Chemotherapy-Induced Oral Mucositis: 11 SNPs Implicated In GWAS Of 972 Patients With Multiple Myeloma

Patients who receive high-dose chemotherapy, such as induction regimens followed by high dose melphalan (HD-MEL) and autologous stem cell transplant (ASCT) to treat multiple myeloma (MM), often develop oral mucositis (OM) that impairs adequate intake of fluids and nutrition, serves as a breeding ground for bacteria, leads to increased complexity and cost of care, and may interrupt or delay cancer treatment. Previous work by members of our team demonstrated that renal function and melphalan dose contribute to risk predisposition for oral mucositis. In this study we explored whether genetic variants along with baseline clinical characteristics contribute to risk for oral mucositis by conducting a genome-wide associations study (GWAS).

Following strict data quality screening and genotyping using Illumina’s HumanOmni1-Quad v1.0 BeadChip, 892,589 SNPs were tested for association to OM. The analyses evaluated oral mucositis risk in three steps: 1) baseline clinical characteristics; 2) genetic factors (SNPs); and then 3) clinical characteristics combined with genetic factors. Among 972 Caucasian patients who participated in Total Therapy clinical trials (TT2, TT3, TT4) with HD-MEL and ASCT for newly diagnosed myeloma, 353 developed grades 2-4 OM and 619 developed grades 0-1 OM. The baseline sample characteristics included mean age 57.65 (SD 9.16); gender (male 63.68); mean estimated GFR 80.44 (SD 28.40); mean serum albumin 3.85 (SD 0.50); and mean melphalan dose 4.44 (SD 0.79)

The combination of estimated glomerular filtration rate (GFR) with melphalan dose (mg/kg), serum albumin, and female gender predicted 67.2% of the risk for oral mucositis (grades 2-4). Analysis of the genetic data identified 11 SNPs significantly associated with Grade 2-4 oral mucositis. The SNPs are located in or near MMP13, JPH3, DHRS7C, CEP 192, CPEB1/CEBP, FBN2, ALDH1A1, DMRTA1/FLJ35282 and have functions related to gene regulation and transcription. We surmise that inflammatory pathways underlie the significance of these SNPs; for example, expression of MMP13 is upregulated in oral epithelial cells during inflammation and FBN2 contributes to wound healing, possibly related to TGFβ signaling and availability. Cyclophospamide (CTX) was part of the total therapy regimens and is known to be detoxified by dehydrogenases. SNP variations associated with aldehyde dehydrogenase 1A1 (ALDH1A1) may alter susceptibility to chemotherapy induced oral mucositis in the presence of HD-MEL, CTX, and other regimens. We postulate that SNPs associated with oral mucositis may influence gene expression through non-coding RNA, thus providing a pathway and novel drug target to prevent this deleterious treatment effect. We will test this hypothesis in future work.