Functional Assessment (FA) in Multiple Myeloma (MM) Patients Predicts Survival: Use of Most Valuable Comorbidity Scores and Functional Tests in a German Study Group (Deutsche Studiengruppe Multiples Myelom [DSMM]) Incentive

Introduction: Treatment options in Multiple Myeloma (MM) patients (pts) have substantially improved, but pts present with heterogeneous health status. Since the incidence of MM is increasing with age and the majority of pts is above 60 years, it is important to assess additional diseases and comorbidities. Measuring frailty, especially in older pts allows to identify those with an increased risk of treatment complications and impaired progression-free- (PFS) and overall-survival (OS). As MM interventions and therapy can be tailored, not only older, but also "intermediate-old" and younger MM pts may profit from such assessments. Therefore, a brief assessment of pts' functional, physiological and psycho-cognitive status was performed within this DSMM incentive, using 12 functional assessment (FA) tests and 5 established Comorbidity Indices (CI). The aim was to determine the most essential stratification for MM, to develop a meaningful, time-effective assessment arsenal for true estimation of pts' fitness and avoidance of both under- and overtreatment.

Methods: We prospectively performed a FA in 301 consecutive newly diagnosed MM pts treated between 2011 and 2017. 12 functional tests comprised of the Karnofsky Performance Status (KPS), physician- and patient-rated fitness, Activity of daily living (ADL), Instrumental Activity of daily living (IADL), pain (VAS), Geriatric Depression Scale (GDS), Mini-Mental State Examination (MMSE), Timed Up and Go Test (TUGT), malnutrition, SF-12-Qualtiy of Life (QoL) Physical (PCS) and Mental Health Composite Scale (MCS). Additionally, 5 established CI were assessed: 1. International Myeloma working group (IMWG)-frailty index, 2. Revised Myeloma Comorbidity Index (R-MCI), 3. Charlson Comorbidity Index (CCI), 4. Hematopoietic Cell Transplantation-specific Comorbidity Index (HCT-CI) and 5. Kaplan-Feinstein-Index (KFI). In order to identify most valuable functional tests and CI, Kaplan Meier estimation and Cox regression for OS with backward variable selection were performed in order to determine which tests provided additional value for both IMWG-frailty index and R-MCI.

Results: Pt characteristics were typical for tertiary centers. Median OS from the time point of assessment and median follow-up were 58 and 19 months, respectively. CI that grouped pts into significantly different risk groups for OS were the IMWG-frailty index, R-MCI and CCI, whereas both HCT-CI and KFI remained insignificant. Those CI that differentiated best between different risk groups were the IMWG-frailty index and R-MCI. 3-year-OS for good-, intermediate and high-risk pts using the R-MCI was 90%, 74% and 43% (p=0.0006) and via IMWG-frailty index 92%, 83% and 57% (p=0.0001), respectively. The 3-year-OS for good- vs. high-risk pts via CCI was 90% vs. 64% (p=0.0063). The 12 functional tests that significantly differentiated pts into different risk groups were the KPS with 3-year-OS for good- vs. high-risk pts of 89% vs 60% (p=<0.0001), physician-rated fitness of 80% vs. 66% (p=0.0085), patient-rated fitness of 82% vs. 64% (p<0.0036), ADL of 80% vs. 66% (p=0.0159), IADL of 78% vs. 48% (p=0.0036), pain of 78% vs. 66%, MMSE of 78% vs. 47% (p=0.0001), TUGT of 79% vs. 63% (p=0.0168) and SF-12 PCS of 86% vs. 66% (p=0.0091). GDS, malnutrition and SF-12 MCS tests proved insignificant. Backward selection with both IMWG-frailty index and R-MCI forced therein revealed 4 other tests to significantly complement these CI, namely the KPS, ADL, MMSE and SF-12 PCS.

Conclusion: Our results suggest that our abbreviated FA, consisting of 2 CI (IMWG-frailty index and R-MCI) and 4 functional tests (KPS, ADL, MMSE, SF-12 PCS) constitutes a feasible, time-effective and highly diagnostic test set to estimate fitness and vulnerability in MM. Next steps include the automatic inclusion of the abbreviated assessment results into our MM-tumor board tool and pts' reports for high availability in medical discussions and therapy selection. Rather than focusing on pts' chronological age alone, our analysis aims to ultimately facilitate treatment decision and allow individualized treatment.