# paper.pdf ## Metadata - PDFFormatVersion=1.7 - Language=en - IsLinearized=true - IsAcroFormPresent=false - IsXFAPresent=false - IsCollectionPresent=false - IsSignaturesPresent=false - CreationDate=D:20260624172913+05'30' - Subject=BLOOD ADVANCES, 10 (2026) 4519-4532. doi:10.1182/bloodadvances.2025018435 - Author=Irene Dogliotti∗ - Creator=Elsevier - Producer=PDFlib+PDI 10.0.2p2 (C++/Win64) - Custom.EPSprocessor=PStill version 1.84.42 - Custom.ElsevierWebPDFSpecifications=7.0.1 - Custom.robots=noindex - Custom.doi=10.1182/bloodadvances.2025018435 - ModDate=D:20260624173328+05'30' - Title=Quantitative MYD88 L265P and flow cytometry levels for outcome determination in IgM gammopathies: the SAL-TO study - dc:format=application/pdf - dc:identifier=doi:10.1182/bloodadvances.2025018435 - dc:title=Quantitative MYD88 L265P and flow cytometry levels for outcome determination in IgM gammopathies: the SAL-TO study - dc:creator=Irene Dogliotti∗,Cristina Jiménez∗,Veronica Peri,Martina Ferrante,Davide Musto,Silvio Mercadante,Gian Maria Zaccaria,Marco Ghislieri,Giulia Benevolo,Enrique María Ocio,Araceli Rubio,Ilda Murillo,Fernando Escalante,Carmen Aguilera,Mateo Aránzazu García,Alfonso García de Coca,Roberto Hernández,Julio Dávila,Federica Cavallo,Noemi Puig,Veronica González de la Calle,Maria Eugenia Sarasquete,Miguel Alcoceba,Simone Ragaini,Michele Clerico,Chiara Consoli,Enrico Amaducci,María García-Álvarez,María del Carmen Chillón,Alejandro Medina,Marcos González Díaz,Norma Carmen Gutiérrez,Benedetto Bruno,Daniela Drandi,Simone Ferrero,Ramon García-Sanz - dc:description=BLOOD ADVANCES, 10 (2026) 4519-4532. doi:10.1182/bloodadvances.2025018435 - dc:publisher=American Society of Hematology - prism:aggregationtype=journal - prism:publicationname=BLOOD ADVANCES - prism:copyright=© 2026 American Society of Hematology. Published by Elsevier Inc. Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0), permitting only noncommercial, nonderivative use with attribution. All other rights reserved. - prism:issn=2473-9529 - prism:volume=10 - prism:number=13 - prism:coverdisplaydate=14 July 2026 - prism:coverdate=2026-07-14 - prism:pagerange=4519-4532 - prism:startingpage=4519 - prism:endingpage=4532 - prism:doi=10.1182/bloodadvances.2025018435 - prism:url=https://doi.org/10.1182/bloodadvances.2025018435 - jav:journal_article_version=VoR - pdfx:elsevierwebpdfspecifications=7.0.1 - pdfx:doi=10.1182/bloodadvances.2025018435 - pdfx:robots=noindex - pdfx:epsprocessor=PStill version 1.84.42 - xap:creatortool=Elsevier - xap:createdate=2026-06-24T17:29:13+05:30 - xap:modifydate=2026-06-24T17:33:28+05:30 - xap:metadatadate=2026-06-24T17:33:28+05:30 - xaprights:marked=True - tdm:reservation=1 - tdm:policy=https://www.elsevier.com/tdm/tdmrep-policy.json - pdf:producer=PDFlib+PDI 10.0.2p2 (C++/Win64) - xapmm:documentid=uuid:bc19f336-9e4e-42f2-9f43-99cf6e55617a - xapmm:instanceid=uuid:3f590481-ef78-432f-860e-99cc469f008e - ali:license_ref=https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode ## Contents ### Page 1 ![Figure]() Quantitative MYD88 L265P and flow cytometry levels for outcomedetermination in IgM gammopathies: the SAL-TO study Irene Dogliotti,1,* Cristina Jimenez, ´2,* Veronica Peri,1 Martina Ferrante,1 Davide Musto,1 Silvio Mercadante,1 Gian Maria Zaccaria,3Marco Ghislieri,4 Giulia Benevolo,1 Enrique María Ocio,5 Araceli Rubio,6 Ilda Murillo,6 Fernando Escalante,7 Carmen Aguilera,8Mateo Aranzazu ´ García,9 Alfonso García de Coca,10 Roberto Hernandez, ´ Julio Davila, ´ Federica Cavallo,1 Noemi Puig,2Veronica Gonzalez ´ de la Calle,2 Maria Eugenia Sarasquete,2 Miguel Alcoceba,2 Simone Ragaini,1 Michele Clerico,1 Chiara Consoli,1Enrico Amaducci,1 María García-Álvarez,2 María del Carmen Chillon, ´ Alejandro Medina,2 Marcos Gonzalez ´ Díaz,2 Norma Carmen Gutierrez, ´ Benedetto Bruno,1 Daniela Drandi,1 Simone Ferrero,1 and Ramon García-Sanz2,13 1 Division of Hematology, Department of Molecular Biotechnology and Health Sciences, University of Torino, Turin, Italy;2 Hematology Department, University Hospital ofSalamanca, Research Biomedical Institute of Salamanca, Centro de Investigacion ´ Biomedica ´ en Red-Cancer and Center for Cancer Research-Instituto de Biología Moleculary Celular del Cancer, Universidad de Salamanca-Consejo Superior de Investigaciones Científicas, Salamanca, Spain;3 Department of Electrical and Information Engineering,Polytechnic University of Bari, Bari, Italy;4 PolitoBIO Med Lab and Department of Electronics and Telecommunications, Politecnico di Torino, Turin, Italy;5 HematologyDepartment, University Hospital of Marques ´ de Valdecilla, Santander, Spain;6 Hematology Department, Hospital Miguel Servet, Zaragoza, Spain;7 Hematology Department,Hospital Complex of Leon, ´ Leon, ´ Spain;8 Hematology Department, Regional Hospital of El Bierzo, Leon, ´ Spain;9 Hematology Department, General Hospital of Segovia,Segovia, Spain;10 Hematology Department, University Clinical Hospital of Valladolid, Valladolid, Spain;11 Hematology Department, Virgen de la Concha Hospital, Zamora,Spain;12 Hematology Department, Nuestra Senora˜ de Sonsoles Hospital, Ávila, Spain; and13 Department of Haematology, Gregorio Maran˜on ´ General University Hospital,Madrid, Spain Key Points • Combining MYD88L265P and flowcytometry data enablesstratification of patientswith IgMgammopathies into riskgroups with distinctsurvival outcomes. Waldenström macroglobulinemia (WM) is a rare, indolent B-cell lymphoproliferativedisorder, often preceded by a history of immunoglobulin M (IgM) monoclonalgammopathy of undetermined significance (IgM-MGUS). In this retrospective, multicenterstudy, we collected real-life data from 577 patients with IgM gammopathy (221symptomatic WM [sWM], 245 asymptomatic WM [aWM], 111 IgM-MGUS) from 22 Spanishcenters with a validation cohort of 166 patients (73 sWM, 71 aWM, 22 IgM-MGUS) from theUniversity Hospital of Torino, Italy. The median overall survival (OS) was 126.7 months forthe Spanish cohort and 202.8 for the Torino cohort. A multivariate analysis identified age>65 years, male gender, diagnosis of sWM, and β-2-microglobulin >3 as significantpredictors of a shorter OS. In addition, age >65 years, bone marrow (BM) biopsyinfiltration, hemoglobin <11.5 g/dL, and platelets <100 000/μL were associated with ashorter time to first treatment (TTFT). Pooling data from both cohorts revealed that abaseline BM quantitative MYD88 L265P to wild-type MYD88 ratio of >0.162 (either bydroplet digital polymerase chain reaction [PCR] or quantitative PCR), together withmultiparameter flow cytometry (MFC) infiltration >4.39%, had a significant impact on OSand TTFT. The combination of MYD88 and MFC levels enabled stratification of patients intohigh-, intermediate-, and low-risk groups with patients with high-risk IgM gammopathyshowing increased disease-related death in a competing risk analysis. Submitted 26 September 2025; accepted 18 February 2026; prepublished online onBlood Advances First Edition 19 March 2026. *I.D. and C.J. contributed equally to this study. Data are available from the corresponding author, Simone Ferrero (), on request. 14 JULY 2026 • VOLUME 10, NUMBER 13 The full-text version of this article contains a data supplement. © 2026 American Society of Hematology. Published by Elsevier Inc. Licensed under, permitting only noncommercial, nonderivative use with attribution. All other rights reserved. 4519 ### Page 2 Introduction Waldenström macroglobulinemia (WM) is a rare, indolent B-celllymphoproliferative disorder that is characterized by bone marrow(BM) involvement through lymphoplasmacytic cells that secretemonoclonal immunoglobulin M (IgM) protein.1,2 Although a familial predisposition has been demonstrated,3 themain risk factor for WM remains a history of IgM monoclonalgammopathy of undetermined significance (IgM-MGUS)4 with aprobability of progression to WM or other lymphoproliferativedisorders of 1.5% to 2% per year.4 Important advances in the understanding of the biology of WMhave been made in the last few years. By using whole-genomesequencing, Treon et al identified MYD88 L265P as a highlyrecurrent somatic mutation in 90% of patients with WM.5 Severalstudies that used different techniques, such as Sangersequencing, polymerase chain reaction (PCR), and allele-specific,quantitative PCR (ASqPCR),6-8 confirmed the presence of theMYD88 L265P mutation in WM. Moreover, it was also demonstrated in BM samples in 50% to 80% of patients with IgMMGUS,7-10 and it can also be identified in peripheral blood (PB),11specifically in the cell-free DNA fraction, of patients with IgMgammopathy.12,13 From a clinical standpoint, WM typically shows an indolent clinicalcourse with ~25% of patients being asymptomatic at initial diagnosis, despite a heterogeneous disease behavior. Indeed, amongpatients with smoldering WM, there are currently no indications forthe initiation of treatment until the development of significantsymptoms.14-16 However, these patients have a high risk fortransformation from smoldering WM to active WM and should bemonitored closely.17 Over the years, many attempts have been made to predict theoutcomes of patients with IgM gammopathy based on clinicalassessments by applying different criteria for WM or IgM-MGUSdiagnosis and treatment initiation., For patients with activeWM who require treatment, the International Prognostic ScoringSystem for WM (IPSS-WM) was designed to assess survival aftertreatment initiation and overall survival (OS)24 by stratifyingpatients into 3 risk categories with corresponding 5-year survivalrates of 87%, 68%, and 36%, respectively. Later on, further revisions of the IPSS-WM refined prognostication in active WM,25,26whereas for patients with asymptomatic WM (aWM), Bustoroset al developed a prognostic score for disease progression thatrelies on BM biopsy (BMB) evaluation and basic clinical findings.27 Some authors have also tried to assess the impact of laboratoryprognostic factors, such as multiparameter flow cytometry (MFC)features,28 cytogenetics, and molecular biology characterization,on patient outcomes. Even in the absence of disease-definingcytogenetic abnormalities, deletions in the chromosome 6q arepresent in 40% to 50% of WM cases and have been shown tonegatively affect patient prognosis29,30 by being associated with ahigher IPSS-WM score and a shorter time to treatment,progression-free survival (PFS), and OS.31-34 The impact of molecular factors in determining the prognosis forpatients with WM is still not clear. Some studies have suggestedthat the MYD88 L265P mutation may have an impact on OS and play a role in disease progression by promoting the transition fromIgM-MGUS to WM or other lymphoproliferative diseases.,,, Incontrast, other studies showed that the MYD88 L265P mutationhad no effect on OS or time to progression14 and that patients withMYD88WT had higher progression rates to symptomatic WM(sWM) and a more aggressive clinical course., Moreover,CXCR4WHIM mutations, the second most frequent genetic alteration in WM, have been proven to affect the response to therapyand PFS,39,40 without affecting OS.33,37 Finally, similar to manyother neoplasms, TP53 alterations confer worse outcome in termsof OS,40-42 irrespectively of the IPSS-WM score; of note, suchalterations were not identified in patients who had IgM-MGUS.41,43 Aims This study aimed to (1) collect a large series of patients with IgMgammopathy who were treated in real-life settings; (2) determinepatient outcomes in terms of OS, time to first treatment (TTFT),and PFS; (3) identify clinical factors that affect disease progression and survival; and (4) evaluate the impact of baseline molecularand flow cytometry analyses on survival and the need for WMtreatment. Materials and methods Patient selection Spanish cohort. The registry of monoclonal gammopathiesbased in Salamanca and in the region of Castilla y Leon (Spain)was investigated for IgM-secreting disorders (); onlypatients with confirmed IgM-MGUS, aWM, or sWM1 wereselected for this study. All patients provided written informedconsent in accordance with the Helsinki declaration. Torino cohort. Electronic health records of patients who wereobserved or treated at the Centre of Torino (Division of Haematology 1, Department of Biotechnology and Health Sciences,University Hospital Citta ` della Salute e della Scienza, University ofTorino, Italy) were scanned to identify patients with IgM monoclonal gammopathy; patients with a confirmed diagnosis of IgMMGUS, aWM, or sWM were selected by the study team usingthe same criteria employed for the Spanish cohort (). Allpatients provided written informed consent in accordance with theHelsinki declaration. Sample collection and laboratory analysis BM and PB samples were extracted from patients at first diagnosisof IgM gammopathy or at disease progression, according to locallyestablished procedures, for both study cohorts. A BMB was performed per clinical practice, mainly when the criteria for treatmentinitiation or suspicion of progressive/transformed disease were met. Genomic DNA was extracted for molecular studies from the evaluated patients, as previously described, and analyzed for the presenceof the MYD88 L265P mutation.44 Alternatively, MYD88 mutationalstatus was determined by droplet digital PCR (ddPCR) on both BMand PB samples, as previously described.12 All patient samplestested in Torino were analyzed by ddPCR. Of note, in neither ofthe cohorts was CD19+ cell selection routinely applied beforethe molecular study, but it was performed for selected cases in theSpanish cohort before fluorescence in situ hybridization analysis. 4520 DOGLIOTTI et al 14 JULY 2026 • VOLUME 10, NUMBER 13 ![Figure]() ### Page 3 Deletions of 6q were assessed in patients with IgM-MGUS andWM by either simple interphase fluorescence in situ hybridizationperformed on cell nuclei from whole BM samples or CD19selected cells using a previously published technique.31 Immunophenotypic evaluation was done using conventional panelsof monoclonal antibodies, as previously described,45 and followingthe general recommendations of the EuroFlow group for theimmunophenotypic evaluation of hematologic malignancies. Thesensitivity of MFC assay was 0.004% for at least 500 000 events. Statistical analyses Statistical analyses were carried out using R (version 4.3.1). Survival curves were plotted using the Kaplan-Meier method andcompared with the log-rank test. The medians between groups forcontinuous variables were compared using the Kruskal-Wallis test(for nonnormal variables) or the 1-way analysis of variance test (fornormal variables); the χ2 test or Fisher exact test for small studysamples was employed for categorical variables. OS wasmeasured from the date of the initial diagnosis to the date of deathfrom any cause. For asymptomatic cases, TTFT was defined as thetime between diagnosis and progression to sWM that requiredactive treatment. PFS was defined as the time between first-lineWM treatment and the date of progression or death from anycause. The Cox proportional hazards model was implemented forthe univariate and multivariate survival analyses. In particular, forthe multivariable model, an Akaike Information Criterion-basedbackward stepwise algorithm (R function stats::step) was used toperform the variable selection (restricting the data set to have nomissing data) to determine the most relevant covariates andanalyze potential confounding factors. Once the covariates ofinterest were obtained from the restricted data set, they wereapplied to the full data set for each endpoint. Given the larger size of the Spanish data set, the statistical modelswere constructed using these data. Subsequently, the Torino dataset was merged to evaluate whether there were differencesbetween the 2 populations. An analysis to evaluate the IPSS-WMand the revised IPSS-WM scores was also performed on bothcohorts (Spain, Torino). Moreover, the aWM score was evaluatedin patients with aWM in the Torino series. In addition, we worked out a cutoff for quantitative MYD88 andMFC infiltration in BM; the cutoff points were identified using the surv_cutpoint function in R on the Spanish data set, and this wasextended to the Torino cohort analysis. Finally, a competing risk analysis was performed for OS using thecuminc function of the tidycmprsk package of R. Institutional review board approval Institutional review board approval was obtained from the Salamanca University Hospital Ethic Committee, Paseo de San Vicente, 58-182 37007 Salamanca, Spain; Ethical Commitee numberPI91/07/2018 (date of approval: 13 August 2018) and 21/742(date of approval: 3 November 2021). Furthermore, the ethicCommittee of the Azienda Ospedaliera Universitaria Citta ` delleSalute e della Scienza di Torino - A.O. Ordine Mauriziano di Torino - A.S.L. Citta ` di Torino, Corso Bramante, 88/90 - 10126 Turin,Italy; Ethical Commitee number 476/2021 (date of approval 25November 2021) also provided institutional review board approval. Results Patient characteristics Data from 903 patients across 22 Spanish centers were collectedbetween 1976 and 2019 in the considered registry; 577 caseswith a local diagnosis of IgM-MGUS, aWM, or sWM wereselected, whereas the remaining cases were excluded because ofa diagnosis of other lymphoproliferative diseases, misregistration,or missing data (). Among patients observed in Torino,Italy, 166 patients with IgM gammopathy with available baselineand follow-up data were selected according to the same criteria. Itshould be noted that these patients were diagnosed from 1988 to2020, although the vast majority (161/166) were diagnosed afterthe year 2000. The demographic and clinical data at initial diagnosis are reportedin and (Spanish and Torino cohorts). When they were first evaluated, 111 Spanish patients were locallydiagnosed with IgM-MGUS, 245 with aWM, and 221 with sWM.The IPSS-WM score at diagnosis for patients with sWM was lowin 30 patients (18.2%), intermediate in 68 (41.2%), and high in 67patients (40.6%) in the Spanish cohort, and low in 10 patients(20.0%), intermediate in 18 (36.0%), and high in 22 patients(44.0%) in the Torino cohort (P = .804 for the distribution in the 2cohorts). ![Figure]() ![Figure]() 14 JULY 2026 • VOLUME 10, NUMBER 13 LABORATORY PROGNOSTIC TOOLS IN WM AND IgM-MGUS 4521 ### Page 4 | IgM MGUS (n = 111) | Asymptomatic WM (n = 245) | Symptomatic WM (n = 221) | P value | | - | - | - | - | | Age (median [IQR]) | 69.5 [29.0, 87.0] | 72.0 [34.0, 93.0] | 71.0 [30.0, 94.0] | 0.070 | | Sex | | Female | 46 (41.4%) | 88 (35.9%) | 67 (30.3%) | 0.120 | | Male | 65 (58.6%) | 157 (64.1%) | 154 (69.7%) | | Hb (median [IQR]), g/dL | 13.9 [8.1, 17.4] | 13.2 [6.0, 20.0] | 10.4 [3.6, 18.0] | <0.001 | | ECOG | | 0 | 68 (61.3%) | 136 (55.5%) | 46 (20.8%) | <0.001 | | 1 | 19 (17.1%) | 55 (22.4%) | 74 (33.5%) | | 2 | 3 (2.7%) | 12 (4.9%) | 48 (21.7%) | | 3 | 00 (0%) | 3 (1.2%) | 10 (4.5%) | | 4 | 1 (0.9%) | 2 (0.8%) | 2 (0.9%) | | Missing | 20 (18.0%) | 37 (15.1%) | 41 (18.6%) | | B symptoms | | No | 92 (82.9%) | 203 (82.9%) | 117 (52.9%) | <0.001 | | Yes | 1 (0.9%) | 7 (2.9%) | 73 (33.0%) | | Missing | 18 (16.2%) | 35 (14.3%) | 31 (14.0%) | | Platelets (median [IQR]), ×109/L | 227 [73, 1970] | 248 [30, 644] | 210 [2, 684] | <0.001 | | B2M (median [IQR]), mg/dL | 2.00 [0.5, 6.4] | 2.46 [0.15, 18.0] | 3.26 [0.4, 18.6] | <0.001 | | Missing | 21 (18.9%) | 55 (22.4%) | 45 (20.4%) | | Albumin (median [IQR]), g/dL | 4.10 [2.9, 5.0] | 3.90 [2.3, 5.2] | 3.60 [1.6, 4.9] | <0.001 | | Missing | 20 (18.0%) | 38 (15.5%) | 32 (14.5%) | | IgM (median [IQR]), mg/dL | 781 [87, 6130] | 1460 [225, 9220] | 3300 [19.0, 13 000] | <0.001 | | Missing | 9 (8.1%) | 23 (9.4%) | 22 (10.0%) | | LDH ratio (median [IQR]), UI/L | 0.73 [0.27, 1.8] | 0.65 [0.28, 1.76] | 0.65 [0.25, 4.0] | 0.043 | | Missing | 18 (16.2%) | 35 (14.3%) | 35 (15.8%) | | Light Ig type | | k | 69 (62.2%) | 160 (65.3%) | 157 (71.0%) | 0.039 | | l | 37 (33.3%) | 62 (25.3%) | 38 (17.2%) | | Negative | 00 (0%) | 2 (0.8%) | 1 (0.5%) | | Missing | 5 (4.5%) | 21 (8.6%) | 25 (11.3%) | | IPSS-WM | | High | 00 (0%) | 13 (5.3%) | 67 (30.3%) | <0.001 | | Int | 45 (40.5%) | 126 (51.4%) | 68 (30.8%) | | Low | 30 (27.0%) | 45 (18.4%) | 30 (13.6%) | | Missing | 36 (32.4%) | 61 (24.9%) | 56 (25.3%) | | FISH | | Abnormal | 4 (3.6%) | 20 (8.1%) | 34 (15.5%) | 0.001 | | Normal | 26 (23.4%) | 93 (38.0%) | 65 (29.5%) | | Not evaluable | 81 (73%) | 132 (53.9%) | 121 (54.1%) | | MYD88 L265P mutation | | Negative | 20 (18.0%) | 25 (10.2%) | 26 (11.8%) | <0.001 | | Positive | 18 (16.2%) | 111 (45.3%) | 98 (44.3%) | | Not evaluable | 73 (65.8%) | 109 (44.5%) | 97 (43.9%) | | MYD88 L265P mutational ratio (median [IQR]) | 0.02 [0.001, 0.0868] | 0.0165 [0.00131, 1.76] | 0.0362 [0.00158, 1.36] | 0.004 | ECOG, Eastern Cooperative Oncology Group; FISH, fluorescence in situ hybridization; Ig, immunoglobulin; Int, intermediate; k, free light kappa chain; l, free light lambda chain; LDH, lactatedehydrogenase. 4522 DOGLIOTTI et al 14 JULY 2026 • VOLUME 10, NUMBER 13 ![Figure]() ### Page 5 | IgM MGUS (n = 22) | Asymptomatic WM (n = 71) | Symptomatic WM (n = 73) | P value | | - | - | - | - | | Age (median [IQR]) | 73.0 [48.0, 91.0] | 67.0 [24.0, 84.0] | 70.0 [43.0, 95.0] | 0.091 | | Sex | | Female | 8 (36.4%) | 33 (46.5%) | 24 (32.9%) | 0.237 | | Male | 14 (63.6%) | 38 (53.5%) | 49 (67.1%) | | Hb (median [IQR]), g/dL | 10.4 [3.60, 18.0] | 14.0 [8.30, 16.0] | 13.0 [0, 18.0] | <0.001 | | ECOG | | 0 | 16 (72.7%) | 59 (83.1%) | 30 (41.1%) | <0.001 | | 1 | 4 (18.2%) | 8 (11.3%) | 15 (20.5%) | | 2 | 1 (4.5%) | 1 (1.4%) | 10 (13.7%) | | 3 | 00 (0%) | 00 (0%) | 7 (9.6%) | | 4 | 00 (0%) | 00 (0%) | 1 (1.4%) | | Missing | 1 (4.5%) | 3 (4.2%) | 10 (13.7%) | | BMB% (median [IQR]) | 0 [0, 0] | 30.0 [10.0, 90.0] | 70.0 [10.0, 95.0] | <0.001 | | Missing | 15 (68.2%) | 8 (11.3%) | 10 (13.7%) | | B symptoms | | No | 20 (90.9%) | 69 (97.2%) | 53 (72.6%) | 0.002 | | Yes | 1 (4.5%) | 00 (0%) | 10 (13.7%) | | Missing | 1 (4.5%) | 2 (2.8%) | 10 (13.7%) | | Platelets (median [IQR]), ×109/L | 209 [114, 315] | 260 [71.0, 632] | 203 [29.0, 543] | 0.013 | | B2M (median [IQR]), mg/dL | 2.52 [1.4, 6.6] | 1.95 [1.1, 13.2] | 3.10 [1.5, 12.6] | 0.002 | | Missing | 10 (45.5%) | 23 (32.4%) | 42 (57.5%) | | Albumin (median [IQR]), g/dL | 3.80 [3.0, 5.1] | 4.00 [3.0, 5.3] | 3.75 [2.5, 4.8] | 0.012 | | Missing | 11 (50.0%) | 34 (47.9%) | 35 (47.9%) | | IgM (median [IQR]), mg/dL | 781 [195, 5360] | 1370 [210, 5480] | 2960 [205, 9620] | <0.001 | | Missing | 00 (0%) | 5 (7.0%) | 11 (15.1%) | | LDH (median [IQR]), UI/L | 0.8 [0.3, 1.8] | 0.8 [0.3, 1.4] | 0.7 [0.4, 2.4] | 0.554 | | Missing | 8 (36.4%) | 25 (35.2%) | 22 (30.1%) | | Light Ig type | | k | 15 (68.2%) | 50 (70.4%) | 52 (71.2%) | 0.982 | | l | 4 (18.2%) | 15 (21.1%) | 15 (20.5%) | | Missing | 3 (13.6%) | 6 (8.5%) | 6 (8.2%) | | IPSS WM | | High | 00 (0%) | 4 (5.6%) | 22 (30.1%) | <0.001 | | Int | 15 (68.2%) | 30 (42.3%) | 18 (24.7%) | | Low | 6 (27.3%) | 23 (32.4%) | 10 (13.7%) | | Missing | 1 (4.5%) | 14 (19.7%) | 23 (31.5%) | | FISH | | Abnormal | 00 (0%) | 2 (2.8%) | 2 (2.9%) | 0.193 | | Normal | 00 (0%) | 2 (2.8%) | 00 (0%) | | Not evaluable | 22 (100%) | 67 (94.4%) | 68 (97.1%) | | MYD88 L265P mutation | | Negative | 3 (13.6%) | 5 (7.0%) | 10 (13.7%) | 0.376 | | Positive | 11 (50.0%) | 50 (70.4%) | 53 (72.6%) | | Missing | 8 (36.4%) | 16 (22.5%) | 10 (13.7%) | | MYD88 L265P mutational ratio (median [IQR]) | 0.00497 [0.000551, 0.0371] | 0.0216 [0.000520, 0.283] | 0.0278 [0.000400, 0.713] | 0.003 | ECOG, Eastern Cooperative Oncology Group; FISH, fluorescence in situ hybridization; Ig, immunoglobulin; Int, intermediate; k, free light kappa chain; l, free light lambda chain; LDH, lactatedehydrogenase. ![Figure]() 14 JULY 2026 • VOLUME 10, NUMBER 13 LABORATORY PROGNOSTIC TOOLS IN WM AND IgM-MGUS 4523 ### Page 6 ![Figure]() Treatments In the Spanish retrospective series that comprised patientsobserved from 1976 to 2019, 78 of 577 (13.5%) patients hadmissing first-line treatment details and were excluded from theanalysis of treatments received. Among 499 patients with availabledetails of treatments received, 321 (64.3%) (diagnosed with IgMMGUS or aWM) were observed in a watch and wait (W&W)approach, whereas 178 of 499 (35.7%) patients with WM receivedinitial therapy (supplemental Table 1). At last follow-up, 208 of 321(64.8%) patients remained asymptomatic, whereas 41 of 321(12.8%) had progressed to WM and required active treatment;follow-up data were missing for 72 patients. Most employed first-line treatments included chlorambucil-basedregimens (n = 98/178, 55%), dexamethasone-rituximabcyclophosphamide scheme (n = 19/178; 10.7%), andfludarabine-containing regimens (n = 6/178; 3.4%). Overall, 51 of 178 patients (28.6%) received rituximab as first-line treatment(single agent or any combination). In the Torino cohort, 5 of 166 patients were excluded from theanalysis of treatment received for missing information; 60 of 161patients (37.3%) remained asymptomatic and did not receivetherapy during follow-up. First-line treatments used at any timeincluded dexamethasone-rituximab-cyclophosphamide (n = 27/73;37%), rituximab-bendamustine (n = 8/73; 11%), and chlorambucilbased schemes (n = 6/73; 8.2%). Rituximab was administered asfirst-line treatment to 69.8% of treated patients. Survival analyses OS. In the Spanish cohort, at last follow-up, 305 (52.9%) patientswere alive at a median follow-up of 100.6 months (interquartilerange [IQR], 55.5-166.6). The median OS was 126.7 months(IQR, 58.7-193.6). The OS at 5 years was 74.0% (95% 4524 DOGLIOTTI et al 14 JULY 2026 • VOLUME 10, NUMBER 13 ![Figure]() ### Page 7 confidence interval [CI], 70.1-78.2), and at 10 years, the OS was52.5% (95% CI, 47.3%-58.2%;A). The median OS forpatients with IgM-MGUS vs aWM vs sWM was 180.6 (IQR,111.3-not reached) vs 143.6 (80.9-199.7) vs 85.0 (38.8-151.6)months (P < .001, hazard ratio [HR] for death in sWM group,2.481) in the univariate analysis (C). In the Torino cohort, at last follow-up, 132 (79.5%) patients werealive at a median follow-up of 71.2 (IQR, 39.6-115.4) months. Themedian OS was 202.8 months (IQR, 109.0-not reached); the OS at 5 years was 85.7% (95% CI, 80.0-91.8) and 71.6% (61.7-83.1) at10 years from initial diagnosis (B). The median OS stratifiedfor initial diagnosis was 131.0 (131.0-not reached) months for IgMMGUS, not-reached (131.4-not reached) for aWM, 174.4 (65.9-notreached) months for sWM (P = .084;). TTFT. The TTFT was evaluated for all asymptomatic patients atbaseline, including those with IgM-MGUS and aWM. For the Spanish cohort, the median TTFT was 228.5 months (IQR, 146.9not reached) with 89.1% of patients not requiring treatment after 5years of follow-up (95% CI, 85.5-92.9). For patients observed inTorino, the median TTFT was 165.6 months (IQR, 49.4-207.9),and the probability to remain asymptomatic was 71.4% at 5 years(95% CI, 60.9-83.7) (). PFS. PFS was evaluated for patients who received treatment forsWM at any time. Among patients with available data, the medianPFS was 42.4 months (IQR, 13.6-91.0) for the Spanish cohort and38.0 months (IQR, 13.9-75.6) for the Torino cohort (,bottom panels). At 2 years, the PFS probability was 64.6%(95% CI, 58.8-70.9) and 63.3% (54.2-73.8) for the Spanish andTorino cohorts, respectively. Importantly, there was no difference inthe PFS outcome between patients who received active treatmentat diagnosis and patients who initially underwent the W&Wapproach and then progressed to WM (P = .34 and P = .24 in the 2 cohorts). ![Figure]() ![Figure]() 14 JULY 2026 • VOLUME 10, NUMBER 13 LABORATORY PROGNOSTIC TOOLS IN WM AND IgM-MGUS 4525 ### Page 8 | OS | PFS | TTFT | | - | - | - | | Variables | HR | 95% CI | P value | HR | 95% CI | P value | HR | 95% CI | P value | ![Figure]() Age ![Figure]() | ≤65 | — | — | — | — | — | — | | - | - | - | - | - | - | - | | 65-75 | 2.221 | 1.514-3.260 | <.001 | 1.389 | 0.931-2.072 | .107 | 1.002 | 0.483-2.083 | .996 | | ≥76 | 4.136 | 2.812-6.085 | <.001 | 2.561 | 1.675-3.918 | <.001 | 0.238 | 0.075-0.757 | .015 | | Sex | | Female | — | — | — | — | | Male | 1.651 | 1.192-2.287 | .003 | 2.115 | 0.971-4.606 | .059 | | Hb | | ≤11.5 | - | - | | >11.5 | 0.353 | 0.136-0.918 | .033 | | MGUS | — | — | | aWM | 1.172 | 0.722-1.902 | .521 | | sWM | 2.086 | 1.277-3.407 | .003 | | BMB | | Not evaluable | — | — | — | — | | Negative | 0.937 | 0.423-2.074 | .873 | 1.498 | 0.523-4.291 | .452 | | Positive | 1.332 | 0.950-1.869 | .096 | 2.902 | 1.337-6.298 | .007 | | Platelets | | ≤100 | — | — | — | — | — | | >100 | 1.567 | 0.950-1.869 | .091 | 0.096 | 0.027-0.345 | <.001 | | B2M | | ≤3 | — | — | — | — | | >3 | 1.617 | 1.200-2.180 | .002 | 1.262 | 0.905-1.761 | .171 | | Albumin | | <3.5 | — | — | — | | ≥3.5 | 0.449 | 0,200-1.010 | .053 | | LDH | | ≤1 | — | — | | >1 | 0.533 | 0.326-0.872 | .012 | | Hyperviscosity | | No | — | — | | Yes | 1.407 | 0.943-2.098 | .094 | | Torino cohort | | Age | | ≤65 | — | — | — | — | — | — | | 65-75 | 2.221 | 1.514-3.260 | <.001 | 1.389 | 0.931-2.072 | .107 | 1.002 | 0.483-2.083 | .996 | | ≥76 | 4.136 | 2.812-6.085 | <.001 | 2.561 | 1.675-3.918 | <.001 | 0.238 | 0.075-0.757 | .015 | | Sex | | Female | — | — | — | — | | Male | 1.651 | 1.192-2.287 | .003 | 2.115 | 0.971-4.606 | .059 | | Hb | | ≤11.5 | — | — | | >11.5 | 0.353 | 0.136-0.918 | .033 | Variables with a statistically significant effect in the multivariable models are indicated in bold. 4526 DOGLIOTTI et al 14 JULY 2026 • VOLUME 10, NUMBER 13 ![Figure]() ### Page 9 | OS | PFS | TTFT | | - | - | - | | Variables | HR | 95% CI | P value | HR | 95% CI | P value | HR | 95% CI | P value | | MGUS | — | — | | aWM | 1.172 | 0.722-1.902 | .521 | | sWM | 2.086 | 1.277-3.407 | .003 | | BMB | | Not evaluable | — | — | — | — | | Negative | 0.937 | 0.423-2.074 | .873 | 1.498 | 0.523-4.291 | .452 | | Positive | 1.332 | 0.950-1.869 | .096 | 2.902 | 1.337-6.298 | .007 | | Platelets | | ≤100 | — | — | — | — | — | | >100 | 1.567 | 0.950-1.869 | .091 | 0.096 | 0.027-0.345 | <.001 | | B2M | | ≤3 | — | — | — | — | | >3 | 1.617 | 1.200-2.180 | .002 | 1.262 | 0.905-1.761 | .171 | | Albumin | | <3.5 | — | — | — | | ≥3.5 | 0.449 | 0.200-1.010 | .053 | | LDH | | ≤1 | — | — | | >1 | 0.533 | 0.326-0.872 | .012 | | Hyperviscosity | | No | — | — | | Yes | 1.407 | 0.943-2.098 | .094 | Variables with a statistically significant effect in the multivariable models are indicated in bold. Finally, 22 of 577 (4.8%) and 5 of 166 (3.1%) patients in the 2examined cohorts experienced transformation from IgM gammopathy to aggressive lymphoma during follow-up and showed verypoor survival (supplemental Figure 1). No significant difference inthe incidence of transformation in the 2 groups was found (P = .51). Clinical prognostic factors The clinical data at baseline, including age, sex, IgM levels,hemoglobin (Hb), platelets, Eastern Cooperative Oncology GroupPerformance Status, the presence of hyperviscosity, BMB infiltration, albumin, β-2-microglobulin (B2M), lactate dehydrogenase,MYD88 L265P mutational ratio, and MFC infiltration in BM, wereevaluated for prognostic significance in terms of OS, TTFT (forasymptomatic patients), and PFS (for patients treated at any time).The significant variables for each cohort are summarized in. A multivariable model for OS, PFS, and TTFT (summarized in) was built, as previously described, for the Spanishcohort. This model was then used to evaluate a merged population (Spain + Torino) to identify differences between the 2 cohorts. Because a P value of > .05 was found, we mayconclude that there was no evidence of differences betweenthese groups. For the OS prediction, the algorithm identified age >65 years,male sex, initial diagnosis of sWM, and B2M >3 as statisticallysignificant variables. In contrast, the multivariable analysis for TTFT was built for asymptomatic patients (IgM-MGUS, aWM) and produced a model with age, BMB infiltration, Hb <11.5 g/dL, andplatelets <100 000/μL as significant variables. Finally, a multivariable analysis for PFS in patients with sWM (sWM at diagnosis +patients progressing from W&W approach to sWM) was performed, and age and lactate dehydrogenase ratio were identifiedas statistically significant. In addition, we evaluated the IPSS-WM and revised IPSS-WMscores in Spanish patients with WM and found that they bothretained statistical significance (P < .001 for both scores;supplemental Figure 2), also confirmed in the multivariable analysis. In contrast, the aWM score could not be evaluated in patientswith aWM from the Torino series because of the low number ofpatients with available complete data (n = 29). Prognostic significance of molecular and flowcytometry factors For patients with available BM MYD88 L265P mutation evaluationdata at baseline (n = 298; 51.6% for the Spanish cohort; andn = 142; 85.5% for Torino cohort), we analyzed the outcome in termsof OS and TTFT using both quantitative PCR methods results. Patients from the Spanish cohort and Torino cohorts were pooledand divided into high vs low MYD88 groups (cutoff point set at0.162, defined as previously described). ![Figure]() 14 JULY 2026 • VOLUME 10, NUMBER 13 LABORATORY PROGNOSTIC TOOLS IN WM AND IgM-MGUS 4527 ### Page 10 OS PFS TTFT HR 95% CI P value HR 95% CI P value HR 95% CI P value ![Figure]() MYD88 ratio: cutoff >0.1624318 ![Figure]() | Neg | — | — | — | — | — | — | | - | - | - | - | - | - | - | | Low | 0.995 | .982 | 0.789 | 0.333 | 0.864 | .754 | | High | 1.823 | .061 | 1.013 | 0.969 | 3.478 | .049 | | Flow ratio: cutoff >4.39% | | Neg | — | — | — | — | — | — | | Low | 0.912 | .792 | <0.001 | 0.995 | 3.215 | .268 | | High | 1.421 | .279 | <0.001 | 0.995 | 7.196 | .056 | | Torino cohort | | MYD88 ratio: cutoff >0.624318 | | Neg | — | — | — | — | — | — | | Low | 1.153 | .822 | 0.811 | 0.575 | 0.724 | .670 | | High | 3.063 | .16 | 1.430 | 0.434 | 2.874 | .395 | | Flow ratio: cutoff >4.39% | | Neg | — | — | — | — | — | — | | Low | <0.001 | .998 | <0.001 | 0.997 | 0.35 | .396 | | High | <0.001 | .998 | <0.001 | 0.997 | 1.48 | .621 | Variables with a statistically significant effect in the multivariable models are indicated in bold. In the univariate analysis, the MYD88low group showed better OS(P = .005; HR, 0.44) and TTFT (P = .024; HR, 0.33) than theMYD88high group (). In the multivariable analysis, statisticalsignificance was not confirmed (P = .622) for OS, although therewas a trend toward significance for TTFT (P = .06). Regarding the prognostic impact of MYD88 on OS specifically inpatients with IgM-MGUS, data were available for 37 patientsclassified as MYD88 low (n = 17) or MYD88WT (n = 20), whereasno patient was classified as MYD88 high; no difference in OS wasobserved in the univariate analysis between these 2 subgroups(P = .2; supplemental Figure 6). In addition, the molecular burden was compared with the percentage of marrow infiltration in the BMB (available only for theTorino cohort), which showed a moderate positive Pearson correlation (r = 0.653; P < .0001). The infiltration reported by BMMFC also showed a strong positive linear correlation (r = 0.73;P < .0001). Patients in the MYD88high group generally showedhigher BMB infiltration (~70%-90%), although rare outliers withlow marrow infiltration (as low as 20%) were also observed, indicating possible sampling errors and biologic heterogeneity. The pooled population was also analyzed for the prognostic significance of baseline clonal B-cell infiltration by MFC in availableBM samples with a cutoff point set at 4.39%. In the univariableanalysis, MFClow patients had increased OS (P = .033; HR, 0.65)and TTFT (P = .008; HR, 0.37) when compared with the MFChighgroup (). In the multivariable analysis, MFC lost statisticalsignificance for OS (P = .51); however, it retained significance forTTFT (P = .025; HR, 0.37). Similarly, for patients with IgM-MGUS,the MFC burden affected the TTFT (P = .006; supplementalFigure 7) but did not show an impact on OS (P = .5). In addition, we combined both the MFC and MYD88 baselinelevels and stratified patients into low-, intermediate-, and high-riskgroups (low: MFClow/MYD88low; intermediate: either MFClow/MYD88high or MFChigh/MYD88low; high: MFChigh/MYD88high).Based on MFC and the molecular results, no patients with IgMMGUS was assigned to the HR group; 16 HR patients hadsWM, whereas 8 had aWM. Moreover, among the patients withsWM with high IPSS-WM risk, 7 patients were in the low-riskgroup, 18 patients were in the intermediate group, and 6 were inthe high-risk group. In the univariate analysis, the high-risk group differed significantlyfrom the others in terms of OS (n = 156; P = .005; HR, 3.28;) and TTFT (n = 92; P = .003; HR, 9.61). For TTFT,the intermediate-risk group (P = .015; HR, 4.76) was alsosignificantly different from the reference group (low;supplemental Figure 3). A multivariable analysis confirmed these results, highlighting thestrong prognostic significance of the combination of molecular andMFC evaluations (for OS: P = .038; HR, 2.81; for TTFT: P = .026;HR, 7.91 for intermediate vs low risk and P = .003; HR, 24.68 forhigh vs low risk). Finally, a competing risk analysis was performed. Among the 545evaluable cases, at 5 years from diagnosis, 12.0% (95% CI, 9.6%-16%) of deaths were not related to WM, whereas 7.1% (95% CI,5.0%-9.6%) were considered related (supplemental Figure 4).Among patients in the MYD88/MFC high-risk group, there was astatistically significant increase in disease-related, but not unrelated, deaths at 5 years (P = .002; 2.2%; 95% CI, 0.17%-10% vs19%; 95% CI, 5.6%-38% for LR vs HR; supplemental Figure 5). 4528 DOGLIOTTI et al 14 JULY 2026 • VOLUME 10, NUMBER 13 ![Figure]() ### Page 11 ![Figure]() Discussion Previous real-life studies that reported on the survival outcomes ofpatients with WM, for instance, the Rory Morrison Registry in theUnited Kingdom and the Swedish Lymphoma Registry, showed a5-year OS that ranged from 60% to 90%, both with a medianfollow-up of ~6 years.20,46 In our series that also included aWMand IgM-MGUS, the 5-years OS was comparable (85% for theTorino cohort, 74% for the Spanish cohort), after a median followup of 71 and 100 months, respectively. It should be noted thatthe slightly superior OS outcome of the Torino cohort might beexplained by the more recent IgM gammopathy diagnoses and themore extensive adoption of rituximab-containing regimens in firstand subsequent lines of treatment. However, in published registry studies, molecular evaluations andMFC data were limited or not available, and thus, prognosticevaluations relied solely on clinical factors. Indeed, prognosticationfor sWM was historically established based on the IPSS-WMscore and its revised version24,25 that stratified patients basedon age and baseline characteristics. In this study, both the IPSS-WM and revised IPSS-WM wereconfirmed to correctly identify the risk of patients with sWM,whereas for those with aWM, the score proposed by Bustoroset al was not reproducible, probably because of the low number ofpatients with data available on the percentage of invasion inBMBs.27 Moreover, analyzing the whole IgM gammopathy series(IgM-MGUS, aWM, and sWM taken together), clinical factorswere confirmed to be important for both OS (age, sex, B2M, anddiagnosis) and TTFT (age, Hb, platelet, and BMB involvement). Indeed, the IPSS-WM was initially developed at a time (before2002) when nucleoside analogs represented around one-third ofthe first-line therapies, whereas rituximab was only administered to4% of patients. Nevertheless, it was still confirmed to be a feasibleand powerful tool. However, in the rapidly changing field of WM inwhich both an understanding of its biology and its treatment wererevolutionized by the discovery of the MYD88 L265P mutation, theneed to incorporate translational knowledge into prognosticassessment is still unmet. Indeed, although MYD88 L265Pmutation testing is already considered part of standard initialevaluation in suspected WM and is particularly helpful in the differential diagnosis of other lymphoproliferative neoplasms, its rolein prognostication is still debated. Indeed, previous findings have highlighted that patients withMYD88WT WM have inferior survival outcomes36 and show slowerand less profound responses with Bruton tyrosine kinase inhibitors, especially with ibrutinib.40 However, a large cohort study byAbeykoon et al did not show inferior OS or time to next treatmentin patients with MYD88WT vs those with MYD88 L265P WM,whereas there was an increased occurrence of transformation todiffuse large B-cell lymphoma.14 For patients with IgM-MGUS, a study by Varettoni et al showedlower progression rates to overt WM in MYD88WT cases than incases with MYD88 L265P.9 In the same study, the MYD88 L265Pallele burden, as determined by ASqPCR on BM CD19+ selectedmononuclear cells, showed a significant correlation with progression from MGUS to WM, albeit with a mild effect. These findingswere later confirmed in a recent study by Moreno et al using ddPCRin IgM-MGUS and aWM cases.47 Of note, among patients with ![Figure]() 14 JULY 2026 • VOLUME 10, NUMBER 13 LABORATORY PROGNOSTIC TOOLS IN WM AND IgM-MGUS 4529 ### Page 12 IgM-MGUS, the reproducibility of the prognostic effect of molecularfactors might be affected by low disease burden; the use of highlysensitive techniques is therefore key to correctly estimate prognosis. Indeed, past studies have shown heterogeneous resultsbecause of the different PCR techniques, the variable application ofCD19+ sorting, and the adoption of different diagnostic criteria.48 To date, however, there were no published data regarding theeffect of high vs low MYD88 L265P baseline levels on OS andTTFT in patients with WM. In this study, highly sensitive techniquesfor MYD88 quantitative evaluation, either by ddPCR or ASqPCR,together with MFC analysis, allowed for a precise definition ofdisease burden at the time of IgM gammopathy diagnosis. Interestingly, the identified MFC cutoff points (4.39% clonal B-cellinfiltration in BM) that separated high vs low disease burden wasactually quite low, close to the median results in the aWM subgroup (3.55%). In contrast, the MYD88 L265P cutoff point (ratiomutated/WT, 0.162; allele frequency, 13.4%) was higher than themedian MYD88 L265P identified in the sWM group (ratiomutated/WT, 0.04; allele frequency, 3.85%), thus identifyingpatients with very high molecular disease burden. We alsoobserved a strong correlation between the MYD88 L265P resultsand BM MFC infiltration, higher than the one between MYD88L265P and the percentage of infiltration in BMB by conventionalhistopathology observation. From a technical standpoint, the molecular analysis was nothomogeneous between the 2 cohorts. All Torino samples wereanalyzed using ddPCR, which offers higher sensitivity, whereas thevast majority (>95%) of the Spanish samples were assessed byASqPCR, reflecting clinical practice at the time of initial diagnosis.Although both techniques correctly identified high molecular riskpatients, this discrepancy represents a limitation of this retrospective, real-world study and highlights the absence of standardized molecular approaches for this disease. This lack ofharmonization underscores the need for future efforts aimed atmethod standardization to ensure comparability and consistencyacross studies and clinical settings. In the multivariable analysis, the combination of MFC and molecular variables identified a high-risk group that consisted of ~15%of the evaluable patients. Interestingly, a minority of aWM cases(~9%), but no patients with IgM-MGUS, were classified as highrisk in terms of the MYD88/MFC score; thus, it is conceivable thatthis kind of evaluation could help to determine the risk for progression and disease-related death, even in asymptomaticpatients. Indeed, a significant fraction (~22%) of the patients whoscored high in the IPSS-WM was reclassified as low risk by themolecular/MFC score. Conclusion To our knowledge, this is the first time that quantitative MYD88L265P evaluation and MFC degree of infiltration in BM were provento be prognostic of patients’ survival and need for treatment in a large IgM gammopathy series. Moreover, molecular and flowcytometry results have never been combined before to generatetranslational data. Validation of these results in prospective studiesmight add greatly to prognostic evaluation of IgM gammopathy,particularly to improve the definition of patients with high-risk WM. Acknowledgments The authors thank I. Isidro, T. Prieto, A. Anton, ´ R. Maldonado, andM. Hernandez ´ for providing technical assistance. I.D. was supported by Societa ` Italiana di Ematologia and Associazione Italiana contro le Leucemie, Linfomi e Mieloma. C.J. wassupported by Fundacion ´ Espanola˜ de Hematología y Hemoterapia(FEHHJAN23/001H). This work has been partially supported bythe Instituto de Salud Carlos III through the project PI21/00568(cofunded by the European Union) and partially by grants PI18/01866 and PI21/00568. Authorship Contribution: I.D., C.J., D.D., R.G.-S., and S.F. conceptualized thestudy; I.D., V.P., M.F., D.M., S.M., S.R., and G.M.Z. contributed todata curation; I.D., C.J., V.P., and E.A. wrote the original draft; D.M.,S.M., G.M.Z., and M.G. developed the methodology; I.D., V.P.,G.B., E.M.O., A.R., I.M., F.E., C.A., M.A.G., A.G.d.C., R.H., J.D.,F.C., N.P., V.G.d.l.C., S.R., M.C., C.C., E.A., M.d.C.C., A.M.,N.C.G., S.F., and R.G.-S. performed investigation; S.F., B.B.,M.G.D., and R.G.-S. provided supervision; and all authorsreviewed and edited the manuscript. Conflict-of-interest disclosure: S.F. is a consultant for Janssen,EUSA Pharma, AbbVie, and Sandoz; is on the advisory boards ofJanssen, EUSA Pharma, Recordati, Incyte, Roche, AstraZeneca,Italfarmaco, and Behring; reports speaker’s honoraria fromJanssen, EUSA Pharma, Recordati, Lilly, BeiGene, Gilead, andGentili; and research funding from Gilead and MorphoSys. G.B.reports serving on the speaker’s bureau and advisory boards ofJanssen, Bristol Myers Squibb, GSK, and Novartis. The remainingauthors declare no competing financial interests. ORCID profiles: C.J.,; M.F.,; G.M.Z.,; M.G.,; G.B.,; E.M.O.,; A.R.,; I.M.,; F.E.,; M.A.G.,; J.D.,; F.C.,; N.P.,; V.G.d.l.C.,; S.R.,; M.G.D.,; N.C.G.,; B.B.,; D.D.,; S.F.,; R.G.-S.,. Correspondence: Simone Ferrero, Department of MolecularBiotechnologies and Health Sciences, University of Turin, viaGenova 3, 10126 Turin, Italy; email:. References | 1. | . | | - | - | | 4530 | DOGLIOTTI et al | 14 JULY 2026 • VOLUME 10, NUMBER 13 | | ### Page 13 - 2.. - 3.. - 4.. - 5.. - 6. . - 7.. - 8.. - 9.. - 10.. - 11.. - 12.L265P mutation detection by droplet digital polymerase chain reaction in Waldenström. - 13.. - 14.. - 15.. - 16.. - 17.. - 18.. - 19.. - 20.. - 21.. - 22.. - 23.. - 24.. - 25.. - 26.. - 27.. - 28.. - 29.. ![Figure]() 14 JULY 2026 • VOLUME 10, NUMBER 13 LABORATORY PROGNOSTIC TOOLS IN WM AND IgM-MGUS 4531 ### Page 14 - 30.. - 31. . - 32.. - 33.. - 34.. - 35.. - 36.. - 37.. - 38.. - 39.. - 40.. - 41.. - 42.. - 43.. - 44.. - 45.. - 46.. - 47.. - 48.. 4532 DOGLIOTTI et al 14 JULY 2026 • VOLUME 10, NUMBER 13 ![Figure]()