Rates of Myeloid Neoplasms in Patients With Unexplained Cytopenias

Patients with unexplained cytopenias may never receive a definitive diagnosis, or may eventually be classified as having myelodysplastic syndromes (MDS, a myeloproliferative neoplasm [MPN] or, if their cytopenias persist, as having idiopathic cytopenia of undetermined significance [ICUS]). Adding mutational profiling to the evaluation of patients with unexplained cytopenias could help identify individuals who have or are at risk of developing a myeloid malignancy, according to results of a prospective cohort study published in Blood.

Lead author Luca Malcovati, MD, from the Department of Molecular Medicine at the University of Pavia in Italy, and colleagues looked for somatic mutations in 40 genes known to be recurrently mutated in myeloid malignancies to determine their predictive value for identifying patients with, or at high risk of developing, a myeloid neoplasm. The researchers found that carrying one or more of these somatic mutations was associated with a high probability of developing a myeloid neoplasm during follow-up (hazard ratio [HR] = 13.9; p<0.001).

“Mutation profiling on peripheral blood cells has a high predictive value for identifying individuals at high risk of developing MPNs,” Dr. Malcovati told ASH Clinical News. “Mutation analyses may be a valuable complement to the current diagnostic work-up of unexplained cytopenia.”

The authors included 873 patients with unexplained cytopenias who underwent evaluation (including peripheral blood and bone marrow examinations) at the Fondazione IRCCS Policlinico San Matteo and the University of Pavia between 2003 and 2015. Patients were split into a learning cohort (those being investigated for unexplained cytopenias; n=683 consecutively seen patients) and a validation cohort (those referred to the same institution for a second opinion for a suspected myeloid malignancy; n=190). Mutation analysis was performed on DNA collected at the time of initial evaluation.

“Mutation analyses may be a valuable complement to the current diagnostic work-up of unexplained cytopenia.”

-Luca Malcovati, MD

After undergoing diagnostic work-up, patients were prospectively followed to establish the diagnosis and to monitor disease.

In the learning cohort, patients were eventually diagnosed with: a myeloid malignancy (n=409), ICUS (n=154), or other cytopenias, including those due to iatrogenic, immune-mediated, or secondary cytopenia (n=120). In the validation cohort, the final diagnosis was a myeloid malignancy in 138 patients, while a myeloid malignancy was ruled out as a cause of cytopenias in 38 patients. ICUS was provisionally adopted as a diagnosis in 14 patients.

Among patients with ICUS after initial diagnostic work-up, those who carried at least one of the mutations had a significantly higher risk of developing a myeloid malignancy compared with those without clonal evidence (HR=13.9; 95% CI 5.40-35.91; p<0.001). The 5- and 10-year cumulative probabilities of progression were 82 percent versus 9 percent and 95 percent versus 9 percent, respectively (p<0.001).

A total of 435 patients (64%) in the learning cohort carried at least one of the 40 somatic mutations included in the panel. Patients with ≥1 mutation had a positive predictive value for a myeloid malignancy of 81 percent (0.81; 95% CI 0.76-0.84), and the positive
predictive value appeared to be even stronger in patients with ≥2 mutations, at 88 percent (0.88; 95% CI 0.84-0.92). Conversely, the absence of any of these mutations had a negative predictive value for myeloid neoplasm (0.76; 95% CI 0.70-0.81; p value not reported).

Within the 40-gene panel, spliceosome genes (SF3B1SRSF2U2AF1), JAK2, and RUNX1 were associated with the highest predictive value for myeloid malignancies, ranging from 0.88 to 0.97.

The positive predictive value of mutations in TET2DNMT3A, or ASXL1 occurring as single lesions ranged from 29 percent to 57 percent, and co-mutation patterns involving these same genes had positive predictive values for myeloid neoplasm ranging from 86 percent to 100 percent. Together, these mutations and patterns “accounted for 73 percent of mutated myeloid malignancies,” the authors wrote.

In multivariate regression analysis, two factors were found to be independent predictors of MDS or another myeloid malignancy:

  • the presence of ≥2 mutations (odds ratio [OR] = 4.69; p<0.001)
  • the presence of an SF3B1 mutation (OR=4.83; p=0.003)

Factors independently associated with diagnoses other than myeloid malignancies included isolated DNMT3A (OR=0.21; p=0.001) or TET2 (OR=0.43; p=0.023) mutations.

“The results of this study suggest that mutation analysis of peripheral blood cells may significantly improve the current diagnostic approach to [patients] with unexplained cytopenias,” the authors concluded. “More effective, non-invasive diagnostic procedures are in turn expected to improve compliance to diagnostic tests.”

“We also identified mutation patterns showing a high specificity for myeloid malignancies with dysplasia,” Dr. Malcovati added. “Notably, patients with ICUS carrying these highly specific mutation patterns have a clinical course not significantly different from that of patients with a myelodysplastic syndrome. This indicates that cases with inconclusive dysplasia but highly specific mutation patterns might be classified as bona fide myelodysplastic syndromes.”

The study is limited by its lack of sequential mutation analysis – the authors were unable to exclude changes in mutation pattern or variant allele frequency as factors potentially contributing to the risks for myeloid malignancies.


Malcovati L, Gallì A, Travaglino E, et al. Clinical significance of somatic mutation in unexplained blood cytopenia. Blood. 2017 April 19. [Epub ahead of print]