2026 ASH®
Annual Meeting & Exposition December 12-15, 2026 ▪ New Orleans, Louisiana

Paper Details

Abstracts, authors, affiliations, and disclosures.

Monday, December 14, 2026
06:00 p.m. - 08:00 p.m.
MCCNO - Hall F
Technology and Procedures, Molecular Testing, Pathology, Treatment Considerations, Combination Therapy, Drug Development, Supportive Care, Biological Therapies, Monoclonal Antibody Therapy, Non-Biological Therapies, Anticoagulant Drugs, Pharmacology, Women’s Health, Adverse Events, Artificial Intelligence (AI), Generative AI, Machine Learning, Natural Language Processing, Biological Processes, Apoptosis, Genomics, Hematopoiesis, Molecular Biology, Clinical Practice (Health Services and Quality), Diseases, Bleeding and Clotting, Bleeding Disorders, Platelet Disorders, Thromboembolism, Thrombotic Disorders, VWD, Genetic Disorders, Immunology, Autoimmune Disorders, Lymphoid Malignancies, Myeloid Malignancies, Acute Myeloid Malignancies, Chronic Myeloid Malignancies, Research, AML, Translational Research, MDS, MPN, CMML, Clinical Research, Clinical Trials, Epidemiology, Health Outcomes Research, Patient-Reported Outcomes, Real-World Evidence, Survivorship, Study Population, Human, Adult, Elderly, Young Adult
PAPER

5527 ASXL1 variant allele frequency and associated driver mutation determine prognostic relevance in chronic phase primary myelofibrosis

Abirame Gopinath, MBBS, Paola Guglielmelli, MD, PhD2, Amina Akram, MBBS3, Saubia Fathima, M.B., B.Ch.4, Giuseppe Gaetano Loscocco, MD, PhD5, Shivani Udyawar6, Merry Nakhleh7, Kaaren Reichard, MD8, David Viswanatha, MD9, Rong He, MD10, Cinthya J. Zepeda Mendoza11, Animesh Pardanani, PhD, MBBS12, Naseema Gangat, MBBS13, Alessandro Vannucchi, MD14, Ayalew Tefferi, MD15
1 Mayo Clinic, Rochester, MN, United States
2 University of Florence, Florence, Italy
3 Mayo Clinic, Rochester, MN, United States
4 Mayo Clinic, Rochester, MN, United States
5 University of Florence, Florence, Italy
6 Mayo Clinic, Rochester, MN, United States
7 Mayo Clinic, Rochester, MN, United States
8 Mayo Clinic, Rochester, MN, United States
9 Mayo Clinic, Rochester, MN, United States
10 Mayo Clinic, Rochester, MN, United States
11 Mayo Clinic, Rochester, MN, United States
12 Mayo Clinic, Rochester, MN, United States
13 Mayo Clinic, Rochester, MN, United States
14 University of Florence, Florence, Italy
15 Mayo Clinic, Rochester, MN, United States

Background:

Prognosis in primary myelofibrosis (PMF) is determined first by disease stage (chronic phase [PMF-CP] vs. accelerated/blast phase myeloproliferative neoplasm [MPN-AP/BP]) and second by genetic and clinical risk factors. Current prognostic models consider high molecular risk (HMR) mutations but do not account for variant allele frequency (VAF) or prognostic interaction with MPN driver mutations. The current study was designed to address this knowledge gap in PMF-CP.

Methods:

Diagnoses were assigned according to ICC criteria (Blood 2022. 140:1200). Cases were recruited from Mayo Clinic MPN databases, based on availability of information on HMR VAF. CLIA-approved NGS testing was used to curate mutational data, collected at diagnosis or prior to progression into MPN AP/BP. Overall survival (OS) analyses were censored at allogeneic stem cell transplantation (ASCT). Optimal HMR VAF cutoff level was determined by time-dependent ROC curve.

Results:

Mayo Clinic enterprise-wide database search identified 611 informative patients with PMF-CP (median age 65 years; males 63%): 60% JAK2, 19% CALR1, 10% MPL, 8% triple-negative (TN), and 3% CALR2. DIPSS-R categories were low (23%), intermediate-1 (39%), intermediate-2 (35%), and high (3%). Karyotype was unfavorable in 18% and mutational frequencies were 35% ASXL1, 16.7% TET2, 16% SRSF2, 9% U2AF1-Q157, 6.4% DNMT3A, 6% SF3B1, 4.9% EZH2, 4.8% CBL, 3.2% TP53, 3.1% IDH2, 2.8% IDH1, and 2.8% NRAS.

At a median f/u of 50 months, 209 deaths (34.2%), 50 (8%) blast transformations, and 149 (24.3%) ASCTs were recorded. In multivariable analysis of genetic risk factors, the previously recognized independent adverse prognostic contributions of ASXL1MUT, SRSF2 MUT, U2AF1-Q157, unfavorable karyotype, and absence of CALR1 were confirmed (p<0.04 in all instances). In univariate analysis, prognostic relevance of VAF, as a continuous variable, was apparent for ASXL1 (p=0.01) but not for SRSF2 (p=0.3) or U2AF1-Q157 (p=0.9). A statistically optimized ASXL1 VAF cutoff level of 30% revealed a significant difference in OS between VAF ≥30% (N=129; median 69 months) vs. VAF <30% (N=84; median 211 months; p<0.01); the latter showed similar OS to that of ASXL1WT (N=398; median 175 months; p=0.33). The results were similar when OS was calculated from time of mutation detection and further validated by an external cohort of 556 informative patients from the University of Florence, Italy. The inferior OS associated with ASXL1 VAF ≥30% and the similar OS between ASXL1WT and ASXL1 VAF <30% were further confirmed by multivariable analysis that included SRSF2 MUT, U2AF1-Q157, CALR1, and unfavorable karyotype, with corresponding p values of <0.01, <0.01, and o.93.

The adverse prognostic impact of ASXL1 VAF ≥30% was most apparent in JAK2-mutated patients (N=87; median 54 months; p <0.01) whose OS was otherwise similar in the presence of ASXL1 VAF <30% (n=51; median 193 months) or ASXL1 WT (n=229; median 196 months). On the other hand, the presence of CALR1 was associated with attenuation of the adverse prognostic impact of ASXL1MUT in general, regardless of VAF, with median OS of 196, 299 and 220 months for ASXL1 VAF ≥30% (N=21), ASXL1 VAF <30% (N=13), and ASXL1WT (N=81), respectively (p = 0.2). In patients lacking both JAK2 and CALR1 mutations (N=129), the presence of ASXL1 MUT was associated with poor survival, irrespective of VAF: (50 vs. 65) vs. 131 months for (VAF ≥30% vs. VAF <30%) vs. ASXL1WT (p=0.03).

Consistent with their divergent prognostic relevance, ASXL1 VAF ≥30% and ASXL1 VAF <30% groups differed significantly in some of their clinical and genotype correlates with the latter being more aligned with ASXL1WT:  ASXL1 VAF ≥30%/ASXL1 VAF <30%/ASXL1WT frequencies were for NRAS 7.8%/2.4%/1.3% (p<0.01), CBL 10%/2.4%/3.5% (p=0.01), IDH1 7.8%/1.2%/1.5% (p<0.01), DNMT3A 1.5%/7.1%/7.8% (p=0.02), SRSF2 22.5%/14.3%/13.6% (p=0.06), absolute monocyte count ≥1 x 10(9)/L (29%/10%/12%; p<0.01), PB blast ≥2% (29%/17%/15%; p<0.01), leukocyte count ≥11 x 10(9)/L (58%/25%/31%; p<0.01), and male gender 74%/64%/59% (p<0.01).  

Conclusions:

The prognostic relevance of ASXL1MUT in PMF-CP is determined by its VAF (≥30% unfavorable vs. <30% similar to ASXL1WT) and further modified by driver mutation context. Consistently, certain clinical and genotype correlates of ASXL1 VAF <30% were more likely to be aligned with ASXL1WT vs. ASXL1 VAF ≥30%. Such information is critical for treatment decision-making, including timing of ASCT. 

Abirame Gopinath, MBBS: No relevant conflicts of interest to declare.