PRV is an MPN characterized by the clonal production of red blood cells (RBCs) that is almost always associated with the JAK2 V617F mutation. Patients often develop a concomitant increase in granulocytes and megakaryocytes (panmyelosis) in the early to middle stages of the disease, followed by cytopenias in the late spent or fibrotic phase. Affected patients are in their 60s, with a greater frequency in males than females. They may present with hepatosplenomegaly, headaches, dizziness, or signs of arterial or venous thrombosis. Gout may occur secondary to the increased cell turnover. Erythromelalgia, which is painful redness of the hands after heat exposure, is a common cutaneous finding.^1,3 Ichthyosis is a cutaneous disorder in which the skin becomes dry and hyperkeratotic, and its acquired form has been described in PRV as well as other inflammatory and neoplastic conditions.⁴ PRV patients often develop vascular thrombosis, thought to be due to increased red cell mass causing vascular stasis and tissue hypoxia. As such, patients may present with leg pain thought to be due to peripheral vascular disease. A livedo-reticularis type of skin reaction may be seen in patients presenting with cyanosis.⁵ Similarly, an acute myocardial infarction may be the initial presentation of PRV. A complete blood count showing increased RBC indices and JAK2 molecular assays may be helpful in reaching a diagnosis.⁶

Cutaneous infections occur in PRV and other MPNs under the rubric of neutrophilic dermatoses (NDs) such as pyoderma gangrenosum (PG) and acute febrile neutrophilic dermatosis (also termed Sweet syndrome [SS]). However, the underlying hematologic disorder of NDs most commonly is an acute leukemia. Patients with PG have solitary or multiple large, necrotic, and painful ulcers that present on the lower extremities, although the trunk, face, and arms may also be affected. SS, on the other hand, presents as multiple circumscribed painful red plaques or nodules. It may present on the extremities as well as the head and neck region. Histologically, both entities are characterized by dermal and subcutaneous neutrophilic infiltrates. One hypothesis for the appearance of pustular dermatosis is a defect in neutrophil superoxide anion (O₂⁻) generation. Support for this hypothesis stems from the fact that ND resolves after neutrophil function is improved via treatment with dapsone.⁷ Magro et al.⁸ also explored the possibility that PG and SS are characterized by clonally restricted neutrophilic infiltrates that develop secondary to an underlying hematologic disease, most often myeloid dyscrasias. Using human androgen receptor assays, they found that 20% to 40% of cases of PG and SS cases contained clonally restricted neutrophils and that this restriction either preceded or occurred subsequent to an underlying hematologic disease. However, after systemic workup, they also found cases that were not associated with hematologic disease. Nevertheless, a diagnosis of SS and PG should alert the clinician to rule out any underlying hematologic condition. An important entity to consider in the differential diagnosis of NDs is that of rheumatoid neutrophilic dermatitis, which presents as nodules and plaques on the extensor aspects of the upper and lower extremities. Histologically, it is identical to other NDs; the key distinguishing characteristic is positive serology for rheumatoid factor.⁹

Patients with PRV also have variable degrees of pruritus, which can precede the disease by years. Dermatologic examination may often reveal no grossly visible lesion, in which case it is attributed to an invisible dermatosis. Pruritus may occur at any time, but is often exacerbated after exposure to warm water. One recent study showed that the degree of epidermal and papillary dermal infiltration by mononuclear cells (including eosinophils) was higher in PRV patients exposed to warm water than in healthy control patients and PRV patients exposed to room temperature water. The degree of mononuclear and eosinophil infiltration correlated with the severity of pruritus, but the infiltrate itself was deemed to be reactive in nature.¹⁰

PRV may present with urticarial vasculitis, a condition that is seen in other entities such as viral infections (e.g., hepatitis B), collagen vascular diseases (e.g., systemic lupus erythematosus), and hematologic conditions (e.g., Hodgkin lymphoma). It is characterized clinically by painful, recurrent episodes of pruritic papules or wheals that last longer than 24 hours, which is different from other urticarial presentations. Histologically, urticarial vasculitis is characterized by endothelial cell swelling, extravasated RBCs, and perivascular neutrophilic infiltrate. Direct immunofluorescence may reveal immunoglobulin and C3 deposits within the walls of the blood vessels and at the dermal–epidermal junction. Treatment of the underlying PRV may resolve the vasculitis.¹¹

ET is characterized by a clonal proliferation of megakaryocytes as evidenced by persistent thrombocytosis (i.e., platelet count higher than 450 × 10⁹/L) in the peripheral blood. It occurs in patients in their 50s and 60s, although it has been reported in younger individuals. Men and women are equally affected. Causes of reactive thrombocytosis must first be excluded before a diagnosis of ET is considered, although demonstration of a JAK2 V617F mutation in the presence of isolated thrombocytosis is highly suggestive of ET. A BCR–ABL1 gene fusion is not detected. Bone marrow examination usually reveals increased atypical megakaryocytes with hyperchromatic, irregular nuclei imparting a “stag-horn” morphology.¹ Rarely, ET may transform into acute leukemia.¹²

CMs of ET show some overlap with those of PRV. Patients often present with signs of thrombosis and/or hemorrhage, which may accompany pain (erythromelalgia), especially in the upper and lower extremities and following heat exposure or exercise. Raynaud phenomenon, necrotizing vasculitis, urticaria, acrocyanosis, superficial thrombophlebitis, pruritis, livedo reticularis, ischemic gangrene, leg ulcers, hematomas, petechiae, ecchymoses, and purpura have all been described in patients with ET.² A hemorrhagic diathesis, although rare, occurs from altered function of von Willebrand factor when platelet levels are very high.¹³ Histologically, small-vessel thrombosis and endothelial hyperplasia are most often noted in the acute setting, with nonspecific findings in the chronic/resolved phase. The pathophysiology behind thrombosis is clonally stimulated dysfunctional thrombocytes with an increased affinity for aggregation and thus thrombosis. The workup of ET includes the exclusion of other causes of coagulopathy such as factor V Leiden deficiency. Treatment is aimed at reducing the number of platelets via cryoreductive agents such as HU, interferon α, pegylated interferon, and anagrelide, as well as preventing aggregation via aspirin. Urticarial vasculitis can also, on rare occasions, be found in patients with ET.¹⁴

PMF is a rare, clonal MPN characterized by the proliferation of atypical megakaryocytes and granulocytes in the bone marrow and is divided into a prefibrotic (cellular) phase and a fibrotic (paucicellular) phase. The fibrotic phase is characterized by bone marrow reticulin or collagen fibrosis, which often results in few, if any, bone marrow elements during aspiration (“dry-tap”). Any MPN can become fibrotic in the latter stages of the disease process and therefore may resemble PMF. It affects approximately 1 per 100,000 persons and occurs in the sixth to the seventh decade of life with an equal sex distribution. The hallmark of PMF is the presence of extramedullary hematopoiesis (EMH), the formation and development of blood cells in sites other than the bone marrow. It most often involves the liver, spleen, and lymph nodes.^1,15,16 It is rarely found in the skin, with only 45 reported patients to date,¹⁷ and occurs in approximately 0.4% of PMF cases.¹⁸ PMF is characterized by infiltration of the bone marrow by atypical megakaryocytes with hyperlobed, “cloud-like” nuclei that are often localized to the sinusoidal spaces. The distinction of PMF from CML is important, as PMF demonstrates a JAK2 V617F or MPL W K/L mutation, whereas CML has a characteristic BCR–ABL1 fusion.¹ More recently JAK2-negative cases of MPN (including PMF) have been shown to harbor mutations of the calreticulin gene.¹⁹

EMH is a normal physiologic phenomenon in the early fetal development up to the fifth month of gestation where dermal mesenchymal cells still have the ability to undergo hematopoietic differentiation.¹⁸ In the postnatal period, however, it is pathologic and associated with TORCH infections (toxoplasmosis, other agents, rubella/German measles, cytomegalovirus, and herpes simplex), twin transfusion syndrome, and rhesus hemolytic disease. Pediatric patients often present with a red- or magenta-colored maculopapular rash due to persistent dermal hematopoiesis, often referred to as “blueberry muffin syndrome.”²⁰ In adults, EMH may be reactive and seen in conditions leading to acquired or functional hyposplenism such as sickle cell disease, hereditary spherocytosis, immune thrombocytopenic purpura, and myelophthisic processes such as metastatic carcinoma involving bone marrow.¹⁶ However, EMH in adults is most often associated with an underlying MPN.²¹ Clinically, patients with cutaneous EMH may present with multiple red–purple nodules, papules, bullae, erythema, rash, and ulcers and thus may mimic inflammatory disorders and leukemia cutis (LC).²² Histologically, EMH is characterized by a nodular dermal or subcutaneous, predominantly intravascular infiltrate of atypical megakaryocytes with typical or atypical mitotic figures as well as a mixture of mature and immature myeloid cells and erythroid precursors. Occasional blasts may be present, but when the predominant cell is a blast, a leukemic transformation must be considered. Immunohistochemical studies may be useful in the distinction of the latter. Blasts are usually CD34, CD117, or TdT positive, whereas megakaryocytes are CD61 and CD42b positive and CD34, CD117, and TdT negative. O’Malley et al.¹⁶ have argued that the term “cutaneous EMH” is misleading and favored the term “neoplastic myeloid proliferations” when EMH is associated with an underlying hematologic disorder. Hoss and McNutt recommended the use of the term “cutaneous myelofibrosis” (CMF) as a more accurate term to describe cutaneous EMH associated with an underlying PMF in comparison to cutaneous EMH of a reactive nature. Support for CMF is provided by the fact that the same clone in the bone marrow has been identified in the skin as in the bone marrow. They postulated that the atypical megakaryocytes travel from the bone marrow to the skin via the vasculature and that this finding accounts for their intravascular location.¹⁵ Distinguishing reactive cutaneous EMH from CMF may be difficult (Table 54-2), yet of paramount significance as the latter finding may portend an aggressive clinical course of the underlying hematologic disease.¹⁷ CMF clinically presents as multiple firm dermal/subcutaneous nodules, similar to metastases, while reactive cutaneous EMH usually presents as mildly elevated papules. In addition, while CMF implies the presence of predominantly atypical megakaryocytes and immature myeloid/erythroid elements in a background of fibrosis, reactive cutaneous EMH is associated with a predominance of mature myeloid/erythroid elements in a nonfibrotic background. While lesions of reactive cutaneous EMH spontaneously resolve, those of CMF rarely do, and as such portend an aggressive clinical course with median survivals of only a few months. Lastly, CMF demonstrates a JAK2 mutation, while reactive cutaneous EMH does not (Figs. 54-1 and 54-2).²³