Monitoring for Response to Antineoplastic Drugs: The Potential of a Metabolomic Approach

Abstract

For most cancers, chemotherapeutic options are rapidly expanding, providing the oncologist with substantial choices. Therefore, there is a growing need to select the best systemic therapy, for any individual, that effectively halts tumor progression with minimal toxicity. Having the capability to predict benefit and to anticipate toxicity would be ideal, but remains elusive at this time. An alternative approach is an adaptive approach that involves close observation for treatment response and emergence of resistance. Currently, response to systemic therapy is estimated using radiographic tests. Unfortunately, radiographic estimates of response are imperfect and radiographic signs of response can be delayed. This is particularly problematic for targeted agents, as tumor shrinkage is often not apparent with these drugs. As a result, patients are exposed to prolonged courses of toxic drugs that may ultimately be found to be ineffective. A biomarker-based adaptive strategy that involves the serial analysis of the metabolome is attractive. The metabolome changes rapidly with changes in physiology. Changes in the circulating metabolome associated with various antineoplastic agents have been described, but further work will be required to understand what changes signify clinical benefit. We present an investigative approach for the discovery and validation of metabolomic response biomarkers, which consists of serial analysis of the metabolome and linkage of changes in the metabolome to measurable therapeutic benefit. Potential pitfalls in the development of metabolomic biomarkers of response and loss of response are reviewed.

Discovery and Predictive Modeling of Urine Microbiome, Metabolite and Cytokine Biomarkers in Hospitalized Patients with Community Acquired Pneumonia

“…Since the microbiome and metabolite profiles are influenced by the environment, nutrition, age, and lifestyle of the host, in addition to genetics, these concatenated profiles provide a unique snapshot of individual health. 19,22,31,[23][24][25][26][27][28][29][30] Indeed, while these complex profiles can be examined independently, changes in the collective abundance patterns of metabolites and microbes may indicate deeper homeostatic disturbances, which may be reflected through changes in interleukin signaling. The membership of the bodies microbial communities have dynamic interconnected relationships with one another and the host that change under states of disease and stress.…”

“…Patients were selected from the University of Louisville Pneumonia Study (ULPS) biorepository with IRB approval. The ULPS was a population-based cohort study of 7,449 unique patients hospitalized with CAP between June 1, 2014 and May 31,2016. From this biorepository, we selected 30 urine samples from patients, ten each with a confirmed etiology of influenza A virus (IAV), S. pneumoniae, or S. aureus infection.…”

“…The metabolome and microbiome 21 are also unique portraits of the individual patient as they are not only influenced by genetics and disease state but also by the environment, nutrition, age, and lifestyle 19,22,31,[23][24][25][26][27][28][29][30] . Thus the microbiome and metabolome are very different from transcriptomic or proteomic biomarkers.…”

“…Thus the microbiome and metabolome are very different from transcriptomic or proteomic biomarkers. Unlike biomarkers that individually vary with health status, meta-biomarkers by definition are so co-related and interwoven that they produce a precise disease signature that evolves with the pathophysiological state of the individual [31][32][33][34] .…”

“…Thus meta-biomarkers are less likely to produce false identifications as they do not rely on changes to a single analyte 31,34 . Temporal and dynamic changes to the microbiome and metabolome along with their connection to phenotype and meta-biomarker characteristics leads us to select these quantifiable components in the urine as indicators of CAP.…”