Personalized Medicine Based on Nasal Epithelial Cells: Comparative Studies with Rectal Biopsies and Intestinal Organoids

Abstract

As highly effective CFTR modulator therapies (HEMT) emerge, there is an unmet need to find effective drugs for people with CF (PwCF) with ultra-rare mutations who are too few for classical clinical trials and for whom there are no drug discovery programs. Therefore, biomarkers reliably predicting the benefit from CFTR modulator therapies are essential to find effective drugs for PwCF through personalized approaches termed theranostics. Here, we assess CFTR basal function and the individual responses to CFTR modulators in primary human nasal epithelial (pHNE) cells from PwCF carrying rare mutations and compare these measurements with those in native rectal biopsies and intestinal organoids, respectively, in the same individual. The basal function in pHNEs shows good correlation with CFTR basal function in rectal biopsies. In parallel, CFTR rescue in pHNEs by CFTR modulators correlates to that in intestinal organoids. Altogether, results show that pHNEs are a bona fide theranostic model to assess CFTR rescue by CFTR modulator drugs, in particular for PwCF and rare mutations.

Keywords: CFTR; CFTR modulators; human nasal epithelial cells; intestinal organoids; rectal biopsies; theranostic.

Figure 1

Comparison of basal CFTR-mediated Cl⁻ secretion in primary human nasal epithelial cells (pHNEs) and rectal biopsies. Measurements of basal CFTR-mediated Cl⁻ secretion in pHNEs (left panels) in samples from one healthy control and five PwCF with different genotypes show similarities with the basal CFTR-mediated Cl⁻ secretion measured in rectal biopsies (right panels) from the same individual. The sample code and genotypes are indicated above the tracings. (A,C,E,G,I,K) Original Ussing chamber (open-circuit) recordings showing transepithelial voltage measurements (Vte) obtained for pHNE monolayers with different CFTR genotypes, after cAMP-dependent activation. Amiloride (20 μM) was kept during the whole experiment duration to avoid interference of ENaC-mediated currents. Lumen-negative transepithelial Vte deflections are observed following luminal stimulation by forskolin alone (Fsk, 2 μM) or together with potentiator ivacaftor (VX770, 3 μM). The latter is fully reverted by application of the specific CFTR inhibitor CFTRInh-172 (30 μM). (B,D,F,H,J,L) Representative original recordings of the effects of cholinergic (by carbachol (CCH), 100 μM, basolateral) and cAMP-dependent (by IBMX/Fsk (I/F), 100 μM/2 μM, basolateral) Cl⁻ secretion on transepithelial Vte in rectal biopsies.

Figure 2

Correlation between maximum activation of CFTR-mediated Cl⁻ secretion in pHNEs and rectal biopsies from the same individual. (A) Summary of CFTR-mediated Cl⁻ secretion in rectal biopsies and pHNE cells in one healthy control and five different individuals with CF. CFTR basal function in pHNEs correspond to the differences in Vte after IBMX/Fsk stimulation. CFTR basal function in rectal biopsies is expressed by the difference in Vte upon IBMX/Fsk+IBMX/Fsk/CCH stimulation. Data is represented as the mean of three different experiments for each model ± SD. (B) Correlation of CFTR-mediated Cl⁻ secretion stimulated by IBMX/Fsk in pHNEs and IBMX/Fsk+ IBMX/Fsk/CCH in rectal biopsies (Vte_I/F+I/F/C) from the same individual (Pearson’s R = 0–95, p-value = 0.0012). The grey line represents the linear regression of the data. The grey area represents the 95% confidence interval.

Figure 3

Comparison of CFTR rescue by CFTR modulators in pHNEs and intestinal organoids. Measurements of CFTR rescue by CFTR modulators show similar rescue in pHNEs and intestinal organoids from the same individual. Quantification of forskolin-induced swelling (FIS) assay in intestinal organoids from (A) PwCF homozygous for F508del (control, individual 2); PwCF with the (D) G85E/1717-1G>A (individual 6); (G) P205S/Y1092X (individual 3); and (J) D614G/F508del (individual 4) genotypes treated with ivacaftor (VX-770, 3 µM), lumacaftor (VX-809, 3 µM), and tezacaftor (VX-661, 5 µM) at forskolin (Fsk) concentrations of 0.02, 0.128, 0.8, and 5 μM. FIS data are expressed as the area under the curve (AUC) of organoid surface area increase (baseline = 100%, t = 60 min). Quantification of these is shown in graphs (B,E,H,K) Data represent the mean of measurements ± SD. p-value is indicated in the figure (** = < 0.01; *** = < 0.001; **** = < 0.0001); original Ussing chamber (open-circuit) recordings showing CFTR-mediated Cl⁻ secretion measured as transepithelial voltage (Vte) obtained for pHNE monolayers from the same individual (C,F,I,L), also treated with DMSO, 3 µM of VX-809 or 5 µM VX-661 for 48 h. Data is represented as the mean of three different technical replicates for each model.

Figure 4

Correlation between CFTR rescue by CFTR modulators in pHNEs and intestinal organoids. (A,B) Summary of responses in intestinal organoids (shown as ΔAUC (DMSO-treated) t = 60, 3 µM VX-809 or 5 µM VX-661) and pHNE cells (shown as transepithelial voltage (Vte)) in five different individuals. WT function in intestinal organoids is not comparable due to pre-swelling (due to CFTR pre-activation by endogenous cAMP). According to standard protocols, pHNEs were treated with DMSO, 3 µM VX-809, or 5 µM VX-661 for 48 h and intestinal organoids for 24 h. Data is represented as the mean of three technical replicates for each model ± SD. (C,D) Correlation analysis of CFTR rescue in pHNEs and intestinal organoids from the same individual (Pearson’s R = −0.91, p-value = 0.03 for VX-661 and Pearson’s R = −0.88, p-value = 0.02 for VX-809). The grey line represents the linear regression of the data. The grey area represents the 95% confidence interval.