Detection of oesophageal cancer biomarkers by plasma proteomic profiling of human cell line xenografts in response to chemotherapy

Abstract

Background: The incidence of oesophageal adenocarcinoma is increasing worldwide but survival remains poor. Neoadjuvant chemotherapy may improve survival, but targeting treatment to patients who respond to chemotherapy could be improved by the availability of markers of response. This study sought proteomic markers of therapeutic response using an adenocarcinoma xenograft model.

Methods: Epirubicin, cisplatin or 5-fluorouracil was administered to severe combined immune-deficient mice bearing OE19 oesophageal adenocarcinoma xenografts. Murine plasma samples from treated and untreated xenografts were analysed by surface-enhanced laser desorption/ionisation time-of-flight mass spectroscopy, and panels of peaks were found using class prediction models that distinguished treatment groups. Proteins in these peaks were identified by mass spectroscopy in tryptic digests of purified fractions. Five paired samples from oesophageal cancer patients before and after chemotherapy were analysed using the same methodology.

Results: Plasma protein peaks were identified that differed significantly (P<0.05, ANOVA) between the treated xenograft and control groups. Marker panels predicted treated vs untreated xenografts with sensitivities of 100%, specificities of 86-100% and test efficiencies of 89-100%. Three of the proteins identified in these panels, apolipoprotein A-I, serum amyloid A and transthyretin were confirmed in the clinical samples.

Conclusion: Plasma protein markers can be detected in response to chemotherapy in oesophageal adenocarcinoma xenografts and in clinical samples, and have the potential to monitor response and guide chemotherapy in oesophageal adenocarcinoma.

Figure 1

Tumour volume data obtained during treatment of OE19 xenograft bearing mice with each drug (epirubicin, n=20, cisplatin, n=16 and 5-fluorouracil, n=18 or control (blank, n=23, 20 and 12 respectively). Mean tumour volumes normalised to day zero (first day of drug injection) and 95% confidence intervals are shown and statistically significant differences (P<0.05, two-tailed t-test or Mann–Whitney test as appropriate) are highlighted (^*).

Figure 2

Principal component analysis of plasma profiles obtained in the mouse xenograft experiments. Each point represents a single sample, colour coded as indicated according to untreated xenograft (epirubicin, cisplatin or 5-fluorouracil) treated xenograft, untreated non-xenograft or treated non-xenograft mouse.