跳到正文
原文
Nature 健康护理主题·· 2026-10-12AI 评分52

骨髓增生在原发性血小板增多症中常见且临床中性

Marrow hypercellularity is common and clinically neutral in essential thrombocythemia

AI 导读

研究发现,38%的ET患者骨髓增生,与预期不符。该研究通过自然语言处理技术从病理报告中提取数据,分析了骨髓细胞ularity的普遍性及其临床意义。

正文

Srisuwananukorn et al. recently reported in Leukemia that artificial intelligence (AI) can distinguish prefibrotic myelofibrosis (preMF) from essential thrombocythemia (ET), highlighting increased marrow cellularity as a potentially discriminatory feature [1]. Normal bone marrow cellularity for age is often invoked as a distinguishing feature of ET, helping to differentiate it from other myeloproliferative neoplasms (MPNs) such as polycythemia vera (PV) and preMF [2]. The International Consensus Classification (ICC) and European LeukemiaNet (ELN) guidance describe normocellularity as typical in ET, with only rare cases of mild hypercellularity [2, 3]. However, normal marrow cellularity is not required for ET diagnosis in the 5th edition World Health Organization (WHO) or ICC criteria [3, 4]. Despite this expectation, hypercellularity is occasionally encountered in patients with otherwise typical WHO-defined ET, prompting possible reconsideration of diagnosis. This paradox motivated us to review the prevalence and clinical significance of marrow cellularity in ET.

To evaluate marrow findings in ET at scale, we employed a natural language processing pipeline (NLP) [5] developed within the Silver MPN Center research data repository (RDR) to extract structured data related to cellularity from all hematopathology reports generated at our center. The diagnosis of ET was established by manual review of all cases according to 5th edition WHO/ICC criteria; cases lacking bone marrow biopsy were excluded. Competing diagnoses, particularly preMF and PV, were excluded by established WHO criteria. Structured query language-based extraction captured clinical, laboratory, pathology, and molecular data from the RDR. Marrow cellularity estimates were available for 338 pathology reports from 273 distinct patients among 497 patients with ET treated at our center (Supplementary Table 1, Supplementary Fig. 1). Age-adjusted hypercellularity was documented in 38% of reports and for 103 of 273 evaluable patients (38%), contrary to expectation. Notably, these estimates were rendered by experienced hematopathologists across multiple decades of institutional practice. This observation prompted systematic evaluation of the prevalence and clinical significance of marrow hypercellularity in ET.

Data availability

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.

References

  1. Srisuwananukorn A, Loscocco GG, Dolezal JM, Kuykendall AT, Santi R, Zhang L, et al. Artificial intelligence differentiates prefibrotic primary myelofibrosis with thrombocytosis from essential thrombocythemia using digitized bone marrow biopsy images. Leukemia. 2026;40:1018–26.

  2. Kvasnicka HM, Orazi A, Thiele J, Barosi G, Bueso-Ramos CE, Vannucchi AM, et al. European LeukemiaNet study on the reproducibility of bone marrow features in masked polycythemia vera and differentiation from essential thrombocythemia. Am J Hematol. 2017;92:1062–7.

    Article  PubMed  Google Scholar 

  3. Gianelli U, Thiele J, Orazi A, Gangat N, Vannucchi AM, Tefferi A, et al. International Consensus Classification of myeloid and lymphoid neoplasms: myeloproliferative neoplasms. Virchows Arch. 2023;482:53–68.

    Article  PubMed  Google Scholar 

  4. Khoury JD, Solary E, Abla O, Akkari Y, Alaggio R, Apperley JF, et al. The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours: Myeloid and Histiocytic/Dendritic Neoplasms. Leukemia. 2022;36:1703–19.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Krichevsky S, Sholle ET, Adekkanattu PM, Abedian S, Ouseph M, Taylor E, et al. Automated information extraction from unstructured hematopathology reports to support response assessment in myeloproliferative neoplasms. Methods Inf Med. 2024;63:176–82.

    Article  PubMed  Google Scholar 

  6. Hartsock RJ, Smith EB, Petty CS. Normal variations with aging of the amount of hematopoietic tissue in bone marrow from the anterior Iliac Crest. A study made from 177 cases of sudden death examined by Necropsy. Am J Clin Pathol. 1965;43:326–31.

    Article  PubMed  Google Scholar 

  7. Thiele J, Kvasnicka HM, Facchetti F, Franco V, van der Walt J, Orazi A. European consensus on grading bone marrow fibrosis and assessment of cellularity. Haematologica. 2005;90:1128–32.

    PubMed  Google Scholar 

  8. Bain BJ, Clark DM, Wilkins B. The Normal Bone Marrow. Bone Marrow Pathology. Fifth edition. ed. Hoboken, NJ: John Wiley & Sons Ltd.; 2019. p. 1–60.

  9. Krichevsky S, Ouseph M, Zhang Y, Abu-Zeinah G, Scandura JM, Gupta R. A deep learning-based pathomics methodology for quantifying and characterizing nucleated cells in the bone marrow microenvironment. Blood. 2023;142:2294.

    Article  Google Scholar 

  10. Wong J, Jackson R, Chen L, Song J, Pillai R, Afkhami M, et al. Determination of age-dependent bone marrow normocellularity. Am J Clin Pathol. 2024;161:170–6.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

This work is dedicated to the memory of Dr. Richard T. Silver, M.D., founder of our MPN Center, whose unwavering commitment to patients and pioneering contributions to the field continue to inspire and guide our mission.

Funding

This research was supported by the American Society of Hematology (ASH) Scholar Award 2024-2027 (GAZ); National Institutes of Health grant R01HL175556 (JMS); and the Johns Family Fund of the Cancer Research and Treatment Fund (CR&T).

Author information

Author notes

  1. deceased: Richard T. Silver.

Authors and Affiliations

  1. Richard T. Silver, MD Myeloproliferative Neoplasms Center, Weill Cornell Medicine, New York, NY, USA

    Katie Erdos, Madhu M. Ouseph, Spencer Krichevsky, Neville Lee Jr, Richard T. Silver, Ghaith Abu-Zeinah & Joseph M. Scandura

  2. Department of Pathology, Weill Cornell Medicine, New York, NY, USA

    Madhu M. Ouseph & Julia T. Geyer

  3. Department of Biomedical Informatics, Stony Brook University, Stony Brook, NY, USA

    Spencer Krichevsky & Rajarsi Gupta

Authors

  1. Katie Erdos
  2. Madhu M. Ouseph
  3. Spencer Krichevsky
  4. Julia T. Geyer
  5. Neville Lee Jr
  6. Richard T. Silver
  7. Rajarsi Gupta
  8. Ghaith Abu-Zeinah
  9. Joseph M. Scandura

Contributions

Study conceptualization and design: JMS, RTS. Data collection: KE, NL, MMO, JG. Data analysis, design of figures/tables: KE, SK, JMS. Data interpretation, writing and reviewing of the manuscript: KE, MMO, SK, JG, NL, RTS, RG, GAZ, JMS. All authors participated in the writing and review of this article, and approved the final submitted version.

Corresponding author

Correspondence to Joseph M. Scandura.

Ethics declarations

Competing interests

GAZ: Chief Medical Advisor and Member, Board of Directors: Cancer Research & Treatment Fund (CR&T). Consultant and Advisory Board Member: PharmaEssentia, Takeda, Silence Therapeutics. Research Funding: PharmaEssentia, Pharma&.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Erdos, K., Ouseph, M.M., Krichevsky, S. et al. Marrow hypercellularity is common and clinically neutral in essential thrombocythemia. Leukemia (2026). https://doi.org/10.1038/s41375-026-03151-6

Download citation

  • Received: 27 May 2026

  • Revised: 28 August 2026

  • Accepted: 15 September 2026

  • Published: 12 October 2026

  • Version of record: 12 October 2026

  • DOI: https://doi.org/10.1038/s41375-026-03151-6

来源:Nature 健康护理主题 · nature.com