Published Evidence for Factors Contributing to the Development Myelofibrosis in BCR-ABL1-Negative Myeloproliferative Neoplasms: A Scoping Review

Background:

In classical myeloproliferative neoplasms (MPN) the presence of fibrosis is associated poor outcome and increased morbidity. The landscape of published evidence is vastly heterogeneous. Hence, it is difficult to estimate the impact of individual risk factors on the development and progression of myelofibrosis (MF) in MPN.

Aims:

This scoping review aims to assess the available evidence on factors contributing to the development of MF in classical MPNs. We set out to determine (1) Which mediators have experimentally tested and quantified in the context of MF? (2) How well and consistently are results reported and published in the field of MPN?

Methods:

Following the PRISMA-guidelines, a comprehensive literature search of the databases Pubmed, Web of Science, bioRxiv, International Clinical Trials Registry Platform and Google Scholar was conducted for the time from October 2022 to December 2023.

Abstracts were screened for inclusion following a pre-defined hierarchical screening process. The search was limited to primary publications written in English and published as of 2005. Each publication needed to disclaim a clearly defined murine or human in vitro or in vivo model resulting in a fibrotic MPN-like phenotype and the level of fibrosis needed to be assessed and reported on any level.

Results:

We identified 11792 publications for the hierarchical screening process. After deduplication and screening, 962 publications met the inclusion criteria, of which 821 studies focused exclusively on clinical data and specimens. In total, 316 results were extracted from the identified 141 experimental publications. Most studies (28.48%) focused on the JAK2V617F driver mutation, followed by 20.57% using MPLW515L and 5.70% using CALR mutations. 8.86% and 6.65% of studies employed the Gata1 low model or overexpression of thrombopoietin to induce MPN, respectively.

In the results we found published evidence for 108 risk factors with 16.14% JAK2, 12.66% TGF-β, 5.70% CALR, 4.11% Pim1, 3.80% Gata1 low. These were associated with perturbations in a total of 38 fibrosis-related signaling pathways. With 23.73%, 11.71%, 6.65%, 5.38%, TGF-β, JAK/STAT, Hh, and NFκB, were the most frequently described pathways involved. 21.52% mediators and 5.06% pathways were only reported once.

There was a relevant heterogeneity in reporting the extent of MF across studies. 62.34% studies accessed MF using reticulin staining but only 28.43% of these reported their results semi-quantitatively in the established grading system as a standard diagnostic criterion. 9.14% reported MF severity as fiber staining area and 62.44% reported only qualitative statements about changes in fibrosis severity. 14.24% of all studies reported collagen expression, 6.01% α-SMA expression, 2.85% fibrocyte frequency and 2.85% stroma cell frequency as readout. Rare studies used indirect and less established surrogate markers such as the expression of TGF-β or fibronectin as measures of fibrosis severity.

Conclusion:

This scoping review revealed that experimental evidence is only published for a limited number of mediators with JAK2 and TGF-β associated with their respective pathways backed by the highest number of studies. Many risk factors were only reported in a single study, which can be interpreted as a lack of confirmative studies in the field.

Inconsistent reporting of results makes it challenging to methodologically pool quantitative measurements of effect size and variance by standard approaches. A standardized convention for the reporting of results and harmonization in the future will be important to establish.

No relevant conflicts of interest to declare.