Dual single‑nucleotide polymorphism biomarker combination for opioid selection to treat cancer pain

Abstract

We have been exploring biomarkers that could help physicians select the appropriate opioid for individualized treatment of cancer pain. Recently, we identified a single nucleotide polymorphism (SNP) of CCL11 (rs17809012) as one such biomarker that was significantly associated with the analgesic effect of morphine. The current study measured the plasma concentrations of chemokines/cytokines in pre-treatment plasma samples of a total of 138 patients who were randomized to receive morphine (n=70) or oxycodone (n=68). Based on the results, one cytokine, IL-16, was identified whose plasma concentrations showed a clear bias between patients with cancer pain who responded well and responded poorly to oxycodone. A genotypic analysis also identified a SNP of the IL-16 gene (rs4778889) as being significantly associated with the analgesic effect of oxycodone. Whether both of the SNPs identified as being significant (CCL11 rs17809012 and IL-16 rs4778889) could be used in combination to accurately predict which opioid might be the most suitable to provide pain relief in patients with cancer was assessed. Morphine tended to provide superior analgesic effect over oxycodone in patients with the IL-16 rs4778889 TT genotype and the CCL11 rs17809012 AG/GG genotype (n=45), while a trend towards a greater analgesic effect of oxycodone was observed in patients with other genotype combinations of these SNPs (n=93) (P=0.0012 for the interaction), suggesting that the IL-16 rs4778889 and CCL11 rs17809012 SNPs could serve as a potential dual-biomarker combination for personalized analgesic therapy in patients with cancer pain.

Keywords: CCL11; IL-16; cancer pain; genotype; morphine; oxycodone; plasma concentration; single nucleotide polymorphism.

Figure 1

Schema of the study design. IL-16, interleukin 16; CCL11, C-C motif chemokine ligand 11; NRS, numerical rating scale; SNP, single-nucleotide polymorphism.

Figure 2

Forest plots comparing the ∆NRS for the two treatments according to the plasma concentrations (low and high) of 39 chemokines/cytokines. The dotted line represents the regression coefficient (estimate) for treatment in the overall subject population. ∆NRS, difference in the score on the numerical rating scale for pain recorded prior to the start of opioid treatment and after opioid titration. CI, confidence interval; L, concentrations (pg/ml) lower than the median values for each cytokine; H, concentrations (pg/ml) higher than the median values for each cytokine; NRS, numerical rating scale.

Figure 3

Forest plots comparing the ∆NRS for the two treatments according to the genotypes of three SNPs (IL-16_rs4778889, IL-16_rs11556218 and CCL11_rs17809012) and the genotype combinations (I-IV for each of the IL-16/CCL11 SNP combinations). The dotted line represents the regression coefficient (estimate) for treatment in the overall subject population. ∆NRS, difference in the score on the numerical rating scale for pain recorded prior to the start of opioid treatment and after opioid titration. NRS, numerical rating scale; SNP, single-nucleotide polymorphism; CI, confidence interval; Others, patients with the other genotype combinations; IL-16, interleukin 16; CCL11, C-C motif chemokine ligand 11.

Figure 4

Significance of the ∆NRS in patients treated with morphine (blue) or oxycodone (orange) is expressed by the LSMs ± standard error separately for the genotype combinations (rs4778889/rs17809012) of: i) TT/AA; ii) TT/(AG/GG); iii) (TC/CC)/AA; and iv) (TC/CC)/(AG/GG). ∆NRS, difference in the score on the numerical rating scale for pain recorded prior to the start of opioid treatment and after opioid titration; Numbers in parentheses indicate the numbers of patients. LSMs, least square means; NRS, numerical rating scale.