Investigation of the influence of germ line polymorphisms as determinants of erlotinib toxicity

Background: Erlotinib is an orally active tyrosine kinase inhibitor used in the treatment of non-small cell lung cancer (NSCLC) (1). Erlotinib toxicity is the sole reason for dose reduction. Erlotinib is metabolised predominately by CYP3A4 and is substrate for the ATP-Binding Cassette Sub-Family B Member 1 (ABCB1) and ATP-binding cassette sub-family G member 2 (ABCG2) efflux transporters (2). Single nucleotide polymorphisms (SNP) found within key enzymes of metabolism are known to affect erlotinib pharmacokinetics. We sought to investigate the effect of germ line variation on toxicity and response to treatment.

Methods: Patients with NSCLC and having received at least one cycle of erlotinib (150mg erlotinib hydrochloride oral once daily) were recruited from Guy’s and St Thomas’ Oncology department, London. Whole blood samples were obtained and genotyped to assess the association between candidate SNPs and the incidence of toxicity/ clinical response using a two-tailed Fishers exact test in additive, dominant and recessivemodels. All toxicities were noted and graded for cycles 1-6 of erlotinib treatment according to Common Terminology Criteria for Adverse Events Version (CTCAE) 4 (14 June 2010) (3).

Results: The mean age of the cohort was 66 years, the majority of subjects had adenocarcinoma and 70% had Stage IV NSCLC. The most commonly recorded toxicity was grade 1 rash. 34% of patients were dose reduced to 100mg, and the most frequent cause for dose reductions was skin rash. 15 single nucleotide polymorphisms within six candidate genes were genotyped and tested for associations with toxicity and clinical response for 33 patients.

A significant protective effect of ABCB1 c.2677 A>C (rs2032582) polymorphism against any grade diarrhoea was found. Variant rs2032582 within ABCB1 were significantly associated with reduced incidence of grade 1 and above diarrhoea within cycle 1 and cycles 1-3 (cycle 1 rs2032582 P=0.003, OR 0.003 (95% CI 0.363-0.899), cycle 1-3 rs2032582 P=0.005, OR 0.45 (95% CI 0.04-0.480)) in a recessive model. After Bonferroni correction for multiple testing the variant was significantly associated with reduced diarrhoea in cycle 1. No significant associations were found between candidate SNPs and incidence of erlotinib related rash or response to treatment.

Discussion: The protective effect of ABCB1 c.2677 A>C may be explained by increasing the affinity of ABCB1 for erlotinib, thereby increasing efflux out of the intestinal cells and reducing erlotinib-related diarrhoea. In this pilot study we have reported for the first time a protective association between SNP ABCB1 c.2677A>C (rs2032582) and increased diarrhoea.

References:

(1) NICE Guidelines on Erlotinib for the first-line treatment of locally advanced or metastatic EGFR-TK mutation-positive non-small-cell lung cancer, www.nice.org.uk/guidance/ta258/chapter/1-guidance.

(2) Erlotinib Pathway, Pharmacokinetics 2012, https://www.pharmgkb.org/pathway/PA154426903#PGG

(3) Common Terminology Criteria for Adverse Events Version 4.0 2010, http://ctep.cancer.gov/protocolDevelopment/electronic_applications/ctc.htm