Daratumumab Based Response Adapted Therapy for Older Adults with Newly Diagnosed Multiple Myeloma: Final Results of a Phase II Study

Introduction

The combination of daratumumab (DARA), lenalidomide (Len) and dexamethasone (Dex) is an accepted therapy for older adults with newly diagnosed multiple myeloma (NDMM). A less intense bortezomib (Bort) lenalidomide dexamethasone combination (RVD lite) is also used. We thought to investigate a personalized daratumumab-based response adapted therapy for older adults with NDMM.

Methods

NDMM patients over age 65 and not deemed candidates for transplant received DARA Dex for 2 cycles: in the event of PR or better, patients continued DARA Dex per usual dose and schedule, in the event of less than a PR, frailty adjusted dosing of Len or Bort was added to the DARA Dex. The choice of Len or Bort was per the treating physician and informed by Ex Vivo drug sensitivity assay. Modified geriatric assessment (IMWG frailty, timed up and go) and bone marrow biopsy (for correlatives) were performed at screening as well as on C2D22. The primary endpoint was best response per IMWG criteria. Secondary endpoints included PFS, ORR after 2 cycles of DARA Dex.

Results

Between 11/2019 and 3/2023, 32 patients were enrolled. The median age was 77.5 years (range 66-88) and 11 (34%) were older than 80 years. The proportion of patients with R-ISS stage I, II and III was 15%, 66% and 19%, respectively and 5 patients (15.6%) had high risk cytogenetics (del17p, or t(4;14) or t(14;16)). Using the IMWG frailty score, 6 (19%), 12 (37%) and 14 (44%) were considered fit, unfit and frail (including 22% of patients with a IMWG score of 3 and greater). After 2 cycles of DARA Dex, 12 patients (37.5%) had a PR or better, 1 patient discontinued therapy after 1 dose of study therapy. Of the remaining 19 patients, 12 patients were treated on arm B (addition of Len) and 7 on arm C (addition of Bort). Using this response adapted approach, the ORR was 94% with 12 patients with CR/sCR and 65% in VGPR. After a median of 31 months of follow up, the median PFS for the response adapted approach was 42.5 months (95% CI 20.3- not reached). PFS was not different between arms A, B and C or by frailty status (p=0.75 and p=0.06 respectively). Patients with del 1p had a worse PFS (log rank p value=0.046). Median OS was not reached (2 year OS was 100%). No unexpected toxicity or grade 4 adverse event was noted. Overall, 2 patients (6%) experienced grade 3 diarrhea and 2 patients had grade 3 lymphopenia only noted on the Len arm. The most common all grade AE were hyperglycemia in 22 patients (69%), diarrhea in 12 patients (38%), fatigue, insomnia, and lymphopenia in 11 patients (34%) each, neutropenia and edema in 8 patients (25%) each, and hypokalemia in 7 patients (22%). Peripheral neuropathy was noted in 4 patients (1 grade 2); 3 occurred in the Bort arm.

Correlatives

DARA AUC in our ex vivo assay was lower in the arm A (responders to DARA) than arm B and C p=0.036. Using single cell RNA sequencing (RNAseq, n= 11) and multiparameter flow cytometry (MPF, n = 32), we were able to examine tumor and immune tumor microenvironment (iTME) factors associated with response to DARA. Arm A tumors have increased PC phenotype, translation, and interferon response; whereas, arm B&C tumors have increased B cell memory phenotype, immune evasion mechanisms, adhesion, iTME interactions, TNF response, and defective apoptosis at baseline. CD38 expression was higher at baseline (1.7-fold) and expressed in a larger proportion of cells in arm A samples. During the initial two cycles of daratumumab, arm A tumors and iTME cells began to adopt acquired resistance phenotypes similar to de novo resistance phenotypes observed in arms B&C at baseline. Analysis of contracting and expanding intratumoral clones demonstrated similar phenotypes to arm A and arm B&C, respectively. Examination of the iTME by scRNAseq revealed that arm B&C samples were enriched for the immunosuppressive populations Treg, pTreg, and mMDSC. In contrast, the cytotoxic populations, Neutrophils, Macrophages, CD4_EM, CD8_EM and NK cells were increased in arm A. Tumor-iTME communication was assessed by CellPhoneDB and demonstrated arm A interactions included more cytotoxic cells compared to arms B&C which had increased immunosuppressive cell interactions.

Conclusion

This response adapted therapy resulted in an ORR of 94% and a median PFS of 42 months in this mostly frail NDMM population with 37% of patients receiving primary therapy with DARA Dex and a very tolerable side effect profile. Further, our data indicate that both tumor and iTME features dictate response to DARA.