Hypofibrinogenemia and Disseminated Intravascular Coagulation Rarely Complicate Treatment-Naive Acute Lymphoblastic Leukemia

Introduction: Hypofibrinogenemia (HF) in acute leukemia has largely been attributed to disseminated intravascular coagulation (DIC). Pathological alterations in hemostasis through various mechanisms by circulating leukemia cells have been implicated in the pathogenesis of DIC, particularly in acute promyelocytic leukemia (APL) (Tallman et al., Leuk Lymphoma, 1993). In adult acute lymphoblastic leukemia (ALL), DIC is rare at presentation, but is significantly more common during induction chemotherapy, with varying degrees of reported morbidity and mortality (Higuchi et al., Leuk Lymphoma, 2005; Sarris et al., Leuk Lymphoma, 1996). Conversely, hyperfibrinogenemia at presentation has also been observed, possibly due to increased synthesis as an acute phase reactant in the setting of inflammation (Al-Mondhiry H et al., Cancer, 1975). Discrepant reports in plasma fibrinogen levels and potentially serious hemostatic complications led us to investigate the incidence of HF and DIC among adults with untreated ALL at our center.

Methods: Using a DataLine search and ICD-9 and ICD-10 codes (204.00, 204.01, 204.02, C91.00, C91.01, C91.02), all patients aged 18 years or older with a diagnosis of ALL treated at Memorial Sloan Kettering Cancer Center (MSKCC) between August 30, 1989 and April 27, 2018 were retrospectively identified. HF was defined as a fibrinogen level < 190 mg/dL. The diagnosis of DIC was made using the scoring system for overt DIC from the International Society on Thrombosis and Haemostasis (Toh et al., J Thromb Haemost, 2007). Patients who received any chemotherapy, including asparaginase and glucocorticoids, or who had a potential alternative cause of acquired HF such as sepsis or significant liver disease prior to their initial encounter were excluded. Data collection was approved by the MSKCC Institutional Review Board.

Results: A total of 275 adult patients with treatment-naive ALL were identified. The median age at presentation was 44 years. The study population consisted of 122 patients with B-ALL, 81 with T-ALL and 72 with ALL of unspecified lineage. Of these, 187 had a fibrinogen level obtained prior to initiation of therapy. Eight patients (4.3%) had HF at initial assessment with a mean level of 141 mg/dL. Of these 8 patients, 5 had B-ALL, 2 had T-ALL and 1 had ALL of unspecified lineage. Four of these 8 patients required fresh frozen plasma (FFP) and/or cryoprecipitate during their hospitalization for prevention of overt coagulopathy. Three of the 4 patients who required FFP and/or cryoprecipitate went on to receive an asparaginase containing induction regimen. One patient with B-ALL met criteria for overt DIC without clinical complications. One patient with HF experienced WHO grade 2 epistaxis and hematochezia. No deaths in our study population were attributable to DIC at presentation.

Discussion: Our study demonstrated that HF and DIC in treatment-naive ALL are uncommon. The rare incidence of DIC in our population is consistent with prior studies in adult ALL (Sarris et al., Blood, 1992). Our findings are also similar to those reported in patients with non-APL acute myeloid leukemia (Weltermann et al., Leukemia, 1998). In contrast, most patients with APL exhibit HF (Lou et al., Leuk Res, 2016). Proposed mechanisms of coagulopathy in APL include primary activation of fibrinolysis, as well as expression of tissue factor and cancer procoagulant by the malignant promyelocytes themselves; these mechanisms appear to be significantly less prevalent in ALL (Breen et al., Br J Haematol, 2012). DIC, when present at initial assessment, is generally subclinical and seldom leads to morbidity and mortality. Although the incidence of HF and DIC was low at diagnosis in our study population, we recommend pre-treatment screening for DIC to identify patients with overt coagulopathy at presentation and to establish baseline values, given the risk of coagulopathic complications during induction chemotherapy for ALL, particularly among patients receiving asparaginase products (Truelove et al., Leukemia, 2012). Further characterization of the ALL microenvironment in patients developing HF and DIC may elucidate mechanisms of procoagulant expression and increased fibrinolysis in this subgroup. Assessing predictors and clinical implications of baseline HF in adults with ALL in a larger series may help to guide early supportive management.