Sensitivity and Specificity of Laboratory Parameters to Detect Early/Prefibrotic Myelofibrosis in 857 Patients with Essential Thrombocythemia. A Diagnostic Algorithm

Patients presenting with a clinical picture of essential thrombocythemia (ET) can actually have an early/prefibrotic myelofibrosis (PMF), according to current WHO criteria, in about 18% of cases. Laboratory tests which are significantly different in early/prefibrotic PMF as compared with histologically confirmed ET (WHO-ET) include decreased gender-matched hemoglobin (Hb), increased white blood cell (WBC), platelet (PLT) counts and lactate dehydrogenase (LDH) values.

To evaluate sensitivity (SE) and specificity (SP) of blood cell counts and LDH, at presentation, for the diagnosis of early/prefibrotic PMF vs. WHO-ET.

Five hundred thirty-six cases (50%) who had complete laboratory data measured at diagnosis constituted the exploratory set of our study and were derived from an international ET database. The discriminatory ability of Hb, WBC, PLT and LDH in correctly classifying patients in the early/prefibrotic PMF or WHO-ET groups was initially tested by plotting their Receiving Operating Characteristic (ROC) curves and comparing the relative Areas Under the Curve (AUC) with the value of 0.50 (which stands for the completely useless of the test). Three parameters with statistically significant discriminatory power were chosen (Hb, WBC and LDH) and thresholds searched in order to guarantee at least 90% of SE or SP. Finally, a diagnostic algorithm was designed.

The validation set of this analysis was constituted by 321 patients with WHO-ET (n=62) or early/prefibrotic PMF (n=259) diagnosed by the same pathologist who confirmed the training set cohort and collected in the Institute for Pathology, University of Cologne, Germany. SE and SP for the same parameters and thresholds as well as the final diagnostic algorithm were applied to this set of patients to demonstrate the results' reproducibility.

Sensitivity and specificity to recognize early/prefibrotic PFM have been evaluated by ROC curves. The best performance was found for LDH (AUC = 0.7059). WBC and Hb had super imposable curves, with AUC of 0.6279 and 0.6257, respectively. The worst performance was registered for PLT count: its AUC was only 0.5628, not significantly different from the reference value of 0.50 (p=0.154). Thresholds of Hb, WBC and LDH were searched to achieve at least 90% of SE or SP. HB < 12 g/dL for women or <13 g/dL for men, and WBC >= 13 x10⁹/L had higher SP (92% and 91%, respectively). High SP is highly related to the presence of early/prefibrotic PMF (true positives).

On the contrary, LDH < 200 mU/mL and WBC < 7 x10⁹/L had good sensitivity (91% and 94%, respectively). High SE is highly related to the absence of early/prefibrotic PMF (true negatives). By applying these SE and SP values in a step-by-step algorithm, nearly half of patients (48%) could be classified as WHO-ET or early/prefibrotic PMF, assuming at each step a margin of error of less than 10%. For the remaining 50% of patients, laboratory results didn't allow to suspect or exclude the presence of early/prefibrotic PMF.

In the validation set of 321 patients classified by WHO 2008 as true ET or early/prefibrotic PMF (Cologne cohort) SP of anemia was 84%, WBC < 7 x10⁹/L or >= 13 x10⁹/L had 91% and 81% of SE and SP, respectively. LDH values < 200 mU/mL had 85% of SE. By applying the same flow-chart, 46% of patients were classified as WHO-ET or early/prefibrotic PMF.

The present study provides clinicians with laboratory parameters that should increase suspicion of early/prefibrotic PMF in a patient with a working clinical diagnosis of ET. In fact, while patients presenting clinically with ET can now be discriminated as true ET or early/prefibrotic PMF by adopting the WHO 2008 criteria that require bone marrow histology, an algorithm including baseline anemia, WBC count and LDH, allows this differentiation in about 50% of patients with a good approximation. However, for a definitive proof, bone marrow histology is still an integral part for final diagnosis.

Vannucchi:Novartis: Honoraria.