Drug-resistant metastatic disease accounts for almost all cancer deaths. The ability of metastatic cancer to develop resistance to almost all chemotherapeutic agents has led many in the oncology community to believe that, for most patients, metastatic cancer is not a curable disease. These sentiments were once echoed in the HIV/AIDS (human immunodeficiency virus infection/acquired immune deficiency) community, where progressive resistance and a lack of available drugs meant that patients could expect to succumb to the virus in just a few short years. This situation has changed dramatically in the past two decades, and HIV patients with access to adequate health care systems can, in many cases, expect to live a normal lifespan. The remarkable success of HIV treatment provides a roadmap for the treatment of multidrug-resistant metastatic cancer. In addition to new agents, the most important element of this roadmap is the development of methods to easily and frequently monitor disease resistance and progression, analogous to viral load and CD4 counts for HIV. Current methods for guiding treatment decisions suffer from limitations related to a lack of ability to account for the in vivo tumor setting and the impracticality of performing repeat biopsies. Even modalities such as MRI (magnetic resonance imaging) and PET-CT (positron emission tomography - computed tomography) suffer from a limited ability to monitor treatment resistance due to the relative insensitivity of anatomic (size) monitoring and the limited ability to perform longitudinal studies. We have recently demonstrated that non-invasive optical monitoring, using Diffuse Optical Spectroscopy (DOS), can not only provide important metabolic signatures of treatment efficacy in patients treated with chemotherapy, but that these signatures are present incredibly early during therapy and across a range of treatment regimens. In this proposal, we will build on this work to develop non-invasive optical metabolic imaging for monitoring disease progression, with a special emphasis on detecting drug resistance. We are rapidly advancing DOS technology towards wearable probes, allowing a level of patient access unparalleled with current modalities, while simultaneously providing unprecedented metabolic, hemodynamic, and morphologic data on the in vivo tumor state. When this wearable technology is paired with other emerging technology such as wireless communications, pattern recognition algorithms, and existing health records, a plethora of new possibilities for better understanding the dynamics of treatment response will be available. Towards this goal, we will first conduct a preclinical screen of chemotherapeutic agents in an array of tumor models. This will allow us to better understand the variation and timescales of optical signatures present during both successful treatment and drug resistance. We will then use these signatures to attempt, for the first time, to use non-invasive optical feedback to schedule combined cytotoxic and antiangiogenic therapies. Finally, we will conduct a clinical feasibility study to test the presence of optical drug resistance signatures in metastatic breast cancer patients. If successful, our proposal will provide physicians a new view of the tumor state in real time, and it has the potential to transform the treatment of metastatic disease.