Introduction: Immune checkpoint inhibition targeting the PD-1/PD-L1 pathway is insufficient to induce clinical response in relapsed or refractory (R/R) multiple myeloma (MM). We postulated that combining atezolizumab (A; anti-PD-L1) with daratumumab (D; anti-CD38), which targets myeloma cells and has immunomodulatory activity, may alter the tumor microenvironment (TME) to favor cytotoxic T-cell activation and clinical activity. To assess the immunologic efficacy of this combination, we studied changes in CD8+ T cells in D-naïve and D-refractory pts from a Phase Ib study (GO29695; NCT02431208).

Methods: Flow cytometry was performed using longitudinal peripheral blood (PB) and bone marrow aspirates (BMA) to characterize CD8+ cytotoxic T cells using 8 color flow panels. RNA sequencing (RNAseq) and dual-plex immunohistochemistry (IHC) (CD138/CD8, CD8/Ki-67, CD138/osteoclast) were performed using longitudinal CD138+ fraction and bone biopsies, respectively. For IHC, CD138+ cell masses of >5000μm2 were defined as tumor clusters. Osteoclasts were enumerated based on TRAP positivity and morphology. Table 1 shows on-treatment changes in Cohorts D1-D3. Table 2 shows baseline data in Cohorts A, B, D1-D3, and E. The median (bootstrap 95% CI) is used to describe the data.

Results: 9/36 (25%) pts in cohorts D1-3 showed clinical efficacy (partial response or better); all were D-naïve. We studied CD8+ T-cell activation and proliferation (%CD8+HLA-DR+Ki-67+), the pharmacodynamic marker for A (Herbst et al. Nature 2014), in PB. All D-naive pts showed on-treatment increase in %CD8+HLA-DR+Ki-67+ cells in the periphery (C1D15-C2D1) compared to baseline, which was not observed in D-refractory pts (Table 1). In BMA, the increase in %CD8+HLA-DR+Ki-67+ (C2D15-C4D1) was observed in D-naïve pts with clinical response to A-D (sensitive), but not in non-responders (resistant) or D-refractory pts (all resistant), suggesting that sensitive pts have an immune-supportive TME. Preliminary IHC staining also showed an increase in CD8+Ki-67+ T cells in two responders after treatment. Gene enrichment analysis (RNAseq data, n=20) showed upregulation of an innate immune response signature, which appeared to be driven by a 'macrophage activation' gene signature post-treatment, in the CD138+ fraction of responders. To understand the mechanisms regulating sensitivity to treatment, we studied the spatial localization of CD8+ T cells with respect to CD138+ tumor cells by IHC. A higher density of CD8+ T cells within tumor clusters was seen at baseline in sensitive versus resistant pts, but this was not observed outside of tumor clusters (Table 1). In addition, the number of osteoclasts in the tumor region was higher in resistant pts, suggesting that these cells may contribute to the inhibition of T-cell function as reported(An et al. Blood 2016). This hypothesis was further supported by higher osteoclast numbers in D-refractory pts at baseline (Table 2), for whom an on-treatment increase in %CD8+HLA-DR+Ki-67+ cells was not observed in PB or BMA. Interestingly, higher median fluorescence intensity of PD-1 on CD8+ T-effector cells and on CD8+ T-effector memory cells was observed at baseline in D-naïve relative to D-refractory pts, while the level of PD-L1 expression on tumor cells was similar. An increase in activated proliferating T cells (%CD8+HLA-DR+Ki-67+) observed after treatment in D-naïve responders suggests that high PD-1 expression in this subset is not a marker of CD8+ T-cell exhaustion, but of functional capability.

Conclusions: Clinical efficacy of A-D therapy in R/R MM pts is associated with higher CD8+ cell density in tumor clusters and lower osteoclast numbers in the tumor region at baseline, and an on-treatment increase in activated CD8+ T-cell populations in the bone marrow. The lack of a D-monotherapy arm in this study makes it difficult to assess the individual contribution of A to T-cell activation. The data presented, albeit a small number of samples from a Phase Ib study, support the hypothesis that the TME, including CD8+ T cells, tumor cells, and cells of myeloid lineage such as osteoclasts, has significant impact on the immunologic and clinical efficacy of combination therapy. A better understanding of the complex interplay between myeloma cells and their immune environment should pave the way for designing better immunotherapies with the potential for long-term disease control.


Raval:Roche: Employment, Equity Ownership. Cho:Agenus: Research Funding; Genentech: Honoraria, Research Funding; Takeda: Research Funding; BMS: Consultancy; The Multiple Myeloma Research Foundation: Employment; Celgene: Honoraria, Research Funding; GSK: Consultancy. Green:Genentech Inc.: Employment. Wassner Fritsch:F. Hoffmann-La Roche Ltd: Employment, Equity Ownership. Ma:Genentech: Employment. Chang:Roche Canada: Employment. Yan:F. Hoffmann-La Roche Ltd, Mississauga, Canada: Employment. Kockx:HistoGeneX: Equity Ownership. Shen:Genentech, Inc.: Employment. Huw:Roche/ Genentech: Employment, Equity Ownership. Balestiere:Genentech: Employment. Lipkind:Roche/Genentech: Employment. Huang:F. Hoffmann-La Roche Ltd: Employment. Byrtek:Genentech: Employment; Roche: Equity Ownership. Colburn:Genentech: Employment; Roche: Equity Ownership. Wong:Celgene Corporation: Research Funding; Genentech: Research Funding; Janssen: Research Funding; Fortis: Research Funding; Juno: Research Funding. Venstrom:F. Hoffmann-La Roche Ltd: Employment. Adamkewicz:F. Hoffmann-La Roche Ltd: Equity Ownership; Genentech, Inc.: Employment.

OffLabel Disclosure:

Atezolizumab (atezo) is a programmed death-ligand 1 (PD-L1) blocking antibody. In the United States, atezo is approved for treatment of pts with locally advanced or metastatic urothelial carcinoma who are not eligible for cisplatin-containing chemotherapy (chemo) and whose tumors express PDââ‚â'¬Ã'ÂL1, or are not eligible for any platinumââ‚â'¬Ã'Âcontaining chemo regardless of PDââ‚â'¬Ã'ÂL1 status, or have disease progression during or following any platinum-containing chemo, or within 12 months of neoadjuvant or adjuvant chemo. Atezo is also approved: in combination with bevacizumab, paclitaxel and carboplatin for first-line treatment of pts with metastatic non-squamous non-small-cell lung carcinoma (NSCLC) with no EGFR or ALK genomic tumor aberrations, and for pts with metastatic NSCLC who have disease progression during or following platinum-containing chemo; in combination with paclitaxel protein-bound for the treatment of adults with unresectable locally advanced or metastatic triple-negative breast cancer whose tumors express PD-L1; and in combination with carboplatin and etoposide, for the first-line treatment of adults with extensive-stage small cell lung cancer. Atezo is not approved for treatment of pts with multiple myeloma.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution