We report four novel families with LPVs in RNF43. Our findings suggest variable penetrance of SPS. It is notable, that only three out of eight carriers (all relatives) had either serrated polyps or CRC. Of the variants none had previously been reported in literature in association with SPS. One of the identified variants occurred in two families.

Our findings demonstrate that not all carriers of LPV in RNF43 fulfill the WHO criteria for SPS. Only two members of one of our families fulfilled the WHO criteria (2019) for SPS despite having undergone a colonoscopy at a relatively high age. In the previously published cases of patients with PV/LPV in RNF43, the patients were diagnosed with SPS between the age of 18‒65 (Supplementary Table 1). Additionally, we found one case of gastric cancer but no other extracolonic cancer corresponding to findings in previous studies [13, 14].

We acknowledge that the classification of variants in RNF43 according to the general ACMG guidelines is a stretch, as RNF43 could be considered a gene of unknown significance (GUS) and Mendelian inheritance is not firmly established. However, we have used these guidelines precisely to highlight the issue in classifying variants where a gene-disease link and Mendelian inheritance have not been firmly established. In 2 patients (in family 2 and 3) we identified the variant RNF43, c.1948 C > T, p.(Arg650Ter), which have not previously been linked to SPS. The variant is classified as likely pathogenic because it in theory results in nonsense mediated decay (NMD). However, the effects of the variant have not been investigated experimentally, and other genes have been found to escape NMD, thus, any disease association is still not proven.

The results from previous studies suggest some degree of ascertainment bias as most patients underwent genetic testing due to a phenotype suggestive of SPS or a family history thereof. Thus, the risk of being a carrier of a PV/LPV in RNF43 and the associated increased risk of SPS/CRC in naive families is unknown.

Based on our findings, we find it speculative if PVs in RNF43 should be regarded as a contributing polygenetic risk factor, more than a monogenetic cause. On the other hand, we do know that other cancer predisposition syndromes express varying degrees of decreased penetrance (MSH6, PMS2, DICER1 etc.), thus a monogenic association is difficult to rule out before more carriers are identified and published. A family published by Chan et al., could support variable penetrance with only one patient with SPS in a large family with many carriers of the family variant in RNF43 [15]. This highlights the complexity of genetic counseling in RNF43 positive families, and whether healthy relatives should be offered predictive testing. This is a well-known issue in the field of hereditary cancer, where the indication and effect of surveillance programs are controversial ‒ to name a few: CHEK2 and ATM. Although some guidelines have proposed surveillance in families that carry a LPV/PV in RNF43, there is a definite risk of overtreatment [16]. Thus, it seems reasonable to rely on the patient phenotype and whether the families fulfill the clinical criteria of SPS and not on the genetic findings alone. This is in accordance with recommendations from the American Consortia Guidelines which conclude that there are insufficient data to support multigene panel testing in patients with SPS without a family history of CRC and/or adenomatous polyposis [17]. British guidelines state that RNF43 can be included in a gene panel for SPS, but that a routine recommendation is not warranted due to the low frequency of LPV/PVs in RNF43 [18]. Thus, it should be considered whether RNF43 should be included in multigene cancer panels, and if LPVs/PVs in RNF43 should lead to colonoscopy surveillance in SPS and CRC naïve families.

In conclusion, we present four novel families with LPVs in RNF43– three of these without serrated polyps and/or CRC. Our results suggest that the rationale for integration of RNF43 in broader cancer gene panels can be questioned– and that the genetic counseling of the family should be handled carefully, considering cancer and SPS family history and with information about reservations. Further research is needed to elucidate the role of RNF43 in the risk of SPS and CRC.