Preview

PDF, English Download (2MB) | Terms of use

Abstract

AL amyloidosis is caused by the misfolding and aggregation of proteins into highly organised amyloid fibrils, which are deposited in the tissues and lead to persistent organ damage. The disease shows a variety of disease manifestations, whereby different organs can be affected. Besides the heart, the kidneys are most frequently affected. And different to healthy individuals and multiple myeloma patients, kappa is less common than lambda and thus data on kappa AL amyloidosis are limited. Aim of this project was the comparative analysis of light chain sequences of kappa AL amyloidosis patients, especially with kidney involvement, in comparison to multiple myeloma patients. 41 kappa AL amyloidosis patients, including 14 patients with dominant kidney involvement, were consecutively enrolled and compared with 83 kappa multiple myeloma patients. Kappa light chain sequences were analysed for specific sequence pattern and their amyloidogenic potential. For this purpose, mutation analyses were performed, but also various biophysical properties such as theoretical isoelectric point, potential N-glycosylation sites, hydrophobicity (GRAVY-score) and the prediction for a β-sheet aggregation (AGG-score) were calculated. The overall aggregation potential was also analysed using an artificial intelligence-based tool. Subsequently, the findings were compared with data from lambda AL amyloidosis patients and for studying specific sequence features the implementation of a next-generation sequences analysis pipeline was another aim of this project. Results revealed a higher frequency of IGKV1 in AL amyloidosis compared to multiple myeloma patients (AL= 80 %, MM= 53 %, p= 0.002) and IGKV3 was higher in multiple myeloma (MM= 30 %, AL= 10 %, p= 0.014). IGKV1/D-33 was significantly more expressed in AL amyloidosis (32 % vs. 13 %, p= 0.016), especially in patients with heart involvement (75 %, p= 0.024). IGKV1/D-39 was more frequently in patients with kidney involvement (36 %). IGKV1-16 was only identified in the AL cohort (10 %, p= 0.010) and IGKV3/D-11 was detected as one of the most abundant subfamilies in multiple myeloma (MM= 14 % vs. AL= 2 %). IGKJ2 was primarily associated with cardiac involvement (63 %) and IGKJ1, IGKJ2 and IGKJ3 was equally presented (29 %) in AL amyloidosis patients with kidney involvement. The analysis of the most abundant representatives for AL amyloidosis subfamilies showed a higher median mutation count of the IGKV-regions of kappa AL amyloidosis compared to multiple myeloma sequences (IGKV1/D-33 (AL= 12.0 vs. MM= 10.0); IGKV1/D-39 (AL= 14.0, MM= 9.5)), which was confirm by the comparative analysis of IGKV1-16 AL amyloidosis sequences and IGKV3/D-11 multiple myeloma sequences (AL (IGKV1-16)= 9.0, MM (IGKV3/D-11)= 5.5, p= 0.033). Interestingly, the lambda AL amyloidosis sequences had a lower median mutation count in comparison to multiple myeloma (IGLV: MM (IGLV2-23)= 10.0, AL (IGLV6-57)= 4.5, p< 0.001). The analysis of the biophysical properties did not provide a clear pattern, however significantly differences were found for the isoelectric point of IGLV6-57 AL amyloidosis and IGLV2-23 multiple myeloma sequences (MM (IGLV2-23)= 10.0, AL (IGLV6-57)= 4.5, p< 0.001), whereby the isoelectric point of the IGLV2-23 multiple myeloma patients sequences (pI: IGLV2 23: 8.12 vs. pI: IGLV6-57: 5.0) was closer to the physiological pH and thus indicates greater stability. Patients with kappa AL amyloidosis had significantly more potential N-glycosylation sites than kappa multiple myeloma patients (42.5 % vs. 16.9 %, p= 0.004), but none of the lambda patients showed potential N-glycosylation sites. Systems based on artificial intelligence becoming increasingly popular in the medical sector and the analysis in this project showed also interesting results. 50 % of the IGKV1-16 AL amyloidosis sequences were classified as amyloidogenic and all the IGKV3/D-11 multiple myeloma sequences as non-amyloidogenic. In the case of IGKV1/D-33, the performance was not consistent, as all AL amyloidosis sequences were classified as amyloidogenic, but also all multiple myeloma sequences. The possibility of the prediction of an increased risk to develop AL amyloidosis would support personalised screening and treatment options. In addition to the increased abundance of IGKV1/D-33 and IGKV1-16 in AL amyloidosis, the calculation of the relative risk of these two subgroups stood out. With a rr of 2.452 (p= 0.013) for IGKV1/D-33 and 18.44 (p= 0.049) for IGKV1-16, patients showing an increased risk of developing AL amyloidosis. In conclusion, this study contributes to a better understanding of AL amyloidosis. The results on the identification of an increased risk of certain patient groups for the tendency to develop kappa AL amyloidosis could contribute to the development of an individual treatment strategy.