Polycythemia vera (PV) and essential thrombocythemia (ET) are chronic Philadelphia chromosome negative myeloproliferative neoplasms (MPN) characterized by clonal hematopoiesis, thrombotic risk, and variable disease evolution 1, 2. Increasing evidence supports a role for immune dysregulation and inflammatory pathways in disease development and clinical heterogeneity, although their clinical implications remain incompletely defined 3. Epidemiological studies have reported an association between prior autoimmune diseases and an increased risk of developing MPN 4. Autoimmune or inflammatory diseases (AID) have also been described in approximately 6%–8% of patients with MPN, with heterogeneous effects on clinical features and survival across cohorts 5, 6. However, the clinical burden of AID in patients specifically affected by PV and ET, particularly concerning disease outcomes and comorbidity profiles, remains poorly defined. We therefore evaluated the prevalence of AID in a large cohort of patients with PV and ET and assessed their association with disease outcomes and comorbidities. Institutional review board approval was obtained from the Mayo Clinic, Rochester, MN, USA, for retrospective data collection in accordance with the Declaration of Helsinki. The study included consecutive patients with PV and ET evaluated between 1970 and 2024 and diagnosed according to the 2022 International Consensus Classification criteria 7. AID were defined according to established disease-specific clinical and serologic criteria. Clinical and laboratory variables were collected at diagnosis or first referral. Thrombotic events were confirmed by imaging studies and classified as arterial or venous; AID, cancers, and thrombotic events were categorized as occurring before/at or after MPN diagnosis using a ≥ 3 month interval. Overall survival was estimated using the Kaplan–Meier method and compared using the log-rank test. Competing-risk analyses were performed for thrombosis, leukemic transformation, myelofibrosis progression, second cancer, and incident AID using Fine–Gray regression models. Categorical variables were compared using the chi-square or Fisher's exact test, and continuous variables using the Mann–Whitney U test. Statistical analyses were performed using SPSS, JMP, and R software. Among 1968 patients with polycythemia vera (PV) or essential thrombocythemia (ET), 121 (6%) had a history of AID before/at the time of MPN diagnosis (Table 1). Compared with patients without AID, those with prior AID were older (median age 62 vs. 60 years, p = 0.016), more frequently female (69% vs. 55%, p = 0.002), and less likely to have PV (40% vs. 52%, p = 0.011). Hemoglobin levels were significantly lower in patients with AID (median 14.9 vs. 16 g/dL; p = 0.003), while platelet and leukocyte counts were similar between groups. A history of venous thrombosis at or prior to diagnosis was more frequent among patients with AID (18% vs. 12%, p = 0.046), whereas arterial thrombosis rates were comparable. Patients with previous AID more frequently had a history of other malignancies before/at MPN diagnosis (25% vs. 14%, p = 0.001), including when non-melanoma skin cancers (NMSC) were excluded (18% vs. 10%, p = 0.006). Additionally, the occurrence of subsequent AID during follow-up was more common among patients with prior AID (6% vs. 2%, p = 0.005). Among the 121 patients with AID before/at MPN diagnosis, 48 (40%) had PV and 73 (60%) had ET. Patients with PV were older (median 65 vs. 59 years, p = 0.043) and females were more common among ET cases (78% vs. 56%, p = 0.011). As expected, PV patients displayed higher hemoglobin levels, leukocyte counts, and neutrophil counts, while ET patients had higher platelet counts. Fibrotic progression was more frequent among PV cases (17% vs. 3%, p = 0.006), whereas leukemic transformation rates were low and comparable (Table S1). The spectrum of AID observed before/at and after MPN diagnosis is detailed in Figure 1A. The most frequently observed conditions included autoimmune dysthyroidism, rheumatoid arthritis, psoriasis, inflammatory bowel disease, and polymyalgia rheumatica. Kaplan–Meier analysis showed no significant difference in survival between patients with and without prior AID (log-rank p = 0.09, Figure 1B). In univariate analysis, prior AID showed a borderline association with inferior OS (HR 1.36, 95% CI 0.96–1.94, p = 0.087). After adjustment for sex, age, PV phenotype, and JAK2 mutation status, the association remained borderline and moved closer to statistical significance (HR 1.42, p = 0.054). Male sex (HR 1.50, 95% CI 1.27–1.77, p < 0.001) and increasing age (per 10 years, HR 2.43, 95% CI 2.25–2.62, p < 0.001) were independently associated with inferior survival, whereas PV (HR 1.15, p = 0.16) and JAK2 mutation status (HR 1.14, p = 0.34) were not significantly associated. To explore potential heterogeneity across AID, patients with prior AID were further stratified into two predefined categories based on the predominant clinical phenotype and anticipated degree of systemic inflammatory burden, distinguishing inflammatory arthritis, vasculitis, or unclassified AID from predominantly organ-specific, connective tissue, or inflammatory dermatosis conditions (Figure S1). In univariate analysis, patients with inflammatory arthritis, vasculitis, or unclassified AID exhibited significantly inferior survival compared with patients without AID (HR 1.89, 95% CI 1.13–3.16; p = 0.015), whereas those with organ-specific, connective tissue, or inflammatory dermatosis AID showed no significant difference (HR 1.10, 95% CI 0.69–1.77; p = 0.679). After multivariable adjustment, inflammatory arthritis/vasculitis/unclassified AID remained significantly associated with inferior survival (HR 1.64, 95% CI 1.01–2.70; p = 0.040), whereas organ-specific/connective tissue/inflammatory dermatosis AID showed no significant association (HR 1.29, 95% CI 0.80–2.08; p = 0.294). The cumulative incidence of major clinical outcomes after MPN diagnosis was assessed using competing-risk analysis (Figure S2). Prior AID was not significantly associated with MF progression (Gray test p = 0.20), AML progression (p = 0.52), venous (p = 0.49), or arterial thrombosis (p = 0.86). Conversely, a history of AID was associated with an increased risk of SC (Gray test p < 0.001, sHR 1.57, p = 0.015, Figure 2A). This association remained significant after excluding non-melanoma skin cancers (NMSC) (sHR 1.70, p = 0.022), whereas after excluding both NMSC and hematologic malignancies, only a non-significant trend was observed (sHR 1.52, p = 0.12). In the multivariable model including all cancers, prior AID was not independently associated with SC risk (sHR 1.31, 95% CI 0.88–1.96, p = 0.183) after adjustment for sex, age, PV versus ET, JAK2 mutation status, prior cancer history, hyperlipidemia, hypertension, and arterial thrombosis. Of interest, after exclusion of NMSC, prior AID showed a trend toward increased SC risk (sHR 1.50, 95% CI 0.93–2.42, p = 0.094). This association was no longer observed in the most restrictive model excluding both NMSC and hematologic malignancies (sHR 1.35, 95% CI 0.79–2.32, p = 0.276). The occurrence of incident AID after MPN diagnosis was subsequently analyzed using competing-risk methods. The cumulative incidence of AID was significantly higher among patients with prior AID compared with those without (Gray test p < 0.001, Figure 2B). In univariate analysis, prior AID (sHR 3.89, p = 0.001) and female sex (sHR 2.63, p = 0.007) were significantly associated with an increased risk of incident AID. Hypertension showed a borderline association (sHR 1.74; p = 0.095), whereas no other clinical, hematologic, or molecular variables were significantly associated (Table S2). In the multivariable model (Figure 2C), prior AID (sHR 3.03, p = 0.012), female sex (sHR 2.39, p = 0.015), and hypertension (sHR 2.18, p = 0.024) remained independently associated with an increased risk of incident AID. In this large cohort of patients with PV and ET, AID were identified in 6% of cases and were associated with a distinct clinical and comorbidity profile. Patients with prior AID were more frequently female, slightly older, and more likely to have a history of venous thrombosis and prior malignancy. These findings are consistent with previous epidemiologic studies supporting shared inflammatory and immune-mediated mechanisms across AID and myeloid neoplasia, as well as the increased risk of MPN development among patients with pre-existing AID 4, 6, 8. Although prior AID was not significantly associated with fibrotic or leukemic progression, nor with arterial or venous thrombosis during follow-up, a borderline association with inferior survival was observed and persisted after multivariable adjustment. Exploratory analyses highlighted heterogeneity across autoimmune phenotypes, with selected AID categories demonstrating a sustained association with inferior survival 9. Collectively, these findings support the concept that immune dysregulation and chronic inflammatory burden contribute to disease heterogeneity in MPN, influencing long-term outcomes beyond classical disease-related risk factors 3. A history of AID was associated with an increased risk of SC in unadjusted analyses, particularly after exclusion of NMSC, although this association was attenuated after multivariable adjustment. This finding suggests that part of the observed risk may be explained by coexisting clinical factors, including prior malignancy history and cardiometabolic comorbidities, while remaining consistent with prior observations linking chronic immune dysregulation with carcinogenesis across multiple disease settings 5, 10-12. A particularly relevant observation was the significantly increased risk of incident AID after MPN diagnosis among patients with prior AID, independent of sex, age, and hypertension. This finding aligns with the well-recognized phenomenon of polyautoimmunity, whereby individuals with one autoimmune disease have an increased risk of developing additional immune-mediated conditions over time 13, 14. Several limitations should be acknowledged. The retrospective design introduces potential selection and ascertainment biases, and the relatively low number of specific outcomes may have limited statistical power in subgroup analyses. In addition, treatment heterogeneity across the long study period was not specifically accounted for and may have influenced clinical outcomes. Furthermore, the intrinsic heterogeneity of AID, encompassing conditions with distinct pathophysiologic mechanisms and clinical trajectories, limited the ability to evaluate outcomes for individual AID entities and required grouping into broader categories, which may have diluted subtype-specific associations. Despite these limitations, the large sample size, extended follow-up, and use of competing-risk methodology strengthen the robustness of the findings. G.G.L., N.G., and A.T. conceived and designed the study. G.G.L., F.A., M.I., P.F., M.S.R., and N.G. collected the data. G.G.L. performed statistical analyses. G.G.L. drafted the manuscript. P.G., A.M.V., A.P., N.G., and A.T. reviewed and edited the manuscript. The authors declare no conflicts of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request. TABLE S1: Clinical, laboratory, and genetic characteristics at diagnosis and during follow-up in 121 patients with autoimmune or inflammatory disease before/at MPN diagnosis, stratified by polycythemia vera and essential thrombocythemia. TABLE S2: Univariate Fine–Gray regression analysis for risk of incident autoimmune or inflammatory diseases (AID) after MPN diagnosis. FIGURE S1: Overall survival according to autoimmune or inflammatory disease (AID) category. Kaplan–Meier curves comparing overall survival among patients with arthritis/vasculitis/unclassified AID, organ-specific/connective tissue disease (CTD)/dermatosis, and patients without prior AID. FIGURE S2: Cumulative incidence of major disease outcomes according to prior autoimmune or inflammatory disease (AID). Cumulative incidence curves comparing patients with prior AID versus no prior AID for (A) myelofibrosis (MF) progression, (B) acute myeloid leukemia (AML) progression, (C) venous thrombosis, and (D) arterial thrombosis. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
Loscocco et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: