Population-specific variations in KCNH2 predispose patients to delayed ventricular repolarization upon dihydroartemisinin-piperaquine therapy

Abstract

Dihydroartemisinin-piperaquine is efficacious for the treatment of uncomplicated malaria and its use is increasing globally. Despite the positive results in fighting malaria, inhibition of the Kv11.1 channel (hERG; encoded by the KCNH2 gene) by piperaquine has raised concerns about cardiac safety. Whether genetic factors could modulate the risk of piperaquine-mediated QT prolongations remained unclear. Here, we first profiled the genetic landscape of KCNH2 variability using data from 141,614 individuals. Overall, we found 1,007 exonic variants distributed over the entire gene body, 555 of which were missense. By optimizing the gene-specific parametrization of 16 partly orthogonal computational algorithms, we developed a KCNH2-specific ensemble classifier that identified a total of 116 putatively deleterious missense variations. To evaluate the clinical relevance of KCNH2 variability, we then sequenced 293 Malian patients with uncomplicated malaria and identified 13 variations within the voltage sensing and pore domains of Kv11.1 that directly interact with channel blockers. Cross-referencing of genetic and electrocardiographic data before and after piperaquine exposure revealed that carriers of two common variants, rs1805121 and rs41314375, experienced significantly higher QT prolongations (ΔQTc of 41.8 ms and 61 ms, respectively, vs 14.4 ms in controls) with more than 50% of carriers having increases in QTc >30 ms. Furthermore, we identified three carriers of rare population-specific variations who experienced clinically relevant delayed ventricular repolarization. Combined, our results map population-scale genetic variability of KCNH2 and identify genetic biomarkers for piperaquine-induced QT prolongation that could help to flag at-risk patients and optimize efficacy and adherence to antimalarial therapy.

Keywords: Kv11.1; QT prolongation; cardiotoxicity; hERG; long QT; malaria.

Fig 1

The landscape of genetic variability of KCNH2 (hERG). The 3D structure of Kv11.1 is shown in transversal (A) and lateral views (B). One of the four identical subunits is shown as opaque and the other three as transparent. The PD, VSD, PAS, and cNBD are shown in red, brown, turquoise, and gray, respectively. (C) The total number of variants (left y-axis; column plot) and the number of variants per amino acid (right y-axis; dot plot) are shown for each domain. (D) Schematic representation of the distribution of variants across protein domains. All common variations in KCNH2 are shown in dark red, all rare variants in magenta. (E) The frequencies of all common KCNH2 variations are shown across five major ethnogeographic groups. Mean and standard deviations across the indicated five populations (Africans, East Asians, Finnish, non-Finnish Europeans, and Latin Americans) are shown.

Fig 2

The KCNH2 gene harbors a multitude of rare and population-specific variations that are predicted to have deleterious effects on channel function. (A) ROC curves of sensitivity (ordinate) vs 1-specificity (abscissa) are shown for 16 computational variant effect predictors. Areas under the ROC curves (AUC) are indicated for each algorithm. Dots indicate the performance when using the original parameters, while diamonds indicate performance after informedness optimization. (B) The increase in informedness upon parameter optimization is shown for each algorithm. (C) The carrier frequency of deleterious KCNH2 variants, obtained by aggregating the population-specific frequencies of all putatively deleterious variants, is shown for different populations. Dashed line indicates the global average based on genomic data from 141,614 individuals. (D) Pie chart showing that the majority of deleterious KCNH2 variations (72%) were only identified in a single population. Due to the population imbalance of available genomic data, the largest number of population-specific variations is presently identified in Europeans (stacked column plot).

Fig 3

Association of identified populations with QT prolongation. (A) Absolute QTc intervals in milliseconds (ms) measured before (day 0) and after 2 days of dihydroartesunate piperaquine combination therapy. **** indicates P < 0.0001 using a Kruskal-Wallis test. (B) The variation in QTc from day 0 to day 2 (ΔQTc) was not associated with measured serum piperaquine levels. The black dashed line indicates no change in QTc intervals (ΔQTc = 0), the red dashed line indicates an increase in QTc above the clinically significant threshold of 30 ms, and the black solid line constitutes the regression line (P = 0.65 for deviation of slope from zero). (C–E) The variation in QTc from day 0 to day 2 (ΔQTc) is shown for carriers of the common variations rs1805121 (C), rs41314375 (D), and rs1137617 (E). The upper inlets show the fraction of variant carriers (red) and controls (blue) who experience QT prolongations >30 ms. The lower inlets show serum piperaquine concentrations in carriers (red) and controls (blue). * and ** indicate P < 0.05 and P < 0.01 using a Kruskal-Wallis test. (F) QTc interval variations are shown for carriers of rare and novel KCNH2 variations for successive episodes. Time difference between episodes is indicated.