads to parallel and convergent evolution, as well as spatial segregation of clones in primary and metastatic sites. Tumor heterogeneity is measured through genomic sequencing, but this approach has limitations in capturing true heterogeneity. Physical barriers, such as the blood-brain barrier, can prevent drugs from reaching certain areas of the body, leading to resistance. Anti-angiogenic agents can normalize vascular structure and function, facilitating drug delivery. However, resistance to these agents can still occur. The immune system and tumor microenvironment play a significant role in resistance. Immunosuppressive microenvironments, or 'immune deserts,' can hinder the effectiveness of checkpoint inhibitors. Techniques to turn 'cold' tumors into 'hot' ones by recruiting immune effectors are being explored. Undruggable genomic drivers, such as MYC, RAS, and TP53, remain challenging targets. New approaches, including miniproteins, allele-specific inhibitors, and small molecules that covalently bind to p53, are being developed to address these targets. Selective therapeutic pressure can lead to adaptive responses or acquired resistance. Adaptive responses can occur rapidly and may be responsible for short-lived clinical responses. Acquired resistance involves the emergence of new activating mutations, pathway alterations, or histological changes. Overcoming resistance requires strategies such as earlier detection, deeper responses, therapeutic monitoring, and exploitation of cancer dependencies. ctDNA testing offers a non-invasive method for detecting cancer and monitoring clonal evolution. It can also help identify high-risk groups for additional therapeutic intervention. Monitoring response and adaptive interventions are crucial for managing resistance. Functional imaging and ctDNA-based methods can provide real-time data on tumor heterogeneity and response to therapy. These approaches can help identify patients who may relapse and inform the next therapy at an earlier time point. Resistance to immunotherapy is different from resistance to targeted therapy. Mechanisms of resistance to immunotherapy include mutations in B2M, JAK1