Over 2100 variants have been identified in the CFTR gene
[10], of which approximately one-third are now classified as CF-causing
[11]. The deletion of a phenylalanine at position 508 (p.Phe508del, legacy:
F508del) is the most prevalent CF-causing variant, accounting for approximately 70% of all CF alleles
[1], while the remaining 30% of CF alleles are represented by an enormous number of CFTR variants and most are (ultra)rare, occurring among few PwCF worldwide
[11]. Due to such CFTR allelic heterogeneity, distinct CF phenotypes exist—on average, PwCF with pancreatic insufficient exhibit more severe forms of the disease, while milder phenotypes are usually associated with pancreatic sufficiency
[12]. Indeed, these variants cause distinct primary defects, comprising CFTR mRNA and protein biosynthesis, anion transport, and/or PM turnover. Therefore, they have been separated into CFTR variant classes, which are characterized by alterations in (I) expression, (II) folding and trafficking, (III) gating, (IV) conductance, (V) abundance, and (VI) PM stability
[13,14][13][14]. Overall, CFTR variants in classes I and II are associated with a minimal (or null) function, while a residual (or some) function is usually observed in those variants in classes IV–VI. This grouping offers the advantage that CFTR variants with similar defects might be tackled by similar therapeutic strategies—i.e., theratyping
[15].
Over the last two decades, precision (or personalized) medicines targeting the fundamental cause of CF have been developed with tremendous accomplishments attained
[14,16][14][16]. Indeed, four CFTR modulator drugs are now approved for clinical use: the correctors lumacaftor (LUMA, VX-809), tezacaftor (TEZA, VX-661), and elexacaftor (ELX, VX-445), which retrieve CFTR folding and trafficking to the PM, and the potentiator ivacaftor (IVA, VX-770), which enhances CFTR channel open probability. These drugs—particularly the ‘highly effective’ ones—have provided impressive clinical benefits, representing thus a new dawn for PwCF with eligible genotypes
[17,18,19,20,21,22,23][17][18][19][20][21][22][23]. However, there is a significant number of PwCF carrying (ultra)rare variants for whom no modulator therapy has been approved
[11,16][11][16].
2. Laboratory Tools to Predict CFTR Modulator Effectiveness In Vivo
Several in vitro assays have been developed using various model systems to comparatively assess the efficacy of CFTR modulators (individually or combinations thereof)
[29][24]. Although heterologous cell lines have been fundamental to enhance the understanding of CFTR biology at genetic, biochemical, and physiological levels, in the context of precision medicine, they can only be used for variant theratyping (i.e., matching single variants to modulators), being unable to predict responses of a determined individual to a specific therapy. Accordingly, alternative translational models have been established by using primary cells from PwCF to inquire about responses at an individual level (
Figure 1).
Figure 1. Translational human CF model systems for personalized/precision medicine. Airway samples can be collected from the nose or bronchi by brushing or biopsies and gastrointestinal samples can be collected from the colon or rectum by biopsies. These samples are cultured in specific in vitro conditions, expanded, and then seeded on porous membranes to grow in polarized monolayers or in matrigel to form organoids/spheroids. Cell monolayers and organoids can be used to assess CFTR function/rescue through Ussing chamber measurements and through forskolin-induced swelling assays, respectively.
While cells heterologously expressing CF-causing variants remain very useful for CFTR studies, primary cell models can provide a more sensitive and reliable prediction of therapeutic responses for several reasons: (i) The last express CFTR in the native genomic context, while cell lines frequently use CFTR cDNA (a copy of the mature mRNA, which lacks the introns); therefore, cells lines may not recapitulate certain cellular mechanisms, including nonsense-mediated decay or splicing effects. For instance, p.Gly970Arg (legacy
G970R) was thought to be a CFTR gating variant based on cDNA expression findings
[30,31][25][26]; however, analysis of cells from PwCF carrying this variant revealed that it actually causes a splicing defect
[32,33][27][28]. (ii) The cellular background has a marked influence on CFTR processing and function as well as its pharmacological sensitivity. As exemplified by the cases of p.Phe508del
[34,35][29][30] and p.Gly1244Glu
[36][31], the complexity of cellular processes related to protein biogenesis and folding, as well as its PM trafficking, may not be completely recapitulated in cell lines—particularly those from non-human and/or non-respiratory epithelium origin
[29][24]. Likewise, CFTR gene regulation can be impacted by epigenetic factors and these are only taken into account by the assessment in primary cells from each individual
[37][32]. (iii) Characterization of single variants can be efficiently accomplished in cell lines; however, two variants indicating to be low responsive (below therapeutic relevant threshold) in separated cell lines can compose a genotype with a good prediction for clinical benefits (if evidenced by assessing responses in this individual’s cells). This is particularly relevant for (ultra)rare CFTR variants that are frequently identified in racial and ethnic minority populations, which are usually excluded from traditional clinical trial designs
[38][33].
Moreover, numerous reports have established correlations of the data on primary cell models with clinical features of PwCF (before and after initiating modulator therapy) to provide a translational perspective of therapeutic effects (
Table 1). Accordingly, these can serve as potential biomarkers to identify which drug(s) could be the best therapeutics for every individual with CF—i.e., “the right therapy for the right person”.
Table 1.
Studies reporting correlations of CFTR modulator-promoted responses in cell models and clinical effects in PwCF.