Solid Tumor Opioid Receptor Expression and Oncologic Outcomes: Analysis of the Cancer Genome Atlas and Genotype Tissue Expression Project

Abstract

Background: Opioid receptors are expressed not only by neural cells in the central nervous system, but also by many solid tumor cancer cells. Whether perioperative opioids given for analgesia after tumor resection surgery might inadvertently activate tumor cells, promoting recurrence or metastasis, remains controversial. We analysed large public gene repositories of solid tumors to investigate differences in opioid receptor expression between normal and tumor tissues and their association with long-term oncologic outcomes.

Methods: We investigated the normalized gene expression of µ, κ, δ opioid receptors (MOR, KOR, DOR), Opioid Growth Factor (OGFR), and Toll-Like 4 (TLR4) receptors in normal and tumor samples from twelve solid tumor types. We carried out mixed multivariable logistic and Cox regression analysis on whether there was an association between these receptors' gene expression and the tissue where found, i.e., tumor or normal tissue. We also evaluated the association between tumor opioid receptor gene expression and patient disease-free interval (DFI) and overall survival (OS).

Results: We retrieved 8,780 tissue samples, 5,852 from tumor and 2,928 from normal tissue, of which 2,252 were from the Genotype Tissue Expression Project (GTEx) and 672 from the Cancer Genome Atlas (TCGA) repository. The Odds Ratio (OR) [95%CI] for gene expression of the specific opioid receptors in the examined tumors varied: MOR: 0.74 [0.63-0.87], KOR: 1.27 [1.17-1.37], DOR: 1.66 [1.48-1.87], TLR4: 0.29 [0.26-0.32], OGFR: 2.39 [2.05-2.78]. After controlling all confounding variables, including age and cancer stage, there was no association between tumor opioid receptor expression and long-term oncologic outcomes.

Conclusion: Opioid receptor gene expression varies between different solid tumor types. There was no association between tumor opioid receptor expression and recurrence. Understanding the significance of opioid receptor expression on tumor cells remains elusive.

Keywords: cancer; immunohistochemistry; neoplasm; opioid receptors; perioperative opioid; surgery; tumor.

Figure 1

Violin (Left panels) and density (Right panels) plots of the expression of opioid receptor genes. Green: tumor samples. Orange: control samples. MOR, µ opioid receptor; KOR, κ opioid receptor; DOR, δ opioid receptor; TLR4, toll–like receptor; OGFR, opioid growth factor receptor. (A, C, E, G, I) Gene expression (Log scale) is on the y axis. (B, D, F, H, J) Gene expression (Log scale) s on the x axis.

Figure 2

Violin and box plot graphs of the expression of opioid receptors genes by tumor type. Green: tumor samples. Orange: control samples. MOR, µ opioid receptor; KOR, κ opioid receptor; DOR, δ opioid receptor; TLR4, toll–like receptor; OGFR, opioid growth factor receptor.

Figure 3

Random effect plot of the mixed logistic model assessing the association between opioid receptors expression and type of tissue. Red dotted line, significance threshold. Dots effect estimates and bar 95% Confidence intervals.

Figure 4

Logistic model fit of opioid receprotrs association with tumor tissue by tumor type. Dotted red line represents no effect. Estimates are reported as red or blue when the odds ratio point estimate is lower or greater than one respectively. Dots are point estimates and bars 95% confidence intervals. Statistical significance is reported as * < 0.05, ** < 0.001, *** < 0.001.