Identification of biomarkers for radiation-induced acute intestinal symptoms (RIAISs) in cervical cancer patients by serum protein profiling

Ying Chai, Jie Wang, Yuan Gao, T. Wang, Fan Shi, Jin Su, Yan Yang, Xia Zhou, Liping Song, Zhihong LiuJournal of Radiation Research 2014 · DOI

TL;DR

This study identifies potential protein biomarkers in the serum of cervical cancer patients that can predict the likelihood of developing acute intestinal symptoms (RIAISs) after pelvic radiotherapy. Using SELDI-TOF MS and a decision tree model, they achieved an 81% accuracy in distinguishing patients at risk, paving the way for personalized treatment and improved patient care.

What Problem Does This Paper Address?

Pelvic radiotherapy is a cornerstone treatment for cervical cancer, but it frequently leads to acute intestinal symptoms (RIAISs). These symptoms, including abdominal pain and diarrhea, can be severe, limit treatment efficacy, and significantly impact a patient's quality of life. Currently, predicting which patients will develop RIAISs is challenging, leading to a one-size-fits-all approach to managing this side effect. The authors aim to address this by identifying reliable serum biomarkers that can predict RIAISs before radiotherapy begins, enabling proactive interventions and personalized treatment strategies.

Methods

The core of this study relies on Surface-Enhanced Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (SELDI-TOF MS). This technique allows for the high-throughput analysis of proteins and peptides in biological samples like serum. Here's a breakdown of their approach:

  • Sample Collection: Serum samples were collected from cervical cancer patients before they underwent pelvic radiotherapy.
  • SELDI-TOF MS Analysis: The collected serum samples were processed and analyzed using SELDI-TOF MS. This technology uses specialized surfaces (ProteinChip arrays) to capture proteins based on their properties, followed by mass spectrometry to determine their mass-to-charge ratio (m/z) and intensity. This generates a "protein profile" for each sample.
  • Bioinformatics and Statistical Analysis: Sophisticated bioinformatics tools were used to compare the protein profiles between patients who developed RIAISs and those who did not. Statistical tests (like the Student's t-test) were employed to identify protein peaks that showed significant differences in abundance between the two groups.
  • Decision Tree Model: To create a predictive model, a decision tree algorithm (specifically, the J48 algorithm, a Java implementation of C4.5) was used. This algorithm learns rules from the data to classify samples. In this case, it used the intensities of specific protein peaks to predict whether a patient was likely to develop RIAISs.
  • Biomarker Identification: Identified protein peaks with significant differences were further analyzed to determine their identity. This involved comparing their m/z values to known protein databases (UniProtKB/Swiss-Prot) to find matching theoretical m/z values and thus identify the proteins.
  • ELISA Validation: For a subset of the identified candidate biomarkers, Enzyme-Linked Immunosorbent Assay (ELISA) was used to validate their serum levels in a separate group of patients.

Datasets

  • Primary Dataset: Serum samples from 66 cervical cancer patients were collected prior to pelvic radiotherapy. Of these, 11 patients developed RIAISs and were matched with 11 control patients who did not develop RIAISs. This dataset was used for SELDI-TOF MS analysis and decision tree construction.
  • Validation Dataset: Serum samples from an additional 23 RIAISs patients and 10 matched control patients were used for ELISA validation of specific candidate biomarkers.

Key Insights and Ideas

  • Proteomics for Predictive Biomarkers: The study demonstrates the power of SELDI-TOF MS in identifying subtle differences in serum protein profiles that can predict a clinical outcome (RIAISs) before it occurs. This moves beyond reactive treatment to proactive management.
  • Multivariate Prediction is Key: A single biomarker might not be sufficient. The development of a decision tree model using multiple protein peaks highlights that a combination of factors is more effective for accurate prediction.
  • Specific Proteins Implicated: The identification of specific proteins like osteopontin (OPN), neurosecretory protein VGF, transthyretin (TTR), hepcidin, and C1-INH as potential biomarkers is a significant step. Their altered levels suggest underlying biological mechanisms related to intestinal health and radiation response.
  • Potential Mechanisms: The downregulation of OPN and TTR, and upregulation of VGF and hepcidin, hint at complex interactions involving immune response, nutrient transport, and inflammation in the context of radiation injury.

Key Results

  • 22 Significantly Different Peaks: Out of 253 detected peaks, 22 showed statistically significant differences (P < 0.05) between the RIAISs group and the control group.
  • Six Identified Biomarkers: Six protein peaks were successfully identified with m/z ratios of 7514.9, 4603.94, 6887.41, 2769.21, 3839.72, and 4215.7. These correspond to OPN fragment, neurosecretory protein VGF fragment (two peaks), TTR, hepcidin, and C1-INH fragment.
  • Decision Tree Accuracy: A decision tree model constructed using three biomarkers (m/z 1270.88, 1503.23, and 7514.90) achieved an accuracy of 81% in separating RIAIS-affected patients from controls after 4-fold cross-validation.
  • ELISA Validation Findings: ELISA confirmed significantly lower levels of TTR and OPN, and significantly higher levels of neurosecretory protein VGF in RIAISs patients compared to controls. Hepcidin and C1-INH levels did not show significant differences in this validation step.

Limitations and Future Work

  • Unknown Peaks: The decision tree model included two peaks (m/z 1270.88 and 1503.23) that remain unidentified. Further research is needed to elucidate their nature, structure, and function.
  • Sample Size: While the study identified promising biomarkers, the number of patients in the RIAISs group (n=11) is relatively small. Larger, independent validation cohorts are crucial to confirm these findings.
  • ELISA Discrepancies: The ELISA validation did not show significant differences for hepcidin and C1-INH, despite their identification as potential biomarkers by SELDI-TOF MS. This highlights the need for careful interpretation and further validation of these candidates.
  • Future Directions: The authors suggest that further investigations are required to explore the precise roles of these identified proteins in intestinal homeostasis and radiation enteropathy. They also emphasize the potential for individualized treatment programs, advanced radiotherapy techniques, and prophylactic measures for patients identified as being at high risk for RIAISs. The identified proteins could also serve as potential therapeutic targets.

Summary updated

Abstract

Radiation-induced acute intestinal symptoms (RIAISs) are the most frequent complication of radiotherapy that causes great pain and limits the treatment efficacy. The aim of this study was to identify serum biomarkers of RIAISs in cervical cancer patients by surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS). Serum samples were collected from 66 cervical cancer patients prior to pelvic radiotherapy. In our study, RIAISs occurred in 11 patients. An additional 11 patients without RIAISs were selected as controls, whose age, stage, histological type and treatment methods were matched to RIAISs patients. The 22 sera were subsequently analyzed by SELDI-TOF MS, and the resulting protein profiles were evaluated to identify biomarkers using appropriate bioinformatics tools. Comparing the protein profiles of serum samples from the RIAIS group and the control group, it was found that 22 protein peaks were significantly different (P < 0.05), and six of these peaks with mass-to-charge (m/z) ratios of 7514.9, 4603.94, 6887.41, 2769.21, 3839.72 and 4215.7 were successfully identified. A decision tree model of biomarkers was constructed based on three biomarkers (m/z 1270.88, 1503.23 and 7514.90), which separated RIAIS-affected patients from the control group with an accuracy of 81%. This study suggests that serum proteomic analysis by SELDI-TOF MS can identify cervical cancer patients that are susceptible to RIAISs prior to pelvic radiotherapy.
BibTeX
@article{chai2014identification,
  author = {Ying Chai and Jie Wang and Yuan Gao and T. Wang and Fan Shi and Jin Su and Yan Yang and Xia Zhou and Liping Song and Zhihong Liu},
  title = {{Identification of biomarkers for radiation-induced acute intestinal symptoms (RIAISs) in cervical cancer patients by serum protein profiling}},
  journal = {Journal of Radiation Research},
  year = {2014},
  doi = {10.1093/jrr/rru081},
}
Figure