Bargou, Ralf C.

Key Points MRD response has value as a prognostic factor for blinatumomab treatment in R/R B-cell precursor ALL. MRD response was associated with better outcomes in terms of OS and RFS in blinatumomab-treated R/R ALL.

Abstract Background: Blinatumomab is an investigational bispecific T-cell engager (BiTE®) antibody construct designed to direct cytotoxic T cells to CD19-expressing B cells. In a study in minimal residual disease (MRD)-positive ALL, 80% of pts achieved MRD-negative status after receiving one treatment cycle of blinatumomab (Topp MS, et al. J Clin Oncol. 2011;29:2493-8). For pts with newly diagnosed ALL, persistence of MRD after first-line treatment predicts disease recurrence. However, only limited data are available about the relationship between MRD and outcome in adult pts with r/r ALL. Therefore, the aim of this secondary analysis was to evaluate MRD responses and their relationships to other efficacy indices in adult pts with r/r ALL receiving blinatumomab in this phase 2 study. Methods: Eligibility criteria included adult pts ≥18 years, with Philadelphia chromosome negative B-cell r/r ALL who were primary refractory or refractory to first salvage, had their first relapse within 12 months of first remission or allogeneic hematologic stem cell transplantation (HSCT), or received second or later salvage. Blinatumomab was dosed via continuous IV infusion at 28 μg/d (cycle 1 only: 9 μg/d on days 1-7, 28 μg/d on days 8-28) for up to five cycles (one cycle: 4 weeks on, 2 weeks off). The primary endpoint was complete remission (CR: ≤5% blasts with platelets >100,000/µL, absolute neutrophil count >1000/µL) or CR with partial hematologic recovery (CRh*: platelets >50,000/µL, absolute neutrophil count >500/µL) within the first two treatment cycles. MRD in bone marrow was assessed at a central laboratory using allele-specific real-time quantitative PCR. Two MRD-based exploratory endpoints were analyzed: MRD response (no PCR amplification at a minimum sensitivity of 10-4 or <10-4 leukemic cells by PCR) and complete MRD response (no PCR amplification at a minimum sensitivity of 10-4) within the first two treatment cycles. Results: Of 189 treated pts (median [range] age = 39 [18–79] years), 81 (43%) pts had a CR (n = 63, 33%) or a CRh* (n = 18, 10%) during the first two treatment cycles. In the subgroup of pts with a CR/CRh* and evaluable MRD data (n = 73), 60 (82%) pts had an MRD response. Of those, 51 (70%) pts had a complete MRD response. The majority of those who achieved a CR/CRh* also achieved a complete MRD response regardless of the number of prior lines of salvage therapy (Table 1). Of all treated pts, 64 (34%) pts had prior HSCT. In pts who achieved CR/CRh*, the rate of MRD response was 81% in pts without and 85% in pts with prior HSCT. In pts who achieved CR/CRh*, the median (95% CI) duration of overall survival was 11.4 (8.5, not estimable) months for those with a MRD response and 6.7 (2.0, not estimable) months for those with no MRD response. The median (95% CI) duration of relapse-free survival was 6.9 (5.5, 10.1) months in patients with a MRD response and 2.3 (1.2, not estimable) months in patients with no MRD response. The HSCT realization rate did not differ significantly between pts with CR/CRh* and a MRD response and those with CR/CRh* and no MRD response (31% and 43%, respectively). Seventeen pts classified as nonresponders per protocol had ≤5% blasts and no evidence of disease, but recovery of peripheral counts did not meet criteria for CRh*. Ten of these 17 pts had evaluable MRD assessments; 50% (5/10) had MRD response and 20% (2/10) had complete MRD response. The clinical significance of this finding remains open. Grade ≥3 adverse events regardless of causality and occurring in >10% of all treated pts during treatment until 30 days after treatment were febrile neutropenia (25%), neutropenia (16%), and anemia (14%). Grade 3 cytokine release syndrome (CRS) was reported in three (1.6%) patients. Three (1.6%) patients had grade 4 neurologic events, all resolved. Twenty-one (11%) patients had grade 3 neurologic events, 17 were resolved, with 4 events unresolved at death or unknown. Conclusion: In this largest study to date of adult pts with r/r ALL with PCR-based central assessment of MRD, pts who achieved CR/CRh* during single-agent treatment with blinatumomab had high MRD response rates. The exploratory analyses suggested that within pts with a CR/CRh*, those who did not achieve MRD responses tended to have shorter durations of overall and relapse-free survival. These results are consistent with prior reports highlighting the importance of achieving an MRD response in first-line therapy. Disclosures Goekbuget: Amgen Inc.: Consultancy, Honoraria, Research Funding. Off Label Use: This presentation will discuss the off-label use of blinatumomab, as this agent is not approved for use by the FDA, EMA or any other regulatory authorities. Kantarjian:Amgen Inc.: Research Funding. Brüggemann:Amgen Inc.: Consultancy, Research Funding. Stein:Amgen Inc.: Membership on an entity's Board of Directors or advisory committees. Bargou:Amgen Inc.: Consultancy, Honoraria. Dombret:Amgen Inc.: Research Funding. Heffner:Amgen Inc.: Honoraria, Research Funding; Biotest: Research Funding; Dana Farber CI: Research Funding; Genentech: Research Funding; Gilead: Research Funding; Idera: Research Funding; Janssen: Research Funding; Pfizer: Research Funding; Pharmacyclics: Honoraria, Research Funding; Onyx: Research Funding; Spectrum: Research Funding; Talon Therapeutics: Research Funding. Rigal-Huguet:Amgen, Inc.: Consultancy; Ariad: Consultancy; BMS: Consultancy; Novartis: Consultancy; Pfizer: Consultancy. Litzow:Amgen Inc.: Honoraria, Research Funding. Zugmaier:Amgen Reseach (Munich) GmbH: Employment; Amgen Inc.: Equity Ownership. Jia:Amgen Inc.: Employment, Equity Ownership. Maniar:Amgen Inc.: Employment, Equity Ownership. Huber:Amgen Research (Munich) GmbH: Employment; Amgen Inc.: Equity Ownership. Nagorsen:Amgen Inc.: Employment, Equity Ownership, Related to blinatumomab Patents & Royalties. Topp:Amgen Inc.: Honoraria, Membership on an entity's Board of Directors or advisory committees.

Abstract Introduction: Current therapies for patients with r/r ALL who have had prior allogeneic hematopoietic stem cell transplantation (alloHSCT) have very poor outcomes. Improvements in the therapeutic options available for adult r/r ALL are required. Blinatumomab is a bispecific T-cell engager (BiTE®) antibody construct that redirects cytotoxic T cells to lyse CD19-positive B cells. The aim of the present analysis was to characterize a subset of patients with r/r ALL and prior alloHSCT before treatment with blinatumomab from a large confirmatory open-label, single-arm, multicenter phase 2 study (Topp MS et al Lancet Oncol 2015;16(1):57-66). Methods: Eligible patients (≥18 years) had Philadelphia chromosome-negative r/r ALL with 1 of the following negative prognostic factors: primary refractory, first relapse within 12 months of first remission, relapse within 12 months of alloHSCT, or second or greater salvage. Patients with active acute or chronic graft-versus-host disease (GvHD) were excluded. Patients were required to stop all immunosuppressive GvHD therapy within 2 weeks before starting blinatumomab. A total of 189 patients were enrolled and received blinatumomab by continuous intravenous infusion (4 weeks on/2 weeks off) for up to 5 cycles. The primary endpoint was complete remission or complete remission with partial hematologic recovery (CR/CRh) within the first 2 cycles. Secondary endpoints included overall survival (OS), relapse-free survival (RFS), and adverse events (AEs). Results: 64 (34%) patients had received alloHSCT before study enrollment; 10 patients had 2 prior alloHSCTs. Donor types primarily included 29 (45%) matched sibling and 31 (48%) unrelated, with 34 (59%) patients receiving myeloablative conditioning regimens (donor chimerism data were unavailable). Among those with prior alloHSCT, median age (range) was 32 (19-74) years. At baseline, 23 (36%) patients had 1 prior relapse, 24 (38%) had 2 prior relapses, and 17 (27%) had ≥3 prior relapses; 28 (44%) patients had relapsed post-alloHSCT. Of the 55 patients who had received previous salvage therapy, 38 (69%) had received salvage therapy after last alloHSCT and prior to blinatumomab. Median time (range) between the last alloHSCT and subsequent relapse was 6 (1-33) months. Median time from last prior alloHSCT to first dose of blinatumomab was 10 (3-40) months. Nineteen (30%) patients had a history of GvHD, and 42 (66%) had ≥50% bone marrow blasts at start of treatment as assessed by a central laboratory. Patients received blinatumomab for a median of 2 (1-5) cycles. Efficacy data are presented in Table 1. Overall, 45% (29/64; 95% confidence interval [CI], 33-58) of patients achieved CR/CRh within the first 2 cycles, with similar rates of remission also observed in the alloHSCT-naïve (42%; 52/125) group. With a median follow-up of 8.8 months, median (95% CI) OS was 8.4 (4.2-9.4) months for the 64 patients with prior alloHSCT treated with blinatumomab. Of the 29 responders (CR, n=18 and CRh, n=11), 22 (76%) had a minimal residual disease (MRD) response and 19 (66%) achieved a complete MRD response. Median RFS (95% CI) was 6.1 (5.0-7.7) months. 9/29 (31%) responders subsequently underwent another alloHSCT. In total, 56 (88%) patients had grade ≥3 treatment-emergent AEs, with the most frequent including neutropenia (22%), febrile neutropenia (20%), anemia (17%), and thrombocytopenia (14%). Six patients reported treatment-emergent GvHD (two grade ≥ 3) during blinatumomab treatment, 3 of whom had GvHD in skin. Eight patients had fatal treatment-emergent AEs, which included 1 due to gastrointestinal hemorrhage, 1 due to respiratory failure, and 6 due to infection/infestation; 1 of these (candida infection) was considered to be possibly related to treatment by the investigator. Of the subjects who had treatment-emergent fatal AEs, none were in remission at the time of death. Summary: In this heavily pretreated group of patients with r/r ALL and prior alloHSCT, single-agent blinatumomab was able to induce a CR/CRh rate of 45%, with an AE profile consistent with that previously reported. Post-alloHSCT patients who had relapsed performed equally as well as those without prior alloHSCT. To prolong remission in this poor outcome patient group, the addition of other immunotherapies to the treatment regimen may be considered for future investigations. Disclosures Stein: Amgen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Seattle Genetics: Research Funding. Topp:Astra: Consultancy; Regeneron: Consultancy; Affimed: Consultancy, Research Funding; Roche: Consultancy, Other: Travel Support; Jazz: Consultancy; Pfizer: Consultancy; Amgen: Consultancy, Honoraria, Other: Travel Support. Goekbuget:Erytech: Consultancy; Gilead Sciences: Consultancy; Kite: Consultancy; Mundipharma: Consultancy, Honoraria, Research Funding; Novartis: Consultancy, Honoraria, Research Funding; Bayer: Equity Ownership; Sanofi: Equity Ownership; GlaxoSmithKline: Honoraria, Research Funding; Bristol-Myers Squibb: Honoraria; Amgen: Consultancy, Honoraria, Research Funding; Eusapharma/Jazz: Consultancy, Honoraria, Research Funding; Medac: Consultancy, Honoraria, Research Funding; Pfizer: Consultancy, Honoraria, Research Funding; SigmaTau: Consultancy, Honoraria, Research Funding. Bargou:Pfizer: Consultancy, Honoraria; Amgen: Consultancy, Honoraria, Patents & Royalties: Patent for blinatumomab; University of Wuerzburg, Germany: Employment; GEMoaB GmbH: Consultancy, Honoraria; Novartis: Consultancy, Honoraria. Rambaldi:Roche: Honoraria; Novartis: Honoraria; Amgen: Honoraria; Celgene: Research Funding; Pierre Fabre: Honoraria. Zhang:Amgen: Employment, Equity Ownership. Zimmerman:Amgen: Employment, Equity Ownership. Forman:Mustang: Research Funding; Amgen: Consultancy.

Abstract Introduction: Proteasome inhibition is the backbone of various Multiple Myeloma (MM) treatment regimens, leading to durable responses and high quality remissions. However, under prolonged therapy patients eventually develop drug resistance and the underlying mechanisms have been poorly understood. Proteasome inhibitor resistant cell lines were generated by continuous exposure to proteasome inhibiting drugs. In these cell line models a number of variants within the PSMB5 gene were observed. This includes single point mutations, leading to conformational changes of the _5 subunit within the 20S proteolytic core of the proteasome, impairing its chymotryptic catalytic function and the binding of inhibitory drugs. However, no PSMB5 mutations could, so far, be identified in human disease, leaving the functional relevance of such mutations to be determined. Methods: We applied the MM Mutation Panel (M3P) and used the Personal Genome Machine (Life Technologies) to sequence CD138+ purified bone marrow plasma cells and peripheral blood germ line control from a MM patient in third relapse that had been previously treated by various proteasome inhibitor containing therapies (VTD-PACE, VCD, PAD-Rev). This patient was subsequently treated with a pomalidomide, bortezomib, adriamycin and dexamethasone combination therapy (Pom-PAD) and achieved a partial remission after 4 cycles of therapy. When being exposed to VCD at earlier relapse a complete remission was induced within 6 cycles of therapy, demonstrating excellent response to proteasome inhibition at this earlier disease stage. Results: We identified clonal missense mutations in this patient at known NRAS hotspot location (p.Gln61Arg) and in MAX (p.Arg35His). Most notably we found and validated four subclonal single point mutations in PSMB5, each of them in subclonal frequencies with variant reads (VR) ranging from 1.9%- 5.9% (average read depth 750X). All PSMB5 mutations occurred in a highly conserved region in exon 2 of the gene (p.Cys122Tyr, p.Met104Ile, p.Ala86Pro, p.Ala79Thr), with three of them being located within the proteasome inhibitor drug binding site. Our 400bp amplicon design allowed us to observe that each mutation identified in PSMB5 is exclusively present on a different sequencing read and no reads are shared between the mutations. This implies that the mutations are present on different subclones of the tumor, which means that, despite low VR frequencies of the single mutation, in sum more than a quarter of the whole tumor might be affected by mutated PSMB5. At a disease stage when the patient was well responsive to proteasome inhibitor treatment (at diagnosis and at first relapse), of note, none of the mutations in PSMB5 is detectable. Conclusion: Here we report the first in human PSMB5 mutation in a MM patient. These mutations evolved in parallel within different subclones of the disease under the selective pressure of bortezomib- containing treatment regimens, representing branching evolution. It is to speculate whether previous investigations negating the existence of PSMB5 mutations in proteasome inhibitor treated patients might not have sequenced deep enough or did set up a sensitivity cut-off too rigid to detect such subclonal mutations. Our finding suggests that the mutations identified may contribute to the development of proteasome inhibitor resistance, emphasizing the need for more detailed genomic characterization of tumors, including minor subclonal mutation detection. Functional analysis of the mutations is ongoing and results will be presented at the meeting. Figure Figure. Disclosures No relevant conflicts of interest to declare.

BACKGROUNDNVP‐AUY922 (AUY; Luminespib) with or without bortezomib showed preclinical activity against multiple myeloma (MM) cells. This phase 1/1B study assessed NVP‐AUY922 alone and with bortezomib in patients with relapsed or refractory MM.METHODSDose escalation was guided by an adaptive Bayesian logistic regression model. In phase 1, patients who progressed after 2 to 4 prior therapies received NVP‐AUY922 intravenously once weekly. In phase 1B, patients who progressed after 2 or fewer prior therapies received NVP‐AUY922 plus bortezomib. The primary objective was to determine the maximum tolerated dose (MTD) of NVP‐AUY922.RESULTSTwenty‐four patients received NVP‐AUY922 monotherapy at doses of 8 to 70 mg/m2. One dose‐limiting toxicity (DLT) was observed (grade 3 blurred vision at 70 mg/m2); no MTD was reached. The recommended phase 2 dose was 70 mg/m2. The most frequent drug‐related adverse events (AEs) were diarrhea, nausea, and ocular toxicities. Grade 3/4 AEs were uncommon (<10%). Eight patients discontinued treatment because of AEs; 5 had ocular toxicities (≥45 mg/m2). The best response was stable disease in 66.7% of the patients. There were no partial or complete responses. Five patients received NVP‐AUY922 (which was started at 50 mg/m2) plus bortezomib (1.3 mg/m2). Three of these patients experienced DLT. No further dose escalation was performed; the MTD for NVP‐AUY922 plus bortezomib was not established.CONCLUSIONSThis study showed disease stabilization with NVP‐AUY922 in patients with relapsed or refractory MM. The MTD for NVP‐AUY922 was not reached, but reversible ocular toxicity has been reported at high dose levels. Bortezomib at the standard recommended dose plus NVP‐AUY922 was not tolerated. Cancer 2015;121:2185–2192. © 2015 American Cancer Society.