A pharmacogenetic candidate gene study of tenofovir-associated Fanconi syndrome

Abstract

Background: Tenofovir disoproxil fumarate (TDF) is a widely used antiretroviral agent with favorable efficacy, safety, and tolerability profiles. However, renal adverse events, including the rare Fanconi syndrome (FS), may occur in a small subset of patients treated for HIV infections.

Objectives: The aim of this study was to identify genetic variants that may be associated with TDF-associated FS (TDF-FS).

Methods: DNA samples collected from 19 cases with TDF-FS and 36 matched controls were sequenced, and genetic association studies were conducted on eight candidate genes: ATP-binding cassette (ABC) transporters ABCC2 (MRP2) and ABCC4 (MRP4), solute carrier family members SLC22A6 (OAT1) and SLC22A8 (OAT3), adenylate kinases 2 (AK2) and 4 (AK4), chloride transporter CIC-5 CLCN5, and Lowe syndrome protein OCRL. The functional effects of a single nucleotide polymorphism (SNP) predicted to alter the transport of tenofovir were then investigated in cells expressing an identified variant of ABCC4.

Results: The case group showed a trend toward a higher proportion of rare alleles. Six SNPs in ABCC2 (three SNPs), ABCC4 (one SNP), and OCRL (two SNPs) were associated with TDF-FS case status; however, this association did not remain significant after correction for multiple testing. Six SNPs, present in OCRL (four SNPs) and ABCC2 (two SNPs), were significantly associated with increased serum creatinine levels in the cases, and this association remained significant after multiple test correction (P < 2 × 10). One synonymous SNP in ABCC2 (rs8187707, P = 2.10 × 10, β = -73.3 ml/min/1.73 m(2)) was also significantly associated with the decreased estimated glomerular filtration rate of creatinine among cases. However, these results were driven by rare SNPs present in a small number of severely affected cases. Finally, a previously uncharacterized, nonsynonymous SNP, rs11568694, that was predicted to alter MRP4 function had no significant effect on tenofovir cellular accumulation in vitro.

Conclusion: Although no single predictive genetic marker for the development of TDF-FS was identified, the findings from our study suggest that rare variants in multiple genes involved in the renal handling of tenofovir, and/or renal cell homeostasis, may be associated with increased susceptibility to TDF-FS.

Figure 1. Characteristics of SNPs Identified in TDF-FS Cases and Controls

(A) Overall SNP frequency in the study subject samples by predicted functional category: “3utr” = 3′untranslated region, “5utr” = 5′untranslated region, “coding” = all synonymous and non-synonymous coding SNPs, “intronic” = all SNPs located within the intron/exon flanking regions (< 200 bp from exon termini). (B) Boxplots represent median MAF ± IQR (y-axis) in cases and controls by predicted functional category (colored squares).

Figure 2. Distribution of Rare Variants

(A) Frequency distribution of alleles in all cohort subjects. (B) Density plot of the cumulative distribution of SNPs in cases (red line) and controls (black line) by number of SNPs (x-axis). (C) Boxplots show median MAF ± IQR (plotted on the y-axis) in the cases and controls.

Figure 3. Functional Characterization of the MRP4-V854F SNP

(A) Time course of tenofovir intracellular accumulation. (B) Time course of tenofovir accumulation in the supernatant (extracellular concentration). Y-axis values represent % of the reference (MRP4-ref) intracellular amount of tenofovir at t=0. Red squares indicate HEK-293-MRP4-reference cell line (reference). Yellow squares indicate HEK-293-MRP4-V854F cell line (variant). Blue squares represent HEK-293-mock cell line (untransfected control). Error bars indicate SEM.