Gnri (Geriatric Nutritional Risk Index) in Multiple Myeloma

Introduction

Multiple myeloma is developed in elderly individuals, and their symptoms lead frailty. Frailty score has been reported as powerful predictive factor for overall survival (OS) in myeloma, and developed using age, activities of daily living (ADL), instrumental ADL (IADL), performance status (PS), and several factors. Meanwhile, nutritious condition might also predict OS in myeloma. Geriatric nutritional risk index (GNRI) predicted OS in not only patients with solid cancers, including lymphomas, but also elderly healthy individual. However, GNRI has never been investigated in myeloma patients. The aim of this retrospective study was investigating the clinical significance of GNRI in myeloma patients.

Methods

We reviewed the medical records of patients with newly diagnosed multiple myeloma who treated with proteasome inhibitor and/ or immunomodulatory drug containing initial treatment at the Jikei University Hospital and the Jikei University Kashiwa Hospital from January 2009 to December 2022, followed up until December 2023. GNRI is calculated by [(14.89 x serum albumin level) + (41.7 x current body weight / ideal body weight)]. The primary endpoint was the OS. The secondary endpoints were the kinetics of GNRI during treatment, and their impact for OS compared to those at baseline.

Results

Three hundred and fifty-seven patients were included in this study. The cutoff of GNRI for high-risk was 92 following by previous article. The percentage of high-risk GNRI was 46.8%. The frequencies of age over 65years, ECOG PS 2 or more, anemia defined by the CRAB criteria, ISS stage 3, high CRP level (cutoff was 1.0mg/dL), and in the high-risk GNRI group were significantly higher than those in the low-risk GNRI group. In a median followed-up time of 37.8months, OS in the high-risk GNRI group was significantly shorter than those in the low-risk GNRI group (3year-OS 66.7% and 83.1%; P >0.001). Predictive impact of GNRI high-risk was observed independently from age and the cause of death, such as myeloma disease and another cause. In multivariate analysis including PS 2 or more, kappa of free-light chain type, ISS stage3, anemia by CRAB criteria, bone disease, high CRP, high LDH (cutoff was 230U/L), simplified frailty score over 2, ASCT, and achieving complete response (CR), which were prognostic factors for OS using univariate analysis, the high-risk GNRI (HR 1.716, 95%CI 1.104 – 2.668, P = 0.016), bone disease (HR 1.745, 95%CI 1.076 – 2.831, P = 0.024), high LDH (HR 1.885, 95%CI 1.231 – 2.886, P = 0.004), and achievement of CR (HR 0.404, 95%CI 0.260 – 0.629, P < 0.001) were prognostic factor, while PS and simplified frailty score and treatment regimen including ASCT was not associated with OS. The GNRI risk was improved during treatment overtime. The frequencies of GNRI high-risk at 3, 6, and 12 months after treatment started were 40.0%, 31.3%, and 29.9%, respectively. The GNRI risk at 12months were assessed in ninety-three patients with the high GNRI risk at base line, and 50.5% of them improved from high GNRI risk to low GNRI risk. The improved GNRI risk at 12 months predicted OS after 12 months; the OS in the patients whose GNRI risk remained high at baseline and 12 months was significantly shorter than those in the GNRI low-risk at baseline group (HR 2.386, 95%CI 1.203 – 4.731, P = 0.013) while there was no significant difference of OS in the patients whose GNRI risk was low-risk at baseline and improved to low-risk at 12 months (HR 1.128, 95%CI 0.501 – 2.539, P = 0.771).

Conclusion

The GNRI at baseline predicted OS in myeloma patients, and GNRI was changing overtime during treatment, suggesting that nutritious condition might be a key to predict survival time in myeloma.