Allocating colorectal cancer patients to different risk categories by using a five-biomarker mRNA combination in lymph node analysis

PLoS One. 2020 Feb 12;15(2):e0229007. doi: 10.1371/journal.pone.0229007. eCollection 2020.

Abstract

Background and aims: Curative surgery saves ≈50% of all patients with colorectal cancer (CRC) while remaining patients have synchronous or will develop metachronous metastases. Presently, the single most important prognostic factor is histopathological detection of disseminated tumor cells in regional lymph nodes. However, the routine method has several limitations. The aim was to identify biomarker mRNAs that could be combined in a formula that would allow better prediction of patients' survival after surgery.

Methods: Screening for biomarker mRNAs overexpressed in CRC was performed by genome-wide hybridization bead array, with verification by qRT-PCR. Specific qRT-PCR assays with copy standards were developed for 5 selected genes and mRNA expression levels determined in lymph nodes from 174 CRC patients (517 nodes) and 24 control patients (118 nodes). Prognostic value of biomarker mRNAs was estimated. A cut-off was set using univariate Cox regression analysis and used for calculation of differences between patient groups in disease-free survival 12 years after surgery (Kaplan-Meier survival model) and risk for recurrent disease (Cox's regression analysis). A formula was constructed for evaluation of the prognostic value of the biomarkers in combination.

Results: Two new biomarkers, SLC35D3 and POSTN with prognostic value were identified. SLC35D3 was expressed in the epithelium derived tumor cells and POSTN in fibroblasts. Combined with CEACAM5, KLK6 and MUC2 they could be used to identify risk groups. A formula was constructed using CEACAM5 as denominator for KLK6, SLC35D3 and MUC2 and 18S rRNA as denominator for POSTN. The formula yielded 5 categories (-1, 0, 1, 2, 3). Categories (-1 and 0) had good prognosis, categories (1 and 2) relatively poor prognosis and category (3) very poor prognosis.

Conclusion: Lymph node analysis using 5 selected biomarker mRNAs and 18S rRNA in combination allowed allocation of CRC patients to different risk categories with respect to recurrent disease.

Publication types

  • Clinical Trial
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Aged, 80 and over
  • Biomarkers, Tumor* / genetics
  • Biomarkers, Tumor* / metabolism
  • Caco-2 Cells
  • Colorectal Neoplasms* / genetics
  • Colorectal Neoplasms* / metabolism
  • Colorectal Neoplasms* / pathology
  • Female
  • Genome-Wide Association Study
  • Humans
  • Jurkat Cells
  • Lymph Nodes* / metabolism
  • Lymph Nodes* / pathology
  • Male
  • Middle Aged
  • Neoplasm Proteins / genetics
  • Neoplasm Proteins / metabolism
  • RNA, Messenger
  • RNA, Neoplasm* / genetics
  • RNA, Neoplasm* / metabolism
  • Risk Assessment

Substances

  • Biomarkers, Tumor
  • Neoplasm Proteins
  • RNA, Messenger
  • RNA, Neoplasm

Grants and funding

Funding was provided by the Swedish Research Council no. 2010-5669 and 2013-4522 (to M-LH); the Swedish Cancer Foundation (to SH); Stig and Ragna Gorthon Foundation (to GL); Birgit and Henry Knutsson Donation, Helsingborg Hospital (to GL); the Medical Faculty, Umeå University, grants including PhD scholarship to LO (to M-LH); the Lion's Cancer Foundation, Umeå (to M-LH); Umeå Biotech Incubator (to LO). The funding agencies had no role in study design, data collection, data analysis, decision to publish, or preparation of the manuscript.