KIT mutation and AHN-associated oncogenic profiles in systemic mastocytosis with an associated hematological neoplasm∗ Authors: Paula Navarro-Navarro, Carlos M Fernández-Giménez, Andrés C García-Montero, Iván Álvarez-Twose, Laura Sánchez-Muñoz, Ana Henriques, Alba Pérez-Pons, María González-Tablas, Cristina Teodosio, María González-González, Sergio Matarraz, Antonio López, Paloma Barcena, Enrique Colado, Jordi Esteve, Mª Leonor Senent, Óscar González-López, Manuela Mollejo, Cristina Morales-Cabeza, María Jara-Acevedo, Alberto Orfao Published: 2026 ABSTRACT Key Points•The KIT mutation and AHN-related genetic alterations found in MC and AHN cells define different oncogenic profiles in patients with SM-AHN.•Patients with SM-AHN and genetically unrelated MC and AHN cells have more indolent disease and better outcomes vs patients with shared alterations. Introduction Systemic mastocytosis (SM) comprises a heterogeneous group of diseases characterized by the accumulation of clonal mast cells (MC) in ≥2 tissues, including the skin and/or 1 extracutaneous organ, such as the bone marrow (BM) and/or gastrointestinal tract.1, 2, 3 Currently, 6 different subtypes of SM are recognized by the World Health Organization (WHO),4, 5, 6 including BM mastocytosis (BMM), indolent SM (ISM), smoldering SM, and 3 variants of advanced SM (AdvSM): aggressive SM (ASM), SM with an associated hematological neoplasm (SM-AHN), and MC leukemia (MCL). Among AdvSM, SM-AHN is by far the most prevalent and heterogeneous subtype.1,7,8 It consists of various SM forms (BMM, ISM, MCL, or more frequently, ASM) associated with a coexisting myeloid neoplasm (AMN), such as chronic myelomonocytic leukemia (CMML),9,10 myelodysplastic syndrome (MDS),11,12 or acute myeloid leukemia (AML),13, 14, 15 according to the WHO and International Consensus Classifications,4,16 and/or an associated lymphoid neoplasm (ALN),9, 10, 11, 12, 13, 14, 15,17, 18, 19, 20, 21, 22 which is recognized only by the WHO. Nowadays, the treatment, prognosis, and outcome of patients with SM-AHN are considered to mostly depend on the subtypes of SM and AHN coexisting in individual patients.23, 24, 25, 26, 27 Similar to other subtypes of SM, most patients with SM-AHN (60%-80%) carry the KIT p.D816V mutation5,24,28,29 and, to a lesser extent, other KIT mutations (≤5%).8,30, 31, 32, 33, 34 Previous observations indicate that in some patients with SM-AHN, the KIT mutation and/or the other AHN-related genetic/molecular alterations might coexist in the SM and AHN cell compartments, with potential involvement of a common progenitor; conversely, in other patients with SM-AHN, the KIT mutation and/or the AHN-associated genetic alterations are restricted to MC and AHN cells, respectively.15,32,35, 36, 37, 38 These findings point to common vs distinct underlying molecular pathways for the 2 coexisting diseases in individual patients, which might help explain the heterogeneous clinical behavior of SM-AHN, even among patients with the same subtypes of SM and AHN.1,29,39 In recent years, the clinical impact of individual genetic markers and combinations of markers has been extensively analyzed in AdvSM, including SM-AHN.23,40,41 However, so far, only a few studies have investigated the clonal relationship between the coexisting SM and AHN cell compartments, mostly based on single case reports and limited series of 7 to 48 patients with SM-AHN.32,34,37,42, 43, 44, 45 Globally, these reports point to the occurrence of different, highly variable oncogenic pathways in SM-AHN. However, the precise frequency of those individual oncogenic pathways, defined by the relationship between the clonal MC and AHN cell compartments, and their impact on disease outcome remain to be determined. Here, we report on the oncogenic profiles of 79 patients with SM-AHN, their frequency, and their impact on patient outcomes. To our knowledge, this is the largest series in which the clinical and prognostic features of patients with SM-AHN carrying distinct oncogenic profiles are described. Methods Patients and samples Seventy-nine patients (52 men and 27 women; median age, 61 years [range, 1-81]; 76 adults and 3 pediatric patients aged 1, 10, and 11 years [patients 78, 47, and 31, respectively]) diagnosed with SM-AHN and carrying a KIT mutation were studied. The diagnostic subtypes of SM and AHN were retrospectively updated according to the 2022 WHO criteria4, 5, 6,16,46 (Table 1). From each patient, EDTA-anticoagulated BM samples were collected at diagnosis after the patient had provided informed consent to participate in this study, and the study had been approved by the local ethics committees of the 2 participating Spanish Network on Mastocytosis (REMA) centers: Instituto de Estudios de Mastocitosis de Castilla La Mancha (Toledo, Spain) and Cancer Research Center (University of Salamanca, Salamanca, Spain). Forty-three patients (54%) were subsequently treated with chemotherapy, 15 (19%) received tyrosine kinase inhibitors, and 3 (3.8%) underwent stem cell transplantation (supplemental Methods). Table 1.Distribution of the KIT mutation and AHN-associated genetic/molecular alterations in BM MC and AHN cells of patients with SM-AHN (n = 79)Patient IDOncogenic profileSM-AHN diagnostic subtypeBM MC compartmentBM AHN cell compartmentKIT mutationAHN genetic/molecular alterationKIT mutationAHN genetic/molecular alteration1BMM + AML-MR+−−del(5q) (60); del(7q) (12%)2Genetically unrelated, 21/79 (27%)BMM + AML-MR+−−−Y (98%); del(RARA) (26%)3BMM + AML-MR+−−SRSF2 (p.P95H) (20%);RUNX1 (p.R481∗) (20%)4BMM + ET+−−CALR (p.L367fs∗46) (NA)5BMM + HES+−−RUNX1 (p.S424P) (30%)6BMM + PV+−−JAK2 (p.V617F) (20%)7BMM + CML+−−BCR::ABL (84%)8BMM + CLL+−−del(RB1) (60%); del(D13S25) (5%)9BMM + MZL+−−t(14q32) (20%)10BMM + LPL/WM+−−—11BMM + CLL + LPL/WM+−−del(RB1) (27%); del(D13S25) (80%); t(14q32) (63%);DNMT3A (p.D652N) (20%)12BMM + AML-NPM1 + MZL+−−NPM1 (p.W288fs) (20%); PIGA (p.H128R) (20%); +12 (94%)13ISM + SMM+−−del(RB1) (80%)14ISM + SMM+−−CCND1::IGH (90%)15ISM + SMZL+−−del(14q32) (76%)16ISM + T-LGLL + LPL/WM+−−—17ISM + CLL + LPL/WM+−−del(RB1) (20%); del(D13S25) (20%)18ASM + AML with ODGA+−−+9q34 (78%)19ASM + AML+−−—20ASM + MDS-IB2+−−—21ASM + MDS-LB+−−—21/21 (100%)0/21 (0%)0/21 (0%)16/21 (76%)22KIT mutation first, 24/79 (30%)BMM + AML-MR+−+BCOR (p.V594I) (20%)23BMM + AML-NPM1+−+NPM1 (p.W288fs) (20%);SRSF2 (p.P95H) (40%)24BMM + CMML+−+−20q (6%)25ISM + AML-NPM1+DNMT3A (p.R882H) (20%)+NPM1 (p.W288fs) (20%);DNMT3A (p.R882H) (20%)26ISM + AML-NPM1+SRSF2 (p.P95R) (20%)+SRSF2 (p.P95R) (20%);NRAS (p.G12R) (20%)27ISM + HES+−+—28ISM + AML-NPM1 + SMM+−+NPM1 (p.W288fs) (20%)29ISM + CMML + MM+−+∗CCND1::IGH (90%); del(RB1) (80%)30ISM + AML with ODGA†+ (60%)RUNX1 (p.R166Q) (14%)+ (90%)RUNX1 (p.R166Q) (20%)31ISM + AML RUNX1::RUNX1T1‡p.D816YRUNX1::RUNX1T1 (NA)+RUNX1::RUNX1T1 (38%)32ISM + CMML†+ (100%)DNMT3A (p.R882H) (30%);BCOR (p.V679I) (24%)+ (100%)DNMT3A (p.R882H) (NA);BCOR (p.V679I) (60%)33ISM + PV‡+−+TET2 (p.E283∗) (NA)34ISM + MM†+−++(1q21) (50%); del(RB1) (74%)35ASM + MDS-LB+−+—36ASM + MDS-IB2+−7 (26%)+−7 (50%); SRSF2 (p.P95R) (50%)37ASM + MDS-IB2+SRSF2 (p.P95R) (20%)+SRSF2 (p.P95R) (NA)38ASM + MDS-IB2+ (100%)+ (20%)+ (100%)SRSF2 (p.P95R) (20%)39ASM + CMML+ (100%)TET2 (p.Q622Pfs∗18) (30%)+ (90%)TET2 (p.Q622Pfs∗18) (76%)40ASM + CMML++ RUNX1 (36%)++ RUNX1 (60%)41ASM + HES+−+—42ASM + T-LGLL+−+—43ASM + MDS-LB + MM+−+CCND1::IGH (68%)44ASM + CMML + SMM+−+—45ASM + CMML + MM++8 (30%); del(RB1) (90%); CCND1::IGH (70%)++8 (35); del(RB1) (90%); CCND1::IGH (70%)24/24 (100%)11/24 (46%)24/24 (100%)19/24 (79%)46AHN-related genetic alteration first, 25/79 (32%)BMM + AML-MR+SRSF2 (p.P95L) (50%)−SRSF2 (p.P95L) (100%)47BMM + AML RUNX1::RUNX1T1+RUNX1::RUNX1T1 (90%)+RUNX1::RUNX1T1 (100%)48BMM + CMML+RUNX1 (p.R201Q) (20%);TP53 (p.N239D) (20%)−RUNX1 (p.R201Q) (20%); TP53 (p.N239D) (20%)49BMM + CMML+SRSF2 (p.95_103del) (60%)+ (10%)SRSF2 (p.95_103del) (44%)50BMM + ET+DNMT3A (p.W306X) (40%); SETBP1 (p.T195P) (60%)−DNMT3A (p.W306X) (30%); SETBP1 (p.T195P) (60%)51BMM + HESp.D816ARUNX1 (p.S424P) (14%);TP53 (p.P128A) (50%)−RUNX1 (p.S424P) (30%); TP53 (p.P128A) (50%)52ISM + AML RUNX1::RUNX1T1p.K558_S584dupRUNX1 (p.R201Q) (20%);TP53 (p.N239D) (20%)−RUNX1 (p.R201Q) (20%); TP53 (p.N239D) (20%)53ISM + AML-NPM1+NPM1 (p.W288fs) (20%); SRSF2 (p.P95R) (20%); NRAS (p.G12R) (20%)+NPM1 (p.W288fs) (20); SRSF2 (p.P95R) (20%); NRAS (p.G12R) (20%)54ISM + MDS-IB1+TET2 (p.N442fs) (100%)+ (82%)TET2 (p.N442fs) (NA)55ISM + CMML+ASXL1 (p.E1102D) (NA)−ASXL1 (p.E1102D) (NA)56ISM + CMML+CBL (p.F418;R420Q) (40%)−CBL (p.F418;R420Q) (40%); del(20q12) (100)%)57ISM + CMML+TET2 (p.S358X) (NA)+TET2 (p.S358X) (NA)58ISM + MPN-NOS†+SRSF2 (p.P95H) (80%)+SRSF2 (p.P95H) (40%)59ASM + AML-MR+del(5q32) (45%); +8 (80%); NRAS (p.G12R) (20%)+del(5q32) (90%); +8 (86%); NRAS (p.G12R) (20%)60ASM + MDS-LB+FLT3 (p.D835Y) (20%)−FLT3 (p.D835Y) (20%)61ASM + CMML+SRSF2 (p.P95L) (30%)−SRSF2 (p.P95L) (80%)62ASM + CMML+SRSF2 (p.P95H) (90%); RUNX1 (p.T104Nfs∗34) (40%); TET2 (p.P1894S) (100%)+SRSF2 (p.P95H) (100%); RUNX1 (p.T104Nfs∗34) (80%); TET2 (p.P1894S) (100%)63ASM + CMML+ (100%)SRSF2 (p.P95R) (100%)+ (30%)SRSF2 (p.P95R) (100%); KRAS (p.A59E) (60%)64ASM + CMML+ (88%)SRSF2 (p.P95L) (34%)+ (24%)SRSF2 (p.P95L) (80%)65ASM + CMML+ (100%)SRSF2 (p.P95L) (100%)+ (60%)SRSF2 (p.P95L) (100%); TET2 (p.G422X) (80%)66ASM + CMML+−7 (100%)+−7 (100%)67ASM + CMML+DNMT3A (p.R882H) (80%)+DNMT3A (p.R882H) (76%)68ASM + CMML+DNMT3A (p.R882H) (40%)+DNMT3A (p.R882H) (60%)69ASM + CMML+TET2 (p.N275I) (20%)+TET2 (p.N275I) (20%)70ASM + HES+SRSF2 (p.P95H) (80%)+SRSF2 (p.P95H) (80%)25/25 (100%)25/25 (100%)16/25 (64%)25/25 (100%)71SimultaneousKIT and AHN-related genetic alterations, 9/79 (11%)ISM + MDS-SF3B1+SF3B1 (p.K666N) (100%)+SF3B1 (p.K666N) (NA)72ASM + AML-MR+ (100%)RUNX1 (p.R162K) (70%);BCOR (p.S1298∗) (80%)+ (100%)RUNX1 (p.R162K) (80%); BCOR (p.S1298∗) (80%)73ASM + MDS-LB+ (100%)SRSF2 (p.P95R) (60%)+ (100%)SRSF2 (p.P95R) (80%)74ASM + MDS-SF3B1+ (100%)SF3B1 (p.K666T) (20%)+ (100%)SF3B1 (p.K666T) (80%)75ASM + CMML+ASXL1 (p.G642fs) (80%)+ASXL1 (p.G642fs) (NA)76ASM + CMML+−Y (96%); NRAS (p.G12R) (20%)+−Y (100%); NRAS (p.G12R) (20%)77ASM + CMML+ (100%)SRSF2 (p.P95L) (60%)+ (60%)SRSF2 (p.P95L) (60%),−7 (100%)78ASM + PV+ (90%)JAK2 (p.N337D) (90%)+ (100%)JAK2 (p.N337D) (90%)79MCL + MDS-IB1+ (100%)del(5q31) (80%); TP53 (p.R175H) (80%); RUNX1 (p.L339Afs∗260) (100%)+ (100%)del(5q31) (100%); TP53 (p.R175H) (80%); RUNX1 (p.L339Afs∗260) (100%)9/9 (100%)9/9 (100%)9/9 (100%)9/9 (100%)Total79/79 (100%)45/79 (57%)50/79 (65%)68/79 (86%)The percentage of mutated cells indicates the percentage of mutated/genetically altered cells for each specific gene/genetic marker. Three pediatric patients aged 1, 10, and 11 years (patients 78, 47, and 31, respectively) were included in this cohort.CLL, chronic lymphocytic leukemia; CML, chronic myeloid leukemia; ET, essential thrombocythemia; HES, hypereosinophilic syndrome; IB, increased blasts; ID, identity; LB, low blast; LPL, lymphoplasmacytic lymphoma; MM, multiple myeloma; MR, myelodysplasia related; MZL, extranodal marginal zone lymphoma; NA, not available; NOS, not otherwise specified; ODGA, other defined genetic alterations; PV, polycythemia vera; SMM, smoldering multiple myeloma; SMZL, splenic MZL; T-LGLL, T-cell large granular lymphocyte leukemia; WM, Waldenström macroglobulinemia.∗Shared only by myeloid cells but not by myeloma plasma cells.†Patients who would meet criteria for SSM (KIT-mutated cells >10%).‡Patients who would meet criteria for SSM (serum baseline tryptase (sBT) >200 ng/mL and hepatosplenomegaly). Distribution of the KIT mutation and AHN-associated genetic/molecular alterations in BM MC and AHN cells of patients with SM-AHN (n = 79) The percentage of mutated cells indicates the percentage of mutated/genetically altered cells for each specific gene/genetic marker. Three pediatric patients aged 1, 10, and 11 years (patients 78, 47, and 31, respectively) were included in this cohort. CLL, chronic lymphocytic leukemia; CML, chronic myeloid leukemia; ET, essential thrombocythemia; HES, hypereosinophilic syndrome; IB, increased blasts; ID, identity; LB, low blast; LPL, lymphoplasmacytic lymphoma; MM, multiple myeloma; MR, myelodysplasia related; MZL, extranodal marginal zone lymphoma; NA, not available; NOS, not otherwise specified; ODGA, other defined genetic alterations; PV, polycythemia vera; SMM, smoldering multiple myeloma; SMZL, splenic MZL; T-LGLL, T-cell large granular lymphocyte leukemia; WM, Waldenström macroglobulinemia. Shared only by myeloid cells but not by myeloma plasma cells. Patients who would meet criteria for SSM (KIT-mutated cells >10%). Patients who would meet criteria for SSM (serum baseline tryptase (sBT) >200 ng/mL and hepatosplenomegaly). Cell purification In each patient, specific populations of BM cells (purity >98%) were purified using a 4-way FACSAria III fluorescence-activated cell sorter (FACS; Becton/Dickinson Biosciences, San Jose, CA) at the Cell Purification Service of the University of Salamanca. Before sorting, samples were stained with ≥4-color combinations of fluorochrome-conjugated monoclonal antibodies for simultaneous identification of phenotypically aberrant MC, AHN cells, and other residual myeloid (ie, maturing neutrophils) and/or lymphoid cells. BM samples were lysed and stained following the EuroFlow standard operating procedures available at www.EuroFlow.org. In 11 patients with SM-AHN, BM mesenchymal stem cells (MSC) were also purified, as described elsewhere.47, 48, 49, 50 KIT mutational analysis KIT p.D816V was assessed in genomic DNA (gDNA) from FACS-purified BM cell populations and reported as the percentage of mutated cells, using previously described semiquantitative peptide nucleic acid–mediated polymerase chain reaction (PCR; sensitivity, <1%-5%)51 and allele-specific oligonucleotide quantitative PCR methods.52 To detect other KIT mutations involving distinct codons of tyrosine kinase domain 2 of KIT, the peptide nucleic acid–mediated PCR method and/or Sanger sequencing (sensitivity, <20%) were used in KIT p.D816V− cases.42,51 iFISH studies Interphase fluorescence in situ hybridization (iFISH) analyses were performed on interphase nuclei from FACS-purified BM cell populations freshly fixed in 3:1 methanol/acetic acid (volume-to-volume ratio) and stored at −20°C until analysis with a Nikon Eclipse iCi fluorescence microscope (Olympus, Hamburg, Germany). For this purpose, a set of probes in double stainings (supplemental Table 1) was used for detection of gene rearrangements and large copy number chromosome alterations recurrently observed in patients with SM-AHN, as described elsewhere.53,54 iFISH results were reported as the percentage of altered cells. Assessment of chromosome-associated CNAs throughout the human genome The CytoScan 750K Array (Thermo Fisher Scientific, Waltham, MA), with probes for 200 436 single-nucleotide polymorphisms, was used to identify copy number alterations (CNA) across the entire genome of individual patients with SM-AHN at the Genomics Unit of the Cancer Research Center. For this purpose, gDNA from whole BM and FACS-purified BM cell populations was used. For single-nucleotide polymorphism/CNA data analysis, the Chromosome Analysis Suite software (Thermo Fisher Scientific) was used, and the hg19 human genome sequence was used as the reference. Gene sequencing The AmpliSeq Comprehensive Cancer Panel was analyzed in BM MC and AHN cells on an Ion Proton platform (Life Technologies, Carlsbad, CA), according to the manufacturer’s instructions. The pathogenic variants identified were subsequently validated by conventional Sanger sequencing (3130XL; Applied Biosystems, Waltham, MA) in gDNA from purified BM MC, non-MC myeloid and/or lymphoid AHN cells, other residual myeloid cells (maturing neutrophils and monocytes), B and T lymphocytes, and MSC. In addition, next-generation sequencing was performed at the Sequencing Service of NUCLEUS (University of Salamanca) to further investigate the presence of AdvSM-related and/or AHN-related single-point mutations, based on a panel of 6 genes frequently altered in SM40,41,55 and 54 AHN-associated genes56,57 (supplemental Methods). For the SM-associated gene panel, DNA libraries were generated from gDNA of BM-purified cell populations using the AmpliSeq Illumina Custom DNA Panel (Illumina, San Diego, CA). In turn, for sequencing of the 54 AHN-related genes, the KAPA HyperCap Custom Panel (Roche, Basel, Switzerland) was used to generate DNA libraries. All libraries were sequenced on Illumina MiSeq and NovaSeq 6000 instruments at 2 × 150–base pair read length, following the manufacturer’s instructions. For data analysis, the Mutation Mining software (https://mutationmining.usal.es; University of Salamanca) was used, and genetic variants were identified following a previously described workflow.23 In addition, hot spot sequences of 17 genes (supplemental Methods) were further analyzed (Biomark HD; Standard BioTools, San Francisco, CA), and results were reported as the percentage of mutated/altered cells. Statistical methods For continuous variables, median (range) values were calculated, whereas for categorical parameters, frequencies were used. Comparisons between ≥2 groups were made using the Kruskal-Wallis and χ2 tests; whenever statistically significant differences were observed, pairwise comparisons were subsequently conducted using the Mann-Whitney U test or the χ2 test with the Bonferroni correction for multiple comparisons. Receiver operating characteristic curve analysis was used to determine the most accurate percentage of KIT p.D816V–mutated cells to predict multilineage KIT involvement. Survival curves were plotted according to the Kaplan-Meier method and compared by the 2-sided log-rank test. For multivariable analysis, the covariate adjustment model58 was applied to variables that showed prognostic impact on progression-free survival (PFS) and overall survival (OS) in the univariable analysis. Disease progression was defined as progression of SM to more advanced disease subtypes, as previously defined,59 and/or clinical progression of the AHN with need for treatment (when it had not been previously required) or after treatment, and/or transformation to more advanced disease. At study closure, median patient follow-up was 4.5 years (range, 0.1-15), and 56% of patients remained alive. For all statistical analyses, the Statistical Package for Social Sciences version 29 (IBM, Armonk, NY) and GraphPad Prism version 8 (GraphPad Software Inc, San Diego, CA) software programs were used. Statistical significance was set at P value < .05. Results Diagnostic subgroups of SM-AHN Most patients with SM-AHN (70/79 [89%]) had SM associated with 1 AHN, whereas in 9 patients (11%), SM was associated with 2 coexisting AHNs (Table 1). In most of the former cases (62/79 [78%]), SM coexisted with an AMN, which consisted of (1) 12 MDS (15%): 6 with low blast counts, of which 2 had SF3B1 mutations, 2 with increased blasts type 1, and 4 with increased blasts type 2; (2) 17 AML (21%): 8 with myelodysplasia-related changes, 4 with NPM1 mutation, 3 with RUNX1::RUNX1T1, and 2 with other defined genetic alterations (+9q34, RUNX1 p.R166Q); (3) 21 MDS/myeloproliferative neoplasia (MPN) cases (27%), all corresponding to CMML; and (4) 12 MPN cases (15%): 3 (4%) polycythemia vera, 2 (3%) essential thrombocythemia, 5 (6%) hypereosinophilic syndrome, 1 (1%) chronic myeloid leukemia, and 1 (1%) MPN not otherwise specified (Table 1). Lymphoid AHNs were less frequently observed (8/79 patients [10%]) and consisted of smoldering multiple myeloma (SMM) in 2 patients (3%) and multiple myeloma (MM), B-cell chronic lymphocytic leukemia, lymphoplasmacytic lymphoma (LPL) of the Waldenström macroglobulinemia type, extranodal marginal zone lymphoma (MZL), splenic MZL, and T-cell large granular lymphocyte leukemia in 1 patient (1%) each (Table 1). Six of those 9 patients (7.5%) diagnosed with SM associated with 2 AHNs had AMN plus ALN, that is, AML with NPM1 and MZL, AML with NPM1 and SMM, CMML plus SMM, and MDS low blast plus MM in 1 patient each, and CMML associated with MM in 2 patients; 2 coexisting ALNs were found in the other 3 patients (4%), namely, B-cell chronic lymphocytic leukemia plus LPL/Waldenström macroglobulinemia in 2 and LPL plus T-cell large granular lymphocyte leukemia in 1 (Table 1). The 2022 WHO subtypes of SM among SM-AHN consisted of 21 BMM (27%), 23 ISM (29%), 34 ASM (43%), and 1 MCL (1%) (Table 1). As expected, a greater frequency of CMML (P = .036) was found among patients with ASM-AHN (14/34 [41%]) vs BMM-AHN (3/21 [14%]). Conversely, a higher proportion of ALN was detected in BMM-AHN and ISM-AHN vs patients with ASM-AHN : 4 of 21 (19%) and 5 of 17 (29%) vs 1 of 34 (3%; P = .037 and P = .007, respectively; Table 1). Frequency of the KIT mutation and AHN-related genetic alterations in different BM cell compartments of patients with SM-AHN KIT p.D816V was present in BM MC from all but 3 patients with SM-AHN, who displayed other KIT mutations (KIT p.D816A, KIT p.D816Y, and KIT p.K558_S584dup). In 30 of 79 patients (38%), the KIT mutation was restricted to BM MC, whereas in 49 of 79 patients (62%), it was shared by MC and AHN cells (Table 1). In turn, AHN-related genetic alterations other than the KIT mutation were detected in 68 of 79 patients with SM-AHN (86%) (Table 1). In 23 of 79 patients (29%), AHN-related alterations were restricted to AHN cells and absent in BM MC; conversely, in another 45 of 79 patients (57%), ≥1 AHN-related alteration was shared by MC and AHN cells, together with (36/79 [44%]) or without (9/79 [12%]) coexistence of the KIT mutation in both cell compartments (Table 1). In patients in whom BM MC and AHN cells shared the KIT mutation (49/79 [62%]), this mutation was detected in both MC and other residual myeloid BM cells; a significant fraction of these patients (27/49 [55%]) also had KIT-mutated BM lymphoid cells. Interestingly, purified MSC from 7 of 9 patients with SM-AHN (78%) with KIT-mutated residual myeloid (ie, maturing neutrophils and/or monocytes) plus B and/or T lymphoid cells had KIT-mutated MSC. In turn, in most of those 45 patients (26/28 [92%]) who shared AHN-related genetic alterations in BM MC and AHN cells and in whom the study of residual cells was performed (25 with AMN and 1 with AMN plus ALN) had those same alterations in other non-AHN residual myeloid, but not lymphoid, BM cells (Table 2). Table 2.Genetic profiles of distinct BM neoplastic cell compartments (MC and AHN cells) and residual myeloid and lymphoid cells according to the KIT mutation and AHN-related genetic/molecular alterations in patients with SM-AHN (n = 79)Mutational statusGenetic profilesGenetically unrelated MC and AHN cells (n = 21), %KIT mutation first (n = 24), %AHN-related genetic alteration first (n = 25), %Simultaneous KIT- and AHN-related genetic alterations (n = 9), %KIT p.D816V+ BM cells∗ MC100100100100 AHN cells1410064100 Other myeloid cells010060100 Other lymphoid cells075†29‡78§AHN-related genetic/molecular alterations in BM cells MC045100100 AHN cells7675100100 Other myeloid cells337594100 Other lymphoid cells000100Results are expressed as the percentage of cases analyzed within each oncogenic group of patients with SM-AHN. MSC were investigated only for KIT p.D816V.∗Three patients showed KIT mutations other than p.D816V (KIT p.D816A, KIT p.D816Y, and KIT p.K558_S584dup).†Two of 5 patients had KIT-positive MSC.‡One of 2 patients had KIT-positive MSC.§Four of 4 patients had KIT-positive MSC. Genetic profiles of distinct BM neoplastic cell compartments (MC and AHN cells) and residual myeloid and lymphoid cells according to the KIT mutation and AHN-related genetic/molecular alterations in patients with SM-AHN (n = 79) Results are expressed as the percentage of cases analyzed within each oncogenic group of patients with SM-AHN. MSC were investigated only for KIT p.D816V. Three patients showed KIT mutations other than p.D816V (KIT p.D816A, KIT p.D816Y, and KIT p.K558_S584dup). Two of 5 patients had KIT-positive MSC. One of 2 patients had KIT-positive MSC. Four of 4 patients had KIT-positive MSC. Oncogenic subgroups of SM-AHN Based on the aforementioned patterns of involvement of BM MC and AHN cells by the KIT mutation and/or AHN-related genetic alterations and the number of genetically altered cells, hypothetical models of clonal evolution were built for individual patients, as illustrated in Figure 1. Those oncogenic models showed that in 21 of 79 patients (27%), the KIT mutation and AHN-related genetic alterations were unrelated events present in different cell compartments. Conversely, in the other 58 of 79 patients (73%), the coexisting SM and AHN cell compartments were genetically related because they shared a KIT mutation (13/79 [16%]), AHN-related genetic alterations (9/79 [11%]), or both (36/79 [45%]). Remarkably, in 11 of the latter cases (11/36 [31%]), the KIT mutation was found in a greater median number of precursors and residual myeloid cells (also including residual BM lymphoid cells in 10/11 patients) compared with the AHN-related genetic alterations: 100% (range, 80%-100%) vs 50% (range, 20%-75%), respectively (P < .0001). In another 16 patients (16/36 [44%]), greater median percentages of BM cells carrying AHN-related genetic alterations than those carrying the KIT mutation were found: 94% (range, 8%-100%) vs 27% (range, 2%-88), respectively (P < .0001). In the remaining 9 of 36 patients (25%), the KIT mutation and AHN-related alterations were found at identically high median percentages (≈100% vs ≈100%, respectively). Based on these results, 4 distinct oncogenic groups of SM-AHN were defined: (1) a group of patients with genetically unrelated SM and AHN cell compartments (21/79 [27%]) and 3 SM-AHN groups with shared genetic alterations in both cell compartments in which (2) the KIT mutation was the first alteration (24/79 [30%]), also associated with greater involvement of hematopoiesis by the KIT mutation vs that observed for AHN-related genetic alterations; (3) the AHN-related genetic markers emerged first, leading to more extensive hematopoietic involvement by those alterations vs the KIT mutation (25/79 [32%]); or (4) both alterations involved virtually all cells, with high (≈100%) percentages of mutated cells (9/79 [11%]; Table 3). Hypothetical clonal evolution pathways of 4 representative patients with SM-AHN illustrating different patterns of acquisition of the KIT mutation and AHN-associated genetic/molecular alterations during hematopoiesis, from HSC (or MSC) to CD34+ and CD34− myeloid and/or lymphoid progenitors, BM MC, and AHN cells. (A-D) The hypothetical clonal evolution pathways of patient 15 (ISM + SMZL) (A), patient 36 (ASM + MDS-IB2) (B), patient 69 (ASM + CMML) (C), and patient 76 (ASM + CMML) (D) are represented. For each patient, the frequency (percentage of mutated cells) of the coexisting genetic alterations detected in CD34+ HSC (and/or MSC), CD34− maturing cells, MC, and AHN cells is shown. Orange dots and backgrounds represent the KIT-mutated neoplastic cell compartments, whereas blue dots and backgrounds represent AHN-associated genetic alterations acquired by genetically altered precursor cells and their descendant neoplastic cell compartments. Percentages indicate the proportion of mutated/genetically altered cells. HSC, hematopoietic stem cells; SMZL, splenic MZL. Figure created with biorender.com. Navarro P. (2026) https://biorender.com/hhmyj27. Table 3.Distribution of different diagnostic subtypes of SM and AHN in patients with SM-AHN grouped according to the pattern of acquisition of the KIT mutation and AHN-related genetic/molecular alterations by BM MC and other non-MC hematopoietic cellsDiagnostic subtypeTotalPattern of acquisition of the KIT mutation and AHN-related genetic/molecular alterations by BM MC and AHN cellsP valueGenetically unrelated MC and AHN cells†KIT mutation first‡AHN-related genetic alteration first§Simultaneous KIT- and AHN-related genetic alterations||SM subtype BMM21 (27)12 (57)∗,∗∗∗,‡,||3 (14)6 (29)0 (0)<.001 ISM23 (30)5 (22)10 (43)7 (30)1 (4)∗,‡.020 ASM/MCL35 (43)4 (11)11 (32)12 (34)8 (23)ns ASM34 (42)4 (12)11 (32)12 (35)7 (21)ns MCL1 (1)0 (0)0 (0)0 (0)1 (100)nsAHN subtype AMN62 (78)11 (18)17 (27)25 (40)∗,∗∗,†,||9 (15).004 AML17 (21)5 (29)6 (35)5 (29)1 (6)ns MDS/MPN45 (57)6 (13)11 (24)20 (44)∗,∗∗,†,||8 (18).004 MDS12 (15)2 (17)4 (33)2 (17)4 (33)ns MPN12 (15)4 (33)3 (25)4 (33)1 (8)ns MDS/MPN21 (27)0 (0)4 (19)14 (67)∗∗,†,‡,§3 (14)<.001 ALN11 (14)9 (82)∗,∗∗,‡,§2 (18)0 (0)0 (0)<.001 Mature B-cell neoplasm6 (8)6 (100)∗∗,‡,§,||0 (0)0 (0)0 (0)<.001 Mature PC neoplasm3 (4)2 (67)1 (33)0 (0)0 (0)ns T-cell neoplasm1 (1)0 (0)1 (100)0 (0)0 (0)ns Mixed B/T neoplasm1 (1)1 (100)0 (0)0 (0)0 (0)ns Mixed AMN/ALN6 (8)1 (17)5 (83)0 (0)0 (0).002Total79 (100)21 (27)24 (30)25 (32)9 (11)∗,‡,§.012Results are expressed as number of cases (percentage). Statistical analyses were performed using the χ2 test for continuous variables and Kruskal-Wallis test for categorical variables. When statistically significant differences were observed, pairwise comparisons were subsequently conducted using the Mann-Whitney U test or χ2 test, with the Bonferroni correction for multiple comparisons. Overall P values are reported in the P value column, whereas the significance (corrected) of individual pairwise comparisons is indicated as follows: ∗P < .05; ∗∗P < .01; ∗∗∗P < .001.MPN, myeloproliferative neoplasm; ns, no statistically significant differences found.†Genetically unrelated MC and AHN cells group.‡KIT mutation first group.§AHN-related genetic alteration first group.||Simultaneous KIT- and AHN-related genetic alterations group. Distribution of different diagnostic subtypes of SM and AHN in patients with SM-AHN grouped according to the pattern of acquisition of the KIT mutation and AHN-related genetic/molecular alterations by BM MC and other non-MC hematopoietic cells Results are expressed as number of cases (percentage). Statistical analyses were performed using the χ2 test for continuous variables and Kruskal-Wallis test for categorical variables. When statistically significant differences were observed, pairwise comparisons were subsequently conducted using the Mann-Whitney U test or χ2 test, with the Bonferroni correction for multiple comparisons. Overall P values are reported in the P value column, whereas the significance (corrected) of individual pairwise comparisons is indicated as follows: ∗P < .05; ∗∗P < .01; ∗∗∗P < .001. MPN, myeloproliferative neoplasm; ns, no statistically significant differences found. Genetically unrelated MC and AHN cells group. KIT mutation first group. AHN-related genetic alteration first group. Simultaneous KIT- and AHN-related genetic alterations group. Clinical, biological, and molecular features of the oncogenic subgroups of SM-AHN Patients who showed genetically unrelated MC and AHN cell populations more frequently had BMM (P = .04) and ISM (P = .02), whereas a trend toward a higher prevalence of ASM was found in patients who had the KIT mutation or AHN-related genetic events emerging first vs those with unrelated genetic markers (Table 3). Interestingly, patients with SM-AHN who first acquired AHN-related alterations more frequently had an AMN (40%) compared with patients with unrelated (18%) or simultaneous genetic alterations (15%) (P = .03 and P = .006, respectively), particularly when the AMN was CMML (P < .001). Notably, all except 1 SM-AHN patient with mixed AMN/ALN corresponded to patients with SM-AHN in whom the KIT mutation emerged first (P = .001) (Table 3). From the clinical point of view, patients with genetically unrelated MC and AHN cell compartments more frequently showed symptoms associated with SM, such as anaphylaxis (P = .01), in addition to lower serum baseline tryptase (sBT) levels (P = .005) and lower percentages of both BM MC by flow cytometry (P = .01) and KIT p.D816V–mutated cells (P = .01), together with signs of less advanced disease, including lower frequency of organomegalies (P = .002), and a lower rate of disease progression (P = .048) (Table 4). Remarkably, sBT levels and percentage of KIT-mutated cells were particularly high among patients with SM-AHN in whom the KIT mutation emerged first (or in parallel with the AHN-related alteration) compared with the other oncogenic groups (P = .03 for both). In turn, patients in whom the AHN-related genetic alterations emerged first showed the lowest platelet counts (P = .02), whereas simultaneous acquisition of the KIT mutation and AHN-related alterations was associated with the highest serum alkaline phosphatase levels (P = .03) and BM MC percentages by flow cytometry (P = .01) (Table 4). Table 4.Clinical, laboratory, and genetic features of patients with SM-AHN grouped according to the pattern of acquisition of the KIT mutation and AHN-related genetic/molecular alterations by BM MC and AHN cellsDisease featuresPattern of acquisition of the KIT mutation and AHN-related genetic/molecular alterations by BM MC and AHN cellsP valueGenetically unrelated MC and AHN cells† (n = 21)KIT mutation first‡ (n = 24)AHN-related genetic alteration first§ (n = 25)Simultaneous KIT and AHN-related genetic alterations|| (n = 9)Age, y59 (43-82)66 (11-90)69 (10-81)70 (1-79)nsSex, male12/21 (57)13/25 (52)19/24 (79)8/9 (89)nsSkin lesion10/19 (47)16/25 (36)17/24 (29)5/9 (55)nsAnaphylaxis7/15 (53)∗∗,∗∗∗,‡,§0/21 (0)1/18 (5)0/8 (0)<.001Organomegaly5/17 (29)∗,∗∗∗,‡,||17/23 (74)15/25 (60)9/9 (100).002Lymphadenopathy1/15 (6)∗,§,||9/21 (43)9/19 (47)5/8 (63).027Splenomegaly5/17 (29)∗∗,||16/24 (67)14/24 (58)9/9 (100).005Hepatomegaly2/15 (12)11/24 (46)8/24 (33)3/9 (33)nsHb, g/L138 (84-154)∗,‡,§,||106 (54-155)113 (69-146)97 (80-134).032WBC count, ×109/L7.9 (1.6-23.3)9.6 (1.9-96.3)6.1 (2.2-39.2)11 (3.2-43.4)nsPlatelet count, ×109/L227 (26-921)∗∗,§,||155 (29-368)∗,§80 (15-644)95 (12-222).003SAP, IU/L65 (30-512)∗,∗∗,‡,§,||139 (60-549)152 (40-587)∗,||375 (155-1160).008sβ2m. μg/mL2.2 (1-6.7)∗,‡,§3.9 (0.6-12.8)3.5 (1.9-11.4)4.1 (3.4-5.9).04†sIgE, kU/L18.6 (2-1370)9.8 (2.1-89)∗∗,§34 (3.8-949)5.3 (3.7-14.2).02‡sBT, μg/L19 (5.3-418)∗,∗∗,‡,||176 (18.2-390)∗,§90 (6.9-1376)147 (50-1328).005Osteoporosis5/11 (45)1/14 (7)2/10 (20)1/5 (20)nsOsteopenia5/11 (45)2/14 (14)3/10 (30)1/5 (20)nsOsteolytic lesions1/11 (9)1/14 (7)0/10 (0)1/5 (20)nsOsteosclerosis1/11 (9)0/14 (0)2/10 (20)0/5 (0)nsHαT3/14) (21)1/16 (6)2/16 (12)1/5 (20)nsKIT p.D816V–mutated cells, %0.168 (0-84)∗,∗∗∗,‡,§,||72 (1.14-100)∗∗,§17 (0.12-100)68 (24-80).003BM MC by FCM, %0.1 (0.01-3.2)∗,∗∗,‡,||0.4 (0.01-11.1)0.13 (0.01-18)∗,||2.1 (0.09-27).01BM MC by cytology, %2.5 (1-8)∗,‡8 (2-16)6 (2-60)8.5 (5-12).03†BM MC by histopathology, %5 (2-70)60 (3-70)10 (5-40)30 (20-40)nsBM blasts, %10 (0.1-87)20 (0.2-83)33 (2-66)21 (4-70)nsDisease progression2/19 (11)∗∗,||8/24 (33)9/25 (36)4/8 (50).02†Deaths4/19 (21)∗,§11/24 (46)14/25 (56)5/9 (56).05†Genetic alterations FLT30/21 (0)0/24 (0)0/25 (0)2/9 (22).001 SF3B10/21 (0)0/24 (0)1/25 (4)2/9 (22).02 SRSF24/21 (19)∗,§7/24 (29)15/25 (53)3/9 (33).03 TET22/21 (10)8/24 (33)12/25 (48)0/9 (0).006 TP530/21 (0)0/24 (0)6/25 (24)1/9 (11).009 D13S253/21 (14)0/24 (0)0/25 (0)0/9 (0).04 t(14q32)¶3/21 (14)0/24 (0)0/25 (0)0/9 (0).04 No. of alterations per patient1 (0-3)1 (0-4)2 (1-4)2 (1-4).01Results are expressed as number of cases from cases analyzed (percentage) or as median (range), unless otherwise specified. This cohort included 3 pediatric patients aged 1, 10, and 11 years. Statistical analyses were performed using the χ2 test for continuous variables and Kruskal-Wallis test for categorical variables. When statistically significant differences were observed, pairwise comparisons were subsequently conducted using the Mann-Whitney U test or χ2 test, with the Bonferroni correction for multiple comparisons. Overall P values are reported in the P value column, whereas the (corrected) significance of individual pairwise comparisons is indicated as follows: ∗P < .05; ∗∗P < .01; ∗∗∗P < .001.FCM, flow cytometry; HαT, hereditary α-tryptasemia; Hb, hemoglobin; ns, no statistically significant differences found; SAP, serum alkaline phosphatase; sIgE, serum immunoglobulin E; sβ2m, serum β2-microglobulin; WBC, white blood cell count.†Genetically unrelated MC and AHN cells group.‡KIT mutation first group.§AHN-related genetic alteration first group.||Simultaneous KIT- and AHN-related genetic alterations group.¶t(14q32) other than (t11;14) involving the CCND1::IGH genes. Clinical, laboratory, and genetic features of patients with SM-AHN grouped according to the pattern of acquisition of the KIT mutation and AHN-related genetic/molecular alterations by BM MC and AHN cells Results are expressed as number of cases from cases analyzed (percentage) or as median (range), unless otherwise specified. This cohort included 3 pediatric patients aged 1, 10, and 11 years. Statistical analyses were performed using the χ2 test for continuous variables and Kruskal-Wallis test for categorical variables. When statistically significant differences were observed, pairwise comparisons were subsequently conducted using the Mann-Whitney U test or χ2 test, with the Bonferroni correction for multiple comparisons. Overall P values are reported in the P value column, whereas the (corrected) significance of individual pairwise comparisons is indicated as follows: ∗P < .05; ∗∗P < .01; ∗∗∗P < .001. FCM, flow cytometry; HαT, hereditary α-tryptasemia; Hb, hemoglobin; ns, no statistically significant differences found; SAP, serum alkaline phosphatase; sIgE, serum immunoglobulin E; sβ2m, serum β2-microglobulin; WBC, white blood cell count. Genetically unrelated MC and AHN cells group. KIT mutation first group. AHN-related genetic alteration first group. Simultaneous KIT- and AHN-related genetic alterations group. t(14q32) other than (t11;14) involving the CCND1::IGH genes. Detailed analysis of the distribution of different genetic alterations among patients with SM-AHN revealed that single-point mutations were the most commonly identified. Lymphoid-associated genetic alterations, such as del(13q14.3; D13S25) and all t(14q32) other than t(11;14), were restricted to patients with genetically unrelated neoplastic cell compartments (P = .04 for both; Table 4). In contrast, ASXL1, BCOR, KRAS, CBLC, FLT3, SF3B1, and TP53 were detected only in patients with shared genetic alterations, with the latter 4 genes restricted to patients with shared AHN-related genetic alterations. Of note, FLT3 and SF3B1 mutations (P = .001 and P = .02) were restricted to patients with simultaneous acquisition of the KIT mutation and AHN-related alterations (Table 4). In addition, mutations in TET2 (P = .006) were significantly more prevalent in patients in whom the AHN-related alteration emerged first (Table 4). Other mutations (DNMT3A, JAK2, NPM1, PIGA, RUNX1, and SRSF2) were found across all oncogenic groups, with SRSF2 present at a higher prevalence among patients in whom the AHN-related alterations emerged first (53% vs 19%, 29%, and 33%; P = .003) (Table 4; supplemental Table 3). Impact of the oncogenic groups on survival in SM-AHN Univariable analysis of prognostic factors for PFS revealed that the pattern of involvement of BM MC and AHN cells by the KIT mutation and AHN-related genetic alterations identified 2 major patient risk groups, particularly among patients with SM associated with a non-AML hematological malignancy (supplemental Figure 1): (1) low-risk patients with SM-AHN carrying unrelated mutational events and (2) high-risk patients in whom the AHN-related genetic alterations and/or the KIT mutation involved both BM MC and AHN cells, with 5-year PFS rates of 85% (95% confidence interval, 52-96) vs 69% (95% confidence interval, 52-81), respectively (P = .015; Figure 2A-B). Other variables that showed an adverse impact on PFS included multilineage involvement of BM hematopoiesis by the KIT mutation (P = .008); the subtype of SM (ASM vs BMM/ISM; P < .001); the subtype of AHN (CMML vs other AHN; P = .019); lower hemoglobin levels (<110 g/L; P = .011); the presence of organomegalies (P = .046), including splenomegaly (P = .017) and hepatomegaly (P = .019); the absence of osteoporosis (P = .048); tyrosine kinase inhibitor administration (P < .001; supplemental Figure 2); the Global Prognostic Score for Mastocytosis60 (GPSM) score for PFS (P = .05); the REMA prognostic score41 (P = .025); and the presence of SRSF2, ASXL1, and/or RUNX1 mutations (P = .01; Figure 3A; supplemental Figure 3). Receiver operating characteristic curve analyses identified 2% KIT-mutated cells in whole blood (n = 21) and 1.4% KIT-mutated cells in whole BM (n = 34) as the most accurate cutoff values to predict multilineage (vs MC-restricted) involvement in our SM-AHN cohort, with a specificity of 100% and 100% and a sensitivity of 100% and 95%, respectively. Multivariable analysis revealed that, among those variables, the REMA prognostic score was the only parameter with independent predictive value for PFS (hazard ratio, 2.29; P = .017; Figure 3A). Impact of the oncogenic profile of BM cells defined by the KIT mutation and AHN-related genetic alterations on the outcome of SM-AHN. (A,C) PFS and OS of patients with SM-AHN grouped according to the 4 oncogenic profiles identified. (B,D) PFS and OS of patients with SM-AHN with genetically unrelated vs genetically related BM MC and AHN cell compartments. Follow-up time is expressed in years from diagnosis to disease progression (PFS), death (OS), or last follow-up. Number at risk indicates the number of patients in each group at diagnosis (time 0) and at 5, 10, and 15 years after diagnosis. Median PFS and OS are defined as the time at which 50% of patients in a given group had experienced disease progression or death, respectively, and correspond to the point at which each survival curve crosses the dotted horizontal line. Statistical significance was analyzed using the log-rank test. Impact of individual prognostic factors on PFS and OS in patients with SM-AHN. (A,B) Univariable and multivariable analyses were performed to evaluate the association of each individual prognostic factor with PFS (A) and OS (B). Data are expressed as hazard ratios (HR; black dots) with corresponding 95% CI (horizontal lines) and P values. Statistical significance was analyzed using the log-rank test and the Cox proportional hazards model for univariable and multivariable analyses, respectively. HR <1 indicate no adverse impact on PFS or OS, whereas HR >1 indicate a significant adverse impact on patient outcomes. The dotted vertical line indicates HR = 1. β2m, β2-microglobulin; CI, confidence interval; IPSM-Adv, International Prognostic Scoring System for Advanced Mastocytosis; Hb, hemoglobin; S/A/R, SRSF2, ASXL1, and/or RUNX1; WBC, white blood cell count. Likewise, the pattern of involvement of BM MC and AHN cells by the KIT mutation and AHN-related alterations (genetically related vs unrelated cell compartments; P = .022), the presence of multilineage (vs MC-restricted) KIT involvement (P = .05), the subtype of SM (ASM vs BMM/ISM; P = .001), and AHN (AMN vs ALN [P = .006]; and CMML vs other AHN [P = .008]), together with age (>60 years) (P = .005), absence of skin lesions (P = .006), splenomegaly (P = .035), lower hemoglobin (≤110 g/L; P < .001) and platelet counts (<100 × 109/L; P = .014), increased white blood cell counts (≥16 × 109/L; P = .022), β2-microglobulin serum levels (>2.50 μg/mL; P = .011), a greater number (≥2) of gene mutations (P = .013) and their type (SRSF2, ASXL1, and/or RUNX1 mutations; P = .036), as well as the GPSM, the International Prognostic Scoring System for Advanced Mastocytosis61 and the Mutation-Adjusted Risk Scores62 (MARS; P ≤ .001 for all 3), were all found to be adverse prognostic factors for OS in the univariable analysis (Figure 3B; supplemental Figure 3). Among those parameters, MARS (hazard ratio, 6.26; P < .001) was the only one that retained independent predictive value for OS in the multivariable analysis (Figure 3B). Discussion SM-AHN is a well-recognized poor-prognosis subtype of SM that accounts for 40% to 60% of all AdvSM.1,4,40,63 From the clinical, genetic, and prognostic point of view, SM-AHN exhibits marked heterogeneity.1,26 Thus, the SM and AHN subtypes that coexist within SM-AHN vary substantially, resulting in multiple combinations of malignant diseases that require distinct treatment approaches (ie, ASM vs ISM and AML vs chronic lymphocytic leukemia).7,8,64 Although few studies based on limited series of patients with SM-AHN have been reported, the SM and AHN subtypes are consistently considered the main determinants of treatment decisions, disease behavior, and patient outcome, and thereby the most relevant prognostic factors in SM-AHN.12,24 Accumulated evidence suggests that patients with SM-AHN might display distinct oncogenic profiles defined by different patterns of acquisition of the KIT mutation and AHN-related genetic alterations by BM MC and/or AHN cells (and potentially also other non-MC hematopoietic cells), even within the same subtype of SM-AHN.1,4,25,40,55,63,65, 66, 67, 68 However, the frequency and potential impact on patient outcome of those oncogenic profiles have not been thoroughly investigated. Here, we confirmed the existence of 4 major oncogenic profiles in SM-AHN, characterized by different patterns of acquisition of the KIT mutation and AHN-related genetic alterations identified in the coexisting BM neoplastic MC and AHN cell compartments, and the degree of involvement of the underlying hematopoiesis by those alterations. These profiles delineated 2 major groups with genetically related and unrelated neoplasms. In patients with genetically related SM and AHN cell compartments, the KIT mutation and AHN-related genetic alterations were asynchronously or synchronously acquired during neoplastic hematopoiesis. In contrast, genetically unrelated patients likely originated from distinct clonal events, although the possibility of an earlier common ancestor clone cannot be excluded. Despite this limitation, more detailed mutational profiling revealed marked differences among these oncogenic groups. Thus, mutations in DNMT3A, JAK2, NPM1, PIGA, RUNX1, and SRSF2 were present across all oncogenic groups. In contrast, lymphoid-associated genetic alterations occurred almost exclusively in genetically unrelated SM-AHN, whereas ASXL1, BCOR, KRAS, CBLC, FLT3, and TP53 gene mutations appeared only in patients with shared genetic backgrounds, with mutations in the latter 4 genes restricted to patients with shared AHN-related genetic alterations. These findings reinforce the existence of different oncogenic pathways with unique molecular signatures among SM-AHN, consistent with prior studies.32,34,37,42, 43, 44, 45 Patients with shared genetic alterations between MC and AHN cells were identified more frequently than those with genetically unrelated neoplasms. The former were predominantly associated with AMN, particularly CMML, and SM subtypes such as ASM, whereas the latter mainly encompassed BMM, ISM, and ALN. These results support previous observations1,69 suggesting that coexistence of SM with specific AHN subtypes is not random but is underpinned by shared molecular mechanisms. Thus, occurrence of an AHN is a relatively common finding in SM, both at diagnosis and during follow-up,1,24,32,69, 70, 71 particularly among patients with SM presenting with greater involvement of BM hematopoietic myeloid and lymphoid cells,42 including MSC,38,41 by the KIT mutation. In turn, KIT mutations have also been identified in a subset of myeloid malignancies, such as CMML (≈5%),72 MDS (≈5%),56,73 and AML, especially those with RUNX1::RUNX1T1 and CBFB::MYH11 gene rearrangements (≈20%),74, 75, 76 typically at lower percentages of mutated cells than those found in ASM and MCL.1,8,27, 28, 29,41,64 This nonrandom overlap underscores the biological link in SM-AHN between SM and certain AMNs, as also found here. Notably, in our cohort, MCL-AHN occurred in only 1 patient, possibly reflecting, at least in part, our inclusion criteria of patients with KIT-mutated SM-AHN, together with the exclusion of secondary MCL-AHN, because both KIT-mutated MCL and MCL itself are relatively rare.77,78 Remarkably, in most patients with ASM-AHN, multilineage KIT involvement was detected across both myeloid and lymphoid cells, as well as MSC, in addition to BM MC and AHN cells, consistent with prior reports.42 These observations might help explain the close association between early emergence of the KIT mutation, extensive involvement of BM hematopoietic myeloid and lymphoid cells, and the coexistence of SM with both myeloid and lymphoid neoplasms such as SMM and MM, which, to our knowledge, is documented here for the first time. Globally, our findings suggest the existence of distinct underlying biological scenarios in SM-AHN. In approximately one-third of patients, AHN-related genetic alterations may precede the KIT mutation triggering secondary SM development. Conversely, in another one-third of patients with SM-AHN, early KIT mutations in hematopoietic stem cells likely promote myeloid-biased clonal expansions and subsequent acquisition of AHN-related genetic alterations, resulting more often in AMN than ALN. This aligns with prior work showing that myeloid-associated mutations in SM are largely confined to AdvSM.40,41,55 Overall, this mechanism would support predominance of myeloid vs lymphoid neoplasms in SM-AHN, despite the general prevalence of lymphoid malignancies in hematology.42 Remarkably, concurrent acquisition of a KIT mutation and AHN-related genetic alterations might occur in only a minority (≈10%) of patients with SM-AHN, which exhibited massive involvement of BM cells by both genetic alterations. Overall, our results contrast with earlier in vitro clonogenic studies suggesting that KIT mutations are typically secondary events.37 However, those findings were based on a limited number of cases (n = 13) and ex vivo CD34+ cell assays that often promote expansion of KIT wild-type vs KIT-mutated precursors,37,79 which might help explain those apparent discrepancies. Future single-cell analyses, which were unavailable here because of retrospective sampling, will be crucial to further elucidate SM-AHN clonal architectures associated with the oncogenic profiles identified here. Although most patients with SM-AHN harbored genetic markers shared by SM and AHN cells, supporting a common neoplastic stem cell origin, in a smaller proportion of our patients with SM-AHN, 2 genetically unrelated compartments of clonal MC and AHN cells were also identified. These latter patients typically presented with unrelated nonadvanced SM and ALN with an MC-restricted KIT mutation and limited involvement of BM hematopoiesis, characterized by symptoms27,80 such as anaphylaxis, skin lesions, and/or osteoporosis; low BM MC burden (supplemental Table 4; Figure 2); absence of organ damage; and significantly prolonged PFS and OS,81,82 particularly among patients with SM associated with a non-AML hematological neoplasm but not SM-AML cases. Conversely, patients with SM-AHN and shared KIT mutations and/or AHN-related genetic alterations exhibited features of more advanced disease, such as higher frequency of organomegalies and cytopenia, together with higher sBT and serum β2-microglobulin levels and poorer outcome.83,60 Whether these features result primarily from the SM, the coexisting AHN, or their interaction remains uncertain. Despite the preceding findings, multivariable analysis of prognostic factors for PFS and OS revealed that the oncogenic profile of BM cells in SM-AHN had no independent predictive value for either PFS or OS once well-established prognostic scores, such as the REMA score41 and MARS,62 were accounted for, respectively. These latter findings were because of the close relationship identified here between the oncogenic profiles and currently available prognostic risk scores (ie, MARS, GPSM, International Prognostic Scoring System for Advanced Mastocytosis), in particular the REMA score and MARS (supplemental Figures 4 and 5). In summary, our findings confirm the clinical, genetic, and prognostic heterogeneity of SM-AHN and reveal its link to distinct oncogenic pathways in neoplastic MC and AHN cells. Although oncogenic profiles correlated with patient outcome, their prognostic value appears to be superseded by well-established risk scores. However, patients in this cohort were treated with various treatment modalities, mostly before the availability of KIT-targeted therapies (ie, avapritinib), and therefore, further studies in larger, uniformly treated cohorts of patients with SM-AHN are warranted to clarify the prognostic impact of the oncogenic profiles described here. Conflict-of-interest disclosure: The authors declare no competing financial interests. Acknowledgments This work was supported by grants from the Instituto de Salud Carlos III (ISCIII, Ministerio de Ciencia e Innovación), European Regional Development Fund (FEDER) grants PI16/00642, PI19/01166, and PI25/00372, Centro de Investigación Biomédica en Red de Cáncer (CIBERONC) program grant CB16/12/00400, and Blueprint Medicines Corporation grants 2023-IST-000117 and 2025-ISR-000513. P.N.-N. was supported by a grant from the Government of Castilla y León, Spain (Orden EDU 875 2021), cofinanced by the European Social Fund (Base de Datos Nacional de Subvenciones: 540787). A.P.-P. was supported by CIBERONC contract number 2907. C.T. was supported by an Andrés Laguna fellowship (Junta de Castilla y León, cofinanced by Fondo Social Europeo Plus; Orden EDU/300/2025). S.M. was supported by ISCIII through grant PI21/01115 and cofunded by the European Union, CIBERONC, and FONDOS FEDER grant CB16/12/00400. Authorship Contribution: P.N.-N. performed experiments, analyzed data, interpreted results, prepared the figures, and wrote the manuscript; C.M.F.-G. and A.C.G.-M. performed experiments, analyzed data, interpreted results, and critically reviewed the manuscript; A.P.-P., C.M.-C., C.T., M.G.-G., M.G.-T., M.M., S.M., A.L., Ó.G.-L., and P.B. performed experiments and critically reviewed the manuscript; I.Á.-T., L.S.-M., A.H., E.C., J.E., and M.L.S. recruited and followed the patients, collected samples, and critically reviewed the manuscript; M.J.-A. designed the study, performed experiments, analyzed the data, interpreted results, supervised the study, and wrote the manuscript; A.O. designed and supervised the study, interpreted results, and wrote the manuscript; and all authors reviewed and approved the manuscript before submission. Footnotes M.J.-A. and A.O. contributed equally to this study and are joint last authors. Data are available from the corresponding author, Alberto Orfao (orfao@usal.es), on request. The full-text version of this article contains a data supplement. Supplementary Material Supplemental Methods, Results, Tables, and Figures References Associated Data Supplementary Materials Supplemental Methods, Results, Tables, and Figures