Exploration of Tumor Biopsy Gene Signatures to Understand the Role of the Tumor Microenvironment in Outcomes to Lisocabtagene Maraleucel

In the TRANSCEND NHL 001 study, 53% of patients with relapsed/refractory large B-cell lymphoma (LBCL) treated with lisocabtagene maraleucel (liso-cel) achieved a complete response (CR). To determine characteristics of patients who did and did not achieve a CR, we examined the tumor biology and microenvironment from lymph node tumor biopsies. LBCL biopsies from liso-cel-treated patients were taken pretreatment and ∼11 days posttreatment for RNA sequencing (RNA-seq) and multiplex immunofluorescence (mIF). We analyzed gene expression data from pretreatment biopsies (N = 78) to identify gene sets enriched in patients who achieved a CR to those with progressive disease. Pretreatment biopsies from month-3 CR patients displayed higher expression levels of T-cell and stroma-associated genes, and lower expression of cell-cycle genes. To interpret whether LBCL samples were "follicular lymphoma (FL)-like," we constructed an independent gene expression signature and found that patients with a higher "FL-like" gene expression score had longer progression-free survival (PFS). Cell of origin was not associated with response or PFS, but double-hit gene expression was associated with shorter PFS. The day 11 posttreatment samples (RNA-seq, N = 73; mIF, N = 53) had higher levels of chimeric antigen receptor (CAR) T-cell densities and CAR gene expression, general immune infiltration, and immune activation in patients with CR. Further, the majority of T cells in the day 11 samples were endogenous. Gene expression signatures in liso-cel-treated patients with LBCL can inform the development of combination therapies and next-generation CAR T-cell therapies.

Lisocabtagene maraleucel (liso-cel) treatment of patients (pts) with relapsed/refractory (R/R) B-cell non-Hodgkin lymphoma (NHL) and secondary CNS lymphoma: Initial results from TRANSCEND NHL 001

7515

Background: No clinical studies have yet evaluated CAR T cell therapies in pts with R/R B-cell NHL who have secondary CNS lymphoma. We report data from this pt subgroup receiving liso-cel (JCAR017), an investigational, anti-CD19 CAR T cell product administered as a defined composition of CD4+/CD8+ CAR T cells, in the phase 1 TRANSCEND NHL 001 study. Methods: Eligible pts had confirmed B-cell NHL with R/R disease after ≥2 prior lines of therapy. Pts with secondary CNS lymphoma could enroll or, if it developed on study, could continue to receive liso-cel. After lymphodepleting chemotherapy, liso-cel was administered at 1 of 2 dose levels (DL): DL1 = 50 × 106 or DL2 = 100 × 106 total CAR+ T cells. Efficacy was evaluated per the Lugano criteria. Pts achieving a complete response could be retreated with liso-cel upon progressive disease. Results: At data cutoff, 9 pts with secondary CNS lymphoma at initial treatment (n = 6), retreatment (n = 2), or cycle 2 (n = 1) received liso-cel. 4 pts were treated at DL1 and 5 at DL2. The median (range) age was 60 (47‒73) years and number of prior lines of therapy was 3 (2‒7). Median time to peak CAR+ T cell expansion was 12.5 (7–112) days. 1 of 9 pts had grade (G)2 cytokine release syndrome (CRS) and 1 of 9 pts had a neurological event (NE; G3 decreased level of consciousness). No retreatment pts had CRS or NE; however, 1 retreatment pt had an NE of G2 temporal edema with initial treatment with liso-cel. 5 pts received prophylactic levetiracetam. 1 pt received corticosteroids and tocilizumab. Other toxicities were predominantly cytopenias. There were no treatment-related deaths. 4 pts responded to liso-cel; all had a best response of complete response, of which 2 are ongoing at 270 and 545 days post-liso-cel. All 4 responses occurred after initial liso-cel treatment; no retreated pts responded. Conclusions: In the ongoing TRANSCEND NHL 001 study, liso-cel continues to demonstrate the ability to be safely delivered to pts with R/R B-cell NHL, including those with secondary CNS lymphoma, a population of pts with a highly unmet medical need. No excess NE was noted in this population. This cohort continues to be evaluated. Clinical trial information: NCT02631044.

Safety and preliminary efficacy in patients (pts) with relapsed/refractory (R/R) mantle cell lymphoma (MCL) receiving lisocabtagene maraleucel (Liso-cel) in TRANSCEND NHL 001

7516

Background: Most pts with MCL relapse after first-line immunochemotherapy, with poor responses to salvage therapy. We report initial dose-finding results from pts with R/R MCL treated with liso-cel (JCAR017), an investigational, anti-CD19 CAR T cell product administered as a defined composition of CD4+/CD8+ CAR T cells, in the ongoing phase 1 TRANSCEND study. Methods: Eligible pts had confirmed MCL (cyclin D1 expression, t[11;14]) with R/R disease after ≥1 prior lines of therapy. After lymphodepleting chemotherapy, liso-cel was administered at 1 of 2 dose levels (DL): DL1 = 50 × 106 or DL2 = 100 × 106 total CAR+ T cells. Results: At data cutoff, 9 pts (DL1, n = 6; DL2, n = 3) had received liso-cel. The median (range) age was 66 (58‒78) years; 7 pts were male. Histologies included blastoid (n = 3) and pleiomorphic (n = 1) variants. 8 pts had documented Ki67 > 30% (40%‒80%); 1 pt had TP53 mutation. Pts had received a median of 5 (3‒7) prior therapies; 3 pts had received prior hematopoietic stem cell transplant. All 9 pts had prior ibrutinib; 4 had a best response of progressive disease on ibrutinib. 6/9 pts (67%) received bridging chemotherapy. 4/9 pts (44%) had serious treatment-emergent adverse events (TEAEs). 5/9 pts (56%) had grade (G) 3/4 TEAEs, primarily anemia, neutropenia, and hypophosphatemia (22% each). 3/9 pts (33%) had cytokine release syndrome (CRS); all were G1. Median time to CRS onset was 6 (2‒7) days; median time to resolution was 6 (2‒6) days. 1 pt received tocilizumab and corticosteroids. There were no neurological events. 4 pts died, all in DL1 (3 from disease progression; 1 after receiving new anticancer therapy post liso-cel). Overall response rate was 78% (7/9 pts; 4/6 in DL1, median follow-up 12.4 [95% CI: 9.2–12.4] mo; 3/3 in DL2, median follow-up 1.4 [95% CI: 1.0–1.4] mo). 2 pts in DL1 maintained a durable CR until last follow-up (day 281 and 378, respectively). Median time to peak CAR+ T cell expansion: 9.5 (9–10) days at DL1 and 17.5 (10–27) days at DL2. Conclusions: In this phase 1 study in pts with R/R MCL, liso-cel treatment showed tolerable toxicity and had clinical activity. Updated DL2 data and longer follow-up will be presented. Clinical trial information: NCT02631044.

The Broad Anti-AML Activity of the CD33/CD3 BiTE Antibody Construct, AMG 330, Is Impacted by Disease Stage and Risk

The CD33/CD3-bispecific T-cell engaging (BiTE) antibody construct, AMG 330, potently lyses CD33+ leukemic cells in vitro. Using specimens from 41 patients with acute myeloid leukemia (AML), we studied the factors that might contribute to clinical response or resistance. For this purpose, thawed aliquots of primary AML samples were immunophenotypically characterized and subjected to various doses of AMG 330 in the presence or absence of healthy donor T-cells. After 48 hours, drug-specific cytotoxicity was quantified and correlated with CD33 expression levels, amounts of T-cells present, and other disease characteristics. AMG 330 caused modest cytotoxicity that was correlated with the amount of autologous T-cells (P = 0.0001) but not CD33 expression, as AMG 330 exerted marked cytotoxic effects in several specimens with minimal CD33 expression. With healthy donor T-cells added, AMG 330 cytotoxicity depended on the drug dose and effector:target (E:T) cell ratio. High cytotoxic activity was observed even with minimal CD33 expression, and AMG 330 cytotoxicity and CD33 expression correlated only at high E:T cell ratio and high AMG 330 doses (P<0.003). AMG 330 resulted in significantly higher cytotoxicity in specimens from patients with newly diagnosed AML than those with relapsed/refractory disease despite similar levels of CD33 on myeloblasts. AMG 330 cytotoxicity also appeared greater in specimens from patients with favorable-risk disease as compared to other specimens. Together, our data demonstrate that AMG 330 is highly active in primary AML specimens across the entire disease spectrum, while suggesting the presence of yet undefined, CD33-independent, relative resistance mechanisms in specific patient subsets.

Preclinical development of AMG 139, a human antibody specifically targeting IL-23

BACKGROUND AND PURPOSE

AMG 139 is a human anti-IL-23 antibody currently in a phase II trial for treating Crohn's disease. To support its clinical development in humans, in vitro assays and in vivo studies were conducted in cynomolgus monkeys to determine the pharmacology, preclinical characteristics and safety of this monoclonal antibody.

EXPERIMENTAL APPROACH

The in vitro pharmacology, pharmacokinetics (PK), pharmacodynamics and toxicology of AMG 139, after single or weekly i.v. or s.c. administration for up to 26 weeks, were evaluated in cynomolgus monkeys.

KEY RESULTS

AMG 139 bound with high affinity to both human and cynomolgus monkey IL-23 and specifically neutralized the biological activity of IL-23 without binding or blocking IL-12. After a single dose, linear PK with s.c. bioavailability of 81% and mean half-life of 8.4-13 days were observed. After weekly s.c. dosing for 3 or 6 months, AMG 139 exposure increased approximately dose-proportionally from 30 to 300 mg·kg(-1) and mean accumulation between the first and last dose ranged from 2- to 3.5-fold. Peripheral blood immunophenotyping, T-cell-dependent antigen responses and bone formation markers were not different between AMG 139 and vehicle treatment. No adverse clinical signs, effects on body weight, vital signs, ophthalmic parameters, clinical pathology, ECG, organ weights or histopathology were observed in the monkeys with the highest dose of AMG 139 tested (300 mg·kg(-1) s.c. or i.v.).

CONCLUSIONS AND IMPLICATIONS

The in vitro pharmacology, PK, immunogenicity and safety characteristics of AMG 139 in cynomolgus monkeys support its continued clinical development for the treatment of various inflammatory diseases.

Mutations affecting the SAND domain of DEAF1 cause intellectual disability with severe speech impairment and behavioral problems

Recently, we identified in two individuals with intellectual disability (ID) different de novo mutations in DEAF1, which encodes a transcription factor with an important role in embryonic development. To ascertain whether these mutations in DEAF1 are causative for the ID phenotype, we performed targeted resequencing of DEAF1 in an additional cohort of over 2,300 individuals with unexplained ID and identified two additional individuals with de novo mutations in this gene. All four individuals had severe ID with severely affected speech development, and three showed severe behavioral problems. DEAF1 is highly expressed in the CNS, especially during early embryonic development. All four mutations were missense mutations affecting the SAND domain of DEAF1. Altered DEAF1 harboring any of the four amino acid changes showed impaired transcriptional regulation of the DEAF1 promoter. Moreover, behavioral studies in mice with a conditional knockout of Deaf1 in the brain showed memory deficits and increased anxiety-like behavior. Our results demonstrate that mutations in DEAF1 cause ID and behavioral problems, most likely as a result of impaired transcriptional regulation by DEAF1.

Evidence for Endotoxin Contamination in Plastic Na+-Heparin Blood Collection Tube Lots

BACKGROUND

Biomarker assays are often conducted on whole blood samples in the course of drug development studies. Because bacterial lipopolysaccharide (LPS) (endotoxin) contamination is known to cause spontaneous cytokine production by monocytes, contamination of blood collection tubes may interfere with biomarker assay results.

METHODS

Whole blood from healthy donors was collected into plastic or glass sodium (Na+)-heparin Vacutainer™ blood collection tubes and heparinized syringes. Samples were analyzed for phosphoprotein response, cytokine production, and RNA expression. Tubes were tested for endotoxin contamination by use of the limulus amoebocyte lysate assay.

RESULTS

Results of phospho-flow cytometry, branched DNA (bDNA), and ELISA assays indicated that a specific lot (#5339582) of plastic Na+-heparin Vacutainer tubes was highly contaminated with an endotoxinlike substance, and contamination was confirmed by the limulus amoebocyte lysate assay. Analysis of multiple-analyte panels revealed that analytes whose changed expression was predictive of LPS stimulation were increased when whole blood was incubated in contaminated tubes for 6 or 18 h. Two additional lots of plastic tubes tested had detectable amounts of endotoxin sufficient to strongly alter phospho-flow cytometry analyses, as determined by the fold change in phosphorylation of p38 mitogen-activated protein kinase in response to tumor necrosis factor α and LPS. In contrast, 3 lots of glass tubes had substantially lower levels of spontaneous blood activation.

CONCLUSIONS

Endotoxin contamination associated with tubes from 3 lots of a particular type of plastic Na+-heparin Vacutainer tube dramatically affected biomarker assay measurements. Prescreening these tubes is suggested before their use in clinical sample analysis.

Dynamic anchoring of PKA is essential during oocyte maturation

In the final stages of ovarian follicular development, the mouse oocyte remains arrested in the first meiotic prophase, and cAMP-stimulated PKA plays an essential role in this arrest. After the LH surge, a decrease in cAMP and PKA activity in the oocyte initiates an irreversible maturation process that culminates in a second arrest at metaphase II prior to fertilization. A-kinase anchoring proteins (AKAPs) mediate the intracellular localization of PKA and control the specificity and kinetics of substrate phosphorylation. Several AKAPs have been identified in oocytes including one at 140 kDa that we now identify as a product of the Akap1 gene. We show that PKA interaction with AKAPs is essential for two sequential steps in the maturation process: the initial maintenance of meiotic arrest and the subsequent irreversible progression to the polar body extruded stage. A peptide inhibitor (HT31) that disrupts AKAP/PKA interactions stimulates oocyte maturation in the continued presence of high cAMP. However, during the early minutes of maturation, type II PKA moves from cytoplasmic sites to the mitochondria, where it associates with AKAP1, and this is shown to be essential for maturation to continue irreversibly.

Deletion of the RIIβ-Subunit of Protein Kinase A Decreases Body Weight and Increases Energy Expenditure in the Obese, Leptin-Deficient ob/ob Mouse

Disruption of the RIIbeta regulatory subunit of protein kinase A (PKA) results in mice with a lean phenotype, nocturnal hyperactivity, and increased resting metabolic rate. In this report, we have examined whether deletion of RIIbeta would lead to increased metabolism and rescue the obese phenotype of the leptin-deficient ob/ob (ob) mouse. Body weight gain and food consumption were decreased, whereas basal oxygen consumption and nocturnal locomotor activity were increased in the double mutant animals compared with ob mice. The ob mice are unable to maintain body temperature when placed in a cold environment due to a loss of brown adipose tissue activation, and this cold sensitivity was partially rescued by concomitant disruption of RIIbeta. These findings indicate that PKA modifies the phenotype of the leptin-deficient mouse, leading to increases in both thermogenesis and energy expenditure.