Is IMWG frailty-score-adjusted dosing the right approach to improve toxicity and outcomes in FRAIL patients? Results from the UK-MRA FiTNEss trial in newly diagnosed myeloma patients ineligible for ASCT.

Abstract Introduction Frail myeloma (MM) patients remain at risk of greater toxicity and shorter survival outcomes. Approaches to improve this by therapy adaptation include delivering different therapeutic combinations or dose adjusting components of therapy. Understanding which of these approaches to adopt is timely as 4-drug combinations are now approved for transplant ineligible (TNE) patients but were only trialled in non-frail patients. Methods The UK-MRA, Myeloma XIV FiTNEss trial (NCT03720041) is a phase III, multi-centre, randomised controlled trial for newly diagnosed TNE MM patients. The primary objectives are to compare standard and frailty-score adapted (FA) induction therapy with an oral PI/IMiD combination (ixazomib, lenalidomide, dexamethasone, IRD) for 12 cycles and, after a second randomisation (R2), to compare maintenance R to IR. Frailty status was assessed using the full IMWG frailty score. The primary endpoint, early treatment cessation (ETC) in UNFIT/FRAIL patients within 60 days of R1, was reported at ASH 2024 with no significant difference, but there was significant heterogeneity in this outcome with UNFIT patients having a benefit from FA treatment (OR 0.34 [0.16,0.72])​ not seen in FRAIL patients (OR 1.33 [0.79,2.25]). With longer follow-up we now present data exploring the pathway and outcomes of UNFIT and FRAIL patients, including updated analysis of EFS, PFS and OS. Results The FiTNEss trial randomised 733 patients from 04/AUG/20 - 01/MAR/24 at 84 UK sites. 239 (32.6%) were UNFIT and 296 (40.4%) FRAIL. In the UNFIT group the median (IQR) age was 77 years (75-78), 56.9% were male, ISS was I in 25.9%, II 43.9% and III 30.1%, 13% had ECOG >=2. In the FRAIL group the median age was 81 years (78-84), 55.4% were male, ISS was I in 14.2%, II 48.6% and III 36.8%, 43.2% had ECOG >=2. Across the primary endpoint population (UNFIT/FRAIL combined) after median follow up of 26 (IQR 16,38) months (m) median EFS was not significantly different between FA and standard dosing (FA 2m [95% CI 1,3] vs standard 1m [1,2]). PFS and OS also appeared similar (median PFS: FA 23m [19,34] vs standard 27m [21,36] HR 1.12 [0.87,1.43], p=0.372; 3yr OS: FA 69.5% [61.9,75.9] vs standard 66.4% [59.2,72.7], HR 0.87 [0.63,1.20], p=0.399). Due to the heterogeneity in the ETC primary endpoint we performed subset analysis in UNFIT and FRAIL patients for EFS, PFS and OS. In UNFIT patients FA therapy was associated with improved EFS and OS (median EFS FA 5m [95% CI 3,9] vs standard 2m [1,4]; median PFS FA 34m [20,46] vs standard 37m [26,NR]; 3yr OS: FA 84.9% [75.4,91.0] vs standard 75.2% [65.1,82.7]). FRAIL patients had inferior outcomes across all endpoints compared to UNFIT patients and did not appear to gain any benefit from FA dosing (median EFS FA 1m [0,1] vs standard 1m [0,1]; median PFS FA 19m [14,26] vs standard 21m [16,33]; 3yr OS: FA 55.2% [43.6,65.3] vs standard 58.7% [48.1, 67.8]). Most EFS events were >=G3 non-haematological toxicities (most frequently infections and skin/subcutaneous disorders). These comprised a lower proportion of events in the FA group (UNFIT 70/104, 67.3%; FRAIL 92/138, 66.7%) compared to standard (UNFIT 78/104, 75.0%; FRAIL 105/140, 75.0%). FRAIL patients were more likely to have >=G4 haem toxicity, withdrawn or died as their EFS event, and these comprised a slightly higher proportion of events in the FA group, leading to the similar EFS in this group. FRAIL patients had a shorter median duration of induction therapy even with FA dosing (UNFIT: FA 12 cycles [95% CI NR] vs standard 12 [10,NR]; FRAIL: FA 7 [4,10] vs standard 9 [6,11]). FRAIL patients in both arms were much less likely to reach R2 (UNFIT: FA 55%, standard 46%; FRAIL: FA 29%, standard 30%). Unacceptable toxicity was the most reported reason for withdrawal from treatment. Conclusions Results from FiTNEss suggest that ETC, EFS and OS were improved by FA dosing in UNFIT but not FRAIL patients. In FRAIL patients, shorter durations of treatment and shorter EFS/PFS/OS persist with PI/IMiD combinations even with prospective treatment modifications. Data from other trials suggest anti-CD38/IMiD combinations may be better tolerated in FRAIL patients, with improved outcomes. Our data suggest the recently approved anti-CD38/IMiD/PI combinations should be used cautiously in FRAIL patients even with FA dosing strategies. Future studies should explore other approaches such as treatment switching for FRAIL patients with a suboptimal response.