Reply: CFTR analysis should not be offered to all patients with unexplained azoospermia in the presence of normal gonadotropin levels

Sir,

We thank Cantineau et al. (2021) for their critical comments on our recent paper (Rudnik-Schöneborn et al., 2021). The authors suggest that the evaluation of andrological variables was incomplete in our study group of 71 patients. We agree that in an ideal world all possible parameters would be assessed in all patients. However, in the real world—the routine diagnostic setup—information often remains incomplete. In our retrospective data from several large fertility centers in Germany and Austria, only sperm count and FSH levels were inclusion criteria for all cases. In contrast, ultrasound of the seminal vesicles is only rarely performed in the andrological workup. Nevertheless, the data available in normal clinical practice can assist in deciding on cystic fibrosis transmembrane conductance regulator (CFTR) genotyping. Nearly all probands (62 of 64, 97%) with presumed CFTR-related infertility had at least two semen abnormalities suggestive of obstructive azoospermia, and genetic analyses are recommended in these individuals. On the other hand, normal semen or imaging parameters do not completely rule out CFTR-related azoospermia, and genetic analyses might be considered in these cases. It is interesting to note that in our cohort three out of the four men with reportedly normal seminal vesicles had further evidence of obstructive azoospermia (data not shown).

We thoroughly analyzed and classified the pathogenic effect of the CFTR mutations in our cohort. Although most mutations were not confirmed to be in trans by segregation studies, we are very familiar with the CFTR allele from cystic fibrosis genotyping. We pointed out that the combination of one well-known pathogenic CFTR mutation and one (rare) variant of unknown significance (VUS) most likely can be interpreted as biallelic. The classification as VUS mostly applies to catagories defined for severe cystic fibrosis. Functional studies of milder CFTR variants with regard to congenital bilateral absence of the vas deferens (CBAVD) or obstructive azoospermia are limited. In the context of our study only a single relevant cis double mutant constellation—discussed by us—is known in our populations. Thus a biallelic constellation can be assumed to be present in at least the great majority of our patients. In addition, the andrological phenotype was strikingly homogeneous in our patient group, and we did not find significant clinical differences between individuals with confirmed and postulated CFTR deficient genotypes. Clearly, we cannot rule out a small number of individuals in whom the presence of two CFTR variants is unrelated to the azoospermia phenotype. Nevertheless, we are confident that the andrological abnormalities observed in our study correctly reflect the phenotype of men with CFTR-related azoospermia.

The authors of the letter are correct that previous guidelines recommend to restrict CFTR mutation analysis to patients with CBAVD, but they also admit that it can be hard to confirm CBAVD. Our study was initiated because of our clinical experience that the andrological manifestation spectrum of CFTR-related azoospermia is broader than previously reported. This was indeed confirmed by our data. Science advances and recommendations need to be updated accordingly.

The authors of the letter also point out that increased CFTR genotyping will result in further detection of VUS and of heterozygous carriers, and they are concerned of concomitant unnecessary costs to the health system. We agree that this is a highly relevant issue but would like to emphasize the importance of making correct diagnoses—particularly in CFTR-related azoospermia—for clinical management decisions and the assessment of offspring genetic risks. Any medical activity requires a balance between potential benefits, risks, and cost. Considering the rapidly falling costs of genetic analysis and the improved understanding of genetic variants and their relevance, we believe that one should not underestimate the value of genetic diagnostics in clinical care.

References