TH1/17 hybrid CD4+ cells in bronchial alveolar lavage fluid from leukemia patients with checkpoint inhibitor-induced pneumonitis

204

Background: Immune checkpoint inhibitor (ICI)-based therapies are showing encouraging results for acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS). 20% of AML or MDS patients receiving an ICI develop lung inflammation (herein, pneumonitis), one of fatal immune related adverse events (irAEs). The mechanisms of pneumonitis, the most important step for risk stratification and early detection, remain elusive. Methods: We analyzed bronchial alveolar lavage (BAL) fluid from 8 AML or MDS patients, who received an ICI, developed respiratory symptoms, and underwent a standard-of-care bronchoscopy. As a control, we analyzed BAL fluid from 5 AML or MDS patients who had never received an ICI or had received last ICI more than 16 weeks prior to the bronchoscopy. We also analyzed matched blood within 72 hours after the bronchoscopy. We stained CD4+ cells with lineage specific markers, including CXCR3, CXCR5, CD25, CD127, CCR4, and CCR6. Proportion of the CD4+ cell subsets within total CD4+ lymphocytes in BAL and blood were compared between the pneumonitis and controls. Results: Th1 (CXCR3hi) CD4+ cells were expanded in controls in BAL (pneumonitis versus control, 4.2 ± 2.5 % versus 17.2 ± 6.3 % within total CD4+ lymphocytes, P= 0.04) and blood (pneumonitis versus control, 0.5 ± 0.3 % versus 4.0 ± 1.3 % within total CD4+ lymphocytes, P= 0.01). In contrast, Th1/17 (CXCR3hi CCR6hi) hybrid CD4+ cells, known to be pathogenic in autoimmune diseases, were expanded in BAL from the pneumonitis group (pneumonitis versus control, 40.3 ± 8.4 % versus 13.7 ± 4.5 % within total CD4+ lymphocytes, P= 0.03). Th1/17 hybrid CD4+ cells were also PD-1hi Ki67hi, suggesting their hyperactive status. Though not reached statistical significance, regulatory T cells were decreased in BAL from pneumonitis group (pneumonitis versus control, 20.8 ± 4.9 % versus 26.2 ± 9.2 % within total CD4+ lymphocytes). Conclusions: These results suggest that Th1/17 hybrid CD4+ cells may play a central role in pneumonitis. Understanding of the Th1/17 hybrid CD4+ cell biology will provide therapeutic targets and reliable biomarkers for pneumonitis.

Factors associated with risk of central nervous system relapse in patients with non-core binding factor acute myeloid leukemia

Central nervous system (CNS) relapse is uncommon in patients with acute myeloid leukemia (AML) with the use of high-dose cytarabine containing chemotherapy regimens. The clinical and molecular features associated with a higher risk of CNS relapse are not well defined. We assessed the incidence and outcome of CNS relapses among 1245 patients with relapsed/refractory AML referred to our institution between 2000 and 2014. CNS leukemia relapse was observed in 51 patients (4.1%). Using a multivariate regression model and after adjusting for age, FLT3-ITD mutation (OR = 2.33; P = .02) and elevated LDH (>1000 IU/L, OR = 1.99; P = .04) were independent predictive factors for CNS relapse. Patients under 64 years of age with 0, 1, or 2 baseline adverse features had a probability of 3.8%, 7.0%-8.0%, and 13.9% for developing CNS disease, respectively. Our study identifies patients with AML at higher risk for CNS relapse in whom prophylactic CNS therapy may be warranted.

Sorafenib plus 5-azacytidine (AZA) in older untreated FLT3-ITD mutated AML

7029

Background: Sorafenib plus 5-azacytidine (AZA) is observed to be safe and effective in relapsed / refractory FLT3-ITD mutated acute myeloid leukemia (AML) patients (pts). Hypothesis: Combining sorafenib with AZA is safe and effective in older untreated FLT3-ITD mutated AML pts. Methods: Eligibility included: untreated FLT3-ITD mutated AML (≥10% mutation burden), age ≥60 yrs, adequate organ function, and ECOG performance status ≤ 2. The regimen was: AZA 75 mg/m2daily x 7 days and sorafenib 400 mg twice daily for 28 days. Results: 26 pts with untreated AML [median age 73 (61-86)] were enrolled: 16 (62%) pts had normal karyotype, 2 (8%) complex karyotype, 4 (15%) other miscellaneous abnormalities, and 4 (15%) with insufficient metaphases. Prior to the initiation of treatment, FLT3-ITD was detected in all pts with a median allelic ratio of 0.3735 (0.009-0.885). The overall response rate (ORR) in 25 evaluable pts was (76%) [7 (28%) with CR, 10 (40%) CRi/CRp, and 2 (8%) PR]. Pts underwent a median of 3 (1-35) treatment cycles. The median number of cycles to response was 2 (1-4), and the median time to achieve response, 1.77 months (mos) (0.689-4.271 mos). The median duration of CR/CRp/CRi is 14.5 mos (1.18—28.74). Three (18%) responding pts (CR, CRp, CRi) have proceeded to allogeneic stem cell transplant. With a median follow-up of 6.8 mos (0.2-18.8), 6 pts are alive, 3 in remission (CR/CRP/CRi). The median overall survival (OS) for the entire group is 8.3 mos; 9.2 mos in 17 responders. Evaluable pts treated with AZA + sorafenib (n = 25) were compared to a matched cohort of historical FLT3-ITD mutated pts > 60 yrs, but treated with hypomethylator-based (HMA) therapy without sorafenib (n = 20); the respective ORR (CR, CRp, CRi, PR) (76% vs. 70%, p = 0.653) and median OS (8.3 and 9.4 mos, p = 0.69) were similar. The remission duration for the responding pts treated with AZA+sorafenib was significantly longer (14.5 mos) than those on other HMA regimens without sorafenib (3.8 mos) (p = 0.01). Adverse events possibly attributable to the regimen included: grade (Gr) 1/2 nausea (n = 3), Gr 1/2 diarrhea (n = 2), Gr 1 dyspnea (n = 1), and Gr 1 breast pain (n = 1). Conclusions: The combination of AZA and Sorafenib is both well tolerated and effective in older untreated FLT3-ITD mutated AML. Clinical trial information: NCT02196857;NCT01254890.

Phase IB/II study of nivolumab with azacytidine (AZA) in patients (pts) with relapsed AML

7026

Background: Blocking PD-1/PD-L1 pathways enhances anti-leukemia responses in murine AML (Zhang et al, Blood 2009). PD-1 positive CD8 T-cells are increased in bone marrow (BM) of pts with AML (Daver et al, ASH 2016). AZA up-regulates PD-1 in AML (Yang et al., Leukemia 2013). Methods: Pts were eligible if they had AML and failed prior therapy, had adequate performance status (ECOG ≤ 2), and organ function. AZA 75mg/m2Days 1-7 with nivolumab 3mg/kg on Day 1 and 14 was established as the recommended phase II dose. Courses were repeated every 4-5 weeks indefinitely. Responses were evaluated at the end of 3 courses. Results: 53 pts with med age 68 years (range, 44 – 90), secondary AML (43%), poor risk cytogenetics (43%), med prior regimens 2 (range, 1-7) have been enrolled. Common mutations included DNMT3A (n = 11), TP53 (n = 11), TET2 (N = 8), CEBPA (n = 8), ASXL1(n = 8). All 53 pts are evaluable for response: 11 (21%) achieved CR/CRi and 7 (14%) had hematologic improvement (HI) for an overall response rate of 35%. Additionally, 14 (26%) had ≥50% BM blast reduction, 3 (6%) had stable disease > 6 months, and 12 (23%) had progression. The CR/CRi have been durable with 9 of 11 (82%) pts with CR/CRi alive at 1 year, after censoring for SCT. Med survival for the 53 evaluable pts was 5.7 months (range, 0.9 – 16.2) and in the 27 salvage 1 pts was 9.3 months (range, 1.6 – 16.2). These compare favorably to historical survival with AZA-based salvage protocols at MDACC. Grade 3/4 and Grade 2 immune toxicities were observed in 7 (14%) and 6 (12%) pts, respectively. These responded rapidly to steroids and 12 of 13 pts were successfully rechallenged with nivolumab. Multicolor flow-cytometry data were available on pretherapy, end of cycle 1, and end of cycle 2 BM aspirates in 9 CR/CRi and 22 non-responders. Pts who achieved CR/CRi had higher pre-therapy total CD3 (P = 0.02) and higher CD8+ T-cells (P = 0.07) infiltrate in the BM. Responders demonstrated progressive increase in BM CD8+ and CD4+ infiltrate. Both responders and non-responders had increase in CTLA4+ CD8+ cells on therapy. Conclusions: Full dose AZA and nivolumab are tolerable and may produce durable responses in relapsed AML. Up-regulation of CTLA4 may be a mechanism of resistance to PD1 based therapies in AML. Clinical trial information: NCT02397720.

Idarubicin and cytarabine with clofarabine or fludarabine in adults with newly diagnosed acute myeloid leukemia: Updated results of a randomized phase II study

7037

Background: The purine nucleoside analogues fludarabine and clofarabine are effective agents in the treatment of acute myeloid leukemia (AML). This study evaluated the efficacy and safety of combining idarubicin and cytarabine with either clofarabine (CIA) or fludarabine (FIA) in adults with newly diagnosed AML. Methods: Using a Bayesian adaptive design, patients (pts) deemed suitable for intensive chemotherapy were randomized to receive CIA (n = 106) or FIA (n = 76). All pts received idarubicin 10 mg/m2 IV daily on Days 1-3 and cytarabine 1 g/m2 IV daily on Days 1-5. Clofarabine and fludarabine were given at 15 mg/m2 and 30 mg/m2, respectively, IV daily on Days 1-5. Pts with FLT3 mutations could receive sorafenib. Up to 6 cycles of consolidation were allowed for responding pts. Results: Baseline characteristics were similar comparing CIA to FIA with a median age of 53 years (range, 20-66) vs 49 years (range, 18-66) respectively and ELN risk intermediate-2/adverse of 57% and 58% respectively. With a median follow-up of 27 months (range, 1-58), the CIA and FIA arms had a similar CR/CRp rate (80% and 82%, respectively). MRD negativity rate by multiparameter flow cytometry at the time of CR/CRp was higher comparing CIA to FIA (80% vs. 65%, respectively, P = 0.07). The median EFS were 13 months and 12 months, respectively (P = 0.91), and the median OS were 24 months and not reached, respectively (P = 0.23). There were more adverse events (all grades) associated with CIA, particularly AST/ALT elevation (29% vs 4%), hyperbilirubinemia (26% vs 9%) and rash (31% vs 9%). Early mortality was similar in the 2 arms (60-day mortality: 4% for CIA vs 1% for FIA; P = 0.32). Comparing the 2 arms to a historical cohort of pts treated with IA showed similar response rates, EFS and OS excluding pts with FLT3 mutations from this analysis. However, in pts < 50 years of age, FIA was associated with improved survival compared with IA (2-year EFS rate: 58% vs 30%, P = 0.05; 2-year OS rate: 72% vs 36%; P = 0.009). Conclusions: CIA and FIA have similar efficacy in younger pts with newly diagnosed AML. FIA is associated with a better toxicity profile and may improve survival compared to IA in pts < 50 years of age. Clinical trial information: NCT01289457.

Ruxolitinib (RUX) in combination with azacytidine (AZA) in patients (pts) with myelodysplastic/myeloproliferative neoplasms (MDS/MPNs)

7064

Background: Clinical trials exclusively focusing on pts with MDS/MPNs are lacking. Combining RUX and AZA may target distinct manifestations of MDS/MPNs. Methods: Pts were treated with single-agent RUX 15 mg or 25 mg (based on platelets count) orally twice daily continuously in 28-day cycles for the first 3 cycles. AZA 25 mg/m2 (day 1-5) was added on each cycle starting cycle 4 and could be increased to 75 mg/m2 (maximum) or started earlier than cycle 4 and/or at higher dose in pts with proliferative features or high blasts. Results: 35 pts with med age 70 years (range, 43-79) were enrolled (MDS/MPN-U, n = 14; CMML, n = 17; atypical CM (aCML), n = 4), 28 (80%) were Int-2/High per MF DIPSS, 14 (41%) had splenomegaly > 5 cm, and 12 (34%) had EUMNET MF-2/MF-3 fibrosis. Common mutations on a 28-gene sequencing panel included JAK2 (29%), RAS (27%), ASXL1 (21%), TET2 (18%), and DNMT3A (12%). All 35 pts were evaluable for response per MDS/MPN IWG criteria and 17 (49%) responded. 6/17 (35%) IWG responses occurred after the addition of AZA (med time after AZA = 1.8 months). JAK2 mutated pts had a trend to higher responses vs those with non-mutated pts (8/10 vs 9/25, P = 0.19). Ten pts had > 5% pretreatment BM blasts and 7 achieved a reduction in blasts to < 5% (70%). A > 50% reduction in palpable spleen length reduction at 24 weeks was seen in 9/12 (75%) pts. New grade 3/4 anemia and thrombocytopenia occurred in 18 pts (51%) and 19 (54%) pts but were manageable with dose modifications. Only one pt discontinued therapy due to cytopenias. At a med follow-up of 17.4 mo (range, 1.2-36.8+), 14 (40%) pts died: pneumonia (n = 4), sepsis (n = 4), progression to AML (n = 4), and cardiac arrest (n = 1). The med survival for all pts was 16.6 mo (1.0 - 36.8). Compared to CMML and aCML, MDS/MPN-U pts had significantly better med survival (26.4+ vs 15.0+ vs 1.5+ mo, respectively; p = 0.01). Conclusions: The combination of RUX and AZA showed an IWG-response rate of 49% in pts with MDS/MPNs, and was well-tolerated. The benefit appears more profound in pts with MDS/MPN-U. This study is ongoing. (ClinicalTrials.gov Identifier: NCT01787487).

Phase I study of lurbinectedin (PM11083) in patients with advanced AML and MDS

e18521

Background: The FDA-approved drug trabectedin is a DNA minor groove binder (MGB) with activity against translocation-associated sarcomas. Lurbinectedin is a next-generation MGB, with activity against myeloid leukemia cells. A phase I clinical trial was initiated to determine recommended doses in MDS/AML. Further assessment of its safety, tolerability, and pharmacogenetics were studied. Methods: In total, 42 patients with relapsed/refractory MDS/AML received lurbinectedin, using a 3+3 study design. It was administered as a 1-hour IV infusion first at 3.5, 5, 6, and 7mg per dose on days 1 and 8. Subsequently, it was administered on days 1, 2, and 3, using 1, 1.5, 2, or 3mg per dose per patient. Results: Three patients had dose-limiting toxicities (DLTs), which were rhabdomyolysis (up to gr 4), hyperbilirubinemia (gr 3), and oral herpes (gr 3). All DLTs occurred in the 6 and 7mg cohorts. Adverse events included febrile neutropenia/infections (n = 3), GI (n = 6) and pulmonary (n = 2) toxicity, hyperhidrosis (n = 1), and QT prolongation (n = 1). Overall, 33 of 42 patients (79%) had reduction of blasts in peripheral blood (PB) or bone marrow (BM) at nadir, including 23 (55%) with > 50% reduction in PB blasts alone (n = 18) or both BM and PB blasts (n = 5). One patient had PR, 2 patients had morphologic leukemia-free survival, and most (n = 30) were discontinued for progressive disease. Early deaths occurred from disease-related causes, not attributable to lurbinectedin. Thirty-two patients treated at MD Anderson were analyzed for clinical characteristics and responses. Notably, among patients with follow-up bone marrow biopsy, those with a cytogenetic abnormality at chromosome 11q21-23 had significantly greater bone marrow blast reduction than those without such abnormality (change in percentage blasts: -31±14% [n = 4] vs. 8±8% [n = 16], respectively; P= 0.04). Conclusions: Lurbinectedin is generally safe and tolerated at dose levels tested. While no sustained remissions > 3 cycles were observed in these highest-risk patients, single-agent lurbinectedin was transiently leukemia suppressive for some patients, including some with abnormal chr11q or TP53 mutation. Rational combinations and situational uses of lurbinectedin are under consideration. Clinical trial information: NCT01314599.

Immune-related gene expression deficit of leukemia stem cells (LSC) in AML

7011

Background: AML LSC are believed to be responsible for residual and resistant leukemic disease leading to relapse. Understanding differences between bulk AML and the LSC subpopulation may allow the identification of novel LSC targets, especially for the most adverse risk AML where few patients are cured. Targeting LSC may be needed to eradicate AML, and immune-based therapies provide an approach for eliminating LSC. The transcriptional landscape of immune-related genes in LSC is not well understood. Methods: Samples were collected at diagnosis from 12 patients with high-risk AML prior to therapy. Bulk (CD45-dim blasts) and LSC (Lin-CD34+CD38-CD123+) AML marrow cells were FACS-sorted and analyzed using whole genome RNA-sequencing. Transcriptomes were analyzed using AltAnalyze software to identify differentially expressed genes in bulk AML cells and in AML LSC populations. These genes were further assessed by gene enrichment analysis using data from Gene Ontology (GO) and the Cancer Genome Atlas Project (CGAP). Results: Sixty-eight genes were identified with greater than 3-fold differential expression between bulk AML and LSC. GO enrichment analysis demonstrated more than 10-fold enrichment of genes involved in the molecular functions, biologic processes, and cell components related to the antigen presentation pathway, with the comparative down-regulation occurring in LSC. Among the top differentially expressed gene clusters, both the MHC class II and interferon-gamma signaling/response pathway gene expression was blunted in LSC. Additional expression analysis revealed that 42% of a CGAP-curated list of 201 antigen-processing and -presentation genes had significantly decreased expression in the LSC subpopulation compared to bulk AML. Conclusions: LSC from primary AML patient samples are characterized by reduction in expression of MHC class II receptor and antigen presentation genes compared to bulk AML. These results suggest that impairment in the presentation and/or processing of tumor associated antigens by MHC class II on LSC, along with tonic sponging of immune response cells and diversion away from LSC by bulk AML, may contribute to LSC evasion of immune surveillance and response.

Clinical relevance of mutations in patients with myelodysplastic syndromes and myelodysplastic/myeloproliferative neoplasms with normal karyotype

7053

Background: Clinical outcomes of patients with myelodysplastic syndromes (MDS) and myelodysplastic/myeloproliferative neoplasms (MDS/MPN) are heterogeneous. Specific mutations and mutation patterns are known to define prognostic groups in normal karyotype acute myeloid leukemia. Whether this is the case in MDS and MDS/MPN remains unknown. Methods: We evaluated 325 previously untreated patients with MDS or MDS/MPN with normal karyotype evaluated from 2012 to 2016. Next generation sequencing (NGS) on whole bone marrow DNA analyzing a panel of 28 or 53 genes was performed at the time of diagnosis. Results: A total of 225 (69%) patients had MDS and 100 (31%) had MDS/MPN including 77 (24%) patients with chronic myelomonocytic leukemia (CMML). Median age was 69 years (31-92). Among patients with MDS, 189 (84%) had lower-risk and 36 (16%) had higher-risk based on IPSS. NGS data was obtained by 53-gene panel in 93 (29%) patients and by 28-gene panel in 232 (71%). A total of 202 (62%) patients had detectable mutations. Median number of mutations was 1 (range 0-6). Detected mutations are detailed in Table 1. A total of 111 (34%) patients, 70 with MDS and 41 with MDS/MPN, received therapy with hypomethylating agents. Median follow up was 12 months (0-167). By univariate analysis, NRAS (HR 3.28, CI 1.25-8.62, p=0.016) and TP53 (HR 4.9, CI 1.44-16.67, p=0.011) predicted for shorter overall survival (OS) among MDS patients. After multivariate analysis including IPSS-R, only TP53retained its impact in OS (HR 5.25, CI 1.44-19.13, p=0.012). Among MDS/MPN patients, no mutation was found to significantly impact OS. Conclusions: With the exception of TP53mutations, no other identified mutation seemed to independently define prognosis of patients with MDS or MDS/MPN with normal karyotype. In view of the high proportion of lower-risk patients, longer follow up is required to better define prognostic impact of mutations in this population. [Table: see text]

Phase IB/II study of lirilumab with azacytidine (AZA) in relapsed AML

e18505

Background: Lack of KIR-HLA class I interactions has been associated with NK-mediated antitumor efficacy in AML patients (pts) in remission upon KIR-mismatched haploidentical stem cell transplantation (SCT) (Ruggeri L et al., Science 2002). Blockade of KIR2DL1, 2 and 3 receptors induced augmented NK-cell mediated lysis of tumor cells (Romagne F et al., Blood 2009). Hypomethylating agents possess anti-leukemia activity, increase MHC II expression, and interferon-gamma signature. Methods: Pts with AML who failed prior therapy, have adequate performance status (ECOG ≤ 2) and organ function are eligible. AZA 75mg/m2 Days 1-7 was given with lirilumab on Day 8 at the dosage of 1 and 3 mg/kg in 2 consecutive cohorts of 6 pts each. Courses were repeated every 4-5 weeks. Lirilumab 3mg/kg established as recommended phase 2-dose with AZA. 13 additional pts treated at the RP2D. Responses evaluated at the end of 3 courses. Results: To date, 25 pts (16 de novo, 9 secondary AML), median age 64 years (range, 30 – 89), 52% adverse cytogenetics, median prior therapies 3 (range, 1-8), and prior alloSCT in 6 (28%) have been enrolled. All 25 pts had baseline next generation sequencing, common mutations included ASXL1 (n = 9), TP53 (n = 7), RAS (n = 4), TET2 (n = 4), and RUNX1 (n = 3). Two (8%) achieved CR/CRi (1 CR, 1 CRi) and 3 (12%) achieved hematologic improvement for an overall response rate of 20%. The median cycles to response was 3 (range, 1-11). The 8-week mortality is 12%, respectively. The median duration of response and overall survival were 2.0 months and 4.0 months, respectively (med f/u = 5.1 months). Grade 3/4 toxicities were similar to those seen with AZA based therapies in salvage including bloodstream infections in 15 (60%), 6 pneumonia, 1 UTI, 1 skin infection, 2 abdominal pain, and 1 mucositis. Immune mediated toxicities were observed in 4 (16%) pts (1 pneumonitis Grade 3, 2 colitis Grade 3, 1 colitis Grade 2). The immune mediated toxicities responded rapidly to steroids and 3 pts could be rechallenged safely with lirilumab. Six pts were postSCT and no Grade 3/4 GVHD flares were noted. Conclusions: Full doses of AZA and lirilumab were well tolerated in heavily pretreated pts with relapsed AML. The efficacy data are still preliminary and the study is ongoing. Clinical trial information: NCT02399917.

Phase 2 study of ruxolitinib in combination with 5-azacitidine in patients with myelofibrosis

7063

Background: Ruxolitinib (RUX) is effective in controlling symptoms and organomegaly in patients with myelofibrosis (MF). Combination with azacitidine (AZA) may further improve its efficacy. Methods: RUX 15 or 20 mg orally twice daily was given continuously since cycle 1. AZA 25 to 75 mg/m2 on days 1-5 of each 28-day cycle was added starting cycle 4. Responses were assessed per International Working Group for Myelofibrosis Research and Treatment 2013 criteria (IWG-MRT). Results: Among 44 pts enrolled between 03/2013 and 06/2016, 39 patients (89%) were evaluable for response. After median (med) follow-up of 20.4+ months (range, 0.5-37+); 24 pts (54%) are on study with a med overall survival of 39+ months. Med age was 66 years (range, 48-87), 36 pts (82%) had int-2/high IPSS score, 29 (66%) had spleen ≥5cm, and 24 (55%) were JAK2 V617F positive. Twenty five pts (57%) were previously treated. Twenty eight (72%) pts had objective response regardless of previous therapy (Table). Med time to response was 1.0 months. 7 (25%) responses occurred after the addition of AZA with med time to response of 4.2 months. In total, 23 pts (79%) had palpable spleen reduction by > 50%, which occurred after AZA was added in 6 (28%) of them. JAK2V617Fallele reduction was noted in 13 (87%) evaluable pts, including > 50% reduction in 3 pts (13%). A reduction in bone marrow fibrosis grade was observed in 12 (31%) responders, including ≥2 and 1 grade reduction in 2 and 9 pts, respectively. Grade 3/4 non-hematological and hematological toxicities occurred in 4 and 16 pts, respectively. The most common reasons for therapy discontinuation (n=17) were stem cell transplantation (n=6), lack of response (n=3) and progression to AML (n=2). Conclusions: Concomitant RUX with AZA was feasible with overall IWG-MRT response rate of 72%, including >50% spleen reduction in 79% of patients, which compares favorably to single RUX. Clinical trial information: NCT01787487. [Table: see text]

Distinct patterns of somatic mutation clearance and association with clinical outcome in patients with AML

7005

Background: Persistence of somatic mutations at the time of complete remission (CR) was associated with poor outcome in patients (pts) with AML. Methods: We studied 95 pts with AML who were treated with frontline induction and subsequently achieved CR. We sequenced pre-treatment and CR bone marrow samples by targeted capture sequencing of 295 genes (median 280x coverage). We defined 3 levels of mutation clearance (MC) based on variant allele frequency (VAF): MC2.5, persistent mutation with VAF<2.5%; MC1.0, persistent mutation with VAF<1%; and complete mutation clearance (CMC). Results: In the pre-treatment samples, we detected 597 mutations in 78 genes in 87 (92%) patients. In the matching CR samples, 62 (10%) and 82 (14%) mutations persisted at VAF≥2.5% and ≥1%, respectively, which corresponded to 43 (49%), 34 (39%), and 30 (34%) patients achieving MC2.5, MC1.0 and CMC, respectively. Table 1 shows the differential patterns of MC based on the mutations and pathways. Mutations associated with clonal hematopoiesis of indeterminate potential (CHIP), DNA methylation, and splicing pathways had low rate of MC, whereas mutations in transcription factors or receptor tyrosine kinase (RTK) had high rate of MC. Pts who achieved MC1.0 (median 31.2 vs. 12.5 months, P = 0.04) or CMC (median 31.2 vs. 12.5 months, P = 0.049) had significantly better relapse-free survival (RFS). Conclusions: Somatic mutations associated with CHIP, DNA methylation, and splicing pathways persisted frequently in CR samples suggesting preleukemic origin. Pts with deeper MC had significantly better RFS. Somatic mutation clearance may help risk prediction of AML. [Table: see text]

Efficacy and safety of pegylated human IL-10 (AM0010) in combination with an anti-PD-1 in renal cell cancer

4567

Background: IL-10 has anti-inflammatory activity and stimulates the cytotoxicity and proliferation of CD8+ T cells at higher concentrations. IL-10 receptors and PD1 are expressed on activated CD8 T cells, providing a rationale for combining AM0010 and an anti-PD1. The efficacy and safety profile for AM0010 alone was established in poor to intermediate risk RCC pts treated in 3rdLOT. Objective responses were observed in 4 of 15 pts with RCC. In the dose escalation of AM0010 plus pembrolizumab, 4 of 8 patients had an objective response. The mPFS was 16.7 months and the mOS has not been reached, median follow up (mFU) is 19.3 mo. Methods: In this Phase 1b, 29 pts. with metastatic RCC were enrolled until Nov. 18 2016 on AM0010 (10 ug/kg daily SC) and nivolumab (3mg/kg, q2wk IV). 2 had favorable, 20 had intermediate and 4 had poor IMDC risk (3 were not available). Pts. had a median of 1 prior therapy (range 1-3). All pts. had received a VEGFR-TKI. Tumor responses were assessed with irRC. Immune responses were evaluated by serum cytokines, activation of blood derived T cells and peripheral T cell clonality. Results: AMO010 plus nivolumab was well tolerated. TrAEs were reversible. There were no autoimmune colitis, pneumonitis, or endocrine disorders. 14 patients had at least 1 G3/4 TrAE, including anemia (9), thrombocytopenia (5), hypertriglyceridaemia (4). 2 pts had a reversible cytokine release syndrome with splenomegaly and increased immune mediated red blood cell phagocytosis most likely precipitated by T-cell activation, as both pts had tumor responses. As of Jan 31 2017, partial responses (PR) were observed in 8 of 26 evaluable pts (31%). An additional 13 of 26 pts had stable disease (41%), 7 pts had tumor reductions of more than 30%. The mPFS and mOS has not been reached with a mFU of 5.2 mo. (range 0.3-10.3). AM0010 + anti-PD1 increased Th1 cytokines in the serum while decreasing TGFb, an expansion of proliferating PD1+ Lag3+ activated CD8 T cells and de-novo oligoclonal expansion of T cell clones in the blood. Conclusions: AM0010 in combination with nivolumab is well-tolerated in RCC pts. The efficacy and the observed CD8 T cell activation is promising and encourages the continued study of AM0010 in combination with nivolumab. Clinical trial information: NCT02009449.

Outcomes with lower intensity therapy in TP53-mutated AML

7017

Background: TP53 mutations confer an adverse prognosis in patients (pts) with AML treated with standard chemotherapy. A recent study reported high response rates using a 10-day regimen of decitabine (DAC10) in pts with TP53-mutated ( TP53-MUT) AML. The question remains whether this benefit is unique to DAC10 or whether the same benefit among TP53-MUT AML applies to other low intensity therapy (Rx). Methods: We reviewed our own experience of pts treated with low intensity Rx from 2012 - 2016. Mutation testing was performed using a whole-exome sequencing panel. We reviewed the clinico-pathologic characteristics of these pts, and compared their outcomes based on the presence/absence of a TP53mutation and by the type of Rx they received. Results: There were 131 pts in our cohort of which 33 (25%) had TP53-MUT. Pt characteristics are outlined in Table 1A. All pts were treated with low intensity Rx and were divided into the following groups: DAC10 [n=34, 26%]; 5-day decitabine, or 7-day azacytidine (DAC5) [n=39, 30%]; or cladribine+low dose araC (CLAD/LDAC) [n=58, 44%]. Response rates and OS by Rx and TP53-MUT status are summarized in Table 1B. While there was no significant difference in response rates or OS by TP53-MUT status within any of the treatment approaches, there was a trend for inferior response rates and OS among pts with TP53-MUT who received either DAC-5 or CLAD/LDAC ; this was not seen in pts receiving DAC10. Conclusions: The presence of a TP53-MUT was associated with a nonsignificant trend towards inferior outcomes among pts receiving DAC5 or CLAD/LDAC, but not among those receiving DAC10. Comparing across groups, the CLAD/LDAC combination was associated with the longest OS, and DAC10 was associated with superior outcomes compared to DAC5, in TP53-MUT cohort. [Table: see text]

Achievement of a negative minimal residual disease state after hypomethylating agent therapy in older patients with AML to reduce risk of relapse

7018

Background: Persistence of minimal residual disease (MRD) post therapy is a powerful predictor of outcome in patients with AML treated with traditional cytarabine and anthracycline based regimens. The clinical relevance of MRD in the context of hypomethylating agents has not been evaluated extensively. Methods: Among 194 patients with AML treated with single agent azacytidine, decitabine, or gaudecitabine, 116 (median age 76, range 60 - 92) had MRD analysis performed on bone marrow specimens obtained at time of assessment of response or thereafter; among them 69 (59%) achieved either morphologic complete remission (CR) or CR with incomplete recovery of platelets (CRp) or counts (CRi), and 61 (53%) had evaluable MRD data; MRD was assessed using an 8-color flow panel, with a detection sensitivity of 0.01%. Results: Median cycles to achieving response was 2 (range, 1-6). Sixty one patients had evaluable MRD data at the time of response, of whom 19 (28%) became MRD negative (-). This was associated with a reduced cumulative risk of relapse (p = 0.012) but did not translate to an improved relapse-free survival (RFS; p = 0.17) or overall survival (OS; p = 0.79) due to high frequency of non-relapse deaths (attributable to comorbidities and infections) in the MRD- group. Patients who achieved a MRD- state at the time of achieving response had a higher mortality [5/8 (62%)] when compared with those who achieved a MRD- state in subsequent cycles [1/13 (7.6%); p = 0.01], resulting in an inferior OS (6.2 months (mo) vs 20 mo, p = 0.012). Similarly, achieving negative MRD at CR and at any time up to 3 months post response was not associated with improved RFS or OS despite a lower cumulative risk of relapse (p = 0.045). There was no impact of MRD, on OS, whether the MRD- state was achieved after 1st, 3rd or 6thcycle of therapy. Association between depth of MRD response at time of remission and RFS was borderline significant (p = 0.08). On multivariate analysis, response (CR vs CRi/CRp), but not a negative MRD, was predictive for RFS or OS. Conclusions: In this cohort of older AML patients treated with hypomethylating agents, achieving a MRD- state was associated with a reduced risk of relapse but not improved RFS or OS.

Impact of the type of first-cancer therapy in therapy-related myelodysplastic syndromes outcomes

e18555

Background: Therapy-related myelodysplastic syndromes (t-MDS) are known to have poor outcomes. The impact of the therapy received for the first cancer remains unknown. Methods: 228 t-MDS diagnosed between July 2001 and December 2015 were retrospectively reviewed. Results: Median age at diagnosis of t-MDS was 66 years (range 13-87). Previous malignancies: 54% hematological cancers and 47% solid tumors. 14% had multiple cancers before the t-MDS. Previous therapy received: 119 (52%) chemotherapy (CT), 17 (8%) radiotherapy (RT), or 92 (40%) both (Table). Median time to develop t-MDS was 73 months (range 5-355). According to IPSS, 13 patients (6%) had low-risk, 65 (29%) intermediate-1 risk, 119 (53%) intermediate-2 risk and 29 (13%) high-risk. 124 patients (58%) had complex karyotype. The most frequent cytogenetic abnormalities detected were: -7/7q (n=99; 46%), -5/5q (n=77; 36%), and +8 (n=13; 6%). TP53was observed in 51% (n=21/41). Median OS was 12 months (CI95% = 10-14). Univariate and multivariate analysis revealed that CT vs RT (p=0.028; HR [CI95%] = 1.988 [1.08-3.67]), and higher IPSS risk (low: p=0.027; intermediate-1: p=0.285 HR [CI95%] = 1.48 [0.72-3.01]; intermediate-2: p=0.055, HR [CI95%] = 1.95 [0.99-3.88]; high: p=0.015 HR [CI95%] = 2.59 [1.21-5.55]), had a poor impact in OS. No statistical differences between OS and type of CT received, molecular or cytogenetic alterations were observed. However, complex karyotype (71% vs 51%, p=0.007) and -7/7 abnormalities (61% vs 36%; p=0.001) were more frequent in patients who had received platinum; -7/7q- abnormalities were also associated to alkylating agents (50% vs 27%, p=0.012), antibiotics (74% vs 42%; p=0.003) and topoisomerase inhibitors (60% vs 37%; p=0.002). Conclusions: t-MDS is associated with poor outcomes. Receiving CT vs RT and higher IPSS risk had a poor impact in OS. The type of CT may be associated with specific cytogenetic alterations. [Table: see text]

Frontline hyper-CVAD plus ponatinib for patients with Philadelphia chromosome-positive acute lymphoblastic leukemia: Updated results of a phase II study

7013

Background: The combination of chemotherapy plus a TKI is highly effective in Ph+ ALL. In this phase II study, we evaluated the safety and efficacy of HCVAD in combination with the third-generation pan- BCR-ABL inhibitor, ponatinib. Methods: Patients (pts) with newly diagnosed Ph+ ALL received 8 cycles of HCVAD alternating with high dose MTX/Ara-C every 21 days. Ponatinib was given at 45 mg daily for the first 14 days of cycle 1. Initially ponatinib 45 mg daily was given indefinitely beginning at cycle 2. Due to concern for vascular events, a protocol amendment was made in which, beginning in cycle 2, pts in CR received 30mg daily and pts in CMR received 15mg daily. Rituximab and IT chemotherapy were given with the first 4 courses. After 8 cycles of HCVAD, pts in CR received maintenance with ponatinib, vincristine and prednisone for 2 years followed by indefinite ponatinib. Results: 64 pts have been treated, 10 of whom had received prior treatment with another regimen (8 in CR, 2 not in CR). Median age was 48 years (range, 21-80) and median follow-up was 33 months (range, 2-62). Median cycles received was 6 (range, 2-8). 63 pts (98%) achieved CR after 1 cycle; 1 pt achieved CRp. CCyR was achieved in 98%, MMR in 97% and CMR in 77%. Median time to CMR was 10 weeks (range, 2-96). Median times to platelet and ANC recovery in cycle 1 were 22 and 18 days, respectively, and for subsequent cycles were 22 and 16 days, respectively. Grade ≥3 pancreatitis was observed in 12 pts (19%), thrombotic events in 4 (6%) and MI in 3 (5%). 8 pts have died, with 2 deaths attributed to ponatinib (both from MI). No grade ≥3 vascular events occurred after the protocol amendment. 38 pts continue to receive treatment (7 in consolidation, 14 in maintenance and 17 post-maintenance). 10 pts (16%) underwent allogeneic SCT in CR1. 7 pts have relapsed, 3 of whom were still receiving ponatinib. The 3-year continued remission and OS rates were 79% and 76%, respectively. In a landmark analysis at 4 months, CR duration and OS did not differ significantly in pts with or without allogeneic SCT. Conclusions: HCVAD plus ponatinib is highly effective in pts with newly diagnosed Ph+ ALL, resulting in high rates of CMR and promising long-term survival. Clinical trial information: NCT01424982.