Medicine Advances (Mar 2024)

Chinese collaborative study of survival analysis in 980 patients with AL amyloidosis

  • Hokhim Yau,
  • Liye Zhong,
  • Sheng Li,
  • Weiting He,
  • Yaxi Zhu,
  • Pengjun Liao,
  • Jianteng Xie,
  • Hongwen Fei,
  • Liwen Li,
  • Hui Liu,
  • Jie Li,
  • Wenjian Wang

DOI
https://doi.org/10.1002/med4.53
Journal volume & issue
Vol. 2, no. 1
pp. 82 – 94

Abstract

Read online

Abstract Background The prognosis of patients with light‐chain amyloidosis (AL) has improved markedly in the past decade in China; whether the current staging systems are suitable to predict the overall survival (OS) of the patients remains undetermined. Methods Based on 980 biopsy‐proved AL patients with 5‐year follow‐up from China Registration Network for Light‐chain Amyloidosis, we evaluated the efficacy of existing staging systems and developed a new stratification model. This involved analyzing parameters such as N‐terminal pro‐brain natriuretic peptide (NT‐proBNP) thresholds and estimated glomerular filtration rate (eGFR). Results We found that 30% patients were classified as stage I, 25% as stage II, 26% as stage III, and 19% as stage IV disease using the Mayo 2012 staging system, with varying median OS values. However, the observed median OS values were notably higher than previously reported. By incorporating NT‐proBNP thresholds and eGFR values, we developed a four‐stage score system. With this new model, 41.6% patients were reclassified to stage I, 34.3% to stage II, 17.8% to stage III, and 6.3% to stage IV disease, with adjusted median OS values. Conclusion The new stratification model exhibited improved consistency compared to traditional staging systems. It effectively identified patients with the best and worst prognoses, even among those receiving comprehensive treatment. Specifically, NT‐proBNP levels exceeding 9000 pg/mL combined with an eGFR less than 60 mL/min/1.73 m2 proved superior in prognosticating patient outcomes. Overall, the median OS of AL amyloidosis patients in China has significantly improved, underscoring the need for tailored prognostic models in clinical practice.

Keywords