Nakhleh, Merry

Abstract Background In our first exploration of mutations in essential thrombocythemia (ET; N=183), we described somatic variants other than JAK2/CALR/MPL in 53% (N=96) of patients with the most frequent being TET2 (16%), ASXL1 (11%), DNMT3A (6%), and SF3B1 (5%) [Blood Adv 2016;1:21]. In a subsequent report (N=270; BJH 2020;189:291), we highlighted the prognostic relevance of SF3B1, SRSF2, and EZH2 mutations for overall survival (OS), SRSF2, EZH2, and TP53 for leukemia-free survival (LFS), and SF3B1 and U2AF1 for myelofibrosis-free survival (MFFS). We have now expanded our NGS-informative patient cohort to 500 and examined the distribution and frequencies of specific mutations at different stages of the disease and corresponding prognostic correlates. Methods Molecularly-annotated ET patients diagnosed at the Mayo Clinic between January 1980 and May 2025 were considered. Diagnoses of chronic phase ET (ET-CP), post-ET MF, and post-ET acute myeloid leukemia (AML) were retrospectively fitted to be in accordance with the International Consensus Classification criteria (Blood 2022;140:1200). Mutation and cytogenetic analyses were performed according to institutional protocols for clinical use that included multi-gene next-generation sequencing (NGS); in a subset of patients, historically archived samples were accessed for analysis in a research setting. Patients were monitored until death or time of last follow-up, determined through electronic medical records or direct communication with patients or healthcare providers. Conventional statistical methods were utilized (JMP Pro 17.1.0, SAS Institute, Cary, NC, USA). Results A total of 500 ET patients (median 57 years; Females 61%) were informative for NGS-derived mutations that were obtained in the chronic (N=401; 80.2%; group A), post-ET MF (N=88; 17.6%; group B), or post-ET AML (N=11; 2.2%; group C) phase of the disease; group C included 9 (1.8%) patients who transformed directly to AML without transitioning through the MF phase. The most frequent mutations with at least 1% (N=5) mutational frequency were (approximately) TET2 (10%), ASXL1 (10%), DNMT3A (7%), SF3B1 (4%), TP53 (5%), SRSF2 (3%), EZH2 (2%), U2AF1 (2%), CBL (1%), IDH1 (1%), IDH2 (1%), RUNX1 (1%), ZRSR2 (1%), and PPM1D (1%). Comparison of distribution of mutations and their frequencies based on timing of NGS (AML phase vs. MF phase vs. ET-CP) revealed significant differences for TP53MUT (72.7% vs. 11.4% vs. 1%; p<0.01), ASXL1MUT (18.2% vs. 25% vs. 7.2%; p<0.01), and EZH2MUT (18.2% vs. 6.8% vs. 0.3%; p<0.01), respectively; borderline significance was also apparent for RUNX1MUT (9.1% vs. 0% vs. 1.3%; p=0.10) and ZRSR2MUT (9.1% vs. 2.3% vs. 0.5%; p=0.08). Notably, among 9 patients in whom NGS was performed at the time of direct transition from ET to AML, without an intermediary MF phase, 8 (88.9%) displayed TP53MUT and only one (11.1%) EZH2MUT and none (0%) ASXL1MUT. At a median follow-up of 66 months, 93 (18.6%) deaths, 26 (5.2%) leukemic transformations (LT), and 108 (21.6%) MF transformations were reported. In group A patients, age-adjusted multivariable analysis (MVA) of mutations identified ASXL1MUT (N=29; HR 3.8, 1.9-7.7; p<0.01) and SF3B1MUT (N=12; HR 3.9, 1.4-11.0; p=0.01) to be associated with inferior OS, ASXL1MUT (HR 4.9, 1.01-24.2; p=0.048) and SRSF2MUT (N=11; HR 11.9, 1.3-104.8; p=0.03) with inferior LFS, and ASXL1MUT (HR 9.0, 2.8-29.5; p<0.01) and SF3B1MUT (HR 15.5, 3.1-77.5; p<0.01) with inferior MFFS, all calculated from time of NGS. A similar analysis in group B patients identified ASXL1MUT (N=22; HR 3.7, 1.2-11.7; p=0.03) to be associated with inferior OS and TP53MUT (N=10; HR 13.4, 1.7-108.6; p=0.02) with inferior LFS. Age-adjusted MVA combining groups A and B (N=489) confirmed the association between inferior OS and ASXL1MUT (N=51; p<0.01) and SF3B1MUT (N=18; p<0.01) while inferior LFS was associated with TP53MUT (N=14; p<0.01) and ASXL1MUT (p=0.03). Conclusions The majority (⁓73%) of patients with post-ET AML harbor TP53MUT, at the time of LT. By contrast, TP53MUT are infrequent in ET-CP (⁓1%) and did not appear to predict LT, as was the case with post-ET MF. ASXL1MUT was shown to be an independent risk factor for OS, MFFS, and LFS in ET-CP and for OS in post-ET MF. SF3B1MUT was associated with inferior OS and MFFS in ET-CP and SRSF2 with inferior LFS in ET-CP. Identifying ET patients with adverse mutations allows for closer surveillance and potentially more tailored therapeutic interventions.

Abstract Background: The recently published BLAST clinical and BLAST molecular (BLAST-Mol) risk models for chronic myelomonocytic leukemia (CMML) incorporate clinical and molecular variables, respectively, for prediction of overall survival (OS), while BLAST-BT (blast transformation) risk model was predictive of BT free survival (BTFS) [Blood 2025;doi:]. In the current study, the Mayo Clinic (USA) CMML cohort used in the aformentioned study was expanded from 457 to 600 patients and merged with an external cohort of 501 patients from the Humanitas Research Hospital (HRH; Italy) in order to validate the BLAST, BLAST-Mol and BLAST-BT risk models. Methods: CMML diagnosis and sub-categorization were according to the International Consensus Classification, which included CMML-1 (BM/PB blasts <10%/<5%) vs. CMML-2 (BM/PB blasts ≥10%/≥5%) and CMML-MP (leukocytes ≥13 x 10⁹/L) vs. CMML-MD (leukocytes <13 x 10⁹/L) [Blood 2022;140:1200]. The BLAST score is based on clinical risk factors including PB blasts ≥2%, leukocytes ≥13 x 109/L, and severe or moderate anemia resulting in BLAST low, intermediate, and high risk groups. BLAST-Mol also accounts for mutations and karyotype and identifies PHF6 and TET2 as “favorable” and DNMT3A, U2AF1, BCOR, SETBP1, ASXL1, NRAS, PTPN11, RUNX1, TP53, as “unfavorable” mutations with adverse karyotype also considered unfavorable. BLAST-BT variables include absence of PHF6MUT, BM blasts ≥10%, presence of DNMT3AMUT, PB blasts ≥2%, leukocyte count ≥13 x 109 /L, and ASXL1MUT. OS analyses were censored for allogeneic stem cell transplant (ASCT). Conventional statistical methods were employed, using JMP Pro 18.0.0 software SAS Institute, Cary, NC, USA. Results: A total of 1,101 patients (median age 73 years; 67% males) with newly-diagnosed CMML were considered: Mayo cohort 600 and HRH cohort 501. Compared to the HRH cohort, the Mayo cohort was more likely to include CMML-MP (44% vs. 28%; p<0.01), CMML-1 (87% vs. 81%; p<0.01), ASXL1MUT (48% vs. 35%; p<0.01), SRSF2MUT (43% vs. 33%; p<0.01), NRASMUT (15% vs 9%; p<0.01), PHF6MUT (7% vs. 4%; p<0.01), and less likely to harbor TET2MUT (50% vs. 58%; p<0.01), and SF3B1MUT (4% vs. 10%; p<0.01) [Table 1]. At a median follow-up of 55 months (range 0-202), 686 (62%) deaths, 218 (20%) BTs, and 142 (13%) ASCTs were documented. CPSS-mol risk categories at diagnosis included low (N=190; median OS 67 months), intermediate-1(N=271; median 38 months), intermediate-2 (N=407; median 26 months), and high (N=231; median 16 months; p<0.01) [AUC at 5 years 0.75]. BLAST clinical risk distribution was low (N=356; median OS 65 months), intermediate (N=393; median 30 months), and high (N= 352; median 16 months; p<0.01) [Figure 1; AUC at 5 years 0.79]. The corresponding AUCs calculated for the Mayo and HRH cohorts, separately, were 0.81 and 0.76, respectively. BLAST-Mol risk assignments included low (N=357; median OS 66 months), intermediate (N=512; median 27 months) and high (N=232; median 15 months; p<0.01) [Figure 2; AUC at 5 years 0.79]. The corresponding AUCs for the Mayo and HRH cohorts were 0.81 and 0.77, respectively. The 3-year cumulative incidence of BT (accounting for death from another cause) stratified by BLAST-BT risk model included 8% for very low risk (N=54), 11% for low risk (N=332), 20% for intermediate-risk (N=281), 27% for high risk (N=250), and 34% for very high risk (N=176) [p<0.01; Figure 3]. Multivariable analysis of the combined Mayo-Humanitas cohort (N=1,101) identified age >65 years (HR 1.7, 95% CI 1.3-2.1), CMML-2 subtype (HR 2.0; 95% CI 1.6-2.5), and a number of other mutations or co-mutations with BLAST-Mol-independent and CPSS-mol-independent prognostic contribution (to be elaborated further at the time of meeting presentation). Conclusions: The current study validates the predictive performance of the BLAST, BLAST-Mol and BLAST-BT risk models in a large two-center international cohort of patients with CMML. The possibility of further enhancement of the BLAST-Mol model is suggested by the demonstration of independent prognostic relevance from additional mutations and co-mutations.

Abstract Background Blast transformation (BT) occurs in approximately 15-30% of patients with chronic myelomonocytic leukemia (CMML) and remains a leading cause of death. Allogeneic stem cell transplantation (ASCT) is currently the only treatment modality with the potential to cure the disease or prolong survival. Optimal timing of ASCT is critical for maximizing benefit while minimizing risks. To that end, contemporary risk models have focused on the prognostic relevance of individual, as opposed to concurrent mutations. In the current study, we looked into the possibility of prognostic prominence from concurrent mutations in predicting BT in CMML. Methods CMML diagnostic criteria were according to the International Consensus Classification (Arber et al. Blood 2022; 140:1200). A machine-learning hierarchical clustering algorithm was developed and tailored for patient stratification using survival outcomes and co-expression of genomic alterations. To reduce complexity and improve interpretability, we generalized each cluster using existing rule induction algorithms, including the in Trees framework and JRip. The final output was a set of mutation-based cluster definitions, each representing a distinct patient subgroup with similar survival trajectories. Competing risk analysis and cumulative incidence functions were used in downstream evaluation to validate the clinical distinctiveness of the clusters. For all survival analysis, patients were censored at the time of ASCT. Time-to-BT was calculated from the date of diagnosis to the date of BT or ASCT or last contact. BT-free-survival was calculated from the date of diagnosis to the date of BT, ASCT, death, or last contact. Results The core study cohort included 605 patients from the Mayo Clinic, USA, and the external validation cohort 501 patients from Humanitas Cancer Center, Milan, Italy. Using the patient cohort from the Mayo Clinic, machine-learning algorithms identified five molecular clusters with 3-year blast transformation (BT) rates ranging from 0% to 100% (AUC at 3 years 0.78): the order of molecular signature assignment (probability of BT/death from another cause) was i) PHF6MUT/ASXL1WT (0%/17% at 3 years; N=32), ii) NPM1MUT OR BCORMUT/ASXL1MUT OR SETBP1MUT/NRASMUT (48%/30% at 1 year; N=24), iii) RUNX1MUT/ASXL1MUT OR SRSF2MUT/NRASMUT OR EZH2MUT/ASXL1MUT OR SETBP1MUT OR BCORMUT (31%/55% at 3 years; N=132), iv) ASXL1MUT/TET2MUT OR DNMT3AMUT OR JAK2MUT (24%/28% at 3 years; N=123), and v) all other permutations (10%/40% at 3 years; N=294) [Figure 1]. Additional analysis confirmed significant differences in survival between RUNX1MUT/ASXL1MUT vs. RUNX1MUT/ASXL1WT (p<0.01) OR RUNX1WT/ASXL1WT (p<0.01) OR RUNX1WT/ASXL1MUT (p=0.038) [Figure 2]. Similar patterns of differences in survival were also documented for NRAS/SETBP1, ASXL1/EZH2, and NRAS/SRSF2 mutation combinations (Figure 2). A subsequent Cox regression analysis confirmed independent prognostic contributions from “PHF6MUT/ASXL1WT” (HR 5.43e-10; p<0.01), NPM1MUT (HR 26.7; p<0.01), “SETBP1MUT/NRASMUT” (HR 12.7; p<0.01), BCORMUT(HR 5.8; p<0.01), “RUNX1MUT/ASXL1MUT” (HR 2.3, p<0.01), JAK2MUT (HR 2.1; p<0.01), and “ASXL1MUT/TET2MUT” (HR 1.7; p=0.02). The prognostic relevance of “SETBP1MUT/NRASMUT”, “RUNX1MUT/ASXL1MUT”, NPM1MUT, and BCORMUT was validated in the external cohort from Italy (N=501). In the Mayo Clinic cohort, presence of any of the latter mutations was associated with 1-, 3-, and 5-year BT (death from other cause) rates of 27% (21%), 44% (49%), and 44% (51%), respectively (Figure 3). The corresponding values in the absence of high risk mutations were 7% (15%), 15% (37%), and 18% (52%) [Gray's p value <0.01 for both BT and death from another cause; Figure 3]. Similarly, the 1-, 3-, and 5-year BT (death from another cause) rates in the Italian cohort were 21% (19%), 37% (47%), and 42% (54%) in the presence and 8% (14%), 20% (35%), and 24% (45%) in the absence of high-risk mutations (Gray's p value <0.01 for BT and 0.18 for death from another cause; Figure 4). Conclusions In the current study, machine-learning algorithms enabled the discovery of concurrent mutations in CMML that were shown to be prognostically more significant than their individual constituents. Such prognostic interaction might have contributed to some of the discrepancies noted in current literature regarding the prognostic relevance of certain mutations in CMML and should be accounted for in the development of future risk models.

Abstract Background Red blood cell distribution width (RDW) is an easily accessible diagnostic test that is underutilized as a prognostic marker, despite substantial supporting evidence in cancer, cardiovascular diseases, and helathy individuals (Pilling et al. PLoS One 2028;13:e0203504). The association between increased RDW and incident acute myeloid leukemia (AML) has long been suggested in healthy volunteers (Abelson et al. Nature 2019;559:400). We looked into a similar phenomenon in patients with polycythemia vera (PV) or essential thrombocythemia (ET) in large discovery and validation cohorts. Methods A machine learning (ML) approach was used to identify baseline clinical and laboratory parameters that were predictive of leukemic transformation (LT) and develop a predictive model in a large discovery cohort of PV and ET, obtained from electronic medical records of the Maccabi Healthcare Services (MHS; Tel-Aviv, Israel). Receiver operating characteristic (ROC) curve analysis was utilized to determine the optimal cutoff points for RDW and other continuous variables. Mayo Clinic databases (Rochester, MN, USA) for patients with PV and ET were utilized as validation cohorts. Identification of study patients in the latter was according to the International Consensus Classification (Arber et al. Blood 2022; 140:1200) while in the discovery cohort was based on presence of ICD-9 codes associated with documentation of disease-specific driver mutations or treatment records that are consistent. Statistical analyses were performed using the R statistical software 4.4.2 (November, 2024) (Foundation for Statistical Computing, Vienna, Austria) or JMP Pro 18.0.0 software (SAS Institute, Cary, NC, USA). Results The discovery cohorts from Israel included 4,592 patients with PV (median age 63 yesrs; 53% males) and 5,968 with ET (median age 61 yesrs; males 40%). The validation cohorts from the Mayo Clinic included 633 PV (median 65 years; 51% males) and 634 ET (median 60 years; females 64%) patients that were informative for RDW. In the discovery cohorts, median follow-up (incident AML) was 6.1 years (N=186; 4.1%) in PV and 6.1 years (N=85; 1.4%) in ET. The corresponding values for the validation cohorts were 6.6 years (N=24; 3.8%) in PV and 7.0 years (N=22; 3.5%) in ET. Discovery cohort analyses Covariate-adjusted Cox regression analysis was utilized to estimate associations between LT and several clinical variables followed by machine learning pipeline to predict LT in a given timeframe. Conversion from continuous to binary variables was accompliched by ROC analysis. Shapley Additive exPlanations (SHAP) application enabled selection of top-ranked features that included RDW >15.5 (HR 3.2, 1.1-9.5; p=0.03), age >60 years (HR 2.9, 1.4-5.9; p<0.01), and LDH >1.5 x UNL (HR 4.0 (1.6-10.0; p<0.01). A similar analysis in ET also identified RDW >15.5 (p<0.01), age >60 years (p=0.01) as well as male sex (p=0.03) as independent risk factors for LT. Subseqent risk models based on these risk factors were externally validated and enabled robust risk stratification with 10-20 year AUCs rangening from 0.75 to 0.81. Validation cohorts analyses The indpendent predictive value of increased RDW observed in the discovery cohort was confirmed in the validation cohorts for both PV and ET and further gender-specified: female/male RDW ≥16.4/15.8 in PV (HR 7.9, 1.9-33.8) and ≥16.1/14.4 in ET (HR 5.1, 2.2-11.9). Also confirmed were the additional prognostic contributions of age ≥60 years (HR 3.8, 1.3-10.9) and LDH >1.5 x UNL (HR 3.9, 1.6-9.4), in PV, while no other clinical risk factor showed RDW-independent prognostic relevance in ET. The prognostic impact of RDW in ET was most apparent in females (p<0.01), as opposed to males (p=0.32) while the effect was equally significant in PV. In multivariable analysis that included previously recognized high risk mutations in PV (SRSF2) and ET (SF3B1), both RDW and mutations sustained significance (p<0.05). Conclusion The current study identifies higher RDW as a key risk factor for LT in both PV and ET, independent of other clinical or genetic risk factors. Possible underlying mechanisms include increased RDW as a marker of long-term pre-leukemic clones with sub-clinical ineffective erythropoiesis or an associated pro-leukemic inflammatory state.

ABSTRACT Survival prediction models in essential thrombocythemia (ET) include the International Prognostic Scoring System (IPSET) and the more recently introduced triple‐A (AAA) prognostic score. The latter enlists age and absolute neutrophil (ANC) and lymphocyte (ALC) counts as risk variables. In the current study, a Mayo Clinic discovery cohort of 658 patients with ET was used to identify AAA‐independent risk variables. Accordingly, multivariable analysis‐derived HRs (95% CI) were 15.7 (8.4–29.5) for age > 70 years (8 points); 4.2 (2.3–7.5) for age 50 to 70 years (2 points); 1.8 (1.2–2.5) for ANC ≥ 8 × 10 9 /L (1 point); 1.4 (1.03–1.9) for ALC < 1.7 × 10 9 /L (1 point); 1.8 (1.2–2.6) for absolute monocyte count (AMC) ≥ 0.5 × 10 9 /L (1 point); 1.8 (1.2–2.3) for male sex (1 point); 1.8 (1.3–2.4) for arterial hypertension (1 point); and 1.6 (1.2–2.3) for arterial thrombosis (1 point). HR‐weighted scoring enabled a 4‐tiered risk classification: ultra‐low (0–1 points; N = 94; median survival 42.7 years), low (2–4 points; N = 297; 23 years), intermediate (5 points; N = 66; 17.3 years), and high (6–14 points; N = 201; 10.8 years). Time‐dependent predictive performance at 20/25 years favored AAA + (AUC 0.92/0.91) vs. AAA (0.86/0.86) vs. IPSET (0.81/0.84). The AAA + risk model was subsequently validated by two external cohorts from Israel ( N = 5968) and Italy ( N = 682). In the cohort from Israel, disease‐specific mortality was assessed by comparing observed survival to an age‐ and sex‐matched reference population, which suggested near‐normal life expectancy in ultra‐low risk patients. The current study highlights host‐related factors as the primary determinants of longevity in ET and provides a composite risk score (AAA + ) that is based on complete blood count‐derived parameters and host‐related factors. Predictive performance of the new model was shown to be superior to that of IPSET and AAA.