Early Diagnosis of Brain Metastases Using a Biofluids-Metabolomics Approach in Mice
James R Larkin, Alex M Dickens, Timothy DW Claridge, Claire Bristow, Kleopatra Andreou, Daniel C Anthony, Nicola R Sibson (2016)
Theranostics 6 12 2161-2169. DOI: 10.7150/thno.16538.
Over 20% of cancer patients will develop brain metastases. Prognosis is currently extremely poor, largely owing to late-stage diagnosis. We hypothesized that biofluid metabolomics could detect tumours at the micrometastatic stage, prior to the current clinical gold-standard of blood-brain barrier breakdown.
Metastatic mammary carcinoma cells (4T1-GFP) were injected into BALB/c mice via intracerebral, intracardiac or intravenous routes to induce differing cerebral and systemic tumour burdens. B16F10 melanoma and MDA231BR-GFP human breast carcinoma cells were used for additional modelling. Urine metabolite composition was analysed by 1H NMR spectroscopy. Statistical pattern recognition and modelling was applied to identify differences or commonalities indicative of brain metastasis burden.
Significant metabolic profile separations were found between control cohorts and animals with tumour burdens at all time-points for the intracerebral 4T1-GFP time-course. Models became stronger, with higher sensitivity and specificity, as the time-course progressed indicating a more severe tumour burden. Sensitivity and specificity for predicting a blinded testing set were 0.89 and 0.82, respectively, at day 5, both rising to 1.00 at day 35. Significant separations were also found between control and all 4T1-GFP injected mice irrespective of route. Likewise, significant separations were observed in B16F10 and MDA231BR-GFP cell line models. Metabolites underpinning each separation were identified.
These findings demonstrate that brain metastases can be diagnosed in an animal model based on urinary metabolomics from micrometastatic stages. Furthermore, it is possible to separate differing systemic and CNS tumour burdens, suggesting a metabolite fingerprint specific to brain metastasis. This method has strong potential for clinical translation.