A New Frailty Assessment Model Incorporating Gait Speed and Grip Strength for Multiple Myeloma and Consistency Analysis with IMWG-FI

Introduction

The International Myeloma Working Group-Frailty Index (IMWG-FI) has the ability to predict survival and treatment toxicity in newly diagnosed multiple myeloma (NDMM) patients. The utilization of activities of daily living (ADL) and instrumental activities of daily living (IADL), as the physical function component of IMWG-FI, is time-consuming and may not accurately reflect the actual assessment due to potential cultural variations. In 2020, a simplified version of frailty assessment (IFM-FI) was developed; however it compromises objectivity and predictive power. To achieve both simplicity and accuracy in assessing Chinese NDMM patients, there is a need for a more objective and applicable frailty assessment tool. Previous studies have demonstrated that gait speed and grip strength are reliable indicators of physical function. Therefore, we aimed to construct a new frailty model using gait speed and grip strength as indicators of physical functional status.

Methods:

We recruited patients aged 60 and above with NDMM who were admitted to Hebei University Affiliated Hospital from October 2021 to January 2024. These patients underwent ECOG assessment, baseline gait speed and grip strength measurements, and were stratified based on IMWG-FI and IFM-FI scores. Additionally, dynamic measurements of gait speed and grip strength were conducted at C3 and C6 in synchronization with IMWG-FI.

Based on NCCN and the important role that gait speed and grip strength play in assessing physical function, we replaced the ADL and IADL with gait speed and grip strength in IMWG-FI, whereby gait speed and grip strength are assigned 1 point each. We compared the consistency of the new model and IFM-FI with the gold standard IWMG-FI. Furthermore, we analyzed the predictive efficacy of the new model，IFM-FI and IMWG-FI for overall survival (OS) in elderly NDMM patients.

Results

The study included a total of 71 elderly NDMM patients. In the baseline IMWG-FI assessment, fit accounted for 29%, intermediate fit for 13%, and frailty for 58%; in the baseline new model assessment, fit accounted for 19%, intermediate fit for 21%, and frailty for 60%; and in the baseline IFM-FI assessment, fit accounted for 6%, intermediate fit for 20%, and frailty for 74%. The baseline new model demonstrated moderate agreement with IMWG-FI (Kappa=0.513); however, compared to the new model, the baseline frailty stratification using IFM-FI had a lower level of consistency with the IMWG-FI (Kappa=0.484).

In the third cycle, dynamic IMWG-FI and new model assessments were conducted. The IMWG-FI group exhibited 34% fit, 14% intermediate fit, and 52% frailty, while the new model group showed 32% fit, 13% intermediate fit, and 55% frailty. The consistency between the two models in the third cycle exceeded that of the baseline (Kappa=0.662).

In the sixth cycle, the ongoing dynamic IMWG-FI and new model assessments were conducted. The IMWG-FI group exhibited 37% fit, 16% intermediate fit, and 47% frailty, while the new model group showed 35% fit, 18% intermediate fit, and 47% frailty. The consistency between the two groups was good (Kappa=0.934). As the number of treatment lines increased, the IMWG-FI and new model showed increasing consistency in classifying patients based on frailty.

Considering the intermediate-fit and fit group as the non-frail cohort, further investigation was conducted to assess the impact of the new model, IFM-FI and IMWG-FI on prognostic factors. In terms of OS prediction, the median OS for IMWG-FI showed a significant difference between frailty and non-frailty: 30 months vs. not reached (κ²=3.928, P<0.05); for IFM-FI, no significant difference was observed: 30 months vs. not reached (κ²=2.158, P>0.05); while for the new model, there was a notable contrast in frailty vs. non-frailty: 23 months vs. 37 months (κ²=8.032, P<0.05). The new model demonstrates the predictive performance for OS.

Conclusions

The new model is based on the application of gait speed and grip strength instead of the complicated and subjective ADL and IADL, which has better consistency with IMWG-FI than IFM-FI. In addition to having consistency with the IMWG-FI baseline frailty assessment, the new model demonstrates significant advantages in dynamic assessment. Furthermore, in terms of prognostic performance for OS, the new model is able to effectively distinguish between frail and non-frail groups, and is worth further exploration.