Proteomic analysis to identify markers for response to neoadjuvant treatment in esophageal and gastroesophageal cancer

Abstract Background Esophageal cancer represents a global challenge. Despite significant evolution of treatment protocols in the past decade, recurrence rates are still high and survival rates are poor. Current treatment paradigm for localized gastroesophageal junction (GEJ) carcinoma remains to be further elucidated as for the role of neoadjuvant chemoradiation versus perioperative chemotherapy. Aim To identify biomarkers for response to chemoradiation in esophageal and gastroesophageal cancer, we performed an in‐depth proteomic analysis of esophageal and gastroesophageal tumors, to describe differences in pathway activation between patients with favorable and poor prognosis following neoadjuvant chemoradiation. Methods Patients with locally advanced esophageal and gastroesophageal cancer following neoadjuvant chemoradiation were included in the cohort. The study cohort was dichotomized into two groups of patients, named “favorable prognosis” and “poor prognosis” according to the postoperative disease‐free interval. We performed a mass spectrometry analysis of proteins extracted from the malignant regions of surgical specimens and analyzed data from electronic medical records. Clinical data was correlated with differences in protein expression between patient with a favorable and poor prognosis using validated gene expression pathways. Results The study included 35 patients with adenocarcinoma. All patients in this cohort had esophageal adenocarcinoma. Patients median age was 62 years. Twenty‐five (71.3%) patients underwent neoadjuvant chemoradiation, and 28.7% underwent neoadjuvant chemotherapy only. A proteomic analysis of our cohort identified 2885 proteins. Enrichment levels of 98 of these proteins differed significantly between favorable and poor prognosis cohorts in patients who underwent neoadjuvant chemoradiation (p < .05) but not in patients who underwent neoadjuvant chemotherapy. The favorable prognosis patients group analysis exhibited differential enrichment of 87 proteins related to cellular respiration and oxidative phosphorylation pathways as well as proteins of the RAS oncogene family. Conclusion In this study we identified differential enrichment of pathways related to oxidative phosphorylation and RAS oncogene pathway in esophageal cancer patients with a favorable response to chemoradiation. Following further validation, our findings may portray potential surrogate signature of biomarkers based upon these pathways.

Conclusion: In this study we identified differential enrichment of pathways related to oxidative phosphorylation and RAS oncogene pathway in esophageal cancer patients with a favorable response to chemoradiation. Following further validation, our findings may portray potential surrogate signature of biomarkers based upon these pathways. follow-up of the Cross-trial participants confirmed the survival benefit. 6 Other milestone trials have demonstrated that perioperative chemotherapy alone may also provide a significant survival advantage for gastroesophageal junction cancer patients. The most significant were the MAGIC trial 7 (with 12% GEJ tumors and 15% lower esophageal tumors) and the French FFCD trial 8  Patients were dichotomized into two groups, named "favorable prognosis" (FP) and "poor prognosis" (PP) based upon the time interval from surgery to disease recurrence. Patients who had no recurrence of disease at least 1 year following surgery were defined as "favorable prognosis" and patients that demonstrated tumor recurrence within 1 year of surgery or less were defined as "poor prognosis."

| Proteomics
Pathological slides from all patients in the cohort were reviewed by a pathologist from our institution, that also chose the relevant tumor areas for analysis. These areas were dissected and used for proteomics analysis.
Key components of the quantitative proteomics approach were employed, allowing the quantitation of targeted proteins from FFPE tumor biopsy. The mass spectrometry data was analyzed using the MaxQuant software 1.5.2.8 13 for peak picking and identification using the Andromeda search engine, searching against the human proteome from the Uniprot database with mass tolerance of 6 ppm for the precursor masses and 20 ppm for the fragment ions. Oxidation on methionine and protein N-terminus acetylation were accepted as variable modifications and carbamidomethyl on cysteine was accepted as static modifications. Minimal peptide length was set to six amino acids and a maximum of two miscleavages was allowed. The data was quantified by label free analysis using the same software. Peptide-and protein-level false discovery rates (FDRs) were filtered to 1% using the target-decoy strategy. Protein table were filtered to eliminate the identifications from the reverse database, and common contaminants and single peptide identifications.
Statistical analysis of the identification and quantization results was done using Perseus 1.6.2.2. 14 T-test was done, and differential proteins were proteins with p value < .1 with at least twofold change (>1 or À1> at the difference).

| RESULTS
Thirty-five patients included in the study cohort underwent neoadjuvant therapy followed by surgery. Twenty-five (71.3%) of these patients underwent neoadjuvant chemoradiation, and 10 patients Mitochondrial oxidative phosphorylation is upregulated in most types of cancer cells and is currently being investigated as a potential target for cancer treatment. [17][18][19] One group of differentially enriched proteins we found in this study were NDUF proteins which belong to  gets. 26 Proteins of the RAB family were shown to be markers for radio-resistance in esophageal cancer, 27 as well as potential targets for radio-sensitization of rectal cancers. 28 Considering this evidence and our findings it seems that following further validation, RAB proteins may serve as biomarkers for response to neoadjuvant chemoradiation in esophageal cancer.
Our analysis identified 11 significantly enriched proteins in the poor prognosis group. Some of these proteins are known to play a significant role in biologically related cancers survival. One example is F I G U R E 1 Protein scatter plot. This graph depicts a representative graph of label free quantification (LFQ) of proteins in favorable and poor prognosis cohorts on a logarithmic scale. The proteins with the most significant differential enrichment in the favorable prognosis group appear on the left side of the curve, and proteins with the most significant differential enrichment in the poor prognosis group appear on the right side of the curve DSG1, a type of desmosomal cadherin, known to be associated with decreased survival in head and neck squamous cell carcinoma. 29 Another example is RAb11-FIP2 which was shown to be increased in an immunohistochemical analysis of 86 gastric cancer patients and was closely correlated with nodal metastasis. 30 Though each of these proteins may be a potential biomarker for unfavorable response to chemoradiation, but our analysis did not identify any common biological process or interaction between these proteins.
Study limitations account for heterogenous chemotherapeutic protocols that may modify proteomic signature of the resected tumor.
Nevertheless, a substantial part received fluoropyrimidine and platinum and the others taxane and platinum. We employed proteomic analysis of surgical specimens following neoadjuvant treatment, that may have altered baseline proteomic landscape. Most importantly, this is a preliminary report, that warrants further validation. Following such validation, the proteins we identified may serve as biomarkers for favorable response to radiation in gastroesophageal cancer. Assuming our findings are indeed validated, future studies that will prospectively employ baseline and posttreatment samples may provide additional insights for esophageal cancer patients stratification.

| CONCLUSIONS
In conclusion, our study revealed a potential predictive proteomic signature to neoadjuvant chemoradiation in patients with esophageal Further studies are warranted to confirm whether these pathways and proteins may serve as a surrogate biomarker for patient selection. Thus, Neoadjuvant chemoradiation may be administered to appropriate patients in an era of optional perioperative chemotherapy as an alternative mode of neoadjuvant therapy for GEJ tumors.

ACKNOWLEDGMENTS
We would like to appreciate our gratitude to the Smoler Proteomics Center at the Technion institute of technology for their significant contribution to this study.

CONFLICT OF INTEREST
The authors have stated explicitly that there are no conflicts of interest in connection with this article.

AUTHOR CONTRIBUTIONS
All authors had full access to the data in the study and take responsi-

DATA AVAILABILITY STATEMENT
Clinical data are stored in an institutional database and will be shared upon request to the corresponding author. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE 31 partner repository with the dataset identifier PXD025192.