Background:

Primary myelofibrosis (PMF) and essential thrombocytosis (ET) are common Philadelphia chromosome–negative chronic myeloproliferative neoplasms (MPNs), characterized by cytoses, splenomegaly, and hypercellular bone marrows with proliferation of myeloid, erythroid, and/or megakaryocytic lineages. However, it is a challenge to distinguish between ET and prefibrotic/early myelofibrosis(Pre-PMF). In the WHO 2016 classification of myeloid neoplasms, Pre-PMF was defined as a distinct entity. It often presented with marked thrombocytosis in the peripheral blood and granulocytic and atypical megakaryocytic proliferation with minimal, if any, fibrosis in the BM. Pre-PMF is the early stage of PMF, and its prognosis differs significantly compared to ET and overt-PMF. Early identification and effective intervention may help improve patient prognosis by delaying disease progression.

Methods:

Clinical data from 249 patients diagnosed with ET, Pre-PMF, or primary myelofibrosis (PMF) were collected. Independent predictors for discriminating ET from Pre-PMF were identified using multivariate logistic regression analysis. The diagnostic performance of individual variables was evaluated using receiver operating characteristic (ROC) curve analysis, and an optimal ET-Pre-PMF discrimination model was constructed. The Hosmer-Lemeshow test was used to assess model goodness-of-fit, while ROC analysis evaluated the predictive value of the diagnostic model.

Results:

This study included a total of 249 MPN patients (comprising 144 ET cases, 68 Pre-PMF cases, and 37 Over-PMF cases), with a median age of 60 years (interquartile range [IQR], 51–71.5 years). Further analysis of the absolute count of peripheral blood CD34+ cells in these patients revealed that patients with Pre-PMF had an intermediate number of circulating CD34+cells (median number: 9.16/μL, interquartile range, IQR: 6.9925–10.7375) between those with ET (3.595/μL, IQR: 2.665-6.945) and those with overt PMF (25/μL, IQR: 12.23-46.4).

Using a peripheral blood CD34+ cell absolute count of 6.8/μL (median) as the cutoff, all MPN patients were divided into two groups: <6.8/μL and ≥6.8/μL. Compared to the <6.8/μL group, the ≥6.8/μL group showed significantly lower total T cell absolute counts (1646 vs 993/μL, P<0.05), helper T cell absolute counts (1048 vs 496/μL, P<0.05), CD4+/CD8+ ratio (2.21 vs 1.22, P<0.05), NK cell absolute counts (156.5 vs 110/μL, P<0.05), and B cell absolute counts (256.5 vs 99/μL, P<0.05).

We further analyzed the correlation between peripheral blood CD34+ cell absolute counts and lymphocyte subsets. The results showed that total T cell absolute counts (r= -0.510, p<0.001), helper T cell absolute counts (r= -0.594, p<0.001), CD4+/CD8+ ratio (r= -0.383, p<0.001), NK cell absolute counts (r= -0.173, p=0.006), and B cell absolute counts (r= -0.392, p<0.001) were negative correlation with peripheral blood CD34+ cell absolute counts.

Multivariate logistic regression analysis showed that: peripheral blood CD34+ cell absolute count (OR=0.055, 95%CI 0.011-0.291, P=0.001), CD4+/CD8+ ratio (OR=52.763, 95%CI 2.641-1053.963, P=0.009), B cell absolute count (OR=9.682, 95%CI 1.357-69.074, P=0.024), hemoglobin level (OR=11.574, 95%CI 2.345-57.12, P=0.003), bone marrow fibrosis grade (OR=4.763, 95%CI 1.017-22.308, P=0.048), and thrombosis history (OR=0.034, 95%CI 0.004-0.338, P=0.004) were independent predictive factors for differentiating ET from Pre-PMF. ROC curve analysis showed that the differential diagnosis model constructed with these variables had an AUC of 0.92, with a sensitivity of 92.6% and specificity of 75.7%. The calibration curve of the nomogram model showed that the predicted curve closely followed the ideal standard curve without significant deviation.Conclusions: Peripheral blood CD34+ cell absolute counts could be used to differentiate between ET, Pre-PMF and PMF. Circulating CD34+cells, lymphocyte subsets, fibrosis grade, history of thromboembolic disease and Hb may be potential biomarkers for the differential diagnosis of ET and prePMF patients. Our results suggest that this nomogram might serve as a tool for the identification of prePMF patient at Early stage . As such, we speculated that the diagnostic nomogram model demonstrated robust discrimination performance in clinical practice, particularly in developing countries.

This content is only available as a PDF.
2025
Sign in via your Institution