In this study, we found SVEP1 SNPs correlated most closely with the response to tolvaptan. This is the first report showing a strong correlation at a genome-wide level of significance between a genetic variant (SVEP1) and the response to tolvaptan. The SVEP1 gene is located on 9q32, spans 214 kb of genomic DNA, and consists of 48 exons, encoding a secreted multi-domain protein which harbors sushi (named also complement control protein; CCP), von Willebrand factor (VWF) type A, EGF, and pentraxin-domain motifs (14). SVEP1 is an extracellular matrix protein involved in lymphatic vessel remodeling and plays a critical rule in epidermal differentiation (13). Previously known as Polydom, SVEP1 is also a high-affinity ligand for integrin a9b1. Polydom knockout mice show severe edema and die immediately after birth as a result of respiratory failure due to dysfunction of fluid drainage; these mice also fail to undergo remodeling and formation of collecting lymphatic vessels (13). SVEP1 is expressed in heart, lung, skeletal tissue, placenta, stomach, intestine, stromal osteogenic tissues and so on. A recent report showed that missense variants of SVEP1 were significantly related to coronary arterial disease (15).
With respect to clinical data, only BUN was identified as a response marker based on multivariate analysis (Table 3), as previously reported (6, 8). This result showed that renal dysfunction, especially caused by dehydration, may worsen the response to tolvaptan. On the other hand, the SVEP1 SNP (rs2991364), which was associated with SVEP1 expression levels, would influence the vulnerability of lymphatic vessels, which could lead to fluid retention. The combination of BUN and SVEP1 SNP showed a stronger correlation with the response to tolvaptan compared with BUN or SVEP1 SNP alone. Previous reports showed that various parameters were associated with the tolvaptan response, including BUN (6, 8), BUN/Cre (16), CRP (17), urine Na/K (18) ratio, urine osmolarity (9), urine AQP-2 (9, 17). As it was difficult to collect urine samples in this multicenter study, we could only compare BUN and BUN/Cre with previous reports. BUN showed a stronger correlation compared with BUN/Cre (data not shown); however other urinary parameters could not be evaluated.
It is important to identify tolvaptan non-responders to avoid a lengthy period of futile treatment. The PPV of the response to tolvaptan based on BUN was 0.86, suggesting that BUN is a good predictive marker for the response to tolvaptan. However, because the NPV of the response to tolvaptan based on BUN was only 0.30, it would be difficult to predict non-responders to tolvaptan using BUN alone. The NPV of the response to tolvaptan based on the combination of BUN and SVEP1 was 0.63, indicating that this parameter was more accurate for predicting non-responders, but not sufficiently so for clinical practice. We next proposed an additional prediction model based on logistic regression analysis of a population with the rs2991364 risk allele. The failure probability based on Na and BUN was identified for detecting non-responders. Assuming a cutoff of failure probability of 38.6%, sensitivity was 84.4%, specificity was 70%, and AUC was 0.774, suggesting that patients with the risk allele and a > 38.6% probability of failure would be expected to have a poor treatment response. This may a good marker for identifying non-responders to tolvaptan and avoid a lengthy period of futile treatment.
We propose the following strategy for the use of tolvaptan in clinical practice (Figure S4). When a patient has a BW decrease after 1 week of tolvaptan treatment, tolvaptan should be continued. However, when a patient has a BW increase after 1 week of treatment and is therefore a possible non-responder, SVEP1 SNP (rs2991364) should be measured. For patients without an SVEP1 SNP risk allele, measures should be employed to improve renal blood flow by reducing diuretics or performing large volume puncture with consideration of the late tolvaptan response (19). For patients with a risk allele, the prediction score should be determined. When the predictive score is over 38.6% (27/32, 84.4%), suspension of tolvaptan treatment should be considered as well as a change to a different treatment option, such as large volume puncture.
Our bioinformatic analysis showed that the rs2991364 genotype affects the expression of the SVEP1 gene in the cerebellum, although the phenotype varies depending on the tissue. Furthermore, decreased SVEP1 expression caused a decrease in the expression of downstream genes. In particular, the expression level of the FOXC2 gene, which is known to play an important role in the development of the lymphatic vascular system (20–22), is reduced with the downregulation of SVEP1 expression. It has also been reported that FOXC2 expression is decreased through the ANGPT2 and TIE1/TIE2 receptor system in Polydor/SVEP1 mutant mice, causing severe edema (20), which is consistent with the results of our public data analysis. Our results indicate that in the rs2991364 risk allele group, tissue fluid is originally difficult to collect due to the weakness of vascular network remodeling, and suggest that rs2991364, which affects SVEP1 gene expression, may serve as an SNP marker for predicting the effect of tolvaptan in other types of edema, especially cardiac edema. A significant association between coronary artery disease and missense variants in the SVEP1 gene was reported based on a large-scale exome-wide association study (15). To elucidate the relationship between the tolvaptan response and rs2991364 allele in cardiac edema patients, further studies are required.
This study had several limitations. First, a limited number of SVEP1 SNP variants were studied. Second, the number of non-responders was relatively small and difficult to draw conclusions from. We excluded patients with a borderline response (BW change − 1.5 < to < 0kg) to tolvaptan. Third, it was not possible to evaluate urine osmolarity or urine AQP-2. Lastly, all enrolled patients in this study were Japanese and further studies are needed to confirm the data in other populations.
In conclusion, we identified an association between SVEP1 SNPs and the response to tolvaptan among patients with difficult-to-treat hepatic ascites in the Japanese population. The combination of BUN and SVEP1 SNP was predictive of the response to tolvaptan, and the use of the predictive score can further help to identify non-responders and avoid prolonged use of tolvaptan in patients who will not ultimately benefit.