The bidirectional relationship between JAK2V617F allele burden and inflammation in polycythaemia vera: Insights from interferon and ruxolitinib therapy

Abstract

Myeloproliferative neoplasms (MPNs), including polycythemia vera (PV), essential thrombocythemia (ET), and primary myelofibrosis (PMF), are clonal hematopoietic stem cell disorders driven by recurrent mutations in JAK2, CALR, or MPL, resulting in constitutive activation of the JAK–STAT signaling pathway.

Although these mutations are central to disease initiation, accumulating evidence indicates that chronic inflammation plays a pivotal role in disease progression and clinical manifestations, making it a relevant therapeutic target.¹

In PV, the interplay between inflammation and clonal hematopoiesis appears particularly important as inflammatory signals may promote the expansion and persistence of the JAK2V617F mutant clone, thereby influencing its accumulation over time. Within this context, the two principal targeted agents, pegylated interferon-α and ruxolitinib, both exert anti-inflammatory effects. Interferon therapy in PV evolved from conventional and pegylated IFN-α2a to ropeginterferon alfa-2b (Ropeg), a long-acting formulation that enables sustained treatment while preserving disease-modifying activity.

Recent data indicate that ropeginterferon and ruxolitinib differ in clonal suppression, likely due to distinct interactions with the inflammatory microenvironment, providing the rationale for this brief review of their biological and therapeutic implications.

Among inflammatory mediators, IL-1 signaling appears particularly relevant. Elevated IL-1 levels have been demonstrated in both MPN patients and JAK2V617F mouse models. Genetic deletion of IL-1R1 normalizes blood counts, reduces splenomegaly and fibrosis, and decreases mutant hematopoietic stem/progenitor cells, whereas exogenous IL-1β promotes myeloid expansion and fibrosis.² Complementary studies showed that IL-1β is produced by JAK2V617F hematopoietic cells and contributes to bone marrow niche damage, myelofibrosis, and osteosclerosis, with additive therapeutic effects observed when IL-1β suppression is combined with ruxolitinib.² Importantly, IL-1β also appears involved in disease initiation, since its inhibition reduces engraftment and clonal expansion of JAK2-mutant HSCs and delays progression from JAK2V617F CHIP to overt MPN.³

Upstream of IL-1β, the NLRP3 inflammasome has emerged as a critical regulator of inflammatory activation. Deletion of Nlrp3 abolishes IL-1β and IL-18 release and attenuates thrombocytosis, granulocytosis, HSPC expansion, splenomegaly, and fibrosis in JAK2V617F models. Pharmacologic NLRP3 inhibition reverses established disease manifestations, supporting its therapeutic potential.⁴

TNF-α also promotes clonal dominance. JAK2V617F induces TNF-α production, while mutant progenitors display resistance to TNF-α-mediated suppression, unlike normal hematopoietic cells.⁵ This selective survival advantage contributes to suppression of normal hematopoiesis and expansion of the malignant clone. Persistent inflammatory activation is further sustained by defective TLR negative regulation in MPN monocytes.

Hyperactivation of NF-κB signaling represents another major pathogenic mechanism, involving both malignant and non-malignant hematopoietic compartments and highlighting the ability of MPN cells to remodel the bone marrow microenvironment. Inhibition of NF-κB reduces inflammatory cytokines and disease burden, while combined JAK/BET inhibition further enhances therapeutic efficacy and reverses fibrosis in vivo.⁶

Additional inflammatory amplification derives from integrin signaling, which activates FAK, SYK, NF-κB, and JNK pathways, promoting cytokine production and thrombosis.⁷

Finally, genotype-specific inflammatory programs further modulate disease biology. JAK2V617F neutrophils display a stronger inflammatory and adhesive phenotype than CALR-mutant cells, potentially explaining the higher thrombotic risk associated with JAK2-mutated disease.⁸

This inflammation-clonal expansion axis functions as a self-perpetuating feedback loop: chronic inflammation confers a selective advantage to mutant HSCs, while the expanding clone further amplifies the inflammatory milieu.⁸ MPN stem cells co-opt mechanisms normally used by healthy HSCs to prevent exhaustion, including signal attenuation, insulation from activating cytokines, and epigenetic reprogramming, in order to outcompete their normal counterparts under inflammatory conditions. An example of mutational cooperation is the interaction between JAK2V617F and Dnmt3a loss, which drives progression from PV to MF through activated enhancer-driven inflammatory signaling.⁹

Collectively, the body of evidence supports the concept that inflammation is not merely an epiphenomenon of MPNs but rather a central pathogenic driver that establishes a permissive microenvironment for clonal expansion and sustained disease evolution. Consequently, inflammation represents a rational and actionable therapeutic target.

Within this context, the two principal non-chemotherapeutic approaches, pegylated interferon-α and ruxolitinib, both exert anti-inflammatory effects; however, current data indicate that their impact on clonal suppression differs, likely reflecting distinct mechanisms of interaction with the inflammatory milieu.

Preclinical studies have shown that interferon-α selectively targets JAK2V617F-mutant hematopoietic stem cells by inducing their exit from quiescence and promoting terminal differentiation, ultimately leading to progressive exhaustion of the malignant clone.¹⁰ Clinical evidence further indicated that Ropeg induces a progressive and sustained reduction in JAK2V617F allele burden over time. In contrast, hydroxyurea (HU) was generally associated with more modest molecular effects, consistent with a predominantly cytoreductive action on proliferating progenitor cells rather than disease-initiating cells.¹¹ The characteristic kinetics of response observed with Ropeg are continuous and time-dependent, a pattern most compatible with progressive depletion of the mutant stem cell pool.

The neutrophil-to-lymphocyte ratio (NLR) has emerged as a key biomarker for identifying PV patients who derive the greatest benefit from Ropeg therapy, further emphasizing the interplay between inflammation and clonal dynamics.¹² To date, no data are available on NLR dynamics during therapy in patients with evolving or PMF.¹³

In the Low-PV trial, patients with baseline NLR ≥ 3.5 exhibited significantly higher long-duration disease from diagnosis and JAK2V617F VAF compared with those with a lower NLR value (56% vs. 20%, P = 0.001) together with a more proliferative disease phenotype.¹² Ropeg proved superior to phlebotomy alone in reducing NLR (P = 0.008), primarily through a marked decrease in neutrophil counts. Importantly, this reduction was linearly associated with suppression of JAK2 VAF and achievement of the primary study endpoint (P = 0.021). Notably, this association was confined to patients with elevated baseline inflammatory status: among patients with NLR < 3.5, no correlation between NLR changes and JAK2 VAF reduction was observed after 12 months, suggesting that in a low-inflammatory context, Ropeg may not exert a measurable early effect on the malignant clone (Figure 1).

These observations are further supported by the PROUD-PV/CONTINUATION-PV studies,¹⁴ in which Ropeg induced a sustained and durable reduction in NLR over time, particularly among patients with a higher baseline inflammatory burden (e.g., NLR ≥ 4.3), an effect not observed with HU. Notably, a clear gradient emerged between the magnitude of NLR reduction and the depth of JAK2V617F suppression, with greater reductions in NLR associated with progressively deeper molecular responses. This apparent dose–response relationship supports the concept that control of inflammation is closely linked to clonal dynamics and may represent a key mechanism underlying disease modification in PV (Figure 2).

Taken together, these findings support a unified two-step model of IFN-α activity: an initial phase dominated by anti-inflammatory and immunomodulatory effects, followed by a delayed yet progressive targeting of the malignant clone. Within this framework, inflammation emerges as a modifiable determinant of therapeutic efficacy; a similar benefit may also be expected with ruxolitinib given its potent anti-inflammatory activity.

An important and timely contribution in this context comes from the study by Guglielmelli et al.,¹⁶ which provides the first longitudinal integrated analysis of NLR dynamics in relation to JAK2V617F allele burden in patients with PV treated with ruxolitinib.

In this cohort of 41 patients HU-resistant/intolerant treated with ruxolitinib for a median of 8.7 years, baseline NLR strongly correlated with JAK2V617F VAF (r = 0.5, P = 0.001). Patients with high-NLR exhibited a substantially greater allele burden compared to low-NLR patients (81.8% vs. 54.8%, P < 0.001). During treatment, only patients with elevated baseline NLR experienced a significant and sustained reduction in NLR, from 9.6 to 4.2 at 3 years (P = 0.001), accompanied by a progressive decline in JAK2V617F VAF (from 81.8% to 52.5% at 4 years, P = 0.008). By contrast, patients with low baseline NLR showed no significant longitudinal changes in either NLR or VAF.

These findings are particularly compelling as they reinforce the biological link between inflammatory activation and clonal expansion, while also suggesting that NLR may serve as a simple, clinically accessible surrogate biomarker of disease activity. Moreover, the consistency of these results with those observed in Low-PV and PROUD-PV/CONTINUATION-PV¹⁴ strengthens their relevance and supports the broader applicability of the inflammation/clone axis across different therapeutic classes.

At the same time, the study does not definitely clarify whether ruxolitinib exerts a direct disease-modifying effect on the malignant stem cell compartment or whether the observed reductions in allele burden are predominantly secondary to its well-established anti-inflammatory activity.¹⁷ Despite limitations related to the relatively small sample size, the Guglielmelli et al. study provides a strong rationale for further investigation. Specifically, it supports the concept that targeting inflammation may influence clonal dynamics, while simultaneously underscoring the need for larger, prospective studies to determine whether these effects translate into genuine disease modification comparable to that observed with interferon.

Gorantla et al.¹⁷ provided experimental support for this interpretation by demonstrating, in ruxolitinib-resistant MPN mouse models, that the therapeutic efficacy of ruxolitinib was independent of direct oncogenic JAK2V617F signaling and is rather mediated by suppression of pro-inflammatory cytokines produced by non-malignant cells within the tumor microenvironment. Similarly, Kleppe et al.¹⁸ showed that JAK-STAT3–driven cytokine production from both malignant and nonmalignant cells contributes substantially to MPN pathogenesis. Their findings further support the concept that the principal therapeutic effect of JAK inhibition derives from dampening the inflammatory milieu rather than eradicating the clone.¹⁵

The MAJIC-PV trial¹⁹ provides important clinical support for this framework. In this randomized Phase II study of 180 HU-resistant/intolerant PV patients, ruxolitinib achieved higher complete response rates than best available therapy (43% vs. 26%, P = 0.02), together with improved duration of response and event-free survival. Molecular response, defined as ≥50% reduction in JAK2V617F allele burden, was also more frequent with ruxolitinib (56% vs. 25%, P < 0.001) and associated with superior progression-free, event-free, and overall survival. However, no information on NLR or other inflammatory biomarkers was reported.

The clinical significance of JAK2V617F reduction remains debated. As emphasized by Tefferi et al.²⁰ co-occurring mutations such as TET2 and ASXL1 complicate attribution of disease biology solely to JAK2V617F. Among these, ASXL1 appears particularly relevant. In MAJIC-PV, ASXL1 mutations independently predicted inferior event-free survival. Experimental studies further indicate that mutant ASXL1 actively promotes inflammation through NF-κB activation, interferon-driven programs, and TNFα signaling, positioning ASXL1 as a key inflammatory amplifier beyond JAK2V617F signaling.

This framework provides a plausible explanation for the inferior outcomes of ASXL1-mutant patients treated with ruxolitinib: while the drug effectively suppresses JAK–STAT–dependent cytokine signaling, it may not fully counteract inflammation driven by alternative pathways such as NF-κB and TNFα. Whether Ropeg can overcome this additional inflammatory layer remains uncertain. Early clinical observations raise caution, as adverse mutations including ASXL1 have been associated with vascular events, suboptimal responses, and clonal evolution during interferon therapy, suggesting that mutation-specific inflammatory programs may influence treatment efficacy.

Current evidence supports the concept that inflammation regulates JAK2V617F allelic burden and disease evolution in PV. Accordingly, Ropeg and ruxolitinib appear complementary rather than interchangeable: Ropeg combines clonal suppression with immune modulation, whereas ruxolitinib primarily targets the inflammatory milieu, highlighting inflammation as a key determinant of clonal dynamics and therapeutic response.

Future research should determine whether PV patients with short telomeres, reflecting a longer replicative history of JAK2-mutant clonal hematopoiesis,²¹ exhibit distinct inflammatory and therapeutic trajectories. Similar longitudinal studies in myelofibrosis should assess the relationship between NLR and JAK2V617F allele burden during ruxolitinib therapy to establish whether this bidirectional interplay extends to advanced MPNs.

In addition, conventional risk stratification should evolve toward integrated models combining clonal architecture (JAK2V617F VAF, ASXL1, and other co-mutations) with quantitative inflammatory markers, including NLR and cytokine profiles.

Finally, statins and metformin are attractive candidates for combination strategies because of their anti-inflammatory and immunomodulatory properties.^{22, 23} Emerging evidence also suggests that statins may enhance the efficacy of interferon-α, potentially contributing to the achievement of minimal residual disease in a subset of patients.²² Combination strategies targeting both JAK–STAT–dependent and independent pathways may further improve long-term clinical outcomes.

The Danish COMBI-I and COMBI-II studies support the complementary roles of interferon and ruxolitinib by demonstrating improved hematologic and molecular responses and restoration of interferon responsiveness.^{24, 25}

Together, these advances support a shift from blood count control toward biology-directed therapy, in which simultaneous targeting of clonal evolution and inflammation may ultimately modify the natural history of PV.

Tiziano Barbui: Conceptualization; writing—original draft. Paola Guglielmelli: Writing—review and editing. Alessandro Maria Vannucchi: Writing—review and editing.

T.B.: received honoraria for lecture from AOP; P.G.: received honoraria for lectures and advisory board from Incyte, AOP, Novartis, and GSK; A.M.V.: received honoraria for lectures and advisory board from Incyte, AOP, Bluemedicine, Menarini Stemline, Novartis, and GSK.

This work is supported by Fondazione per la Ricerca Ospedale di Bergamo – ETS, Bergamo, Italy.

Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.