Incidence

The incidence of SOS varies greatly in the literature, because of controversies in diagnosis, variable diagnostic criteria, and confounding factors, such as the patient population being studied. The published rates of SOS in all populations range from 0% to 60% of HSCT patients, but the mean incidence in a recent systematic review was 13.7%. The incidence of SOS clearly varies by type of transplant; rates in fully myeloablative allogeneic transplants are 10% to 60%, whereas in reduced-intensity conditioning regimens and autologous transplants, the incidence is much lower, ranging from 5% to 30%.

Adults

The average incidence of SOS in adults is 8% to 14% depending on the series and in a few studies is as high as 40%; this is much lower than observed in children. The incidence of SOS has steadied now that there is a greater emphasis on use of reduced intensity conditioning regimens.

Pediatrics

In contrast to adults, SOS is a common complication of HSCT in children, occurring in approximately 20% to 30% of patients overall and as many as two-thirds of those at high risk for this complication. One of the challenges of making this diagnosis in the pediatric population is the fact that one-third of children with SOS do not have hyperbilirubinemia; ^, therefore dependence on strict fulfillment of the Baltimore or modified Seattle criteria may miss the diagnosis, and children are more likely to present with late-onset SOS (20%). ^{, ,} Pediatric patients who do develop hyperbilirubinemia usually do so late in their disease progression or as a manifestation of more severe SOS. An early diagnostic clue in the pediatric population is thrombocytopenia that is refractory to platelet transfusion; this usually precedes the diagnosis by 1 week. Children also can have more severe disease, with 30% to 60% developing multiorgan dysfunction. ^,

Patient related

Extremes of age increase the risk of SOS in both adults and children, ^, as does poor Karnofsky performance status (< 90%) ^, or dependence on parenteral nutrition before transplant. Advanced malignant disease, and the etiology of the underlying malignancy, are well-recognized risk factors. ^, In children, a diagnosis of osteopetrosis, familial hemophagocytic lymphohistiocytosis, juvenile myelomonocytic leukemia, thalassemia major, sickle cell disease, neuroblastoma, Wilms tumor, or rhabdomyosarcoma have all been associated with a higher risk of SOS. Laboratory abnormalities indicative of liver injury or inflammation, such as elevated transaminases, ferritin, or bilirubin, also are associated with a higher SOS risk. Several genetic mutations predisposing to the condition have been identified, including GSTM1-null, C282Y allele, and the MTHFR 677CC/1298CC haplotype. Patients with preexisting liver disease caused by viral hepatitis or iron overload are also at elevated risk for this complication. Although it had been suggested in the past that there may be a gender difference in SOS, the increased risk for SOS development seems to be limited to women maintained on norethisterone to prevent gynecologic bleeding.

Pretransplant treatment related

There are several pretransplant treatment characteristics associated with an elevated risk of SOS, including prior abdominal or liver radiation and baseline abnormalities in transaminases or bilirubin. ^{, ,} Certain salvage chemotherapies used in leukemia to control disease and allow patients to proceed to HSCT deserve special mention. There are two antibody-drug conjugates that contain the same toxin, ozogamicin, that have been associated with SOS. Gemtuzumab ozogamicin (anti-CD33) was previously used in acute myeloma leukemia patients and was associated with SOS. ^, In patients exposed to gemtuzumab ozogamicin before HSCT, the risk of SOS was increased and fatalities resulting from SOS noted. However, the use of the lower dose regimen may be associated with an acceptable SOS risk of around 8%. This agent was taken off the market because of lack of use and toxicity concerns, but has recently been reapproved at a new dose. Inotuzumab (anti-CD22) ozogamicin has been approved for the salvage treatment of acute lymphoblastic leukemia and is associated with a 17% risk of SOS after HSCT.

Transplantation related

In general, more intensive consolidating regimens and less well-matched transplants increase the risk of SOS. SOS is more common in fully myeloablative regimens than in reduced-intensity regimens, but has been described as a complication of autologous transplants. Use of fludarabine, cyclophosphamide, high-dose melphalan, and busulfan for consolidation have all been associated with SOS; intravenous administration of busulfan may decrease the risk because of improved pharmacokinetic profiling of the dose. High-dose or unfractionated total body irradiation increase SOS risk. Second HSCT transplant recipients are also at increased risk for SOS because of their larger cumulative chemotherapeutic and radiation burden.

Although SOS can occur in autologous SCT, it is much more common in allogeneic transplants. Unrelated donors, donors with a greater degree of human leukocyte antigen (HLA) mismatch, non-T cell depleted grafts, and peripheral blood stem cell transplants are all associated with an increased risk for development of SOS. ^, Posttransplant factors, such as choice of graft-versus-host disease (GVHD) prophylaxis, can also have an effect, with patients maintained on a calcineurin inhibitor, especially if given in combination with sirolimus and methotrexate, being at higher risk.

Laboratory tests

Adult patients with SOS often have elevated transaminases, hyperbilirubinemia, prolonged prothrombin time, elevated international normalized ratio, and low serum albumin. Approximately one-third of pediatric patients with SOS do not have hyperbilirubinemia.

Biomarkers

Although the diagnosis of SOS remains a purely clinical one, there have been attempts to identify noninvasive biomarkers for the disease. Plasminogen activator inhibitor-1, an inhibitor of the fibrinolytic system, was one of the first biomarkers identified; levels increase in SOS patients concurrent with the rise in serum bilirubin. Levels of thrombomodulin, von Willebrand factor (vWF), and soluble intercellular adhesion molecule-1 have all been shown to be elevated early in patients with SOS maintained on sirolimus for GVHD prophylaxis. In one study, levels of vWF 1400 IU/mL or greater and thrombomodulin 100 ng/dL or greater on day +7 were 100% sensitive and specific for the diagnosis of SOS. Using quantitative mass spectrometry based proteomics, Akil et al. developed a diagnostic biomarker panel for SOS; suppression of tumorigenicity-2, angiopoietin-2, L-ficolin, hyaluronic acid, and vascular cell adhesion molecule-1 (VCAM-1) levels together were diagnostic of SOS with an area under the curve (AUC) of more than 0.81. These markers were able to identify patients at risk for SOS as early as the day of stem cell transplantation. In addition, L-ficolin, hyaluronic acid, and VCAM-1 levels predicted the severity of SOS disease with greater than 80% accuracy, when combined with clinical data.

Liver biopsy

Transjugular liver biopsy with portal pressure measurement can be used to make a definitive diagnosis of SOS. Classic biopsy findings include acinar zone 3 necrosis and a wedged hepatic venous pressure gradient more than 10 mmHg is highly suggestive of the diagnosis. However, liver biopsy is an invasive procedure with significant risks of bleeding and the possibility of misdiagnosis because of sampling error; therefore this approach is not commonly used.

Transient elastography

In much of general hepatology practice, liver biopsy has been supplanted by noninvasive measures of liver fibrosis, including transient elastography, which uses ultrasound waves to measure liver stiffness; this test is highly congruent with biopsy results, especially for patients with advanced fibrosis. ^, At least two studies have investigated the utility of transient elastography in the diagnosis of SOS in pediatric HSCT patients. In one study of 25 patients, five developed SOS. Elastography measurements preconditioning did not differ between the groups, but as early as day +5, there was a significant increase in measured liver stiffness in the patients who eventually developed SOS; SOS was diagnosed at day +14 to 19 among these patients on the basis of modified Seattle criteria. The authors report that a velocity increase of 0.26 m/s over the baseline reading on day +5 was 60% sensitive and 90% specific for SOS. A second pediatric study with 22 patients (four of whom developed SOS) noted similar findings; there was again no difference in baseline elastography among patients but those who developed SOS had an increase in liver stiffness 3 to 6 days before their clinical diagnosis. In the adult population, a pretransplant liver stiffness of greater than 8 kPa was associated with significant hyperbilirubinemia posttransplant and posttransplant liver dysfunction, suggesting a role for transient elastography in identifying patients at risk for SOS in whom prophylactic defibrotide could be considered.

Ultrasound

Although ultrasound findings are not part of the official diagnostic criteria, they can assist in making the diagnosis and are useful for ruling out other etiologies in the differential diagnosis. On ultrasound, patients can have ascites, hepatomegaly, gall bladder wall thickening, small portal vein diameter, elevated hepatic artery resistive index (> 0.75), or reversal of hepatic venous flow.

Disease severity grading

SOS can be classified as mild, moderate, or severe ( Table 12.4 ). Patients with mild SOS can be managed with supportive care and their disease is usually self-limiting; in one series, 9% of patients with mild SOS died of their disease. Those with moderate disease will benefit from analgesics, diuretics to manage fluid balance, and supportive measures; mortality in this group is around 25%. Patients with severe SOS often have renal dysfunction and/or multiorgan system failure, with attendant mortality rates upwards of 80%.

The EMBT, in addition to updating the diagnostic criteria for SOS, have also proposed an updated grading system for establishing the severity of disease in adults, which is modeled after the Common Terminology Criteria for Adverse Events reporting structure, ranging from grade 1 (mild) to grade 5 (death). This new system not only considers absolute laboratory values in grading disease severity, but considers the kinetics of laboratory and symptoms development; it also upgrades patients who are in between categories to a more serious classification of disease. The rationale for these changes was to prompt clinicians to institute therapy sooner, as earlier treatment with defibrotide has been associated with improved survival in SOS. Patients with mild disease have had a slow onset of symptoms, with bilirubin between 2 and 3 mg/dL, transaminases that are less than twice the upper limit of normal, weight gain less than 5%, and creatinine that is less than 20% above their baseline. Moderate disease has a quicker onset (5–7 days), with a bilirubin between 3 and 5 mg/dL, transaminases 2 to 5 times the upper limit of normal, weight gain between 5% and 10%, and no more than a 50% increase in serum creatinine. Patients with severe disease have a rapid onset, less than 4 days, with bilirubin in the 5 to 8 mg/dL range, with doubling in the past 48 hours, along with transaminases 5 to 8 times normal, weight gain less than10%, and creatinine may have doubled. Very severe disease is characterized by the presence of organ failure/dysfunction, with bilirubin greater than 8 mg/dL, transaminases eightfold above normal, more than 10% weight gain, and the need for renal replacement therapy.

There is also an updated EBMT severity grading scale for SOS in children, with disease states categorized as mild, moderate, severe, and very severe/multiorgan failure. In mild disease, liver function tests (LFTs; aspartate transaminase, alanine transaminase, and glutamate dehydrogenase) are ≤ 2 times the upper limit of normal, thrombocytopenia has been present for less than 3 days, there is minimal ascites, glomerular filtration rate (GFR) is 89 to 60 mL/min, and there is minimal to no oxygen requirement. In moderate disease, LFTs are 2 to 5 times normal, thrombocytopenia has been present for 3 to 7 days, there is moderate ascites, GFR is 59 to 30 mL/min, and supplemental oxygen is required. Severe disease presents with LFTs greater than 5 times normal, more than 7 days of resistant thrombocytopenia, bilirubin 2 mg/dL or more, ascites requiring paracentesis, coagulopathy, GFR 29 to 15 mL/min, and ventilatory support is required. Very severe disease is similar with doubling of bilirubin in 48 hours, replacement of coagulation factors, renal replacement therapy, and cognitive impairment.