How to meet the challenge of individualized tumor therapy? This goal is increasingly possible with advances in characterized tumor high-throughput technologies and the continued expansion of molecularly targeted therapy programs. Individualized cancer therapy is based on the knowledge of the fine molecular characteristics of a patient’s tumor and its microenvironment, and this therapy makes it possible to personalize treatment to improve prognosis and reduce toxicity. The goal of individualized therapy is to achieve treatment that targets mutations that promote tumor growth and survival by giving the right combination of drugs to the right person. Due to the increased understanding of the underlying tumor progression mechanisms and drug sensitivities, as well as the increasing utility of targeted therapies and cost-effective integrated multiple trials, the time appears to have come for the final submission of the option of individualized cancer therapy as the standard of care. In any case, there are many challenges to overcome to successfully achieve individualized cancer therapy. In a Perspectives article published in the September issue of Nature Reviews Clinical Oncology, Funda Meric-Bernstam, director of the Institute of Medicine for Individualized Cancer Therapy, and Gordon Mills, professor of medical systems biology in breast oncology and of the immunology unit at the University of Texas Medical Branch Anderson Cancer Center, provide insight into what it will take to achieve individualized cancer treatment, they analyzed those challenges. In their article, Dr. Meric-Bernstam and Mills argue that individualized cancer therapy will not be available as a standard of practice, and therefore will not be fee-based and profitable, until its clinical utility is widely demonstrated. Biological Challenges Tumor Heterogeneity Over the past few years, there has been a growing acknowledgement of the existence of tumor heterogeneity. First, there are significant intra-tumor differences, and cancer cells within tumors may also have functional heterogeneity. In addition, tumors may differ in their ability to metastasize and invade, while spreading cancer cells can survive, remain dormant or metastasize. Molecular evolution and resistance can be altered by the clonal array of specific mutations under the selective pressure of targeted therapy and activation as mutagenic activity of radiotherapy and chemotherapy. Resistance to therapy may have existed previously due to the presence of concurrent mutations, which can arise through an adaptive response induced by targeted therapy, or, alternatively, new mutations can be acquired through activation of signaling pathways required for tumor cell survival. There are two general concepts studied to deal with the emergence of intratumoral heterogeneity and resistance: in-depth characterization of tumorigenesis and recurrence to identify rare and dominant proliferations, and in-depth characterization of tumor epigenetic biology to identify dominant clones. Due to the high cost of obtaining multiple biopsies and their association with potential prevalence, alternative diagnostic tools such as molecular imaging, analysis of circulating tumor cells or circulating free DNA are also being used in ongoing studies. Targeting of drugs To date, most targeted therapies have been able to target gain-of-function mutations in oncogenes, but many proteins are currently “out of reach” and manipulation of loss-of-function mutations in oncogenes in many tumor tissues is currently inaccessible. Technical challenges The identification and validation of marker sensitivity and resistance is a critical step toward individualized cancer therapy. However, molecular analysis of abnormal responses should be an integral part of all individualized cancer therapy programs due to the increased understanding of mechanisms of sensitivity and resistance to therapy. Comprehensive analysis of not only the genome is altered, but also the epigenome, transcriptome, and proteome, and gene-gene, protein-genome, and genome-environment interactions may have important clinical implications for biomarker development. New trial designs are needed Biomarker discovery and validation must be integrated into all aspects of drug development, from discovery to clinical trials. However, current clinical trial designs for biomarker discovery and validation are often not optimal. Biomarker identification and validation, as well as clinical trial design for targeted therapies, require special approaches such as obligatory studies of tissue biopsies, comprehensive tumor and germ cell characterization. The study of tissue biopsies is optional rather than mandatory because tissue biopsies are not always available. In many trials, the amount of scientific information obtained from tissue biopsy studies outweighs the risk of tissue biopsy because of the relatively rare complications associated with tissue biopsy studies (prevalence of approximately 1.4%). Integrated multiple marker analysis Currently, individual biomarker-related trials are often completed before standardized therapies are initiated for treatment or patients are included in clinical trials. This approach helps to reduce overall trial spend, it allows for the use of extensions at the start of treatment, and requires a large amount of valuable sample tissue, however, for patients with tumors that test negative for biomarkers, this approach has the potential to delay the achievement of effective therapies. Retrieval of tissue blocks stored in the archive can take days to weeks, especially if they are stored at different institutions. In therapy implementation, this can delay treatment delivery, making determinations on testing and analysis that may take weeks, whole exome sequencing (WES), whole genome sequencing (WGS), or even extending the testing process beyond the clinically acceptable period for many patients. When a patient is admitted for the first time, especially for patients at high risk of progression or recurrence, one option to avoid biomarker shift delays is to implement a comprehensive and integrated multiplex tumor screening. However, especially for many patients whose mutations are not localized and for only a subset of patients who have the potential for cancer recurrence, paying for early screening has been an issue until investigators have demonstrated that this approach is helpful for caring for patients. Moreover, upfront screening does not take treatment-induced tumor progression or heterogeneous changes into account. Alternatively, for patients with recurrence or progression of cancer, a new tissue biopsy is necessary to save the time required to retrieve the tissue blocks stored in the archive and to buy time for molecular analysis to be performed. However, this approach entails an increase in cost and morbidity due to the need for patients to undergo tissue biopsies, and the quality of the tissue biopsies and whether they are representative is of concern because, even with dedicated radiologists and pathologists, the percentage of meaningful tissue biopsies is not sufficient for biomarker assessment. Three comprehensive multiplex conceptual studies are under review: the first, a “hot spot” analysis of frequent mutations; the second, an evaluation of open reading frames for many identified oncogenes; and the third, WES and WGS, where whole-genome sequencing will be less expensive than storing, processing, and analyzing the data. Prioritization of targets and treatments High-quality bioinformatics requires the use of data from molecular expression profiles to inform decisions about which mutations targeted therapies are effective for patients. And today, only a few dozen mutations are actually confirmed by high-quality trial data. Pharmacological Challenges It is not yet known what the optimal level and duration of targeted inhibition is for formulating the optimal effect. Will short-term, high-level inhibition be more effective or less toxic than long-term inhibition? Such combinations are particularly challenging due to the toxicity of additional agents and the need for appropriate dosing and sequence of administration of both drugs. A major effort in preclinical studies is the need to link systems biology approaches to animal models and clinical trials. Regulatory challenges Clinical laboratory accreditation Biomarkers for clinical decision making (including patient selection or stratification for clinical trials) requires an assured laboratory for identification, but this can significantly increase costs and delays for implementation of biological markers. Since the scope of “actionable” items is rapidly evolving with the emergence of new drugs and marker pharmacophores, determining which targets and analyses should be done in an accredited laboratory requires a prospective plan. Reimbursement In the early stages of genomic testing, it is unclear which tests will be reimbursable, and what level of proof needs to be considered to perform a “standard of care” test, rather than considering the purpose of the study. This is a meaningful activity and debate, and it also requires the use of guidelines and reimbursement for molecular diagnostics. Patient and physician challenges Patients are very willing to participate in the tests required for the study, especially for tests that are only minimally invasive, such as urine, blood, CT and ultrasound imaging, however, patients are less willing to undergo tumor and skin tissue biopsies. Regardless, most patients are willing to undergo at least one tumor biopsy. The new era of genetic diagnosis is also uniquely challenging due to patients’ need for molecular testing and their unrealistic expectations. WES and WGS will present unique challenges regarding the translation of research results, such as the widespread sharing of biospecimens and data from genomics studies