Citations

INTRODUCTION: The Stromal Antigen 2 ( STAG2), on Xq25, is the most common cohesin complex gene mutated in myeloid neoplasms (MN) (5-10%). STAG2 mutations ( STAG2m) were incorporated in the ELN2022 as AML-MDS defining genetic mutations inferring poor prognosis and is a genetic marker in IPSS-M risk stratification. The PHD-finger protein 6 ( PHF6) gene, on Xq26.3, is a tumor suppressor gene, mutated in 3% of MN, and is more common in T-ALL. PHF6m is believed to be a late event in MN with little data on its prognostic implications in MN. Both STAG2m and PHF6m are more prevalent in males, but their implications on the clinical presentation of MN have not been compared prior. METHOD: Our study is a multicenter study at Mayo Clinic (Rochester, Florida, Arizona). After IRB approval, we retrospectively analyzed charts of 7935 patients who had NGS between 2016-2022. STAG2m was found in 96 patients, and PHF6m in 116 (including 3 with concurrent STAG2m and PHF6m). We excluded 4 PHF6m patients who had T-ALL. Data was captured at time of NGS. MDS risk stratification was by IPSS-M and AML by ELN2022. Overall survival (OS) was calculated from date of NGS to date of last follow-up. BlueSky Software was used for statistical analysis. RESULTS: Characteristics: We compared 96 STAG2m vs 112 PHF6m MN patients. Of STAG2m cases, 78% were male vs 76% in PHF6m MN. Median age was 72 vs 73 years, respectively. Diagnoses included AML (29% of STAG2m vs 34% of PHF6m MN), MDS (54% vs 28%), MDS/MPN (10% vs 21%) and MPN (3% vs 9%). MDS was more common among STAG2m MN (p<.001) while MDS/MPN was more common among PHF6m MN (=.05). MDS-IB was the most common subtype in both (69% vs 52% in STAG2m vs PHF6m MDS). MDS-LB and 5q subtypes were only found in PHF6m MDS, and there was more PHF6m MDS-RS (13% vs 2%). Median IPSS-M score was 0.9 vs 0.8 in STAG2m vs PHF6m MDS (both high-risk). Majority of STAG2m AML were adverse risk (96%) while PHF6m AML were intermediate (79%) (p<.001). AML cases included 43% vs 30% secondary AML (sAML) in STAG2m vs PHF6m AML. Among STAG2m and PHF6m MN, 17% of each were tMN. Cytogenetic abnormalities were more common in PHF6m MN (50% vs 31%, p=.005). The median bone marrow (BM) blasts count among STAG2m MN was 7% vs 4% in PHF6m MN (p=.03). Genetic characteristics: STAG2m were nonsense, frameshift and splice site (54%, 36% and 11%). PHF6m were nonsense, frameshift, splice site, missense and start loss variant mutations (44%, 33%, 12%, 10% and 1%). The median variant allele frequency (VAF) of STAG2m MN was 50% compared to 32% in PHF6m MN (p=.008). Co-mutations: The median number of co-mutations was 3 in STAG2m and 4 in PHF6m MN (p=.04). The most common were TET2m (37% in STAG2m MN vs 45% in PHF6m MN), RUNX1m (31% in both), SRSF2m (39% vs 20%), ASXL1m (66% vs 38%), BCORm (19% vs 6%), U2AF1m (13% vs 14%) and IDH2m (18% vs 7%). ASXL1m, SRSF2m, BCORm, IDH2m and SF3B1m were more common among STAG2m MN (p=<.001, .003, .006, .02, .09). JAK2m, IDH1m are more common among PHF6m MN (p=.02, .06). Survival: Of STAG2m and PHF6m MN, 70% and 72% of patients received treatment. HSCT was performed in 23 (24%) and 23 (21%) of STAG2m and PHF6m patients, including AML (n=9, n=10), MDS (n=12, n=8), MDS/MPN (n=2, n=2). OS among STAG2m and PHF6m MN was 20.5 and 22.1 months, respectively. OS among AML cases was 13.5 vs 21 months (p=.6) in STAG2m vs PHF6m, was NR vs 22 months (p=.4) in MDS, 11.7 vs 17.6 months in MDS/MPN (p=.4), and 10.4 vs 27.4 months (p=.5) in MPN. There was no difference between males and females in either STAG2m or PHF6m MN. HSCT led to longer OS among both STAG2m (NR vs 19.9 months, p=.01) and PHF6m MN patients (NR vs 17.6 months, p=.07). Among AML patients, HSCT did not significantly impact OS (18 vs 9.6 months, p=.3 and 25.1 vs 21 months, p=.2, in STAG2m and PHF6m). HSCT improved OS in STAG2m MDS (NR vs 20.5 months, p=.03). In PHF6m MDS, OS was 22.1 vs 13.1 months in HSCT vs non-HSCT (p=.3). CONCLUSION: STAG2m and PHF6m are 2 X-linked mutations that occur in MN. Both commonly occur in elderly males. MDS was the most common diagnosis in STAG2m MN. STAG2m co-occurred particularly with SRSF2m ASXL1m and BCORm, which are associated with poor prognosis. In our cohort, STAG2m AML carried a poorer prognosis than PHF6m AML. Our study confirms the ELN2022 AML risk stratification. MDS risk classification and OS did not differ between STAG2m and PHF6m in our cohort. A larger cohort is required to confirm our findings and may provide stronger validation.

Introduction The advent of venetoclax (VEN) based chemotherapy has revolutionized the treatment of elderly acute myeloid leukemia (AML), with improvement in remission rates from historical 10%-20% to 60%-70%. However, remission duration is short-lived in patients (pts) with high-risk AML with consequent dismal outcome. As our knowledge of genomic landscape of AML expands and more immunotherapies become available, exploration of the immune tumor microenvironment (TME) of AML may allow identification of subsets of pts that may benefit from immune based therapy. We conducted an investigator initiated prospective study to evaluate TME of pts receiving VEN plus hypomethylating agent (HMA; azacitidine or decitabine) combination. We hypothesize that these pts will cluster into immune-infiltrated vs. immune-depleted sub-types by high-definition immune cell profiling. Methods In 26 newly diagnosed AML pts, who were planned to receive VEN plus HMA therapy, we isolated bone marrow (BM) and peripheral blood (PB) derived mononuclear cells (MC) by density gradient centrifugation and cryopreserved. We performed CyTOF (cytometry time-of-flight) analysis to study their immune signatures. Thawed cells were labeled with metal-conjugated antibodies that bind CD33, CD196, CD19, CD127, CD11b, CD4, IgD, CD11c, CD14, CD56, CD123, TCRgd, CD185/CXCR5, CD45RA, CD27, CD28, CD66b, CD183 (CXCR3), CD161, CD45RO, CD197 (CCR7), CD8a, CD25, CD3, CD20, CD38, HLA-DR, CD194 (CCR4), CD57, CD16 and CD45. We performed a Rphenograph clustering on 25 paired BMMC and PBMC samples (one patient only had PBMC) to create 35 unique clusters for each ( Figure 1). Then we broke down the distribution of cluster by two different groupings: pts with TP53 mutation (m) and those who achieve complete remission. Statistical significance of clusters calculated using t-test. We performed Pearson correlation test on the paired BMMC and PBMC samples to assess correlation between BMMC and PBMC samples. Results Baseline characteristics The median age of the pts was 74 years (range, 53-88), and 17 (65%) pts were male. Ten (37%) pts had secondary AML, and 12 (46%) pts had ELN adverse risk disease (n=6; TP53m). There was no significant difference in median age (p= >0.99), proportion of pts with secondary AML (p= 0.64), acute myelomonocytic leukemia (p= 0.14) white blood cell count (WBC) ≥ 100 (10 9/L) (p= >0.99), and frequency of commonly occurring co-mutations ( ASXL1 [p=0.35]), FLT3 ITD [p=0.54], splicing [p=0.29], IDH1/2 [p=0.28], DNMT3A [p=0.29], NPM1 [p=0.29], TET2 [p=0.35]) in TP53m and TP53 wild-type group. Response 12/20 (60%) evaluable pts achieved complete remission with or without count recovery (CR/CRi). There was no significant difference in median age (p= 0.06), proportion of pts with secondary AML (p= 0.16), therapy related AML (p= 0.60) WBC ≥ 100 (10 9/L) (p= 0.40), and frequency of commonly occurring mutations ( TP53 [p=0.60], ASXL1 [p > 0.99]), FLT3 ITD [p=0.54], splicing [p > 0.99], IDH1/2 [p=0.60], DNMT3A [p=>0.99], NPM1 [p=0.61], TET2 [p >0.99]) in CR/CRi and no-CR/CRi group. CyTOF analysis and Rphenograph clustering on 25 paired BMMC and PBMC samples We found significant clustering with high expression of B-cell activated (cluster 2; p= 0.005), NK cell rich (cluster 1; p= 0.005, cluster 35; p= 0.0009), CD8 T-cells (cluster 31; p= 0.01) and CD4 effector cells (cluster 17; p= 0.005) phenotype among TP53m AML ( Table 1). Among pts who achieved CR/CRi, significant clustering with low expression of B-cell activated (cluster 29; p= 0.01), CD8 cytotoxic (cluster 14; p= 0.01) CD4 activated (cluster 21; p= 0.03) phenotypes observed. Correlation between BMMC and PBMC clustering Among clusters which showed significance for a particular phenotype, we performed Pearson correlation (R 2) and observed strong correlation between BMMC and PBMC samples clustering in cluster 1 (R 2 = 0.99), cluster 2 (R 2 = 0.95), cluster 17 (R 2 = 0.96), cluster 21 (R 2 = 0.99) and cluster 29 (R 2 = 0.93). Conclusion In this single institutional study using CyTOF analysis we have identified novel clustering with immune- infiltrated phenotype among pts with TP53m AML and conversely immune-depleted phenotype among AML pts who responded to VEN plus HMA. Moreover, we found strong correlation in CyTOF analysis conducted from BMMC and PBMC samples. Future studies with larger sample size needed to validate our findings.

Background: BMT CTN 1506/MORPHO was a randomized study of the FLT3 inhibitor gilteritinib versus placebo as post-transplant (HCT) maintenance therapy for patients (pts) with FLT3-ITD acute myeloid leukemia (AML). Patients with FLT3-ITD AML in first remission underwent HCT and then were randomized, in double-blind fashion, to either gilteritinib or placebo for 24 months. The primary endpoint was relapse-free survival (RFS), and a pre-specified secondary endpoint was the effect of measurable residual disease (MRD) on survival as detected by a highly sensitive assay for FLT3-ITD mutations. In the primary analysis, while RFS was higher for pts randomized to gilteritinib, the difference did not reach the pre-determined threshold for significance (HR: 0.679; 95% CI: 0.459, 1.005; 2-sided p-value: 0.0518). However, in the secondary analysis, the 50.6% of pts who had MRD detectable pre- or post-HCT had significantly higher RFS with gilteritinib (HR of relapse or death=0.515, 95% CI: 0.316, 0.838, p = 0.0065). In this post-hoc analysis, we examined 1) the impact on RFS of different levels of FLT3-ITD variant allele frequency (VAF) pre-randomization (immediately before or after HCT but prior to randomization to gilteritinib or placebo); 2) the impact of the presence of multiple mutations detected as MRD pre-HCT; and 3) the eradication of FLT3-ITD clones detected post-HCT during follow up on gilteritinib or placebo. Methods: First-pull marrow aspirates were collected from pts at two time points pre-randomization (pre-HCT and between 30-90 days post-HCT), as well as at 3, 6, 12, 18, and 24 months post-randomization. For MRD detection (performed at Invivoscribe; San Diego, CA), 700 ng input DNA was amplified by polymerase chain-reaction (PCR) using 25 cycles and primers flanking exons 14 and 15 of FLT3, followed by next-generation sequencing (NGS) analysis of the amplicons. Variant allele frequency (VAF) was calculated as FLT3 mutant reads/total FLT3 reads. For pts with multiple FLT3-ITD mutations, the VAF used in analysis was the sum of the VAFs for each FLT3-ITD variant. The lower limit of blank (LOB) of the assay is estimated to be 1 x 10 -6 VAF. For terminology, VAF > 1 x 10 -6 and < 1 x 10 -5 is referred to as MRD6, VAF > 1 x 10 -5 and < 1 x 10 -4 is MRD5, VAF > 1 x 10 -4 or greater is MRD4, and MRD0 equals no detectable MRD. Results: MRD was evaluated in 350/356 (98.3%) pts pre-HCT and 347/356 (97.5%) pts post-HCT. MRD was detected in 46% of pts pre-HCT and 19.9% prior to randomization, including 4.5% who did not have detectable MRD pre-HCT. The variant allele frequency (VAF) of detectable MRD ranged from a low of 1.09 x 10 -6 to a high of 3.0 x 10 -1. More than a single mutation was detected in 51 pts pre-HCT. Most of these 51 pts had 2 clones, but the total ranged from 2-9. The presence of more than one mutation detected as MRD immediately pre-HCT was associated with worse RFS compared with all other pts (MRD-positive with only one clone and MRD-negative), irrespective of treatment arm (Figure 1A; HR=2.428, 95% CI: 1.555, 3.792, P < 0.001). There was no significant difference in age, mutation length, VAF, or karyotype between pts with more than one mutation and those with single mutations. However, if the VAFs of pts with multiple FLT3-ITD mutations were summed and treated as a single value, then the median VAF for pts with multiple mutations was higher than for those with single mutations (2.14 x 10 -4 vs. 3.12 x 10 -5). The VAF (including the summed VAFs of pts with multiple mutations) at multiple levels correlated very closely with RFS after HCT. To illustrate this quantitative effect of peri-HCT FLT3-ITD VAF on RFS, Kaplan-Meier analysis of pts on the placebo arm grouped by MRD level is displayed in Figure 1B. In the 71 pts with MRD detectable post-HCT, the MRD was eradicated in 69% of pts on gilteritinib versus 44% of pts on placebo. Conclusions Pts with multiple FLT3-ITD mutations have a worse prognosis, which may simply be a reflection of increased disease burden (VAF). The quantity of FLT3-ITD MRD appears to highly correlate with outcome. Gilteritinib appears to augment the effect of HCT, as evidenced by increased eradication of MRD post-HCT. These data illustrate the potential utility of the PCR-NGS FLT3-ITD MRD assay in the management of pts with FLT3-ITD AML.

Background: Most older patients (pts) with relapsed or refractory (R/R) AML cannot tolerate intensive treatment and are not eligible for curative allogeneic hematopoietic cell transplant (alloHCT). 131I-apamistamab, an anti-CD45 radioimmunoconjugate, delivers high dose targeted radiation to hematopoietic cells, allowing for myeloablation and eradication of leukemic cells. 131I-apamistamab led induction and conditioning can provide these pts with access to alloHCT. Methods: SIERRA (NCT02665065) was a multi-center, randomized, controlled phase 3 study comparing the efficacy of 131I-apamistamab led induction and conditioning versus physician's choice of conventional care (CC) in pts ≥55 years of age with active, R/R AML. Pts were randomized (1:1, n=153) to CC or 131I-apamistamab with fludarabine and total body irradiation (2 Gy) followed by alloHCT. Primary endpoint was durable complete remission (dCR), defined as CR ≥6 mos with or without platelet recovery (CRp). Pts not achieving CR in CC could crossover (CO) to receive 131I-apamistamab. Treatment with 131I-apamistamab was prescribed as an individualized dose for each pt following a dosimetric infusion and biodistribution analysis, with the prescribed activity of 131I set to deliver a maximum estimated radiation dose to the liver as the dose limiting organ of 24 Gy (determined as the maximum tolerable dose (MTD)) in earlier studies), thereby providing the highest dose to the diseased bone marrow and circulating leukemic cells without exceeding individual organ radiation dose tolerances. Here, we performed an analysis to determine whether a radiation dose-response relationship could be established based on either 1) administered radiation dose to liver and the rate of achieving the primary endpoint of dCR (with 95% binomial confidence intervals) or 2) dCR rate and the ratio of bone marrow/liver absorbed radiation dose as an indicator of favorable biodistribution between target/risk organ. Results: All pts receiving the therapeutic dose of 131I-apamistamab (n=66) were able to receive an alloHCT, and the primary endpoint (dCR) was achieved in 13 pts receiving 131I-apamistamab vs 0 pts in CC (p<0.0001). The median absorbed liver dose was 21.6 Gy (IQR: 21.0 - 22.8 Gy), median absorbed dose to bone marrow was 16.0 Gy (IQR: 11.5 - 22.3 Gy), and median ratio between marrow/liver dose was 0.79 (IQR: 0.52 - 1.02). Table 1 shows the distribution and rate of dCR for pts who received above or below 22 Gy administered dose to liver (24 Gy to liver was the established MTD from earlier studies, in which doses were escalated in 2 Gy increment and thus, 22 Gy represented MTD-1), and for pts stratified by marrow/liver dose ratio (higher ratio representing more favorable biodistribution). Pts with ≤22 Gy liver dose had a dCR rate of 13.5% vs. 27.6% for pts with >22 Gy liver dose, and pts with <median dose to bone marrow had a dCR rate of 12.1% vs. 27.3% for pts with ≥median estimated marrow dose. Importantly, pts with increasingly favorable biodistribution had nominally higher dCR rates of 10.0%, 20.0% and 26.9% for a marrow/liver ratio <0.6, 0.6-0.9 and >0.9, respectively, as illustrated in Figure 1. Taken together this demonstrates a radiation dose-response for the primary endpoint in this study. Overall, safety was similar across doses to the liver. Conclusion: 131I-apamistamab led induction and conditioning followed by alloHCT resulted in statistically significant improvement in dCR at 6 mos vs conventional care. A dose-response was demonstrated for pts receiving 131I-apamistamab with those receiving a liver dose closer to the MTD of 24 Gy having about twice the dCR rate compared to pts 22 Gy (MTD-1) or less. We also found pts with higher marrow/liver ratio experienced considerably higher rates of dCR which highlights the importance of maximizing the dose to target tissues, within the limits of established risk organ dose tolerances.

Background: Ph-like ALL is a high-risk subset of B-lineage ALL that responds poorly to standard chemotherapy regimens. While allogeneic hematopoietic cell transplantation (HCT) is routinely recommended for patients with Ph-like ALL, large multicenter data addressing its role is limited. Here, we conducted a multicenter study to evaluate the role of HCT in adolescent and young adult (AYA) and adult patients (pts) with Ph-like ALL and compare their outcomes to pts with Ph-pos and Ph-neg (non-Ph-like) ALL. Methods: The study included pts (age ≥18) with B-lineage ALL who received HCT in first complete remission (CR1) at 3 academic centers (MD Anderson Cancer Center, City of Hope and Mayo Clinic) between 2006 and 2021. Pts were divided into 3 main subgroups: Ph-like, Ph-pos and Ph-neg ALL. Ph-like ALL cases were identified using cumulative data obtained by targeted RNA sequencing, conventional cytogenetics, FISH studies, whole genome chromosomal microarray and flow cytometry. Measurable residual disease (MRD) prior to HCT was measured using flow cytometry with a sensitivity of 0.01% for Ph-neg and Ph-like ALL, and with a PCR assay for BCR::ABL1 for Ph-pos ALL. MRD-negativity was defined as <0.01%. Predictors of relapse were evaluated in univariate and multivariate analysis using Fine-Gray sub-distribution hazard models considering non-relapse mortality (NRM) as a competing risk. Results: 673 patients were identified, 83 (12.3%) had Ph-like ALL, of which 73% had CRLF2-alterations. The remaining patients were classified either as Ph-pos (N=271, 40.3%) or Ph-neg ALL (N=319, 47.4%). Patients with Ph-like ALL were younger (median age 40 years vs 48 and 46 years for Ph-pos and Ph-neg ALL, respectively, p=0.01), more likely to be of Hispanic ethnicity (69% vs 32% and 48% for Ph-pos and Ph-neg ALL respectively, p<0.001). Patients with Ph-like ALL were also more likely to have induction failure (24% vs 2% and 11% for Ph-pos and Ph-neg ALL, respectively, p=0.001). A summary of pts characteristics for each subgroup can be found in Table 1. The median follow-up was 30, 52 and 48 months for patients with Ph-like, Ph-pos and Ph-neg ALL, respectively. In univariate analysis, 3-yr OS and PFS did not differ significantly (66% and 59% vs 59% and 54%, p=0.1) between patients with Ph-like and Ph-neg ALL, respectively. Similarly, the cumulative incidence of relapse was comparable between patients with Ph-like and Ph-neg ALL (22% vs 20%, respectively, p=0.7). In contrast, the 3-yr OS (75%, p<0.001) and PFS (70%, p=0.001) were significantly higher and relapse rate significantly lower (12%, p=0.003) in pts with Ph-pos ALL compared with Ph-like or Ph-neg ALL (Figure 1). The inferior survival in patients with Ph-like and Ph-neg ALL was likely attributable to higher relapse rates (22% and 20% vs 12%, respectively, p=0.003) compared to Ph-pos ALL. Finally, the 3-yr NRM rate in patients with Ph-like ALL was comparable (20% vs 18% and 23%, respectively, p>0.1) to Ph-pos and Ph-neg ALL. We confirmed that the relapse rate in patients with Ph-like ALL was comparable to the rate in pts with Ph-neg ALL (HR=1.02, 95% CI 0.6-1.8, p=0.9) on multivariate analysis (MVA). Ph-pos ALL was associated with a significantly lower relapse rate compared with Ph-neg ALL (HR=0.4, 95% CI 0.3-0.7, p<0.001). MRD positivity prior to allo-HCT was the only additional significant predictor for relapse on MVA (HR=2.9, 95% CI 1.8-4.7, p<0.001). Conclusions: HCT appears to overcome the poor prognosis of patients with Ph-like ALL. Despite higher induction failure rates, patients with Ph-like ALL undergoing HCT in CR1 had similar outcomes to other Ph-neg ALL subsets. The impact of effective salvage therapy leading to MRD-negativity in CR1 in this group is under investigation and will be reported. Further studies are needed to assess the impact of different Ph-like alterations on HCT outcomes. Outcomes for patients with Ph-pos ALL are significantly better than other groups, likely due to the use of TKIs pre- and post-HCT.

Background With the advent of tyrosine kinase inhibitor (TKI)-based therapy the outcome of Philadelphia chromosome-positive (Ph +) acute lymphoblastic leukemia (ALL) has significantly improved. Moreover, patients (pts) who achieved undetectable (u) BCR::ABL1 at 3 months (mo) from the time of initiating therapy, tend to have improved outcomes (Short et al. Blood 2016; 128 [4];504-507). We conducted a retrospective multicenter study to analyze the predictors of achieving uBCR::ABL1 in adult (> 18 years) Ph + ALL pts. Methods A total of 458 adult Ph + ALL pts from 10 US academic institutions, who were diagnosed between May 2003 and April 2023, were evaluated to assess the rates of uBCR::ABL1 at 3 mo after initiating therapy and factors that predict the achievement of molecular response. Continuous variables were summarized as median (range), while categorical variables were reported as frequency (percentage). Unadjusted comparisons of patient/ treatment related characteristics and molecular response were made using a non-parametric test (continuous variables) or Fisher's exact test (categorical variables). Survival curves were estimated using the Kaplan-Meier method and compared between molecular responses via the log-rank test. Results Baseline characteristics The median age of the pts was 53 years (range [R] 19-85), median white blood cell count (WBC) was 17.6 (10 9/L) (R, 1.2-571), 249 (55%) pts had additional cytogenetic (CG) abnormalities, 98 (22%) pts had p210 fusion protein, and 35 (8%) pts had CNS disease at diagnosis. A higher proportion of pts received TKI in combination with intensive chemotherapy (IC) (69%), than TKI + steroids (15%), TKI + low intensity chemotherapy (LC; combination of vincristine, steroids +/- rituximab) (9%), TKI + blinatumomab (3%) and IC without TKI (4%) during induction. During induction, the highest proportion of pts received 2 nd generation TKI combinations (67%; majority dasatinib 98%), compared to imatinib (19%) or ponatinib (10%): 4% of pts did not receive TKI with induction. The complete remission (CR) rate was 94%; measurable residual disease (MRD) negative by flow cytometry 51% (181/352 evaluable pts) and 58.5% (209/354 evaluable pts) of pts had uBCR::ABL1 by PCR at 3 mo from induction. A total of 212 (46%) pts received allogeneic stem cell transplantation (allo-HCT) in CR1, among them 61% of pts received TKI maintenance post allo-HCT. Predictors of uBCR::ABL1 at 3 months Predictors for uBCR::ABL1 at 3 mo are summarized in Table 1. The proportion of pts with WBC > 100 (45% vs 60%, p= 0.04), p210 fusion protein (35% vs 85%, p= <0.001), additional CG abnormality with monosomy 7 (40% vs 61%, p= 0.03), additional somatic myeloid co-mutation (39% vs 61%, p= 0.09; 94 pts evaluable), received induction with steroid plus TKI (41% vs 63%, p= 0.002) or received imatinib based induction (47% vs 61%,p= 0.07) had inferior uBCR::ABL1 rates at 3 mo. Conversely pts who received ponatinib based induction had significantly better uBCR::ABL1 rates (79% vs 57%, p= 0.009). Survival The median relapse free survival (RFS) was 72.3 mo (95% CI; 50.1-94.4), RFS was numerically higher among pts who achieved uBCR::ABL1 at 3 mo but not statistically significant (84.2 mo vs 62.9 mo, p= 0.20). The median overall survival (OS) was 129.2 mo (95% CI; 81.2-177.2), OS was not significantly different among pts who achieved uBCR::ABL1 at 3 mo (129.2) versus no uBCR::ABL1 (not reached; 50% alive at 94 mo), p= 0.47. In sub-set analysis, among non-transplanted pts, RFS was not statistically significant between those who achieved uBCR::ABL1 vs those who did not (38.9 vs 22.0 mo, respectively p= 0.24). However, the median OS was significantly higher in pts who achieved uBCR::ABL1 at 3 mo compared to pts who did not achieve uBCR::ABL1 at 3 mo: 149.3 vs 77.5 mo, p= 0.04 ( Figure 1). Conclusion In this large multicenter study, we observed significantly inferior uBCR::ABL1 rates among pts with WBC > 100, p210 fusion protein, monosomy 7 CG abnormality, pts receiving induction with steroid plus TKI or imatinib based induction. Conversely, uBCR::ABL1 rates were significantly better in pts receiving ponatinib based induction therapy. Overall, we did not see significant impact of uBCR::ABL1 at 3 mo, on RFS or OS. It is likely subsequent therapies and allo-HCT abrogates the negative impact of not achieving uBCR::ABL1 at 3 mo. However, among non-transplanted pts achievement of uBCR::ABL1 at 3 mo showed a significantly improved OS.

Introduction: Next-generation sequencing (NGS) represents a promising modality for measurable residual disease (MRD) testing in patients with acute myeloid leukemia (AML), but systematic efforts are required to define the appropriate targets, test characteristics, and clinical implications. We recently reported that detection of persistent NPM1 or FLT3 internal tandem duplication (ITD) mutations in adults with AML in first complete remission (CR1) prior to allogeneic hematopoietic cell transplantation (alloHCT) is associated with increased relapse and death compared with those testing negative (JAMA 2023, PMID: 36881031). FLT3 tyrosine kinase domain (TKD) variants are present in 7-10% of adult patients with AML but the utility of this target for AML MRD testing in CR1 prior to alloHCT was previously unknown. Methods: Patients aged 18 or older who underwent alloHCT for AML in CR1 between 2013 and 2019 reported to have a FLT3-TKD variant at diagnosis and a pre-conditioning remission blood sample available in the CIBMTR biorepository were eligible for this study. Single-amplicon NGS (SA-NGS) library generation utilizing UMI-tagged primers targeting the D835 and I836 codons of FLT3 was performed on 400ng genomic DNA sequenced with unique dual indices, with error-corrected variant calling. Primary outcomes were overall survival (OS) and cumulative incidence of relapse (CIR). Secondary outcomes were relapse-free survival (RFS) and non-relapse mortality (NRM), with NRM being considered as a competing risk. Results: A total of351 patients met inclusion criteria for this study, and 342 (97.4%) had sufficient DNA for SA-NGS and were analyzed for clinical outcomes. Of these patients, 81 (23.7%) experienced relapse at a median of 5 months post-alloHCT. Using the current ELN-recommended variant allele fraction (VAF) threshold of 0.1% for NGS-based AML MRD, 14 patients (4.1%) tested positive. Patients with FLT3-TKD MRD VAF≥0.1% experienced higher rates of relapse (73.8% vs. 20.5%, p < 0.0001) and decreased OS (11.4% vs. 66.8%, p < 0.001) compared to VAF<0.1%. The highly sensitive SA-NGS method allowed for exploration to determine if a deeper VAF threshold of 0.01% (previously validated for NPM1 and FLT3-ITD in the Pre-MEASURE study) was prognostic for FLT3-TKD as an MRD target. This led to an additional 20 patients being reclassified as positive (n = 34 with VAF above 0.01%, =9.9%). However, there was no significant difference in CIR (33.6% vs. 19.7%, p = 0.089) or OS (65.2% vs. 66.9%, p = 0.267) between patients with 0.01%≤VAF<0.1% and VAF<0.01% ( Figure). Out of 28 SA-NGS positive calls, 6 could not be validated by orthogonal digital droplet PCR testing, and interestingly, none of those 6 patients experienced relapse. Site-reported pre-transplant remission multiparametric flow cytometry (MFC) was available for 335 patients (98%). Results of MFC were not associated with differences in CIR (HR 1.2, 95% CI 0.52 - 2.76, p = 0.67) or OS (HR 0.78, 95% CI 0.34 - 1.77, p = 0.551).In multivariable analysis, FLT3-TKD NGS-MRD VAF≥0.1% remained prognostic for CIR (HR 6.1, 95% CI 3.1 - 12.0, p < 0.001) and OS (HR 3.2, 95% CI 1.7 - 5.9, p < 0.001). Conclusions: Detectable persistence of FLT3-TKD variants in blood from AML patients in first remission prior to first alloHCT is rare, but at a VAF level of 0.1% was strongly associated with increased relapse and death after transplant compared to those testing negative. In contrast to FLT3-ITD NGS-MRD, no evidence was found to support lowering the NGS-MRD VAF threshold from 0.1% to 0.01% for FLT3-TKD. Alternative methodology may further improve NGS-MRD test performance. Site-reported MFC prior to alloHCT did not stratify for post-transplant relapse or survival. Optimal NGS-MRD testing for patients with AML will likely require a multi-target approach and testing for persistence of FLT3-TKD MRD is shown here to represent an important component of this strategy.

Background : Most older patients (pts) with relapsed or refractory (R/R) AML cannot tolerate intensive treatment and are not eligible for curative allogeneic hematopoietic cell transplant (alloHCT). 131I-apamistamab, an anti-CD45 radioimmunoconjugate, delivers high dose targeted radiation to hematopoietic cells, allowing for myeloablation and eradication of leukemic cells while sparing toxicity to healthy organs. 131I-apamistamab led induction and conditioning can thus provide these pts with access to alloHCT. Methods: The SIERRA trial (NCT02665065) is a multi-center, randomized, controlled Phase 3 study comparing the rate of durable complete remission (dCR) lasting >6 months (mos) after complete remission with/without platelet recovery (CR/CRp) between two groups: 131I-apamistamab led induction and conditioning followed by alloHCT vs physician's choice of conventional care (CC). Pts were randomized (1:1) to CC or 131I-apamistamab with fludarabine and total body irradiation (2 Gy) followed by alloHCT. CR/CRp assessment was 28-56 days post alloHCT or 28-42 days post initiation of therapy on the CC group. Pts in the CC group not achieving leukemia-free state could crossover (CO) to 131I-apamistamab. Here we report the results of a post hoc analysis to determine if the presence of various risk factors (i.e., Karnofsky Performance Status (KPS) <90, HCT Comorbidity Index >3, age >65, adverse-risk cytogenetics, venetoclax failure prior to randomization) influenced achievement of dCR in pts treated with 131I-apamistamab. Results: In total, 153 pts were randomized (CC, n=76; 131I-apamistamab, n=77). All pts who received the therapeutic dose of 131I-apamistamab (n=66) underwent alloHCT vs 14 (18.2%) in the CC group. Of evaluable pts, dCR rates at 6 mos were 22% in the 131I-apamistamab group vs 0% in the CC group (95% CI;12.29, 34.73; p<0.0001). A total of 19 pts (13, 131I-apamistamab group; 6, CO) achieved dCR. Table 1 shows the pt, disease, and transplant characteristics of dCR vs. non-dCR pts. Eight of the 19 pts (42.1%) had primary induction failure and the median time to randomization from diagnosis was 5.4 mos. Approximately 50% of pts (9/19) had adverse-risk cytogenetics. Over 35% (7/19) of the pts had failed prior venetoclax and the median number of prior treatments was 3 (range 1-5). A total of 10/19 (52%) pts had KPS of <90% and a comorbidity index of >3. There was no difference in the rate of dCR when stratifying by number of risk factors for a given pt (0 to 5) with 27% for pts with 0-1 risk factors, 14% for pts with 2-3 risk factors, and 11% for pts with 4-5 risk factors (p=0.251). Conclusion: Pts with R/R AML who have multiple risk factors such as adverse risk cytogenetics, age >65, venetoclax failure, high comorbidity index or poor KPS are typically not considered for alloHCT due to high transplant-related mortality and post-transplant relapse rates. 131I-apamistamab was effective in achieving durable responses in R/R AML pts irrespective of the presence of multiple risk factors and successfully enabled alloHCT in such pts due to its targeted mechanism of action.

Introduction: Detection of measurable residual disease (MRD) prior to allogeneic hematopoietic cell transplant (alloHCT) in patients with acute myeloid leukemia (AML) in remission is associated with increased relapse and inferior survival after transplant. We recently reported that next-generation sequencing (NGS) detection of mutated NPM1 or FLT3-ITD in first complete remission (CR1) blood prior to alloHCT is strongly associated with increased relapse and death (PMID: 36881031). However, the prognostic implications of detecting persistence of other common AML-associated mutations at this treatment landmark remain incompletely defined. Around 20% of patients diagnosed with AML have variants in isocitrate dehydrogenase ( IDH) genes (~7% for IDH1) and clinical outcomes are known to differ due to co-occurrence of other mutations, IDH subtypes, treatment strategies, and patient factors. Studies have examined the relationship between residual IDH1 variants ( IDH1m) in CR and subsequent clinical outcomes, but definitive standardized evidence to inform decision-making for the clinical utility of IDH1m as a target for AML MRD testing was not previously available. Methods: Adult patients undergoing alloHCT for IDH1 mutated AML in CR1 at CIBMTR sites in the USA between 2013-2019 were eligible for the study if remission blood sample collected within 100 days before alloHCT and outcome data were available. Ultrasensitive error-corrected NGS (NGS-MRD) targeting IDH1, NPM1, and FLT3 genes was performed on DNA from pre-conditioning blood to identify residual variants. Detected IDH1m were validated by digital droplet PCR (ddPCR). The day of transplant was considered as day 0. Median follow-up time was calculated for censored patients. Overall survival (OS) was estimated using Kaplan-Meier (log-rank tests) and Cox proportional hazards models (forward selection); cumulative incidence of relapse was examined using Fine-Gray regression models with non-relapse mortality as a competing risk. Results: A total of 148 patients were included in the study, with a median follow-up time of 24 months. Relapse was observed in 37 (25%) patients, with the majority occurring within 1 year after alloHCT (n = 28, 76%). Pre-transplant flow cytometry results were reported by the transplant centers for 97% of patients (n = 144), but no statistically significant differences in clinical outcomes were observed between the positive and negative groups (OS: p = 0.3; relapse: p = 0.07). 53 patients (36%) tested positive for IDH1m persistence in CR1, with a 100% validation rate by ddPCR for variants with an orthogonal assay available. Logistic regression indicated that IDH1m persistence in remission was more likely to occur in patients older than 40 years (OR: 1.02 with every 1-year increase after 40, p = 0.04). Clinical outcomes showed no statistically significant difference between the pre-alloHCT IDH1m positive and negative groups (Figure 1A), and this remained true after further subgroup analysis by age. The variant allele fraction (VAF) of the residual IDH1m ranged from 0.09% to 48.5% (median: 1.5%), and no differences in clinical outcomes were observed when stratifying by high (≥2.5%) or low VAF (<2.5%). Among patients with co-mutated NPM1 and/or FLT3-ITD at baseline (n=69), those with residual mutated NPM1 and/or FLT3-ITD in pre-transplant remission had significantly higher rates of relapse compared to those with residual IDH1m only or NGS-negative groups (p = 0.01, Figure 1B). The cohort excluding patients with co-mutated NPM1 and/or FLT3-ITD did not show any significant clinical outcome differences based on IDH1m persistence. Conditioning intensity did not show an association with clinical outcomes, and multivariable analysis for overall survival and relapse did not identify residual IDH1m as an important marker. Conclusion: Detection of persistent IDH1 variants in the blood of adult patients with AML in CR1 prior to first alloHCT is common, but not associated with increased relapse or death after transplant compared with those testing negative. For those patients with IDH1 mutated AML co-mutated with either NPM1 and/or FLT3-ITD, detection of persistent NPM1 and/or FLT3-ITD was associated with higher rates of relapse. These findings, from the largest study to date, do not support the detection of isolated IDH1 variants in CR1 blood prior to alloHCT as evidence of AML MRD or increased post-transplant relapse risk.

Background TP53-mutated (m) acute myeloid leukemia (AML) is an adverse risk disease, which continues to have dismal outcome in the contemporary era of novel therapies. Venetoclax (VEN) in combination with hypomethylating agents (HMA; azacitidine or decitabine) have shown to improve complete remission (CR) rates compared to HMA alone, however no significant differences in overall survival (OS) have been observed. Methods We conducted a retrospective study through the COMMAND consortium (a collaboration of acute leukemia experts from 10 US academic institutions) to analyze the impact on survival outcomes of HMA plus (+) VEN compared to HMA-based therapies in adult (≥18 years) pts with TP53m AML. The AML Database of 381 pts with TP53m AML diagnosed between 2012-2022 was queried and 154 (40%) (HMA [n=50] and HMA +VEN [n=104]) pts were identified eligible for this analysis. Continuous variables were summarized as median (range) while categorical variables were reported as frequency (percentage). The Kaplan-Meier method was used to estimate OS, defined as time from diagnosis to death or last follow-up. Logistic regression models were used to determine the univariate and multivariate of OS. Multivariable models included all significant univariate predictors. All tests were two-sided with a p value <0.05 considered statistically significant. Results Baseline characteristics and responses The median age (years [yrs]) was 74 (range [R], 38-87) and 71 (R,29-88) yrs in the HMAand HMA + VEN groups (gps) (p= 0.23), respectively ( Table 1). Thirty-four (68%) and 57 (55%) patients (pts) were ≥ 70 yrs in the HMA and HMA + VEN gps, respectively (p= 0.16). Twenty-six percent vs 31% ( p=0.57), 86% vs 90% ( p=0.58) and 68% vs 73% ( p=0.56) of pts had secondary AML, had complex cytogenetics (CG), and had multi-hit TP53m in the HMA and HMA + VEN gps, respectively. The proportion of pts with most frequently occurring (> 5%) myeloid co-mutations ( RUNX1, ASXL1, TET2, DNMT3A, RAS, PTPN11 and JAK2) were not significantly (NS) different between the HMA and HMA + VEN gps, as shown in Table 1. The early treatment related mortality in first 30 days of induction was higher in the HMA+VEN gp (25%) when compared with the HMA gp (6%), p=0.004. The complete remission rates (CR/CRi) were higher in the HMA + VEN gp compared to the HMA gp (35% vs 18%, p=0.05). Similarly, the proportion of pts receiving allogeneic stem cell transplant (allo-HCT) after induction was higher in the HMA + VEN gp when compared with the HMA gp (13% vs 4%; p=0.14). Duration of response and overall survival The median follow up was 9.8 months (mo) (R, 1-56) and 5.80 mo (R, 1-30.4) in the HMA and HMA + VEN gps, respectively ( p=0.12). The median duration of response (DOR) was higher in the HMA +VEN gp (15.6 mo) when compared with the HMA gp (7.93 mo), but NS ( p=0.26). The median OS was NS different when comparing the HMA and HMA + VEN gps (9.23 vs 7.3 mo, p=0.80 [ Figure 1]). Multivariable analysis for duration of response and overall survival We did multivariable analysis (MVA) for DOR and OS using baseline variables ( Table 1) that showed significance/trend towards significance in univariate analysis ( p<0.1). Allo-HCT (HR; 0.11, 95% CI: 0.02-0.52, p=0.005) retained significance for better DOR, whereas age ≥ 70 yrs (HR; 1.7, 95% CI: 0.63-4.65, p=0.28) and complex CG (HR; 3.05, 95% CI: 0.79-11.73, p=0.10) did not retain significance in MVA. In MVA for OS, achievement of CR/CRi (HR; 0.35, 95% CI; 0.20-0.59, p=<0.001) and allo-HCT after induction (HR; 0.06, 95% CI; 0.09-0.46, p=0.007) retained favorable significance for OS, whereas complex CG (HR; 2.36, 95% CI: 1.11-4.99, p=0.02) retained significance for shorter OS. PTPN11m (HR; 1.57, 95%CI: 0.75-3.28, p=0.23) and TET2m (HR; 1.02, 95% CI: 0.56-1.87, p= 0.92) did not retain significance in MVA. Conclusion In this multi-center real-world study, while HMA+VEN led to improvement in CR rates and a higher proportion of pts were bridged to allo-HCT, it did not associate with an improvement in OS when compared with HMA monotherapy. More effective treatment strategies are warranted to improve outcome of pts with TP53m AML.